•^: 'A.. ■ I , » THE VOYAGE OF H.M.S. CHALLENGER. ZOOLOGY-VOL XX. REPORT ^ ^ ON THE SCIENTIFIC RESULTS OF THE VOYAGE OF H.M.S. CHALLENGER DURING THE YEARS i87 3-7 6 UNDER THE COMMAND OF Captain GEORGE S. NARES, R.N.. F.R.S. AND THE LATE Captain FRANK TOURLE THOMSON, R.N. PREPARED UNDER THE SUPERINTENDENCE OF THE LATE Sir C. WYVILLE THOMSON, Knt, F.R.S., &c. REGIUS PROFESSOR OF NATURAL HISTORY IN THE UNIVERSITY OF EDINBUROH niRECTOR OF THE CIVILIAN SCIENTIFIC STAFF ON BOARD AND NOW OF JOHN MURRAY ONE OF THE NATURALISTS OF THE EXPEDITION Zoology — Vol. XX. Pul)lis|)erj bp (QrUev of \^zx jWajestp's (Soliernmeut PRINTED FOR HER MAJESTY'S STATIONERY OFFICE AND SOLD BY LONDON :— EYRE & SPOTTISWOODE, EAST HARDING STREET, FETTER LANE EDINBURGH :— ADAM & CHARLES BLACK DUBLIN :— HODGES, FIGGIS, & CO. 1887 Price Forty SJnlhngs. PRINTED BY NEILL AND COMPANY, EDINBURGH, FOR HEK majesty's STATIONERY OFFICE. CONTENTS. I. — Report on the Monaxonida collected by H.M.S. Challenger during the years 1873-1876. By Stuart 0. Ridley, M.A., F.L.S., late Assistant in the Zoological Department of the British Museum, and Arthur Dendy, B.Sc, F.L.S., Assistant in the Zoological Department of the British Museum. {The Alanuscrijit loas received in Instalments hetiveen 19 th November 1886 and I3th May 1887.) II. — Report on the Myzostomida (Supplement) collected by H.M.S. Challenger during the years 1873-1876. By Dr. L. von Graff, Professor in the University of Graz, Austria. (The Manuscript loas received on 10th March 1887.) III. — Report on Cephalodiscus dodecalophus, a new type of the Polyzoa, procured on tlie Voyage of H.M.S. Challenger during the years 1873-1876. By William C. M'Intosh, M.D., LL.D., F.R.S., &c.. Professor in the University of St. Andrews. {The Manuscript ivas received in Instalments between 25th April 1887 and 19th May 1887.) 790^)2 EDITOEIAL NOTES. Tins Volume contains Parts LIX., LXI., and LXII. of the Zoological Series of Eeports on the Scientific Eesiilts of the Expedition. Part LIX. — The Collection of Monaxonida was in the first instance intrusted to Mr. Stuart 0. Ridley, F.L.S., of the British Museum, for examination and description. Mr. Ridley had made very considerable progress in the work when he was compelled, owing to ill health, to inform me that he could not complete the Report within the specified time. In these circumstances I suggested that Mr. Arthur Dendy, B.Sc, F.L.S., should be associated with him in the work, to which proposal Mr. Ridley at once assented. The result has been the present highly satisfactory Memoir on this group of sponges, consisting of about 350 pages, illustrated by 51 lithographic plates, woodcuts, and a map. Part LXI. — This Part is a Supplementary Report on the Myzostomida by Dr. L. von Graff, Professor of Zoology in the University of Graz, Austria, and contains descriptions of seven new species, and also references to the litera- ture of the group which has appeared since the publication of the First Part of the Report. It contains sixteen pages and four plates. The First Part of Dr. L. von Grafl''s Report forms Part XXVII. of the Zoological Series of Reports. Part LXII. — This Report consists of a description by Professor William C, M'Intosh, F.R.S., of Cephalodiscus dodecalophus, one of the most curious Vlll THE VOYAGE OF H.M.S. CHALLENGER. and interesting organisms dredged by the Expedition, with an Appendix on its affinities by Sidney F. Harmer, B.A., B.Sc., Fellow of King's College, Cambridge. It contains forty-eight pages and is accompanied by seven plates and woodcuts. JOHN MURRAY. Challenger Office, 32 Queen Street, Edinburgh, 22nd July 1887. ERRATA.— PART LIX. Page 12, line 19 from top, for " SchmicUia variabilis" read " Petrosicc variaUlis." Page 17, line 2 from bottom, omit the words "and from." Page 29, line 21 from top, for " Tuhulodigitus " read " Patuloscula" and for " p. 867 " read " p. 365." Page 38, lines 9 and 13 from top, for " Gellius pyriformis " read " Gdlius sp." Page 46, line 10 from bottom, /or "oxeote" read " strongylote." Page 82, line 15 from bottom, >• "lat. 48° 30' O" S." rmd "lat. 0° 48' 30" S." Page 102, line 18 from top, and elsewhere, for "Meliiderma stiintata" read "Meliiderma stipitatum." Page 129, line 12 from bottom, for " Espcrinte" read " Espercllinm." Page 220, line 7 from top, and elsewhere, for " Tricliostemma irrcgidaris " read " Trichostemma irrajularc." Page 225, line 18 from top, for " Suberites ra7nulosa " read " Suberites ramulosus." Page 242, line 13 from top, for " Myxilla plumosa " read " MyxAlla (?) ptlumosa." Page 242, line 15 from top, omit the words " Plumohalicliondria p)lumosa." Page 243, last line, for " Cladorhiza tridcntata " read " Cladorhiza (?) tridcntata." THE YOYAGIE OF H.M.S. CHALLENGEK. ZOOLOGY. REPORT on the Monaxonida collected by H.M.S. Challenger during the Years 1873-76. By Stuart O. Ridley, M.A., F.L.S., late Assistant in the Zoological Department of the British Museum, and Arthur Dendy, B.Sc., F.L.S., Associate of the Owens College, and Assistant in the Zoological Department of the British Museum. PREFACE. In presenting this Eeport it seems desirable to give some explanation with regard to the arrangement of the work. We have always felt strongly the great importance of minute anatomical and histological investigation in describing sponges. Owing, however, partly to the unfavourable condition for histological purposes of a large proportion of the material at our disposal and partly to the limited space of time available for the completion of the work, it has been found impossible to work out the minute anatomy and histology of all the species in the collection. Hence we have found it advisable to investigate in detail some of the leading types only, and to confine our remarks on the subject to a special chapter in the Introduction, making them at the same time rather of a general nature. We have to express our sincere thanks to Mr. John Murray, the Editor of the Challenger Reports, in the first place for placing so valuable and interesting a collection in our hands, and in the second place for the great indulgence which he has shown us while preparing the Report. We have also especially to thank Professor Flower, F.R.S., Director of the Natural History Department of the British Museum, and (ZOOL. CHALL. EXP. — PART LIX — 1887.) NuD « li THE VOYAGE OF H.M.S. CHALLENGER. Dr. Giinther, F.R.S., the Keeper of the Zoological Department, for the facilities which they have placed at our disposal while working in the Museum, and for the use of the rich national collection, without which it would have been scarcely possible to carry on the work. We are further under great obligations to Professor A. M. Marshall, F.R.S., of the Owens College, and to the authorities of the Normal School of Science and Royal School of Mines, especially Mr. G. B. Howes, F.L.S., for kind permission to make use of their laboratories and appliances for purjaoses of anatomical and histological investigation. With reference to his share in the preparation of this Report Mr. Ridley desires to add the following note : — " Works like the present, produced by a joint authorship, differ in some respects from those otherwise produced, and I wish to say a few words with regard to my own connection with the Report, and especially to the views expressed in it. " At the time at which it became necessary for another worker to share the labour in order satisfactorily to complete the work, I had done no more than study as carefully as my somewhat limited time and means allowed the classificatory characters of most of the species, and in a few cases also their minute anatomy and histology, and had decided on the novelty or the nomenclature (in the case of the old species) of most of the forms below treated of The results thus obtained proved to require considerable modification, and the credit of presenting the work in its present form is mainly due to Mr. Dendy, who has studied afresh all the species and mastered their characteristics, besides undertaking by far the greater part of the labour involved in the ' production ' of the Report. " With regard to theoretical views, where any such are referred to in the following pages, either explicitly or by implication, my own j^osition is that references to 'affinity,' 'genetic relationship,' 'development' 'acquisition,' &c., are only to be taken, so far as I am individually concerned, as convenient methods of expressing the phenomena presented to us, and as involving merely hypothetical interpretations of the processes which may be conceived to have taken place in the past history of these organisms ; the theory of evolution appears to me to give an extremely reasonable and very possibly true, but as yet not fuUy demonstrated explanation of these phenomena, except, perhaps, in the case of some book-species, which ought not to be separated from each other. I hold the view that ' theories are convenient bases on which to group facts,' but further consider that facts acquire an additional interest when they are studied with the object of testing theories, and perhaps attain their greatest importance when it is possible to use them inductively, viz., for the construction or confirmation of general laws." TABLE OF CONTENTS. PAGE INTRODUCTION, v-lxviii Chapter I.- — The History of the Monaxonida as a group, V Chapter II. — Anatomy and Histology, and Embryological Notes xii Chapter III. — The Classification of the Monaxonida, liii DESCEIPTION OF GENERA AND SPECIES, 1-240 Suborder I. Halichondrina, .... 1 Family I. Homorrhaphidse, 1 Subfamily 1. Renierinas, 1 Halichondria, Fleming, 1 Petroda, Vosmaer, 9 Bemera, Nardo, 14 Subfamily 2. Chalininoe, 18 Pachijchalina, Schmidt, 19 Clialina, Grant, 25 Siplionoclialina, Schmidt, 29 Family II. Heterorrhaphidse, 31 Subfamily 1. Phloeodictyinse, 31 Rhizochalina, Schmidt, 32 Oeemiapia, Norman, . 36 Subfamily 2. GeUiinae, .... 37 Gelliits, Gray, 37 Gelliodes, Ridley, 47 Toxoclialina, Ridley, . 49 Subfamily 3. Tedaniinse, 50 Tedania, Gray, 50 Trachytedania, Ridley, • 57 Subfamily 4. Desmacellinas, 58 Desmaeella, Schmidt, . 58 Subfamily 5. Hamacanthinae, 59 Vomerula, Schmidt, . ■ 59 Family III. Desmacidonidae, 62 Subfamily 1. Esperolliuce, 62 Esperella, Vosmaer, 62 Esperiopsis, Carter, 76 Cladorhiza, M. Sars, . 86 Axoniderma, Ridley and Dendy, 96 IV THE VOYAGE OF H.M.S. CHALLENGER. Family IIL Desmacidonidae. Subfamily 1. Esperellinas — continued. Clwndrodadia, TVyville Thomson, Meliiderma, d. gen., Desmacidon, Bowerbank, Artemigina, Vosmaer, . Phelloderma, Eidley and Dendy, Sideroderma, Eidley and Dendy, loplion. Gray, . Ampliilectus, Vosmaer, Subfamily 2. Ectyoninae, Myxilla, Schmidt, Clathria, Schmidt, Rhaplndophlus, Ehlers, Plu'/nohah'chondria, Carter, Plocamia, Schmidt, Acamus, Gray, Eeliinoclathria, Carter, Affelas, Duchassaing and Michelotti, EcMnodidijum, Eidley, Family IV. Axinellidse, Hymeniacidon, Bowerbank, PJialellia, Bowerbank, Ciocalypta, Bowerbank, AeantheUa, Schmidt, . Axinella, Schmidt, Paspailia, Nardo, Dendropsis, Eidley and Dendy, Tlirinacopliora, Eidley, Suborder II. Clavclixa, Family I. Suberitidae, Suherites, Xardo, Polymastia, Bowerbank, Protdeia, Dendy and Eidley, Trichostemma, M. Sars, Tentorium, Vosmaer, . Styloconlyla, Wyville Thomson, Quasillina, Norman, . Cliona, Grant, Family II. S p i r a s t r e 1 1 i d se, . Spirastrella, Schmidt, . Latrunculia, Bocage, . GEOGRAPHICAL AND BATHTMETEICAL DISTEIBUTION, APPENDIX — Synonymy of the Spicular Terminology, INDEX, ....... P.\GE 99 102 103 112 113 lU IIG 123 128 128 li6 151 155 157 159 159 163 164 166 166 169 173 176 178 188 191 193 196 197 197 210 214 216 221 222 225 227 229 229 233 241-263 . 264 265 INTRODUCTION. CHAPTEE I.— THE HISTORY OF THE MONAXONIDA AS A GEOUP. The Monaxonicla comprise by far the most commonly met with and abundant of all sponges. They occur in greater or less profusion in all parts of the world, but are more especially shallow-water forms. They may be collected between tide-marks almost anywhere, and on parts of our own coast are thrown up in great quantities after every gale. With all these points in favour of the collection of abundant material for study, we might naturally be led to expect that our knowledge of the group would be in a corre- spgndingly satisfactory condition. Yet this is not so. On the contrary, there are few (if any) groups of equal importance in the whole Animal kingdom which have been so little studied and with so little success. If we seek a reason for this state of things it is to be met with largely in the great and exceptional difficulties which beset all attempts at a satisfactory classification. According to one of our oldest and most experienced workers at sponges, " the subject is actually repulsive from its difficulties ; " ^ but we do not, ourselves, take quite such an extreme view of the case. In the first place it is impossible to classify the group according to external characters, for often the external appearance of one and the same species varies almost indefinitely with its varying surroundings, and often two species, sometimes even of distinct genera, resemble each other so closely that it is impossible to distinguish them with certainty without microscopic examination. Hence nothing could be expected of the days before microscopic research, and, as a matter of fact, we find that nothing of importance was done. In the next place many of the distinguishing characters are so minute and difficult to observe, depending as they do upon the forms of the smaller spicules, that even with the aid of recent improvements in the microscope our best observers have often overlooked or misinterpreted them. Lastly, we must remember the fact that aU the Monaxonid sponges which an ordinary observer is likely to meet with appear at first sight very uninteresting, and off"er very little inducement to careful study. It is not 1 Carter, Notes Introductory to the Study and Classification of tlie Spongida, Ann. and Mag. Nat. Hist, ser. 4, vol. xvi. p. 3. VI THE VOYAGE OF H.M.S. CHALLENGER. until we are able to examine them microscopically, after an elaborate and troublesome course of preparation, that the real interest of the sponge, namely, the minute anatomy and histology of the soft parts, shows itself. It is probable that one of the chief causes which have prevented the group from being hitherto almost entirely ignored is the singular beauty of the spicules as "objects for the microscoiDC " ; now, however, many observers have taken up the subject from a more scientific point of view, and we may hope for a rapid advance in this department of science. Only within quite recent times have the Monaxonida begun to have a history of their own apart from the history of sponges at large, and as the latter is a subject which has already been more or less fully treated of by previous authors, such as Schmidt,^ Vosmaer,^ and von Lendenfeld,^ we shall here mention only those facts which bear more directly upon our group. To Zittel is due the credit of having first separated the Monaxonida from the Hexactinellida and Tetractinellida, as a distinct group of siliceous sponges, under the name Monactinellidae. This he did in 1878, in his paper Zur Stammegeschichte der Spongien ; * where he proposes the following classification of the sponges : — Classe : Sponfjise oder Porifera. \. Ordnung : Myxospongix. Haeck. (Carnosa Cart.). 2. Ordnung : Cei-aospongix. Bronn < ^ a .;. ( (Psammonemata Cart). 3. Ordnung: Monadinellids. Zitt. Kieselspongien mit einaxigen Nadeln. (Kaphidonemata, Ecliino- nemata und Holoraphidota z. Th. Carter). 4. Ordnung : Tetractinellidx. Marshall. Kieselspongien mit aiikerformigen Nadebi oder mit den Nadeln des pyramidalen typus (Holoraphidota z. Th. Carter). 5. Ordnung : Lithistidx. 0. Schmidt (Holoraphidota z. Th. Carter). 6. Ordnung : HexadinelUdx. 0. Schmidt (Carter). 7. Ordnung : Calcispongiee. Blv. (Calcarea Cart.). In 1883^ the name Monactinellid* was altered by SoUas to Monaxonidse, as the former implied a viTong idea, viz., that the chief spicules of the group are " monactinal," that is to say, consisting of only a single ray, while they are just as often " diactinal " or composed of two rays diametrically opposite to one another. Both these forms of spicule are, however, " monaxonid," that is to say they have each only one axis, which, in the case of the diactinal forms passes through both the rays. Hence the slight alteration in name appears to be justified. It has been further emended by subsequent writers to " Monaxonida," as " ida3 " is the accepted termination for the name of a family. ' Spong. Adriat. Meeres, p. 1. 2 Article Porifera, in Bronn's Klassen und Ordnuugen des Thierreichs. 3 Proc. Linn. See. N.S.W., vol. ix. pt. i., p. 122. * Festgabe der philosophischen Facultiit zum 50 jiihrigen Doctorjubilaum des Professor von Siebold, Miinchen. * Cassell's Natural History, 1883, vol. vi. p. 326. REPORT ON THE MONAXONIDA. Vll It is worth our while to investigate in this place the treatment which the Monaxonida, thus constituted by Zittel as a distinct group of sponges, have met with at the hands of other recent authors. Commencing with Bowerbank, we find that his classification of sponges has the one advantage of simplicity, though, based as it is almost solely upon the arrangement of the skeleton, without any regard to the anatomy of the soft parts or to the forms of the spicules (except as specific characters), it has led to the most absurd results and has been followed by no one. Briefly, then, it is as follows:^ — Class Porifera. Order 1. Calcarea. Order 2. Silicea. Order 3. Keratosa. The " Silicea " and " Keratosa " are further divided into suborders (but these have no names) and then straightway into genera. The Monaxonida are chiefly to be found amongst the Silicea, being dispersed through five suborders ; but two genera, Clialina and OpJilitaspongia, are relegated to the Keratosa. Gray took the field armed with a much more elaborate scheme, which cannot be said to possess even the advantage of simplicity ; it was as follows :" — 1. Spongiadse. 2. Ceratelladae. 3. Hirciniadce. Class Pcn-iphora. Subclass 1. P. Silicea. Section A. Thalassospongia. Subsection 1. Leiospongia. Order 1. Keratospongia. Families — • 4. Dysideidae. 5. Chalinidae. 6. Phakelliadae. 7. Halichondriadce. 8. Polymastiadse. 9. Ophistospongiadse. 1. Suberitidse. Order 2. Suberispongia. Families— 2. Eaphiopboridae. 3. Clioniadffi. 1 Vide Men. Brit. Spong., vol. ii. p. xx. ' The arrangement of the subclass Silicea here quoted, is to be found in Gray's paper— Notes on the Classification of Sponges, in Ann. and Mag. Nat. Hist, ser. 4, vol. ix. p. 446, 1872. The Calcarea are not dealt with in this, Gray's second, scheme, but his anangement of them will be found in his first scheme in Proc. Zool. Soc. Lond., 1867, p. 553. Vlll THE VOYAGE OF H.M.S. CHALLENGER. 1. Esperiadfe. 2. Desmacidonidffi. Order 3. Arenospongia. Family 1. Xenospongiadce, Subsection 2. Acanthospongia. Order 4. Hamispongia. Families — 3. Hamacanthidee. 4. Gelliadse. L PteronemadsB. 2. Lanuginellidse. 3. EuplectelladEB. 4. Hyalotliaumadae. Order 5. Coralliospongia. Families — 5. Macandrewiadse. 6. Farreadge. 7. Dactyloealycidfe. 8. Aplirocallistidae. 9. Corbitellidfe. 10. Askonematidse. IL Carteriadse. 12. Axidae. 1. Geodiadfe. 2. Placospongiadae. 3. Tethyadse. Order 6. Spliserospongia. Families — 4. Donatiadse. 5. Theneadee. 6. Lophurellidae. 7. Casuladse. 8. ChondrilladK. 9. AncorinidEe. 1. Granfciadae. Section B. Potamospongia. Family 1. Spongilladse. Subclass 2. P. Calcarea. Families — 2. AlcyonceUidee. 3. Aphroeerasidaj. Yet in spite of its cumbersome and largely artificial character, good features in this classification are not wanting ; the best, perhaps, being that the Keratosa are here already included amongst the Silicea, a step which has been fully justified by recent research. To Gray also is due the credit of having appreciated the value of spicular forms as a means of distinguishing genera. Of Gray's families the following belong to the Monaxonida : — Clialinidx. PhaheUiadx. Halichondriadx. Polymastiadai. OpMstospongiadee. Suberitidx. Raph iophoridx. Clioniadse. Xenospongiadx. Esperiadx. Desmacidonidx. Hamacanthidx. Gelliadx. Spongilladx. 1 Axidx. REPORT ON THE MONAXONIDA. ix Schmidt, in his work on the Atlantic Sponges/ makes the following main divisions, which represent his latest independent classification : — 1. Hexactinellidse. 2. Lithistidx. 3. Halisareinx and Gumminex. 4. Geraospo7igix. 5. Chalinex. 6. Renierinse. 7. Suheritidinse. 8. Desmacidinx. 9. Ulialinopsidinx. 10. Ancorinidx. 11. Geodinidx. 12. Calrisjinngix, Here Groups 5, 6, 7, 8 and 9 correspond to Zittel's Monactinellidfe (Monaxonida), and, at a later date (1880), as has already been pointed out by Yosmaer, Professor Schmidt adopts Zittel's modification, grouping together Nos. 5 to 9 as Monactinellidse, and 10 and 11 as Tetractinellidse.^ We next come to Mr. Carter's classification of the Sponges, which is as follows : ^ — Class SPONGIDA. Order 1. Cuniosa. Families — 1. Halisarcida. 2. Gumminida. Order 2. Geratina. Families — 1. Luffarida. 2. Aplysinida. 3. Pseudoceratida. Order 3. Fsarnmonemata. Families^ — 1. Bibulida. 2. Hircinida. 3. Pseudohircinida. Order 4. BapMdonemata. Families — 1. Chalinida. 2. Cavoolialinida. 3. Acervochalinida. 4. Pseudochalinida. Order 5. Echinonemata. Families — 1. Ectyonida. 2. AxinelUda. Order 6. Holoraj-iliidota. Families — 1. Eerderida. 2. Suberitida. 3. Pachytragida. 4. Pachastrellida. 5. Potamospongida. Order 7. Hexactinellida. Families — 1. VitreohexactineUida. 2. SarcohexactineUida. 3. Sarcovitreohexactinellida. In this scheme the following families are Monaxonida : — Chalmida. Cavochalinida. A eervochalinida. Pseudochalinida. Ectyonida. AxinelUda. Renierida. Suberitida. Potamospongida. All these schemes of classification have been fully discussed by Dr. Vosmaer in his useful and comprehensive work on the Porifera, in Bronn's Klasseu u. Ordnung. des Thierreichs, a work which has been of great service to us in compiling these pages ; hence no discussion is needed here, and we give them, as shortly as possible, only 1 Grundzuge einer Spongien-Fauna des Atlantischen Gebietes, 1870. 2 Spoug. Meerb. von Mex., pt. 2. 2 Ann. and Mag. Nat. Hist., ser. 4, voL xvi. p. 126. (zooL. chall. EXP. — part LIS. — 1887,) Nnn h X THE VOYAGE OF H.M.S. CHALLENGER. for the sake of showing what has lieeu the fate of the Monaxonida at the hands of the more recent authorities on the subject. We must now investigate the position assigned to the group by Dr. Vosmaer himself. His system, based as it is upon all that was good of the various old schemes, is naturally a great improvement upon any of its pre- decessors. In brief outline it is as follows : — PORIFERA. Class I. POEIPEEA NON-CALCAEEA. Order L Hyalospongise ( = Hexactinellida). Order 2. Spiculispongife. Suborder I. Lithistina. ,, II. Tetraotina. „ III. Oligosilicina. „ IV. Pseudotetraxoiiia. ,, V. Claviilina. Order 3. Cornacuspongise. Suborder I. Halichondrina. „ II. Ceratina. Class II. POEIFERA CALCARBA. Order 1. Homoccela. Order 2. Heterocoela. Here the two suborders Clavulina and Halichondrina correspond to the Monaxonida, and in the present work these two suborders are maintained as the two primary divisions of the group, though it has been found necessary, after a careful study of the large amount of material at our disposal, to modify somewhat the arrangement of the smaller subdivisions. Dr. Vosmaer, we may here mention, subdivides the Clavulina and Halichondrina as follows : — Suborder Clavulina. Families — 1, Polymastidse. 2. Suberitidae. (3. ClioniadEe.) Suborder HaUchondrina. Families — 1. Halicbondridse. 2.^Spoiigillid8e. 3. Desmacidonidae. 4. Ectyonidse. Quite recently ■* Professor Sollas has published the outlines of a slightly different modification of already existing schemes, in which the meaning of the name Monaxonida ' Sci. Proc. Roy. T)vh. Soc, vol. v. part ii. p. 112. REPORT ON THE MONAXONIDA. XI is enlarged to include the Keratosa, while the name Monaxona is made more or less exactly synonymous with the older Monaxonida ; thus : — Subclass 1. Hexactinellida, Subclass 2. Demospongiae, , Subclass 3. Myxospongiee, PORIFEEA (Phylum). Class I. Plbthospongi*. { Tribe a. Monaxonida, I Tribe h. Totractinellida, Class II. Calcispongia. Order 1. Lyssakina (Zittel). „ 2. Dictyonina (Zittel). Order 1. Monaxona. „ 2. Ceratosa (Grant). Order 1. Choristida (Sollas). „ 2. Lithistida (O. Scdt.). Order 1. Halisarcosa(O.Scdt.). „ 2. Chondrosiosa. Finally, Dr. E. von Lendenfeld has, in a paper recently read before the Zoological Society of London,-^ given an elaborate scheme of classification of Sponges agreeing more or less closely in its main features with that of Dr. Vosmaer, and adapted, as regards the smaller subdivisions, from the most recently published works on the subject. To this paper we would refer the reader who seeks for further information on this much-vexed question. As the portion of the scheme bearing upon the Monaxonida is derived chiefly from our own Preliminary Report we need not here discuss it in detail. We must, however, point out one very serious error in Dr. von Lendeufeld's j^aper, which appears to be a misjjrint. We refer to the use of the name " Monaxonida " on p. 583, instead of " Monaxonia," as opposed to the "Tetraxonia" on p. 580. We point this out as such a mistake is likely to lead to great confusion, for the term Monaxonida as used by Dr. von Lendenfeld in this place has a significance very diff"erent from that of the same term as used by ourselves. ' Proc. Zool. Soc. Land., December 21, 1886. xii THE VOYAGE OF H.M.S. CHALLENGER. CHAPTER II.— ANATOMY AND HISTOLOGY, AND EMBRYOLOGICAL NOTES. A. Anatomy and Histology. We propose in this chapter to treat of the anatomy and histology of the Monax- onida under three principal headings — (I.) the skeleton, (II.) the soft tissues, and (III.) the canal system. This will be found to be on the whole a very convenient division, although not an absolute one, for we shall have to forestall our account of the soft tissues in so far as these are directly concerned with the formation of the skeleton. We hope in this manner to pave the way for a somewhat detailed discussion of the classification which we propose to give in the next chapter. I. The Skeleton. (a) The Spicules. If any apology be needed for dealing in some bttle detail with a subject which has already received perhaps more than its share of attention at the hands of previous authors, it is to be found in the present state of confusion with regard to the true shapes of some of the minuter and more complex forms of spicules, and with regard to the terminology employed to describe them. Perhaps we may be allowed to say a few words here in self justification. The leading, and in fact the only really comprehensive work on the subject published in English is Bowerbank's Monograph of British Spongiadse, and this work leaves the matter in such an unsatisfactory condition that it is absolutely necessary to go over the ground again in order to facilitate a proper understanding of the subsequent pages. Let us see how far these remarks are supported by a brief examination of Dr. Bowerbank's system. In the first place we are tokP that " the spicula may be conveniently classed under the following heads : — 1. The essential skeleton spicula. 2. The auxiliary spicula." The auxiliary spicula are subdi \dded into — ' ' Connecting spicula. Prehensile spicula. Defensive spicula. Tension spicula. Retentive spicula. Spicula of the sarcode. Spicula of the ovaries and gemmules." This appears to us an extremely artificial arrangement, more- over it would be very inconvenient in practice, for how can anyone decide whether a given "skeleton spicule" is "essential" or " auxiliary ? " Or how are we to know whether a spicule is a "Tension spiculum" or a "Retentive spiculum'?" We are told that these two classes of spicules are both appropriated to the " membranous tissues of sponges," and that " the ofiice of the first of these is simply to strengthen and support those ' Mon. Brit. Spong., voL i. p. 13. REPORT ON THE MONAXONIDA. xiii delicate tissues when necessary, and to communicate to tliem a certain amount of tension when it is required," while " the office of the second class is that of assisting in the retention and protection of the sarcode on the interstitial and other membranous structures." Leaving aside the question as to what an interstitial membranous structure may be, we may point out in passing that these statements are based upon unproved theory. As to the forms assumed by many of the spicules, Bowerbank's work is also unsatisfactory. No one could well go wrong with regard to the larger " essential skeleton " spicules, and we consequently find these correctly figured in the plates, but the terminology employed to describe them is most unfortunate ; who, for example, could bring himself to make use of the term " Fusiformi-attenuato-cylindrical," ^ or " Exflected elongo-equiangulated triradiate,"^ in describing the spicules of a sponge'? To discover the true form of many of the more minute spicules (" auxiliary " spicules) is, it must be confessed, a difficult task, still it is hard to see how such a careful observer as Dr. Bowerbank can have been so far misled as he was in this respect. The imperfections of his descriptions and figures show themselves in the very important grouj) of so-called " anchorate " s^^icules [ = cliela3, nobis). Quite correctly he distinguishes between two main divisions of these, the equal ended and the unequal ended, but as to the true shape of either of these he seems to have had very little idea. He subdivides both categories into "bidentate," "tridentate," and "palmate." The tridentate and palmate forms, as he himself indicates, run into one another, while a mere tyro, by the examination of Dr. Bowerbank's figures and comparison of these with actual specimens, may satisfy himself in a very short time that the " bidentate " form is neither more nor less than a side view of either of the other two. Yet again and again do we find sponges described by Dr. Bowerbank as possessing two kinds of " anchorates," the two views being carefully figured as distinct spicules.'' On pi. xlvi., vol. iii. of the Monograph of British Sponges, for example, fig. 12 is obviously the side view, and fig. 13 the front (or back?) view of the same spicule, but we find the following descriptions:— "Fig. 12. — A bidentate, inequi-anchorate, retentive spiculum, from the dermal membrane. x 530 linear. Fig. 13. — A dentato-palmate, inequi-anchorate, retentive spiculum, from the interstitial membranes, x 530 linear. This form of spiculum was not observed in the specimen of the sponge first examined. In the specimen figured they are about equal in number to the bidentate spicula." Moreover, the figures of the spicules are rarely complete, the " anterior palm," owing to its great transparency difficult to make out, being almost invariably omitted. It is in our opinion of the greatest importance that these errors should be corrected, ' Op. cit.., vol. i. p. 231. 2 Op_ „■(_, vol. i. p. 233. 3 The shape of the " bidentate equianchorate " spicule will be found elaborately, though somewhat unintelligibly, described on p. 46, vol. i., Mon. Brit. Spong. xiv THE VOYAGE OF H.M.S. CHALLENGER. and we point them out here because no one else seems to have taken the trouble to do so, and lest the student should be led to waste valuable time in endeavouring to reconcile the observed facts with Dr. Bowerbank's descriptions and figures. It might be urged that Mr. Carter has abeady given good descriptions and figures of the leading forms of spicules, including even the most complex ; this is true, but unfortunately these occur scattered through a number of separate papers which arc not likely ever to come into the hands of any but sj)ecialists. Vosmaer has also given some general account of the spicules, but his figures of the more complex and minute forms are far from perfect, being for the most part copied from other authors. Perha2:)s we need say no more in excuse of giving here a general account of the minute anatomy of the spicules, and, more especially, of discarding the old and cumbersome nomenclature, and endeavouring to bring into use a more scientific and rational terminology. The terminology here adopted is the result, in so far as that result afiects the Monaxonida, of the mutual agreement of several spongologists, who consulted together with a view to formulating a set of terms which might have some chance of being generally accepted, as an experience of over twenty years has shown that this cannot be hoped of the Bowerbankian nomenclature, and there exists no other system of scientific terms to fall back upon. We give in the form of an appendix a list of the terms here employed, with the more important synonyms of other authors. Concerning the chemical composition and the development of the spicules very little is as yet known. It is, however, a generally recognised fact that each (at any rate of the larger forms, and probably, judging from analogy, of the smaller forms also) consists of a central, axial, organic thread (the " axial thread," commonly, though incorrectly, referred to as the "central canal"), surrounded by concentric layers of a peculiar siliceous substance, vaguely spoken of as " organic silica." That the spicules are the products of the activity of special cells (silicoblasts),^ within which they originate, is now also an admitted fact; but it is also pretty certain that the larger forms at any rate become free from the parent cell (silicoblast) before attaining their full size. Whether it be within or without the cell, the spicule grows by apposition. The axial thread appears to Ije the portion first developed, and upon the form assumed by this in the course of its growth, depends the form of the fully developed spicule. Not infrequently, in the case of the minuter forms of spicules, several originate in one and the same silicoblast, and they then remain together in bundles for a longer or shorter period ; to such bundles the general term dragmata^ is here applied. This brief account of the nature and growth of siliceous spicules is necessary for the 1 Vide PI. XXI. fig. 13., and cf. Vosmaer, Bronn's Klasseu u. Ordnungen d. Tluerreiclis, Porifera, p. 436, where the name " Silicoblast " is first introduced. - Greek, o^a'/y.x, a sheaf. REPORT ON THE MONAXONIDA. XV proper understanding of the various forms which they assume. For further details the reader is referred to the works cited below. ^ The spicules of the Monaxonida may be very conveniently divided into two classes. Those of the first class constitute (with or without the aid of horn-like cementing substance, or spongin) the true skeleton of the sponge. They commonly occur associated together in tracts or in definite fibres, and are usually, when full grown, much larger than those of the second class ; consec^uently the name megasdera is applied to them. The spicules of the second class occur, as a rule, irregularly scattered through the soft tissues without taking part in the formation of the true, continuous skeleton. Their function is in most cases very doubtful. Owing to their minute size they are termed microsclera. Megasclera. In the Monaxonida the megasclera are (with few exceptions), as the name of the group implies, uniaxial ; that is to say, the imaginary axis of the spicule, which finds its nearest visible representative in the axial thread, remains unbranched, although it may be more or less curved, or even bent at rather a sharp angle. Such a spicule may either grow in two directions from its point of origin, or in one direction only. In the former case two equivalent and usually similar rays arise more or less diametrically opposite to one another, and the spicule is consequently said to be diactinal, the point of origin remaining at or near the centre of the spicule. In the latter case one ray only is formed, and the point of origin, commonly represented by a l^ulbous enlargement of the axial thread, remains at the end of the spicule, which may or may not be swollen ; in this case the spicule is said to be monactinal. Although it is impossible in many cases to demonstrate with certainty the manner in which the growth of a spicule takes place, yet the division into monactinal and diactinal is a very convenient one for all practical purposes. The diactinal appears to be the simpler type, from which the monactinal is derived by abortion of one of the rays. A. Diactinal megasclera. Here we may distinguish four fundamental types, which, however, run into one another by imperceptible gradations, but for which it is convenient to have distinct names for purposes of description. 1. Oxea (woodcut, Fig. I., 1); these are straight or curved spicules, more or less fusiform, and gradually pointed at each end. 1 Gf. Bowerbank, Mon. Brit. Spong., vol. i. p. 6, &c.; Carter, Ann. and May. Nat. Hist., ser. 2, vol. xx, p. 23, pi. i.; and ser. 4, vol. xiv. p. 100, pi. x.; Vosmaer, Bronn's Klassen u. Ordnungen d. Tliierreiclis, Porifera, pp. 187, 436, &c. xvi THE VOYAGE OF H.M.S. CHALLENGER. 2. Tornota (woodcut, Fig. I., 2) ; the tornote form differs from the oxeote in having the ends abruptly and suddenly pointed. 3 c <- > C 0 Fig. I. — Diactinal megasclera. 3. Strongyla (woodcut, Fig. I., 3) ; here the spicule is not pointed at all, but the ends are evenly rounded off. 4. Tylota (woodcut, Fig. I., 4); in these forms the spicule has a swelling at each end, so that it is divisible into a central, elongated, cylindrical shaft and two terminal heads. B. Monactinal megasde7-a. Of these we need only distinguish two fundamental t}^3es : — 1. Styli (woodcut, Fig. 11. , 1); pointed at one end and evenly rounded off at the other without any swelling. Fig. II. — Monactinal megasclera. 2. Tylostyli (woodcut, Fig. II., 2) ; in which a head is developed at one end of the spicule. In the case of monactinal spicules it is convenient to distinguish between a " base," which is the blunt end of the spicule, and an " apex," which is the pointed end. C. Branched megasclera. We have now to speak of certain rarely occurring branched forms, which constitute an exception to the general rule that the spicules of the Monaxonida are uniaxial. The Fig. ni.— Branched forms of megasclera. angle which the branches make with the shaft may be either greater than 90°, as in woodcut. Fig. III., 1, or less than 90°, as in woodcut. Fig. III., 2. 1. Cladostrongyla (woodcut. Fig. III., 1); when the spicule is branched at one end and REPOET ON THE MONAXONIDA. xvii simply rounded off at the other it is said to be cladostrongylote, and the sj)icules are cladostrongyla. Such spicules occur in Thrinacopliora funiformis, nobis {vide PL XXIII. figs, le, If). 2. Cladotylota (woodcut, Fig. III., 2); when it is branched at one end and swollen into a knob at the other, it is said to be cladotylote, and the spicules are cladotylota. Such forms occur in the genus Acarnus. Any of the above forms of megasclera may become spinose, as frequently happens in the case of styli (e.g., in the genus Myxilla), but this in no way affects their fundamental forms or the names applied to them. Swellings, also, may be developed elsewhere than at the ends of the spicule ; hence, when there is a swelling in the centre of the spicule it is said to be centrotylote (PI. IX. fig. 2, h, c), and when there are several swellings on a spicule (no matter where they are situate) it is said to be polytylote (PI. XIX. figs. 9, 9'). Microsclera. The microsclera of the Monaxonida exhibit considerable variety in form, and it is no easy matter to arrange them satisfactorily in natural groups. The simplest plan, and that which we shall adopt here, is to consider them under three heads — (A) simple linear forms, (B) hooked forms, and (C) stellate forms. This may be an artificial arrangement, but it is a convenient one in practice, and as yet we hardly know enough about the question to allow of a more philosophical classification. A. Simple Linear' Forms. 1. Minute oxea (woodcut. Fig. IV., 3), which are generally if not always spined, as in Dendropsis bidentifera, nobis (PL XL. fig. 7c), Spongilla lacustris, &c. 5 Fig. IV. — Linear forms of microsclera. 2. Ehap>hides (woodcut. Fig. IV., 1) ; long, hair-like spicules, not in sheaves. These occur in great abundance in the genus Tedania. (ZOOL. CHAIjL. EXP. PART LIX. — 1887.) ^'^^ '^ XVIU THE VOYAGE OF H.M.S. CHALLENGER. 3. Trichodragmata (woodcut, Fig. IV., 2) ; hair -like spicules arranged in more or less comjjact bundles. Such forms are very common in the genus Es2)erella. 4. Toxa (woodcut, Fig. IV., 4) ; bow-shaped spicules, tapering towards either extremity and with the ends often spinose. These occur in many sponges. A slight modification of the tyj^e is found in the " forcipiform " spicule of Halichondria forcipis} In one sponge, Ampliilectus pilosus, nobis, we have found the toxa, which in the young condition are very distinct and strongly curved, passing by gradual transitional stages into simple, much elongated, slender oxea of such a size, that, did they occur apart from the small toxa, they would certainly be classed amongst the megasclera {vide PI. XIX. figs, ba-ba'""). This fact well illustrates the difficulty in classifying spicules according to size and form. 5. Toxodragmata (woodcut, Fig. IV., 5) ; more or less compact bundles of toxa, which have aU developed in one and the same cell. B. Hooked Forms. 1. Sigmata (woodcut. Fig. V., 1, 2) ; each consisting of a slender, cylindrical shaft, which is curved over so as to form a more or less sharp hook at each end. The two terminal hooks may curve both in the same direction, when the spicule is said to be simple (woodcut, Fig. V., 2), or they may curve in different dii-ections, when it is said to be contort (woodcut. Fig. V., 1). There is, however, no real distinction between the two, and, as a matter of fact, the spicules are nearly always contort to some extent. Fig. v.— Hooked forms of microsclera. Sigmata are perhaps the commonest of Monaxonid microsclera, being very characteristic of the Desmacidonidse. 2. Sigmadragmata (woodcut, Fig. V., 3); more or less compact bundles of sigmata, which have all developed in one and the same cell. 3. Diancistra (woodcut. Fig. V., 4); hooked forms of very peculiar shape ; character- istic of the subfamily Hamacanthiuse. The spicule resembles a large stout sigma, but the inner margin of both shaft and hook thins out into a fine knife-edge, notched as ^ Vide H. J. Carter, Ann. and Mag. Nat. Hist., ser. 4, vol. xiv. p. 247, pi. xiv. fig. 32 ; cf. also tlie same spicule in Forcepia colonensis, torn, cit, p. 248, pi. xv. fig. 47. REPORT ON THE MONAXONIDA. XIX shown in the figure. Usually, at any rate, the spicule is more or less contort, the two hooks lying in two diflferent planes. 4. Chelse (woodcuts, Fig. V., 5, 6, and Fig. VI.); these are the most complex of all the microsclera, and present us mth a considerable range of variation in detail. The fundamental shape will be best understood from the accompanying figures, so we shall make the description short. It will be seen that each spicule consists of a more or less curved " shaft " (.s), bearing at each end a variable number of sharply recurved processes {at, at' , It, It') which may be conveniently spoken of as the •'' teeth," or, when they are broad and much expanded, the "jaalms." Each tooth or palm is connected with the shaft by a buttress-like projection of the latter, called by Mr. Carter-^ the " falx," and generally itseK con- sists of a thin, flattened, oval lamella, so transparent as to be very difficult to make out. The terminal portion of the falx, upon which the tooth or palm directly rests, has been called by Mr. Carter the " tubercle " {t, t'); it is generally very conspicuous through the transparent lamella. Very commonly there is a single central or anterior tooth [at, at'), and two lateral teeth {It, It') placed one on each side of it. When the teeth are comparatively narrow and the two lateral teeth are completely cut away from the shaft, as in Fig. VI., 1, la, the spicule is said to be " tridentate;" when, however, they are broad (forming palms) and the lateral palms remain adherent to the shaft for their entire length, the central palm alone being completely separated, as in Fig. VI., 2, 2a, the spicule is said to be " palmate." Numberless gradations exist between these two types, depending upon the extent to which the incisions between the lateral teeth and the shaft are carried, and upon the breadth of the teeth or palms. The distinction between the two types is, as has already been pointed out, a purely artificial one, but for the sake of convenience it may be retained. In some species {e.g., Chondrocladia concrescens (?)) there may be as many as seven teeth completely cut away from the shaft and from one another (PL XXI. fig. 12), while in one species {Sideroderma navicelligerum) there is a very remarkable little chela (PI. IX. fig. 8) in which there is only a single (the median) tooth at each end. The shaft itself is frequently expanded laterally into wing-like processes, or fimbriae (woodcut, Fig. VI.,/), which may extend along its whole length, but are more generally confined to one (or the two) ends. 1 Ann. and. May. Nat. Hist., ser. 4, vol. xiv. pp. 208, 209. Fig. VI. — Structure of the chelae. 1, tridentate anisochela, front \-iew ; 1«, .same, side view ; 2, palmate isochela, front view ; 2a, same, side view ; t, t', tubercle; at, at', anterior tooth or palm; It, It, lateral tooth or palm ; s, shaft ; /, fimbria. XX THE VOYAGE OF H.M.S. CHALLENGER. The whole group of chelse, with their almost endless modifications, may be divided into two subgroups according to whether the two ends of the spicule are equal (Fig. v., 5 ; Fig. VI., 2, 2a) or unequal (Fig. V., 6 ; Fig. VI., 1, la); to the former class the name isochelse is applied, and to the latter the name anisochelw. The most complex form of chela as^yet known to us appears to be that of Melonanchora elliptica, figured and descrilied by Mr. Carter.^ Not infrequently the anisochelse are found in " rosettes" (PI. XVII. fig. 7), all adhering together by their small ends, which are attached to a central, granular (?) mass; no satisfactory explanation has, so far as we are aware, as yet been given of this state of things, but possibly the rosettes are com- parable to the toxodragmata and sigmadragmata, which we have described above, and orio-inate like them by the development of a number of spicules in one and the same cell. Mr. Carter also records this phenomenon in the case of the isochelse of Desmacidon tituhans, Schmidt.^ In the embryo of Esperella nnammiformis we have succeeded in tracing the development of the chelse in a very interesting manner. This developmental history throws an important light on the relationships of the chelae to other forms of micro- sclera. In the earliest stage observed the sj)icule consists of a sim^jle, slender shaft, very slightly curved in the same direction at each end, and also pointed at each end. It appears that the two pointed extremities then curve sharply inwards so as to form each an acute angle with the shaft, thereby giving to the spicule the appearance of a simple sigma with short, sharply recurved hooks and almost straight shaft. At first, as already noted by Carter,^ the two ends are equal, but this condition does not persist for long. The teeth or palms are now developed ; the central or anterior palm is formed by a delicate, oval, flattened outgrowth from the end of the hook (possibly also in part from the sides of the hook), lying in a plane at right angles to the plane of the original, simple, hooked shaft. The lateral palms are formed as lateral outgrowths from the two ends of the straight portion of the shaft. The hooks of the original spicule form the median falces and tubercles of the adult. Thus it appears that the chelate spicules develop from sigma-like forms by the formation of flattened outgrowths or buds at the two ends ; hence we are justified in grouping sigmata and chelse in the same category. In Esperella mammiformis the lateral outgrowths of the shaft (lateral palms) remain connected with the shaft even in the adult, the spicule (PI. XV. figs. 18, 18a) being palmate; but in other species, as already noted, they often become cut away from the shaft and form distinct teeth (e.g., Cladorhiza tridentata, woodcut. Fig. VI., 1, la ; PL XXI. fig. 20); new lateral outgrowths may then be developed on the shaft, and these may again become cut away as teeth {e.g., Meliiderma stipitatum, PI. XXI. fig. 14), or 1 Ann. and Mag. Nat. Hist, ser. 4, vol. xiv. p. 212, pi. xiii. figs. 9, 10, 11, 12. 2 Ann. and Mag. Nat. Hist., ser. 5, vol. ix. p. 298, pi. xii. fig. 24, h. In this paper will be found a good deal of in- teresting information concerning the chelfe. ^ Ann. and Mag. Nat. Hist., ser. 4, vol. xiv. p. 102, pi. x. fig. 13. UEPORT ON THE MONAXONIDA. xxi remain connected with it as " fimbriae " (Cladorhiza tridentata, Jigg. cltt.), and so on. 5. Bipocilli (woodcut, Fig. V., 7) ; these are curious forms which, owing to their minute size, it is very difficult to make out satisfactorily. They occur only ia a single genus, lophon. Each consists of a shaft, with a termiaal, cup-like expansion at each end. A modification of this type of spicule, which occurs in a new species of lophon, lophon chelifer, nobis (PI. XVII. fig. 3), has the cup-like expansions divided into teeth, and thus seems to connect the bipocilli with the chelate forms. C. Stellate Forms. Microsclera of the stellate group are rare amongst Monaxonid sponges. In the Tetractinellida and in the Tethyse they are, however, met with in great abundance and under a great variety of modifications. We shall describe here only those types which are certainly known to occur in the Monaxonida. ^-^^Jl^^^ijli^^^cvs^ Fig. VII. — Stellate forms of microsclera. 1. Spirulse (woodcut. Fig. VII., 1, la); these ai'e more or less elongated, spiral or subspiral forms, which may be either smooth or provided with more or less numerous spines. The spinose forms are very characteristic of the genus Spirastrella. 2. Discastra (woodcut. Fig. VII., 2) ; spicules with a straight, elongated shaft, usually with a spinose base, and surrounded by a greater or less number of usually spinose whorls, altogether much resembling a miniature chess-man, or, in some species, a little fir tree. These spicules occur in the genus Latruncidia. 3. Amphiastra, consisting each of a cylindrical shaft bearing a single toothed whorl at each end ; occurring, for example, in Axoniderma mirabile, nobis (PI. XXI. fig. 9). {h) The Spongiu, the Spon go blasts and the Connective Tissue Sheath of the Skeleton Fibres. Before proceeding to treat of the arrangement of the skeleton it is necessary that we should first speak of a second very important constituent thereof, viz., the spongin. The spongin is the horn-like cementing material which, in many, though by no xxii THE VOYAGE OF H.M.S. CHALLENGER. means all, Monaxouid sponges unites the individual spicules into a coherent skeleton, and ultimately (in the so-called Keratosa) constitutes by itself the entire skeletal system. In chemical composition spongin has been found to resemble silk. Krukenberg has recently investigated it, and assigns to it the chemical formula CsoHisNgOig.^ Its physical characters and mode of occurrence in the sponge, however, concern us more in this place than its chemical composition. It is usually pale yellow or amber-coloured, and is arranged in concentric layers around the spicules which it unites together. Concerning its mode of formation in any sponges but true so-called Keratosa we have no actual observations. There can, however, be no doubt, from the analogy of the true Keratosa, and from its arrangement in concentric layers, that it is a secretion of glandular cells (spongoblasts). Our own histological investigations on the Challenger material throw some light upon this point. We have found, in a number of different Monaxonid sponges, sheaths of more or less fibrous connective tissue cells accompanying the fibres of the skeleton. This condition has been carefidly observed by us in the following- species: — Es]periopsis challengeri, Axinella (f) paradoxa, Raspailia tenuis, Acanthella pidcherrima and Suherites perfechis. In Esperiopsis challengeri, where there is a fair amount of very pale coloured spongin, these cells are fairly abundant ; they accompany the larger bands of spiculo- fibre in dense tracts, in which the individual cells lie close together side by side. They are fusiform, but apparently not usually A^ery much elongated, measuring about 0*024 by 0*0096 mm.; occasionally, however, they seem to be drawn out into a fine, transparent thread at each end ; they are highly granular and stain well with borax-carmine. In Raspailia tenuis the cells in question are uncommonly well developed ; and here there is also a very large amount of spongin. They occur in a thick but not densely packed sheath around the central axis, with which their longer axes are of course parallel (vide PI. XLIX. figs. 1, 1«) ; they closely resemble those of Esperiopsis challengeri just described, and their form will be best understood from the figures. They average in size about 0"033 by 0'0096 mm. We have not succeeded in distinguishing a nucleus ; the cells appear to be highly and uniformly granular throughout. In Acanthella p)idGherrima we have found similar granular cells scattered about fairly plentifully amongst the spicules, but not forming a definite zone around a central axis. The amount of spongin present is very small, but our thin, stained sections have demonstrated its presence beyond a doubt. Hence we cannot entirely agree with Schmidt, when he says ^ in his generic diagnosis " Parenchyma spisse impletum spiculis simplicibus longioribus, substantia firmiori non inclusis." In Axinella [?) paradoxa, the fibrous tissue accompanying the bands of spicules is very highly developed (PL XLIX. fig. 2a). It is composed of very much elongated, ' Cf. Vosmaer, Bronn's Klass. n. Ordnuiig. d. Thieirreichs, Porifera, p. 434. ^ Spong. Adriat. Meeres, p. 64. REPORT ON THE MONAXONIDA. xxiii rather slender, fusiform cells. These cells are highly granular, and often multipolar, differing really but very slightly from the ordinary mesodermal cells which occur in the ground tissue of this sponge, save for their great elongation in one particular direction {cf. PI. XLIX. fig. 2). They commonly measure about 0'048 by 0-007 mm., and in some cases there appears to be a nucleus in the centre of the cell. The spongin uniting the spicules together is distinct but not very abundant. In Suberites perfectus there is, as is usually the case in the Clavulina, no trace of spongin. But the skeleton fil:)res are accompanied and apparently held together l^y dense bands of very strongly fibrous connective tissue (PI. L. figs. 2, f.t, 2a, 2b). Scattered fairly abundantly through this tissue are a number of fusiform, elongated, more or less granular bodies, more numerous in some parts than in others (PI. L. figs. 2«, 2b, g). The bands of fibrous connective tissue are directly continuous with the fibrous cortex. The data before us are perhaps too scanty to justify any wide generalisation, but it is a significant fact that in Suberites perfectus, where there is no spongin to hold the spicules together, the fibrous element of the surrounding connective tissue is very strongly developed, and the granular element comparatively feebly developed, while in Raspailict tenuis, in which there is a very great deal of spongin, and hence no need for fibrous tissue to unite the spicules, exactly the reverse is the case. The other species mentioned, viz., Esperiopsis challengeri, Axinella {?) paradoxa and Acanthella pulcherrima, occupy intermediate positions between these two extremes, both with regard to the amount of spongin present and with regard to the degree of develop- ment of the granular and fibrous elements of the connective tissue respectively. In explanation of these facts we would venture to suggest that in Raspailia tenuis and other species where spongin is present, surrounded by highly granular, specially modified, mesodermal elements, the latter secrete the former, and were themselves originally ordinary mesodex'mal cells, which have become slightly modified for the fulfilment of this special function. In other sponges {e.g., Dendrilla), where there is a still greater development of spongin, these cells have become still more highly specialised and constitute the " spongoblasts " of the so-called Keratosa. These suggestions are in perfect harmony with the views now generally held as to the late appearance of spongin in the Porifera. To sum up briefly, we may state that spongin is a secretion of mesodermal cells, which, originally indistinguishable from other mesodermal cells, have become under suitable conditions sjjecially modified in form in order to fulfil their special function ; and it is probable that, in at any rate some cases, these cells were derived from bands of fibrous connective tissue which accompanied the bands of spicules for the purpose of holding them together. Coincidently with the development of the spongin the filirous tissue in some cases XXIV THE VOYAGE OF H.M.S. CHALLENGER. gradually disappeared ; but in other cases it persisted, and in this connection it is of especial interest to note that in certain so-called Keratose sponges, i.e., species in which the sj)ongin has entirely usurped the place of the spicules, there is around the horny fibre — (1) a sheath of special spongoblasts, and (2) outside this, a sheath of fibrous connective tissue, composed of elongated, fusiform cells, and much resembling (according to the figures) that found by us in Suberites perfectus. This condition is described and figured by von Lendenfeld in several species, e.g., Dendrilla rosea,^ Dendrilla aeroj^hoba,^ and Aulena villosa? Possibly in these cases there was originally only one thick sheath of connective tissue around the skeleton fibres, and by degrees the cells of the inner layer, next to the fibre, became specially modified as spongoblasts, while those of the outer layer retained their original form unaltered, or, very possibly, became still more fibrous in order to form a protective sheath to the layer of spongo- blasts. We must imagine that the outer layer still has some important function to fulfil in order to account for its very great degree of development in certain forms {e.g., Dendrilla aerophoba). (c) The Arrangement of the Skeleton. Although in the majority of Monaxonid sponges (the Halichondrina) there is more or less spongin present in the skeleton, yet this is, as a rule, subsidiary to the siliceous element, and serves merely as a cement to bind the spicules together in continuous fibres. Under certain conditions, however, the spongin forms by far the most important part of the skeleton, but cases in which this happens have been dealt with elsewhere. (1) The combination of the spicides to form fibres. This may take place according to three main plans, which we propose to distinguish as the Renierine, the Axinellid, and the Ectyonine type respectively. 1. The Renierine type (PI. XLVI. figs. 2, 3, 6). — Here each individual fibre is com- posed of a central axis of parallel spicules, arranged either uniserially or polyserially, which may or may not be enclosed in a distinct sheath of spongin. The surface of the fibre is smooth. (Found in the Homorrhaphidae, Heterorrhaphidse, Esperellinse, &c.) 2. The Axinellid type. — Here all the spicules of the fibre are, in a sense, " echinating." Each spicule has its base in the centre of the fibre (usually embedded in spongin), while its apex projects obhquely outwards and forwards in the direction in which the fibre is running {i.e., more or less towards the surface of the sponge). There is no central core of longitudinally disposed spicules, the centre of the fibre being occupied by the bases of 1 Zeitsdir.f. wiss. ZooL, Bd. xxxviii. p. 286, pi. xiii. figs. 25, 26. ^ Loc. cit, p. 304, pi. xiii. fig. 28. ' Proc. Linn. Soc. N.S.W., vol. x., pt. iii. p. 315, pi. xxx. fig. 18. REPORT ON THE MONAXONIDA. XXV the outwardly projecting spicules. This arrangement prevails amongst the AxinellidEe, of which grou}} it is very characteristic. 3. Tlie Ectyonine type (PI. XLVI. fig. 9 ; PL XLVII. figs. 4a, 5).— This agrees with the Renierine type, except that the surface of the fibre, instead of being smooth, is " echinated " by spicules projecting outwardly and usually more or less at right angles. The central axis of longitudinally placed spicules is generally present, but may, in a few instances, disappear, leaving only the horny fibre and the " echinating " spicules {e.g., in Clathria lendenfeldi, PI. XLVII. fig. 5). The echinating spicules are generally diS'erent from those in the centre of the fibre, but may, though very rarely, be the same {e.g., Efhinoclathria favus). Such fibre is characteristic of the subfamily Ectyoninae, and is apparently confined to this group. These three principal types of fibre all occur in the Halichondrina. In all three the amount of spongin present is extremely variable. In the Clavulina on the other hand, w^here, it will be remembered, there is no spongin present, we are acquainted only with the Renierine type of arrangement ; the fibres, when distinct fibres are present, being com- posed of longitudinally placed spicules lying parallel with one another. In aU Monaxonid sponges there are usually a great number of megasclera scattered irregularly through the tissues, in addition to those which enter into the composition of the fibre. (2) The Arrangement of the Skeleton at large. In this case we have only two main types of arrangement to deal with. First, Reticu- late, and second, Radiate. 1. The Reticidate arrangement, characteristic of nearly aU the Halichondrina. In cases where this arrangement obtains the spicules may either be arranged (at any rate partly) in distinct fibres, or scattered separately through the soft tissues without any arrangement whatever (as in Halichondria solida, nobis). The fibres, when present, may belong to any of the three types above described. The essential character of a reticulate skeleton is that the fibres, or individual spicules in cases where no fibres exist, should cross and recross one another, so as to give rise to a network with closed meshes (PL XLVI. figs. 2, 3, 4, 6, 9). AU gradations exist between a reticulation of separate and quite irregularly scattered spicules and a reticulate skeleton with perfectly definite rectangular meshes whose boundaries are formed of compact fibres (PL XLVI. figs. 3, 6). When the skeleton is rectangularly arranged (as in Renicra, Chalina, Esperella, &c.), it is easy to distinguish between two sets of fibres — (1) Primary fibres (PL XLVI. fig. 3, a), running vertically towards the surface of the sponge, and (2) Secondary fibres (PL XLVI. fig. 3, V), crossing the primary fibres at right angles. When the primary fibres are composed of longitudinally arranged monactinal spicules these have, as a general rule, their apices all pointing in one direction, towards the surface of the sponge. At the (zooL. CHALL. EXP. — PART Lix. — 1887.) Nnn d XXVI THE VOYAGE OF H.M.S. CHALLENGER. surface of the sponge, also, the spicules (whether monactinal or diactinal) composing the primary fil)re very commonly spread out so as to form divergent brushes, upon which the dermal membrane rests (PI. XLVII. fig. 6). In another important, though less common type the skeleton is regularly reticulate without its being possible to distinguish between primary and secondary fibres. In this case the meshes of the reticulation are triangular, with equal sides each of exactly one spicule's length. Hence there can be no long, continuous fibres such as exist in a rectangularly meshed skeleton, the greatest continuous length of fibre possible being determined by the length of the spicules. Such a skeleton is found in Myxilla roscicea, var. jcqMnica, nobis (vide PI. XLVII. fig. 3). 2. The Radiate arrangement. — This is characteristic of the Suberitidae, and, to a less Fig. VIII. — Styloconlyla stipitata, var. ylobosa. Sections sho%ving the arrangement of tlie skeleton. A, longitudinal section of a specimen ■which exhibits no spiral arrangement of the fibres, x 6. A', transverse section of do. , x 4. B, transverse section of a specimen in which the fibres are spirally disposed, x 4 ; o, osculum ; c, cortical spicules ; a, stem ; cm, central mass of spicules derived from the breaking up of the stem ; s, sections of do. ; rf, the radiating skeletal fibres. extent, of the Axinellidae. The fibres or bands of spicules radiate to the surface from a common centre, and there are no crossing fibres {vide woodcut. Fig. VIII.). In the Axinellidse the common centre of radiation is usually much elongated, so as to give rise to a central axis of fibre from which other fibres radiate to the surface. The same may take place in the genus Subeintes (e.g., Suberites elongatus, nobis, Suherites axiatus, nobis, &c.). Thus the radiate type of skeleton difiers principally from the reticulate type with regular rectangular meshes in that there are primary but no secondary fibres present. The skeleton of most ]\Ionaxonid sponges, whether reticulate or radiate in its arrange- ment, is generally divisible into tw^o parts — (l) the Dermal skeleton (PI. XLVII. fig. 3, d.s.), at the surface of the sponge, specially modified to support and protect the REPORT ON THE MONAXONIDA. XXVll ectosome, and (2) the Mcdn skeleton (PL XLVII. fig. 3, m.s.), which forms the main mass of the skeleton, supporting the canal system, &c., in the interior of the sponge. The dermal skeleton is very frequently distinguished from the main skeleton by a different arrangement of the component parts, which is often very striking. It may be either reticulate, composed of more or less definite fibres, or of scattered spicules laid horizontally, or radiate, composed of spicules arranged vertically to the surface with their apices projecting outwards. Thus in Plumohalichondria mammillata, the dermal skeleton is a very compact and regular, polygonally-meshed reticulation of spiculo-fibre (PI. XLVII. fig. 4), while the main skeleton is radiately arranged, consisting of ascending columns of plumose fibres (PI. XLVII. fig. 4a, p.c.) running vertically to the surface; this is an excellent example of a reticulate dermal skeleton combined with a radiate main skeleton. In Myxilla rosacea, var. jaiDonica, on the other hand, we have a radiate dermal skeleton combined with a reticulate main skeleton ^ide PL XLVII. fig. 3). A radiate dermal skeleton and a radiate main skeleton, however, generally occur together. Thus in the genera Suberites and Stylocordyla (woodcut, Fig. VIII.) the dermal skeleton is composed of brushes of small, outwardly projecting spicules, often so closely packed together as to form a dense, velvet-like pile. Very commonly, also, the individual spicules of the dermal skeleton differ from those of the main skeleton. This difference may concern merely the size, as is the case in most SuberitidaB, or it may concern the shape of the spicules. In Tedania, and again in Myxilla, we always find a diactinal dermal spicule (oxeote, tornote or tylote) combined with a monactinal main skeleton spicule (stylote), and in the Spirastrellida3 (Spirastrella and Latruncidia) there is usually a special dermal crust of quite peculiar spicules (spirulse and discastra) which we here consider as belonging to the category of microsclera. In two genera (both new) of deep-sea Desmacidonidse, we have discovered very extraordinary special dermal spicules in addition to the ordinary spicular complement of the group to which they belong. In the one case, Axoniderma, the spiculation would be that of the genus Cladorhiza, were it not for the existence of a dense external armature composed of a thick layer of large amphi- asters ; and in the other case, Meliiderma, we have a stipitate sponge agreeing in spiculation with the genus Chondrocladia, except for the presence of a layer of quite peculiar spicules (woodcut. Fig. IX. ) which encrust the stem. In both cases there can be little doubt that the extra spicules have been acquired in order to guard against the attacks of enemies, a purpose for which they are admirably adapted. Not infrequently the true dermal skeleton is replaced, more or less entirely, by foreign bodies, such as grains of sand, sponge-spicules, &c. These foreign bodies may be Fig. IX. — Meliiderma stipitatum. Spicules encrusting the stem. X 400. XXVlll THE VOYAGE OF H.M.S. CHALLENGER. collected and cemented together into definite fibres as in Desmacidon reptans, nobis, or they may be glued on to the surface of the sponge without any definite arrangement, as in Polymastia agglutinans, nobis (PI. XLI. fig. 6 ; PL XLII. fig. 1). (The main skeleton also may be replaced to a greater or less extent by foreign bodies, but this subject will be treated of more fully in our discussion of the so-called " Phoriosjwngm.") We ought not to leave this part of our subject without a few words as to the arrange- ment of the mierosclera. These, as we have already pointed out, do not, as a rule, take part in the formation of the skeleton proper, and their function is in most cases extremely diflicult to determine. Occasionally, however, it is sufiiciently obvious, and in such cases it nearly always appears to be protective.^ Thus we know of several instances in which hooked mierosclera are arranged around the walls of the canals, with one end embedded in the wall and the other projecting freely into the lumen of the canal. In Esj)erella murrayi, nobis, the inhalent, and apparently some of the exhalent canals also, are plenti- fully armed with sigmata, each with one hook projecting into the canal (PI. XLVIIL figs. 2, 2c), and in lojylion chelifer, nobis, we have observed the very peculiar bipocUli of that species arranged in a similar manner. In these cases there can be little doubt that the mierosclera in question serve to protect the sponge from the ingress of noxious parasites, such as minute Crustacea, to whose attacks sponges seem to be peculiarly subject, by way of the canal system ; just as the dermal skeleton serves to hinder these creatures from boring their way into the sponge at any part of the surface, which is a common occurrence in cases where the dermal skeleton is not sufficiently strongly developed. The arrangement of special mierosclera {e.g., amphiastra in Axoniderma, discastra in Latrunculia, spirulse in Spirastrella) to form a dermal armour, has already been referred to and needs no further comment. More difficult to understand are certain cases in which the mierosclera are attached to the fibre of the proper skeleton. This arrangement has been described by Bowerbank * in the case of the diancistra oiHamacantha johnsoni, and the diancistra of our Vomerula esperioides are arranged in a similar manner, each being cemented on to the spiculo-fibre by the back of the shaft, while the sharp, cutting teeth project outwards. Moreover we have observed a similar arrangement of the large anisochelse of E.spereUa simonis, nobis, the spicules being attached to the skeleton fibre in groups by their small ends. Probably in these cases also the spicules in question are really defensive. It is much more common, however, for the mierosclera to occur simply scattered irregularly through the mesoderm or in the limiting membranes without any pretence of arrangement, and they are often so minute and insignificant that we cannot believe that they now fulfil any function whatever. 1 We have, however, suggested elsewhere (p. xxxi) a different function for the triehodragmata, viz., that they serve, like straw in mortar, to bind together the soft gelatinous tissues in which they lie. - Men. Brit. Spong., vol. i. p. 35. REPOET ON THE MONAXONIDA. xxix II. The Soft Tissues. Professor SoUas, in his Prelimiaaiy Report on the TetractiueUida of the Challenger Expedition,^ has proposed to divide the entire body of a sponge into two parts. The first of these, or " ecfosome," he defines as " the outer layer of the sponge, not containing flagellated chambers," and the second, or " choanosome," as " the ' mark ' or ' parenchyma,' distinguished by the presence of flagellated chambers." This arrangement we find to be a very convenient one for practical purposes, and accordingly we shall make use of it in this Report ; although applied in the first instance to Tetractinellid sponges it is equally applicable to the Monaxonida, or, indeed, to most groups of sponges. So far as the de- finition goes, the terms ectosome and choanosome might be used to include the skeletal elements as well as the soft tissues, but, although the two primary divisions of the skeleton into " dermal " and " main " do, roughly speaking, correspond to the two primary divisions of the soft tissues into " ectosome " and " choanosome," we have found it much the most convenient plan to treat of the skeleton all together and quite separately from the soft tissues. (a) The Ectosome. Although varying very much in details of structure the ectosome invariably consists of two kinds of elements, ectodermal and mesodermal. The ectodermal element always consists, at any rate so far as is known, of a single layer of epidermic cells — usually flat, as in EspereUa murrayi, nobis, but in one instance (according to Vosmaer^) columnar. The sponge in which Dr. Vosmaer describes and figures the columnar epidermic cells is Tentorium semisuberites, but this must be considered as a great exception to the general rule. In our own sections of Tentorium, although they seem to be fairly well preserved, we have not succeeded in detecting the columnar cells. We have nothing new to add concerning the epidermis ; it is very difficult to study it satisfactorily in any but specially prepared specimens. The mesodermal constituents of the ectosome, unlike the ectodermal, vary very much in difi"erent genera both as regards quantity and character ; hence it is upon these that the great diversity of the ectosome in difi"erent sponges depends. They may be very small in quantity, forming only a thin layer of connective tissue immediately beneath the epidermis, in which case the ectosome is reduced to a thin, in most parts pore-bearing membrane (PI. XL VI. fig. 4, d.m.), for which the term dermal membrane is a very convenient one,^ or they may be very strongly developed, forming in some instances (Suberitidee) a thick, fibrous cortex (PI. L. fig. 1, ect.). 1 Sci. Proc. Hoy. Dub. Soc, vol. v., pt. vi. p. 177. 2 Sponges of the " Willem Barents" Expedition, 1880-81, p. 19., pi. iii., fig. 23. 3 The term " dermal memhrane " is also applicahle in cases where there is a thick ectosome, of which the outermost portion forms a separable membrane distinct from the remainder. XXX THE VOYAGE OF H.M.S. CHALLENGER. A thin ectosome, or dermal membrane, is usually associated with a reticulate dermal skeleton, as in Pachychalina lobata {vide PI. XL VI. fig. 4); or it may simply be supported on the ends of brushes of spicules, as already described when speaking of the skeleton. It may perhaps be studied to the best advantage in the Homorrhaphidag, but is also weU illustrated in the Axinellidse. Thus in Phakellia ventilabrum, var. connexiva (vide PI. XLIX. fig. 3), it is everywhere a thin membrane, perforated on the one surface of the sponge by the pores and on the other by the oscula, and in this case there is no dermal skeleton at all. The mesodermal constituents of a thick ectosome are, as we have already indicated, very various in nature. We have found it a very difficult matter to classify them, but have finally decided to distinguish between stellate, amceboid, fibrous, vesicular and glandular f These elements may be present in various combinations. It is probable that stellate and amceboid cells are present in greater or less quantities in every ectosome, just as they are normally present in the ground tissue of the choanosome between the flagellated chambers. When they are largely developed but unaccompanied by any fibrous tissue, or with only a small proportion of it, they give rise to a thick, gela- tinous ectosome, such as we have found in Spirastrella massa, nobis, Axinella\V}paradoxa, nobis, and in the genus Esj^erella {Esperella gelatinosa, Eidley, Esperella murrayi, nobis, and Esperella lapidiformis, nobis) ; in such cases there is, at any rate very often, a thin outer layer separable from the remainder of the ectosome as a dermal membrane. As an example of a thick gelatinous ectosome we may take for more detailed descrip- tion that of Esperella murrayi. In this sponge the ectosome varies much in thickness in different places, interdigitating in an irregular manner with the choanosome (PI. XL VIII. fig. 2), a condition which appears to be rather characteristic of the genus, and strongly contrasting with the sharp distinction between • ectosome and choanosome in most of those cases where the former forms a true cortex [Suberites caminatus, nobis, Stylocordyla stipitata, var. globosa, &c.). Immediately beneath the epidermis, which consists of the usual flattened epithelial cells, there is a thin, fibrous layer, usually about O'Ol mm. thick, resting upon a strongly developed, reticulate dermal skeleton. Below the dermal skeleton comes another thin, fibrous layer, resembling that just mentioned, and the soft tissues between the two, in which the dermal skeleton is embedded, are also more or less fibrous. Down to the bottom of the lower fibrous layer may be considered as " dermal membrane," between the ends of the primary fibres it is about 0'14 mm. thick. The dermal membrane covering over the subdermal cavities {vide infra), is, however, excessively thin, and contains no dermal skeleton; it is perforated by the pores (PL XL VIII. fig. 2b), and in these areas the fibrous tissue is concentrated in special bands running from side to side between the pores (PI. XIII. fig. 16, 6 ; PL XLVIIL fig. 2&,/). These fibrous bands are composed of densely packed, very much elongated cells, with rather faintly discernible nuclei scattered here and there. They do not run straight from side to REPORT ON THE MONAXONIDA. XXxi side of the elongated pore-areas, but are often obliquely placed and often branched. Judging from their relation to the surrounding parts we have little hesitation in pro- nouncing them to be muscular in function. Immediately below the dermal membrane thus constituted, we find, except of course where the subdermal cavities are situate, the gelatinous tissue which makes up all the rest of the ectosome. This is very loose indeed, and appears to be composed of a frame- work of stellate cells, each with a very small round nucleus ; amongst which are scattered numerous other cells of a very diiferent kind. These (PI. XL VIII. fig. 2c, c. ) are large, averaging about 0*0 15 mm. in diameter, irregular in outline and usually granular, though not very distinctly so. They stain very deeply with borax-carmine and are especially abundant around the ends of the inhalent canals. It is very diflicult to decide under which of the five heads above mentioned these elements should be classed, pro- bably with the amoeboid cells. Another important constituent in the ectosome of Esperella murrayi is furnished by very numerous, irregularly scattered trichodragmata (PI. XLVIII. fig. 2c, t) ; it seems very probable that in this and similar cases the tricho- dragmata, like straw in mortar, serve to bind together the loose, gelatinous tissue in which they lie. In Esperella lapidiformis we have found, embedded in the gelatinous ectosome, numerous very young ova, each consisting of a rounded, granular cell, commonly about 0"02 mm. in diameter, with a small, but very distinct, central nucleus; while in the deeper parts of the sponge (choanosome) there are numerous developing embryos as well as young ova. When, instead of being gelatinous, the ectosome is firm and tough, and contains a large proportion of fibrous tissue, we have a true cortex, such as may be studied to the best advantage in the Suberitidse. Although the general direction of the fibres is usually more or less parallel with the surface, we commonly find also bands of fibrous tissue irregularly surrounding the subdermal cavities, as, for example, in Stylocordyla stipitata, var. glohosa (PI. L. fig. la, f.t), and Suheintes perfectus (PI. L. fig. 2, f.t). In Suherites perfectus also, as we have already noted, the fibrous tissue dips into the choanosome and ensheaths the skeleton-fibres. In Tentorium semisuberites (PI. L. figs. 3, 3a) the ectosome on the upper surface of the sponge is more than usually fibrous and is divisible into three distinct layers. There is (1) an external layer (PL L. fig. 3a, a) about 0'05 mm. thick, wherein the fibres are mostly placed parallel with the surface ; (2) a much thicker, but less dense, intermediate layer (PI. L. fig. 3a, h), which is honeycombed by the very numerous, elongated sub- dermal cavities (s.c), and in which the fibres are mainly placed vertically to the surface ; and (3) a dense internal layer (PL L. fig. 3a, c), rather thicker than the external layer and, like it, with the fibres mostly arranged parallel to the surface. We know that this species has a remarkable power of contraction, which chiefly afi"ects the upper, pore- XXXll THE VOYAGE OF H.M.S. CHALLENGER. bearing surface. Indeed, so much does the appearance of the upper surface differ according to whether the sponge be contracted or expanded, that Sir Wyville Thomson recognised a distinct species for each of these two conditions.^ The figures (PI. L. figs. 3, 3a) repre- sent the sponge in a partially contracted condition. Judging then from these facts, we are probably correct in assigning to the fibrous tissue of the ectosome a muscular function. In Tentorium the thick, fibrous ectosome is present only on the upper, pore- bearing surface (PL L. fig. 3, ^j.«.). Elsewhere the columnar body of the sponge is encased in a cylinder of densely packed, vertically disposed, large tylostyli. Inside of this sheath of spicules comes a thin layer of gelatinous mesoderm, containing stellate ceUs, and inside this again a very thin zone of fibrous tissue in which the fibres are arranged in a circular manner, running round and round the sponge ; very possibly these circular fibres also are muscular in function ; but, judging from then- slight degree of development, we cannot suppose that they are very powerful. The fibrous tissue of the ectosome appears to be always very much the same in histo- logical character, composed of very slender, very much elongated cells, closely packed together, with elongated nuclei scattered here and there. The same kind of tissue is not infrequently found surrounding the larger exhalent canals {e.g., Stylocordyla stijntata, var. globosa, nobis ; Latrunculia apicalis, nobis) and in these cases also it ought possibly to be regarded as part of the ectosome. As in the case of a gelatinous ectosome, there are usually several elements entering into the composition of the fibrous cortex. Thus, in Stylocordyla stipitata, var. glohosa, there are present an enormous number of rounded or oval, nucleated cells (PL L. fig. la, g.c.) lying in and amongst the fibrous tissue; each cell measures about 0 007 mm. in diameter, is somewhat granular and stains fairly deeply with borax-carmine. They are most abundant in the outer portion of the cortex, being closely packed together and making up by far the chief portion of the ectosome in this place. What their function may be is very doubtful, but fi-om their form and their position in the sponge we are inclined to place them in the category of "glandular (?)." We have observed similar, but more highly granular and deeply staining cells in the ectosome of Suherites ramulosus, var. cylindrifera, nobis. In Latrunculia apicalis the ectosome has a somewhat diff"erent composition. In as much as it is tough and firm and sharply marked off from the unc^erlying choanosome {vide PL LI. fig. 1), and contains, moreover, a considerable amount of fibrous tissue, it may be considered as forming a true cortex. Here, as is also the case in the genus Suherites, its toughness and firm consistency are largely due to the presence in it of numerous supporting spicules, the arrangement of which is described elsewhere, and wiU be seen by reference to the figures. There is a thin external layer of somewhat fibrous tissue (PL LI. fig. Ih, f.t.), in which are embedded the bases of the closely- 1 The Depths of the Sea, figs. 23, 24. REPORT ON THE MONAXONIDA. XXX 111 placed, outwardly-projecting discastra (PI. LI. fig. ih, d). Beneath this comes a much thicker layer of very peculiar, vesicular tissue. This is composed of closely packed, perfectly transparent, globular cells (PI. LI. fig. Ih, v.c.) about 0"018 mm. in diameter, and each with a small, round nucleus excentrically placed. (It is interesting to observe that these cells are identical in form and size with those described and figured by Polejaeff in the cortex of his Cacospongia vesiculifera} We are, however, rather surprised that Polejaeff should speak of the cells in question as " thoroughly identical with the vesicular cells of many Desmacidonidse — undescribed indeed hitherto, but undoubtedly very well known to every spongiologist who has had to deal with the repre- sentatives of the family just mentioned." So far as our experience goes we know of no such vesicular tissue in the Desmacidonidfe.) In the outer portion of the cortex, embedded in the layer of vesicular tissue, there occur also numerous, deeply staining, nucleated, granular cells (PL LI. fig. Ih, g.c.) of irregular shape, of about the same size as or a little smaller than the vesicular cells. These are, for the most part, arranged in not very regular groups (PL LI. fig. 1, g.c). It appears very possible that, like the somewhat similar cells in Stylocordyla, they have a glandular function. This layer of vesicular tissue with its embedded groups of glandular (?) cells gradually gives place below to a gelatinous looking tissue (in which the vesicular cells are still abundant), which, as it approaches the choanosome, becomes more or less strongly but irregularly fibrous. The lower two-thirds (roughly speaking) of the cortex are strengthened and supported by abundant spicules, as already mentioned. In the genus PheUoderma, nobis, which forms a great exception to the rule that the Halichondrina are never corticate, the ectosome is about 0'25 mm. thick, and the fibrous tissue is concentrated in a zone of varying thickness in its lower part ; unfortun- ately the histological condition of the single specimen at our disposal is not sufficiently good to allow of a more detailed description. To sum up our observations on this head, we may briefly tabulate the chief different modifications of the ectosome in the Monaxonida as follows : — Ectosome. Thin; forming only a dermal membrane (e.g., Halichondria, Raspailia, Phakellia, Esperiopsis, &c). Thick Gelatinous, commonly with an external, separable dermal membrane (e.g., EqKrella, SpirastreUa). Tough and fibrous, sharply marked off from the choanosome, no separable dermal membrane (e.g., Suberitidae. 1 Zool. Chall. Exp., part xxxi., Report on the Keratosa, p. 59. (ZOOL. CHALL. EXP. — PART LIX. 1887.) Nnn e XXxiv THE VOYAGE OF H.M.S. CHALLENGER. (b) The Choanosome. According to Schulze -^ the ectoderm of the larval sponge furnishes not only the epithelium clothing the surface of the sponge, but also the epithelial lining of the canal system from the pores on the surface to the inhalent apertures of the flagellated chambers ; while the endoderm furnishes the collared cells and also the epithelial lining of the canal system, from the exhalent apertures of the flagellated chambers to the margin of the oseulum ; the remainder of the sponge is mesodermal. If this \dew be correct, all three germinal layers, viz., ectoderm, mesoderm and endoderm, take part in the formation of the choanosome. Nevertheless the choanosome presents us with much less variety in histological structure than does the ectosome ; a fact which is probably to be accounted for by the circumstance that it does not come into such direct contact with external conditions, and is hence less subject to modification. All the diS'erent histological elements (viz., stellate, amoeboid, flagellated, pavement and fibrous cells) which take part in the formation of the choanosome in the Monaxonida, are, with the exception of the flagellated cells, found also in the ectosome, and we may, accordingly, dismiss this part of our subject very briefly. The extent to which the ground substance, in which the flagellated chambers are embedded, is developed varies much. In the Clavulina there is, at any rate as a general rule, very little indeed, and the flagellated chambers are packed very closely together (vide PI. L. fig. lb). In the Halichondrina (e.g., Esperella murrayi, PI. XLVIII. fig. 2d, and Axinella (f) paradoxa, PI. XLIX. fig. 2) there is commonly more, and the chambers lie further apart from one another. It is moreover sometimes stated that there is a distinction between these two groups in the character of this fundamental tissue ; that in the Clavulina it is finely granular, while in the Halichondrina it is clearer and more gelatinous. That there is some difi"erence of this nature there can be no doubt, but it has probably also been somewhat confounded with the diff"erence already mentioned, viz., the much smaller total amount of ground substance present in the Clavulina as compared with the Halichondrina. "We have, in short, no new facts to add concerning the histological structure of the choanosome, except with regard to the fibrous tissue accompanying the bands of spiculo-fibre, and this has already been fully dealt with in treating of the skeleton. Special Structures. We may conveniently describe in this place certain very remarkable and quite unique structures present in an abyssal sponge, Cladorhiza (?) tridentata, nobis. The shape of I Zeitschr. f. wiss. Zool, Bd. xxxiv. p. 438. REPORT ON THE MONAXONIDA. XXXV the sponge is very peculiar and wUl be best understood by reference to the annexed woodcut (Fig. X.) (see also the description on p. 95, and figs. 9, 9o on PI. XX.). Embedded in the soft tissues all round the upper margin of the sponge are found, arranged in a single circle, a large number of small, yellow, globular bodies (woodcut Fig. X., a, and PL XLIX. fig. 4, g.h.\ What these bodies may be is an ex- tremely difficult question to decide. That they are proper to the species and not foreign objects or parasites is rendered almost certain by the fact that they occur in the same position in all three of the specimens present ; and this view of the case is ^'t^^^'t^^^^.t't^. further strengthened by their histological structure, which we thr™o°nge \ »" giobuia^ToS of 1 IT iiiiii/" 1-T,-, • unknown function, x 2. have worked out to the best oi our ability in two specimens. Unfortunately the sponges are in a very bad state of preservation, but the following details of structure are C[uite certain. In radial vertical sections through the margin of the sponge (PI. XLIX. fig. 4), we find that each globular body consists of a central, more deeply staining and granular portion (PL XLIX. fig. 4, (/.&.), surrounded by and embedded in a matrix of faintly staining, perfectly hyaline ground substance (PL XLIX. fig. 4, s). The central portion is sharply marked off from the matrix in which it lies by a much more darkly staining ring, or several concentric rings of a similar substance. Within this ring we find a clear, jelly-like ground substance containing very numerous embedded cells, which give to the central part of the structure the granular appearance already mentioned. The cells themselves (PL XLIX. fig. 4a, c) are irregular in shape, nucleated and highly granular, averaging about 0"018 mm. in diameter. Outside the darkly staining ring we find a very similar hyaline ground substance, but less deeply staining than the inner portion and without the cells. Towards the centre of the sponge this ground substance becomes gradually vacuolated and finally merges into the network of tissue constituting the choanosome (PL XLIX. fig. 4, c/i.). Above, and towards the periphery of the sponge, it appears to extend almost or quite to the dermal membrane, becoming, however, considerably vacuolated (? owing to the action of the spirit). In these parts also, but chiefly towards the periphery of the sponge, there occur, embedded in the gelatinous-looking matrix, a great number of very peculiar, cup-shaped organs (PL XLIX. fig. 4, gl., and fig. 46). Each of these is about 0*1 mm. in diameter, and consists of a large number of small (■? spherical), deeply staining cells packed close together so as to form a thick-walled cup, the mouth of which is turned towards the centre of the globular body. Judging from the number to be seen in a single thin section there must be a very great many of these organs present to each of the globular bodies. Around them the hyaline matrix is frequently shrunk away from their walls (probably owing to the action of the spirit), but usually remains connected at the mouth. The gelatinous XXXVl THE VOYAGE OF H.M.S. CHALLENGER. matrix outside the deeply staining ring contains, in addition to the cup-shaped bodies, numerous other objects, such as spicules and a few large cells, which have apparently been enveloped in it accidentally, and which are arranged with none of the constancy or regularity which characterises the former. It appears to us highly probable that the cup-shaped bodies are aggregations of glandular cells, similar to the glandular cells in the ectosome of other sponges, and that they secrete the tough, jelly-like substance, or, at any rate, that portion of the jelly-like substance outside the deeply staining ring. This is the only hypothesis on which we can explain their constancy in form and arrangement, and it seems to be greatly strengthened by two facts — (1) that they have their mouths turned towards the centre of the globular body, and (2) that the gelatinous-looking substance remains connected with the cup-shaped organ at the mouth when it has shrunk away from it (probably owing to the action of the spirit) in other places. What may be the function of the whole structure is enigmatical in the extreme ; we can but describe them as we find them, and suggest, but merely suggest, that they may be phosphorescent, and serve to attract minute organisms upon which the sponge feeds. Unfortunately the specimens, which came from a depth of 1600 fathoms, are in such an unsatisfactory state of preservation that the remainder of the anatomy and histology cannot be worked out at all ; the superior preservation of the structures described is probably due to their envelopment in the tough, gelatinous-looking matrix. III. The Canal System. (1) The Pores. The arrangement and character of the pores are, as might be expected, most intimately correlated with the condition of the dermal skeleton and of the ectosome. The pores themselves, i.e., the actual openings on the surface of the sponge, through which the water enters the sponge, present us with but little variation either in form or size. They are round or oval, and measure about from 0'05 to 0"2 mm. in diameter. We have in the systematic portion of this work given the measurements of the pores in a large number of different species, which it would be superfluous to repeat in this place ; the average diameter, both in Halichondrina and in Clavulina, may be taken as about 0*1 mm. The pores may be studied to the best advantage in those sponges (viz., the Halichondrina) in which there is a separable dermal membrane, which can be removed and examined under the microscope as a transparent object. They then appear as definitely l^ounded openings, often so closely placed as to reduce the dermal membrane to a mere network (PI. XL VI. fig. 4). When, however, the sponge is corticate and there REPORT ON THE MON AXON IDA. XXXVii is no separable dermal membrane, the pores must usually be sought for in sections taken at right angles to the surface. Whether the arrangement of the pores is primarily dependent upon the arrangement of the subdermal cavities or vice versa, or whether both are dependent upon the arrangement of the dermal skeleton, is a very difficult question to decide. Probably the arrangement of the dermal skeleton, which is of great importance for the jjrotection of the sponge, to a large extent determines the arrangement both of the pores and of the subdermal cavities ; we shall, later on, give strong reasons for beheving this to be the case. In the meantime it is sufficient to remark that aU three are most intimately correlated with one another. In the first place, it is obvious that the pores, in order to fulfil their function of admitting water into the sponge, must always be placed over the subdermal cavities, from which the inhalent canals take their origin. This, together with the arrangement of the dermal skeleton, usually results in a more or less regular grouping of the pores in "pore-areas." Sometimes, however, it is impossible to detect any regular grouping of the pores whatever ; they are simply scattered over the surface, here and there. We have thus two principal types of pore-arrangement to distinguish — the pores may be scattered, or they may be collected in more or less definite "pore-areas." We shall consider these two cases separately. (o) Pores scattered, not collected in definite Pore-areas. It is not very often that we get the pores quite irregularly scattered, for usually they are constrained to group themselves more or less regularly, either by the exigencies of the subdermal cavities or of the dermal skeleton. Still, in cases where there is no dermal skeleton present, or where this is of such a nature as not to interfere with their arrangement, we sometimes find this condition occurring; as for example, in Petrosia his2nda (no dermal reticulation), Reniera suhglobosa (dermal skeleton unispicular), Vom'erula esperioides (dermal skeleton reticulate, but meshes very wide), Esperella mammiformis (dermal skeleton loosely reticulate), Esperella lapidiformis (dermal skeleton absent), Esperella simonis (dermal skeleton irregularly reticulate), Desmacidon (Homosodictya) kerguelenensis (skeleton irregular, very loose), Myxilla rosacea, var. japonica (dermal skeleton radiate), Myxilla frondosa (dermal skeleton loose and irregular), and most Suberitidse (dermal skeleton radiate). But here, as in so many other cases, we can draw no hard and fast line, and in many cases it is impossible to decide whether the pores should be classed as irregularly scattered or collected in areas. Perhaps the best cases of irregularly scattered pores are to be found in the Suberitidse, e.g., Stylocordyla stipitata, var. glohosa, where the dermal skeleton is radiately arranged and therefore does not interfere with the arrangement of the pores (PI. L. fig. 1, 25.). XXXviii THE VOYAGE OF H.M.S. CHALLENGER. (b) Pores arranged in more or less definite Pore-areas. We may consider this part of our subject under tliree heads, according to whether the pores arc (1) localised in areas owing to the arrangement of the dermal skeleton, (2) localised in areas owing to the arrangement of the subdermal cavities, the dermal skeleton being either absent or so feebly developed as not to interfere with their arrangement ; or (3) localised in areas owing directly to the habit of the sponge. (l) Pores localised in areas oiving to the arrangement of the dermal skeleton. — The simplest cases of this kind (if such they can be called) are those in which the dermal skeleton is well developed and regularly reticulate, and the surface is broken up hj the fibres of the reticulation into a number of polygonal areas, in which the pores are scattered. Such a condition we find in Pachychalina lobata (PL XLVI. fig. 4), l)ut, as a glance at the figure will show, these cases might be classed almost as well under the heading of " Pores scattered." We have, however, to consider certain very remarkable cases in which, omng to the arrangement of the dermal skeleton, the pores are collected into j^erfectly definite, confined areas, strongly contrasted with the remainder of the surface of the sponge. We shall describe this condition as it occurs in four distinct s^^ecies belonging to four very difi"erent families, viz., Halichondria latruncidioides, Esperella murrayi, Latruncxdia ajncalis, and Tentorium semisuberites. In Halichondria latruncidioides (PL I. fig. 5) the dermal skeleton is composed of a dense, continuous layer of large oxeote spicules, placed in such a manner (PL XLVI. fig. 5) as to leave no interstices, or very few, through which it would be possible for water to obtain entrance to the sponge. The consequence is that the pores are collected together into perfectly definite, round or oval, raised pore-areas (PL I. fig. 5, p.a.), from which the dermal skeleton is absent and in which they are so abundant as to reduce the dermal membrane to a mere sieve (PL I. fig. 5a). In the genus Halichondria the dermal skeleton is usually reticulate and the pores irregularly scattered or placed in the meshes of the reticulation. Our sponge agrees with the other species of the genus in all essentials, and we are obliged to consider the peculiar arrangement of the pores as a case of special adaptation, consequent upon the unusual density of the dermal skeleton over the general surface. This view of the case is supported by the fact that, where there do happen to be odd gaps in the almost continuous skeleton, we sometimes find j^ores in them, although they are nearly all confined to the special raised areas. This case of adaptation has an unusual interest owing to the curious resemblance which the species in question bears to another sponge coming from the same locality (Station 320), viz., Latrunculia apicalis, and careful investigation has shown us that the similar external form of these two species is due to the same cause in both, as will be seen later on. REPORT ON THE MONAXONIDA. XXxix In Esperella murrayi (PL XIV. fig. 1) we find a still more remarkable arrange- ment of the pores. Here again there is an exceedingly dense and compact dermal skeleton, and the pores are collected in definite grooves from which alone the dermal skeleton is absent. These grooves appear on the surface of the sponge like cracks (PI. XIV. fig. 1, p.a.; PI. XLVIII. fig. 2a, p.a.), and the floor of each is formed by a thin and dehcate dermal membrane, overlying an elongated subdermal cavity (PI. XLVIII. fig. 2a, s.c), and pierced by numerous round pores (PI. XIII. fig. 16, c; PI. XLVIII. fig. 25, p.). They are guarded on either side by dense tufts of large stylote spicules (PI. XIII. fig. 16, a), and, by the action of appropriate bands of muscular fibre, running from side to side between the pores, the two sides can be brought together and the cracks closed (see also pp. xxx, 68). In this case also we find that in other species of the genus {e.g., Esperella lapidi- formis, nobis, Esperella mammiformis, nobis, &c. ), where the dermal skeleton is arranged in such a manner as not to interfere with the distribution of the pores, these latter occur all over the surface of the sponge ; and so we must regard the arrangement of the pores in Esperella murrayi hb due to the exceptional arrangement of the dermal skeleton. And here again this view is strengthened by the fact that, where there is room for them, there are a few odd pores scattered over the general surface of the sponge.^ We have now to consider a similar condition of things occurring in a corticate sponge, viz., Latrunculia. We have only succeeded in satisfactorily working out the relations of the pores in one species of the genus, viz., Latrunculia apicalis, but we have strong reasons for believing that the same arrangement obtains in at least several species of the genus, if not in all {cf. p. 237, &c.). It will be seen from the figure (PI. XLIV. fig. 4) that the upper surface of the sponge is covered with mammiform proj ections. Those nearer the top are larger than the remainder and not nearly so abundant ; each one is conical and has a distinct oscular opening at the summit, to which a wide exhalent canal leads up (PI. LI. fig. 1). The smaller processes, on the other hand, are cylindrical and abruptly truncated at the top, which is commonly slightly concave and without any opening visible to the naked eye ; in longitudinal section (PL LI. fig. 1) we see, however, that numerous slit-like pores (p) lead from the surface vertically 1 Vosmaer has described in Esperella lingua, Bk. sp., a localisation of the pores in grooves very similar indeed to that which occurs in Esperella murrayi, only in the former case the localisation appears to be due to the fact that the general surface of the sponge is covered with sand ; he says, '• The most remarkable thing in Esperella lingua is perhaps the incurrent canal-system. The sponge, covered with sand, shows on its surface numerous fissures that are not covered with sand. In examining the sponge with a lens it becomes clear that in these places the pores are situated (fig. 22, pi. iv.)," Sponges of the "Willem Barents" Expedition, 1880-81, p. 30. Bowerbank in his description mentions no such pore-cracks, and we were inclined to think that Vosmaer's species could not be the same ; but on examining Bowerbaok's dried type (Mon. Brit. Spong., vol. iii., pi. LxxviL fig. 1) we found the pore-cracks in one or two places in a most perfect condition, and were enabled to make a microscopic preparation which showed them to be identical with those of Esperella murrayi, even down to the presence of the transverse bands of muscular (?) tissue. Unfortun- ately, we were not aware of these important facts until after our description of Esperella murrayi was printed, so that we have made no comparisons in that place. The two species are, however, perfectly distinct, as shown by the external form, the arrangement of the skeleton and certain details of spiculation. xl THE VOYAGE OF H.M.S. CHALLENGER. downwards, between radiating brushes of sj)ieules, into one large inhalent canal (i.e.), from which smaller canals or lacunse originate. Here again we find an intimate connection between the arrangement of the pores and the nature of the dermal skeleton, for the entire surface of the sponge, except on the tops of the raised pore-areas, is covered with a dense, continuous armour of outwardly projecting discastra (PI. LI. figs. l,d; lb, d), between which there is no room for the pores. We have already seen how a similar arrangement of the pores, consequent upon the denseness of the dermal skeleton (which is, however, very diff'erently arranged in the two cases, as will be seen by comparison of PL XLVI. fig. 5, with PI. LI. fig. 16), in a totally different sponge, viz., Halichondria latrunculioides, has given rise to a similar external form. We ought perhaps also to consider under this head one more case, viz., that of Tentorium semisuberites, in which there is, however, only a single, large pore-area. The arrangement of the dermal skeleton and ectosome in this sponge has been already referred to (p. xxxi), the columnar body (PI. L. fig. 3) is encased in a quite impenetrable sheath of spicules, except at the top (PL L. fig. 3, p.a), where the dermal skeleton is differently (radiately) arranged. In the centre of the cushion-like top arises usually a single oscular tube (PL L. fig. 3, o), and all around this are numerous pores, seen best in longitudinal section (PL L. fig. 3a), leading into the elongated subdermal cavities between vertical brushes of spicules, much as in other Suberitidse. There are no pores except on the summit of the sponge. (2) Pores localised in areas in accordance with the arrangement of the subdermal cavities; the dermal skeleton being either absent, or so feebly developed as not to interfere with their arrangement. — This condition is found most frequently in non-corticate sponges, and here, as before, it is impossible to say exactly where the scattered condition ceases and localisation begins. We shall content ourselves with taking a few examples in which the localisation is more or less distinctly marked. In the genus Myxilla it is exceedingly common to find the surface of the sponge marked out into more or less regular, oval or rounded pore-areas, marking the position of the underlying subdermal cavities, and this without any relation to the dermal skeleton, which is commonly very vaguely and poorly developed ; as examples we may cite Myxilla crihrigera, nobis, Myxilla mariana, var. massa, nobis, Myxilla nobilis, nobis, and Myxilla compressa, nobis. Fig. 2a on PL XXX. shows this condition as seen in Myxilla nobilis. For further details the reader is referred to the descriptions of the several species mentioned. We also find this condition occurring in the genus Phahellia, e.g., Phahellia flabel- lata, nobis, where the pores occur in smaU groups over the ends of narrow inhalent canals ; and in Phahellia ventilabrum, var. connexiva, we have much the same thing (vide PL XLIX. fig. 3). In neither of these cases is the skeleton arranged so as in any way to determine the arrangement of the pores {cf p. 171, and PL XLIX. fig. 3). REPORT ON THE MONA.XONIDA. xU In short, all that is necessary iu order to bring about such a localisation of the pores is that the subdermal cavities should be arranged at some little distance from one another and have a number of pores leading into each. (3) Pores localised in areas owing directly to the hahit of the sponge. — We have only one example to adduce belonging to this class, although very likely future researches will show such a condition to be not uncommon. The example in question is a deep-sea sponge, Tedania actiniiformis, nobis, obtained at Station 299 (off Valparaiso), at a depth of 2160 fathoms and on a bottom of blue mud. The external form of the sponge will be best understood from the figure (PI. XL fig. 2 ; cf. also p. 55) ; it is attached to a piece of rock. At a short distance below the flattened top there is a definite narrow zone (PI. XI. fig. 2, p.z., and fig. 2a) of pores encircling the sponge. In this zone the pores are so abundant as to reduce the dermal membrane to a mere network. Judging from the general appearance of the sponge and the nature of the bottom upon which it lives, we have little hesitation in saying that in life it was buried in the mud up to within a short distance of the top, and hence the pores, in order that clean water might gain access to them, became confined to that portion of the surface which was above the mud, namely, a narrow zone immediately below the flattened top. The genus Tedania is characteristically an inhabitant of shallow water, and in no other species do we find an arrange- ment of the pores similar to that which characterises this deep-sea form. There can be no doubt that in this case the arrangement of the pores is dependent directly upon the habit of the sponge, and not, as in most cases, upon the arrangement of the dermal skeleton or of the subdermal cavities. Before leaving the question of the pores we must consider briefly the condition of flabellate sponges in this respect. It is an almost invariable rule, that in flabellate sponges the pores are to be found on one surface and the oscula on the other. Thus in Phakellia ventilabrum, var. connexiva (PI. XXXV. figs. 3, 3a ; PL XLIX. fig. 3), and Phakelliajlahellata, nobis (PI. XXXIV. figs. 2, 3, 3a), this arrangement is very well illus- trated, and the same condition occurs in Myxilla frondosa, nobis (PI. XXVI. figs. 1, la); and Gellius Jlabelliformis, nobis (PI. XXVI. figs. 5, 5a). Again in that very remarkable sponge Esperiopsis challengeri (PI. XVIII.) the pores occur only on the concave surfaces of the lamellae (PI. XVIII. fig. 4), while the oscula are all on the convex surfaces. By far the most remarkable instance of this kind is, however, afibrded by a boring (zooL. CHALL. EXP. — PAET Lix. — 1887.) Nnn / Fig. XI. — Climia dissimilis. Vertical section of a decalcified .specimen ; 0,0, osciila ; pa, pore-areas ; c, ectosome ; ic, iuhaleut canals ; ec, exhalent canals, x 4. xlii THE VOYAGE OF H.M.S. CHALLENGER. Suberite which we have described under the name Cliona dissimilis (p. 227; PL XXV. fig. 5, &c.). Here the sponge has bored its way into a flattened coral which it completely surrounds; hence it has itself acquired a flattened, lamellar form, and we find the pores collected in areas (woodcut. Fig. XL, pa), on one side of the sponge and the oscula (woodcut. Fig. XL, o) on the other. There is no other known example, so far as we are aware, of a lamellar Suberitid sponge, and even the species in question is lamellar only because it has bored into a lamellar coral, and yet the pores and oscula are arranged just as they would be in a free living, frondose sponge such as Phakellia. There must be some strong reason why, as soon as a sponge, for any cause, acquires a lamellar form, the oscula become confined to one surface and the pores to the other, and to account for the occurrence of this condition in genera so widely separated as Gellius, Myxilla, Phakellia and Cliona. What this reason may be we cannot at present say. (2) The Subdermal Cavities. The subdermal cavities are the spaces into which the pores directly lead, and from which the inhalent canals proper take their origin. They might be regarded merely as the proximal portions of the inhalent canals, but inasmuch as they are sometimes very distinctly marked ofi" from the remainder of the inhalent canal system, both by position and structure, it is convenient to treat of them separately and to retain for them a special name. Although the term subdermal is hardly a suitable one to apply to these structures in corticate sponges, as we shall see better later on, yet it is advisible to have the same term for the same structures in the two groups Halichondrina and Clavulina, and as the term subdermal cavity is already in use we shaU retain it in preference to inventing a new one. In the Halichondrina the subdermal cavities are usually expanded horizontally and more or less lacunar in form, being roofed in only by the thin, pore-bearing dermal membrane, and many pores usually lead into one and the same subdermal cavity. The extent to which they are developed varies much in difi"erent sponges, attaining its maximum in the genus Ciocalypta amongst the Axinellidse. Here, it will be remem- bered, the thin dermal membrane is supported at some distance from the choanosome on slender pillars of spiculo-fibre, and the numerous pores lead into one large, continuous subdermal space, interrupted only by these supporting pillars.^ In certain other Axinellidee, e.g., Phakellia, the subdermal cavities are also largely, but not nearly so largely developed, but this not does appear to be a constant family character. In most of the Halichondi-ina, however, the subdermal cavities are rather vague and ill-defined, and distinguishable from the remainder of the lacunar canal system more by their position than by any peculiarities in structure ; they show no great regTilarity either ^ BowerLank gives a good figure of this arrangement in Mon. Brit. Spong., vol. i., pi. xxx. fig. 360. REPORT ON THE MONAXONIDA. xliii in size or form. Such appears to be the case in Halichondria panicea, Myxilla nohilis, &c., and in many cases it is extremely difficult to say where the subdermal cavities end and the inhaleut canals or lacunae jDroper begin. In Phakellia and m Esperiopsis challengeri (both of which, it will be remembered, are fiabellate sponges with the pores on one surface and the oscula on the other), we find a somewhat peculiar condition. The pores lead directly into large, irregularly conical chambers (vide PI. XLIX. fig. 3), which often e^ftend almost through the entire thickness of the sponge, and from these great chambers, which we may regard as the subdermal cavities in that they are in direct communication with the exterior, the system of small inhalent lacunse takes its origin. These large, vertically elongated subdermal cavities interdigitate with similar exhalent chambers opening on to the opposite surface {vide PI. XLIX. fig. 3). In those cases where, owing to special circumstances, the pores are confined to special areas, we find, of course, that the subdermal cavities undergo a corresponding modification. Thus in Esperella murrayi the subdermal cavities are narrow, horizontally elongated, branching channels (PL XLVIII. figs. 2, 2a) underlying the pore-areas, and in this case they are sufiiciently distinctly marked ofi" from the true inhaleut canals, which lead out of them by narrow openings (PI. XLVIII. figs. 2, 2a). In the Clavulina (especially the Suberitidse) we find a good deal more regularity in the form and arrangement of the subdermal cavities, a fact which is undoubtedly due to the presence of a distinct cortex, and to the radiate arrangement of the dermal skeleton. The vertical disposition of the spicules in the cortex prevents extension of the subdermal cavities in a horizontal direction, and they consequently become A'ertically elongated, and we have here an important distinction between the Suberitid^ and the Halichondrina. The pores usually lead in the first instance each into a narrow, slit-like canal, between radiating bundles of spicules (PI. L. figs. 1, 3a) ; these canals then expand, several of them often uniting together, and give rise to a wide space which is the chief part of the subdermal cavity (PI. L. figs. 2, 3a, s.c). Sometimes, as in Latrunculia apiccdis (PI. LI. fig. 1) and Stylocordyla stipitata, var. glohosa (PI. L. fig. 1), this cavity cannot be sharply distinguished from the remainder of the inhalent canal system. At other times, however, it is definitely bounded, its vertical limit coinciding with that of the cortex. Of this latter condition it will sufiice to give two examples, viz., Siiberites caminatus and Tentorium semisuherites. In Suberites caminatus the condition of afi'airs is very simple ; each pore leads into a separate, elongated subdermal cavity, which, at first very narrow, gradually increases in diameter, and then, as it reaches the commencement of the choanosome, contracts again more suddenly, being thus somewhat flask-shaped ; occasionally two of these flask- shaped subdermal cavities appear to run into one another. In Tento7-ium semisuherites the arrangement is a much more complex one. The subdermal cavities are each divided into three parts corresponding to the three divisions xliv THE VOYAGE OF H.M.S. CHALLENGER. of the cortex which we have already described (p. xxxi). The pores (PI. L. fig. 3a, ])) lead by narrow canals through the outer layer of the cortex into expanded, elongated cavities in the middle layer of the cortex, and these on reaching the lower layer of the cortex again contract into narrow canals (PI. L. fig. 3a, i/), which at the commence- ment of the choanosome open into the much wider inhalent canals (PI. L. fig. 3a, i.e.). This is the most complex type of subdermal cavity with which we are acquainted.^ These are the principal modifications of the subdermal cavities which are met with in the Monaxonida. It will be seen from what has been said that the term is incapable of a sharp definition, and that the sul:)dermal cavities of one sponge do not necessarily correspond strictly to those of another. In the Halichondrina all we can say of them by way of definition is that they are the spaces beneath the dermal membrane into which the pores directly lead, while in the Suberitidse they are usually represented by so much of the inhalent canal system as is enclosed within the ectosome ; and in both cases they may or may not be sharply marked off from the remainder of the inhalent canal system. (3) The Inhalent Canal System below the Subdermal Cavities. In this section of the canal system we meet with but little variation amongst the Monaxonida. As a very general rule the inhalent canal system, upon entering the choanosome, becomes converted into a system of lacunar spaces, rather than definite canals. Such we have found to be the case in the genera Halichondria, Reniera, Esperella, Myxilla, Axinella, Latrunculia. If, however, as in the genus Esperella, there is a thick, gelatinous ectosome, then the inhalent canals may, after leaving the subdermal cavities, remain perfectly definite while in the ectosome and run straight down- wards to the choanosome, where they at once break up into a system of lacunse. This condition is very well illustrated in Esperella murrayi; each inhalent canal (PI. XLVIII. fig. 2, i.e.) leaves the horizontally elongated subdermal cavity by a narrow exit (PI. XLVIII. fig. 2a, i.e.) and dips vertically downwards towards the choanosome, upon arriving at which it is at once lost in a system of irregular inhalent lacunge lined by flagellated chambers. Another interesting feature of the inhalent canals in this sponge is the manner in which their walls are guarded by projecting spicules (PI. XLVIII. fig. 2c), as we have already described. In Latruneulia apicalis the lacunar arrangement is also well illustrated ; here we have no long, straight canal interpolated between the subdermal cavity and the choanosome, as in Esperella murrayi, but the subdermal cavity into which the pores lead passes directly into a very wide, more or less lacunar channel (PL LI. fig. 1), from which numerous ' Dr. Vosmaer has also described and figured the subdermal cavities in Tentoriuvi semisuberites (Sponges of the " Willem Barents" Expedition, 1880-81, p. 19, pi. iii. fig. 22, &c.). His description agrees in the main with ours, but he appears to have overlooked the important division of the cortex into three distinct layers. REPOET ON THE MONAXONIDA. xlv very much smaller lacunar channels take their origin, and from these again smaller ones ; and here again the lacunae are surrounded by the flagellated chambers (PI. LI. fig. la). In Stylocordyla stipitata, var. glohosa, the inhalent canals are also, at any rate near the surface of the sponge, represented by a system of lacunar channels surrounded by the flagellated chambers (PL L. fig. 1). We have in no case found the inhalent canals breaking up into a system of finer and finer canals, of which the ultimate ramifications end each in a single flagellated chamber. This, however, is probably due to the insuSiciency of the time and material at our disposal, for such a condition is described by Vosmaer in Trichostemma {Polymastia) hemisphxricum,^ and, judging from other accounts, of which, however, none are very definite, would appear to occur also in some other Suberitidse. That it does not occur in all the Clavulina we have already shown ; for in Latrunculiawe have found the lacunar condition as above described, and Pol^jaefl" informs us^ that Suberites domwncula also is "characterised by an entire absence of special cameral canaliculi." (4) The Flagellated Chambers. The flagellated chambers appear, from our researches, to be nearly always globular or subglobular in form in the Halichondrina, and either globular or oval in the Clavulina. In size they vary from about 0-024 to about 0-058 mm. in average diameter. The following table of measurements and forms as observed by us in eighteen speciesmay be of use in showing the relative sizes and shapes of the chambers in the difierent groups : — Species. Average Diameter of Flagellated Chambers. Form of Flagellated Chambers. Halichondria panicea, Petrosia hispida, Reniera sp., Esperella gelatinosa, . Esperellafusca, .... Enperella lapidiformis, Esperella miirrayi, Esperiopsis challengeri, Myxilla nohilis, .... Hymeniacidon earuncida, . Phakellia venHlahrum, var. connexiva,^ Axinella (?) paradoxa, Raspailia tenuis, Suberites eamincdits, . Stylocordyla stipitata, var. glohom, Tentorium semisuherites, Spirastrella massa, Latrunculia apicalis, . 0-0336 mm. 0-0288 „ 0024 „ 0-0336 „ 0-0288 „ 0-0288 „ 0-0240 „ 00432 „ 0-048 „ 0-024 „ 0-0384 „ 0-0336 „ 0-0336 „ 0-0336 „ (but variable) 0-0336 „ 0-058 „ 0-0288 ,. 0-0336 „ Subglobular Globular Globular Subglobular Subglobular Subglobular or oval Subglobular Subglobular Subglobular Subglobular Subglobular Globular Oval or subglobular Oval or subglobular Subglobular or elongated Oval Subglobular Subglobular 1 Sponges of the " Willem Barents" Expedition, 1880-81, p. 13. 2 Zool. Chall. Exp., part sxsi., Report on the Keratosa, p. 80. 3 It is of especial interest to have succeeded in finding the flagellated chambers in this sponge, as both Vosmaer and xlvi THE VOYAGE OF H.M.S. CHALLENGEE. Owing to the shrinking and distortion of the chambers from the action of the spirit it is not easy to arrive at very certain conclusions either as to their form or size ; probably in many cases in which a chamber now appears to be oval this shape is chiefly due to lateral pressure ; and it is noteworthy in this connection that the elongated form is most commonly to be observed in corticate sponges, which, owing to the presence of the fibrous tissue, probably contract more strongly when put into spirit than the non-corticate species. Within the species the size of the chambers appears usually to be very fairly constant, but the above table seems to indicate that it is not likely to prove of much value for systematic purposes, except perhaps occasionally in the distinction of species or at the most of genera. Mr. Carter^ has already pointed out that the diameter of the chambers ("ampul- laceous sacs") is about l-600th of an inch ( = 0'042 mm.)* in the siliceous sponges, and also that in this group they are for the most part globular, results which agree very well with those above given by us. As a rule the chambers in the Monaxonida are embedded in trabeculae of mesodermal tissue which separate the ultimate inhalent lacunae from the ultimate exhalent lacunae, and they open into the latter by wide mouths {e.g., Esiyerella ofiurrayi, PI. XLVIII. fig. 2d ; Axinella (?) paradoxa, PI. XLIX. fig. 2 ; Latrunculia apicalis, PI. LI. fig. 1«). In Stylocordyla stipitata, var. globosa, we have, however, at any rate occasionally, narrow exhalent canaliculi leading away from the flagellated chambers (cf. PI. L. fig. la), a condition which appears to be much more distinctly pronounced in Polymastia ( Weberella) bursa, as described and figured by Vosmaer.^ Hansen have suggested that they are absent. Vosmaer, speaking of Pliakellia howerhanki, expresses himself very guardedly : — "Also in Phakellia howerhanki I could not detect ciliated chambers. As long as I have not studied PhakelUee which I have preserved myself, being sure that they were living in the moment they came in alcohol, I will not pretend that they really do not exist. But it may be suggested here as a possible fact, that those thin fan- shaped sponges are destitute of them because they do not want them. Every good section shows us that the water can flow through and through the body, the natural movement of the water being probably sufficient for bringing new living material to the sponge (fig. 12)" (Sponges of the "Willem Barents" Expedition, 1880-81, p. 24, 1885). The idea that the water can flow right through from side to side, as indicated here and in Vosmaer's figure, must have arisen from the badness rather than the goodness of the sections (cf. PI. XLIX. fig. 3). Hansen expresses himself with much more confidence concerning his researches on the subject : — " Nach diesen Resultaten der Untersuchung muss man annehmen, dass das Wasser durch die Kanale stromt ohne durch Flimmerbewegung fortgeleitet zu werden ; die Oefi'nungen an der Oberflache kann man demnach als Ostien oder Poren bezeichnen je nach Belieben ; sie functionieren wahrscheinlich als beides zugleich. Da die Phakellien meistens diinne Flatten bilden, die fiist wie ein Sieb durch- lochert sind, wird es kaum schwierig zu verstehen, dass ihnen Nahrung zugefiihrt werden kann einfach durch die Wasser- strbmung ohne dass es ndthig ware diese Stromung durch besondere Einrichtungen zu fordem." (Bergens Museums Aarsberetning for 1885, Bergen, 1886). And this, notwithstanding the fact that Hansen has "keine Schwierigkeit gehabt frisches Material zu beschaffen." The flagellated chambers in Phakellia ventilabrum, var. connexiva are perfectly distinct and closely crowded together {vide PL XLIX. fig. 3), and there cannot be the slightest doubt that they are present also in other species of the genus. ' Ann. and Mag. Nat. Hist, ser. 4, vol. xvi. p. 22. 2 Not 0'42 mm. as given by Vosmaer (Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 129), which is probably a misprint. 3 Sponges of the " Willem Barents " Expedition, 1880-81, p. IV, pL iii. fig. 9. EEPOET ON THE MONAXONIDA. xlvii We have never succeeded iu detecting the inhalent openings of the chambers, nor, so far as we are aware, has any one else done so in a Monaxonid sponge. Possibly this is due partly to the fact that they are excessively minute and close up almost or quite entirely when the sponge is treated with reagents ; and also to the difficulty in obtaining very thin sections of siliceous sponges. That they exist there can be no reasonable doubt, and there is no need for us to take refuge in the novel theory, recently pro- pounded by Mr. Carter,^ that the flagellated chambers (" ampuUaceous sacs") have each only a single opening. (5) The Exhalent Canal System. We have been obliged, in speaking of the flagellated chambers, somewhat to forestall our observations on the exhalent canal system, and we have little to add. It is usually, at any rate in its finer ramifications, lacunar like the inhalent system ; so that we have a system of interdigitating, ramifying lacuna, some inhalent and some exhalent, separated from one another by the mesodermal tissues in which the flagellated chambers are embedded. This appears to be always the condition in the Halichondrina (PI. XLVIII. fig. 2cZ ; and PL XLIX. fig. 2), and, sometimes at any rate, in the Clavulina (PI. LI. fig. la). But, as already pointed out, we may also, in the Clavulina, find narrow exhalent canaliculi leading away from the flagellated chambers, as iu Stylocordyla (PL L. fig. la) and Polyinastia ( Weberella) bursa. The small exhalent channels gradually unite together into wider and wider and usually more and more definite canals, which ultimately open on to the surface at the oscula. Even in species in which the ultimate ramifications of the exhalent system are lacunar, the larger canals are usually perfectly distinct and definite, as, for example, in Esperella murrayi (PL XLVIII. fig. 2, e.c.) and Latrunculia apicalis (PL LI. fig. 1) ; and they are occasionally (e.gr., Spirastrella solida) provided with very definite, circular diaphragms, occurring at intervals, for the purpose of regulating the outflow of the water; while at other times {e.g., Latrunculia apicalis and Stylo- cordyla stipitata, var. glohosa) they are surrounded by a sheath of fibrous tissue which probably serves the same purpose. (6) The Oscula. The structures described under this name by various authors are not, in many cases, as pointed out abeady by Vosmaer, homologous one with another. Indeed, in the present state of our knowledge it is impossible to unravel the intricate question of the homologies of the oscula. In this work we shall, therefore, use the term in a purely ' Ann. and Mag. Nat. Hist, ser. 5, vol. six. p. 203, et seq. xlviii THE VOYAGE OF H.M.S. CHALLENGER. empirical sense to designate the openings on the surface of the sponge through which the water is discharged from the canal system. In this sense the word is now generally understood."' Thus in the case of tuljular sponges, like Siphonochalma (PI. VII.), the osculum is the large opening at the top of each tube. In the case of cup-shaped sponges, however, such as Tedania infundihuliformis (PI. XI. fig. 1), we cannot call the wide expanded mouth of the funnel an osculum, although probably it is homologous with the opening of the tube in Siphonochaliyia. In flabellate sponges, as we have already had occasion to point out (p. xxxix), the oscula are usually confined to one surface; these flabellate forms by folding round give rise to funnel-shaped sponges — e.g., PhakeUia ventilahrum and Phyllospongia (which may be either flabellate or cup-shaped), — in which the oscula are usually situated on the inner surface of the funnel. By a further extension of the same process the sponge becomes tubular, with a comparatively narrow opening at the top which w^e now call the osculum. How far this folding and complication may be carried we do not know. The oscula in the Monaxonida vary very greatly in size ;" they may either be small and abundant {e.g., Petrosia hispida, PL III. fig. 2 ; PhakeUia ventilahrum, var. connexiva, PI. XLIX. fig. 3 ; Esperiopsis challengeri, PI. XVIII. fig. 1 ; Homceodictya grandis, PI. XXII. figs. I, \b, &c.), or large and comparatively few {e.g., Esperella murrayi, PI. XIV. figs. 1, la, Spirastrella papillosa, PI. XLI. fig. 5, &c.). They are very commonly found on the uppermost part of the sponge, as in Halichondria latruncidioides (PL I. fig. 5), JEsp>ereUa murrayi (PL XIV. figs. 1, la), Sp)trastrella solida (PL XLI. fig. 7), &c., but their distribution is very variable. Their form also varies much ; they may either have their margins level with the general surface of the sponge and not surrounded by any raised collar, as, for example, in Pacliychalina {?) punctata (PL VI. fig. 2), or they may be surrounded each by a membranous collar, as in Esperella lapidiformis (PL XVI. figs. 2, 2b), or they may be raised on the summits of papiUse, as in Latrunculia apicalis (PL XLIV. fig. 4 ; PL LI. fig. 1). They may be merely the openings of shallow, basin-like depressions, into which numerous exhalent canals discharge Ijy numerous small mouths, as in Poxhyclialina fibrosa (PL IV. fig. 3); or they may be the openings of long, wide canals coming up from deep in the body of the sponge, as in Latruncidia apicalis (PL LI. fig. 1). Sometimes they are arranged in stellate groups, a condition which appears to be most common in flabellate sponges {e.g., Homceodictya grandis, PL XXII. ; and PhakeUia flabellata, PL XXXIV. fig. 3a); or, again, there may be only a single osculum to each individual, as in Stylocordyla (PL XLIII. fig. 10). ' Cf. Vosmaer, Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 127. 2 For measurements see the Description of Genera and Species. EEPORT ON THE MONAXONIDA. xlix Lastly the sponge may be lipostomous, i.e., without any visible oscular openings at all ; such appears to be the case, commonly at any rate, in the deep-sea genera, Cladorhiza and Chondrocladia (PI. XX.). General Kemarks on the Canal System. We may briefly sum up our more important conclusions with regard to the canal system of the Monaxonida as follows : — (1) The arrangement of the pores varies almost indefinitely, being to a large extent dependent either directly or indirectly (through the dermal skeleton), upon the external conditions under which the species lives. They may be scattered all over the sponge or localised in more or less definite areas. (2) The canal system in the Halichondrina, usually if not always, belongs to Vosmaer's third type,^ that is to say, it is more or less lacunar, and the flagellated chambers open by wide mouths into wide exhalent lacunae. In the Clavulina the canal system may belong either to the third or fourth ^ types ; in the latter case the flagellated chambers are provided with special " cameral canaliculi." Thus our general conclusions with regard to the type of the canal system are quite in accordance with those of Vosmaer'' and Polejaefi".* It is true that two naturalists have attempted to establish the existence of a racemose type of canal system in Halichondrine sponges. Keller's attempt^ in the case of Reniera semituhulosa has been already severely criticised by Polejaefi'," who comes to the conclusion that " the observation of this naturalist on the structure of Reniera semituhulosa, executed under the influence of Professor Haeckel's statements on the non-existing racemose type of the canal system " is " unreliable"; a conclusion which we cannot but endorse. The second attempt does not appear to have attracted much attention, and certainly does not deserve to do so ; for, whatever may be the sponge which Saville Kent' mentions under the name Esperia sp., we are confident that it never had a canal system of the quite impossible kind figured, in which the diameter of the flagellated chambers is about twice that of the pores, and there are (to judge from the figure) no exhalent canals whatever. 1 Vide Bronu's Klass. u. Orduuiig. d. Thierreichs, Porifera, p. 130. '^ Loc. cit. ^ Bronn's Klass. ii. Ordnung. d. Thierreichs, Porifera, p. 335. * Zool. Chall. Exp., part xxxi., Eeport on the Keratosa, p. 80. '' Zeitschr. f. wiss. ZooL, Bd. xxx. p. 579, pi. xxxvii. fig. 1. ^ Zool. Chall. Exp., part x.xxi.. Report ou the Keratosa, p. 79. ' Manual of the Infusoria, vol. i. p. 169, pi. vii. fig. 2. (zool. chall. Exr. — part lix. — 1887.) Nnu g THE VOYAGE OF H.M.S. CHALLENGER. B. Embryological Notes. We had. at first intended to give a somewhat detailed account of a mimber of embryos which we have met with in the course of our work, but after consulting the literature ^ bearing upon this much vexed portion of our subject, we have decided that it is desirable to say as little as possible. There are already so many contradictory ojiinions with regard to the development of the sponges, and, according to Heider, it is so extremely unsafe to draw conclusions from specimens that have been preserved in spirit, that, for fear of adding to the ah-eady existing confusion, we naturally feel very diffident about recording our observations ; we shall, therefore, make our remarks on this head very short. Concerning the position in which the embryos develop in the sponge, the most interesting point seems to be that, as might be expected, they always develop in the position of greatest security. In large, massive sponges it is obvious that, so long as they do not lie very near to the surface, the position is a matter of no very great importance, and accordingly in Esperella lapidiformis, e.g. , we find them scattered through the choanosome in groups, commencing a short distance below the surface of the sponge (PI. XVI. fig. 2a, e). In smaU and delicate species, however, the position becomes a matter of considerable importance ; thus in Esperella hiserialis, where the sponge consists of a central spicular axis coated by only a thin layer of soft tissues, we find the embryos (PI. XIV. fig. 3, e) taking refuge in the centre of the spicular axis. Again in Esperella mammiformis, a hemispherical sponge (PL XIV. figs. 5, 6) with flat base attached to stones, the embryos are found grouped close to the stone near the centre of the base. In Chondrocladia crinita, a "Crinorhiza" form (PI. XX. fig. 4), the embryos again occur in the soft tissues near the centre of the sponge, and each appears to be surrounded by a dense mass of the characteristic isochelse of the species. The ova are, of course, developed from amoeboid mesodermal cells, and it appears from our observations on Esperella lapidiformis {vide p. xxxi) that they may originate either in the choanosome or the ectosome (the latter being gelatinous), but that they develop only in the choanosome, whither we must imagine that those which originate in the ectosome migrate. The free-swimming larva escapes from the parent sponge through the exhalent canals, as is well shown in some of our preparations of Esper-ella murrayi {vide PI. XLVIII. fig. 2, e). The commonest type of embryo met with by us is that R'hich occurs in Esperella 1 Gf. Carter, Development of Marine Sponges, Ann. and Mag. Nat. Hist, ser. 4, vol. xiv. p. 321, et seq.; Keller, Studien tiber Organization u. Entwickelung der Chalineen, Zeitschr. f. wiss. ZooL, Bd. xxxiii. p. 317 ; Marshall, Die Ontogenie von Reniera filigrana, Zdtschr. f. iviss. Zool, Bd. xxxvii. p. 221 ; Schulze, Die Faniilie der Spongidte, Zeitschr. f. wiss. Zool, Bd. xxxii. p. 593 ; Sollas, On the Development of Halisarca lobularis, Quart. Journ. Micr. Sci., K.S., vol. xxiv. p. 6U3 ; Heider, Zur Metamorphose der Oscarella lobularis, Arb. zool. Inst. Wien, Bd. vi.. Heft 2, p. 175 ; &c. REPORT ON THE MONAXONIDA. U mammiformis, nol:)is. In the particular stage referred to the larva is more or less spherical in form, and composed of an outer (? single) layer of small, ciliated, ectodermal cells, and a central mass of much vacuolated tissue. The central mass of tissue has probably been a good deal affected by the action of the spirit ; it appears now to consist of strings of plasma (apparently composed largely of stellate cells) containing numerous spherical nuclei and nucleoli. Probably in the living embryo this central mass was composed of a continuous, structureless, gelatinous matrix with embedded stellate cells, like the ground substance of the adult sponge, but more delicate. Conspicuous in it are numerous stylote and anisochelate spicules, arranged apparently without any order. The palmate anisochelae are more numerous and appear to have reached a higher state of development than the styli. They average about 0'05 mm. in length as against 0"072 mm. in the adult sponge. The styli on the other hand measure only about 0'34 by 0'005 mm., as against 1"0 by 0"019 mm. in the adult. The entire embryo measures up to nearly I'O mm. in diameter, and is enclosed in a membranous capsule. The ectodermal layer in three specimens which we have examined at about this stage exhibits foldings or invaginations which may or may not be due to the action of the spirit. In a younger embryo of the same sponge we have been able to trace the development of the palmate anisochelse, as recorded on p. xx. In Esperella hiserialis the embryos found in the spicular axis of the sponge agree essentially with those just described for Esperella mammiformis, i.e., they consist of an irregular sphere of small ectodermal cells enclosing a central mass of tissue with numerous spicules. In one embryo, howcA'-er, the ectodermal layer appears to be absent from one pole, at which the central mass of tissue -comes to the surface. The embryos are again enclosed in membranous capsules. In Myxilla nohilis we have found numerous embryos in various stages of develop- ment, amongst which a stage corresponding with that first described is again common. The embryo, about 0"24 mm. in diameter, consists of a sphere of ectoderm, which may be invaginated in many places, composed of small, prismatic (?) cells, enclosing a mass of tissue, in this case rather compact, containing numerous megaselera and microsclera. It is important to observe that the styli of the embryo, which are usuall}' arranged in a single dense sheaf with their bases all together at one side of the embryo and their apices pro- jecting into the centre, are all entirely spined and resemble the echinating styli of the adult sponge, and not the large styli, which in the adult are entirely smooth or only very slightly spined at the base [vide p. 141). They are straight, and gradually and sharply pointed, sometimes with distinct heads, and measure about O'll by 0-005 mm. The chelae appear to develop before the styli. The characteristic arrangement of the megaselera in the embryo appears to have already been noted by Carter in an embryo of a siliceous sponge.^ 1 Ann. and Mag. Nat. Hist, ser. 4, vol. xiv. p. 333, pi. xxii. fig. 28, e. lii THE VOYAGE OF H.M.S. CHALLENGER. It thus appears that there is one developmental stage fairly constant in the Halichondrina, in which the embryo consists of a sac of small, probably prismatic, ciliated ectodermal cells, enclosing a central mass of mesodermal (?) tissue containing the developing spicules. In some cases, perhaps in all (for it would be difficult owing to the nature of the case to make sure of the contrary) the ectoderm appears to be absent from one pole, at which the mesoderm (?) comes to the surface. This stage probably nearly corresponds with one figured by Keller for his Chalinula fertilis,^ by Carter for Halicliondria simulam and Esperia wgagropila,^ by Marshall for Reniera Jiligrana,^ and by Schulze for Euspongia officinalis adriatica.* There is one other type of embryo, observed by us in Esperella lapidiformis only, which seems worth noticing in this place. The embryos (PI. XVI. fig. 2a, e) are very numerous, more or less spherical, measuring about 0'5 mm. in diameter. Each is enclosed in a membranous capsule, and is composed mainly of a dense, solid, finely granular mass of tissue (PL XLVIII. fig. 1, a), with small nucleoid bodies scattered through it. This structure prevails throughout the embryo excepting at one pole, where there is a hemispherical cap of large cells (PI. XLVIII. fig. 1, c), appearing in sections to be polygonal from mutual pressure, each with a large oval nucleus (?i), about 0"01 mm. in longer diameter, and a nucleolus. The cap of cells appears sometimes to be slightly shrunk away by the action of the spirit from the remainder of the embryo. This would seem to be a much earlier stage of development than those above described ; there are as yet no traces of spicules, 1 Zeiischr.f. iciss. Zool., Bil. xxx., pi. xix. fig. 15. 2 Ann. and Mag. Nat. Hist., ser. 4, vol. xiv., pi. xxi. figs. 21, 22, 25. ' Zeitschr. f. wiss. Zool, Bd. xxxvii., pi. xiv. figs, b, c. ^ Zeitschr. f. wiss. Zool., Btl. xxxii., pi. xxxviii. fig. 4. REPORT ON THE MONAXONIDA. liii CHAPTER III.— THE CLASSIFICATION OF THE MONAXONIDA. I. The Data of Classification. That the detailed classification of any group of sponges would be a difficult task, one might be led to suppose on d priori grounds alone, from the consideration of the low- position which sponges occupy in the animal kingdom, or, in other words, from the con- sideration of their low degree of specialisation and the consequent scarcity of definite distinguishing characters. No two sponges are exactly alike, and yet all the so-called species in a genus resemble one another so nearly, and are so often still connected by intermediate forms, that the distinction of species often becomes purely an arbitrary question and a matter of individual judgment. The characters of the species do not appear to be nearly so firmly fixed as in higher groups, they seem to be in a very plastic condition, and capable of almost infinite modification according to their surroundings. This being so it is obvious that we must, for purposes of classification, endeavour to discover those structures in a sponge which are least subject to modification. That external form is useful for this purpose only to a very limited extent, and when the utmost caution is exercised, is now agreed on all hands. If further proof be needed of this fact the Challenger collection supplies it in abundance, and it is perhaps worth our while to give one remarkable illustration. We shall describe in the systematic portion of the present work a number of species shown by their spicules to belong to three distinct genera, viz., Cladorhiza, Chondro- cladia, and Axoniderma, all characterised by the same external form, and that a very remarkable one. In brief, each sponge consists of a small, subglobular or conical body provided with an equatorial zone of long, stiff, supporting processes. To this peculiar external form {vide PL XX. figs. 2, 4, 5, 7, 8) we have given the name " Crinorhiza form " ; all sponges as yet known to possess it come from very great depths and live on a bottom of soft mud or ooze, and there can be no doubt that the long, radiating processes, usually associated with a single vertically descending, central, root-like process, are for the purpose of supporting the sponge and preventing it from sinking into the soft mud on which it lies. We only find the Crinorhiza form in species which live in very deep water ; species of the same genera which inhabit shallower water do not possess it. It would be an easy matter to adduce further instances in support of the same law, namely, that similar external conditions beget similar external forms even in species of distinct genera, but it is needless to multiply proof. Moreover, the external form of one and the same species (as shown by its spiculation) varies greatly, sometimes with no apparent reason at all, as, for example, in Petroda similis, nobis (PI. III. figs. 3, 4). We must then look elsewhere than to external form for a guide to classification. liv THE VOYAGE OF H.M.S. CHALLENGER. Bowerbank tliought that he had discovered the great desideratum in the arrangement of the skeleton, and his subfamilies and genera were consequently made to depend solely upon this character. But a glance at his sj^stem shows the utter failure of such an attempt. Let us take an example from his work •} — " Order II. Silicea. Suborder I. Spiculo-radiate skeletons. Not reticulate. Composed of spicula radiating in fasciculi or separately from the base or axis of the sponge." In this suborder Bowerbank places the following genera : — 1. Geodia (Tetractinellid). 2. Pachymatisma (Tetractinellid). 3. Ecionemia (Tetractinellid). 4. Alcyoncellum (Hexactinellid). 5. Polymastia (Monaxonid). 6. Ilalyphysema (Foraminifer). 7. CiocaJypta (Monaxonid). 8. Tethea (partly Tetractinellid). 9. Halicnemia (Monaxonid ?). 10. Dictyocylindrus (Monaxonid). 11. Phakellia (Monaxonid). 12. Microciona (Monaxonid). 13. Hymeraphia (Monaxonid). 14. Hymedesmia (Monaxonid). Such is the result of applying the arrangement of the skeleton as a sole guide to the distinction of genera ; we need say no more. But although the arrangement of the skeleton cannot be applied by itself to the distinction of genera, yet it is by no means valueless as a guide to classification when taken in conjunction with other characters, especially when applied to larger sul)divisions than genera. Thus, in the more careful and discriminating hands of Mr. Carter it has led to much better results ; the Axinellidee, Ectyoninse, and Suberitidse, being all characterised by special types of skeleton arrangement fairly constant within the group ; still it is at the best an uncertain guide. Hitherto a great deal of reliance has been placed upon the greater or less amount of spongin, or horny cementing material, present in the skeleton, but of late years this character also has been losing favour amongst spongologists. Indeed, as has already been pointed out, it is the great merit of Dr. Gray's system that in it the horny sponges are included amongst the SUicea. Dr. Gray was, however, a good deal in advance of his times in this respect. AVe hope to show in this place that even within the group of Monaxonida the amount of spongin present varies in such an irregular fashion as to be of comparatively little use for purposes of classification. In all the Halichondrina spongin is usually present, and we find that the amount stands, roughly speaking, in inverse proportion to the number of spicules. Thus, taking first the Homorrhaphidse, the division between the Renierinse and Chalininse is a purely arbitrary one ; both groups contain more or less spongin, very few (if any) species even of Renierinse containing none, while in the Chalininse the amount is relatively very large. 1 Mon. Brit. Spong., vol. i. pp. 159, 167, ka. REPORT ON THE MONAXONIDA. Iv The form of the spicules in both groups is exactly the same. Hence there is good reason for supposing that the Chalininse have been derived from the Eenierinse by the development of spongin at the expense of the spicules, and, moreover, they seem to be of polyphyletie origin, descended from several distinct genera of Eenierine sponges. In the Heterorrhaphidfe we find examples of exactly the same thing. The genus Gelliodes differs from the genus Gellius only in the greater development of horny fibre, and the genus Toxochalina contains so much spongin in the fibre that it has hitherto been placed amongst the Chalininse, although its spiculation shows it to be one of the Heterorrhaphidae. Amongst the Desmacidonidfe we have abundant instances. We have species of the genus Esjierella with little or no horny matter, and other species with a well-developed horny skeleton, and precisely the same thing occurs in the genus Myxilla. Clathria and Echinoclathria are other instances of genera of the Desmacidonidse in which there is a very highly developed horny skeleton. In the Axinellidse the same facts may again be observed, and the Challenger dredgings have brought to light a highly instructive example of an Axinellid sponge [Axinella Jibrosa, nobis) possessed of a very strongly developed horny skeleton. We have abeady remai'ked the fact that the siliceous skeleton decreases in proportion as the horny skeleton increases, until i;ltimately it disappears and leaves us face to face with a horny sponge. No wonder then, considering their probable origin from many distinct groups of siliceous sponges, that Polejaeff found such great difficulty in classifying his Keratosa. We must mention in this jilace another very suggestive and important fact with regard to spongin, and that is, that its degree of development depends upon locality. Sponges with horny fibre are far and away more abundant in tropical or subtropical seas than in temperate or frigid areas. When a sponge gets into a warm area it tends to develop horny fibre. This is true at least of the Halichondrina amongst Monaxonida, as is sufficiently shown by a glance at our tables of geographical distribution and the description of those species obtained from the warmer areas. As regards the value of the minute anatomy and histology from a systematic point of view we are hardly able as yet to form any conclusions, for our knowledge of this subject is at present in its earliest infancy. We have in the last chapter given details on this head which will, we hope, ultimately be of use in settling the question ; at present all we can do is to collect information. So far as our researches yet go it seems that the nature of the ectosome, i.e., whether it forms a fil^rous cortex or exists as a thin dermal membrane or as a thick gelatinous layer, is likely to be of great systematic value. Amongst the Halichondrina we know of only a single instance, the genus Phelloderma, nobis, in which the ectosome forms a distinct, fibrous cortex, whilst amongst the Clavulina a distinct, thick, fibrous cortex is abamst invariably present. The arrangement of the Ivi THE VOYAGE OF H.M.S. CHALLENGER. pores we have shown to be almost utterly untrustworthy except for specific distinctions, and we are inclined to set very little value, except for purposes of a most general nature upon the fundamental arrangement of the canal system, that is to say, whether it belongs to Vosmaer's third or fourth type, for the simple reason that in nearly every Monaxonid sponge as yet worked out (and we believe also in the Keratosa) it belongs to the former. Still, although the same fundamental type of arrangement will probably be found to prevail throughout the group, yet it is also probable that many minor differences in the canal system of difi"erent genera will be discovered, and will prove to be of great systematic value. As examples of such minor differences we may jwiut to the character- istic arrangement of the larger inhalent and exhalent canals in the genus PhakcUia, and the arrangement of the inhalent canal system in the genus Esperella, as already described by us, in both of which genera the canal system differs in the respects mentioned from the more common Halichondrine type found in Halichondria, &c. As another example we may note the presence of distinct sjihincters or diaphragms in the larger exhalent canals of the genus Spirastrella, which seems to be a fairly constant generic character. We come now to the consideration of the spicules, and of the two categories of these we wiU take first the megasclera, which we shall find to be of very great service in classifica- tion. The same form of megasclera is found to run through large divisions ; thus, in the family Homorrhaphida3, comprising the Eenierines and Chalininge, we find only smooth oxeote megasclera and no microsclera. In the Clavulina, again, the megasclera are almost all tylostylote or stylote. But here we occasionally meet with startling exceptions, one of the most striking being the genus Stylocordyla, a corticate sponge, agreeing in skeleton arrangement, &c., with other Clavulina. but always with oxeote sjiicules. In theHeteror- rhaphidse and Desmacidonidse the form of the megasclera is very variable, but we can always use it for purposes of generic distinction. The size of the megasclera is of very slight use save for distinguishing species, but here it is of the greatest value, especially in groups like the Homorrhaphidse, where even the different genera have all the same form of spicule. The microsclera, when present, are of still greater systematic value than the megasclera ; indeed they form the best guides to the classification of the Monaxonida. The reason of this is probably that they ai-e not subject to modification to suit the external conditions of the sponge ; and, further, they are usually more complex in form, and thus present more points of possible diS"erence than do the megasclera. Their use in the economy of the sponge is probably very slight ; in many cases we can hardly believe, from their minute size and irregular arrangement, that they have any, and hence there appears to be no reason why they should undergo much modification with changing conditions.^ The structures which we have hitherto considered are, on the other hand, ^ " It is a strange result wliich we thus arrive at, namely, that characters of slight vital importance to the species, are the most important to the systematist ; " Darwin, Origin of Species, ed. 6, p. 176. REPOKT ON THE MONAXONIDA. Ivii all of the greatest importance to the sponge, and subject to great modification with every change in external conditions. Of this we have abundant proof ; we have seen already how sponges in warm climates tend to develop much spongin, while the megasclera gradually become fewer in number and less in size. We have also shown that the arrangement of the pores is largely dependent on the arrangement of the dermal skeleton and varies with this, and so forth. But the form of the microsclera does not appear to be dependent upon external conditions or upon any other part of the organisa- tion, and hence we find in them great constancy. Let us take, for example, the large family of the Desmacidonidse. In this famUy we find megasclera of several distinct types, and their arrangement in the skeleton varies much ; the arrangement of the pores varies enormously, and the external form if possible varies even more. But throughout the whole family we find one and the same fundamental form of microsclera (chelse), commonly associated with others. These chelse are such remarkable and complex forms that we cannot possibly imagine that they have arisen independently in the diS"erent sub- families and genera, and hence we are obliged to place all these together and separate them from the remainder of the Monaxonida, none of which possess chelae. To sum up briefly, the result of our investigation is as follows : — (1) AVe must always endeavour to classify by an assemblage of characters ; (2) of individual characters the form of the microsclera (when present) is the most important, and this owing to the fact that these are subject to less modification than other parts of the organism. 11. The Classification of the Monaxonida here adopted. Having thus investigated the data of classification and arrived at certain conclusions with regard thereto, we must now set forth the results to which these conclusions have led us in classifying the Challenger collection, and then discuss the mutual relations of the various subdivisions. In the following scheme of classification we shall mention only those genera which are present in the collection, as the time at our disposal is not sufficient to enable us to give a complete resume of this very intricate subject. The Challenger collection is, how- ever, very fairly complete as regards genera, and quite sufiiciently so to justify this proceeding, and although other genera are not here mentioned it must not therefore be thought that we have ignored them in considering the question. The present scheme is based upon the schemes of previous writers, which have been more or less assimilated and modified in accordance with our increase of knowledge. For the sake of brevity we have also decided to omit in this place all diagnoses ; these will be found given fully in the Description of Genera and Species, to which the reader is referred. (ZOOL. CHALL. EXP. — PART LIX. — 1887.) Nnn h Iviii THE VOYAGE OF H.M.S. CHALLENGER. Order MONAXONIDA (p. !).» Suborder I. HALICHONDEINA (p. 1). Family I. H o M o R R H A p h i D ^ (p. 1). Subfamily 1. Eenierin^ (p. 1). Genera — Halichondria (p. 1). | Petrosia (p. 9). Reniera (p. 14). Subfamily 2. Chalinin.e (p. 18). Genera — Pachychalina (p. 19). | Chalma (p. 25). Siplionochalina (p. 29). Family II. Heterorrh aph ID^ (p. 31). Subfamily 1. Phlceodictyin^ (p. 31). Genera — Rhizochalina (p. 32). | Oceanapia (p. 36). Subfamily 2. Gelliin^ (p. 37). Genera — Gellius (p. 37). I Gelliodes (p. 47). • Toxochalina (p. 49). Subfamily 3. Tedaniin^ (p. 50). Genera — Tedania (p. 50). | Trachytedania (p. 57). Subfamily 4. Desmacellin^ (p. 58). Genus Desmacella (p. 58). Subfamily 5. Hamacanthin.e (p. 59). Genus Vomerula (p. 59). 1 The pages referred to are those on which the diagnoses will be found. REPORT ON THE MONAXONIDA. lix Family III. D e s m A c i d o N i d je (p. 62). Subfamily 1. Esperellin^ (p. 62). Genera — Esperella (p. 62). Esjyeriopsis (p. 76). Cladorhiza (p. 86). Axonidernia (p. 96). Chondrocladia (p. 99). Desmacidon (p. 103). {Homoeodictya (p. 108)). Artemisina (p. 112). Fhelloderma (p. 113). Sideroderma (p. 114). lophon (p. 116). Amphilectus (p. 123). Subfamily 2. Ectyonin^ (p. 128) Genera — Myxilla (p. 128). Clathria (p. 146). Rhaphidophlus (p. 151). Plumohalichondria (p. 155). Plocamia (p. 157). Acarnus (p. 159). Echinoclathria (p. 159). Agelas (p. 163). Echinodictyum (p. 164). Family IV. Axinellidje(p. 166) Genera — Hymeniacidon (p. 166). Phakellia (p. 169). Cioccdypta {p. 173). Acanthella (p. 176). Axinella (p. 178). Raspailia (p. 188). Dendropsis (p. 191). Thrinacophora (p. 193). Suborder II. CLAVULHSTA (p. 196). Family I. S u b e r i t i d /E (p. 197) Genera — Suherites (p. 197). Polymastia (p. 210). Proteleia (p. 214). Trichostemma (p. 216). Tentorium (p. 221). Stylocordyla {]>. 222). Quasillina (p. 225). CTiona (p. 227). Family II. S p i R a s t r e l l i d ^ (p. 229). Genera — Spirastrella (p. 229). | Latrunculia (p. 233). Ix THE VOYAGE OF H.M.S. CHALLENGER. III. Discussion of the Different Subdivisions. The Monaxonida thus comprise two fairly natural suborders, the Halichoudrina and the Clavuhna. But whether these two suborders are sufficiently closely related to one another and sufficiently distinct from other suborders to admit of their Ijeing united in one natural order, the Monaxonida, as opposed to the Tetractinellida, is very doubtful. In the classification of Dr. Vosmaer the order Monaxonida finds no place, and we are inclined to agree with this authority, who accepts the suborders Halichondrina and Clavulina as themselves natural, but refuses to allow a special order for the reception of these two groups to the exclusion of others. There is much evidence in favour of this view. The mere possession by the Tetractinellida of tetraxonid spicules is no safe guide. Imperfectly developed " grapnel spicules," which in a more highly developed condition are so characteristic of Tetractinellid sponges, are now known to occur in the Suberitidas {Fi'oteleia) , and polyaxonid megasclera are occasionally met with in Desmacidonidae and Axinellidse (Acarnus and Thrinacophora). Again, the test whether a sponge is corticate or not breaks down utterly in this case, for the Clavulina, like some of the Tetractinellida, are nearly all corticate and have mostly a radially disposed skeleton. We even find a cortex, associated with a radially disposed skeleton, in one genus of Desmacidonidse (Phelloderma). It is quite certain that there exists no sharp line of division between the Monaxonida and the Tetractinellida, for the Suberitidse, Spirastrellidge and Tethyadae supply us with abundant connecting links. That either the Tetractinellida have been derived from the Monaxonida, or vice versd, is now a generally accepted fact, but which is the parent group is a matter of much controversy, and there is much to be said on both sides. We have already^ advanced strong reasons for supposing the Tetractinellida to be derived from the Monaxonida, whilst Vosmaer upholds the contrary hypothesis. The time has as yet scarcely arrived when a satisfactory discussion of the question is practicable. We must wait for more evidence, and the evidence chiefly to be desired is of an embryological character. In any case it appears to us that the Clavulina and Tetractinellida are at any rate as intimately connected with one another as are the Clavulina and Halichondrina. That the Keratosa are most closely connected with the Halichondrina is also now a generally accepted fact, which finds its expression in recent classifications. They are probably Halichondrine sponges, which, living in warm seas, have developed a large amount of spongin, and suff"ered a correspondingly great reduction in the proportion of spicules present in the skeleton. We can trace this development of spongin through all intermediate stages ; through the Eenierinse to the Chalininae, and thence to aspiculous forms. But this is not the only path by which the same results may have been arrived at. The Challenger dredgings teach us that a horny skeleton may be developed in the ' Ann. and Mag. Nat. Hist., ser. 5, vol. sviii. p. 152, et seq. EEPOET ON THE MONAXONIDA. Ixi Homorrliaphidse (Chalininag), in the Heterorrbaphiclse {Gelliodes, Toxochalina), in the Desmacidonidse {Esperella, &c.), and in the Axinellidse {Axinella fhrosa) , hand in hand with a corresponding reduction in the siliceous element. We thus know four distinct paths along which the Keratosa may have developed, and the group is thus shown to be probably of polyphyletic origin, and, consequently, unnatural.' This fact probably accounts for the singular difficulty which Polejaeff found in classifying the group, and appears to us to be a much more probable explanation of this difficulty than the assumption, for we can caU it little else, that " the whole group is nothing more than a simile family."'^ This method of cutting the Gordian knot, simple as it is, is hardly satisfactory. Having come to the conclusion that the Monaxonida do not constitute a very natural order, although the two suborders therein included are probably natural enough, we ought perhaps to attempt some justification of our conduct in retaining the name at all. The real fault lay in the original distribution of the Challenger collections, and this could not be avoided, for it is only since this distribution was made, and since two of the Eeports on Sponges have been published, while others have been far advanced, that the great mass of facts necessitating the recent modifications in classification has been brought to light. At one time it was believed that the Monaxonida were a natural group, and by the time that the error was discovered the work was so far advanced that it was impossible to efi'ect a redistribution. Hence we were left with two suborders, the Halichondrina and the Clavulina, and for these we have been forced to retain the name 1 Marshall's Plioriospongise, which have given rise to so much discussion, are to be similarly explained as having originated polyphyletically from the Halichondrina. Marshall gives the following diagnosis of the genus : — " Kieselschwiirame mit sehlanken, einfachen Nadeln mit einer Spitze, Stecknadeln und Doppelhaken durchziehen imd umspinnen Sandmassen, sie zu Klumpen vereinigend; das Ganze ist mit einer ahziehbaren Haut bedeckt" {Zdtschr.f. wiss. ZooL, Bd. xxxv. pp. 122-126). The view proposed by him that they are siliceous sponges which penetrate and unite together masses of sand appears to us hardly to bear investigation ; the fact that the whole mass is enclosed in a definite, pore-bearing, dermal membrane, as Marshall himself describes, is opposed to this idea. Von Lendenfeld {Proc. Linn. Soc. N.S.W., vol. X. p. 81) advocates a very different hypothesis; he says " I do not hesitate to consider the Phoriospongiai as belonging to the horny sponge as well as those Porifera which, like Dysidea, possess an arenaceous skeleton but no flesh- spicules I consider the Phoriospongia3 not as boring sponges living in sand ; but as Ceraospongiaj belonging to the groupwith arenaceous irregular fibres,"and again (p. 84) — "I believe that the flesh-spicules in thePhoriospongiajand horny sponges on the one hand, and those of the silicifilired sponges on the other have been produced independently of each other." It seems to us that von Lendenfeld also has here placed a wrong interpretation upon the facts before him. It is quite unnecessary to assign such a polyphyletic origin to the microsclera ("flesh-spicules ") in question, and we regard the Phoriospongise not as forming a separate genus at all, but as derived from several distinct genera of Monaxonida, in which, probably owing to the influence of similar external conditions, the proper siliceous skeleton has been replaced to a greater or less extent by sand and other foreign bodies. The fact that when proper megasclera occur in these sand sponges they are small and slender, and to all appearance degraded forms, argues in favour of our view. It is well known that sponges have a strong tendency to take in foreign bodies of all kinds with which to build up a skeleton. In the horny sponges (e.g., Euspongia, Dysidea) this very frequently occurs, and also in the Heterorrhaphidffi (Tedania commixta, nobis), the Desmacidonidje {Esperella parasitica, Carter, Esperella arenicola, nobis, lophon omnivorus, nobis), the Suberitidre {Polymastia agglntinans, nobis), and the Spirastrellidae {Spirastrella solida, nobis). In some cases these sand sponges have sufiicient spicules remaining to enable us to say from what genus they have been derived, while in other cases this is no longer possible. 2 Zool. Chall. Exp., part sxxi.. Report on the Keratosa, p. 81. Ixii THE VOYAGE OF H.M.S. CHALLENGEE. Monaxonida, altliougli we now feel that the group thereby designated has no place in a natural classification. The Halichondrina are, taken as a whole, strongly contrasted with the Clavulina. Members of the former group very rarely indeed possess a true, fibrous cortex, while those of the latter are rarely without one. In the former group again the skeleton is usually more or less reticulate, in the latter group it is usually radiately disposed, with no secondary, crossing fibres. Thirdly, the Halichondrina have always, or nearly always, a greater or less amount of spongin in the skeleton, which is absent in the Clavulina. All these are important distinguishing characters which, taken in connection with the difier- ences in spiculation, are quite sufficient to separate the two suborders. Yet here, as in other cases, the line of division is not absolute, for we have, in the Axinellidse, transitional forms. Hence we have placed the Axinellidje next to the Suberitidse and at the end of the Halichondrina. In the Axinellidse the skeleton shows a strong tendency to radiate arrangement, and the spicules are generally monactinal, as in the Clavulina ; moreover, the microsclera, when such are present, appear to be sometimes of a distinctly Clavulinid type ; viz., stellate. As yet we know no Axinellid sponge with a true fibrous cortex, but, on the other hand, there are species referred to the genus Suherites, in which the presence of a fibrous cortex has not yet been demonstrated, although they possess the characteristic tylostylote spicules. The two genera Suherites and Axinella appear to be nearly related. This connection may be a less intimate one than we at present think, but this can only be decided when our now very imperfect knowledge of these sponges has been greatly augmented. This is the most apparent point of contact between the Halichondrina and the Clavulina, unless Schmidt's description of his Sceptrella regcdis^ should prove to be correct, and then we should be forced to imagine a connection between the Spirastrellidas and the Esperellinse. Sceptrella regalis is described as j^ossessing discastra (spicules characteristic of the genus Latrunculia) and, at the same time, chelae of a very peculiar type. It is placed by its founder amongst the Desmacidines. There is, presumably, only a single specimen of this sponge. It is described as " eine graue, diinne Kruste," and no account is given of the megasclera. Having regard to the extremely unlikely combination of discastra with chelae, and to the fact that the sponge is known only by a thin crust, and bearing in mind also the frequent occurrence in sponges of spicules belonging to other species, we are strongly inclined to believe that Schmidt had before him a youn» Spirastrellid sponge growing over the remains of some species belonging to the Esperel- linae, and that the discastra belong to one sponge and the chelae to the other.^ ^ Spong. Atlant. Gebiet., p. 58 ; cf. also p. 234 of the present work. 2 This view of the case is very strongly supported by our examination of a preparation of the spicules of Sceptrella regalis (laljelled in Schmidt's handwriting) in the British Museum. We find that the discastra and megasclera are typically those of a Latrunculia. The latter are smooth, sharp-pointed styli, agreeing with those of other species of Latrunculia even down to the characteristic slight crookedness of the shaft {cf. PI. XLV. figs. 8, 9, 10), and measuring EEPORT ON THE MONAXONIDA. Ixiii The Halichondrina are divided by us into four families, the Homorrhaphidse.the Heteror- rhaphidae, the Desmacidonidas, and the Axinellidse. To these von Lendenfeld adds a fifth famUy, the SpongUUdaj/ but with the fresh-water sponges we have in this place nothing to do, and we cannot afford the time and space to discuss their systematic position. The Homorrhaphidse are, obviously, a fairly satisfactory group. All have a reticulate skeleton, all have diactinal megasclera, and all are without microsclera of any kind. The group includes two subfamilies, the Renierinse and the Chalininse, each of which has long been recognised, but hitherto these two subfamilies have been far too widely separated, and that merely on the ground that in the Chalinin^e there is a greater development of spongin than in the Renierinfe. Nearly all Eenierine sponges have more or less spongin in the skeleton, and, as a matter of fact, it is impossible to say where the Renierinse end and the Chalininse begin. The Chalinina3 appear to be derived from Renierinae (probably polyphyletically, i.e., from several distinct genera), which, living in warm seas, have developed a horny skeleton more or less at the expense of the spicular element. The Chalininge proper, as we have already noted, are simulated by genera belonging to quite distinct groups, which, living under similar conditions, have suffered a simdar change in their skeletons. We have thus in the family Heterorrhaphidse the two genera Toxochalina and Gelliodes, each with a distinct, well developed horny skeleton exactly similar to that of true Chalininas, and also with diactinal megasclera, but each betraying its true position by the presence of microsclera. By some authors (e.g., von Lendenfeld) the former genus is included amongst the Chalininse, but we cannot agree to such an arrangement.^ The mere possession of a horny skeleton is not sufficient guide to the systematic position of a sponge, and this fact cannot be too strongly enforced. The Chalininse are very poorly represented in the Challenger collection, and this is accounted for by the fact that they are essentially shallow- water forms, and are rarely if ever met with in deep-sea dredgings. Moreover, their range is rather a restricted one. On the other hand, in the large collection of sponges obtained in Australian seas by Dr. R. von Lendenfeld, and now lodged in the British Museum, the species of Chalininse are extremely plentiful and varied. Dr. von Lendenfeld's descriptions of these sponges are now in course of publication, and we hope that when they appear much will have been done towards putting the group on a more satisfactory footing with regard to classification. about 0-44 by 0-011 mm. The chel.ie, on the other hand, are characteristically those of the genus Myxilla. Many, or most of them, it is true, possess four teeth, but this condition is known to occur in at least one species of the latter genus {Myxilla mariana, nobis ; cf. p. 138, footnote). There are a fair number of them in the preparation, and they are all, or nearly all, equal ended and not, as stated by Schmidt, generally unequal ended. In short, Schmidt's figures and description of these spicules are misleading in the extreme ; there can scarcely be a doubt that they are present as foreign spicules, and such an occurrence is not at all uncommon. 1 Proc. Zool. Soc. Land., December 21, 1886, p. 584. 2 Mr. Carter carries this line of argument a step further, and actually includes a species of Homccodictya, characterised by its remarkable chelate microsclera, in the genus Chalina, presumably because of the amount of spongin present in the skeleton {Ann. and Mag. Nat. Hist., ser. 5, voL x. p. 111). Ixiv THE VOYAGE OF H.M.S. CHALLENGER. The Heterorrhapliidae constitute, perhaps, the least natural of our four families. In many respects they occupy an intermediate position between the Homorrhaphidae and the Desmacidonidge. As their name implies, they are possessed of different kinds of spicules, as opposed to the Homorrhaphidas, in which we never find more than one kind. Micros- clera are present in the group, but these are never of the chelate type. The five sub- families, viz., Phloeodictyinse, Gelliinse, Tedaniinse, Desmacellinse and Hamacanthinse, into which the Heterorrhaphidse are divided, are not so intimately related inter se as are either the two subfamilies of the Homorrhaphidae or of the Desmacidouidae. The Phloeodictyinse form a compact and very remarkable group. Both in external form and in anatomical characters they appear to be strongly contrasted with other subfamilies of the Heterorrhaphidse. The first known specimen of the group was dredged by the Rev. A. M. Norman off the Shetland Islands in 1861. It was evidently only a fragment, and was named by Bowerbank Isodictya rohusta} In 1864 Mr. J. G. Jeffreys dredged another specimen, also off the Shetland Islands, which was much mutilated by the dredge, and was named by Bowerbank Desmacidon Jeffrey sii.^ Later discoveries showed that these two fragments belonged to one and the same species, of which the Rev. A. M. Norman was the first to describe an entire specimen. He established the genus Oceanapia for this interesting form.^ In 1870 Schmidt founded his genus Rhizochalina* the type species being Rhizo- clialina oleracea. Although in the present Report we keep these two genera distinct, yet we have very strong reasons for believing that they ought to be united, and then the genus Oceanapia would be the sole representative of the subfamily. Finally, in 1882, Carter established ^ his " Group " " Phlceodictyina," and included therein nine species, all partaking more or less of the characters of the original Oceanapia rohusta. Unfor- tunately the material at our disposal is not in very good condition for anatomical investigation, but an examination of thin, stained sections has led us to believe that future researches, under more favourable circumstances, will lead to very interesting results. Concerning the Gelliinse we have very little to say. The subfamily is characterised by the extreme simplicity of the spiculation, which in the case of the genus Gellius is identical with that of the genus Oceanajna. This identity might be thought to indicate a close relationship between the two subfamilies. Possibly it does indicate some connection, but we do not think that this can be a very intimate one, for, although the Gelliinse present us with great diversity in external form, yet we know no instances of anything like an approach to the highly characteristic form of the Phlceodictyinge ; in other words the sponge is never divisible into body and fistulas, and this is a very important 1 Mon. Brit. Spong., vol. ii. p. 304. ^ Mon. Brit. Spong., vol. ii. p. 347. 3 Brit. Assoc. Report for 1868, p. 334. * Spong. Atlant. Gebiet., p. 35. ' Ann. and. Mag. Nat. Hist., ser. 5, vol. x. p. 117. REPORT ON THE MONAXONIDA. Ixv distinction between the two. Moreover, the sigmata, which occur in both subfamilies, are such a common and wide sjaread type of microsclera that their presence cannot be safely relied upon in estimating probable relationships. From their extreme simplicity it appears quite possible that they may have originated independently in several groups.' We distinguish three genera in this subfamily, and we have already pointed out the interesting fact, that here, as in the Homorrhaphidae, there is a very strong tendency towards the development of horny fibre, which asserts itself as soon as the necessary surroundings (viz., moderately warm and shallow seas) are attained. The Tedaniinas, which we have placed next in the series, are also a very isolated group. The spiculation is, so far as the megasclera are concerned, identical with that of the genus Myxilla amongst the Desmacidonidse, even down to the relative positions of the two forms of spicules in the sponge. A very wide distinction lies, however, in the fact that the Tedaniinas have no chelas, but, in place thereof, simple rhaphides, which are unliuown in the genus Myxilla. The differentiation of the megasclera into " main " and "dermal" forms, differing fundamently from one another in shape, must be con- sidered as an indication of a high degree of specialisation and alone quite sufficient to necessitate the removal of the Tedaniinse from amongst the Renierinse, where they have been placed by previous authors. It seems not unlikely that here, as in the Phlceo- dictyinse, the only two described genera will have to be merged into one ; for the distinguishing character {i.e., the spination of the styli in Trachytedania) can hardly be considered of generic importance {cf. the similar spination of the styli in some species of the genera lophon and Myxilla). The Desmaeellinse are not a very satisfactory subfamily, for as yet we know hardly anything about them. There is only a single established genus, viz., Schmidt's Desmacella, and they differ from the Gelliinse only in the replacement of the diactinal megasclera by monaetinal forms ; it is very possible that these two subfamilies should be united, though we have found it convenient for the present to keep them distinct. We come now to the last subfamily of the Heterorrhaphidse, viz., the Hamacanthinge. In this subfamily we meet with microsclera of a very remarkable and entirely unique form, the diancistra. These spicules seem to be peculiarly constant in shape in the different species. That they have some connection with the ordinary sigmata of other sponges is very probable, but they are always distinguished from these by the presence of a sharp, cutting inner edge with three notches, one in the centre of the shaft and one at each angle where the shaft bends round to form the terminal hooks. The nearest approach which we have found to this spicule in any other genus is the large sigma of Esperella simonis (PI. XV. fig. 13), which differs, however, in the absence of the characters just mentioned, though resembling the diancistra in its unusually large size and in the shape 1 Cf. Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., 1884, p. 371. (ZOOL. CHALL. EXP. — PAET LIX. — 1887.) Nnii i Ixvi THE VOYAGE OF H.M.S. CHALLENGEE. of the hooks. So far as we know certainly, the diancistra do not occur amongst the Desmacidonidse. Schmidt, however, describes^ a sponge under the name Vomerula tibicen, which is stated by him to contain " Doppelankern " and " Pflugscharspangen ; " unfortunately there are no figures, and the entire description is comprised in the fol- lowing words : — " Vomerula tibicen. Neu — Vom Habitus des Desmacidon tunicatum, also incrustirend, mit Oberhaut, welche in ^ bis 1 Cmtr. lange Rohren iibergeht. Nach den Kieseltheilen ware diese Art ein Desmacidon zu nenuen, das ausser den characteristischen Korpern, nJimlich den etwas gebogenen Nadeln, den Doppelankern van 0'0257 Mmtr., und den etwa halb so langen Spangen, auch noch die grossen Pflugscharspangen von 0-157 Mmtr. enthiilt. " Es ist damit eine abermalige Combination und Uebergangsform verwirklicht. Ob wir es mit einer constanten Form oder mit einer individuellen Abschweifung von Vomerula zu Desmacidon oder umgekehrt zu thun haben, ist nicht zu entscheiden. — Fundort : Grenada, 170 Faden." The combination of spicules here indicated would be so unusual were the " Pflug- scharspangen " true diancistra, that we cannot help feeling some doubt on this point, and thinking that possibly they may be only large contort sigmata like those of E&perella simonis, nobis, and somewhat resembling the true diancistra ; or, supposing that they are true diancistra, then perhaps the chelae occur as foreign bodies, as in the analogous case of Sceptrella regalis {vide supra). Further information concerning this point is much needed ; if diancistra are really present, and also chelje, then the species in question forms a most important link between the subfamilies Hamacanthinse and Esperellinse ; it would, of course, be referable to the latter. The subfamily Hamacanthinse is divided into two genera, according to whether the megasclera are diactinal (as in Hamacantha) or monactinal (as in Vomei^la). The DesmacidonidfB, constituting our third family, are characterised by the presence of chelate microsclcra. These spicules are, unlike the sigmata, of such very complex and remarkable form that we cannot believe them to have originated poljrphyletically, that is, independently in several groups. Hence we are constrained to class together in one family all those species which possess them, and this arrangement will, we hope, be found to be a satisfactory one. It is but a slight, though an important modification of the arrange- ment followed by previous authors. Thus, the Desmacidonidse of Vosmaer include, on the one hand, more than our family, in that he admits most of our Heterorrhaphidae, i.e., species without chelse, while, on the other hand, they include less, in that he keeps separate the Ectyoninte, which do possess chelae. This is an arrangement of which we cannot at all see the justification, more especially with regard to the separation of the " Ectyonidse," and in the present work we consider the old " Ectyonidfe " as a subfamily of the Desmacidonidse under the name Ectyoninse. The presence in the Ectyoninse of 1 Spong. Meerb. von Mex. pt. ii. p. 83. REPORT ON THE MONAXONIDA. Ixvii echinating spicules is, as we shall proceed to show, quite insufficient to justify such a wide distinction between the two groups as Vosmaer admits. We divide the Desmacidonidae into two subfamilies — (1) the Esperellinse, in which the skeleton fibre is not echinated by projecting spicules, and (2) the Ectyonin^e, in which echinating spicules are present. These two subfamilies are, however, by, no means sharply marked off from one another, for we actually have, in the genus Myxilla (placed by Vosmaer as well as by ourselves amongst the Desmacidonidse), both species with and species without the echinating spicules, and these different species are so closely allied that we have found it impossible to separate them generically. This question is again referred to in the systematic portion of the work, under the genus Myxilla, to which we refer the reader for further details. Here then lies the justification, and we think it will prove to be a sufficient one, of the present arrangement. It is probable, nay certain, that the first division of the family, the Esperellinae, will require subdivision, and in our opinion it will fall into several very natural groups, but we cannot, in the present work, make such an arrangement, as we have not the necessary time at our disposal for a complete revision of the group. The Ectyoninae are a very difficult group to deal with ; most of them are evidently sponges which are developing a strong horny skeleton at the expense of the spicular element, and the poor degree of development exhibited by the spicules makes their classification unusually difficult. Possibly, as held by Mr. Carter, they have some connection with the Axinellidae, but we cannot believe that it is nearly so close a one as he seems to indicate in uniting the Ectyoninae and Axinellidse in one group by them- selves, the " Echinonemata." To our minds their intimate connection with the Esperellinas far outweighs in importance any relationship which may exist in the Axinellid direction, while the presence of a strongly developed horny skeleton is to be connected here, as in the Chalininse, with the fact that they flourish almost exclusively in warm areas. We consider the Axinellidse as constituting a family by themselves, which we place at the end of the Halichondrina, and hence after the Ectyoninae and near to the Suberitidae, with which they seem to have striking affinities. These affinities have already been discussed in speaking of the relations of the Halichondrina to the Clavulina, so we shall not dwell upon them in this place. The family is a very difficult one to deal with, and our knowledge of it is in a most unsatisfactory condition. Its most striking positive character, and that in which it most nearly approaches the Ectyoninae, is the arrangement of the spicules to form the skeleton-fibre ; but though there is a strong superficial resem- blance between the two groups in this respect, yet the arrangement is in reality very distinct in the two cases, for in the Axinellidse all the spicules have their apices projecting very obliquely outwards and forwards from the centre of the fibre, while in the Ectyoninas there is a central core of longitudinally placed spicules cemented together by homy Ixviii THE VOYAGE OF H.M.S. CHALLENGER. matter, the fibre thus constituted being echiuated by special spicules which project from its surface more or less at right angles, or the spicular core may be absent, leaving only the horny fibre and the echinating spicules. Yet here again, we have at least one species, viz., Myxilla frondosa, nobis, which seems to indicate a transition between these two types of skeleton arrangement. In the universal absence of chelae, however, the Axinellidse are sharply defined from the Ectyoninae and other Desma- cidonidae. We come now to the second of our suborders, viz., the Clavulina. We have already indicated the probable relations of this group to the Halichondrina. The four most prominent features of the group (none of which, however, will probably be found to be quite constant throughout) arc (1) the presence of radiately arranged tylostylote spicules ; (2) the cork-like, granular ground substance ; (3) the presence of a distinct fibrous cortex ; and (4) the absence of spongin from the skeleton. As we have already hinted, we have no doubt whatever that this group is closely related, through the Tethyadse, to the so-called Tetractinellida. We shall not further discuss this question, but refer the reader to Vosmaer's latest remarks on the subject.^ According to the presence or absence of microsclera we divide the Clavulina into two families ; (l) Suberitidse, in which there are no microsclera, and (2) Spirastrellidie, in which microsclera are present. This appears to us to be a much more natural arrangement than that proposed by Dr. Vosmaer,' viz., according to the presence or absence of mammiform projections on the surface of the sponge.^ This difference in ideas is perhaps partly accounted for by the fact that Vosmaer has omitted from his scheme the two very important genera Spirastrella and Latrunculia, both of which are well repre- sented in the Challenger collection. The Clioniadse, of which group the members, save that they ajjpear to be mostly Clavulina, have only the boring habit in common, will have to be distributed chiefly between the Suberitid^ and Spirastrellidse, as there is no reason at all for supposing that species of both these families (and of others also) may not have independently acquired the boring habit ; and the spiculation of the difierent boring sponges seems to indicate that this is the case. 1 Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 474 (translated by Dendy, Ann. Mid Mag. Nat. Hist., ser. 5, vol. xix. p. 249). 2 Bronn's Klass. u. Ordnung. d. Tliierrichs, Porifera, p. 328, et seq. 3 Vosmaer also gives in the diagnosis (loc. cit.) of his Polyniastida; " Faserrinde meist sehr deutlich," and of his Suberitidje "eine echte Faserrinde fehlt immer;" hut, inasmuch as he includes the genera Siiherites and Stijlocordyla in his Suberitidae, we must emphatically deny the correctness of the latter statement ((/. our own researches on the minute anatomy of these genera, supra, and PI. L.). DESCRIPTION OF GENERA AND SPECIES. Order MONAXONIDA. 1878. Monactinellidx, Zittel, Zur Stammegcschiclite der Spongien,i p. 9. 1883. Monaxonidx, SoUas, Cassell's Natural History, vol. vi. p. 326. Siliceous sponges with uniaxial megasclera. Suborder I. HALICHONDRINA. 1885. Halichondrina, Vosmaer, Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 335. Typically non-corticate ; skeleton usually reticulate ; megasclera usuaUy either oxea or styli. Family I. Homorrhaphid^.^ 1886. HomorrhapMdx, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 326. Megasclera all diactinal, either oxea or strongyla ; no microsclera. Subfamily 1. RENiERiNiE. 1870. Renierinse, pars, Schmidt, Spoug. Atlant. Gebiet., p. 39. 1886. Renierina, Ridley and Deudy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviiL p. 326. The spicules may be united together by a small proportion of spongin, but are never completely enveloped in it. Genus Halichondria, Fleming (Pis. I., II., XLVI.). 1828. Halichondria, Fleming, History of British Animals, p. 520. 1862. Reniera, pars, Schmidt, Spong. Adiiat. Meeres, p. 72, &c. 1870. Amorphina, Schmidt, Spong. Atlant. Gebiet., p. 40. Skeleton confused, may be fibrous but never regularly reticulate. Spicules oxea or strongyla, usually long and slender. Spongin scarcely appreciable. 1 Festgabe der pbilosopMschen Facultat zum 50 jahrigen DoctorjubUaum des Professor von Siebold, Munchen. 2 From Gr. o>oV, one and the same ; and pct0k> needle. (ZOOL. CHALL. EXP. PART LIX. 1887.) } 2 THE VOYAGE OF H.M.S. CHALLENGER. In the simple form and arrangement of the spicules this is the most primitive of all genera of siliceous sponges. It is characteristic of littoral regions in most or all parts of the world, but never seems to occur in deep water. Schmidt was the first to really restrict the genus within reasonable limits/ but Vosmaer ^ has shown that he was in error in overlooking the earlier name Halichondria. The original type of the genus is the well known and cosmopolitan species Halichondria panicea. Halichondria 2oanicea, Johnston (PL II. figs. 2, 3). 1842. Halichondria panicea, Johnston, ^ British Sponges, p. 114, pi. x. and pL xi., fig. 5. 1870. AmorpJiina panicea, Schmidt, Spong. Atlant. Gebiet., p. 77. 1881. Amorplmia megalorhaphis, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. vii. p. 368. 1884. Amor2Jhina megalorrhaphis, Eidley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 416. 1885. Amorphina megalirrhaphis, Eidley, Narr. Chall. Exp., vol. i. pt. ii. p. 571. Vosmaer * has already given a good abstract of the history of this old and well- known species, to which we must refer the reader. A good description of the species is to be found in Johnston's British sponges [he. cit.), and stiU further details are given by Bowerbank.^ Carter^ has already described a specimen of Halichondria panicea (var.) from Kerguelen, which was obtained by the Transit of Venus expedition. Later {loc. cit. sup.) he describes his Amorphina megalorhajMs, from the Basse Eocks, Ceylon, and in so doing makes the following observations : — " This seems to be a variety of the common British species Halichondria panicea, chiefly difi'erentiated by the size of its largest spicules, which is double that of the English one. The spicules also of the specimens brought home by the Rev. A. E. Eaton from Kerguelen's Island, and others dredged up by H.M.S. "Porcupine" in the Atlantic Ocean, are much larger than those of the common British species ; so that this variation may extend even to our own shores, while the single form, great variety in size, and long attenuation towards the end of the spicule generally characterise the species everywhere." We have now to record a series of specimens '' from Kerguelen, which to our mind completely establish the identity of the British Halichondria p)anicea with Mr. Carter's Amorphina megalorhaj)his. While agreeing remarkably well in external form these specimens (of which there are seven or eight of fair size, in addition to smaller pieces) exhibit a great range in the 1 Amorphina, loc. cit. 2 Bronn's Klass. u. Ordnung. d. TMerreichs, Porifera, p. 337. ^ For probable synonyms of older writers, vide Johnston, loc. cit. * Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 336. ' Mon. Brit. Spong., vol. ii. p. 229. « Phil. Trans. Roy. Soc, vol. ckviil p. 286. ' One of these specimens was figured in the preliminary account, Narr. ChaU. Exp., vol. i. pt. ii. p. 571, fig. 188. REPORT ON THE MONAXONIDA. 3 size of the spicules. The following table shows the length of the largest spicules in various specimens. 1. A specimen from Ilfracombe, in the Bowerbank collection, Brit. Mus.; average maximum length about 0"35 mm. 2. Mr. Carter's specimen from Kerguelen ; average maximum length about 0*5 mm. 3. One of the Challenger specimens from Kerguelen ; average maximum length about 0'6 mm. 4. Another of the Challenger specimens from Kerguelen ; average maximum length about 077 mm. 5. Amorphina megalorhaphis (as given by Carter, loc. cit.), "varying in leno-th from l-128th to l-23rd inch" ( = about from 0-2 to 1-0 mm.). Under these circumstances we consider ourselves justified in including aU the above mentioned specimens under the name " Halichondria j^anicea." Some of the Challenger specimens from Kerguelen are infested by a very abundant Oscillatorian Alga ; especially one which is coloured of a dark chocolate brown, but whether the colour is due to the Alga or not is an open question ; several sponges from Kerguelen were similarly coloured. There is also a single small specimen, noteworthy on account of its dark coloration at and near the surface, from a pool at Oosima, Japan ; this agrees in every essential respect with the common British form ; the oscula are distinct and rather large, the dermal skeleton beautifully reticulate, and the spicules up to about 0'5 by 0-0126 mm. in size. The species thus appears to be cosmopolitan. Localities. — "Royal Sound, Kerguelen, 20-60 fathoms;" "Kerguelen, 10-100 fathoms;" "Christmas Harbour, Kerguelen, 60 fathoms;" Oosima, Japan, 14th March 1875, "Pool," one specimen. Habitat. — British Seas (Johnston, Bowerbank, &c.); Basse Rocks, ofi' south-east coast of Ceylon (Carter); Kerguelen (Carter, Challenger); Torres Strait (Ridley, "Alert"); Atlantic (Carter); Japan (Challenger). Halichondria caduca, Bowerbank (PI. II. fig. 15). 1866, &c. Halichondria caduca, Bowerbank, Mon. Brit. Spong., vol. ii. p. 234, and vol. iii. pi. xU. figs. 9-11. We identify with this species a single small, cavernous specimen from off" the south- west coast of Patagonia (Station uncertain). It agrees well with Bowerbank's species in external form, colour, rugged surface and probably also in its cavernous structure ; also in the size and form of the spicules (PI. II. fig. 15), which measure about 0'28 by O'Ol mm. 4 • THE VOYAGE OF H.M.S. CHALLENGER. in both, and in the arrangement of the fibre, though this is rather more definite in the main skeleton in Bowerbank's sponge. Locality. — Off the south-west coast of Patagonia. One specimen. Habitat. — British seas (Bowerbank); off south-west coast of Patagonia (Challenger). Halichondria solida, Eidley and Dendy (PI. II. figs. 5, 5a). 1886. Halichondria solida, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 326. Sponge massive, sessile, encrusting, growing over a mass of nuUipore. The single specimen in the collection is about 62 mm. long by 37 mm. broad and 31 mm. thick. Colour in spirit greyish -yellow.^ Texture extremely compact ; firm, but slightly compressible. Surface uneven, rather rough to the touch ; may be minutely hispid in places. Dermal membrane difficult to make out, inseparable from the underlying tissues. Oscula doubtful ; few if any ; possibly the sponge is Hpostomous. Skeleton. — There is no special dermal reticulation distinguishable as such from the main skeleton. The main skeleton consists of a densely felted mass of long oxeote spicules, densely packed, but without order and without discernible fibres. Hence there is no great difi"erence in appearance between a surface section and a section taken at right angles to the surface. Sjncules. — More or less curved, large, smooth, fusiform oxea (PI. II. figs. 5, 5a), almost invariably rounded off instead of jDointed at each end (thus becoming strongylote), and with the end often rather wrinkled or roughened in appearance. Size extremely variable, both as regards length and thickness. The largest spicules are by no means the thickest ; length up to about 1"1 mm., thickness up to about 0*038 mm.; the degree of curvature also varies very much. This species may be distinguished by its very dense and compact structure, the absence of any skeleton fibres, and the rounding off of the ends of the spicules. If it were not for the absence of a definite skeletal fibre, it would be rather referable to Petrosia, especially considering the tendency of its spicules to become blunted (c/l Petrosia truncata, infra). Locality. — Reefs, Tahiti ; September 28, 1875 ; depth, 30 to 70 fathoms. One specimen. Halichondria solida, var. rugosa. We propose the above name for a single, large, massive, sub-columnar, sub-lobose specimen of a dark chocolate-brown colour. It differs from the type of the species (1) in its dark colour; (2) in the nature of the surface, which is roughened by little ' Tinged with violet owing to the presence of a darkly-coloured horny sponge on the same mass. REPORT ON THE MONAXONIDA. 5 prominences resembling miniature cliains of peaked mountains, over and between which it is glabrous ; (3) in having the skeleton less compact (this does not affect the tough, compact character of the sponge, which in this case is not dependent on the skeleton arrangement) ; (4) in having the ends of the spicules more commonly pointed, but still roughened and irregular and very different from the long drawn out, sharply and evenly pointed ends in Halichondria 2)cinicea. The specimen in question is 106 mm. in height by about 37 mm. in average diameter. It bears several distinct, circular oscula, about 3 mm. in diameter. Locality. — Off Api, New Hebrides, 60 to 70 fathoms. One specimen. Halichondria 2)ellicidata, Ridley and Dendy (PI. I. figs. 1, la ; PI. II. fig. 9). 1886. Halichondria pelliculata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 326. The single specimen in the collection (PL I. figs. 1, la) is erect, lobose, increasing gradually in width from below upwards, and marked by a series of transverse grooves and swellings into a number of segments, each of which probably represents a stage in the growth of the sponge ; the top is flattened, and the sponge is just beginning to branch into two lobes. In the middle of the flattened top there is a large compound osculum, and a similar one occupies the end of the incipient branch (PI. I. fig. 1, o, o). The height of the specimen is 62 mm., and the diameter at the top not quite 25 mm. Colour in spirit yellow. Texture, internally soft and friable, but the surface is hard and chitinous. Surface corrugated as above described, but smooth and glabrous. A thin, hard, chitinous membrane covers the entire surface of the sponge, and appears to take the place of the dermal membrane. Whether this is only a post-mortem condi- tion or not, we are unable to say. Just below the surface are great numbers of round or oval, highly granular bodies, about 0*3 mm. in diameter ; these may be the gland-ceUs which secrete the chitinous envelope;-^ they occur in less numbers in the deeper parts of the sponge. Oscula (PI. I. fig. 1, o, o) ; at the summit of the sponge is a single large round opening, about 6 mm. in diameter, subdivided by a number of vertical partitions which separate the diff'erent exhalent canals from one another ; a similar, but smaller osculum occurs at the top of the incipient branch. Skeleton. — (a) Dermal; a very abundant reticulation of irregularly scattered, horizontally placed, large oxeote spicules, (h) Main ; so far as we have been able to ascertain this is rather sparse, not very regular, and with few distinct fibres, being com- posed of large scattered oxeote spicules. It is, however, very difficult to obtain a satis- factory vertical section, owing to the fragility of the internal tissues. In parts at any rate the skeleton is rectangular in its arrangement, the fibres containing few spicules. 1 For this very probable suggestion we are indebted to Dr. E. von Lendenfeld. 6 THE VOYAGE OF H.M.S. CHALLENGER. Spicules. — Large, stout, fusiform oxea (PI. II. fig. 9), curved, and gradually and fairly sharply pointed ; size about 0'45 by 0*028 mm. This species is at once distinguishable from all others of the genus by its external form and its characteristic chitinous envelope (hence the specific name). The size and shape of the spicules are also very characteristic. In the presence of an external chitinous membrane it resembles Rhizochalina Jistulosa. Like the preceding species, it has a great general resemblance to Petrosia, but, on the other hand, the skeleton arrangement shows some approach to the rectangular character of Reniera; but as the dermis is subreticulate, and the spicule is a large tapering oxeote, it seemed best to regard it as a decidedly aberrant member of the genus Halichondria. Locality. — Amboina ; depth, 100 fathoms. One specimen. Halichondria latrv,nculioides, Ridley and Dendy (PI. I. figs. 5, 5a ; PI. II. fig. 1 ; PI. XLVI. fig. 5). 1886. Halichondria latruncuUoides, Eidley and Dendy, Ann. and Mag. Nat. Hist., .ser. 5, vol. xviii. p. 326. Sponge erect, lobose ; two out of the three specimens collected are pear-shaped, and have apparently been attached by the narrow end to the substratum on which they grew. The third specimen, which is the largest, is lobose and compressed in a vertical plane ; it has a much constricted base, by which it has evidently been attached. The largest specimen measures 44 mm. in height, by 72 mm. in greatest breadth, and about 12 mm. in average thickness ; the larger of the two pear-shaped specimens measures 50 mm. in height by 27 mm. in diameter near the top. Colour in spirit light grey. Texture rather soft and spongy internally. Surface uneven, deeply corrugated, but glabrous ; with numerous large, round or oval, flat, sieve-like pore-areas, elevated above the general surface of the sponge, and constituting by far the most important character of the species (PI. I. fig. 5, p.a., and fig. 5a). Dermal membrane forming (except in the pore- areas), together with its supj^orting layer of spicules, a thickish, parchment-like crust, readily separable from the underlying tissues. In the pore-areas, on the other hand, the dermal membrane is very thin, delicate, and transparent, and reduced to a mere sieve by the very numerous pores. Oscida at the summits of conical projections of the parchment-like dermal membrane at the top of the sponge (PL I. fig. 5, o). Pores almost entirely confined to the pore-areas (PI. I. fig. 5a) ; we have, however, detected a few in gaps in the almost continuous dermal skeleton. Skeleton. — (a) Dermal ; a very dense, in most parts perfectly continuous layer of large oxeote spicules laid horizontally side by side (PL XLVI. fig. 5). {h) Main; loose, composed of rather irregularly arranged spiculo-fibre ; the fibre itself is often fairly stout, but not very definite or compact. REPORT ON THE MONAXONIDA. 7 Spicules. — Large, straight, or very slightly curved, fusiform oxea (PL II. fig. 1), sharply and usually gradually pointed. Size commonly about 0'7 by 0'022 mm., but often much larger, measuring about 1'25 by 0"031 mm. In these larger spicules one end is usually markedly larger than the other, and the points are irregular. This is a very interesting species, and forms another example of the extraordinary diversity and richness of the sponge-fauna at Station 320. We have called the species " latrxinculioides" owing to the curious resemblance which it bears in external form to species of the genus Latrunculia, of which two were obtained from the same Station. As we have already had occasion to remark, the most noteworthy character of the species is the arrangement of the pores in definite, raised pore-areas. We at first thought that this feature would prove of generic importance, but concluded finally that it is not so, being apparently only a case of adaptation, which, however, at once separates the species from all others of the genus. The arrangement of the pores is, as usual, closely correlated with the arrangement of the dermal skeleton. The dermal skeleton in this sponge is so closely packed together that there is literally no space left for the pores {vide PI. XLVI. fig. 5), hence these are almost entirely confined to special areas set apart on purpose for them (PI. I. figs. 5, 5a). It is very interesting to notice how precisely the same adaptation in the arrangement of the pores has taken place in a totally diS'erent and widely separated sponge, viz., Latruncidia cqnccdis (vide infra); here also the dermal skeleton, although of quite a difi"erent kind, is so dense as to allow no room for the pores, which are confined to special raised areas. It is further interesting to notice how these two sponges consequently resemble one another in external form, and that both were obtained from the same station. Locality. — Station 320, February 14, 1876 ; lat. 37° 17' S., long. 53° 52' W. ; off' the mouth of the Rio de la Plata; depth, 600 fathoms; bottom, green sand; bottom temperature, 37°"2. Three specimens. " Halichondria sp. (a). Under this head we must mention two amorphous masses of sponge, apparently belonging to the genus Halichondria, each encrusting a delicately branched white Gorgonia from Station 320. The specimens do not present sufficient characters to make a specific determination desirable. They are very soft, spongy, and cavernous, and the skeleton is confused and lax in the extreme. Spicules, smooth, slightly curved oxea, rather abruptly pointed at each end ; size, very fairly constant, about 0-35 by 0-017 mm. Locality. — Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; off the mouth of the Rio de la Plata; depth, 600 fathoms; bottom, green sand; bottom temperature, 37°'2. Two specimens. THE VOYAGE OF H.M.S. CHALLENGEE. Halichondria sp. (6). We include under this head two fragmentary, massive, amorphous specimens and a few small pieces from off Marion Island, all in a very bad state of preservation, to which we are not prepared to assign any specific name. They are certainly not worthy of the erection of a new species for their reception, and we know of no old one in which they might be included. They are characterised chiefly by a coarse, brittle, dermal layer, readUy shaling off, composed of the dermal membrane and a more or less close reticula- tion of oxeote spicules supporting it. Texture brittle, cavernous. Main skeleton, an irregular loose reticulation of oxeote spicules. Spicules, smooth, slightly curved oxea, rather abruptly pointed; size, about 0*42 by 0'017 mm. Locality. — Off Marion Island, December 26, 1873; depth, 50 to 75 fathoms. Fragments. Halichondria (?) sp. (c), (PI. II. fig. 4). Sponge apparently free, lobose. The single specimen consists of three short, stout lobes, united together at their bases. All the lobes are hoUow and rather thin-waUed, the large central cavities all communicatiag with one another.* The specimen is 37 mm. in greatest diameter (from apex to apex of two lobes), and about 12 mm. thick in the middle. Colour in spirit pale greyish -yellow. Texture rather brittle. Surface uneven, but smooth and slightly glabrous, with a minutely reticulate appearance. Dermal membrane delicate and transparent, but with a chitinous look. Oscula (?); there are several smaU round apertures which lead right through from the exterior to the internal cavities, and one which seems to communicate only with a canal system within the thickness of the wall. Skeleton. — (a) Dermal; an irreg-ular but very abundant reticulation of closely packed, long oxeote spicules, (h) Main; very dense, composed of closely packed, long oxea, not very regular, but stout fibres may be distinguished running towards the surface, and others crossing them; separate spicules again are irregularly strewn throughout, so that the whole thickness of the sponge-wall consists of little else but spicules. Spicules. — Very long, curved (often slightly crooked), fusiform oxea (PI. II. fig. 4), gradually and finely pointed. Size variable, measuring up to about 0'87 by 0'022 mm. Considering that there is only a single specimen of this sponge, which may be abnormal or injured, we have thought it better to describe it without creating a new 1 In the blind end of one of the lobes were found two small Crustaceans, but it is hardly likely that these can have caused the very extensive excavation of the interior, though the possibUity of their being concerned in it must be borne in mind. REPORT ON THE MONAXONIDA. 9 species for its reception. In spiculation it comes very close to Halichondria panicea, while its general form suggests Rhizochalina. Locality. — Station 203, October 31, 1874; lat. 11° 6' K, long. 123° 9' E.; Philippine Islands; depth, 12 to 20 fathoms; bottom, mud. One specimen. Genus Petrosia, Vosmaer (Pis. II., III.). 1862. Reniera (?), Sckmidt, Spong. Adriat. Meer., p. 76. 1863. Schmidtia, Balsamo-Crivelli, Atti Soc. Ital. Sci. Nat., vol. v. p. 293. 1864. Thalysias, pars, Duchassaing and Michelotti, Spongiaires de la mer Caraibe, &c. 1885. Petrosia, Vosmaer, Bronn's Klass. u. Ordnung. d. Thierreiclis, Porifera, p. 338. Sponge usually hard or even stony ; generally with numerous, well-defined, large oscula. Skeleton more or less confused ; spicules oxeote to strongylote, usually short and thick, packed close together in tracts. The most obvious feature of this genus is its hard, often stony texture. Vosmaer has changed the older name Schmidtia into Petrosia, because the former was found to be already occupied. Petrosia similis, Ridle)^ and Dendy (PI. II. fig. 10 ; PL III. figs. 3, 4). 1879. Tlialysias suhtriangularis. Carter, Phil. Trans., vol. clxviii. p. 287. 1886. Petrosia similis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol xviii. p. 327. Sponge (PI. III. figs. 3, 4) variable in external form, repent, branching ; branches sometimes anastomosing, subangular or cylindrical, with large circular oscula arranged in series on the upper surface ; or erect and lobate with large circular oscula at the summits of the lobes. We have here two very distinct types of external form in the same species. The repent specimens (PI. III. fig. 3) are very much broken ; the best measures 93 mm. in length by about 12 mm. in diameter, and bears nine large circular oscula arranged in a single row along its upper surface. There is only one lobate specimen (PI. III. fig. 4), which in external form bears some resemblance to Clathria elegantula, nobis (^cf. PL XXVIII. fig. 3), and measures 81 mm. in height by 87 in width. Colour in spirit yellowish-grey. Texture (of the lobate specimen) firm and almost incompressible, fibrous ; (of the repent specimens) much more yielding to the touch, easUy compressible, fibrous but brittle. Surface smooth but uneven and harsh to the touch ; often with a reticulate appearance owing to the reticulations of the skeleton fibre, which show through the dermal membrane. Denned membrane distinct, translucent, firmly adherent to the underlying tissues. Oscida large, circular, numerous ; sometimes more than 6 mm. in diameter, like great circular pits in the surface. Pores, round oj)enings in the dermal membrane, very few seen. (zooL. CHALL. EXP. — PAET Lix. — 1887.) Nnn 2 10 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton. — There is no special dermal skeleton, but spicules project in irregular, sparse tufts from the uppermost portion of the main skeleton, and serve to support the dermal membrane, commonly projecting slightly l)eyond it ; these tufts are nothing but the projecting ends of the primary skeleton fibres. The main skeleton (in the lobate specimen) is a very well-developed reticulation of very stout, compact spiculo-fibre, in which there is a well-marked distinction between primary and secondary fibres crossing one another at right angles and running vertically to, and parallel with, the surface of the sponge. The rectangular meshes between the fibres vary much in size, usually they are large. In the repent specimens, the skeleton is not so well developed, being laxer and more confused. Spicules. — Slightly curved oxea (PI. II. fig. 10), fairly sharply and fairly gradually pointed, size about 0-225 by 0-016 mm. The above description is taken from a series of specimens, aU from Station 142, which must be considered as the types. There is also in the collection a single small piece from Station 150, without oscula and j)robably young, which we identify with the species ; it diflFers from the types in having a very compact skeleton with very indistinct fibre, and in having the spicules rather larger, measuring about 0-3 by 0-016 mm. The specimen obtained from Kerguelen by the Transit of Venus Expedition, and described by Mr. Carter (loc. cit. supra), must also, having regard to the locality and measurements of the spicules, be referred to this species rather than to Tlmlysias suhtriangularis. Here again the skeleton reticulation is closer, but the fibre not so well developed as in the types of Petrosia similis ; the spicules average about 0-19 by 0-0126 mm. in size. The synonymy of the diflPerent species of Petrosia is very difficult to unravel. Undoubtedly this species comes near to Tlialysias subtriangularis, Duchassaing ^ ; the difficulty is to find out exactly what that species is. Even supposiug that Schmidt had not examined Duchassaing's type specimens, yet, having regard to the localities, it is probable that his identification of his Schmidtia aulopora'^ with Duchassaing's Thalysias suhtriangularis is correct. Thalysias repens, Carter,^ also from the West Indies, which Mr. Carter caUs " a repent form of the white species sithtriaiigularis, viz., Tlmlysias repens, mihi," appears from the measurement of his figure of the spicule (which is thus shown to be about 0-18 mm. long.) to be the same species as Schmidt's aulopora. In Schmidtia aulopora, Schmidt, the spicules measure about 0-175 by 0-0078 mm., being thus shorter and not half so thick as in the types of Petrosia similis.* 1 Animaux radiaires des AntiUes, p. 26. See also Duchassaing and Michelotti, Spongiaii-es de la mer Caraibe, Haarlem, 1864, p. 85, pi. xvii. fig. 1, pi. xviii. fig. 1. ^ Spong. Atlant. Gebiet., 1870, p. 44, Taf. v. fig. 8. 3 Ann. and Mag. Nat. Hist, April 1882, p. 282, pi. xi. fig. 10. * The measurements of the spicules in Schmidtia aulopora are taken from a preparation in the British Museum, labelled in Schmidt's handwriting, from St. Thomas. REPORT ON THE MONAXONIDA. 11 No doubt the Kerguelen specimen forms a connecting link, but we think it advisable to distinguish between two closely allied species, Petrosia subtriangularis and Petrosia similis, the former characteristic of West Indian seas, and the latter of the seas south of the Cape. Two weU-marked varieties of the latter are described below, one of which occurs as far north as the Philippine Islands. Localities. — Station 142, December 18, 1873 ; lat. 35° 4' S., long. 18° 37' E.; south of the Cape of Good Hope; detph, 150 fathoms; bottom, green sand; bottom tem- perature, 47°'0. Several specimens. Station 150, February 2, 1874 ; lat. 52° 4' S., long. 71° 22' E. ; between Kerguelen and Heard Island; depth, 150 fathoms; bottom, coarse gravel; bottom temperature, 35°'2. One small piece. Habitat. — South of the Cape of Good Hope (Challenger) ; Kerguelen (Carter) ; between Kerguelen and Heard Island (Challenger). Petrosia similis, var. massa (PI. II. fig. 11; PL HI. fig. 6). Sponge (PI. III. fig. 6) massively lobose, or consisting of very stout cylindrical branches. The largest specimen is a flattened lobose mass, which, judging from the fact that it bears oscula on both surfaces, has probably grown erect; it is 162 mm. high by about 50 in average breadth (narrowing from above downwards) and 18 mm. thick. Colour in spirit greyish-yellow. Texture rather soft and spongy. Surface smooth but uneven. Dermal membrane distinct, in parts with a lace-like reticulate appearance owing to the supporting skeleton reticulation. Oscula large, circular ; irregulai-ly scattered over the surface (yet with some tendency to form series) ; about 4 mm. in diameter; commonly surrounded by a slightly prominent rim. Pores, rounded openings in the dermal membrane as usual. This variety differs from the types in its more slender spicules, in the looser skeleton arrangement and less compact and well-developed fibres (hence its soft, spongy character), and in the more irregular arrangement of the oscula. The spicules (PI. II. fig. 11) measure up to about 0-28 by 0-012 mm. The dermal membrane is not suj^ported on projecting tufts of spicules, but rests dii-ectly upon a reticulation of spiculo-fibre, which is, however, only the uppermost layer of the main skeleton. The character of the dermal skeleton in th different species of Petrosia seems to be of very little use in separating them. Locality.— Station 314, January 21, 1876; lat. 51° 35' S., long. 65° 39' W. ; between the Strait of Magellan and the Falkland Islands ; depth, 70 fathoms ; bottom, sand ; bottom temperature, 46°'0. Several specimens. 12 THE VOYAGE OF H.M.S. CHALLENGER. Petrosia similis, var. compacta (PI. II. fig. 13 ; PI. III. fig. 5). Sponge (PI. III. fig. 5) cylindrical, ramose, erect (?), or repent (?). TTie single specimen in the collection consists of a stout, irregulai'ly cylindrical basal (?) portion, which bifurcates at one extremity into two but slightly divergent, more slender branches, about equal in length to the original piece. Total length 143 mm. Diameter of unbranched portion about 18 mm., and of branches about 10 mm. Colour in spirit light brownish-yellow. Texture stony hard but brittle. Surface uneven, smooth in appearance, but harsh to the touch. Dermal membrane rather coarse, translucent, allowing the subjacent skeleton reticulation to show through. Oscula large, conspicuous, circular, with their margins level (or almost so) with the general surface of the sponge, diameter about 4 mm. Pores scattered through the dermal membrane. Skeleton. — The main skeleton is a very dense and compact, but rather irregular reticulation of spiculo-fibre and spicules, in which primary fibres are readily distinguishable, running vertically to the surface. Seen from the surface, the uppermost layer appears as a reticulation of stout spiculo-fibre with rounded meshes about 0-35 mm. in diameter, on which the dermal membrane rests directly. Spicules. — Short, rather stout, slightly curved, fairly gradually and fairly sharply pointed oxea (PI II. fig. 13), measuring about 0-22 by 0-0145 mm. This sponge is not unlike Schmidtia variabilis, var., from the same station, in external form, but may at once be distinguished by the much smaller spicules and denser texture. It is distinguished from the type of the species by its much more compact structure, and by the absence of the tufts of spicules supporting the dermal membrane, but agrees weU in the size of the spicules. In its stony texture it resembles Petrosia dura, Nardo, but differs from typical specimens of that species in its external form, which is ramose, instead of tuberous or tuberously ramose. The skeleton-fibre also is by no means so compact and well developed, and the meshes of the main skeleton are much smaller. The spicules, again, are smaller, and do not exhibit that remarkable variability in size which characterises those of Petrosia dm-a.^ Locality. — Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E. ; Philippine Islands; depth, 18 fathoms; bottom, blue mud. One specimen. Petrosia truncata, Ridley and Dendy (PI. II. fig. 14 ; PI. III. fig. 1). 1886. Petrosia truncata, Ridley and Dendy, Ann. and Mag. jSTat. Hist., ser. 5, vol. xviii. p. 327. Sponge (PI. III. fig. 1) massive, sessile. The single specimen in the collection is unfortunately only a fragment. It consists of a massive basal portion about 25 mm. 1 Our knowledge of Petrosia dura, enabling us to make the above comparisons, is derived chiefly from specimens in the British Museum, obtained from Professor Schmidt. For descriptions of the species, vide Nardo, his, 1833, p. 522 ; Schmidt, Spong. Adriat. Meeres, p. 76, Taf. vii. fig. 13 ; Balsamo-CriveUi, Atti Soc. Ital. Sci. Nat., vol. v. p. 293, Taf. y\ figs. 1, 2, 3. REPORT ON THE MONAXONIDA. 13 square and 18 mm. thick, from the upper surface of which arises abruptly a siugle, stout, thick-walled oscular tube with a large circular osculum at the summit. Height of oscular tube about 25 mm., diameter 14 mm., thickness of wall 4 mm. Colour in spirit greyish-yellow. Texture hard and stony, but rather brittle ; internally rather cavernous. Surface smooth but uneven, and harsh to the touch. Dermal membrane distinct, but firmly adherent to the underlying tissues, translucent, very granular, frequently reduced to a mere network by the enormous number of pores. Oscula, one only present, a little over 4 mm. in diameter, at the summit of a large tubular pro- jection. Pores, very numerous rounded openings about 0'07 mm. in diameter. Skeleton. — There is nothing that can be properly termed a distinct dermal skeleton, but the dermal membrane is supported by a reticulation of stout spiculo-fibre, which is not to be distinguished from the main skeleton. The main skeleton consists of a reticulation of very well developed, stout spiculo-fibre in which one distinguishes primary fibres running vertically to the surface and secondary ones crossing them more or less at right angles. This distinction into primary and secondary fibres is not so evident in the deeper parts of the sponge as near the surface, the skeleton arrangement becoming here a more or less confused reticulation of stout spiculo-fibre. The fibre is compact, and averages about 0"1 mm thick. Sjncules. — Short, stout, smooth, slightly curved strongyla (PL 11. fig. 14), evenly rounded ofi" at each end, size about 017 by 0'0094 mm. This species comes near to Schmidtia aulopora, but is distinguished from that species by the strongylote instead of oxeote spicules, by the stouter skeleton fibre, and possibly also by the arrangement of the oscula. If it should prove that the oscula in our species are always situated on the ends of prominent oscular tubes, this would be a very important character. This Petrosia approaches the most closely of all the Chal- lenger species to that named Reniera crateriformis by Mr. Carter, characterised by a cup- or bowl-shaped form and sausage-shaped spicules. Locality. — Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E. ; Philippine Islands; depth, 18 fathoms ; bottom, blue mud. One piece. Petrosia variabilis, Ridley, sp., var. (PI. II. fig. 12). 1884. Schmidtia variabilis, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 41.5, pi. xxxix. fig. N, pL xlL fig. t. With this species we identify as a variety a single specimen from the Philippine Islands. It is irregularly lobate in form, and, though hard in texture, is very brittle and cavernous ; the surface is uneven and harsh to the touch, and has also a minutely reti- culate appearance. This variety differs from the types of the species in its pale greyish- 14 THE VOYAGE OF H.M.S. CHALLENGER. yellow colour/ and in having slightly larger and decidedly stouter oxeote spciules (PL II. fig. 12), which measure about 0'45 by 0-022 as against 0-4 by 0 '01 9 mm. There do not, however, appear to us to be sufficient differences to merit a distinct varietal name. Locality. — Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E. ; Philippine Islands ; depth, 18 fathoms ; bottom, blue mud. One specimen. Habitat. — Port Darwin, North Australia (Ridley, " Alert ") ; Philippine Islands (Challenger). Petrosia hispida, Ridley and Dendy (PI. II. fig. 16 ; PL III. fig. 2). 1886. Petrosia Idspida, Eidley and Dendy, Ann. and Mag. Nat. ilist., ser. 5, vol. xviii. p. 327. Sponge (PI. III. fig. 2) massive, sessile, subglobular or lobate, narrowing towards the base, bearing on the upper surface numerous smaU mammiform processes, each with a single round osculum at the summit. The larger of the two specimens measures 62 mm. in height by the same in breadth. Colour in spirit yellowish-grey. Texture firm and fairly compact, rather brittle. Surface uneven, minutely hispid. Dermal membrane difficult to make out, owing to the hispidity of the surface and the absence of a special dermal skeleton reticulation. Oscula very conspicuous, numerous small round openings, each on the summit of a volcano-like projection, remarkably constant in size, diameter about 1'5 mm. Pores, rounded openings scattered through the dermal membrane. Skeleton. — No dermal reticulation. The main skeleton consists of a reticulation of spiculo-fibre, the primary lines being fairly distinct (sometimes five or six spicules broad), and running vertically to the surface, near which they are more distinct than elsewhere, and having their terminal spicules projecting for a short distance beyond the surface, thus causing the characteristic minute hispidity. The secondary lines are very confused and almost obliterated by numerous scattered spicules. Sincules. — Smooth, slightly curved oxea (PL II. fig. 16), not very sharply but rather gradually pointed; size about 0-37 by 0'021 mm. This species is distinguished by its beautiful and very characteristic external form and by its hispid surface ; the spicules are rather large for the genus. Locality. — Royal Sound, Kerguelen, January 17, 1874 ; depth, 25 fathoms. Two specimens. Genus Reniera, Nardo (Pis. I., II.). 1847. Reniera, Nardo, Prospetto della fauna marina volgare del veneto estuario, Venice. 1864. Isodidya, pars, Bowerbank, Mon. Brit. Spong., vol. i. p. 197, &c. Skeleton composed of definite, rectangular (sometimes triangular or polygonal), typically unispicular meshes. Spicules short oxea or strongyla, usually united together at the ends only by spongin. ' Perhaps due to a different mode of preservation. REPOET ON THE MONAXONIDA. 15 This genus was first defined within its present limits hj Schmidt.^ Here, as in the case of other Eenierinfe, it is impossible to give an absolute diagnosis separating it from other genera, but no one who has studied the group will fail to recognise the fact that it is a very natural genus. By the development of more spongin uniting the spicules this genus appears to have given rise to those Chalinine sponges which possess a definite rectangular skeleton, the fibre of which is cored by few and small oxeote spicules, e.g., Chalina (s. str.). We cannot agree with Schmidt in referring those species which have a separable dermal membrane (" zusammenhiingende Oberhaut") to a distinct genus, Pellina, as we do not regard this character as being of generic importance. His original t3rpe of Pellina is Reniera semituhulosa, Schmidt. The so-called dermal membrane is also a very distinct feature of Halichondria panicea, yet Schmidt keeps this species out of his genus Pellina, into which it ought certainly to fall according to his definition, if such it can be called.^ Reniera cinerea. Grant, sp. 1827. Spongia cinerea, Grant, Edin. New Phil Journ., vol. ii. p. 204. 1828. Halichondria cinerea, Fleming, History of British Animals, Edinburgh, p. 521. 1866. Isodietya cinerea, Bowerhank, Mon. Brit. Spong., vol. ii. p. 274 {vide also vol. iii. pi. xlviii.). We identify with this species a single small, massive specimen from Station 209, measuring only about 12 mm. in diameter, and of a pale yellow colour. The spicules measure about 0"14 by 0"008 mm. In connection with this specimen the remarkably high temperature (71°'0) of the water in which it lived is very noteworthy. XocaZiVy.— Station 209, January 22, 1875 ; lat. 10° 14' N., long. 123° 54' E.; Philippine Islands; depth, 95 fathoms; bottom, blue mud; bottom temperature, 7l°"0. One specimen. Habitat. — British Seas (Grant, Bowerbank, &c.) ; Philippine Islands (Challenger). Reniera implexa, Schmidt, var. (PL I. fig. 4). 1868. Reniera implexa, Schmidt, Spong. d. Kiiste v. Algier, p. 27. This species is well represented in the collection by a series of specimens from the Azores. These, though identical in external form (PI. I. fig. 4) with Schmidt's species, as shown by comparison with specimens in the British Museum from Trieste, purchased from Professor Schmidt, difier slightly from it in skeletal characters. Thus the oxeote spicules are a little longer and more slender in the Challenger variety, measuring about 0'16 by 0'0063 mm. as against about 0'138 by 0-007 mm. This difierence is but very 1 Spong. Atlant. Gebiet, p. 39. 2 Spong. Atlant. Gebiet., p. 41. 16 THE VOYAGE OF H.M.S. CHALLENGER. slight ; a more noteworthy one lies in the arrangement of the spicules, for in our variety the primary fibres are well marked oif from the secondary ones, and contain more than one row of spicules, usually about two or three. We cannot, therefore, say of our sponge, as Schmidt does of his types of the species, " Die doppelspitzigen Nadeln bUden das charakteristische einreihige Netzwerk der achten Renieren." There is not, however, by any means sufficient diflference to justify the erection of a new species. As no one has ever figured this interesting form, we take the present opportunity of doing so ; no one besides Schmidt seems hitherto to have met with it. Locality. — Station 75, Ji;ly 2, 1873 ; lat. 38° 38' N., long. 28° 28' 30" W.; ofi" the Azores ; depth, 450 fathoms ; bottom, volcanic mud. About a dozen specimens. Habitat. — Adriatic Sea (Schmidt) ; off the Azores (Challenger). Reniera aqu^dvctus, Schmidt, var. infundihularis, nov. (PI. I. fig. 2 ; PI. II. fig. 8). 1862. Reniera aquxdudus, Sclimidt, Spong. Adriat. Meeres, p. 73, pi. vii. figs. 6, 6a, 66. 1884. Reniera sp., Eidley, Zool. Coll. H.M.S. " Alert," Brit. Mus., p. 410. Sponge (PI. I. fig. 2) consisting of a thin lamella, folded so as to form irregular funnels and tubes which may anastomose. The largest specimen is a complex, hollow, thin-walled mass growing together with a Gorgonia ; the shape of the whole is very irregular, and its greatest length is about 100 mm. The wall of the sponge is only about 2 mm. thick. Colour in spirit pale yeUow. Texture very delicate, spongy and fragile. Sm-face uneven but smooth. Dermal memhixine very thin and transparent, allowing the round or oval subdermal cavities to show through, and thus giving to the surface a beautiful reticulate appearance. Skeleton. — Typically Renierid in arrangement, i.e., forming a rectangular, unispicular reticulation. Spicules. — Small, smooth, very slightly curved oxea (PI. II. fig. 8), rather slender and sharply pointed at each end; size about 0"17 by 0'008 mm. This variety differs from the types of the species mainly in external form ; forming funnel-shaped or only irregularly tubular masses instead of regular cylindrical tubes. Th& -namQ infundihularis w^^ suggested by Ridley ^ in 1884 for some fragments of a Reniera obtained by the " Alert " in Torres Strait, which may probably be included under this species, and which also differ from the types in not forming definite sym- metrical tubes ; this sponge agrees so closely in spiculation and external form (though, from the fact that fragments only were obtained, we cannot say certainly that it was infundibular) with the Challenger variety, that it seems desirable to include both under the same name. 1 Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 411, 1884. REPORT ON THE MONAXONIDA, 17 Locality. — Station 307, January 4, 1876 ; lat. 49° 24' 30" S., long. 74° 23' 30" W. ; off the south-west coast of Patagonia; depth, 140 fathoms; bottom, blue mud. One specimen. Off the south-west coast of Patagonia (Station uncertain). Two or three specimens. Habitat. — Adriatic Sea (Schmidt); Port Darwin, Australia (Pddley, "Alert"); Torres Strait (Ridley, " Alert ") ; off the south-west coast of Patagonia (Challenger). Reniera suhglohosa, Eidley and Dendy (PL I. figs. 3, 3a ; PI. II. fig. 6). 1886. Reniera suhglohosa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 327. Sponge (PI. I. fig. 3) sessile, subglobular, hollow, thick-walled, with a wide, circular, cloacal opening at the summit. Height 31 mm. ; diameter about 25 mm. ; diameter of cloacal opening 12 mm. ; thickness of wall 8 mm. Colour in spirit yellowish -grey. Texture firm but very brittle ; cavernous, traversed by canals which end beneath the dermal membrane on the outside of the hollow cup, their terminations being here irregularly dispersed. On the inside of the cup (PL I. fig. 3a) the openings of the canals (? oscula) are arranged in regular series, separated from one another by ridges as in the figure ; these ridges, or strands of tissue, form a rect- angular lattice-work. Surface even, and probably smooth in the living sjjonge. Dermal membrane delicate, transparent ; largely rubbed off ; where it still remains it extends right over the openings of the wide canals on the outer surface of the sponge ; sometimes also it appears to extend over the inner openings ; the oscula were very likely confined to the inside of the cup as in other flabeUate and infundiljuliform sponges. Pores found in the dermal membrane on the outside of the cup in the form of numerous, irregularly scattered, small, rounded openings variable in si^e, probably confined to the outside of the cup. Skeleton. — The main skeleton consists of a confused but subrectangular reticulation composed of very loose spiculo-fibre, two or three spicules in width. There is also a definite imispicular dermal reticulation supporting the dermal membrane. Spicules. — Smooth, slightly curved, subhastately and sharply-pointed oxea (PI. II, fig. 6), measuring about 0*3 by 0*013 mm. This species may be recognised by its fine, sul^giobular, hoUow form ; by the striking serial arrangement of the openings on the inner surface, and by its firm though brittle consistence. These characters, combined with the size of the spicules, are sufficient to distinguish it from all previously known species. Of possibly related species, Reniera compacta, Schmidt,-^ differs from Reniera subglobosa in its slender and longer spicules, and from Reniera dura in its bast-like cortex and spicules, which are twice as broad as here ; they resemble it in their density of structure, but are possibly referable to Petrosia. 1 Spong. Adriat. Meer., Supp. i. p. 38. (zooL. CHALL. EXP. — PART Lix. — 1887.) Nnn 3 18 THE VOYAGE OF H.M.S. CHALLENGER. Locality. — Station 307, January 4, 1876 ; lat. 49° 24' 30" S., long. 74° 23' 30" W. ; off the south-west coast of Patagonia ; depth, 140 fathoms ; bottom, blue mud. One specimen. Reniera tufa,^ Eidley and Dendy (PL I. figs. 6, 6a ; PL IL fig. 7). 18S6. Reniera tufa, Eidley and Dendy, Ann. and Mag. Nat. Hist., sor 5, vol. xviii. jx 328. Sponge (PL L fig. G) massive, sessile, cake-like ; represented in the collection by two p)ieces, each measuring about 62 mm. long by 43 broad, and up to 20 mm. thick. Colour in spirit greyish-yellow. Texture firm, almost stony, l^ut somewhat brittle, the sponge being traversed by wide canals {vide PL I. fig. 6a). Surface smooth where the dermal membrane is intact, but uneven. Dermal membrane distinct, rather coarse, readily peeling off, strengthened by a close reticulation of spicules. Oscula rather small, circular, with their margins flush with the general surface of the sponge, and each surrounded by a membranous diaphragm which greatly reduces the diameter of the exhalent canals at the surface. Pores, as usual, rounded openings scattered through the dermal membrane. Skeleton. — A compact, but rather irregular, almost unispicular reticulation of oxeote spicules, with triangular meshes ; no special dermal skeleton. Spicules. — Smooth, very slightly curved, subhastately-pointed oxea (PL IL fig. 7), measuring about 0*2 by O'Ol mm. This species seems to come near to Schmidt's Cribrochalina cretacea,^ but differs in the size of the spicules and in the arrangement of the oscula, the latter point being the most remarkable character in Schmidt's species. Keller's Reniera litoralis ^ also resembles our species very strikingly, especially in external form ; but here again we find differences in the arrangement of the oscula and the proportions of the spicules. In Reniera litoralis the spicules of the main skeleton are described as being nearly twice the length of those of the dermal skeleton and much more slender, a feature which does not obtain in Reniera tvfa. Locality. — Porto Praya, St. lago, Cape Verdes, 100 to 128 fathoms. Two pieces. Subfamily 2. Chalinin^. 1868. CJialinex, Schmidt, Spong. d. Kiiste v. Algier, p. 7. 1886. Chalinina, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 328. A considerable amount of spongin is present, typically forming a thick sheath completely enveloping the spicules and uniting them into strong fibres. 1 So-called from the resemblance to a piece of pumice-stone. ^ Spong. Atlant. Gebiet., p. 36. ^ Zeitschr. f. wiss. ZooL, Bil. xxx. p. 579. REPORT ON THE MONAXONIDA. 19 Genus Pachychalina, Schmidt (Pis. IV., V., VL, XLVL). 1868. Pachychalina, Schmidt, Spong. d. Kiiste v. Algier, p. 8. 1886. Dasijchalina, p.j]. Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 329. Lobose or digitate, .solid. Fibres stout, with spicules numerous and arranged polyserially. Schmidt's original diagnosis {loc. cit. supra) runs " Fasern aus vielen Reihen von Nadeln gebildet, welche durch Hornsubstanz fest verbunden sind." This diagnosis is not nearly exclusive enough and would include most species of Chalininte. Two j^ears later (in 1870) Schmidt describes the genus thus: " Nicht rohrige, knollige oder dickiistige Chalineen mit dichtem festerem Faserwerk, das in den jiiugeren Theilen nur eine oder wenige Nadebeihen, in den alteren und dichteren deren viele enthalt. Die Fasern werden von innen nacli aussen zu feiner, gehn nicht in Pinsel aus, sondern bUden ein zarteres Oberflachennetz, von welchem die Faserenden wie ein feiner Sammet abstehn. Zahlreiche Oscula erscheinen oft reo-elmassicr in Reihen eeordnet, oft ohne Ordnung gestellt und liegen entweder flach oder erheben sich mit kurzen Schornsteinen. Alle diese Merkmale und die Dichtigkeit des Gewebes variiren aber in dem Grade und zeigen in den von einander entferuteren Theilen eines und desselben Stockes solche Abstufungen, dass ich mir darauf hin nicht auf cctostante Species zu schliessen getraue." -^ As an example of the genus he here cites Spongia ruhens, Duch. et ]\Iich. However difficult Professor Schmidt may have found it to distinguish between different species, and indeed it would appear from the passage here quoted that he abandons the attempt, we ourselves have found it no easy matter to make anything of the genus. Thinking that it would be both practicable and convenient to distiDguish between two distinct groups of solid Chalinine sponges with stout fibres and many, polyserially arranged spicules, viz. (1) those with a smooth surface, and (2) those in which the surface is rugged and spinose, we have, in our preliminary report, proposed the name Dasychalina for the latter, and retained the name Pachychalina for the former. This we did under the impression that Spongia onibens was the first example mentioned of Schmidt's Pachychalina ; we now find that this is not the case, but that the type of the genus is Pachychalina rustica, Schmidt, which is described as follows : — " Die Gattungsdiagnose ist von Exemplaren entlehnt, von denen das eine drei centimeter hoch, papillenformig, mit etwas verdicktem oberen Ende und ganz unregelmassiger, fein krausiger Oberfliiche, das andere langlich Kegelformig, struppig ist. Die spitz-spitzen geraden oder etwas gebogenen Nadeln sind, 0"16 bis 0'175 Mmtr. lang." ^ Unfortunately, then, this species, the type of the genus, may have a spinose (struppig) surface, and so our Dasychalina falls to the ground. It might be possil^le '■ Spong. Atlant. Gebiet., p. 37. ^ Spong. d. Kiiste v. Algier, p. 8. 20 THE VOYAGE OF H.M.S. CHALLENGER. to make a genus for the other grouj) of species, viz., those with smooth surface, but experience has shown that it will be advisable to leave this question to some naturalist who has more opportunity for studying a large series of forms. Hence we include both groups under the name Pachychalina. Pachyclialina fragilis, Ridley and Dendy (PL IV. figs. 2, 2a). 1886. Dasychalina fragilis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 330. Sponge (PI. TV. figs. 2, 2a) irregularly ramose, subcylindrical, aculeated ; branches averaging about 1 8 mm. in diameter. Colour in spirit light greyish- or brownish -yellow. Texture hard and brittle. Surface very uneven, covered with sharp aculeations ; delicately striate, with the strias radiating from the summits of the aculeations.^ Oscula large and scattered, but almost entirely confined to one side of each branch ; about 4 mm. in diameter. Skeleton. — (a) Dermal ; an irregular reticulation of individual spicules and spiculo- fibre, backed up behind by a much coarser reticulation of much stouter spiculo-fibre, the nodes of which form the apices of the surface aculeations. (h) Main ; an irregular reticulation of very stout, compact spiculo-fibre and abundantly scattered spicules ; the fibres are about 0"35 mm. thick, and the main fibres terminate on the surface at the summits of the aculeations. There is no doubt a good deal of spongin present uniting the spicules, but it is difficult to make out, and does not form a distinct sheath around the fibres as in typical Chalininse. The fibre resembles that of Petrosia. Spicules. — Oxea, very large for a Chalinine sponge ; rather stout, slightly curved, suddenly and more or less bluntly pointed ; size about 0"42 by 0'02 mm. Taken by itself this species might have been described as a Petrosia, characterised by the aculeations on the surface, but it forms the first of a series of allied forms which seem to lead up from the Eenierinse to more typical species of Paehychalina and Spinosella. Locality. — Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E.; Philippine Islands; depth, 18 fathoms; bottom, blue mud. Several large pieces. Pachychalina melior, Ridley and Dendy (PI. IV. fig. 1). 1886. Dasychalina melior, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 330. Sponge (PL IV. fig. 1) irregularly branched; branches subcylindrical or angular, coarsely aculeated ; aculeations less abundant and weU marked than in Pachychalina fragilis. Average diameter of branches about 12 mm. Colour in spirit greyish- or brownish-yellow. Texture compressible, fibrous, but rather hard. Surface uneven, ' In one small specimen only this striation of the surface is not visible. REPORT ON THE MONAXONIDA. 21 subglabrous, with a faintly reticulate appearance, due to the subdermal cavities showing through the dermal membrane. Dermal membrane distinct, thin, transparent, perforated by numerous scattered pores. Oscula rather small and shallow ; chiefly on one side of each branch ; diameter little over 2 mm. Skeleton. — («) Dermal; a close, small -meshed reticulation of rather loose spiculo-fibre, echinated at the nodes by small bundles of outwardly projecting spicules, which are, at any rate in part, the projecting ends of the primary filjres. (h) Main; a rather irregular reticulation of spiculo-fibre and scattered spicules, in which primary and secondary fibres are distinguishable but not very well marked. Fibre much slenderer than in Pachychalina fragilis, but still without a very evident external sheath of spongin as in typical Chaliniaae. Spicules. — Eather slender, slightly curved, gradually and usually sharply pointed oxea; size about 0"175 by 0"0126 mm. This species, though very distinct from both, forms a connecting link between Pachychalina fragilis and Pachychalina^ Jihrosa, which is next described, and which has much slenderer spicules, and a greater amount of spongin in the fibre. Locality. — Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E. ; Philijjpine Islands ; depth, 18 fathoms ; bottom, blue mud. Two specimens. Pachychalina Jihrosa,, Ridley and Dendy (PL IV. figs. 3, 4). 1886. Dasyclialina fibrosa, Ridley and Dendy, Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 330. Sponge (PL IV. fig. 3) erect or rej)ent ; branched. Branches stout, irregular, coarsely aculeated ; bearing large circular oscula, usually on one side only. Diameter of branches ranging from about 6 to 25 mm. Length of spines varying greatly in different specimens, up to about 4 mm. Colour in spirit greyish-yeUow. Texture coarsely fibrous, compressible and elastic. Surface coarsely aculeated, reticulate. Dermal membrccne distinct, thin, translucent. Oscula large, cii'cular, shallow (i.e., the exhalent canals terminate at the bottoms of shallow depressions, vide figs. 3, 4). Skeleton. — (a) Dermal; a very coarse reticulation of stout spiculo-fibre with usually triangular meshes, broken up by a much finer and rather irregular reticulation of very slender spiculo-fibre. The coarse fibre contains a very great number of spicules, and comparatively little spongin ; measuring about O'l mm. in total thickness. Tlie finer fibre is very slender, contains a larger proportion of spongin, and usually only a single row of spicules; thickness about O'OIS mm. (&) Main; a reticulation of very stout, branching and anastomosing spiculo-fibre; measuring up to about 0*14 mm. thick, and almost or quite filled with densely packed spicules. Spicules occur also very abundantly outside the fibre ; irregularly scattered through the choanosome. 22 THE VOYAGE OF H.M.S. CHALLENGER. Spicules. — Small, slender oxea, very slightly curved ; abruptly and often bluntly pointed; size about O"! by 0-0032 mm. This species diflfers from Pachychallna {Cladochalina) diffusa, Ridley,^ in the inferior amount of spongin in the fibre, in the slenderer spicules (0-0032 as against 0-0063 mm. thick) and in the much larger oscula. Localities. — " Off Bahia, 7-20 fathoms"; several pieces. " Off Bahia" (inside label); •■' off Bermuda, shallow water " (outside label) ; several pieces. Station 208, January 17, 1875; lat. 11° 37' K, long. 12.3° 31' E.; Philippine Islands; depth, 18 fathoms; bottom, blue mud. One small specimen (PL IV. fig. 4) of a vai-iety, differing slightly in external appearance, but agreeing closely in micro- scopical structure, size of spicules, &c. Pachychalina lobata, Ridley (PL V. fig. 1 ; PL XLVI. fig. 4). 1884. Pachychalina lohata, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 404. The ChaUenger obtained one specimen of this sponge (PL V. fig. 1), which, being in much better condition than the single specimen obtained by the " Alert," and described by Ridley {he. cit.), enables us to add some interesting details. The specimen in question consists of a single, long, compressed, digitate process ; it has apparently grown erect and measures 200 mm. in length, 16 mm. in breadth, and a little over 8 mm. in thick- ness. There is an almost uniserial row of circular oscula down each margin {vide fig. 1), each osculum being about 2 mm. in diameter. The dermal membrane is delicate and transparent, supported l3y a very beautiful and regular reticulation of spiculo-fibre. The pores are very numerous, reducing the dermal membrane in the meshes of the dermal reticulation to a mere network (PL XLVI. fig. 4). This specimen differs from the type of the species ; (1) in the size of the spicules, which measure only about 0-16 by O'OllS mm. as against 0-21 by 0-0128 mm.; (2) in the possession of a more highly developed and regular dermal skeleton reticulation. That there is a similar dermal reticulation in the original type we have convinced ourselves by a re-examination of the specimen, but it is not so well developed as in the Challenger sponge ; and thus the original description is incorrect in so far as it is stated that the dermal skeleton is " a closely-set coat of subparallel spiculo-fibres." The specific name lohata was given by Ridley on the supposition that the species was identical with Esper's Sjwngia lohata,^ but we have now arrived at the conclusion that this identification is untenable; hence the "Alert" specimen must stand as the type of Pachychalina lohata. Locality. — Port Jackson ; 30 to 35 fathoms. One specimen. Ilahltat.—? oxt Darwin, Australia (Ridley, "Alert"); Port Jackson, Australia (Challenger). ' Ann. ;ui(1 Mag. Nat. Hist., ser. 5, vol. xiv. p. 183. ^Hie. Pflanzenthiere, vol. ii. p. 273, pi. xlvi. REPORT ON THE MONAXONIDA. 2S Pachyclialina megalorrhaphis, Ridley and Dendy (PI. V. fig. G). 1886. Pachyclialina megalorrhapliis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 328. Sponge (PL V. fig. 6) composed of rather slender, long, cylindrical branches, which may anastomose. Maximum diameter of branches in the single specimen nearly 1 2 mm. , but average diameter much less. Colour in spirit pale yellow. Texture compressible and elastic. Surface almost quite smooth, very minutely hispid. Dermal membrane thin, transparent, allowing the subdermal cavities to appear through it, and give to the surface a faintly mottled or reticulate appearance. Oscula rather small, subuniserially arranged along the branches. Skeleton. — (a) Dermal; not very distinct. A small-meshed reticulation of loose spiculo-fibre, echinated at the nodes Ijy small tufts of projecting spicules which are the ends of the primary fibres, (h) Main ; a subrectangular reticulation of spiculo-fibre and single spicules, with very distinct primary lines running vertically to the surface. The skeleton fibre is fairly strong and polyspiculous, but with no distinct sheath of spongirt surrounding it. The main lines are five or six sjjicules broad. Spicules. — Slightly curved, gradually sharp-pointed oxea; size about 0'245 by 0-016 mm. The spicules of this species are large for a Chalinine sponge, and the amount of spongin is correspondingly small, but the elastic, fibrous texture, the external form, and the arrangement of the skeleton leave no doubt as to its systematic position. Zoca%.— Station 163d, June 12, 1874 ; lat. 33° 57' 30" S., long. 151° 39' 15" E. ; off the coast of New South Wales ; depth, 120 fathoms ; bottom, green sand. One specimen. Pachychalijia elongata, Ridley and Dendy (PI. VI. fig. 1). 1886. Pachychalina elongata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 329. Sponge (PI. VI. fig. 1) digitate, ramose ; branches very long, varying in diameter ia different parts of their length, up to about 12 mm., which is, however, unusually wide, the average being about 8 mm. Greatest length about 300 mm. Branches sometimes anastomosing. Colour (?) (now dark reddish-brown, but this is probably due to packing in a metal case). Texture very compressible, elastic, tough and fibrous. Surface smooth, subglabrous. The dermal membrane, together with its supporting skeleton reticulation, forms a very tough, distinct skin, but very firmly adherent to the deeper portion of the sponge which it encloses. Oscula small, scattered, chiefly on one side of the sponge. Skeleton. — (a) Dermal ; a close-meshed reticulation of stout, compact, spiculo-fibre, echinated by projecting spicules; meshes small, only about O'll mm. wide, {h) Main; a rectangularly meshed reticulation of spiculo-fibre, very well developed, with a great 24 THE VOYAGE OF H.M.S. CHALLENGER. deal of spongin ; fibre about 0'07 mm. thick, polyspiculous, spicules not confined to tlie centre of tlae fibre. Sjncules. — Slightly curved oxea, fairly sharply and gradually pointed ; size about 0-1 mm. by 0-0065 mm. ioca%.— Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' 0" E. ; Bass Strait ; depth, 38 fathoms ; bottom, sand and shells. One specimen. Pachychalina (?) punctata, Eidley and Dendy (PL VI. figs. 2, 2a, 2h, 2c; PI. XLVI. figs. 1, 2). 1886. Paehijchalina (?) punctata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. .5, vol. xviii. p. 329. Sponge (PI. VI. fig. 2) erect, flattened, lobose. The single specimen in the collection •consists of a large flattened lobe, bearing near the summit two smaller ones. The large lobe may itself have been formed by the coalescence of smaller ones, as there are two large holes right through it. Height of sponge 143 mm. ; breadth 87 mm. ; thickness about 6 mm. Colour in spirit rather dark greyish-yellow. Texture very tough and leathery, compressible and elastic. Surface very uneven but glabrous. Oscula (PI. VI. figs. 2, 26, o), small round openings, scattered at about equal distances from one another on one side only of the sponge; diameter about 1 mm. Pores (J) (PI. VI. figs. 2, 2a, 2b, p, 2c, p, and PI. XLVI. fig. 1) unusually large ; visible to the naked eye as minute openings scattered singly but very abundantly on both sides of the sponge (PL VI. figs. 2, 2a) ; diameter up to about 0-35 mm., but varying, and usually smaller. They appear to be lined by spongin (?), which projects into their cavities in large bosses, frequently arranged so that the cavity has the form of a cross, the four angles between the arms being filled up by four projections of spongin (?) (PL XLVI. fig. 1). Possibly this arrangement is to guard against the ingress of noxious parasites, &e.^ Skeleton. — (a) Dermal ; a very well-developed, close-meshed reticulation of spiculo- fibre (PL XLVI. fig. 1), with plenty of spongin, and echinated abundantly by projecting sjaicules. Meshes only about 0-07 mm. wide, (b) Main; also very well developed, with strong primary fibres, about 0-07 mm. thick, radiating to the surface, and crossed by secondary fibres in a lattice-like manner (PL XLVI. fig. 2). Fibres polyspiculous, but with a thick sheath of spongin. Also numerous scattered spicules between the fibres. Spicules. — Slightly curved, sharply and rather gradually pointed oxea; size aliout 009 by 0-0055 mm. The species is very weU characterised, the most characteristic and interesting point being the special means adopted for guarding the inhalent apertures (if such they be). The amount of spongin present in the sponge is very large, whence its tough and leathery 1 Probably pores also occur as small openings through the thin dermal membrane, but we are not certain about this point. REPOET ON THE MONAXONIDA. 25 consistence. It is important to observe that though the oscula are confined to one surface of the frond, the pores occur on both. ioca^.— Station 1G2, April 2, 1874; kt. 39° 10' 30" S., long. 146° 37' 0" E. ; Bass Strait; depth, 38 fathoms; bottom, sand and shells. One specimen. Pachyclialina (?) pedunculata, Eidley and Dendy (PL V. fig. 5). 1886. Pachychalina (?) pedunculata, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 329. Sponge (PI. V. fig. 5) erect, stipitate, cylindrical ; tapering to a blunt point above, and with a short, rather stout stem attached to a worm tube. Total height about 127 mm., diameter of body a little over 18 mm. ; of stalk 6 mm. Colour in spirit greyish- yellow. Texture very soft and spongy, fairly elastic. Surface even, very minutely hispid ; readily fraying out into tags. Dermal membrane very thin and delicate. Oscula rather small, scattered. Skeleton. — Composed of loose fibres and scattered spicules, the main fibres only distinct, running vertically to the surface. No special dermal skeleton. Fibres very loose, main about six spicules broad. Only a small amount of spongin present. Sjncules. — Eather slender, slightly curved, gradually and sharply pointed oxea ; size about 0-5 by 0-017 mm. This species resembles in several respects Vosmaer's Pachychalina caulifera,^ but it is cylindrical instead of flattened, and the shape of the spicules is difi"erent, being slender instead of broadly fusiform. The fibres in Pachychalina caulifera appear to contain a good deal more spongin than in the present species ; indeed, it is only doubtfully that we include the latter in the genus at all; it forms another connecting link between the Renierinae and Chalininae, and shows how little value can be placed upon the amount of spongin present for purposes of classification. Locality. — Kerguelen Island, 10 to 100 fathoms. One specimen. Genus CJialina, Grant (Pis. V., XLVL). 1861. Chalina, Grant, Tabular View of the Animal Kingdom, p. 76. Form various, not tubular, surface smooth. Skeleton reticulation rectangular, with much spongin and few spicules ; fibres typically slender, with a single axial series of spicules. Concerning the name CJialina, we quote as follows from the " Alert " Report : — " This genus was merely mentioned by name in Grant's Tabular View of the Animal Kingdom 1 Eeport on the Sponges dredgedup in the Arctic Sea by the "WiUem Barents," in the years 1878 and 1879, p. 33, pi. i. fig. 14 ; pi. iii. figs. 64-66. (ZOOL. CHALL. EXP. — PAET. LIX. 1887.) ^^^ ^ 26 THE VOYAGE OF H.M.S. CHALLENGER. (1861). In 1864 Dr. Bowerbank^ (first) defined the genus correctly, assigning to it Spongia oculata, Pallas, as its type species. Schmidt therefore appears to me to be in the wrong when he (Spong. Atlant. Gebiet., p. 32) removes this species to his genus Chalinula, of 1868, and restricts Chalina to species which have the habitus of Euspongia and Cacospongia." ^ We have now come to the conclusion that the characters of the genus Chalina should be made to depend rather upon the structure and arrangement of the fibre than uj^on any mere growth form. Thus we should now include in the genus recent species, such as Chalina 7-ectangularis, nobis, which would formerly have come under the genus Acervochalina, and forms like Chalina pergamentacea, Ridley, which have hitherto been assigned to Cladochalina. Chalina p)almata, Lamarck, sp. (PI. V. fig. 4). 1813. Spongia i>almata, Lamarck, Arm. Miis. Hist. Nat. Paris, vol. xx. p. 453. Sponge (PL V. fig. 4) of erect habit ; much branched, palmo-digitate ; branches compressed. The single specimen in the collection measures 68 mm. in height by 100 in breadth. Colour in spirit pale j'ellow. Texture soft and spongy, tough and fibrous. Surface fairly even, very slightly hispid ; Avith a reticulate appearance due to the underlying structures appearing through the thin, transparent dermal membrane. The dermal membrane itself has also a very minutely reticulate appearance due to the dermal skeleton fibres. Oscula small, round, irregularly scattered, averaging a little over 1 mm. in diameter. Pores, rounded openings, varying considerably in size, scattered very profusely through the dermal membrane. Skeleton. — {a) Dermal; a triangularly and polygonally-meshed reticulation of very thin spiculo-fibre. The fibre contains a very large proportion of horny matter, and has a core of slender oxeote spicules uni- or biserially arranged ; the fibre is also " armed " wdth oxeote spicules stuck through it more or less at right angles, [h) Main ; composed of a rectangularly-meshed reticulation of stouter fibre, containing a very large amount of horny matter, but with a core of slender oxeote spicules arranged in the primary fibres polyserially (not strictly parallel but in a somewhat plumose manner), but in the secondary, crossing fibres, usually only uni- or biserially. Spicides. — Small, usually slightly curved oxea, measuring about 0"077 by 0'0024 mm. The most characteristic features of this species are its external form and the arrangement of the dermal skeleton. We have satisfied ourselves fairly well, by examination of Lamarck's probable tj^^e in the Jardin des Plantes, that ours is the same sjaecies, but as it has never been thoroughly described, we give a full description. Lamarck's Spongia pahnata is not to be confused 1 Mon. Brit. Spong., vol. i. p. 208. 2 Ridley, Zool. Coll. H.M.S. " Alert," Brit. Mus., 1884, p. 393 ; Cf. also Vosmaer, Broim's Klass. u. Ordnung. des Thierreiohs, Porifera, pp. 211, 341. REPORT ON THE MONAXONIDA. 27 with the Spongia palmata referred to by Johnston ; ^ the hitter seems to be a HomceodictyaJ Locality.— Station 18G, September 8, 1874; kt. 10° 30' S., long. 142° 18' E. ; Torres Strait ; depth, 8 fathoms ; bottom, coral mud. Habitat. — Indian and European Seas (Lamarck) ; seas of New Holland (" var. (3 ") (Lamarck) ; Torres Strait (Challenger) Chalina pergamentacea, Eidley (PL V. fig. 2). 1881. GlaiJochaKna armigcra, var. iwrgamentacea, Ridley, Proc. Zool Soc. Lond., January, 1881, p. 112, pL X. fig. 4. 1884. Cladochalina pergamentacea, Ridley, Zool. CoU. H.M.S. "Alert," Brit. Mus., p. 398. Eepresented in the collection by a single small, compressedly lobate specimen (PI. V. fiig. 2) bearing a row of circular oscula of various sizes all round the margin. The specimen is characterised by a very glabrous surface (very likely the surface of the type would have been glabrous had it not been dried). The fibre is not so stout as in the type, and the spicules are very much more abundant, and in the larger fibres arranged polyserially. As the external form of this sponge is very characteristic, and has never yet been figured, we give an illustration of it. ioca%.— Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' 0" E. ; Bass Strait ; bottom, sand and shells. One specimen. Habitat. — Off the east coast of Brazil (Ridley, " Alert ") ; Torres Strait (Ridley, " Alert ") ; Bass Strait (Challenger). Chalina rectangularis, Ridley and Dendy (PI. V. fig. 3; PI. XLVI. fig. 6). 1886. Chalina rectangularis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol xviii. p. 331. Sponge (PI. V. fig. 3) sessile, encrusting, thin ; throwing up here and there low, mound-like prominences, each of which normally bears a single osculum at the summit. The single specimen covers half a valve of a PecMfri(Zosa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol.xviii.p. 332. Sponge (PI. VIII. fig. 5) massive and lumpy, subspherical. An upper and a lower half are distinguishable ; from the upper half arise very numerous' short fistulas, scattered irregularly over the surface and all pointing more or less vertically upwards, not radiating in aU directions as in the preceding species. Lower half almost without trace of fistulas. Fistulse nearly all broken off close to the surface ; usually (when perfect) short, broad, finger-like (PI. VIII. fig. 5a), closed at the ends, measuring about 25 mm. in length by 6 to 12 mm. in width. Size of body, 93 to 137 mm. in diameter. Colour in spirit, pale yellow. Texture, solid and very dense throughout, penetrated hy ' Ann. and Mag. Nat. Hist., ser. 5, vol. x. p. 121. (zooL. CHALL. EXP. — PART LIX. — 1887.) Nnn 5 34 THE VOYAGE OF H.M.S. CHALLENGER. wide ramifying tubes wliicli open into the fistulte. Surface, very uneven, covered with numerous bladder-like, almost glabrous swellings. The whole much encrusted with foreign organisms, Serpulas, Polyzoa, Sponges, &c. Skeleton. — Arranged as in Rhizochalina fistulosa into (a) a dermal network of has- tately-pointed oxeote spicules, not very closely aggregated, and (6) a rather delicate reticu- lation of spiculo-fibres, forming subequal polygonal meshes (the bast-layer), the fibres com- posed of oxeote spicules similar to those of the dermal layer, (c) A fairly close reticulation of stouter fibres filling up the interior of the sphere, and composed of spicules like those of the dermal layer, with, as usual, numerous unattached spicules lying between the fibres. Spicules. — Hastately pointed oxea (PI. IX. fig. 1), slightly curved and abruptly, but fairly sharply, pointed, measuring about 0'195 by 0"013 mm.; size and shape very uniform throughout. The rind, composed of the dermal membrane and the subjacent bast-layer, is much more firmly attached to the underlying tissues in this species than in, for example, Rhizo- chalina jyedunculata (below). At the ends of the fistulse (PL IX. fig. 7) the arrange- ment of the dermal spicules is difi'erent to what it is on the body (as shown in the figure). These large specimens present many features of great interest. The one figured (PL VIII. fig. 5) is firmly attached to a slab of rock by one side. There is a very distinct divi- sion into upper and lower surfaces, the fistulse being chiefly confined to the upper surface. We at first identified this well-marked form with Alcyonium putridosum, descril^ed by Lamarck {loc. cit.) from Australian Seas ("port du roi Georges"). There can be no douljt that Lamarck's Alcyonium pxitridosum is some species of Rhizochalina, and that it is nearly allied to Rhizochalina p)utridosa, but as there is not sufiicient evidence to make the identification a certainty, we have given the synonym with a query. Localities.— Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' E. ; Bass Strait ; depth, 38 fathoms ; bottom, sand and shells. Three large specimens. OffPort Jackson, 30 to 35 fathoms. One large specimen. Off Bahia, 7 to 20 fathoms. Some fragments of fistulse, which seem to be referable to this species (?). Habitat. — Australian Seas (Challenger) ; off Bahia (Challenger). Rhizochalina singaporensis. Carter, sp. 1883. PhlceoiUcfyon singapcyrense. Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. xii. p. 326, pL xiii. fig. 17. 1884. Rhizoclialina singaporensis, Kidley, Zool. Coll. H.JI.S. "Alert," Erit. Mus., p. 421, pi. xli. fig. «. With this species we identify a single fistula, in a bad state of preservation and of doubtful locality. The species is described by Carter (loc. cit.), and with only a fragment before us it does not seem ad^dsable to go into further details. REPORT ON THE MONAXONIDA. 35 Localltij.—Q.) Station 135, October 15, 1873 ; lat. 37° 1' 50" S., bug. 12° 19' 10" AV. ; Tristan da Cunha ; depth, 360 fathoms ; bottom, volcanic sand. Habitat. — Singapore (Carter); Prince of Wales Channel, West and Alert Islands, Torres Strait (Ridley) ; (?) Tristan da Cunha (Challenger). RMzochalina pedunculata, Eidley and Dendy (PI. VIII. fig. 1 ; PI. IX. figs. 2, 6). 1886. RMzochalina pedunculata, Ridley and Dendy, Ann. and Slag. Nat. Hist., ser. 5, vol. xviii. p. 332. Sponge (PI. VIII. fig. 1) irregularly rounded, slightly elongated, with one end drawn out into a short, stout peduncle, to the extremity of which is firmly attached a quantity of black sand. The sponge has evidently lived in an erect position, attached to the bottom by the peduncle. Height, inclusive of peduncle, 40 mm. ; breadth 25 mm. Un- fortunately, a portion of the peduncle is broken ofi" obliquely, leaving only a fragment of the base. Irregularly scattered over the surface of the sponge occur a number of fistul^e, about 4 mm. in diameter, all broken off (? some naturally terminating) close to the surface of the sponge, so as to present the appearance of so many oscula ; nine such fistulte are present, varying a little in width. Colour of the body a dirty brownish-yellow. The bast-layer and dermal membrane together form a very thin coating to the sponge, no thicker than a sheet of writing paper, which readily peels off from the underlying tissues. Surface slightly rough and wrinkled. Texture firm and compact throughout. Wide canals penetrate right to the centre, converging and opening into the bases of the fistula, which probably act as oscular tubes. Skeleton. — Arranged as in Rhizochalina Jistulosa, with (a) a dermal network of chiefly oxeote spicules, lying immediately below the thin, transparent external membrane. The spicules of the dermal network vary very greatly in size and form. We have (l) sausage- shaped (strongylote) spicules (PI. IX. fig. 2) with rounded ends, nearly straight or very slightly curved, measuring 0'106 by O'Ol mm., 0*23 by 0'013 mm., 0'19 by 0'017 mm., &c. It is not uncommon to find smaU strongylote spicules with a central bulbous inflation (PL IX. fig. 2, h, c) ; these are probably young forms. (2) More numerous, oxeote spicules, nearly straight, or slightly curved, sometimes becoming blunted at the end, varying greatly in size, from 0"09 by 0"0047 mm. to 0'25 by 0'009 mm. (There can be little doubt that the smaller spicules are here, as in other cases, simply young forms, which have not yet attained their maximum size.) (6) A bast-like reticulation of spiculo-fibres, which in this species forms only a very thin layer, consisting of a uni- lamellar network of fibres, comj)osed of oxeote (and strongylote ?) spicules similar to those forming the dermal network. The fibres are very compact, averaging about O'l mm. in diameter. (c) A rather scanty reticulation of spiculo-fibres, ramifying through the interior of the sphere and supporting the canal system. Also a great number of loose 36 THE VOYAGE OF H.M.S. CHALLENGER. si)icules, mainly arranged iDerpeudieularly to the walls of the smaller canals. The latter spicules are mostly slender, slightly curved oxea ; sometimes they become blunted ; average maximum size, 0'23 by 0'009 mm. Sincules. — Oxeote to strongylote, size variable {vide supra). We were at first inclined to refer this sponge to Rhizochalina singaporensis, Carter, but closer examination shows certain well-marked diflferences, although in some respects the two forms approach one another. Thus in Rhizochalina p)edunculata, the dermal membrane contains a number of cylindrical spicules, some of which attain great size {vide measurements), but which are, however, nearly or quite straight, while the deeper spicules sometimes show a strong tendency to become blunted. In Carter's species there are a great number of cylindrical spicules, but they are mostly very short, stout, and strongly bent. The presence of a distinct peduncle (whence the specific name) may possibly be regarded as characteristic, although, as there is only one specimen, it is impossible to say whether it is a constant feature. It is perhaps noteworthy that a great number of the oxeote spicules of this species show a tendency to become slightly rough and jagged at the points, suggesting the idea that they have begun to undergo absorption (that is to say, as a vital process, not the absorption occurring in dead spicules, which these are not). Locality. — Api, New Heljrides, 60 to 70 fathoms. Genus Oceanapia, Norman (PI. IX.). 18G9. Oceanapia, Norman, Brit. Assoc. Rep. for 1868, p. 334. Microsclera (viz., sigmata) present. Oceanapia rohusta, Bowerbank, sp. (PI. IX. fig. 3). ' 1866. Isodidya rohusta, Bowerbank, Mon. Brit. Spong., vol. iL p. 304. 1866. Desmacidon Jeffreysii, Bowerbank, Mon. Brit. Spong., vol. ii. p. 347. 1867. Gellius robushis, Gray, Proc. Zool. Soc. Lond., p. 538. 1867. ? Biemna jeffreysii, Gray, Proc. Zool. Soc. Lond., p. 539. 1869. Oceanapia Jeffreysii, Norman, Last Rep. Dredging Shetland Isles, Brit. Assoc. Rep. for 1868, p. 334. Sponge more or less globular, solid, with a number of tubular processes or jistulsB coming off abruptly from all parts of the surface. Colour in spirit yellowish to nearly white. Size (of the body) very variable. Fistidse 75 to 100 mm. long by 6 mm. wide, closed at the ends, where they exhibit an irregular branching into four or five or more short, finger-like processes, resembling the tubes of Rhizochalina. oleracea, figured by Schmidt (Spong. Atlant. Gebiet., pi. iv. fig. 1). Texture in well-preserved specimens firm and compact throughout. Surface smooth. REPORT ON THE MONAXONIDA. 37 Skeleton. — The skeleton consists, as in the Rhizochalinse, of three kinds of network, viz., (o) a dermal network of oxeote spicules lying immediately beneath a distinct, thin dermal membrane. (6) A coarse, bast-like reticulation of very stout and dense fibres, composed also of oxeote spicules, (c) A less dense reticulation of spiculo-fibre filling up the centre of the sphere and giving support to the canal system ; in form and size the spicules are the same as in the preceding. Spicules. — (a) Megasclera; oxea (PI. IX. fig. 3, a, h, c), size about 0"19 by O'OllS mm. ij)) Microsclera, extremely abundant in the deeper sarcode in the form of small sigmata (PI. IX. fig. 3, d), measuring 0-038 by 0-0032 mm. This common Shetland species is very poorly represented in the Challenger collection, only two specimens of the body, a small one about 16 mm. in diameter, and a larger one about 31 mm. in diameter, and a few detached fistuliB being present. The oxeote spicules are somewhat shorter than usual, measuring 0-19 by O'Oll mm., while those of the type measure 0-19 to 0-25 by 0-011 mm. Unfortunately, in the case of all the specimens, including the two bodies, there is some slight doubt as to the exact locality where they were obtained. According to the label on the outside of the bottle they were obtained " off Bermuda," whUe according to a label inside the Ijottle, they were obtained " ofi" Bahia, shallow water." Probably the Rhizochalina (f) fihulata, described in four lines by Schmidt (Spong. Meerb. Mexico, p. 7&) as a new species, from Barbados, 288 fathoms, and possessed of numerous sigmata, is also referable to Oceana2:)ia robusta, although the description, which is taken only from a fragment, is too short to make an identification practicable. Locality. — Bahia (?). Hcibitat. — Shetland Isles (Norman) ; ? Barbados (Schmidt) ; ? Bahia (Challenger). Subfamily 2. GELLiiKiE. 1872. Gelliadae (pars), Gray, Ann. and Mag. Nat. Hist, ser. 4, voL ix. p. 449. 1886. Gelliina, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii p. 332. Megasclera all diactinal, oxea or strongyla. Microsclera present, viz., sigmata or toxa. No rind nor fistulse. Genus Gellius, Gray (Pis. VIII., XIIL, XXVI.). 1867. Gellius, Gray, Proc. ZooL Soc. Lond., p. 538. 1867. Asychis, Gray, Proc. Zool. Soc. Lond., p. 539. 1870. Desmacodes, Schmidt, Spong. Atlant. Gebiet., p. 5. 1880. „ Vosmaer, Notes Leyden Mus., vol ii. p. 104. 1882. Fihularia, imrs, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. ix. p. 282. 1886. ? Fihulia, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. xvii. p. 51. Very little spongin present, never forming distinct fibres. Gray's original diagnosis runs as follows : — " Sponge massive, minutely hispid. 38 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton regularly netted. Spicules of two kinds: — (l) Fusiform. (2) Bihamate, simple, iind contorted." The increase of our knowledge, due to recent discoveries, has necessitated several not very serious alterations in the original diagnosis. Thus, it has been found advisable to omit the words " sponge massive," for we now know at least one species, Gellius calyx, nobis, which has a beautifully symmetrical and elegant form. Again, the genus has had to be enlarged to admit certain forms with toxa, or with long, straight, hair-like microsclera, although at present only a few such are known, viz., Gellius angulatus, Bowerbank, sp., Gellius arcoferus, Vosmaer, Gellius Jlabelliformis, nobis, Gellius pyriformis, nobis ; and it is possible that these and kindred forms may subsequently require a separate genus or subgenus to include them. The smooth oxeote character of the megasclera appears to be a very good and constant point, but it is important to notice that the ends of the oxea may become completely rounded oflF, as in Gellius pyriformis and Gellius carduus, the spicules then becoming almost cylindrical (strongylote), but showing their real nature by the marked manner in which they still taper to each rounded extremity, and also by their close correspondence in shape and curvature with those of allied forms. Intermediate stages in this process of rounding off are also sometimes met with. Vosmaer, in the Notes from the Leyden Museum {loc. cit.), does not accept the genus Gellius at all, but has in its place Desmacodes, Schmidt, ^n his later work on the Porifera in Bronn's Klassen und Ordnungen des Thierreichs, he, however, substitutes Gellius for Desmacodes, but gives the genus a much wider scope than we are inclined to admit ; his diagnosis runs as follows : — " Stabnadelu glatt oder gedornt ; Spongin wenig entwickelt. Keine Anker. Statt dessen Haken und Bogen. Atlantischer und Arctischer Ocean (Mittelmeer V). Untief bis 1 8 0 Fad. " ^ The Challenger dredgings extend the range of depth at which Gellius occurs to 600 fathoms, and show it to be as well established in the deep-sea as in shallow water. Gellius varius, Bowerbank, sp. (PI. VIII. fig. 4). 1875. Ealiclwndria varia, Bowerbank, Proc. Zool. Soc. Lond., p. 292. 1875. Isodictya virgata, Bowerbank, Proc. Zool. Soc. Lend., p. 294. 1880. ? Desmacodes fihulatus, Vosmaer, ^)ars, Notes Leyden Mus., vol. ii. p. 105. 1884. Gellius varius, Ridley, Zool. CoU. H.M.S. "Alert," Brit. Mus., p. 424. With this species we identify several good-sized fragments from Station 208. They are all more or less cylindrical, and the longest is a branch about 80 mm. long by about 6 mm. in diameter, slightly forked at the extremity. Other pieces are thicker, much more distinctly branched, and the branches show a strong tendency to anastomose. The measurements of the spicules are as follows : — Oxea about 0-22 by 0-013 to 0-0145 mm. > Vide Klass. u. Ordnung. des Thierreichs, Porifera, p. 349. REPORT ON THE MONAXONIDA. 39 Sigmata about 0"04 by 0"0015. It is needless to give furtber details conceruiug au already well-known species; the reasons for uniting Halichondria varia, Bowerbank, and Isodictya virgata, Bowerbank, bave already been given by Eidley in the " Alert" Report {loc. cit.). Locality.— Station 208, January 17,1875; lat. 11° 37' N., long. 123° 31' E.; Philip- pine Islands ; depth, 18 fathoms ; bottom, blue mud. A number of fragments. Habitat. — Strait of Malacca (Bowerbank) ; Port Darwin (Ridley) ; Philippines (Challenger), Gellius carduus, Ridley and Dendy (PL VIII. fig. 3 : PI. XIII. fig. 7). 1886. Gellius carduus, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xvLii. p. 333. Sponge (PI. VIII. fig. 3) sessile, more or less oval, or egg-shaped. Size of the two most perfect specimens about 33 mm. long by 20 mm. broad. Colour in spirit pale greyish-yellow. Texture loose but firm. Interior of the sponge cavernous, owing to the numerous wide canals by which it is traversed. Surface raised into numerous angular prominences, many of which have oscula at their summits. Dermal membrane distinct, hyaline, covering large subdermal cavities, and pierced by numerous scattered pores, about 0"052 mm. in diameter. Oscida scattered singly on the summits of angular prominences. Skeleton. — (a) Dermal; consisting of a very beautiful unispicular reticulation of large, stout, blunted oxeote spicules (exhibiting occasionally a tendency to form loose, long fibres), {b) Main; consisting of a loose reticulation of similar spicules, not united to form definite fibres. Hence the firmness of the sponge is due rather to the large amount of tough membranes present, supported by loose spicules, than to the presence of any definite spiculo-fibre. Spicides. — (a) Megasclera ; of one kind only, viz., blunted oxea (PL XIII. fig. 7), slightly bent, and tapering to a rounded extremity at each end. Size 0*6 by 0'023 mm. (b) Microsclera ; numerous small sigmata of the usual shape, measuring 0'02 by 0"0012 mm. This species is readily distinguished by its very characteristic external form, the surface resembling that of a large thistle-leaf, whence the specific name. The shape of the skeleton spicules is also very characteristic ; they might be described as strongjda, tapering slightly towards each end ; but there can be no doubt that they are simply blunted oxea. Localities.— Station 148a, January 3, 1874 ; lat. 46° 53' S., long. 51° 52' E.; depth, 240 to 550 fathoms ; bottom, hard ground, gravel, shells. One specimen, attached to a pebble. Ofi" Prince Edward Island, December 27, 1873; depth, 85 to 150 fathoms. Three specimens. Olf Marion Island, 50 to 75 fathoms. One specimen. 40 THE VOYAGE OF H.M.S. CHALLENGEE. Gellius cardmis, var. magellanica, nov. (PL XIII. fig. G). Sponge massive, sessile. Shape elongately oval, size about 62 mm. by 25 mm.; tapering irregularly towards each end. Colour in spirit pale brownish-yeUow. Texture rather firm and compact, but brittle. Surface glabrous but uneven. Dermal membrane distinct, rather coarse. Oscula irregularly scattered over the upper surface. Skeleton. — (a) Dermal ; in some parts a unispicular reticulation of oxeote spicules ; in other parts broad bands of loose spiculo-fibre are also present, (b) Mai?i ; a reticula- tion of loose but distinct spiculo-fibre, main fibres running more or less vertically to the surface and others at right angles to these ; but these two main directions are only very roughly marked out. The fibres themselves are about 0"175 mm. in diameter. Spicules. — (a) Megasclera; stout oxea (PL XIII. fig. 6), tapering rather gradually to a rather blunt point at each end ; measuring about 0"49 by 0*022 mm. (6) Microsclera ; numerous very small sigmata measuring 0"019 by 0"0008 mm. This is a very interesting geographical variety from the Strait of Magellan. The main features in which it differs from the type specimens of Gellius carduus concern the oxeote spicules which, in the variety in question, are much more pointed and a good deal shorter than in the type ; the skeleton-fibre also seems to be better developed. LocalitT/.— Station 311, January 11, 1876; lat. 52° 45' 30" S., long. 73° 46' W.; Strait of Magellan, west end ; depth, 245 fathoms ; bottom, blue mud ; bottom tempera- ture, 46°*0. One specimen. Gellius Isevis, Ridley and Dendy (PL XIII. fig. 8). 1886. Gellius Ixvis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol xviii. p. 333. This species agrees very closely in spiculation with the typical specimens of Gellius carduus. The spicules, both megasclera (PL XIII. fig. 8) and microsclera, are of the same shape and almost of the same size. The chief points of difference concern the surface and size of the specimens. The surface is not nearly so uneven as in Gellius carduus; the oscula are large and round, and their margins flush with the general surface of the sponge. The specimens are also much larger than the types of Gellius carduus, the largest being massive and cake-like, and measuring 143 mm. in length, 81 mm. in breadth, and about 31 mm. in average thickness. Two specimens were obtained, both of which were infested by a very beautiful and peculiar little Isopod, which occurs also on a fine Esperella from the same Station. Zoca%.— Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; oft' the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°'2. Two fine specimens. EEPOET ON THE MONAXONTDA. 41 Gellius glacialis, Eidley and Dendy (PI. VIII. fig. 7 ; PI. XIII. figs. 1, 15, 19). 1886. GcUius glacial is, Eidley and Dendy, Ann. and jMag. Nat. Hist., sur. 5, vol. xviii. p. 333. Sponge (PI. VIII. fig. 7) massive, sessile; shape very various, globular, lobate, or cylindrical. Size, ranging up to 81 mm. long by about 31 mm. broad (a roughly cylindrical specimen). Colour in spirit very pale greyish-yellow. Texture firm, but very brittle. (The sj^onge has very much the appearance of a mass of half-melted snow, whence the specific name.) Surface even, but rather rough ; dermal layer distinct, readily flaking ofi"; composed of the dermal membrane supported by a reticulation of spicules. The dermal membrane itself is extremely delicate and has almost everywhere been removed, probably washed ofi' in coming up in the dredge. It is further supported by abundant, large sigmata. Pores ; as far as can be made out these are rather large and not abundant. Oscula large, irregularly scattered, having their margins flush with the surface of the sponge. Skeleton. — {a) Dermal (PL XIII. fig. 19); consisting of a very beautiful, almost unispicular reticulation of large, stout, oxeote spicules, {h) Main ; consisting of a very loose and irregular reticulation of spicules like those forming the dermal layer, and of about the same size (PI. XIII. fig. 1). Sometimes a few spicules lie more or less parallel and close to one another, so as to form a rudimentary spiculo-fibre. The whole reticulation is very delicate and fragile, and a very small amount of cementing substance (spongin) is present at the points where the spicules touch one another, at or near their apices. Spicules. — («) Megasclera ; large oxea (PL XIII. fig. 1), slightly curved, and tapering rather abruptly to a sharp point at each end ; size, 0*65 by 0'036 mm. (h) Microsclera ; numerous, large sigmata (PL XIII. fig. 15) of the usual shape, occurring very abundantly in the dermal membrane ; size up to 0"07 by 0'0063 mm. Smaller ones, probably young forms, also occur. This species may be most easily recognised by its external appearance. It is probably nearly related to Gelliiis carduus, nobis, but has not the sharp and projecting edges to the oscula which characterise that species. It is exceedingly delicate. In the specimens before us the soft parts have been almost entirely washed away, leaving a very fragile skeleton. Owing to the absence of soft parts and of any coarse fibre the specimens have a pecuhar translucent look. Locality.— ^i&tion 142, December 18, 1873; lat. 35° 4' S., long. 18° 37' E.; AgulhasBank; dejjth, 150 fathoms; bottom, green sand; bottom temperature, 47°'0. About twenty-five specimens. (ZOOL. CHALL. EXP. — PAET LIX. — 1887.) Nun 6 42 THE VOYAGE OF H.M.S. CHALLENGER. Gellius glacialis, var. nivea, nov. (PI. VIII. fig. 8 ; PI. XIII. figs. 4, 12). Sponge (PI. VIII. fig. 8) incrusting, amorphous, in the form of a small round cushion, filling up the angle between the branches of a Polyzoon. Diameter about 12 mm. thickness about 8 mm. Colour in the dry state white ; appearance like fleecy snow, whence the varietal name. Texture brittle. Surface even, but rough, owing to the dermal membrane being absent. Oscula and 2^ores unknown. Skeleton. — A very loose reticulation of oxeote spicules, not united together into distinct fibres. This reticulation is very like that of Gellius Jlagelli/er, nobis, but in that species it is just possible to speak of spiculo-fibre, while here the reticulation is so loose that the term is no longer admissible ; there is, however, but little real difierence in this respect between the two. Spicules. — (a) Megasclera; oxea (PI. XIII. fig. 4), tapering rather suddenly to a sharp point at each end. Slightly curved and measuring 0"53 by 0'02 mm. (6) Microsclera are present in the form of numerous sigmata (PI. XIII. fig. 12) of the usual shape, but attaining a very great size, up to 0'145 by 0"0063 mm. Smaller ones are also present, probably young forms of the others. It will be seen that both in the arrangement and form of the mes-asclera and in the great size of the sigmata this variety of the species comes very near to Gellius Jlagelli/er, nobis ; it is, however, easily distinguished by the very different shape and superior thickness of the large sigmata. The sponge has unfortunately been dried, so that the skeleton is the only part the structure of which can be satisfactorily made out. Judging from its small size it is probabl}^ a young form. Xoe«%.— Station 145, December 27, 1873; lat. 46° 43' 0" S., long. 38° 4' 30" E.; Prince Edward Island ; depth, 140 fathoms ; bottom, volcanic sand. One small specimen. Gellius fiagellifer, Eidley and Dendy (PI. XIII. figs. 5, 10). 1886. Gelliiis fiagellifer, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 333. Sponge massive, sessile, roughly triangular in shape,^ the side of the triangle measuring a little over 25 mm. in length ; thickness of the sponge from the centre of one triangular surface to the centre of the other 16 mm. Colour in spirit pale greyish- yellow. Texture soft, brittle. Surface even but shghtly rough, probably owing to the dermal membrane having been in most parts rubbed off. Dermal membrane thin and transparent, with numerous scattered i^ores about 0"07 mm. in diameter. Only one distinct osciilum visible at one side. A second specimen, very likely only a fragment, is cylindrical, 29 mm. long by 8 mm. in diameter. Skeleto7i. — There is a dermal netw^ork of spiculo-fibre indistinguishable from the spiculo-fibre forming the skeleton at large. The deeper skeleton is composed of a loose, ^ Perhaps a young specimen, incipiently lobose. REPORT ON THE MONAXONID.^. 43 irregular reticulation, of very loose spiculo-fibre, the meshes of which are about one spicule in breadth. Spicules. — (a) Megasclera; stout, slightly curved oxea (PI. XIII. fig. 5), tapering not very suddenly to a fairly sharp point at each end ; size 0"42 by 0"018 mm. (b) Micros- clera; sigmata (PI. XIII. fig. 10) are present in great abundance, and attain to a great length and a high degree of curvature, being folded once on themselves, so as to resemble little whip-lashes (whence the specific name), the greatest length from curve to curve of these spicules is about 0'06 mm., thickness about 0'0021 mm., the real length is con- cealed by the curvature, and is, of course, much greater than that given. In addition to these long sigmata there are also present numerous smaller ones of the usual shape, possibly young forms of the others, size 0"044 by 0'0025 mm., &c. These smaller sigmata are especially abundant in the dermal membrane. Only two small specimens of this sponge were obtained. The most characteristic features about it are the size and shape of the microsclera, which, taken together, are quite sufficient to distinguish it from any other Gellius. Vosmaer mentions under " Gellius vagabundus (0. S.)," in The Sponges of the "Willem Barents" Expedition, 1880 and 1881, p. 29 (sejs), a variety of that species possessing oxea and sigmata, similar in form to those of our species. His specimen, though containing a few styli, is obviously a true Gellius {Gellius vagabundus being Desmacella for us), and it is not improbably referable to Gellius jiagelliger. It was obtained by the "Willem Barents" expedition of 1880, and hence probably in the Arctic Sea, though the exact locality is unknown. Having regard to the want of definite characters in this species, other than the form of the sigmata, we cannot further insist on the strong resemblance which this form bears to our species, as its locality is so far removed from that of Gellius Jlagellifer. Locality. — Ofi" Marion Island, 50 to 75 fathoms. Two small specimens. Gellius calyx, Eidley and Dendy (PI. VIII. figs, 6, G« ; PI. XIII. figs. 2, 9). 1886. Gellius calyx, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 334. Sponge (PL VIII. fig. 6) consisting of a pyriform body attached by the narrow end to a long slender stalk, which is very slightly expanded at the base ; body hollow, with a single large rounded opening at the summit (PI. VIII. fig. 6a). The whole sponge closely resembling, in size and shape, the flower of a crocus. Length of stalk 50 mm. Mean diameter of stalk 2 mm. Length of body 33 mm., greater breadth of body 10 mm. Diameter of mouth 4 mm. Colour in spirit pale greyish-yellow. Texture of the body very soft, fragile and crumbling ; of the stem hard and stringy. The fibres of the stalk appear to spread out on reaching the body, so that the lowest part of the latter is firm and dense. Surface minutely hirsute, even, but with numerous minute holes, probably 44 THE VOYAGE OF H.M.S. CHALLENGER. due to the dermal membrane having been in most parts rubbed off, exposing the subdermal cavities. Dermal membrane very thin and delicate, with numerous rounded pores, about 0'53 mm. in diameter. Skeleton. — (1) Of the stalk. The stalk is composed of a thick, dense core of spiculo-fibre composed of closely packed, longitudinally arranged, oxeote spicules. Around this is a thin layer of soft tissues, containing numerous loosely placed spicules, both megasclera and microsclera. (2) Of the body, (a) Dermal; a very loose and irregular reticulation of oxeote spicules supporting the dermal membrane. (Jj) Main; a loose reticulation of oxeote s^Dicules in which one can, however, distinguish certain loose main bands of spicules running in a direction vertical to the surface of the sponge. At the surface these spicules project freely, giving it a characteristic hirsute appearance. There are also numerous spicules placed transversely, running from one to another of the above mentioned main bands, the whole skeleton forming an irregular and but slightly coherent network. Spicules. — (o) Megasclera; sharp-pointed oxea (PI. XIII. fig. 2) tapering rather gradually towards each end. In the body they show a tendency to become blunted, and in the stalk to become irregularly pointed as if in process of absorption. Size about 0"5 to 0"7 mm. by 0"022 mm. (&) Microsclera; very numerous small sigmata (PI. XIII. fig. 9) of the usual shape, but often scarcely, if at all, contorted, measuring 0-02 by 0-002 mm.' The presence of a distinct stalk, the cup-like shape of the body, and the hirsute appearance of the surface, are all characters which serve readily to distinguish this species from all other known species of the genus. In the spicules themselves there is nothing characteristic. Owing to its well-defined and elegant external form this is perhaps the most remarkable species of the genus as yet known. Zoca%.— Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W. ; off the mouth of the Rio de la Plata; depth, 600 fathoms; bottom, green sand; bottom temperature, 37°'2. One specimen. Gellius angulaUis, Bowerbank, sp. 1866. Halicliondria angulata, Bowerbank, Mon. Brit. Spong., vol. ii. p. 233; (vol. iii. jx 101, pi. xli. figs. 4-8). 1880. Desmacodes angidatus, Vosmaer, Notes Leyden Mus., vol. ii. \). 107. Three small globular specimens of this sponge, the smallest about the size of a pea (8 mm. in diameter) and attached to a Polyzoon, the other two about 6 to 18 mm. in diameter, were obtained by the Challenger at Station 75, at a depth of 450 fathoms. The species has already been fully described and figured by Bowerbank {loc. cit), with the exception that he omits all mention of the sigmata amongst the microsclera. That ' There are also a very few long, thin, hair-like spicules, very possilily foreign, but interesting, because simOar ones occur in great abundance in another .species of Gellius, from the .same locality {vide Gellitis sp., p. 46). REPORT ON THE MONAXONIDA. 45 these occur iu liis species, is shown b}'' a preparation in the Bowerbank collection in the British Museum. Vosmaer, presumably following Bowerbank, also omits to mention the sigmata. The discovery of these sigmata fully justifies the reference of this species to the genus Gellius. The measurements of the spicules are as follows : — Oxea (megasclera) 0"29 to 0'34 by 0'009 to O'Ol mm. Toxa (microsclera) 0"08 by 0'0012 mm. Sigmata (microsclera) 0-019 by 0-0012 mm. Zoca%.— Station 75, July 2, 1873; lat. 38° 38' 0" N., long. 28° 28' 30" W. ; off the Azores ; depth, 450 fathoms ; bottom, volcanic mud. Three small specimens. Habitat. — British Seas (Bowerbank) ; off Azores (Challenger). Gellms jlahelliformis, Ridley and Dendy (PI. XXVI. figs. 5, 5a). 1886. Gelliits jlahelliformis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 4, vol. xviii. p. 334. Sponge (PI. XXVI. figs. 5, 5a) erect, compressed, forming thin laminae. The largest piece forms an erect, flattened, broadly expanded lobe, arising from a small woody base and rapidly expanding upwards; margin entire, thin, turned back; in addition to this back- ward curvature of the margin, which would appear most plainly in longitudinal section, the lameUa is also curved in a direction at right angles to the former, the second curvature being most evident in horizontal section, and causing the sponge to become convex on one side and concave on the other. The specimen, in fact, resembles a piece broken off from a cup, with the lip turned back; it is not improbable that the sponge may some- times be caliciform, but we do not think the present sjDecimen has ever been so. Height of the largest piece 68 mm., greatest breadth 79 mm., thickness about 4 to S mm. Colour in spirit greyish-yellow. Texture very fragile. Surface even, with a deeply, though minutely, pitted appearance all over. On the convex side each little pit is covered over by the pore-bearing dermal membrane, but on the concave side the pits seem to represent the oscula, for we have been able to find none other. Dermal membrane very delicate indeed. Oscula (? minute, very abundant over the concave surface). Pores, very numerous rounded openings iu the dermal membrane on the convex surface, measuring about 0-085 mm. iu diameter, reducing the dermal membrane to a mere network. We have found none on the concave surface, but should not like to say for certain that they do not exist here, though proliably, from the analogy of other flabellate sponges, they are almost or quite confined to the one surface. Skeleton. — (a) Dermal; there is no very definite dermal skeleton on either side of the sponge, but on the pore-bearing side the toxa are enormously abundant and arranged in loose, irregularly reticulating, fibre-like tracts. (Jj) Main; a loose and cpite irregular reticulation of oxeote spicules. 46 THE VOYAGE OF H.M.S. CHALLENGEE, Spicules. — (a) Megasclera ; large, smooth, fairly stout oxea, rather gradually and sharply pointed at each end, and with a slight bend in the centre; size, about 07 by 0"03 mm. (JS) Microsclera; (1) large, stout, very strongly curved, sharply pointed, simple and contort sigmata, very abundant, measuring about 0"07 by 0"0063 mm. ; (2) smooth, sharp-pointed toxa, with very obtuse central angle, unusually large, measuring about O'lS l)y 0'0063 mm. ; exceedingly abundant all through the sponge, more especially so lining the subdermal cavities. Vosmaer's GcUius arcoferus,^ an Arctic form, certainly comes very near to this species, especially if, as Dr. Vosmaer suggests, the real form is that of a fan; it differs, however, in the proportions of the sigmata and toxa, inter se, the former being very much smaller as compared with the latter than in Gellius jlabelUformis ; this much we can deduce from the figures, but as to the actual sizes of the spicules in Gellius arcoferus, we are unfortunately able to say nothing, for Dr. Vosmaer gives no measurements. Gellius cmgulatus, Bowerbank, also approaches Gellius Jlahelliformis in spiculation, being also possessed of toxa, but differs in external form and in the very much smaller size of the microsclera. Locality. — Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W. ; off the mouth of the Rio de la Plata; depth, 600 fathoms; bottom, green sand; bottom temperature, 37°'2. One large piece and several small ones. Gellius sp. (PL XIII. fig. 3). Sponge consisting of a pear-shaped body, tapering to a point at one end, which has a fibrous appearance, and where possibly a stalk has been attached. Body solid. Length 29 mm., greatest breadth 12 mm. Colour in spirit pale greyish -yellow. Texture very soft and fragile. Surface even, slightly hirsute. Oscula small and scattered. Dermal memhrane thin and delicate. Skeleton. — Arranged much as in the body of Gellius calyx, altogether very loose. Spicules. — (a) Megasclera, of one kind only, viz., oxeote (PI. XIII. fig. 3), slightly curved and rounded ofi'at each end; size 0"6 by 0*022 mm. (b) Microsclera ; (l) small sigmata, size 0"019 by 0"0015, mm.; (2) very long, thin, hair-like spicules, apparently not arranged in bundles ; size 0'7 by 0'003 mm. Again we have only one specimen of the sponge present, and that possiljly imperfect. As it comes from the same station and agrees in several particulars with Gellius calyx it is very likely a close ally of the latter; this hypothesis will be strengthened if the examination of more specimens shows that the fine hair-like spicules are also really proper to Gellius calyx. In the form of the oxeote spicules and the presence of hair- like ones, together with the sigmata, the species in question agrees pretty closely 1 Sponges of the "Willem Barents" Expedition, 1880 and 1881, p. 2D, pi. iv. figs. 18, 19 ; pi. v. figs. 87-90. REPORT ON THE MONAXONIDA. 47 with Carter's Fihularia massa,^ which is possibly also a Gellius. The hair-like spicules in our sponge are, however, much longer than in Fihularia massa, and in the latter they appear to originate in trichite bundles. Locality.— ^i&iion 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; off the mouth of the Eio de la Plata; depth, 600 fathoms; bottom, green sand; bottom temperature, 37°'2. One specimen. Genus Gelliocles, Ridley (Pis. X., XIL, XLVIL). 1884. Gelliodes, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 426. A distinct and well-developed skeleton fibre is present, containing more or less spongin. Microsclera sigmata. Ridley's original diagnosis runs — " Desmacidinidae of erect habit and well-defined form, fibre distinct and compact ; outer surface of sponge beset with pointed eminences. Spicules smooth ; skeleton acerate and bihamate." This diagnosis must now be enlarged, in order to admit Gelliodes liclieniformis, Lamarck, sp., and Gelliodes poculum, nobis, and, omitting all description of external form, we may say that the genus differs from Gellius merely in the possession of a distinct and well-developed fibre with more or less horny matter, and from Toxochalina in the presence of sigmata in the place of toxa as microsclera. It is perhaps doubtful whether the last character is of generic value, and whether Toxochalina and Gelliodes should not be merged in one genus, but as no species is yet known whose spicular comjalement comprises both toxa and sigmata, they may at present be kept apart. Gelliodes filulata, Ridley (PI. XII. fig. 2). 1813. ? S]]oiujia ruhispina, Lamarck, Ann. Mus. Hist. Nat. Paris, vol. xx. p. 450. 1881. ? Axos fihulata, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. vii. p. 383, pL xviii. iig. 4. 1884. Gelliodes fibidata, Ridley, Zool. Coll. H.M.S. " Alert," Brit. ]\Ius., p. 427, pi. xxsix. fig. 1 ; pi. xli. figs, hb-bb". The Challenger obtained four good specimens of this sponge in Torres Strait, where also it was obtained abundantly by the " Alert." It is not necessary to redescribe the species here, but we give a figure of one very beautiful specimen, illustrating how the spines on the surface may bear secondary smaller spines, a feature which appears to have been overlooked hitherto (vide PI. XII. fig. 2). Locality. — Cape York, Torres Strait, August 7, 1874; depth, 3 to 11 fathoms. Four specimens. Habitat. — ? Bass Strait (Carter). Torres Strait (" Alert " and Challenger). ' Ann. and Mcuj. Nat. Hist., ser. 5, vol. ix. p. 282. 48 THE VOYAGE OF H.M.S. CHALLENGER. Gelliodes licheniforrnis, Lamarck, sp. (PL XII. fig. 3 ; PL XLVII. figs. 1, la). 1813. Spongia liclienifonnis, pars, Lamarck, Ann. Mus. Hist. Nat. Paris, vol. xx. p. 371. 1836. Spongia licheniforrnis, Lamarck, Hist. Anim. sans Vert., t. ii. p. 543. To this species we refer a single, encrusting, very lichen-like specimen from Station "135 (?) 60 fathoms." The Challenger sjiecimen difi"ers from the type (as evidenced by a dry specimen in the British Museum, in the series of Lamarckian sponges), in having slightly larger sigmata and a more slender fibre ; in its present condition the surface is also more even, but no doubt this is in part due to the fact that the specimen is preserved in spirit. As so little is known of the species we may add a few details derived from an examination of the Challenger specimen. Sponge (PL XII. fig. 3) sessile, encrusting. Colour in spirit light yeUowish-brown. Texture (in spirit) soft, spongy, very fibrous. Surface uneven. Dermal meonbrane verj- thin and transparent. Pores small, scattered. Oscula f Skeleton. — A not very regular reticulation of strongly developed, very distinctly lamellated, horny fibre cored by oxeote spicules, and of oxeote spicules united together by more or less ho]"ny substance (spongin). In addition to the main skeleton thus constituted there is a very well-developed and beautiful dermal reticulation (PL XLVII. figs. 1, la) of spiculo-fibre, consisting of a unispicular network of oxeote spicules with a very large proportion of spongin ; backed up behind by a network of much stouter spiculo-fibre, with very few spicules and still more spongin. Spicules. — (a) Megasclera ; of one kind only, viz., straight, smooth oxea, measuring about 0'23 by 0'0096 mm. (b) Microsclera ; of one kind only, viz., smooth, slender sigmata, usually measuring about 0*032 mm. in length, very rarely up to about 0"09 mm ; thickness about 0*0012 mm. The Challenger specimen might readily be mistaken for a young form of Gelliodes pocidum, nobis, but this idea is precluded by the difi"erent character of the dermal skeleton. There is no doubt that Lamarck's Spongia licheniforrnis includes more than one species, both as shown by the specimens in the British Museum and by his own statement (loc. cit.) : — "Habite dans difierentes mers, et ofire beaucoup de varietes." Our specimen agrees sufficiently closely with that in the British Museum to be ranked in the same species with it, but regarding Lamarck's other varieties we are not able to speak with certainty. Gelliodes poculum, Eidley and Dendy (PL X.). 1880. Gelliodes poculum, Eidley and Dendy, Ann. and ]\Iag. Nat. Hist., ser. 5, vol. xviii. p. 334. Sponge (PL X.) consisting of a thin, spreading lamina, which, in the case of the Challenger example, grows over a large, massive Tetractinellid sponge ; from this EEPOET Oi\ THE MONAXONIDA. 49 spreading lamina arise large funnel-sliaped calices with wide mouths. The basal lamina, ia the case of the single specimen in the collection, covers a large, but very uneven surface ; it averages about 4 mm. in thickness ; this is also the thickness of the walls of the calices. The specimen bears in all five calices of various sizes, the largest measuring about 50 mm. in height, and 25 to 62 mm. in greatest width of mouth (which is com- pressed). Colour in spirit brownish-yellow. Texture soft, spong)^, but very tough and fibrous. Surface very uneven but fairly smooth. More or less numerous, very small openings on the inside of the calices appear to be the true oscula. Skeleton. — (a) Main; a reticulation of stout horny fibre, cored sparsely by oxeote spicules arranged uniserially. {b) Dermal ; a similar but much closer reticulation of stout horny fibre, with very few axial spicules, but echinated very abundantly by tufts of oxea which project outwards. Spicules. — (a) Megasclera ; short, fusiform oxea, usually slightly curved, sharply and gradually pointed, measuring about 0'2 by 0'014 mm. {h) Microsclera ; rather large and slender, much curved sigmata, not very abundant, measuring about 0'12 by 0-004 mm. This interesting species is distinguished chiefly by its remarkable external form. It resembles a Chalinine sponge, described by Ridley under the name Tuba (Sijjhonochalina) bullata,^ but is distinguished as regards external appearance by the absence of the numerous sharp conuli on the surface, and microscopically, of course, by the presence of the sigmata. Dr. E. von Lendenfeld also obtained a specimen of Gelliodes poculum from the east coast of Australia. Locality. — Port Jackson, 30 to 35 fathoms. One specimen. Genus Toxochalina, Ridley. 1884. Toxochalina, Eidley, Zool. ColL H.M.S. "Alert," Brit. Mus., p. 402. A distinct, rectangularly arranged horny fibre present, cored by oxeote spicules, as in typical Chalinin^e. Microsclera toxa. Ridley's original diagnosis runs " Chalinidse with well-developed horny fibre arranged rectangularly. Spicules, a skeleton acerate and a tricurvate acerate (' Bogen,' German) flesh-spicule." Having regard to the close relationship of this genus to Gelliodes as indicated by the presence of the toxa, in conjunction with a distinct horny fibre, we have decided to remove it from amongst the " Chaliuinse " and place it in the " Gelliinse," amongst the Heterorrhaphidse. It has appeared to us preferable throughout to classify rather by the presence or absence and the form of the microsclera than by the greater or less amount of spongin present. 1 Zool. Coll. H.M.S. " Akrt," Brit. Mus., 1884, p. 399, pi. xxsix. fig. c. (zool. CHAiiL. EXP. — PART Lix. — 1887.) Nnn 7 50 THE VOYAGE OF H.M.S. CHALLENGEE. Toxochalina robusta, Eidley. 1884. Toxochalina robusta, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., 1884, p. 403, pi. xsxix. fig. G ; pi. xli. figs, n, n. We identify with this species a single small, irregularly branched specimen from Bahia, which agrees very closely in microscof)ical structure, though of not nearly such robust growth as the type. It may be only a young specimen. Locality. — Off Bahia, 7 to 20 fathoms. One specimen. Habitat— Fort Jackson (Ridley, " Alert ") ; off Bahia (Challenger). Subfamily 3. Tedaniin.e. 1886. Tedaniina, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 335. Megasclera always of two forms : — (1) Monactinal, styli, forming the main skeleton ; (2) Diactinal, tylota or tornota, dermal. Microsclera always present in the form of rhaphides. Two genera of this subfamily are distinguished, viz., Tedania, in which the stylus is smooth, and Tracliytedania, in which it is more or less spined. In both genera the form is usually massive, amorphous, and both genera are characteristically shallow-water forms. The dermal spicules are susceptible of very great variation in the different species, being sometimes bicapitate (tylota), and at other times hastately pointed (tornota). Genus Tedania, Gray (Pis. XL, XXIII. , XXIX.). 1862. Reniera, 2-iars, Schmidt, Spong. Adriat. Meer., p. 75, &c. 1867. Tedania, Gray, Proc. Zool. Soc. Lend., May 9, p. 520. Styli smooth. Gray's original diagnosis runs — " Sponge lobed, crested, with a lateral tube ending in an open mouth. Spicules of three kinds : — 1. Clavate, needle-shaped. 2. Fusiform, very slender, elongate, sometimes flexuous. 3. Cylindrical, with rather thicker, blunt ends." The range of external form exhibited by this genus is shown by the Challenger dredgings to be a very remarkable one indeed ; hitherto known only by more or less massive or digitate specimens, we have had to add to the genus two new species, Tedania infundibuliformis and Tedania actiniiformis, characterised by very specialised, though quite different, external forms ; the former (as yet known by only one specimen) being funnel-shaped, and the latter " actiniiform," ^ with oscular projections on the top and a definite zone of pores. ' Like an Actinia. EEPORT ON THE MONAXONIDA. 51 The species of this genus are very difficult to separate satisfactorily from one another ; future researches may, very probably, by the discovery of intermediate forms, render possible the union of some which are at present described as distinct. Tedania digitata, Schmidt (PI. XL fig. 3). 1862. Reniera digitata, Schmidt, Spong. Adriat. Meer., p. 75, PI. vii. fig. 11. 1864. „ amhigua, Schmidt, Spong. Adriat. Meer., Suppl., p. 39. 1867. Tedania digitata et amUgiia, Gray, Proc. Zool. Soc. Lond., May 1867, p. 520. 1868. Reniera muggiana, Schmidt, Spong. d. Kiiste v. Algier., p. 28. This widely distributed species has been sufficiently described by previous writers;* a number of specimens, varying in form from massive to digitate cylindrical (PI. XI. fig. 3), were obtained by the Challenger, off Port Jackson. Locality. — Off Port Jackson, 6 to 35 fathoms, several specimens, mostly rather fragmentary. Habitat. — Mediterranean (Schmidt) ; Atlantic (Schmidt) ; Antigua (Carter) ; Mozam- bique, Marie Louise Island, Amirante Group (Ridley, " Alert") ; Kurrachee (Coll. Brit. Mus.). Port Jackson (Challenger) ; Alert and Thursday Islands and Prince of Wales Channel, Torres Strait ; Port Darwin (Ridley, " Alert "). Tedania digitata, 'v&v. fibrosa, nov. A single rather large, massive specimen, distinguished from the type by the very slender spicules and the distinct primary fibre. Spicules. — (1) Smooth styli, measuring about 0*18 by 0"005 mm.; (2) double headed cylindricals (tylota), with minutely spined heads, measuring about 0'21 by 0"0024 mm.; (3) very fine, scattered rhaphides. Locality. — Off Port Jackson, 7 fathoms. One specimen. Tedania digitata, var. hermudensis, nov. A single small, massive specimen from Bermuda, of a pale yeUow colour and with small scattered oscula. The skeleton is somewhat " Isodictyal" in arrangement but confused. The chief point of distinction concerns the styli, which are more slender than in either of the two preceding forms and ajjpear to be blunted. Locality. — Bermuda, West Indies. One specimen. J Vide supra, also Eidley, Zool. ColL H.M.S. "Alert," Brit. Mus., 1884, pp. 417, 607. 52 THE VOYAGE OF H.M.S. CHALLENGER. Tedania tenuicapitata, RicUey (PI. XI. fig. 5). 188L Tedania ienmcapitata,TLidlej, "Alert" Coll., Proc. Zool. Soc. Lond., January, p. 124, pi. xi. fig. 1. This species has already been fully described (loc. cit.), it is, therefore, unnecessary to give much further description. The "Alert" obtained only a single sj^ecimen of the sponge while the Challenger obtained a considerable number. The Challenger specimens show plainly that the usual form of the sponge is massive (PI. XI. fig. 5). They also show a very considerable range of variation in the character of the ends of the tylota, which vary from a very elongate, slightly demarcate and acutely pointed form to a short, regularly defined head, rounded terminally ;^ the roughening of the fine rhaphides alluded to in the original description (loc. cit.) was made out in one specimen, but only in a spicule which was not adult. So great is the variation of the diactinal spicules and so nearly do they approximate the spiculation of this form to that of Tedania suctoria, Schmidt,^ that it must be considered doubtful whether it is more than varietaUy distinct from it. As yet, however, the structures described as excretory papillae by Schmidt (loc. cit.), have not been found here, and perhaps may prove distinctive. Localities.— Station 308, January 5, 1876; lat. 50° 8' 30" S., long. 74° 41' 0" W.; south-west Patagonia; depth, 175 fathoms; bottom, blue mud. Several specimens. Station 311, January 11, 1876; lat. 52° 45° 30" S., long. 73° 46' 0" W.; oS" south- west Patagonia; depth, 245 fathoms; bottom, blue mud; bottom temperature, 46° "0. Several specimens. Station 314, January 21, 1876 ; lat. 51° 35' S., long. 65° 39' W.; east of the Strait of Magellan ; depth, 70 fathoms ; bottom, sand ; bottom temperature, 46°-0. One fine specimen (PI. XI. fig. 5). Habitat. — Off" the south-west coast of Patagonia ("Alert," Challenger); between Strait of Magellan and Falkland Islands (Challenger). Tedania commixta, Ridley and Dendy (PI. XXIII. figs. 9, 9a, 9b, 9c). 1886. Tedania commixta, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 335. Sponge massive, amorphous. Colour in spirit creamy-yellow. Texture soft and compact, with a large admixture of foreign grit. Surface slightly corrugated. Dermal membrane thin but distinct. Oscula small, scattered, with their margins flush with the surface. Pores scattered (?). Skeleton. — Consisting of very loose, whisp-like fibres, apparently without any definite 1 Tlie rounded heads of the tylota were observed only in one of the specimens, and even here a large number of these spicules have the heads pointed. 2 Spong. Atlant. Gebiet, p. 4.3. REPORT ON THE MONAXONIDA 53 arrangement, running towards the surface, where they break up into divergent tufts of slender bicapitate cylindrical spicules (tylota). Sincules. — (a) Megasdera; (1) long, slender, slightly curved, hastately pointed styli ; measuring about 0'3 by 0"0042 mm. These spicules may approach the tylostylote form (PI. XXIII. figs. 9, 9a) ; (2) long, slender tylota, with usually only slightly developed, oval heads; size about 0'35 by 0-004 mm. (PI. XXIII. figs. 96, 9c). {b) Mkrosclera ; exceedingly fine, hair-like, scattered rhaphides, about 0"13 mm. long. This species is represented in the collection by a thick triangular lump with square- cut edges, which appear to be natural ; the upper surface is almost flat. The sponge, especially near the lower surface, contains large quantities of foreign bodies, sand, fragments of shells, &c. It is quite possible that here, as in other cases, the foreign bodies to a certain extent replace the proper skeleton of the sponge. The species is characterised by the presence of these foreign bodies, whence the specific name, and by the proportions of the megasclera and rhaphides, but it comes near to Tedania temdcapitata, being distinguished from it chiefly by the rhaphides being only about half as long as in that species, and by the presence of the foreign bodies ; hence it may perhaps have to be united with it at some future time. Locality.— Station 162, April 2, 1874 ; lat. 39° 10' 30" S., long. 146" 37' 0" E.; Bass Strait ; depth, 38 fathoms ; bottom, sand and shells. Tedania massa, Ridley and Dendy (PL XL figs. 4, 4a ; PI. XXIII. figs. 2, 2a, 26). 1886. Tedania massa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 335. Sponge (PI. XI. fig. 4) massive, flattened and cake-like, amorphous or lobose. Size apparently indefinitely great, the largest specimen measuring as much as 325 mm. long by 200 broad, and from 12 to 25 mm. thick ; other specimens are as much as 50 mm. thick. Colour in spirit, grey, brown, or pale yellow. Texture very soft and spongy. Surface fairly even, very minutely hispid. Dermal membrane fairly distinct, with a reticulate appearance. Oscida more or less numerous, scattered, or in rows along the margins of the flattened lobes (PL XL fig. 4, o), flush with the surface or with slightly raised margins ; diameter about 4 mm. Pores scattered, but commonly in groups (PL XL fig. 4a). Skeleton. — Divisible into three parts : — ( 1 ) a very loose, isodictyal reticulation of large, stout, stylote spicules. (2) A loose reticulation, with very large meshes, of stout, compact fibres like bands of horse-hair, composed of very long, fine, flexible, oxeote rhaphides, with more or less admixture of the stylote and diactinal spicules. One can distinguish primary fibres running more or less vertically to the surftice, and secondary fibres crossing these at right angles ; but the arrangement is very irregular and the fibres run for long distances without being crossed by others. (3) On approaching the surface of the sponge both these kinds of skeleton give place to a reticulation of hastately pointed cylintbicals •54 THE VOYAGE OF H.M.S. CHALLENGER. (tornota). These spicules form a network immediately below the dermal membrane, giving to the surface of the siwnge a characteristic lace-like appearance, and great numbers of them also project from the surface in divergent brushes. Spicules.— {a) Megasdera ; (1) the large, stout, smooth styli (PI. XXIII. fig. 2), confined almost exclusively to the deeper parts of the sponge ; often rather strongly curved and somewhat hastately pointed at the apex, but with the point often blunted ; size about 07 by O'OS mm. (2) The dermal spicules (PL XXIII. figs. 2a, 26) ; these' are rather slender, straight, and sub-hastately but not very sharply pointed at each end ; sometimes they have distinct, though small, heads, but very frequently the heads cannot be distinguished from the shaft. Size about 0-45 by O'OIS mm. (6) Microsdera; very numerous long, fine, oxeote rhaphides, up to about O'S mm. in length; often collected into fibres as above stated, but also found abundantly scattered about in the soft tissues of the sponge (? due to displacement and breaking up in cutting sections), and in the dermal membrane. They often exhibit a roughening of the surface similar to that remarked on in the corresponding spicules of Tedania tenuicajntata. This species was obtained by the Challenger in great quantities ofi' the mouth of the Ptio de la Plata and the eastern end of the Strait of Magellan. In the form of the diactinal megasclera it comes near to Tedania tenuicapitata, Eidley, but it may be readily recognised by the arrangement of the skeleton, the size and shape of the styli, and the lace-like appearance of the surface. The most remarkable feature of the species is the enormous size to which it attains. Judging from the quantity brought home from Station 320, we must imagine it in that locality coating portions of the sea-bottom in great spreading sheets. Localities.— Station 163d, June 12, 1874 ; lat. 33° 57' 30" S., long. 151° 39' 15" E.; depth, 120 fathoms ; bottom, green sand. One small, dark-coloured specimen, which has apparently been taken from a crab's back, and seems to be referable to this species. Station 313, January 20, 1876 ; lat. 52° 20' S., long. 67° 39' W.; east of the Strait of Magellan ; depth, 55 fathoms ; bottom, sand ; bottom temperature, 47°-8. Several fair-sized pieces. Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; off the mouth of the Eio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°"2. A large quantity ; some specimens of very large size. Tedania infundibuliformis, Ridley and Dendy (PI. XL fig. 1; PL XXIX. figs. 2, 2a). 1886. Tedania infundihuliformis, Eidley and Dendy, Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 335. Sponge (PL XL fig. 1) erect, lamellar, funnebshaped.' The single specimen present consists of a deep hollow cup, slightly compressed, so as to have an oval section, and ' This, as in Phakellia ventilabrum, is very possibly not a constant cliaracter of the species. REPORT ON THE MONAXONIDA. 55 with moderately thick walls ; it has been broken off at the bottom, so that the nature of the base cannot be determined. Height 72 mm. ; greatest breadth across the mouth about 50 mm.; thickness of walls about G mm., thinning towards the lip. Colour in spirit pale yellow. Texture soft and very fragile. Surface (the specimen is so much washed that it is difficult to determine the true nature of the surface, but it seems to have been fairly smooth both inside and out). Dermal memhrane ; for the most part washed off; thin and transparent. Oscula small, scattered over the inside of the cup (? only). Pores ? Skeleton. — A very loose, slightly fibrous reticulation of rather slender stylote spicules, with bicapitate cylindricals (tylota) rather sparingly arranged at the surface, in irreo-ular tufts or scattered. Spicules. — (a) Megasclera ; (1) slightly curved or straight, slender, smooth styli (PL XXIX. fig. 2), sharply and rather suddenly pointed at the apex; size about 0"54 by 0"015 mm.; (2) slender, smooth, straight tylota (PL XXIX. fig. 2a), much smaller than the styli, with well-developed oval heads often somewhat pointed at the ends, and (?) with slight traces of minute spination. Size about 0"28 by 0"0063 mm. (b) Micro- sclera ; very abundant, long, hair-like rhaphides, very slender, and scattered through the tissues like hair in plaster ; occurring both separately and in loose bundles ; size about 0-35 by 0-002 mm. Had not this species differed somewhat in spiculation, as well as in external form, from those previously described, we should scarcely have ventured to describe it as new ; unfortunately tLere is only a single specimen in the collection ; as regards spiculation the most characteristic feature is the form of the bicapitate cylindricals (tylota) (PL XXIX. fig. 2a) ; the rhaphides are remarkably large, though not coming up to those of Tedania massa. Locality. — Off the south-west coast of Patagonia. One specimen. Tedania actiniiformis, Ridley and Dendy (PL XL figs. 2, 2a; PL XXIX. figs. 1, la). 1886. Tedania actiniiformis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 336. Sponge (PL XL fig. 2) sessile, cylindrical, shape like that of a sea-anemone (whence the specific name) attached by a narrowed base to a fragment of stone; abruptly and flatly truncated above so as to give rise to a broad oval plateau which bears four or five small oscular tubes. On the side of the sponge at a distance of about 2 mm. below the top is a circular zone of pores, about 2 mm. in breadth, running round the sponge (vide fig.). Height of sponge 18 mm.; greatest diameter at top 29 mm., at base 18 mm. Colour in spirit greyish-brown. Texture soft and spongj^ (the specimen is in bad condition). Surface rather uneven. Dermal membrane thin and transparent. Oscula 56 THE VOYAGE OF H.M.S. CHALLENGER. few, rather small, on the summits of small oscular tubes at the top of the spono-e. Pores, observed only in the well-defined "pore-zone" above mentioned, here they are very abundant (PL XI. fig. 2a). Sl-eleton. — The main skeleton is extremely diffuse and irregular, consisting of an almost unispicular reticulation of stylote spicules with no definite spiculo-fibre. Below the pore-zone there is an irregular dermal reticulation of thickly scattered stylote and tornote spicules. In the pore-zone itself one finds the stylote spicules absent from the dermal membrane, and the tornota coming to the surface in thick brushes. Above the pore-zone comes a low wall of thickly packed, and for the most part vertically disposed tornota (lying parallel with the surface), terminating above flush with the top of the sponge.^ On the top itself the dermal reticulation is of the same character as below the pore-zone {i.e., irregular). Spicules. — («) Megasclera; (l) stout, smooth, slightly curved, rather bluntly pointed styli (PI. XXIX. fig. 1), measuring about 0'87 by 0"03 mm., forming the main skeleton and occurring also in the dermal reticulation ; (2) straight, smooth, slightly fusiform, hastately pointed cylincbicals (tornota) (PI. XXIX. fig. la) with rather peculiarly shaped points, size about 0"56 by 0'019 mm., occurring, as usual, at or near the surface, {h) Microsclera ; long, straight, oxeote rhaphides, size about 0"56 by 0'0031 mm., commonly in whisp-like bundles. It seems to us not at all impossible that the rhaphides in this and in other species of Tedania may be in reality only young oxeote (tornote) megasclera; certain it is that they are very different from the " trichodragmata " such as occur in Esperella murrayi, &c. This is a very important and well-characterised species ; it is distinguished from all previously known by its external form and the arrangement of the pores in a definite zone. Its stylote spicule is the largest in the genus. It affords a really splendid instance of the manner in which sponges, which are shapeless masses when occurring in shallow water, assume in abyssal depths (in this case 2160 fathoms) a definite, symmetrical external form ; this is its chief interest, for the species of the genus hitherto known, from comparatively shallow water, are, par excellence, amorphous sponges.^ Unfortunately there is only one specimen in the collection and that in bad condition, so that we are unable to give any details as to the minute anatomy. LocaZ%.— Station 299, December 14, 1875 ; lat. 33° 31' S., long. 74° 43' W. ; west of Valparaiso; depth, 2160 fathoms; bottom, blue mud; bottom temperature, 35°'2. One specimen. ' This arrangement gives rise to a faint vertical striation on the surface of the sponge above the pore-zone {vide fig. 2). It is also present to a less extent just below the pore-zone. 2 We must make an exception to this statement in the case of Tedania infundibuliformis, nobis, while Tedania massa, nobis, is an amorphous form ranging down to 600 fathoms. E.EPOET ON THE MONAXONIDA. 57 Genus Trachytedania, Eidley (PI. XXIIL). 1881. Trachytedania, Ridley, Proc. Zool. Soc. Lond., January 4, 1881, ji. 122. Styli, in part or all, spined. The first species of the genus was obtained by the " Alert " in Portland Bay, Chili ; in it the characteristic spination of the stylote spicule, which distinguishes this genus from the closely allied genus Tedania, is only very faintly marked. The Challenger adds one new species, from off the south-west coast of Patagonia, in which the spination of the stylus is very strongly marked ; it is, however, an open question whether this character is in itself sufficient to separate the two genera ; Vosmaer,^ however, accepts the genus Trachytedania as distinct. We have found it advisable to make some alterations in the generic diagnosis. Trachytedania patagonica, Eidley and Dendy (PI. XXIIL figs. 6, 6«, Gh, Gc). 1886. Trachytedania patagonica, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 336. Sponge represented by two small, irregularly shaped specimens ; massive, amorphous. Length of largest fragment about 37 mm. Colour in spirit pale yellow. Texture soft and crumbling. Surface uneven, with slight traces of hispidation. Dermal membrane thin but distinct. Oscula (?) small, scattered. Pores scattered; in some parts of the dermal membrane almost or entirely wanting, in other parts fairly abundant. Skeleton. — Very loose and irregular, consisting of a somewhat " Isodictyal " reticula- tion of spicules, chiefly of sf)ined styli, but with a small admixture of oxea (toruota). No distinct fibre. Immediately beneath the dermal membrane are a number of irregular, divergent tufts of oxeote (tornote) spicules, while similar spicules are scattered through the dermal membrane itself, forming, together with some of the stylote spicules, a very sparse and irregular dermal reticulation. Spicides. — (a) Megasclera; of two kinds — (l) Eather stout, slightly curved, spined styli (PI. XXIIL figs. 6, 6a, G6) ; spined all over, except perhaps at the extreme apex ; but with the spination most marked at the base ; size about 0"35 by 0'0125 mm. (2) Straight, short oxea (tornota) (PI. XXIIL fig. 6c), stoutest towards the centre and tapering gradually to a slightly hastate point at each end; size about 0'245 by O'OOZ mm. Both the form of these spicules and their position in the sponge show that they are homologous with the " bicapitate cylindrical " (tylota) in the typical Tedania, and with the corresponding " cylindricals " of Trachytedania spinata, Eidley.- (h) Ilicro- sclera ; very fine, scattered, oxeote rhaphides, slightly thicker at one end than at the other; length about 0"2 mm., often (? always) spined near the Ijroader end with fine spines directed towards that end of the spicule. 1 Porifera, Bronn's Klasi?. u. Ordn. d. Tliierreiehs, p. 340. " Prnc. Zool. Soc. Loncl, .January 4, 1881, p. 123. (zoOL. CHALL. EXP. — PART Lix. — 1887.) Nnn 8 58 THE VOYAGE OF H.M.S. CHALLENGEE. The species is distiuguislied from Trachytedania sjiinata, Ridley, the only species hitherto known, by the superior dimensions of all the spicules, by the distribution of the spines practically all over the stylote spicules, and by the fine spination of most of the rh aphides. Localities.— ^taXion 308, January 5, 1876 ; lat. 50° 8' 30" S., long. 74° 41' 0" W. ; ofi" the south-west coast of Patagonia; dcj^th, 175 fathoms; bottom, blue mud. One small specimen. Station 308 (?) or 311 (1), off the south-west coast of Patagonia. The larger specimen. Subfamily 4. Desmacellin^. 1886. Desmacellina, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 336. Megasclera all monactiual, stylote to tylostylote. Microsclera sigmata or toxa or both. Genus Desmacella, Schmidt. 1870. Desmacella, Schmidt, Spong. Atlant. Gebiet., p. 53. 1870. ? Desmacodes, Schmidt, Spong. Atlant. Gebiet., p. 54. This being the sole genus of the subfamily the diagnosis is the same. Schmidt's original diagnosis {Joe. cit.) runs thus: — " Spongien, welcher ausser den gestreckten cinfachen Nadeln nur Bogen oder Spangen besitzen. Die Nadeln entweder in iindeutlichen Ztlgen oder faserig geschichtet." With regard to Desmacodes he states (loc. cit.) of Desmacodes subereus, the type of the genus, that it unites the habit of Papillina suherea with the spicules of Desmacella, and describes the spicules as " Spin- deluadeln," " Stifte," " Stecknadeln " and " Spangen," but his description is very scanty. Vosma3r ^ limits the name Desmacella to species with diancistra or " trenchant biliamate" spicules {Vomerida of this Report, &c.), but subsequently" places these under Hamacantlia, Gray, stating that Desmacella pumilio and Desmacella vagabunda of Schmidt, the types of the genus Desmacella, could easily be referred to Desmacodes (although he proceeds inconsistently to place Desmacella vagahunda under Gellius directly afterwards). But the name Desmacella has the advantage of priority over Desmacodes ; and therefore, while we agree with Vosmser in keeping this little group of species distinct from Gellius and its allies on the one hand and Vomerida and its allies on the other, we retain the name originally conferred upon them. Desmacella are not commonly met with. A species is described by Carter in the " Porcupine " Report as Desmacella pumilio of Schmidt, but as yet very little is known about the group. 1 The Family of the Desmacidinida^, Notes from the Leyden Museum, vol. ii. p. 109. 2 Sponges of the " Willem Barents " Expedition, 1880-81, p. 28. REPORT ON THE MONAXONIDA. 59 Desmacella annexa, Schmidt. 1870. Desmacella vagahunda, var. annexa, Schmidt, Spong. Atlant. Gebiet., p. 53. 1874. Desmacella pumilio, Carter, Ann. and Mag. Nat. Hist., ser. 4, vol. xiv. p. 250. 1880. Desmacodes vaffahundics, var. annexa, Vosmaer, Notes from the Ley den Museum, vol. ii. p. 108. AVitli this species we identify a small fragment from Station 24. The most characteristic feature of the species is the presence of numerous very fine toxa, measuring 0"09 by O'OOl mm. These toxa also occur in Schmidt's type specimen, as attested by a preparation in the British Museum, although Schmidt and Vosmaer both refer to them simply as " acerates." The identification of Mr. Carter's species was made after a comparison of his account with the sUdes representing Schmidt's species in the British Museum. ioca%.— Station 24, March 25, 1873 ; lat. 18° 38' 30" N., long. 65° 5' 30" W.; West Indies ; depth, 390 fathoms ; bottom, Pteropod ooze. Habitat. — Florida (Schmidt); English Channel (Carter); West Indies (Challenger). Subfamily 5. Hamacanthin^. 1872. Hamacanthidx, Gray, Ann. and Mag. Nat. Hist., ser. 4, vol. ix. p. 449. 1886. Voviefi-idinx, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 337. Megasclera oxea or styli. Microsclera large diancistra, to which others may be added. We 23ropose to recognise in this subfamily two genera, according as the megasclera are oxeote (ranging to strongylote) or stylote. In the one genus, Hamacantha, Gray,^ the only species yet known to us is Bowerbank's Hymedesmia johnsoni,^ a species which appears to have been never yet described, although Dr. Bowerbank figures the characteristic diancistra (loc. cit.). Examination of the preparations in the British Museum has shown us that the megasclera are oxeote. The second genus is Vomerida, Schmidt, in which the megasclera are stylote. Unfortunately the genus Hamacantha is not represented in the Challenger collection. Genus Vomerula, Schmidt (Pis. XII., XVII.). 1874. Haliclwndria, pars, Bowerbank, Mon. Brit. Spong., vol. iii. p. 208. 1880. Vomerula, Schmidt, Spong. Meerb. von Mex., p. 82. 1885. Hamacantlia, pars,\osma.ei; Sponges of the " Willem Barents" Expedition, 1880-81, p. 28 {sep) ; nee id., Bronn's Klass. u. Ordn. d. Thierreichs, Porifera, p. 352, wee Gray, Proc. Zool. Soo. Lond., 1867, p. 538. Megasclera styli. This genus has been wrongly characterised by Vosmaer as distinguished by the 1 Proc. Zool. Sac. Lond., 1867, p. 538. ^ jzirf^ Mon. Brit. Spong., vol. i. figs. 112, 293. 60 THE VOYAGE OF H.M.S. CHALLENGER. presence of " anchorate " microsclera (chelse)/ while Schmidt's typical species of the genus, Vomerula tencla,^ does not possess these; on the other hand, the genns differs from Hamacantha, Gray, in a point unnoticed by Vosmaer, viz., the possession of well- marked stylote megasclera, and on this account the two genera Hamacantha and Vomerula may at present remain separate. This difference is, however, perhaps not so important as would at first sight appear, as the so-called oxeote of Hamacantha johnsoni is frequently slightly blunted at one (or both) ends, and therefore seems to be possibly derived from a stylote spicule or to have given rise to that of Vomerula ; (this tendency to bluntness is indicated, but perhaps too strongly, in Bowerbank's original figure).^ On the other hand, there is one preparation in the Bowerbank collection, from a species identified by him with Hamacantha johnsoni, in which the oxeote spicules are well pointed at both ends, with no noticeable tendency to blunting. This greatly supports us in distinguishing two genera of the subfamily. Vomerula esperioides, Eidley and Dendy (PL XII. fig. 1; PI. XVII. figs. 2, 4, 12). 1886. Vomerula esperioules, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol xviii. p. 337. Sponge (PL XII. fig. 1) large, erect, forming tall, thick, leaf-like expansions, attached by the base, reaching about 250 mm. in height by about 50 mm. or more in breadth and 25 mm. in thickness. Colour in spirit pale yellow. Texture tough and strono-, owino- to the very coarse fibres of the skeleton ; but the interior is cavernous. Surface uneven, with numerous small prominences (conuli) caused l»y the projecting ends of the underlying fibres of the skeleton. Supported upon, and stretched tightly between these prominences is a thin transparent dermal membrane, covering over the large, irregular, subdermal cavities. The dermal membrane (PL XVII. fig. 12) contains very numerous round pores, 0"07 mm. in diameter, which lead from the exterior into the subdermal cavities ; the boundaries of these pores are supported by numerous small sio-mata arranged around them. The dermal membrane is further supported by a reticulation of fine fibres, yet easily visible to the naked eye, which gives a highly characteristic appearance to the surface. The oscida are placed on the summits of small, conical, bladder-like papiUae (PL XII. fig. 1), bounded only by the dermal mem- brane, in which there are here no pores. Over these papillae the dermal reticulation of fine fibres is absent and its place is taken by a single layer of closely placed spicules, in which the styli are mostly arranged longitudinally and more or less parallel v/ith one another. 1 Bronn's Klass u. Ordn. A. Thierreichs, Porifera, p. 352. ' Spong. Meerb. von Max., 1880, p. 82. 3 Mon. Brit. Spong., vol. i. pi. xviii. fig. 293. REPORT ON THE MONAXONIDA. 61 Skeleton. — (a) Dermal; a reticulation of spiculo-fibre (PI. XVII. fig. 12) varying from one to nine or ten spicules in breadth, and composed of stylote spicules. Over the osculum-bearing pajjillae this reticulation is replaced by a single dense layer of closely packed stylote spicules, mainly placed longitudinally and more or less parallel with one another, {li) Main; composed of a coarse reticulation of thick, stout spiculo- fibre, composed of stylote spicules like those of the dermal skeleton, but usually a little shorter. Spicules. — (o) Megasclera; smooth styli (PL XVII. fig. 4) tapering to a not very sharp point at the apex and also slightly towards the base, measuring about 0"7 by 0*0 19 mm. {h) Microsclera ; two kinds are present — (1) numerous large, contort diancistra, measuring 0"1 77 mm. long by 0"019 mm. in breadth of shaft. The exact shape of these spicules will be best understood from the figure (PI. XVII. fig. 2, a, h). They occur both in the deeper tissues and in the dermal layer. (2) Very numerous, small, contort spicules, many of which appear to be young forms of the large diancistra (PI. XVII. fig. 2, d), while others (PI. XVII. fig. 2, ) Main; a rather irregular but compact reticulation of dense spiculo-fibre, in which one may distinguish main fibres running vertically to the surface of the sponge, and secondary fibres crossing these at right angles. As the main fibres approach the surface the spicules composing them spread . REPORT ON THE MONAXONIDA. 69 out into broad tufts, and their ends penetrate into the dermal layer and project very slightly beyond the surface ; it is generally the pointed ends of the styli which thus project. The peculiar arrangement of the spicular tufts along the edges of the pore- bearing cracks has already been described. Spicules. — (a) Megasclera ; of one kind only, viz., styli (PI. XIII. fig. 18), very nearly straight, tapering from near the middle towards both ends, and measuring about 07 by 0"019 mm. (b) Microsclera; these are enormously abundant, both in the dermal membrane, including the pore-bearing areas, and in the deeper tissues. Three kinds occur, all plentiful: — (l) Large palmate anisochelse (PI. XIII. fig. 17); especially abundant in the dermal membrane, up to 0"072 mm. long, with the large end 0"019 mm. broad. These chelate spicules are often found in very beautiful rosettes, which seem to occur chiefly, if not solely, just beneath the dermal membrane. Numerous small anisocheljB are also present, perhaps young forms of the large ones. (2) Numerous sigmata (PI. XIII. fig. 11) of the usual shape, frequently much contorted, measuring about 0'053 by 0*0024 mm. (.3) Trichodragmata (PL XIII. fig. 14), occurring in great profusion in the deeper tissues, measuring about 0'076 by 0"013 mm. For further details as to the minute anatomy and histology the reader is referred to the Introduction (Anatomy and Histology). This is one of the finest and most interesting sponges in the collection, and forms a splendid example of the manner in which the pores may be collected into certain definite " j^ore-areas," a phenomenon which has already been noted by SoUas^ for other species of the genus Esperella. But we have here a still further development, for not only are the pores, for the most part, definitely localised, but the grooves in which they occur may be opened or closed by appropriate muscles, and thus the supply of water regulated with great precision. The transverse muscle-fibres of the pore- bearing cracks may be compared with the fibres which sometimes form sphincters around the indiAddual pores ^ in other sponges. In the maeandering pore-areas and the almost stony spicular and non-reticulate dermis this species stands markedly apart from all other described species, so far as they are known to us ; the spiculation, on the other hand, as often happens in such cases, is of an ordinary type. Locality. — Ofi" Port Jackson, 30 to 35 fathoms. One specimen. Esperella porosa, Eidley and Dendy (PI. XV. figs, 6, 9, 17 ; PI. XVI. fig. 5). 188G. Esperella porosa, Ridley and Dendy, Ann. and Mag. Nat, Hist., ser. 5, vol. xviii. p. 338. Sponge (PI. XVI. fig. 5) cylindrical ; length of largest specimen about 50 mm. Diameter about 6 mm. Colour in spirit nearly white. Texture fibrous, but open and 1 Vide Ann. and Mag. Nat. Hist., ser. 5, vol. ix. p. 437. ' Ficfe Vosmaer, Bronn's Klass. u. Ordn. des Thierreichs, Porifera, p. 125. 70 THE VOYAGE OF H.M.S. CHALLENGER. rather soft. Surface even but minutely hispid ; with a porous appearance due to the close reticulation of the dermal skeleton. Dermal membrane (1) (nearly all rubbed off). Pores {Tj. Oscula; one specimen shows two small, conical, tubular projections, each with an osculum at the summit (PL XVI. fig. 5, o). Skeleton. — (a) Dermal ; a very close and compact reticulation of dense spiculo-fibre, the meshes of the network being about 0"3 mm. and the fibres themselves about 0"08 mm. in diameter. Numerous spicules project more or less vertically from the fibres, thus causing the hispid appearance of the surface, (b) Mam; a loose reticulation of dense, stout spiculo-fibre, which branches and anastomoses. As they approach the surface the fibres expand into brushes of spicules upon which the dermal layer rests. The spicules composing the fibres are united together by a fair proportion of spongin. Spicules. — (a) Megasclera ; of one kind only, viz., tylostyli (PI. XV. fig. 6), stoutest in the middle, with a small head at one end, separated by a slight constriction from the shaft; tapering rather suddenly to a sharp point at the apex. Size about 0'38 by 0"016 mm. (6) Microsclera ; of two kinds; (1) palmate anisochelse (PL XV. fig. 17), rather small for an Esperella, about 0"05 mm. long; with a long, narrow palm at the large end ; (2) large, smooth, simple and contort sigmata (PL XV. fig. 9), measuring about 0-16 by 0-0085 mm. This species may he most readUy recognised by the character of the dermal skeleton reticulation. It is represented in the collection by only two fragments, one of which has an interesting and peculiar habit. It has grown around the stem of an Alcyonarian, which it has covered with a very thin horny layer containing the proper spicules of the sponge, the tylostyli being mostly placed longitudinally. This horny coating is easily separable from the Alcyonarian stem, and forms by itself a hollow cylinder from which the skeleton fibres arise. This species appears to resemble Esperella velutata, Lbkn. sp., as far as spiculation is concerned, but the descriptions given of the latter are too imperfect for identification and the great difference in locality, the one being found at Port Jackson and the other at Venice, renders it improbable that they are the same species.^ Locality. — Off Port Jackson, 30 to 35 fathoms. Two small specimens. Esperella nuda, Ridley and Dendy (PL XV. figs. 5, 11, 14 ; PL XVI. fig. 1). 1886. Esperella nuda, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 339. Sponge (PL XVI. fig. l) encrusting an irregular mass of dead spiculo-horuy fibre belonging to an Esperella, probably of the same species. Sometimes the naked amber- coloured fibre projects far beyond the surface of the sponge. Colour in spirit pale yeUow. 1 For Esperella velutata, vide O. Schmidt, Spongien des Adriatischen Meeres, p. 57; also Vosmaer, Notes from the Leyden Museum, vol. ii. p. 141. REPORT ON THE MONAXONIDA. 7X Texture soft. Surface minutely hispid. Dermal memhrane distinct, thin, transparent. Pores collected into groups, forming small, more or less circular pore-areas, each containing up to about twelve pores, but generally less. The groups themselves are placed close together, scattered over the surface of the sponge. Diameter of the pore-areas about 0'3 mm., of the pores about 0'07 mm. The pores are placed very close together within the pore-areas, being separated only by narrow strands of tissue. Skeleton. — The mass of fil)re on which the sponge is based, and which it encrusts, is coarse, amber-coloured, branching and anastomosing, with a very large proportion of horny matter and a core of tylostylote spicules of the characteristic Esj^erella type. This fibre was certainly dead before the present sponge grew over it, as it projects far beyond the soft j^arts, and the central canals of the enclosed spicules are enlarged ; it appears, however, probable that it belongs to an older specimen of the same species. The specimen has in addition its own proper fibre. There is, however, no dermal reticula- tion. The deeper skeleton consists of a very loose reticulation of spiculo-fibre, very densely spiculous and with very little horny matter. The main fibres are very broad, and as they approach the surface they subdivide and subdivide again, forming spreading tufts of fibres, each ultimate fibre of which breaks up into a brush of separate spicules whose ends project freely beyond the surface of the sponge, penetrating the dermal membrane. The proper fibre of the sponge is probably thus poorly developed owing to the fact that it makes use of the dead skeleton of another sjDonge. Spictdes. — (a) Megasclera ; of one kind only, viz., tylostyli (PI. XV. fig. 5), with small head and slightly constricted neck, tapering rather suddenly to a very sharp point and broadest near the apex; size 0"24.5 by 0'016 mm. The spicules mentioned above as occurring in the horny fibre on which, the sponge is based are of about the same length but slenderer, (b) Microsclera ; (1) rather small palmate anisochelae (PL XV. fig. 11), with a long, narrow palm at the large end; length of spicule about 0'025 mm.; (2) rather large, smooth, simple and contort sigmata (PI. XV. fig. 14), measuring about 0'12 by 0"0063 mm. The chelate spicules seem to be almost confined to the dei-mal membrane and are not very abundant even there, while the sigmata are plentiful, both in the dermal layer and in the deeper tissues. The sigmata are often found in groups, in which the spicules are placed close together and parallel with one another. A noteworthy feature about this species is the absence of a dermal skeleton reticu- lation. In spiculation it agrees closely with Esperella porosa, nobis. The habit of the sponge is very remarkable, the specimen in question having made use of an old and dead skeleton, perhaps of the same species. Locality. — Off Bahia, shallow water. One specimen. 72 THE VOYAGE OF H.M.S. CHALLENGER. Esperella fusca, Ridley and Dendy (PI. XIV. fig. 4 ; PI. XV. figs. 3, 3o, 15, 15a). 1886. Esperella fusca, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 339. Sponge (PI. XIV. fig. 4) small, incipiently lobose, rounded, penetrated through and through by the branches of some Alga. Greatest diameter rather under 25 mm. Colour in spuit dark greyish-brown. Texture soft, resihent. Surface very slightly roughened by the presence of the dermal skeleton reticulation, but not hispid; uneven, with rounded elevations and depressions. Dermal membrane distinct, transparent. Oscula few, small, with their margins produced into small elongated tubes. Pores few, small, scattered ; diameter about 0'025 mm. Skeleton. — (a) Dermal; composed of two layers: — (1) an upper layer, consisting of a rather close, irregular spicular reticulation, in which several spicules often lie side by side forming rudimentary fibres ; and (2) a lower layer, consisting of a reticulation of fairly stout, compact, spiculo-fibre, with rounded or polygonal meshes. (&) Main; composed of loose, branching spiculo-fibre. The main fibres do not, as in so many Esiyerellw, break up into distinct divergent brushes of spicules on approaching the surface, hence the surface is not hispid. Spicules. — {a) Megasclera; of one kind only, viz., tylostyli (PL XV. figs. 3, 3a), often curved, with distinct oval heads and usually much blunted at the apex ; size about 0'455 by 0'0126 mm. (h) Microsclera; (l) a few palmate anisochelse (PL XV. figs. 15, 15a), of the usual Esiierella type (sometimes in rosettes?); length about 0'063 mm. ; numerous smaller ones also occur, probably young forms of the larger, (2) Small, slender sigmata, usually contort, length about 0"044 mm. (3) Trichodragmata, forming small, compact, oblong bundles, measuring about 0'03 by 0*063 mm, ; especially abundant in the dermal layer. This sponge is remarkable for the ease with which the flagellated chambers can be made out. If a fragment, without any staining, be teased up in a drop of Canada balsam, the flagellated chambers are separated out in enormous numbers from the surrounding matrix, and appear under a low power of the microscope as spherical or oval bodies, composed of aggregations of large granules. Further details concerning them will be found in the Introduction. They are about 0'03 mm. in diameter. Locality. — Ofi'Bahia; depth, 17 fathoms. Four specimens. Esperella arenicola, Ridley and Dendy (PL XV. figs. 4, 4a ; PL XVI. fig. 8). 1886. Esperdla arenicola, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol xvui. p. 339. Sponge (PL XVI. fig. 8) massive, flat, cake-like. Length of largest piece 175 mm. ; breadth 81 mm. ; thickness 16 mm. Colour in spu-it light brown. Texture fragile, REPORT ON THE MONAXONIDA. 73 brittle, extremely sandy. Surface rougli and uneven, owing to the numerous foreign bodies, Mollusc shells, Polyzoa, &c., with which the s^^onge is thickly encrusted. Dermal membrane thin and transparent. Oscula scattered, small and round, with their margins level with the general surface of the sponge, about 2 '5 mm. in diameter. Skeleton. — There is no dermal reticulation at all. The main skeleton is very loose. The primary fibres are thin, and give off branches at very sharp angles, recalling the ramification of the Broom. As they approach the surface of the sponge the primary fibres break up into branches, each of which terminates in a sHghtly divergent tuft of long, slender spicules which project little, if at all, beyond the dermal membrane. Spicules. — (a) Megasclera; of one kind only, viz., tylostyli (PI. XV. figs. 4, 4a), long and very slender, each with a distinct head, and tajaering gradually to a fine point at the apex; size about 0*4 by 0'0072 mm. (b) Microsclera ; (l) small, slender, palmate anisochelse, about 0"028 mm. long ; (2) rather large, simple and contort sigmata, measuring about 0"077 by 0'0048 mm. ; (3) trichodragmata, size about 0*35 by 0"07 mm. As already noticed, this sponge is very full of foreign bodies, and this renders it difficult to obtain a satisfactory section. Indeed it seems as if the poor development of the skeleton might be due to its partial replacement by foreign bodies. If we imagine the process of degeneration proceeding a few steps further, we arrive at the condition of Marshall's Phoriospongi^, the real nature of which has been discussed elsewhere. The small size of the chelae and the absence of a dermal skeleton are good characters of this species. There can be little doubt that the modifications in the skeleton are due to the sponge having lived on a shelly bottom. Locality.— Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' 0" E. ; off East Moncoeur Island, Bass Strait ; depth, 38 fathoms ; bottom, sand and shells. One large •specimen and some fragments. Usperella simonis, Ridley and Dendy (PI. XV. figs. 7, 7a, 12, 12a, 13, 16, 16a, 16?>, 16c). 1886. Esperella simonis, Ridley and Dendy, Ann. and Mag. Nat Hist., ser. 5, vol. xviii. p. 339. Sponge branching, cylindrical (but rather angular), or more or less massive. Diameter of branches about 8 mm.; length of largest fragment 45 mm. Texture fibrous, elastic. Surface somewhat uneven and minutely hispid. Dermal membrane thin, transparent, distinct. Oscula (?) small, scattered. Pores distinct, numerous, round openings in the dermal membrane, about 0'063 mm. in diameter; irregularly scattered, but very close to one another. Skeleton. — (a) Dermal; a rather irregular reticulation of spiculo-fibre, in some places very much closer than in others, the extra closeness being due to numerous loose spicules irregularly scattered between and across the fibres. In the looser parts, the (ZOOL. CHALL. EXP. — PART LIX. — 1887.) Nil 10 74 THE VOYAGE OF H.M.S. CHALLENGEE. dermal reticulation is seen to be arranged in a rotulate manner, similar to the dermal reticulation of Bowerbank's Desmacidon rotalis ^ ( = Esperella rotalis), but Hardly so regular as shown in his figure, (h) Main; it is difficult to make out any definite arrangement. In the cylindrical pieces there appears to be a central, rather loose core of thick spiculo-fibre ; one can also distinguish thick bands of spiculo-fibre running more or less vertically to the surface ; on approaching the surface the fibre opens out slightly and the points of the spicules composing it project a little way beyond the dermal membrane, thus causing the minute hispidation already noticed. The spiculo- fibre of the main skeleton is generally dense and thick. Spicules. — (a) Megasclera; of one kind only, viz., tylostyli (PL XV. figs. 7, 7a), tapering towards the base so as to form a neck and then enlarging slightly to form the small head ; towards the apex the spicule tapers very gradually to a sharp point ; size about 0'4 by 0"0145 mm. {h) Microsclera; (1) large and peculiarly-shaped, palmate anisochelse (PI. XV. figs. 16, 16a, 166, 16c), measuring about 0'072 mm. in length, with the large end 0"036 mm. wide; for further particulars as to their shape see the figures [loc. cit.). These spicules are frequently seen to echinate the skeleton fibre, being themselves arranged in groups, with their truncated small ends placed close together upon the spiculo-fibre and their large ends radiating outwards (perhaps the truncation of the small ends may be due to this peculiar habit). Plenty of the large anisochelse are also found lying freely in the tissues. Numerous smaller anisochelse also occur, which difi"er from those just described in having the large end relatively of greater size ; they difler also slightly in general shape, but they have the cup-shaped, truncated small end found in the larger ones, and we are inclined to think that they are young forms of the latter {vide PI. XV. fig. 16c). There are also some minute anisochelse, probably still younger forms of the same. Neither of the small forms appear to be attached to the skeleton fibre, as the large ones are. (2) Very large, thick, smooth, contort sigmata (PL XV. fig. 13), something like the diancistra found (e.g.) in Vomerula esperioides, nobis, but without the membranous expansions and with no notch in the centre of the shaft; size 0*24 by 0"019 mm. These spicules are also very often found adhering, usually by their backs, to the skeleton fibre, which thereby acquires a very formidable appearance. (A similar arrangement of the large diancistra may be seen in Vomerida esperioides, nobis, and in Hamacantha johnsoni, Bk., sp.) (3) Very fine, single, scattered, smooth toxa (PL XV. figs. 12, 12a), measuring about 0"145 by 0-003 mm. This is a highly interesting species ; unfortunately, it is represented in the collec- tion by only a few fragments. With, so far as we can judge, a very insignificant external form, it combines a spiculation which in beauty and variety is hardly sur- passed among the known Monaxonida. One of the most remarkable features about it ' Mon. Brit. Spong., vol. iii. p. 327, pi. sc. figs. 8-14. REPORT ON THE MONAXONIDA. 75 is tlie way in which two out of the three microsclera attach themselves to the skeleton fibre. It is a rather remarkable fact that we can find no spicules which we can put down as young forms of the large sigmata, except a very few, which somewhat resemble them in shape and size but are much more slender, and appear as if they might form links connecting the large with the ordinary, small sigmata. Locality. — Simon's Bay; depth, 10 to 20 fathoms. Three fragments. JEsperella hiserialis, Ridley and Dendy (PI. XIV. figs. 2, 3 ; PI. XV. figs. 8, 8a, 86). 1886. EspereUa hiserialis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 340. Sponge (PI. XIV. figs. 2, 3) consisting of a long, slender axis, perfectly straight and somewhat flattened in one plane, from which arise very short, slender, straight spicular processes. These processes are arranged in two opposite series, at either end of the longer transverse diameter of the sponge, forming on each side a single long row of closely placed, hair-like projections. ■ The axis of the sponge is somewhat twisted, so that the two rows are not quite straight. It is covered with a thin crust of brown, transparent tissue, containing many spicules, and terminates abruptly at the lower end, having apparently been broken oif short, while at the upper extremity it ends in a slightly rounded apex. Length of sponge 92 mm. ; longer diameter 1 mm. Surface hispid. Dermal mem- hrane fairly distinct. Skeleton. — ((() Dermal; from an imaginary line, drawn longitudinally down the middle of each flattened side of the sponge, a large number of tylostylote spicules diverge more or less at right angles towards either side. These spicules have their bases placed along the middle line, while their apices are directed towards the bases of the hair-like processes, which they are just about able to reach. This arrange- ment is easily recognisable, though not very regular, for many of the spicules are placed so as to project from the surface of the sponge, giving to it the hispid appearance described. (6) Main; a central, dense spicular axis composed of very long styli firmly united together, from two sides of which come ofl" short, hair-like fibres as already described (the spicular processes). Outside the axis there are also longitudinal bands of spiculo-fibre, apparently not arranged in any very definite manner. Spicules. — (a) Megasclera; of two kinds, (l) Very long, slender styli (PI. XV. figs. 8a, 86), of extremely elegant shape, becoming gradually inflated in the centre, but very thin at both ends. Apex not very sharply pointed ; no head. These spicules may reach over 2 mm. in length, with a diameter in the middle of 0"038 mm.; they occur only in the main axis. (2) Rather short tylostyli (PL XV. fig. 8), broadest near the apex, each hastately pointed and with a distinct, elongated head narrowing towards the top; size about 0'44 by O'Ol mm.; these occur in the dermal skeleton, in the longitudinal fibres outside the axis, and in the hair-like projections. (6) Microsclera; 76 THE VOYAGE OF H.M.S. CHALLENGER. (1) very minute palmate anisoclielse, length about 0'0126 mm ; (2) small, slender sigmata, often contort; length about 0'025 mm. This species forms a most interesting and important link between the two oenera Esperella and Cladorhiza, especially as regards external form. It is also particularly interesting in that it exhibits a distinctly bilateral symmetry ; in this respect it is approached by Cladorhiza pennatula, Schmidt;^ the latter has, however, a much coarser and less delicate form. The species may be readily recognised by the very minute microsclera, the extraordinary external form, and the two kinds of megasclera, though the latter are both, doubtless, merely modifications of one and the same type. Localities.— Btation 281, October 6, 1875 ; lat. 22° 21' S., long. 150° 17' W. ; South Pacific ; depth, 2385 fathoms ; bottom, red clay ; bottom temperature, 34°'9. One specimen, which still retains the soft parts. Station 291, October 27, 1875; lat. 39° 13' S., long. 118° 49' W. ; South Pacific; depth, 2250 fathoms ; bottom temperature, 34°'6 ; bottom, red clay. One specimen, with the soft parts washed off", but still retaining a few microsclera and exhibiting very beautifully the biserial arrangement of the lateral processes.^ Esperella sp. We have to record the occurrence of a single specimen of an Esperella (species unknown) from Bass Strait, which is growing over the fibre of a Dysidean sponge ; both the tylostylote megasclera and the anisochelse are very much reduced, and we have seen no other kind of spicules. Under these circumstances it has not seemed to us desirable to found a new species for the reception of this sponge. The tylostyli are smooth, straight, slender, sharp-pointed, with slight oval heads, and commonly with much enlarged central canals ; size about 0-28 by 0-0042 mm. The palmate anisochelse are of very delicate appearance and about 0"022 mm. long. ZocoZ%.— Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' E. ; Bass Strait ; depth, 38 fathoms ; bottom, sand and shells. One specimen. Genus Esperiopsis, Carter (Pis. XVIIL, XIX., XXVL, XXX., XLVL). 1882. Esperiopsis, Carter, Ann. and Mag. Nat. Hist., ser. 5, voL ix. p. 296. Form various, amorphous or symmetrical. Megasclera all monactinal, styli or tylostyli, smooth ; microsclera isochelse, to which sigmata may be added. 1 Jdhresb. Comm. wiss. Untersvch. deutsch. Meere, Jahrg. ii., iii. p. 119, pi. i. figs. 14, 15. ^ Lodged right witliin the spicular axis of one of the specimens were found a number of developing embryos, con- cerning which further details will be found in the Introduction. For this interesting discovery we are indebted to the careful observation of onr artist, Mr. P. Higbley. REPORT ON THE MONAXONIDA. 77 This genus differs from Esj^erella only in having the chelate spicule equal-ended instead of unequal-ended. The original type of the genus is EsjJeriopsis villosa, Carter, a " Porcupine " sponge/ first described under the name Esperia villosa, and this has been hitherto the sole species included therein. The Challenger adds six new species, and we must also include one old one, namely, Esperiojms edwardii, Bowerbank, sp., represented in the collection by a new variety. The synonymy of the latter sponge will be found under the species. Several of the species of this genus are remarkable for their well-defined external forms, which serve as excellent guides l)y which to separate them from one another and from others of the genus. The distribution of the genus is very wide, 1joth vertically and horizontally, but it appears to prefer deep water in temperate or boreal seas. Esperiopsis symmetrica, Ridley and Dendy (PI. XIX. figs. 6, 6', 6o, 66, 6f; PI. XXVI. figs. 4, 4a ; PI. XL VI. fig. 7). . 18S6. Esperiopsis STjmmetrica, Ridley and Dendy, Ann. and. Mag. Nat. Hist., ser. 5, vol. xviii. p. 340. Sponge (PL XXVI. figs. 4, 4a) erect, straight, slender, cylindrical, unbranched (in so far as evidenced by the Challenger specimens), covered with numerous, long, slender sj)icular tufts, giving it the appearance of a bottle brush. The largest specimen is 44 mm. in length by 4 mm. in diameter (including the projecting processes, which are themselves about 1 mm. in length). Colour in spirit dark chocolate-brown. Texture fibrous and soft between the tufts of spicules. Surface closely beset with the projecting, seta-like processes already mentioned. [Oscula and Pores unknown.) Skeleton. — Well seen in transverse or longitudinal section (PI. XLVI. fig. 7). Radiately arranged. In transverse section the sponge is circular, and from its centre radiate numerous loose bands of spiculo-fibre, after the manner of the spokes of a wheel ; these fibres (PI. XLVI. fig. 7, d) project far beyond the surface of the sponge, thus causing the hairy appearance. At a short distance within the circumference of the sjionge itself is a regular circle, formed by the cut ends of longitudinal bands of spicules; each group of cut ends occupying one of the segments between two- radiating fibres. At a short distance outside this circle, just beneath the surface of the sponge, is another similar but much less regular circle of cut spicules. The longitudinal fibres of the skeleton are confined almost exclusively to these two sheaths, placed one within the other; the inner (PI. XLVI. fig. 7, h) Ijeing fairly compact and well defined, but the outer (PL XLVI. fig. 7, c) irregular and diffuse, while sometimes the two run into one another. There are no definite longitudinal fibres in the centre of the sponge, but a few loose, more or less longitudinally placed spicules. 1 Ann. and Mag. Nat. Hist, ser. 4, vol. xiv. p. 213, pi. xui. figs. 13-15 ; pi. xv. fig. 36. 78 THE VOYAGE OF H.M.S. CHALLENGER. Spicules. — (a) Megasclera; of one kind only, viz., long, smooth, slender, fusiform styli (PI. XIX. figs. 6, 6'), size variable, about 0'8 by 0"028 mm., usually, however, slenderer, {h) Microsclera; of two kinds; (1) large palmate isochelse (PI. XIX. figs. 6^, 6c), of the Esperella type, only with equal instead of unequal ends, length about 0"037 mm.; (2) very minute and slender sigmata, usually much contort, enormously abundant, length about 0"013 mm.; (3) much larger, very slender sigmata (PI. XIX. fig. 6a). This is a very remarkable sponge, the most noticeable feature in which is the radiately symmetrical arrangement of the skeleton. In external appearance it somewhat resembles a Cladorhiza. In one of the specimens a great number of embryos (PL XLVI. fig. 7, e) were found just beneath the surface of the sponge, lying, for the most part, between the two sheaths of longitudinal fibres. As the dark brown colour of the specimens is shared by other things in the same bottle it is possibly due to a discoloration of the spirit in whicli they were j^reserved; it should be noted that the tissues, even to the centre of the sponge, are all diffusely stained with the same rich chocolate tint. Carter's Esperia horassus (" Porcupine " collection) has an almost identical external form, but our sponge stands alone in this resjject among its immediate congeners. The megasclera are large, but only about half as long as those of Esperiopsis profunda (see below), and the species is further remarkable for having sigmata. Xoca/%.— Station 145a, December 27, 1873; lat. 46° 41' S., long. 38° 10' W.; off Prince Edward Island ; depth, 310 fathoms ; bottom, volcanic sand. Three pieces. Esperiopsis edwardii, Bowerbank, sp., var. americana (PL XIX. figs. 7, 7', 7a). 1866. Isodictya Edwardii, Bowerbank, Mon. Udt. Spong., vol. ii. p. 325 ; vol. iii., pi. IvLii. figs. 15-18. 1866. (1) Isodictya paupera, Id., Op. cit., vol. ii. p. 328 ; vol. iii., jil. Iv. figs. 4-7. 1866. (?) Isodictya gracilis. Id., Op. cit, voL ii. p. 331 ; vol iii., pi. Iviii. figs. 23-26. 1880. Amphilectiis gracilis, Vosmaer, Notes Leyden Mus., vol ii. p. 111. 188.3. Amphilectiis Edwardi, Eidley, Journ. Linn. Soc. Lond. (ZooL), vol. xviL p. 106. Sponge massive, erect, in the form of a somewhat pyramidal column, 137 mm. in height and about 44 mm. in diameter at the base ; based upon a stone, which is encrusted with Polyzoa. Colour in spirit greyish-yellow. Texture rather soft and spongy. Surface glabrous, but covered with low conuli. Dermal membrane distinct, thin, transparent. Pores small, and scattered in irregular groups over the subdermal cavities. Oscula not very numerous, scattered, about 3 mm. in diameter. Skeleton. — (a) Dermal ; the dermal membrane has no special skeleton ; the ends of the primary fibres of the main skeleton abut against it, and it also contains a few scattered stylote spicules, (b) Main ; this is rather diffuse and Halichondrioid, but with distinct primary fibres running vertically towards the surface. REPORT ON THE MONAXONIDA. 79 Spicules. — (a) Megasdera; of one kind only, viz., smooth, sliarji pointed, slightly curved styli (PL XIX. figs. 7, 7') ; size about 0-33 by 0-0126 mm. (6) Microsclera; of one kind only, viz., palmate isochelse (PL XIX. fig. 7a), with only slightly curved shaft and with the extremity of the anterior palm slightly everted ; length about 0'03 mm. This sponge differs from Bowerbank's original mainly in its much more robust growth and (as evidenced by his preparations in the British Museum) in the superior stoutness of the stylote spicules (about 0"0126 as against 0'0078 mm.). The isochelate spicule in the Challenger sponge is also a little larger than in the British form. From an examination of Bowerbank's preparations we have been led to the conclusion that the isochelate spicule which he describes as tridentate is rather palmate, and that there is only one kind of megasclera, which varies somewhat in size. Examples of the same species obtained by the " Triton," off the east coast of Scotland, approach more nearly in external form to the Challenger variety, but still are not nearly so robust and massive.^ Xoca%.— Station 313, January 20, 1876; lat. 52° 20' S.., long. 67° 39' W ; east of the Strait of Magellan; depth, 55 fathoms; bottom, sand; bottom temperature, 47°"8. One fine specimen. Esperiopsis cylindrica, Eidley and Dendy (PL XIX. figs. 2, 2a, 2h). 1886. Esperiopeis cylindrica, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 340. Sponge erect, cylindrical, branching dichotomously ; consisting of a flattened, branching base, about 19 mm. in diameter, from which arises a simple, erect, cylindrical stem, about 4 mm. in diameter. At a height of 187 mm. above the base the main stem, after becoming slightly flattened, divides into two branches, each of which again divides into two ; the total height of the specimen being 275 mm. Colour in spirit yellowish- grey. Texture hard and tough. Surface minutely hispid. Dermal membrane not very distinct ; thin and transparent. Neither pores nor oscula observed. Skeleton. — The centre of the sponge is occupied by a dense core of horny fibre, formed apparently by the almost complete fusion of several longitudinal fibres which are probably branches of the same. This horny core is covered only by a thin rind of granular tissue. A very large proportion of the spicular skeleton is, of course, embedded in the homy substance. There is one central, axial fibre of spicules, which does not, however, form a simple axis, but branches in a tree-like manner, the branches coming oflf at very acute angles ; doubtless this main spicular axis and its branches are to be regarded as the proper core of the horny axis. There is also a system of primary fibres radiating from the central axis to the surface of the sponge ; these, like the axial fibres just described, are composed for the most part of large stylote spicules ; the inner end of each fibre is 1 Ridley, Journ. Linn. Soc. Land. (Zool.), vol. xvii. p. 106. so THE VOYAGE OF H.M.S. CHALLENGER. embedded in the horny axis, while the outer end projects beyond the surface in the form of a tuft of smaller stylote spicules ; these tufts are sometimes seen with a single large stylote spicule in the centre, surrounded by the smaller ones. The primary skeleton lines are crossed by both vertical and horizontal secondary fibres, which are probably in part formed by the branching of the central axis. The whole system of fibres is rather confused. Spicules. — (a) Megasclera ; smooth styli, of two chief sizes; (1) stout, almost straight, unusually sharply pointed, variable in length, measuring up to about 0'7 by 0"023 mm. (PI. XIX. fig. 2); (2) slender, straight, sharply pointed, variable in length, measuring up to about 0'7 by 0'0063 mm. (PL XIX. fig. 2a). The distribution of these two kinds of spicules has already been indicated ; in addition to the tufts of slender spicules which project from the surface there is a zone of very slender, longitudinally placed styli at a short distance below the surface ; indeed it is here that the slender styli attain their greatest length and slenderness, being both longer and more slender than those of the tufts, and it is from one of these longest that the above measurements have been taken. There is little doubt that we have in all these only slight modifications of one form of spicule, (h) Microsclera; (1) small palmate isochelse (PI. XIX. fig. 2h), extremely abundant in the dermal membrane but scanty below; length up to about 0'025 mm. (2) A very few smooth toxa, about 0'07 mm. long; we have only seen a few of these, so they may be foreign, but we think not. The most remarkable feature of this sponge is the extraordinary development of horny matter. The whole sponge, its megasclera, and their arrangement, much recall the genus Ra»pailia. Locality. — Off Port Jackson; depth, 30 to 35 fathoms. One specimen. Espenopsis challengeri^ Ridley, sp. (PL XVIII. ; PL XIX. figs. 8, 8a, 8i). 1885. Amphilectus ehallengeri, Eidley, Karr. Chall. Exp., vol. i. pt. 2, p. 570, fig. 187. 1886. Espenopsis ehallengeri^ Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 341. Sponge (PL XVIII. figs. 1, 2, 3) of very definite, beautiful and symmetrical external form, a good idea of which will be obtained by reference to the figures.^ From a slightly expanded base, which has evidently served to attach the sponge to the bottom upon which it grew, arises a straight or slightly curved stem, composed of densely packed and firmly united stylote spicules. This stem is very much compressed laterally, and from its anterior edge (as we shall term it) arise a number of short, simple branches, placed one above the other at gradually increasing intervals, thus dividing the stem into a series of unequal " internodes," the longest of which is at the top. The stem itself, and each of ^ These drawings are by Dr. J. J. Wild, artist to the Expedition. REPORT ON THE MONAXONIDA. 81 the branches terminate in a thick, fleshy lamella, of which there appear to be usually six or seven in a full-grown example, increasing gradually in size from below upwards. Each lamella has the form of a deeply concave, transversely elongated cup. The concave surface is directed towards the stem (posterior) while the convex surface is anterior, and the supporting branch, or peduncle, is inserted into the centre of the lower margin of each lamella ; sometimes the weight of the large upper lamellse causes the stem to di-oop considerably, and the lamellse to hang down with their concave surfaces uppermost. There are in the collection thirteen specimens, the stem of the smallest (PI. XVIII. fig. 3) being only 38 mm. high, and bearing two rounded, cup-shaped lamellse about 13 mm. in diameter. The foUowiug are the measurements of a very fine and apparently full-grown specimen, with six cup-shaped lamellae. Diameter of expanded base 19 mm., thickness of same 8 mm. Height of stem, from the point where it rises out of the expanded base to its insertion into the margin of the uppermost cup- shaped lamella, 200 mm.; longer diameter of stem, at a point half way between the origins of the third and fourth branches, about 8 mm.; diameter at right angles to the above about 3 mm. Longest diameter of the uppermost lamella, 63 mm.; diameter at right angles to the above, 35 mm.; thickness of same, about 4 mm. Colour in spirit light yellow. Texture, of the stem, very dense and tough ; of the lamella, rather soft and fragUe, but firm. Surface of the stem markedly hispid. Concave surface of lamellse slightly glabrous in appearance, but really very minutely hispid ; convex surface, minutely hispid ; both, as a rule, evenly rounded. Dermal membrane obvious only on the concave surface, where it is distinct, thin and very transparent. Oscula confined to the convex surfaces of the lamellse, over which they are thickly and evenly scattered, small (averaging little over 0*5 mm. in diameter), round, and having their margins flush with the general surface of the sponge. Pores confined to the concave surfaces of the lamellse, where they are enormously abundant, reducing the dermal membrane to a mere network (PI. XVIII. fig. 4) ; they are almost circular openings and of very uniform size, averaging about 0"11 mm. in diameter. Indications are not wanting that these " pores " were, in the living condition, broken up into still smaller openings by bands of delicate membrane, but on this point we are not certain. Skeleton. — (1) The skeleton of the stem; this consists simply of a dense core of closely packed stylote spicules, for the most part placed longitudinally, but with numerous spicules projecting at right angles, and thus giving to the stem its hispid character. (2) The skeleton of the lamellse. (a) Dermal; supporting the dermal membrane on the concave surface is a fairly regular reticulation of rather stout spiculo-fibre, from which numerous spicules project outwards and thus give to the surface its hispid character ; this reticulation is absent from the convex surface, (b) Main; each branch breaks up, at the point where it enters the lamella, into a number of radiating fibres, arranged in a fan-like manner, and in addition to the skeleton thus constituted there is a Halichon- (ZOOL, CHALL. EXP, — PAET LIX. — 1887.) Nllll 11 82 THE VOYAGE OF H.M.S. CHALLENGEK. drioid reticulation of stylote spicules in which one can distinguish lines of fibre running vertically towards the surface and there breaking up into projecting tufts of spicules. Spicules. — {a) Megasdera; of one kind only, viz., smooth, curved styli (PI. XIX. fig. 8), gradually and sharply pointed, usual size about 0"35 by 0'0126 mm. (sometimes a little larger). (b) Microsclera ; rather rare and of one kind only, viz., palmate isochelse (PI. XIX. figs. 8a, 8&) ; with almost perfectly straight shaft, length about 0'03 mm. As regards external form this is one of the most remarkable sponges in the collection, although its spiculation is extremely simple. Coming from a depth of 825 fathoms it affords a capital example of the manner in which deep-sea Monaxonida tend to assume a definite and symmetrical external form. To judge from the number of specimens brought home by the Challenger it would appear to be a very abundant species at the place where it was found. From the fact that the main stem always terminates in a lamella, and that the smallest lamella is always found at the bottom of the series, we may pretty safely assume that growth takes place not from above but from below, and that whUe the upper lamellae are gradually increasing in size, new small lameUse are being one by one intercalated between the lowest previously existing one and the base. The correctness of this view as to the mode of growth is practically proved by the exist- ence in the smallest specimen of a very minute bud on the stem below the lowest of the two lobes {vide PI. XVIII. fig. 3, «)• The only described sponge at all resembling this species in its remarkable external form which we can recall is Foliolina peltata, Schmidt,^ but even here the resemblance is not very great. Zoc«%.— Station 196, October 13, 1874; lat. 48° 30' 0" S., long. 126° 58' 30" E.; east of Celebes Island ; depth, 825 fathoms ; bottom, hard ground ; bottom temperature, 36° "9. Thirteen specimens and some fragments. Esperiopsis challengeri, var. meangensis, Eidley and Dendy. 1886. Esperiopsis cliallengm, var. meangensis, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. sviii. p. 341. We propose the above name for a fragment of a stem and two fragments of a lamella which appear to have come from a specimen closely resembling the types of Esperiopsis challengeri in external appearance and also in spiculation. The differences, so far as we can at present judge, lie in the respective sizes of the spicules, and it is probable that when an entire specimen is found they will prove to be sufficiently great to justify the erection of a new species, when the varietal name given above will stand as a specific name. In favour of the view of the close relationship of the two is the nearness of their respective localities. The spiculation of the variety in question is as follows : — (a) Megasdera ; smooth, slightly curved styli, sharply pointed, varying a good deal in size, but much longer than ' Spong. Atlant. Gebiet., p. 42, yA. iv. fig. 4. EEPORT ON THE MONAXONIDA. S3 those of Esperiopsis chaUengeri proper. Length (in the lamella) up to about 0-63 mm.; breadth usually about 0'015 mm. In the fragment of stem the megasclera are much stouter, measuring about O'G by 0-025 mm. A few spicules of the stouter kind also occur in the lamella, their presence being apparently due to the breaking up and branching of the supporting stem ; this shows that the fragments of lamella and stem probal:)ly belong to the same thing, (b) Microsdera ; palmate isochelse, large and numerous, length about 0'05 mm. (often less, and sometimes up to 0'072 mm.). It will be seen that both microsdera and megasclera are considerably larger than in the types of Esperiopsis chaUengeri. Locality.— Station 214, February 10, 1875 ; lat. 4° 33' N., long. 127° 6' E. ; east of Meangis Island, south of Philippine Islands ; depth, 500 fathoms ; bottom, blue mud ; bottom temperature, 41°'8. Fragments only. Esperiopsis profunda, Ridley and Dendy (PI. XIX. figs. 1, 1«, lb). 1886. Esperiopsis profunda, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 341. There are in the collection two deep-sea sponges which we unite under the above name ; neither is in very good condition ; indeed the larger specimen is very imperfect and hence our description of the external form is taken from the smaller. Sponge consisting of a slender stem, composed of spiculo-fibre, expanded slightly below so as to form a flattened base, and gradually dilating above into a narrow, tubular, elongated head ; we have found no definite opening from the exterior into the central cavity of this head, but the condition of the specimen does not warrant the assertion that none such exists. Total height of sponge, 44 mm.* Longer diameter of tubular head (which in its present condition is flattened) 4 mm.; diameter of stem near base 0'7 mm. Colour in spirit light yellowish-grey. Texture very soft and spongy. Surface hispid. Dermal 'membrane, Oscula, and Pores unknown. Skeleton. — The skeleton of the head consists of a loose and confused reticulation of long stylote spicules, derived from the breaking up of the stem, at the point where it joins the head, into a number of thin, loose fibres, which radiate upwards through the walls of the tube and are crossed irregularly by other stylote spicules. Sp>icules. — (a) Megasclera; of one kind only (PI. XIX. fig. 1), viz., very long, rather slender and often very slightly curved, smooth styli, with evenly rounded base (sometimes, however, becoming tylostylote), and tapering very gradually to a sharp point at the apex; size about 1*4 by 0'0157 mm. (6) Microsdera; of one kind only, viz., large, palmate isochelae (PL XIX. figs, la, lb), with almost straight shaft and rather narrow front palm; these spicules may attain a length of 0'09 mm., but smaller ones about 0"05 mm. long are much more common. ' The larger specimen is a little over 100 mm. in height. 84 THE VOYAGE OF H.M.S. CHALLENGEE. This sponge is interesting on account of the great depth (1600 fathoms) from which it was obtained, and with this must again be associated the presence of a definite external form. It is distinguished from other species of the genus by its very long megasclera and by the external form. Locality. — Station 147, December, 30, 1873; lat. 46° 16' S., long. 4S° 27' E.; between Prince Edward Island and Crozet Islands, Southern Ocean ; depth, 1600 fathoms ; bottom, Diatom ooze ; bottom temperature, 34° -2. Two specimens, in bad condition. Esperiopsis anomala, Ridley and Dendy. 1886. Esperiopsis anomala, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 341. Sponge digitate, irregularly ramose ; branches subcylindrical, varying in diameter. One of the two specimens in the collection measures nearly 125 mm. in length but only about 6 mm. in average diameter. The second is about 50 mm. long and 8 broad, being somewhat compressed and giving ofi" a few short, stumjay branches. Colour in spirit light greyish-yellow to dark grey. Texture very soft and compressible, but elastic, fibrous, and fairly tough. Surface uneven. Dermal membrane thin, delicate, trans- parent, rather loosely attached to the underlying tissues. Oscula (?). Pores, very abundant rounded openings through the thin dermal membrane, of various sizes. Skelet07i. — (a) Dermal; represented only by sparse, radiating brushes of spicules with outwardly projecting points, {h) Main; a rectangularly meshed reticulation of stout spiculo-fibre with primary and secondary lines distinct. Fibre containing a very large amount of spongin and few spicules. Sjncules. — {a) Megasclera; long but very slender styli, sometimes tending to become tylostylote, sharply pointed; size about 0-25 by 0'005 mm. {h) Microsclera; very minute, very slender isochelse, very scarce, found in the dermal membrane, seen only in side view ; length about O'Ol mm. Owing to their great slenderness these spicules are very diflicult to make out; they give one the impression of being on the verge of dis- appearance. ■ They become visible after prolonged soaking in Canada balsam. The most remarkable feature about this sponge is the very strong development of spongin, which gives it quite a Chalinine appearance. It forms a very good instance of the manner in which horny fibre may be developed in any genus. Locality. — Honolulu, 16 to 20 fathoms. Two pieces. Esperiopsis pulchella, Ridley and Dendy (PL XIX. figs. 9, 9', 9a, 96, 9c, 9d ; PI. XXX. fig. 3, e). 1886. Esperiopsis (?) pulchella, Eidley and Dendy, Ann. and Mag. Kat., ser. 5, vol. xviii. p. 341. Under the above name we propose to describe a very small, encrusting sponge from Station 192. The sponge occurs in very small and excessively thin patches of a blackish REPORT ON THE MONAXONIDA. 85 colour, growing upon Myxilla jxiucispinata, nobis. These patches present a faii'ly even surface, but are marked with minute oval spots of a much lighter colour; these spots at first sight look like pits in the surface, but on minute examination each is found to be caused by the presence of a subdermal cavity covered over by a very thin dermal membrane pierced with numerous pores; the pores are thus arranged in definite pore-areas, each of which is about 0'45 mm. in diameter. The pores themselves are oval openings about 0"07 mm. in diameter, reducing the dermal membrane in the pore-areas to a mere network. The dermal membrane and also the deeper parts of the sponge are heavily loaded with very numerous, minute, round cells of a blackish-green colour, each about 0*004 mm. in diameter ; it is to these cells that the dark colour of the sponge is undoubtedly due. Oscula not observed, possibly some of the pores mentioned are exhalent openings. Skeleton. — Consisting of brushes or wisps of small stylote spicules running more or less vertically from the base to the surface of the sponge. Spicules. — (a) Megasclera; smooth, straight styli or subtylostyli (PI. XIX. figs. 9, 9'), sharply but only faii-ly gradually pointed, and usually with several slight bulbous inflations along the shaft (^Jolytylote); size about 0"3 by 0"0063 mm. {h) Microsclera ; these constitute by far the most noteworthy and characteristic feature of the species, being of extraordinary size and of equally remarkable and very beautiful form. They are isochelse, and as the shape is sufficiently illustrated by the figures (PI. XIX. figs. 9a, 96) we shall not enter into a description of it in this place ; the full-grown spicule measures about 0"1 mm. in length. In addition to these there are numerous much smaller spicules of rather diff"erent shape (PI. XIX. fig. 9c), being a great deal shorter and much broader in proportion to their length; length commonly about 0'044 mm. Still smaller ones (PI. XIX. fig. ^d) occur, which look just like the minute, slender isochelge of the ordinary " Amphilectus" t3^e ; length about 0'015 mm. Although it is quite possible that there may be here at least two different kinds of microsclera, yet we are inclined to regard the two smaller forms as young stages of the large one. If this view be correct, then, from the absence of a complete series of intermediate sizes, we must conclude that the microsclera are produced periodically in batches. The chief points of interest in this sponge concern its microsclera and their large size, especially when compared with the small size of the megasclera. It must, however, be borne in mind that the specimens may be merely young, encrusting stages of some sponge of which the adult is yet unknown. It will be seen by comparison of fig. 9, &c., PI. XIX., with fig. 8, &c., PL XXIIL, that Phelloderma radiatum, nobis, comes very close to the present species as regards spiculation ; it is possible that Esperiopsis ptdchella is a young form of some species of PhellodermtL, but there is not sufficient evidence to justify us in placing it in that genus. ioca%.— Station 192, September 26, 1874 ; lat. 5° 49'15"S., long. 132° 14' 15" K; south-west of New Guinea ; depth, 140 fathoms ; bottom, blue mud. 86 THE VOYAGE OF H.M.S. CHALLENGER. Genus Cladorhiza, M. Sars (Pis. XX., XXL). 1872. Cladorhiza, M. Sars, Eemarkable Forms of Animal Life from the great deeps off the Norwegian Coast, pt. i. p. 65. Sponge of varying, but usually symmetrical external form. Skeleton usually consisting of a central, erect axis of spiculo-fibre, which may or may not be branched, and from which arise longer or shorter processes also composed of spiculo-fibre. Spicules. — (a) Megasclera ; chiefly stylote and often attaining a great length. (b) Microsclera ; auisochelse, characteristically with three or more claw -like teeth at each end and with a curved shaft expanded laterally into wing-like processes, which are especially developed near the larger end of the spicule. Sigmata may be present. The genus was first characterised as follows : — " Spongia sUicea ramosa, fasciculis densis spiculorum acuformium axem soHdam formantibus sustentata, radiculis numerosis arborescentibus ex spiculis ejusdem generis formatis in limo afiixa. Parenchyma axem interuam corticis instar circumdans spiculis superficialibus anchoratis et bihamatis ornatum. Oscula et pori nulla. Ova in apicibus dUatatis ramorum se evolventia " (Sars, loc. cit.). This diagnosis, drawn up for but a single species, may now be emended as above. Schmidt ^ includes in the genus also those forms which have isochelate microsclera. It has, however, seemed preferable to us to retain the name Chondrocladia, origin- ally proposed by Wyville Thomson for his Chondrocladia virgata, for the species with isochelaB, and to confine the name Cladorhiza to those with anisochelate microsclera. The two genera thus distinguished are undoubtedly very closely allied, as may be seen by the arrangement and form of the spicules. Although the cliS"erent species of Cladorhiza vary very much in external form, yet the difi"erent modifications of the main skeleton, ujaon which the external form of the sponge depends, are easily derivable from one common primary type, and afi'ord inter- esting instances of adaptation. In its simplest condition the main skeleton consists of a straight, slender axis of spiculo-fibre, from which short spicular processes (or pinnae) proceed in all du-ections. Cladorhiza ahyssicola, var. rectangular is, nobis (PI. XX. fig. 10), still retains almost this primitive type of skeleton, which in other species and varieties becomes very variously modified. In Cladorhiza ji^nnatula, Schmidt," the pinnse are borne only on two opposite sides. In the typical Cladorhiza ahyssicola, Sars, the main axis becomes branched ; this may take place both above and below, giving rise to branches in the one case and to anchoring rootlets in the other. The pinnae may be confined to a certain part of the stem only, where they attain a great development. A good example of 1 Jahresh. Comm. JFiss. Untersuch. d. deutsch. Mecre, &c. 2 Jahresh. Comm. iViss. Untersuch. d. deutsch. Meere, Jahrg. i., ii. p. 119, pi. i. figs. 14-16. REPORT ON THE MONAXONIDA. 87 this is afforded by Cladorhiza moruliformis, nobis, in which the piunse are placed close together near the top of the main axis (vide woodcut, Fig. 1) ; being all of the same length, and radiating in every direction, while the soft tissues of the sponge occupy the spaces between them, they give rise to a spherical head perched on the end of a stalk. But by far the most remarkable of the modifications which the main skeleton thus undergoes are exhibited in the various species which we have called " Crinor- hiza" forms ^ (vide woodcut, Fig. 2). When this remark- able adaptive form is present in its most typical condition we have the central axis represented by a stiff, straight, tapering root ; the bulk of the soft parts are condensed into a small subglobular or cap-shaped " body," while the pinnae are very long and slender, and are arranged in a single whorl close to the top of the axis ; these long processes are very numerous and radiate in all directions, extending far beyond the body of the sponge ; they are directed outwards and downwards, and their function is doubtless to prevent the sponge from sinking into the soft mud on which it lies, for which purpose they are admirably adapted. Fig. 1, Cladorhiza morulifonnis. Longi- tudinal section of tlie head showing the arraugenient of the skeleton ; «, axis of stem and radial bands of fibres ; ca, capsule of loosely disposed spicules, x 4. ^%v ■\ ^^ m FlQ. 2.—Axoniderina mimhile. Vertical section, showing the skeleton arrangement in a typical " Crinorhiza " form ; a, axis of stem ; a', axis of supporting processes ; c, cortical layer of amphiasters ; ch, choanosome (?). x 5. The Crinorhiza forms appear to be without oscula and pores, nor have we succeeded finding flagellated chambers, although some of the specimens were in very fair 1 Anil, and Mag. Nat. Hist., ser. 5, vol. xviii. p. 342. 88 THE VOYAGE OF H.M.S. CHALLENGER. condition. It seems just possible, therefore, that, as originally suggested by Sars ^ in the case of the first known species of the genus, Cladorhiza abyssitola, these sponges have some method of obtaining their supplies of nutriment which is quite difi'erent from that found in other sponges ; this is, however, extremely unlikely. As regards the genus Crinorhiza itself we are necessarily in doul^t. It was founded by Schmidt, and aU that he wi'ote about it is contained in the five lines in which he describes his Crinorhiza araphactis} The description is too short to be of much service for identification, but inasmuch as he mentions and figures an osculum at the summit of the sponge we are quite satisfied that none of our forms are specifically identical with his. The external reseml)lance is however striking, and while the genus Crinorhiza will probably have to be abolished, we gladly make use of the name for that particular external form. In this course we are supported by the example in cognate circumstances of Haeckel in the case of the Calcarea, and of Smitt and Hincks in that of the Polyzoa. Indeed, the existence of well-marked external forms running through difi'erent genera has already been thoroughly recognised in Haeckel's classical work Die Kalkschwamme, and lies at the foundation of the new departure in Polyzoan classification taken by the authors we have named. Crinorhiza forms also occur in the closely allied genus Chondrocladia, Wj."v. Thomson, and in the still more closely allied genus Axoniderma, nobis, but to these cases we shall recur later on. They certainly have a very wide geographical range, for the Challenger brought home species of Cladorhiza, having this form from the North Pacific (2385 fathoms ; bottom, red clay); the South Pacific (3000 fathoms ; bottom, red clay); and the South Atlantic (2200 fathoms ; bottom, Globigerina ooze); as will be found fully recorded under the various species. The geographical distribution of the genus is now known to be very wide, and includes the North and South Atlantic, the Southern Ocean and the North and South Pacific. The Cladorhiza are essentially deep-sea sponges, and are not unfi-equently obtained at depths of between 2000 and 3000 fathoms, and they afford some of the most striking examples known of the symmetry and beauty which characterise Monaxonida living in very deep water. Cladorhiza ahyssicola, Sars, var. rectangular is, nov. (PI. XX. fig. 10). 1872. Cladorhiza ahyssicola, Sars, Remarkable Forms of Animal Life from the great deeps off the Norwegian Coast, pt. i. p. 65. Sponge (PI. XX. fig. 10) consisting of a straight, slender, cylindrical stem, unbranched ; terminating above in a rounded extremity and below in several delicate, forking rootlets. Length of stem 50 mm.; diameter 1 mm.; greatest length of roots 25 mm. All the way up the stem, coming off at right angles, from four sides, arise 1 Remarkable Forms of Animal Life, pt. i. p. 68. ^ Spong. Meerb. von Mexico, pt. ii. p. 83. EEPORT ON THE MONAXONIDA. 89 numerous, long, slender, hair-like processes, broadest at the base, where they join the stem. They are arranged very regularly in four rows, so as to lie in two vertical planes which intersect each other at right angles. The interval between each two successive processes in a row is about 1-5 mm.; the processes themselves may attain the great length of 17 mm. Colour in spirit pale yellowish. Sheleton. — A stout axis of spiculo-fibre runs through the stem, branching to form the roots and giving off above the hair -like processes (pinnag). Spicules. — (a) Megasclera ; smooth, slender styli, broadest in the middle and tapering rather abruptly to a fairly sharp point at the apex. They may reach the great length of over 2 mm., and have a diameter of about 0'019 mm. Their length, however, varies much, and is generally less than that given. They are very firmly bound together in fibres, and it is difficult to get one separate for purposes of measurement, (h) Micro- sclera ; of two kinds ; (1) anisochelae, with long, curved shaft and very unequal, claw-like extremities ; they are very small, measuring about 0"025 mm. in length, and closely resemble those figured by Sars^; (2) rather large, smooth, simple sigmata, measuring about 0-13 by 0'007 mm. The anatomy of the soft parts of this remarkable species has always been a mystery and must still remain so, for unfortunately the only specimen obtained by the Challenger was found dried up when we came to describe it. It will be seen that, as regards external form, the Challenger variety diifers very considerably from those described and figured by Sars.^ It also differs widely in appearance from the " Porcupine " specimens, now in the British Museum. It resembles in slenderness and delicacy Cladorhiza pennatula, Schmidt,^ which has, however, the lateral pinnte developed only in one plane. Two remarkable features about this sponge are the very great depth at which it occurs and the locality ; Sars* says : — " I have only found this remarkable sponge in one single locality, namely at the fishing station Skraaven in Lofoten at the great depth of 300 fathoms on soft clay bottom"; whereas the Challenger variety comes from a depth of 2750 fathoms, in the middle of the Pacific Ocean. The "Porcupine" specimens were obtained between Scotland and the Fseroe Islands, Schmidt's in the Skagerrack, and Hansen's (those of which the localities were preserved) either between Norway and the Fgeroe Islands, or near the west coast of Spitzbergen, one of these stations having a depth of 1215 fathoms; but uncertainty overhangs the identifi- cations of the species by both these authors {e.g., Hansen perhaps includes Carter's so-called variety corticocancellata,, which may probably be regarded as a distinct species). Locality.— Station 274, September 11, 1875; lat. 7° 25' S., long. 152° 15' W.; 1 Remarkable Forms of Animal Life, pt. i. pi. vi. figs. 31, 32, a. 2 Loc. cit, figs. 16, 17. 3 Jahresb. Comm. Wiss. Untersuch. d. deutsch. Mecre, Jahrg. i., ii., p. 119, pi. i. figs. 14-16. * Loc. cit., p. 68. (zooL. CHALL. EXP. — PART Lix. — 1887.) N'nn 12 90 THE VOYAGE OF H.M.S. CHALLENGER. Mid-Pacific; depth, 2750 fathoms; bottom, Radiolarian ooze; bottom temperature, 35 "'l. One specimen. Cladorhiza abyssicola, Sars, var. linearis, nov. (PI. XX. fig. 6 ; PI. XXI. figs. 2, 2a). Under this name we include three denuded skeletons of a Cladorhiza (PI. XX. fig. 6) obtained in the South Pacific at a depth of 2385 fathoms. Each consists of a long, straight, slender spicular axis about 106 mm. in length, with a few remaining short tufts of spicules projecting fi-om it in places ; entangled in some of these tufts are a few microsclera, the size and shape of which induce us for the present to consider the three specimens as a variety of Cladorhiza abyssicola, Sai\s, though very probably the examination of more perfect specimens may ultimately show them to be specifically distinct, or more nearly allied to Esperella biserialis, nobis, from the same station. The anisochelate spicules are of fair size and have a strongly curved shaft, expanded towards the large end, and with only three short teeth at each extremity ; they measure about 0"032 mm. in length. The sigmata are large, apparently not contort, measuring about 0'17 by O'Ol mm. The megasclera (PI. XXI. figs. 2, 2a) are long, slender styli of very beautiful shape, much resembling those of Esi^erella biserialis, but the short tylostyli of the latter are not present. In the stem the styli may attain a length of nearly 3 mm. Locality. — Station 281, October 6, 1875 ; lat. 22° 21' S., long. 150° 17' W.; South Pacific; depth, 2385 fathoms; bottom, red clay; bottom temperature, 34°'9. Cladorhiza moruliformis, Eidley and Dendy (PI. XX. fig. 3 ; PI. XXI. figs. 1, 15, 19). 1886. Cladorrhiza moruliformis, Ridley andDeady, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 342. Sponge (PL XX. fig. 3) consisting of a small, globular head, perched on the summit of a stalk. A large number of short, stout, conical processes, all of about the same size, arise from the head in every direction, giving to it the appearance of a mulberry (whence the specific name). The main stalk, or axis of the sponge, is prolonged through the head, and projects for a short distance vertically above it {vide woodcut. Fig. 3). Diameter of the head, exclusive of the conical processes, 10 mm. Length of conical processes 2 mm.; transverse diameter of same 1 mm. Length of stalk still attached to the head (it has evidently been broken off short) 2 mm. ; diameter of same 2 mm. Length of part of axis projecting above the head 2*5 mm. Colour, when dried, white. Texture fragile. Skeleton. — (l) A main axis of spiculo-fibre is continued from the stalk, of which it forms the chief part (if not the whole), right through the globular head, projecting for a short way beyond the top of the sponge. (2) From this stalk, starting from near REPORT ON THE MONAXONIDA. 91 Fig. 3. — Cladnrhiza mondi/onnis. Longi- tudinal section of the head, showing the arrangement of the slieleton ; a, axis of stem and radial bands of fibres ; ca, capsule of loosely disposed spicules, x 4. the centre of the head, radiate a large number of dense, stout spiculo-fibres (Fig. 3, o) ; each runs out to the surface of the sponge and is continued to the end of one of the conical processes already mentioned. Each conical process is thus supported by an axial fibre. (3) At about one-third of the distance between the surface and the centre of the head is a reticulate layer of megas- clera, which forms a kind of capsule (Fig. 3, ca), dividing the soft parts of the sponge into an inner and an outer portion. This capsule is, of course, pierced in numerous places by the stout radiating fibres already mentioned. Sjyicides. — (a) Megasclera ; long, slender, very straight styli (PI. XXI. fig. 1), which may reach a length of over 2 mm.; diameter about 0"05 mm. They are broadest in the centre and taper gradually towards each end, are flatly rounded at the base and hastately pointed at the apex. (6) Microsciera ; (l) anisochelce (PL XXI. fig. 19), of moderate size, length 0-063 mm. ; these spicules have a slightly curved shaft, much expanded laterally towards the large end ; the ends are of very unequal size and each has three sharp, prominent teeth. (2) Large, very sharply pointed, contort .sigmata (PI. XXI. fig 15), measuring up to 0-35 by 0-0145 mm. The anisochelate spicules are especially abundant in the smaU conical processes, forming a dense crust around the axial fibre, with their large ends directed outwards. They are also very abundant in the tissues both inside and outside the " capsule." The sigmata appear to be very much more abundant outside the " capsule " than inside it. Unfortunately this sponge was dry when the bladder was removed from its bottle, and we have therefore been obliged to confine our observations to the arrangement of the hard parts. It is a form of the very greatest interest, and aS"ords a good example of radial symmetry in a Monaxonid sponge. It was a question with us whether or not this species should form the type of a new genus, but we finally decided not, as no essential changes are necessary to derive it from the more typical species, such as Cladorhiza ahyssicola, Sars. Imagine an unbranched Cladorhiza ahyssicola in which the points of attachment of the lateral pinnse to the axis come close together ; if the pinnae are all of the same length the sponge will then have a globular form, and little further change is needed to convert it into Cladorhiza moruliformis. The chief difficulty is the capsule, and this is not an insuperable one. In some of the " Porcupine " Cladorhiza' s (viz., Cladorhiza ahyssicola, var. corticocancellata, Carter), there is an irregularly reticulate skeleton, chiefly placed at some distance beneath the surface, between the main radial fibres, and if we imagine this to become concentrated into one &2 THE VOYAGE OF H.M.S. CHALLENGER. definite layer, at a little depth beneath the surface, we arrive at the condition of Cladorhiza moniliformis. It is, of course, by no means impossible that perfect specimens of Cladorhiza moruliformis may have a branching stem with a head at the end of each branch ; this remains to be seen. In spiculation it agrees very closely with other species of the genus. Locality. — Station 157, March 3, 1874 ; lat. 53° 55' S., long. 108° 35' E.; Southern Ocean, south-west of Australia ; depth, 1950 fathoms ; bottom. Diatom ooze ; bottom temperature, 32°'l. One specimen. Cladorhiza longijnnna, Eidley and Deudy (PL XX. fig. 2 ; PI. XXI. figs. 4, 21). 1886. Cladorrhiza longipimia, Ridley and Dendy, Aun. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 342. This sponge (PI. XX. fig. 2) has acquii-ed the " Cririorhiza " form. It is composed of an almost globular body, somewhat flattened on the lower surface, and also, though over a less area, on the uj)per surface. The circumference of the flattened lower surface is fringed with very long, fine, supporting processes, twenty-five or thirty in number, projecting outwards and downwards, while a cii'clet of very short, stiff", stumpy processes crowns the summit of the head. From the centre of the lower surface depends a long, rather stout, slightly tapering root or stem. Diameter of head 5 mm. Length of root 27 mm. Average length of supporting processes 19 mm. Colour in spirit pale yellow. Dermal membrane distinct, transparent. Skeleton. — Arranged in the usual " Crino7^hiza" iorm.; i.e., bauds of spiculo-fibre form the axes of the various processes given off" from the body ; these are composed of the usual, long, fusiform, stylote spicules. Spicules. — (a) Megasclera ; fusiform styli, reaching over 3 mm. in length by about 0'05 mm. thick, and usually much blunted at the apices. Very numerous smaller styli also occur scattered about in the soft parts of the sponge, both in the dermal membrane and in the deeper tissues ; these spicules vary greatly in size (measuring say about I'O by 0"015 mm.). Sometimes they show a tendency towards the development of heads, but it is very rarely that these are developed to the extent shown in the figure (PL XXI. fig. 4), which has been unfortunately chosen. They diff"er from the larger styli in being much more sharply pointed, but it would probably not be difficult to pick out an intermediate series both in shape and size, (b) Microsclera ; we have found only anisochelse (PL XXI. fig. 21); these are of the usual Cladorhiza form, but on the lower surface of the sponge attain an unusually large size. Length from 0"034 to 0"06 mm. (on the lower surface of the sponge). The species may be recognised by the shape of the body, the great length of the supporting processes and the presence of a second cii'cle of short processes around the EEPORT ON THE MONAXONIDA. 93 summit. The scattered tylostylote spicules of Cladorhiza similis and Cladorhiza inversa are replaced by styli. Imbedded in the lower surface of the sponge was a large, spherical, hard, brown body ; probably some parasite, perhaps encysted. This illustrates well the protective use of the dense external armour present in Axoniderma mirahile, nobis, and absent in this species. Locality.— ^tn.tio^ 264, August 23, 1875 ; lat. 14° 19' N., long. 152° 37' W.; North Pacific Ocean; dej)th, 3000 fathoms; bottom, red clay; bottom temperature, 35°-2. One specimen. Cladorhiza similis, Eidley and Dendy (PL XX. fig. 7; PI. XXI. figs. 5, 5«, bh, 18). 1886. Cladorrhiza similis, Eidley and Dendy, Ann. and Mag. Kat. Hist., ser. 5, vol. xviii. p. 343. The sponge (PI. XX. fig. 7) has assumed the " Crinorhiza " form, and consists of a cap- shaped, conical head, perched on the end of a stem or root, which appears to have been broken ofi" short, so that its proper length cannot be ascertained. From the margin of the cap project a considerable number of long spicular processes, radiating outwards and downwards. The apex of the sponge is pointed, and apparently devoid of short spicular processes. The lower surface is shallowly concave, and only just G mm. in diameter. Colour in spirit dirty yellow. Skeleton. — AiTanged as usual in " Crinorhiza " forms. Spicules. — (a) Megasclera; (l) very long, slender styli of the usual type; (2) short, fusiform tylostyli (PI. XXI. figs. 5, 5«, 5?^), with peculiarly shaped heads ; usually very sharply pointed, and with the head and neck inclined at an angle to the main shaft. These spicules vary greatly in size, measuring from 0*21 to 0'6 mm. in length, usual thickness about 0'0157 mm. They occur chiefly scattered about near the surface of the sponge, where they are very thickly placed, and apparently represent in function the large, five-toothed, special protective spicules of Axoniderma mirahile. They are also found in the skeleton-fibres. (&) Microsclera; we have seen only the anisochelse, but in our anxiety not to destroy such a unique and so small a specimen, it is not improbable that we have overlooked the sigmata. The anisochela (PI. XXI. fig. 18) is of the usual Cladorhiza type. The shaft is curved and much expanded laterally towards the larger end, and there are three teeth at each extremity. Length about 0'03 mm. It will be seen from the foregoing description that, as regards external appearance, this species comes very close indeed to Axoniderma mirahile, nobis ; it differs, however, in the replacement of the additional microsclera, so characteristic of the latter, by small tylostyli, and also in the absence, if they be really absent, of the sigmata. The anisochelse are, however, almost identical both as regards shape and size in the two species. Locality. — Station 281, October 6, 1875; lat. 22° 21' S., long. 150° 17' W.; South Pacific; depth, 2385 fathoms; bottom, red clay; bottom temperature, 34°'9. One specimen in moderately good condition. 94 THE VOYAGE OF H.M.S. CHALLENGER. Cladorhiza inversa, Eidley and Dendy (PI. XX. fig. 8 ; PI. XXI. fig. 13). 1886. Cladoirhiza inversa, Eidley aud Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 343. Sponge (PI. XX. fig. 8) of the " Crinorhiza " form, consisting of a very small, conical head, produced upwards into a long, slender, tapering process. The base of the cone is nearly flat and faces downwards, and from its circumference there project outwards and downwards sixteen rather short, stifi", radiating processes. From near the centre of the lower (flat) surface of the sponge a single very short, stifi" process projects, which is probably a downward prolongation of the main spicular axis, representing, in a much reduced condition, the usual " Crinorhiza " root or stem, while the maui axis is developed to an unusual extent above. Diameter of base of cone 3 mm. Colour in spirit yellow. Surface even. Skeleton. — The main skeleton consists, as usual, of a number of bands of spiculo- fibre, forming the axes of the various processes given ofi" from the main mass. Spicules. — (a) Megasclera ; long slender styli, bluntly pointed and gradually swelling out towards the centre ; size about 2*0 by 0'037 mm. (there is, however, a wide range of variation). In addition to these, which form the main skeleton, there are present a considerable number of loose tylostylote spicules. These are thickest in the middle, have a club-shaped head, and taper rather suddenly to a sharp point at the apex; size about 0"63 by 0'019 mm. (b) Microsclera; (1) anisochelse (PI. XXI. fig. 13), of the ordinary Cladorhiza type ; with three teeth at each end and a curved shaft very much expanded, especially near the large end ; length about 0*03 mm. (2) Sigmata (?). The anisochelse are exceedingly abundant; on the radiating j)rocess they are arranged close together round the spicular axis with their larger ends pointing- outwards. This appears to be a very common mode of arrangement in the genus.^ The most remarkable feature about this species concerns its external form ; com- pared with other known " Crinorhiza " forms it appears to be upside down ; ^ nor can we be certain that the surface which we have called "lower" in the description is not really the upper, and vice versd. Having regard, however, to the function of the long radiating processes — which is, without doubt, to support the sponge in the soft mud on which it lies — we see that this would be best effected if the sponge lived in the position we have assigned to it. We must imagine, then, that this sponge agrees with its relatives in the shape of the conical head and the arrangement of the supporting processes, but that the main axis is developed in an upward instead of a downward direction ; being barely represented below, while in most cases it is barely represented above. ' Cf. Sars, Remarkable Forms of Animal Life, pt. i. p. 6". 2 Cf., however, Clwndrocladia crinita (p. 101), which makes an approach to this condition. REPORT ON THE MONAXONIDA. 95 As regards spiculation, it will be seen that the species agrees fairly closely with Cladorhiza similis, nobis, though the localities from which the two species were obtained are very widely separated. Locality. — Station 332, March 10, 1876 ; lat. 37° 29' S., long. 27° 31' W.; South Atlantic ; depth, 2200 fathoms ; bottom, Globigerina ooze ; bottom temperature, 34°'0. One specimen, in good condition. Cladorhiza (f) tridentata, Ridley and Dendy (PI. XX. figs. 9, 9a ; PI. XXI. figs. 16, 20). 1886. Cladorrliiza tridentata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 343. Sponge (PI. XX. figs. 9, 9a, and woodcut Fig. 4) small, invertedly dome-shaped. Upper surface circular, concave, with slightly inwardly turned margin ; lower surface convex; may be attached. Height 6 mm.; diameter of upper surface 12 mm. Colour in spirit pale greyish -yellow. Texture soft and jdelding. Lower surface even but minutely hispid. Upper surface even and smooth. Dermal membrane most distinct over the smooth upper surface. (Until we found the specimen figured in the woodcut, which, being attached to a stone, gives the true position of the sponge, we thought that the convex surface was the upper one, hence the position given in PI. XX. fig. 9c* should be inverted ; when it was drawn the attached specimen was missing.) Skeleton. — (a) Dermal ; a very loose and irregular reticu- lation of megasclera. (b) Alain ; with no distinct fibre, a very loose and irregular reticulation of spicules. Very numerous y,,,i,_ciaaorMzai?) tridentata. spicules project more or less vertically outwards from the lower ^^^ZTthe'^lti'lotmlZt surface of the sponge for a considerable part of their length, S unC"l^'fS> 'tl^on^'^x 2!^^''' giving to it its hispid character. Spicules. — (a) Megasclera ; of one kind only, viz., long and very slender tylostyli, with only very small heads ; broadest in the middle and tapering very gradually to a very fine point at the apex; size about 07 by 0'0155 mm. (6) Microsclera ; (l) Large, tridentate anisochelse of very characteristic form (PI. XXL fig. 20), with three stout, sharp teeth at each end, deeply separated from one another ; the shaft of the spicule is curved, with large, expanded, wing-like, lateral processes, tapering off" from the large end downwards. Length of spicule about 0'076 mm. These chelate spicules are especially abundant in the dermal membrane on the top of the sponge, forming in places an almost continuous layer. (2) Slender sigmata (PI. XXI. fig. 16), frequently more or less contort, measuring about 0'09 by 0"0032 mm. WOODS HOLE, MASS 96 THE VOYAGE OF H.M.S. CHALLENGER. In external appearance this species at first sight closely resembles Esperella mammiforviis, and as it comes from the same station one is liable to confuse the two before examining them microscopically. It really bears some resemblance to a typical Esperella in having heads to the megasclera. The resemblance, however, does not go much further. We have seen that there is no definite skeleton fibre such as exists in EsjyereUa mammiformis, although the surface is hispid. The chelate spicule also difi"ers much from that of an Esperella in not being properly palmate and in having a widely expanded shaft, both these features being very characteristic of Cladorhiza. Imbedded in the soft tissues of the sponge all around the margin, and at fairly regular intervals from one another, are a number of small, round, yellow bodies (Fig. 4, a, and PI. XX. fig. 9), varying in size. Possil^ly they are foreign objects, but they occur in the same position in all three specimens, so this view is highly improbable. Full details concerning them will be found in the Introduction. In the arrangement of its skeleton this sponge stands quite alone in the genus, and as this is generally such a good guide it seems very doubtful whether it ought to be admitted as a Cladorhiza ; but for the sake of convenience we shall retain it here for the present, placing it at the end as a doubtful species. Only three examples were obtained, one of which represents only half of the sponge, while another is damaged on the lower surface, having evidently been torn from its attachment. The third specimen is, however, still attached to the stone on which it grew (woodcut, Fig. 4). Locality. — Station 147, December 30, 1873 ; lat. 46° 16' S., long. 48° 27' E.; between Prince Edward Island and Crozet Island; depth, 1600 fathoms; bottom, Diatom ooze ; bottom temp»erature, 34°-2. Three specimens. Genus Axoniderma,^ Ridley and Dendy (Pis. XX., XXL). 1886. Trochoderma, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 344. 1886. Axoniderma, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 493. Megasclera stylote to tylostylote. Microsclera anisocheh-e of the Cladorhiza type, to which may be added sigmata. In addition to these spicules there is also present another form, peculiar to the genus. These may, for the present, be classed as microsclera. They are amphiasters, consisting each of a long, cylindrical shaft with five equal teeth arranged in a star at each end (PI. XXI. fig. 9). This genus is probably descended from a " Crinorhiza " form of a Cladorhiza, which, as a protection against parasites and other enemies, has acquired an additional kind of 1 From Greek a.iu«, a wheel, and lipfj.a., the skin. We are indebted to Professor F. J. Bell for pointing out to us that the name Trochoderma was already in use for a genus of Echinodemis, so we have altered the name while endeavouring to retain the meaning. REPORT ON THE MONAXONIDA. 97 spicule, the amphiaster ; the additional spicules being arranged in a dense external layer. Or, on the other hand, it may have been derived from a genus of sponges, characterised by the presence of an amphiaster, which, having taken to deep water and a soft bottom, has acquired the " Crinorhiza " form independently. In any case the genus seems to be very nearly related to Cladoi-hiza, standing to it in much the same relation as Meliiderma does to Clwndrocladia (vide p. 102). Axoniderma mirahile, Eidley and Dendy (PL XX. fig. 5 ; PI. XXI. figs. 8, 9, 10). 1886. TrocJwderma mirahile, Eidley and Dendy, Ann. and Mag. Nat. Hist., sen 5, vol. xviii. p. 344. 1886. Axoniderma mirahile, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 493. Sponge (PI. XX. fig. 5) of " Crinorhiza " form, consisting of a conical, cap-shaped body perched on the end of a slender stalk or root, which is inserted into the centre of the lower, concave surface of the body. The root sweUs out considerably just before Fig. 5.—Axonidenna mirabile. Vertical section, showing tlie skeleton arrangement ; a, axis of stem ; a', axis of supporting process ; c, cortical layer ofamphiasters ; ch, clioanosome (?). x 5. its insertion into the head. From the free circular margin of the body numerous (thirty or forty) very long, filamentous, spicular processes are given off" obliquely outwards and downwards. The summit of the body is produced into a papilla, from which project numerous very short and slender spicular processes. Transverse diameter of the body, from base to base of the long spicular processes, a httle over 6 mm. Length of root nearly 50 mm.; it thins out to hair -like proportions at the lower end, and may very possibly have been broken off". Length of the long spicular processes (zooL. CHALL. EXP. — PAKT Lix. — 1887.) Nnu 13 98 THE VOYAGE OF H.M.S. CHALLENGER. about 31 mm. Colour in spirit yellow. Texture hard and firm owing to the extraordinarily small amount of soft tissues present. Surface almost smooth. Oscula and pores apparently absent. Skeleton. — A band of spiculo-fibre forms the axis of the stem or root, and is continued in a looser condition to the summit of the head, there giving rise to the short, slender, projecting, spicular processes already described, each of which is composed of a single large spicule. From the centre of the head bands of spiculo-fibre radiate outwards and downwards, giving rise to the very long spicular processes which fringe the margin of the sponge. At their origin these bands are very broad, but they rapidly taper away to hair-like thinness {vide woodcut, Fig. 5, a'). Spicules. — (a) Megasclera ; of one kind only, viz., straight, slender styli, which may attain a length of over 3"5 mm. (b) Mic7-o.sclerct ; (l) anisochelae (PI. XXI. fig. 8) of the usual Cladorhiza type, with the curved shaft much expanded towards the larger end, and with three teeth at each extremity ; length about 0'038 mm. (2) Sigmata (PL XXI. fig. 10), more or less contort, of fair size, and with the ends produced into long, slender, whip-like processes, much as in Cladorhiza ahyssicola, var. co7'tico- cancellata, Carter; length of spicule (from bend to bend) about 0-0756 mm.; these spicules are rather rare. (3) Amphiasters (PL XXI. fig. 9), large and very remarkable spicules, each consisting of a straight shaft with a rosette of five teeth at each end ; these spicules vary considerably in size, reaching about 0'23 mm. in length. They form a very thick, dense layer [vide woodcut. Fig. 5, c), encrusting the entire body, both upper and lower surfaces, and the upper portions of the root and of the long radiating processes. They are confined to this external layer, in which they are very closely packed together without much order. As regards external form this species is almost indistinguishable from a species of Cladorhiza obtained at Station 281, which we have called Cladorhiza similis, so that without microscopic examination one would at once put them downi as identical ; microscopic examination, however, shows the spiculation to be widely diff'erent, as will be seen by reference to the description on p. 93. We are inclined to regard the presence of extra microsclera (if so they can be called) as a special modification introduced to suit the special requirements of the species, and hence they would be of less value in classification than the ordinary microsclera, the fuU complement of which is also present. The very dense external armature of spicules is doubtless a most efficient protection against the inroads of parasites or other enemies. That allied sponges are subject to such attacks Mall be seen by reference to the description of Cladorhiza longipinna (p. 93). Locality. — Station 291, October 27, 1875 ; lat. 39° 13' S., long. 118° 49' W. ; South Pacific ; depth, 2250 fathoms ; bottom, red clay ; bottom temperature, 34°"6. Two good specimens. REPORT ON THE MONAXONIDA. 99 Genus Chondrocladia, Wyville Thomson (Pis. XX., XXL). 1873. Chomlrodadia, Wyville Thomson, The Depths of the Sea, p. 188. 1 880. Clailorhua (j)artt), Schmidt, Die Spongien des Meerbusen von Mexico, pt. ii. p. 83. Of varying, but usually symmetrical form. Skeleton usually consisting of a central, erect axis of spiculo-fibre, which may or may not be branched, and from which arise longer or shorter processes, also composed of spiculo-fibre. The chief megasclera are stylote and often attam a great length. The characteristic microsclera are isochelse, with three or more teeth at each extremity and a curved shaft expanded laterally near each end into wing-like processes. Sigmata may also be present. The genus was founded in 1873 by Sir Wyville Thomson (loc. cit.) for his Chondrocladia virgata ; but he gives no generic diagnosis. The two genera Chondrocladia, Wyville Thomson, and Cladorhiza, M. Sars, appear to be closely allied, differing only in the form of the chelae, which in the one case have the two ends unequal and in the other equal, and, as already pointed out (p. 86), Schmidt includes both under the name Cladorhiza. The Chondrocladia^ are also deep-sea sponges, and may acquire the " Crinorhiza " form ; they are not nearly so well represented in the collection as is the genus Cladorhiza. The geographical distribution of the genus is extended by the results of the Challenger voyage from the North Atlantic to the Southern Ocean and the Pacific, and the vertical distribution to nearly 3000 fathoms. Chondrocladia concrescens (?), Schmidt, sp. (PI. XX. fig. 12 ; PI. XXI. figs. 7, 7 a, 12). 1880. Cladorhiza concrescens, Schmidt, Die Spongien des Meerbusen von Mexico, pt. ii. p. 83, pi. X. figs. 8, 9. Want of evidence, due both to the fi-agmentary nature of the Challenger specimens and to the incompleteness of Schmidt's original account, makes the identification of this species doubtful. The Challenger specimens (PL XX. fig. 12) consist of five stalks, four of which have their ends inserted into sockets in a thick, fleshy mass strongly resembling mammalian hyaline cartilage in appearance, while their other ends are free. The fifth stalk is separate from the others and has a lobe-like, fleshy mass adhering to one end. It is the narrower ends of the stalks which are inserted into the soHd fleshy masses, while the broader ends are free, but have evidently been attached ( ? to the main stem) when the 100 THE VOYAGE OF H.M.S. CHALLENGER. sponge was living. Length of the stalks about 50 mm., with a diameter at the broad end of about 2'5 mm. Spicules. — [a) Megasclera ; slender styli (PI. XXI. figs. 7, 7a), which may attain in the stalks a length of over 4*5 mm., with a diameter of about 0-07 mm. (h) Micro- sclera; isochelae (PI. XXI. fig. 12), with long shaft and six or seven teeth at each end ; length about 0-094 mm. Shaft slightly expanded towards each end. The sigmata are virtually absent though two or three very slender ones were seen, measuring about 0'056 mm. in length, which may or may not be proper to the sponge. Oscar Schmidt also mentions no sigmata in his specimens (loc. cit.) ; his remarks about the shape and size of the microsclera run as follows : — " Es sind zwei Sorten von Doppelankern vorhanden. Die eine, kleinere von 0-02857 bis 0-0311 Mmtr. zeichnet sich durch unver- haltnissmassig lange Zahne aus, deren Enden sich fast beriihren. Noch eigenthiimlicher ist die andere, besonders grosse, von 0-07142 bis 0-12 Mmtr. Sie besitzt namlich keinen Mittelzahn, sondern statt dessen ein Paar Ziihne, daneben jederseits noch zwei, also im Ganzen sechs Zahne. Im Schaft, der in der Nahe der Zahne, wie so oft, mit seitlichen Ausbreitungen versehen ist, sieht man schon bei massiger Vergrosseruug den Axeneanal." Possibly the small chelae here described are young forms of the others, we have not found any in our specimens. Localihj. — Station 248, July 5, 1875 ; lat. 37° 41' N., long. 177° 4' W.; North Pacific ; depth, 2900 fathoms ; bottom, red clay ; bottom temperature, 35°-l. Fragments. Chondrocladia clavata, Ridley and Dendy (PL XX. figs. 1, la; PI. XXI. fig. 11). 1886. Clwndrocladia clavafa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 345. Sponge (PI. XX. figs. 1, la) club-shaped, consisting of a very small, globular body perched on the end of a slender stalk. From various parts of the body radiate long, .slender processes. The stalk is short, and at the bottom breaks up into a tuft of rootlets. Diameter of body 2 mm.; length of stalk and rootlets 12 mm. (these measurements were taken from the smaller of two specimens). Colour in spirit pale yellow. Skeleton. — The skeleton consists of a main axis of spiculo-fibre, breaking up below into a number of smaller fibres forming the rootlets, and giving off above a number of radiating fibres which project for a considerable distance beyond the body of the sponge. Spicules. — (a) Megasclera ; slender styli, sometimes slightly tylostylote, varying in length according to their position in the sponge. In the main fibre of the skeleton they measure about TO by 0-022 mm.; numerous smaller ones occur scattered loosely through the soft parts of the sponge. (b) Microsclera; (l) numerous isochelae EEPORT ON THE MONAXONIDA. 101 (PL XXI. fig. 11) of the usual Cliondrocladia type, each with a curved shaft expanded towards each end and with three claw-like teeth at each extremity. Length about 0'057 mm. (2) A few sigmata, simple and contort, about 0*044 mm. long. It wiU be seen that this sponge makes a near approach to the " Crinorhiza " form, but does not quite attain to it. It is chiefly remarkable on account of its minute size and the exceedingly small amount of soft tissues present. It was obtained in com- paratively shallow water (140 fathoms), but on a bottom of mud. Locality. — Station 174, August 3, 1874 ; lat. 19° 6' 0" S., long. 178° 14' 20" E.; Kandavu, Fiji Islands; depth, 140 fathoms; bottom, coral mud. Two specimens. Chondrocladia crinita, Ridley and Dendy (PI. XX. fig. 4 ; PI. XXI. figs. 3, 17). 1886. Chondrocladia crinita, Ridley and Dendy, Ann. and. Mag. Nat. Hist., ser. 5, vol. xviii. p. 345. Sponge (PI. XX. fig. 4), of the " Crinorhiza " form, consisting of a rather large, conical body, terminating above in a stout, spike-like projection, composed of spiculo- fibre. The base of the sponge is fringed by a number of long, coarse, hair-like processes, and from near its centre projects a short papilla. Diameter of body at base 15 mm.; height of body 12 mm. Colour in spirit brownish-yellow. Surface ribbed by the projection of the upper parts of the fringing processes, which are very prominent. Dermal memhrane distinct, rather thick ; containing numerous microsclera, especially isochelae, but no megasclera. Skeleton. — Of the " Crinorhiza " type, consisting of a central axis of dense spiculo- fibre, forming the spike at the summit of the sponge and not developed below ; with radiating bands of spiculo-fibre which form the axes of the fringing processes. Spicules. — (a) Megasclera ; slender styli (PI. XXI. fig. 3), measuring in the main fibres about 2'2 by 0"044 mm. (6) Microsclera ; (l) numerous large isochel^ (PI. XXI. fig. 17) of the usual Chondrocladia form, with curved shaft expanded towards each end and three teeth at each extremity. Length about O'l mm. (2) Slender, more or less contort sigmata ; very abundant in the dermal membrane ; length about 0'07 mm. This sponge is of especial interest on account of its having acquired the " Crinorhiza " form. Unfortunately only a single specimen was obtained, and that in very bad con- dition. On cutting it open a number of embryos were found imbedded in the soft tissues near the centre of the sponge. Localitij. — Station 216a, February 16, 1875; lat. 2° 56' K, long. 134° 11' E.; north of New Guinea ; depth, 2000 fathoms ; bottom, Globigerina ooze ; bottom temperature, 3 5°' 4. One specimen. 102 THE VOYAGE OF H.M.S. CHALLENGER. Genus Meliiderma,'^ n. gen. (Pis. XX., XXL). 1886. Cliondrocladia {/ ars), Ridley aud Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 3i4. Sponge stipitate. Megasclera stylote to tylostylote. Microsclera, isochelse of the ordinary Cho7idrocladia form, to which may be added sigmata. In addition to these forms of spicules a sjDecial protective spicule occurs encrusting the stem, each spicule consisting of a slightly expanded base of attachment, a constricted neck, and a fusiform, outwardly projecting shaft, terminating in a pointed apex {vide annexed woodcut, Fig. 6). This genus stands in much the same relation to Chondrodadia as does Axoniderma to Cladorhiza, having, like Axoniderma, developed a special spicule for external defensive purposes. The form of the extra spicule is, however, totally different in the two cases ; and in Meliiderma it appears to be confined to the stalk, while in Axoni- derma it is most abundant in the body of the sponge. In both these cases it is very difficult to say whether the extra spicule ought to be classed amongst megasclera or microsclera ; for the sake of convenience both will here be regarded as microsclera, though standing on quite a different footing from other microsclera. The form of the extra microsclera in Meliiderma stipitata is as j'et unparalleled. From their re- semblance to spears or darts we have derived the generic name. Meliiderma stipitata, Ridley and Dendy (PL XX. fig. 11 ; PI. XXI. figs. 6, 6a, 6&, Qc, 14). 1886. Chondrodadia stipitata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 344. Sponge (PL XX. fig. 11) consisting of a spherical head perched on the end of a long stalk, which in the one perfect specimen is much bent. Diameter of head about 12 mm. Length of stalk in the perfect specimen, 25 mm., diameter about 2 mm. At the base the stalk terminates in a flattened expansion which is firmly attached to a fragment of black volcanic cinder. Colour in spuit pale yeUow. Texture of head soft. Surface hispid. Dermal membrane fairly distinct. Skeleton. — There is no dermal reticulation, and the main skeleton consists chiefly of a very dense band of spiculo-fibre, forming the greater part of the stem ; this penetrates the spherical head for a short distance and then expands slightly into a knob-like extremity, from which radiate thin bauds of loose spiculo-fibre in aU directions, running vertically to the surface. Numerous spicules jaroject for some distance beyond the surface of the sponge, giving to it its hispid appearance. Spicules. — (a) Megasclera ; of one kind only, viz., long tylostyli (PL XXI. figs. 6, 6a, 66, 6c), often a little crooked, measuring up to 2-2 by 0'038 mm.; head, as a rule, only very slightly indicated, apex sharply pointed ; the spicules are broadest in the 1 Greek litkla,, a spear, and otj^ta, the skin. REPORT ON THE MONAXONIDA. 103 middle and taper gradually towards both ends, (b) Microsclera ; (l) large Lsochelfe (PI. XXI. fig. 14) with curved shaft and five prominent claw-like teeth at each end ; the shaft is considerably expanded laterally near each end, but contracts again before the points of attachment of the teeth. Length of spicule aljout 0-085 mm. The chelate spicules are especially abundant in the dermal membrane and around the embryos. (2) Slender sigmata, often contort, measuring about 0'055 by 0'016 mm.; these s]3icules are com- paratively scarce. (3) The spear-like spicules (vide woodcut annexed) ; these are densely packed in a single layer around the stalk, each with its apex projecting vertically outwards. Only one perfect specimen of this sponge and one damaged one are in the collection ; in both of these the stalk is encrusted by a thin, yellow, velvet-like layer of the densely packed, spear- like spicules. At first we thought that these spicules belonged . . . . ™ ^"'- 6' — Meliidenna ' stipitata. to some foreign encrustmar sponsfe, and hence no mention of spicules encrusting the stem. , . . . X 400. them was made in our Preliminary Report, but subsequent examination has convinced us that they are proper to the species ; the fact of their occurring in both specimens being strong evidence in favour of this view. An interesting point about the sponge is the position in which the embryos develop. In the specimen which we cut open, a number of round yellow bodies were found, each enclosed in a membranous caj)siile around which the chelate spicules were lying in very great numbers. These embryos were placed in a zone of tissue occupying about the centre of the spherical head, and lying immediately above the expanded termination of the stalk, between the bands of spiculo-fibre which radiate from it ; this being obviously the position of greatest security. Localitij. — Station 147, December 30, 1873 ; lat. 4G° 16' S., long. 48° 27' E.; between Prince Edward Island and Crozet Island ; depth, 1600 fathoms ; bottom, Diatom ooze ; bottom temperature, 34°'2. Two specimens. Genus Desmacidon, Bowerbank (Pis. XXIIL, XXIV., XXX., XLVIL). 1842. Halichondria (pars), Johnston, British Sponges, p. 88. 1864. Desmacidon, Bowerbank, Men. Brit. Spong., vol i., p. 200. Form various. Distinct spiculo-fibre, with a large amount of uniting spongin, may or may not be present. Megasclera diactinal, oxea or strongyla ; microsclera isochelse, and, at any rate usually, sigmata. The genus is usually made much more comprehensive than the above diagnosis would permit. It has thus become unwieldy and artificial, including forms which are much too dissimilar to be placed together. It was founded by Bowerbank, who gives the following generic diagnosis (loc. cit.) — "Skeleton fibrous, irregularly reticulated. 104 THE VOYAGE OF H.M.S. CHALLENGER. Fibres composed entirely of spieula arranged in accordance with the axis of the filjre, cemented together and thinly coated with keratode." This diagnosis is even more indefinite than that of Vosmaer, which runs as follows : — " Rods smooth or spined. Anchors bi- or tridentate, also palmato -dentate, equiended. ' Keratode-fibre ' very conspicuous." -^ Under these circumstances the only thing left for us to do was to take Bowerbank's type of the genus, viz., Desmacidon fruticosa, and unite with it those sponges which have a similar spiculation (as given above). We are conscious of the jjossibility that our diagnosis may prove to be too limited, but there can, we think, be little doubt that whatever sponges fall within its limits will form a natural group, which can scarcely be said of the two previous diagnoses. The genus, even as above restricted, has a very wide geographical range, being found in British and Australian Seas and off the east coasts of North and South America, and, if we include the subgenus Homceodictya, as far south as Kerguelen Island. Desmacidon fruticosa, Montagu, sp., var. (PI. XXIII. figs. 10, 10a, 106, 10c, lOc^ ; PL XXX. fig. 1). 1818. Sporujiu fruticosa, Montagu, Mem. Wern. Soc. EcLin.,vo). ii. p. 112, pL xiv. figs. 3, 4. 1842. Haliehondria fruticosa, Johnston, British. Sponges, p. 103. 1864. Desmacidon fruticosa, Bowerbank, Mon. Brit. Spong., vol. L p. 200. AVith this species we unite, as a variety, an interesting specimen from the south-east of Australia. The sponge (PI. XXX. fig. 1) arises from a short, stout peduncle, and expands into a broad, somewhat compressed palm ; its height is about 62 mm. It agrees with Bowerbank's specimens in having a distinct fibre, and, still more closely, in spicu- lation. The spicules are oxea (PI. XXIII. figs. 10, 10a, lOcZ), tridentate isochelae (PL XXIII. figs. 106, 10c), and sigmata. The oxeote is of about the same size (about 0'25 mm. long) in both, but in the Challenger variety is jjerhaps not quite so abruptly pointed and without any marked inequality of the two ends such as is often visible in Bowerbank's specimens. In the Challenger variety, again, the chelae are slightly larger and of stouter build than in the British form (measuring about 0"038 as against 0"0315 mm.), and the sigmata also appear to attain a slightly greater size (reaching up to about 0'063 mm. in length). The surface of the sponge is scored with a few deep, broad, longitudinal grooves or channels (PL XXX. fig. 1), mostly confined to one side. It is doubtful whether these grooves are natural or whether they have been caused by the sponge growing up against some cylindrical, branching organism. Localitij. — Station 163a, April 4, 1874; lat. 36° 59' S., long. 150° 20' E.; off the south-east coast of Australia ; dejDth, 120 fathoms ; bottom, green mud. One specimen. 1 Notes from the Leyden Museum, vol. ii., 1880, p. 130 ; see also Porifera, in Bronn's Klass. u. Ordnung. des Tliierreichs, p. 350. ■REPORT ON THE MONAXONIDA. 105 Desmacidon reptans, Ridley and Dendy (PI. XXIII. figs. 7, 7a, 7h). 1886. Des7naculon reptans, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 345. Sponge encrusting, creeping over other sponges of different species, or free ; massive, amorphous, or digitate. Colowr in spii-it greyish-yellow. Texture fairly firm, resilient. Surface glabrous-looking, but slightly rough to the touch, marked with shallow and for the most part longitudinal grooves. Dermal membrane thin and delicate. Oscula small, scattered. Pores small, scattered. Skeleton. — (o) Dermal; varying in its degree of development according to the particular specimen examined. Sometimes it is composed of a comparatively wide- meshed reticulation of stout fibre, composed of broken spicules and foreign bodies, while the polygonal meshes of this network are again filled in by the much smaller polygonal meshes of a finer network, composed of more slender fibre, but of the same character. The meshes of the finer reticulation are only about 0'07 mm., while those of the coarser reticulation may be 0'35 mm. in diameter. The finer reticulation appears to be at a slightly higher level than the coarser one. At other times the two reticulations are not distinguishal^le from one another, the whole dermal skeleton is more irregular, and the foreign bodies are in great part or entirely replaced by the proper oxeote spicules of the species ; while again at other times the number of foreign bodies present in the fibres may be very great, consisting chiefly of sand grains. (6) Main ; an irregular, fairly close, somewhat Isodictyal reticulation of oxeote spicules, often several lying parallel side by side, but rarely united into distinct fibres. Spicules. — («) Megasclera ; of one kind only, viz., smooth oxea (PI. XXIII. figs. 7, 7a), tapering rather abruptly to a sharp point at each end; size about 0'18 by 0-008 mm. (b) Microsclera ; of two kinds; (l) isochelse (PI. XXIII. fig. 7b), with three sharp teeth at each end, length about 0'019 mm., rather rare ; (2) numerous simple or contort sigmata, about 0'038 mm. long. In spiculation this species comes near to Desmacidon fruticosa, Montagu, sp., the type of the genus.-^ It difi"ers, however, in the much slighter development of horny matter in the fibre, and in tlie much smaller size of the chelate spicule (0'032 mm. long in Desmacidon fruticosa as against 0"019 in Desmacidon reptans). The habit appears also to difi'er considerably. The sponge afi"ords a good example of the way in which foreign bodies may replace the proper skeleton, being arranged along the true skeleton lines. Locality. — Ofi" Bahia, shallow water ; two specimens. Off Bahia ; 7 to 20 fathoms ; three specimens and some pieces. ' Cf. Bowerbank, Mon. Brit. Spong., vol. i. p. 200, &c. (ZOOL. CHALL. EXP. — PART Lix. — 1887.) Nnn 14 106 THE VOYAGE OF H.M.S. CHALLENGER. Desmacidon conulosa, Eidley and Dendy (PI. XXIII. f:gs. 5, 5a, 56, 5c; PI. XXIV. fig. 2). 1886. Desmacidon conulosa, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 346. S^wuge (PI. XXIV. fig. 2) consisting, in the case of the larger specimen, of a tough peduncle, expanding and dividing above into two broad, flattened, palmate lobes. Height of specimen 94 mm.; thickness of the lobes 8 mm. Colour in spirit greyish- yellow. Texture firm, tough, resilient. Surface rugose, owing to numerous, thickly placed conuli, which project above the dermal membrane ; these are absent from the peduncle. Dermal onembrane thin, delicate, transparent. Pores scattered. Oscula small, about 1-5 mm. in diameter, numerous, scattered on both flattened surfaces and a few along the upper margin of the sponge. Skeleton. — Composed of a coarse reticulation of stout spiculo-fibre ; the fibres being formed of the oxeote spicules with only a very small proportion of uniting horny substance. The main fibres run more or less vertically to the surface and there terminate in compact, slightly divergent tufts of spicules — the conuli above mentioned. Spicules. — (a) Megasclera ; of one kind only, viz., very stout, fusiform oxea (PL XXIII. figs, 5, 5a), tapering gradually to a very sharp point at each end. Size about 07 by 0-057 mm. These spicules occur loosely scattered in the soft tissues of the sponge, as well as in the fibres. (b) Microsclera ; small palmate isochelaj (PI. XXIII. figs. 5b, 5c), with large anterior palms and lateral palms only slightly developed. Length about 0'032 mm. A great number of small, spherical, yellow embryos occur in the deeper tissues of the sponge. This species exhibits a close relationship to Desmacidon compressa, Esper (Ehlers), but difi"ers in the arrangement of the oscula, which, in Desmacidon compressa, are said to be large and confined to the upper margin of the sponge. In the description of Desmacidon compressa we find also no mention of the remarkable and very characteristic conuli present in the Challenger species. This species forms a connect- ing link between the more typical species of the genus and the subgenus Homceodictya, and in some respects comes very near to Desmacidon {Homceodictya) grandis, nobis {vide p. Ill), from the same locality. Locality. — Simon's Bay, Cape of Good Hope, 10 to 20 fathoms. One large specimen, from which the measurements given above are taken, and one small one. REPORT ON THE MONAXONIDA. 107 Desmacidon (f) ramosa, Ridley and Dendy (PI. XXIII. figs. 4, 4fl, 46, Ac ; PI. XXIV. fig. 4 ; PI. XLVII. fig. 6). 1886. Desmacidon {?) ramosa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 346. Sponge (PI. XXIV. fig. 4) consisting of numerous, irregular, long, vermiform, anas- tomosing branches, about 6 mm. in diameter. One piece in the collection is about 212 mm. long, but there seems no reason why it should not grow to a much greater length. Texture tough and leathery. Colour in spirit pale greyish-yellow. Surface uneven and minutely hispid, often with a reticulate appearance. There is a distinct, tough, outer rind, which, however, adheres rather firmly to the underlying tissues. Dermal memhrane very distinct, transparent. Oscula scattered up and down the branches at irregular intervals ; their margins are produced into conical protuberances about 2 mm. long. On some branches the oscula are numerous, while in other parts they appear to be almost or entirely wanting. Pores scattered between the radiating brushes of spicules which support the dermal membrane. Skeleton. — Of a distinctly radiate type. There is a more or less definite axis of spiculo-fibre running through each branch, from which radiate, though in no very definite manner, numerous bands of spiculo-fibre ; these run towards the surface of the sponge, and on approaching it, break up into divergent tufts of spicules which support the dermal membrane and sometimes project beyond it, causing the hispidity of the surface (PI. XLVII. fig. 6). There is no dermal reticulation, Spictdes. — (a) Megasclera ; of one kind only, viz., sharp-pointed, fusiform oxea (PI. XXIII. figs. 4, 4«), measuring about 0'6 by 0"022 mm. (h) Microsclera ; of one kind only, viz., tridentate isochelse (PI. XXIII. figs. 4&, 4c), the shafts of which appear to be extended into slight lateral fimbriae. Length about 0*02 mm. It will be seen from the above that in spiculation this species would be reaUy a typical Desmacidon, were it not for the absence of the sigmata. These, however, appear from the study of other genera to be such very variable elements in the s2:)iculation that they cannot be considered as of generic importance. More important, indeed, is the radial arrangement of the skeleton, very difi'erent from that of other species of the genus and, indeed, of most of the family, and making a near approach to the arrangement of the skeleton in the genus Rasi>ailia. Loccdities. — Station 142, December 18, 1873, lat. 35° 4' S., long. 18° 37' E.; south of the Cape of Good Hope; depth, 150 fathoms; bottom, green sand; bottom temperature, 47°"0. Several specimens. Off Marion Island, 50 to 75 fathoms. One specimen, of much darker (greyish) colour than the previous ones. 108 THE VOYAGE OF H.M.S. CHALLENGER. Subgenus Homceodictya, Ehlers (Pis. XXIL, XXIIL, XXIY., XXIX.). 1797. Spongia (pars), Esper, Die I'flanzenthiere, Fortsetz. 1 p. 190, pL 1. 1842. HaHcho7idna {parit), Johnston, British Sponges, p. 92, pi. ii. 1866. Isodidya (pars), Bowerbank, Mon. Brit. Spong., vol. ii. p. 311 ; vol. iii. pi. lii. 1867. „ (pare), Gray, Proc. Zool. Soc. Lond., May 1867, p. 534. 1870. Hormjeodietya, Ehlers, Die Esper'schen Spongien, Eriangen, 1870, p. 17. 1882. Ghalina {pars), Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. x. p. 111. Form various, usually lobate or palmate. Megasclera oxea. Microsclera isochelse, of a very peculiar type. The latter are the characteristic spicules of the subgenus ; the shaft is slightly curved and may be expanded all the way along, from end to end, into continuous, dehcate, lateral fiinbria3, which may be more or less curved ; in other words, the two lateral palms on each side of the spicule may remain united together, and form merely terminal portions of a continuous lateral fimbria. The anterior palm of the spicule is distinct and somewhat oval in shape, usually slightly curved outwards at the free end ; from the median line of the posterior surface of each anterior palm there projects backwards, i.e., towards the shaft, a delicate, flat fimbria, pear-shaped in out- line ; this gives to the anterior palm the appearance of being forked when the spicule is viewed laterally, and is the distinguishing feature of the subgenus {vide PL XXIX. fig. 7a). Although of a very peculiar form, the isochela of Homceodictya does not really differ so much from an ordinary isochela as might at first sight be expected. We have only to cut away the central portion of each lateral fimbria (in those cases where it persists), leaving the ends as lateral palms (connected, of course, as usual, with the shaft for their whole length), and to abolish the pear-shaped process which remains adherent to the back of the anterior palm, and probably represents the remains of a connection between the latter and the shaft, and we have left a typical palmate isochela, like that, for example, of Desmacidon compressa.^ The history of this interesting subgenus is not a little involved, and wdl give some idea of the literary difficulties which the spongologist has to encounter. Johnston, in his British Sponges (p. 92, pi. ii.) describes and figures a sponge under the name Halichondria palmata, and gives in his list of synonyms, " Spongia palmata, Sihh. Scot. ill. ii. 'lib. iv. 55, &c." With some difficulty we found the book thus referred to,^ but the only reference we found to sponges consists of the two words " Spongia palmata;" and we can hardly retain this as a synonym on so little evidence. Bow^erbank, in his Monograph of British Spongiada3 {loc. cit.), describes and figures his Isodictya p)almata, which he identifies with Halichondria pahnata, Johnston. He 1 Cf. also the descriptions given by Mr. Carter of the isochela oiHcmueodictya (Chalina) palmata and of Desmacidon cpmpressa, Ann. and Mag. Nat. Hist., ser. 5, vol. x. p. Ill, fig. 1, a and b, and p. 113. 2 Nuncius Scoto-Britannus, sive Admonitio de Atlante Scotico &c., by Sir Robert Sibbald, Edinburgh, 1683. The booli is also known as Scotia illustrata. REPORT ON THE MONAXONIDA. 109 here for the first time describes and figures the peculiar chelate spicule, but makes uo mention of such spicules in Johnston's sponge, nor does Johnston himself mention them in his original description [loc. cit.). Fortunately, ]\'Ir. H. J. Carter ^ has found these spicules in Johnston's type in the British Museum, and thus the identification of Bowerbank's with Johnston's species is confirmed ; though Mr. Carter places them both in the genus Chalina. Dr. Gray in his remarkable paper in the Proceedings of the Zoological Society of London (May 1867), appears to recognise the fact (which is undoubtedly true) that Dr. Bowerbank's genus Isodictya is much too comprehensive ; he therefore very wisely splits the genus up into several genera, but unfortunately in so doing he retains the name Isodictya for two of these distinct genera, giving to each a distinct generic diagnosis. With his first genus Isodictya (loc. cit., p. 512) we have here nothing to do. Of his second (p. 534) he gives the following diagnosis. " Sponge sessile, minutely hispid, regularly reticulated. Spicules of three kinds : — 1. Needle- shaped, fusiform. 2. Bihamate, bicalcarate (Bowerb., f. 121). 3. Equianchoratc or palmate." Of this genus he gives Isodictya normani, Bowerbauk, as the type, and says, " see also ..../. palmata, Bowerb." Unfortunately Isodictya palmata, Bowerbank, has only two kind of spicules. Dr. Ehlers (Die Esper'schen Spongien, p. 16) identifies Esper's old species, Spongia digitata,'^ with Johnston's Halichondria 2Jabnata and Bowerbank's Isodictya p)cdmata. He recognises (pp. 35, 36) that both Gray's genera Isodictya cannot stand, and creates for Esper's species, digitata, the new genus, Homceodictya, without, however, giving a fresh generic diagnosis, and apparently intending the new genus to replace Gray's second Isodictya. Here, then, arises a considerable uncertainty as to the correct nomenclature, not only of the genus, but also of the species known as palmata. For the present we retain the name Homceodictya for the subgenus, and include therein only three species, viz., Desmacidon (Homceodictya) palmata, Johnst. (1 = Spongia digitata, Esper), and the two new species Desmacidon (Homceodictya) kerguelen- ensis, nobis, and Desmacidon (Homceodictya) grandis, nobis, all characterised by the peculiar form of the isochelate microsclera. The above does not pretend to be a complete history of the subgenus, which time and space do not permit, and it will be seen by reference to Johnston (loc. cit.) that there is still a considerable amount of literature to which we have not alluded. It appears to us that the characters separating the three species of Homceodictya from Desmacidon are only of subgeneric value, one very strong piece of evidence in favour of this view being the intermediate position of Desmacidon conidosa, nobis. The peculiar backward process of the anterior palm of the chelate spicule is a 1 Ann. and May. Nat. Hist., ser. 5, vol. x. p. 110. 2 Esper, Die PflanzentHere, Fortsetzimgen, i. p. 190, pi. 1. 110 THE VOYAGE OF H.M.S. CHALLENGEE. character wliieh would seem to be not confined to this subgenus, for Mr. Carter fisrures a similar structure in the isochelate spicules of his Microciona affinis;^ further information concerning the latter species is, however, much needed, especially with regard to the megasclera. The new Challenger species are very closely related to that of Johnston and Bowerbank, and it is exceedingly interesting to meet with such intimately related forms in such distant localities as Great Britain, Kerguelen, and the Cape of Good Hope. Desmacidon [Homceodictya) herguelenensis, Ridley and Dendy (PI. XXIII. figs. 3, 3a, 36, 3c ; PL XXIV. fig. 3). 1886. Hommodidya herguelenensis, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5. vol. xviii. p. 346. Sponge (PI. XXIV. fig. 3) lobate or digitate. The larger of the two specimens obtained consists of a short peduncle, narrower towards the base and then expanding and dividing into two divergent, stout, finger-like lobes of very unequal size. Height of the sponge 69 mm. Diameter of lobate processes 13 mm. Colour in .spirit light brownish-yellow. Texture soft, spongy and resilient. Surface rough to the touch, woolly-looking and minutely hispid. Dermal membrane delicate, transparent. Oscida few, small and scattered. Pores irregularly scattered, abundant in parts, diameter about O'l mm. Skeleton. — Very loose and Ul-defined, consisting of an irregular, somewhat Isodictyal reticulation of oxeote spicules, occasionally forming fibres. Spicules. — (a) Megasclera; of one kind only, viz., simple oxea (PI. XXIII. figs, 3, 3a) short and rather stout, tapering gradually to a sharp point at each end ; size about 0'35 by 0"019 mm. (6) Microsclera; the peculiar palmate isochelae of the genus {vide supra; and PI. XXIII. figs. 3&, 3c); length about 0'028 mm. We were at first inclined to regard this sponge as a variety of the British species, Desmacidon [Homceodictya) palmata, which it very nearly approaches both in external form and in spiculation. There can be no doubt that the two are closely related, but on the whole it appears better to separate the Kerguelen form as distinct. The main difi'erences between the two concern (l) the size of the oxeote spicules, which in Desmacidon (Homceodictya) j)almata (according to Bowerbank's preparations in the British Museum) are little more than half as long as in the Kerguelen form ; (2) the skeleton fibre, which is more distinct and highly developed in Bowerbank's than in the Challenger specimens. The isochelse are of about the same size in the two specie.*, and present no important structural difi"erences. The ends of the anterior palms are, as a rule, more turned out in the British species, and the angle which they make with 1 Ayin. and Mag. Nat. Hist., ser. 5, vol. vi. p. 41, pi. iv. fig. 15. REPORT ON THE MONAXONIDA. Ill the backwardly projecting process is greater than in the Kerguelen form. JMr. H. J. Carter ^ has given a very good description of this spicule in Desmacidon [Homceodictya) palmata, with two excellent figures, the correctness of which was most satisfactorily verified by reference to the Bowerbank collection in the British Museum ; it appears to us, however, that the lateral fimbrise of the shaft are usually broader than he has repre- sented them, and the spicule thus makes a still closer approach to that of Desmacidon (Homoeodictya) kerguelenensis than might be judged from his figure. Locality. — Royal Sound, Kerguelen, 25 fathoms. Two specimens. Desmacidon {Homceodictya) grandis, Ridley and Dendy (PI. XXII.; PL XXIX. figs. 7, 7a). 1886. Homceodictya grandis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 347. Sponge (PI. XXII. fig. 1) compressed, lamelliform ; presumably of erect growth. The single specimen in the collection consists of a large, very broad, very much flattened lamella, at one end proliferating into compressed, digitate branches, at the other con- sisting of a simple, flat expansion with incipiently lobate margin. One side of the frond bears very numerous, steUately arranged oscula, placed at very regular intervals from one another ; these form a very prominent feature of the sponge ; they are entirely absent from the other side (cf. PI. XXII. figs, la, Ih). The specimen is of rather irregular shape and has been broken ofi" along its lower margin, so that we cannot tell how far it extended in a vertical direction ; in its present condition it is much broader than it is high, the greatest breadth being about 275 mm. (the lower margin, however, is only about 162 mm. long) and the greatest height about 150 mm. The thickness at the broader end of the piece is about 6 mm. and at the narrower end about 2"5 mm. Colour ia spirit greyish-yellow. Texture firm, tough, fibrous, resilient. Surface on both sides very fairly even, but minutely conulose and very minutely hispid ; conuli most prominent towards and on the free margin ; very harsh to the touch. Dermal membrane thin, transparent, delicate, stretched between the conuli. Oficula minute, arranged in stellate groups of about four each, on one side only of the frond (PL XXII. fig. lb); each group about 2 mm. in diameter; these groups of oscula are very abundant and arranged at fairly constant intervals of about 6 mm. from one another. Pores not found. Skeleton. — Very well developed ; composed of stout, branching, Axinella-Yike spiculo- fibre, made up of stout oxeote spicules. These spicules have each one end imbedded in the fibre, while the other projects outwards, towards the surface of the sponge, at a slight angle. The fibres trend towards the surface, where they terminate in tufts of oxeote spicules, thus giving rise to the surface conuli. Parallel fibres may in places be connected by loose crossing spicules. 1 Ann, and Mag. Nat. Hist., ser. 5, vol. x. p. Ill, fig. 1, a and h. 112 THE VOYAGE OF H.M.S. CHALLENGER. Spicules. — (a) Megasclera ; stout, smooth, fusiform oxea (PI. XXIX. fig. 7), bent at more or less of an angle in the middle, sharply and gradually pointed, size about 0*45 by 0"04 mm. (a few small, slender styli were also observed, but as these are almost certainly foreign we shall say nothing further about them), (b) Microsclera ; of one kind only ; viz., very abundant isochelse (PI. XXIX. fig. 7a), with well-developed front palms, giving off each a distinct, stout, backward process. The lateral palms are united with the shaft up to their ends (i.e., up to a level with the end of the front j)alm), but the middle portion of the shaft is not fimbriated as in Desmacidon {Homceodictya) herguelenensis ; they are large, handsome spicules about 0"063 mm. long; not infre- quently there is an irregular swelling in the central portion of the shaft. This species differs from Desmacidon (Homceodictya) Tcerguelenensis , nobis, in the arrangement of the skeleton, in the larger size of both megasclera (which are here much stouter) and microsclera, in the non-fibriation of the shaft of the latter, and very markedly in external form. In the arrangement of the skeleton it agrees exactly with Desmacidon conidosa, nobis (vide p. 106), and this gives to the two a very similar look ; the two species differ, however, externally, in the arrangement of the oscula and, microscopically, in the form of the isochelate spicules ; yet we think the two come very near together, and in general appearance they are much more like one another than are Desmacidon (Homceodictya) grandis and Desmacidon (Homceodictya) ker- guelenensis (cf. PI. XXII. fig. 1, with PL XXIV. figs. 2 and 3). Locality. — Simon's Bay, Cape of Good Hope, November, 1873 ; 10 to 20 fathoms. One specimen. Genus Artemisina, Vosmaer. ? 1885. Artemisina, Vosmaer, Sponges of the " Willem Barents" Expedition, 1880 and 1881, p. 25. Body compact, texture like that of Suberitidse. Megasclera monactinal ; styli or subtylostyli. Microsclera isochelae and spined toxa, Vosmaer's original diagnosis runs as follows : — " Body covered with a thin dermis forming here and there thin oscular tubes. Sponge-mass rather compact, about as Suberites. Generic spicules tr. ac | A sp. | anc^. | The only species which we include in the genus is the original type, Artemisina suheritoides. Artemisina suheritoides, Vosmaer. ? 1885. Artemisina suheritoides, Vosmaer, Sponges of the "Willem Barents" Expedition, 1880 and 1881, p. 25, pi. i. fig. 16; pi. iv. figs. 11-14; pi. V. figs. 51-55. As the species has already been fairly fully descril^ed and figured, we can dismiss it very briefly. It is extremely unfortunate that Dr. Vosmaer gives no spicular measure- EEPORT ON THE MONAXONIDA. 113 ments ; without knowledge of this important character identifications are usually a very difficult matter ; the figures and description of the external form and spiculation of this sponge are, however, sufficient to convince us that we have really got hold of the same thing, and this conviction is strengthened by the fact that both specimens come from high northern latitudes. The spiculation (taken from the Challenger specimen, for we have had no opportunity of examining the original type) is as follows : — (a) Megasclera ; of one kind only, viz., smooth, slender, straight subtylostyli, with slightly developed oval heads, and tapering gradually to a fine point at the apex ; size about 0*6 by 0"0075 mm. (6) Microsclera ; of two kinds ; (l) minute isochelse, very abundant, about 0'0r26 mm. long ; (2) slender toxa, with spined ends, size very variable, apparently reaching as much as about 0'4 mm. in length (as shown by a single broken spicule, of which one limb was measured and doubled to give the total length), commonly much smaller, about 0"15 mm. long. Locality. — Station 49, May 20, 1873 ; lat. 43° 3' N., long. 63° 39' W.; south of Halifax, Nova Scotia ; depth, 85 fathoms ; bottom, gravel, stones ; bottom temperature, 35°"0. One specimen, attached to a stone. Habitat. — Arctic Seas (Vosmaer) ; south of Nova Scotia (Challenger). Genus Phelloderma,^ Eidley and Dendy (PI. XXIII.). 1886. Phelloderma, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 347. Corticate, with a cork-like ectosome. Megasclera aU monactinal, styli or tylostyli, radiating towards the surface, with the points directed outwards ; microsclera isochelse. In the presence of a distinct cortex, and in the radiate arrangement of its skeleton, this genus approaches the Suberitidae, but its stylote megasclera and isochelate microsclera leave no doubt that its real place is amongst the Desmacidonidse. Although there is only a single specimen, yet its characters are so well marked that they appear to us to justify the erection of a new genus for its reception. Phelloderma radiatum, Eidley and Dendy (PI. XXIII. figs. 8, 8«, 86, 8e, 8c7). 1886. Phelloderma radiatum, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 347. Sponge subglobular, with a concave base of attachment (originally based upon a Sporadopora, from which it has been removed), about 13 mm. in diameter, with a 1 From Greek (ps^Xof, cork ; and oi^/ia, .skin. (ZOOL. CHALL. EXP. — PAET Lix. — 1887.) Nnn 15 114 THE VOYAGE OF H.M.S. CHALLENGER. distinct, cork-like cortex lying immediately below the dermal membrane, about 0"24 mm. thick. Colour in spirit light brown. Texture corky, internally fibrous. Surface fairly smooth. Dermal membrane fairly distinct and peeling off readily in strips. Oscula apparently few, scattered, each on a small papilla, but the condition of the specimen is such as to make it difficult to be certain of this. Pores (?). Skeleton. — Eadiately arranged. From about the centre of the base bands of spiculo-fibre radiate to the surface, where they terminate in brushes of stylote spicules with their points directed outwards and embedded in the dense cortex. Spicules. — (a) Megasclera ; of one kind only, viz., straight, smooth styli (PI. XXIII. figs. 8, 8a), verging upon tylostyli, fairly gradually and sharply pointed, often with the shaft slightly bulbously dilated at intervals ; size about 0"65 by 0*0126 mm. (b) Microsclera ; of one kind only, viz., isochelge, of very peculiar form (PI. XXIII. figs. 86, 8c, 8(:/), with three distinct, rather palmate teeth at each end and with a diamond-shaped " tubercle " (Carter) ; often the two anterior teeth are seen to be connected together by their apices (PI. XXIII. fig. 8c); length about 0"044 mm. In addition to spicules the sponge contains a large amount of sand scattered irregularly through it. Unfortunately there is only a single small specimen and that not in very good condition. Locality. — Station 320, February 14, 1876 ; lat. 37° 17' S., long. 53° 52' W.; ofi" the mouth of the Eio de la Plata ; depth, 600 fathoms ; bottom, green sand. One specimen. Genus Sideroderma,^ Eidley and Dendy (Pis. VIII., IX.). 1886. Sideroderma, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 348. Sponge massive, with mammiform processes on the upper surface, provided with a more or less dense external rind, composed of tylote spicules horizontally arranged. Megasclera, tylota ; microsclera, isochelse of various forms, and, at any rate usually, sigmata and trichodras;mata. Only a single species of the genus, and of that only a single specimen, is present in the collection ; but this difi"ers so markedly from all other known Desmacidonidse that it seems advisable to found a new genus for its reception, more especially as we have been enabled, through the kindness of Dr. R. v. Lendenfeld, to examine a second species which occurs in his large collection of Australian sponges. The description of this second species has not yet been published, but an examination of it has enabled us to give a much more satisfactory generic diagnosis than would otherwise have been the case. 1 From Greek alinQo;, iron and 3ij,«a, skin. REPORT ON THE MONAXONIDA. 115 Histoderma a2'>]pendiculatnm, Carter/ appears to resemble this genus more strongly than any other described sponge, having a similar form (though the tubes on the surface are very long), a similarly arranged skeleton, and stout sigmata ; but as the chela is of a normal form, and the sponge possesses no tylote spicules, the resemblance may have no classiiicatory importance. Sideroderma navicelligerum, Eidley, sp. (PI. VIII. fig. 9 ; PI. IX. figs. 5, 8, 9). 1885. Crella navicelligera, Ridley, Narr. Chall. Exp., vol. i. pt. 2, p. 571. 1886. Sideroderma navicelligerum, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 348. Sponge (PI. VIII. fig. 9) hemispherical, adhering by its broad surface; consisting of a very hard, dense, external rind (whence the generic name), about 1 mm. in thickness, and of an underlying soft mass of tissues containing numerous microsclera and loose megasclera but no fibre. From the surface of the sponge arise a great number of papillse, about 6 mm. in height, some with a (contracted) osculum at the extremity. Size of the specimen 44 mm. in diameter at the base by 31 mm. thick. Colour in spirit pale, dirty yellow. Texture, externally very hard and dense ; internally soft, pulpy, amorphous (possibly a good deal decomposed). Surface smooth, except for the numerous papillse. Oscida on the summits of papillse. Skeleton. — No horny fibre is present. The main part of the skeleton lies in the external rind. This consists of a very dense, felted mass of tylote spicules arranged with some degree of order in two main directions parallel to the surface. The spicules lie in layers, touchiug one another, the spicules of the same layer lying more or less in the same direction and at right angles to the spicules of the layer above or below. In a vertical section through the rind (PL IX. fig. 9) we sometimes see very plainly the alternating layers of longitudinally placed spicules, and of transverse sections of the spicules. Occasionally spicules are found more or less vertical to the surface. The spicules of the external rind are aU dumb-bell-shaped (tylota), and this adds very much to their efiiciency, for by the interlocking of the heads of such .spicules one with another a far firmer structure is produced than if the spicules were simply oxea or strongyla. Spicides. — (a) Megasclera; tylota, each spicule consisting of a long, cylindrical shaft, with an oval head at each end (PI. IX. fig. 5, a, h). Length 0-28 to 0-595 mm.; diameter of shaft in middle 0-0063 to 0-0126 mm. The shaft is thickest in the middle, and tapers slightly and gradually towards the knob at each end. These spicules also occur abundantly scattered about in the soft tissues beneath the rind, and it is from these that the measurements given above are taken. (6) Microsclera ; these are exceed- ingly abundant in the deeper tissues (the isochelte also occur in the riud) ; and vary I Ann. and Mag. Nat. Hist, ser. 4, vol. xiv. p. 220; pi. xiv. figs. 23-25 ; pi. xv. fig. 39. 116 THE VOYAGE OF H.M.S. CHALLENGER. considerably in shape and size ; there are five kinds present ; (1) trichodragmata (PI. IX. fig. 5, c); bundles of A^ery fine hairs, measuring (the bundles) about 0'455 by 0'175 mm. ; these occur in great abundance ; (2) large, stout sigmata (PL IX. fig. 5, d), some only very slightly contort, others considerably so ; size 0"06 by 0"0047 mm.; (3) small, thin, contort sigmata^ (PI. IX. fig. 5, e), size 0-0189 by 0'0015 mm.; (4) a few tridentate isochelas, about 0"019 mm. long ; (5) minute isochelse (PL IX. figs. 5,/, and 8), O'Ol mm. long, of very peculiar form. The shaft is much expanded laterally and, when viewed from the side, is seen to be notched in the centre. There is only one tooth at each end, which is sharply recurved ; owing to the minute size it is difiicult to make out further details. Only one specimen of this species is present. It has evidently been cut from a stone or some other attachment, so that in place of the base of the sponge a large circular aperture is left ; this leads into a large cavity, caused by the shrinking up, and perhaps partly by the loss of the internal soft tissues. Perhaps the most remarkable feature about the sponge is the exceedingly hard and tough external rind, formed as described above, a feature not at all common amongst the Desmacidonidge. The great number and variety of microsclera is also noteworthy. Locality. — Station 188, September 10, 1874 ; lat. 9° 59' S., long. 139' 42' E.; ofi" the south-west corner of New Guinea ; dej)th, 28 fathoms ; bottom, green mud. One specimen. Genus lophon, Gray (Pis. XVI., XVII.). 1867. Ioj3lion, Gray, Proc. Zool. Soc. LoncL, May 1867, p. 534. 1867. AleUon, Gray, Proc. Zool Soc. Lend., May 1867, p. 531.2 Sponge soft and crumbling, usually dark coloured. Skeleton divisible into (a) a dermal layer of diactinal spicules, usually tylota, i.e., with knobbed extremities which may be smooth or spined, and (6) a deep-lying reticulation of loose spiculo-fibre, composed of stylote spicules which are generally more or less spined. (? The proper skeleton may be replaced by foreign bodies, such as sand-grains, foreign spicules, &c.) Microsclera generally of two kinds. («) Palmate anisochelse, the small end of which terminates in a sharp sj)ur ; these appear to be constant. (6) Bipocillate spicules, very characteristic of the genus and almost always present, consisting of a curved shaft terminated by a cup-shaped expansion at either end, which is sometimes divided into lobes in a trefoil-like manner. Species of this genus may generally be recognised by the combination of several of the above characters. A good external guide is the dark colour, many species being black. The sponge is usually an amorpihous mass of crumbling substance, but it may ^ Possibly young forms of tlie larger ones. 2 For further generic synonyms, see under lophon patUrsoni. REPORT ON THE MONAXONIDA. 117 occasionally assume a definite shape, as in lophon 2yiceum, Vosmaer, which is cup-shaped, and in Io2:>hon cylindricus, nobis, which is cylindrical. A common feature of the genus is the distinctness of the dermal membrane and pores. Of the spicules the most remarkable are the bipocilli, which are, so far as we are aware, confined to the genus. The genus has a wide distribution, as will be seen by reference to the localities given for the various species, of which six were obtained by the Challenger. lophon pattersoni, Bowerbank, sp. 1866. Halichondria Patteisoni, Bowertank, Mort. Brit. Spong., vol. ii. p. 255; vol. iii. p. 115, pi. xlvi. •1866. Halichondria scandens, Bowerbank, Mon. Brit. Spong., vol. ii. p. 259; vol. iii. p. 112, pi. xlv. 1866. Halichondria Hyndmani, Bowerbank, Mon. Brit. Spong., vol. ii. p. 264 ; vol. iii. p. 115, pi. xlvi. 1866. Halkltomlria nigricans, Bowerbank, Mon. Brit. Spong., vol. ii. p. 266; vol. iii. p. 113, pi. xlv. 1867. Alehion hijndmani, Gray, Proc. Zool. Soc. Lond., May 1867, p. 534. 1867. lophon scandens, Gray, Proc. ZooL Soc. Lond., May 1867, p. 534. 1867. lophon nigricans, Gray, Proc. Zool. Soc. Lond., May 1867, p. 534. 1880. Esperia Pattersoni, Vosmaer, Notes from lieyden Museum, vol. ii. p 143. 1880. Esperia Hyrulmani, Vosmaer, Notes from Leyden Museum, vol. ii. p. 144. 1880. Esperia scandens, Vosmaer, Notes from Leyden Museum, vol. ii. p. 144. 1880. Esperia nigricans, Vosmaer, Notes from Leyden Museum, vol. ii. p. 144. 1881. AleUon proxinmm, Ridley, Proc. ZooL Soc. Lond., January 1881, p. 117. Sponge massive, amorphous, crumbling. Colour dark brown. Surface uneven, corrugated^ The skeleton consists of (1) a dermal layer of long, dumb-bell-shaped spicules (tylota) with the knobs u.sually slightly spined and the shaft smooth. (2) A rather loose reticulation of spined stylote spicules penetrating all through the body. Two kinds of microsclera are present; (l) palmate anisochelse, with pointed smaller ends, very fine, measuring up to 0"03 mm. long. (2) Minute bipocilli. Both of these vary greatly in numbers in difi'erent specimens, and in diflerent parts of the same specimen. For measurements of the styli and tylota see the table given below. The Challenger specimens of lophon from Stations 308, 311, and from Tristan da Cunha form connecting links between several previously known, though Hi-defined, forms. We are thus enabled to unite these several forms as varieties of a single sjjecies. In all of them the external appearance seems to be the same, and the distinction has lain chiefly in slight variations in form and size of the spined styli and of the tylota {vide table below). The microsclera vary considerably in the numbers in which they occur, but this cannot be considered as a specific distinction. We thus have a single species 118 THE VOYAGE OF H.M.S. CHALLENGEE. iu place of five, for which we adopt the name loattersoni, that being the earliest we cau find {vide synonymy above). The species has a wide range, and presents considerable variation in its spiculation. It is not improbable that other forms, hitherto regarded as distinct, will also sooner or later have to be included in this species. lophoa 2)iceum, Vosmaer, comes near to it in spiculation, but is described as cup-shaped ; it is UD usual for an lophon to have a well-marked external form, and this is perhaps a good specific character. Variety. Spined Sty li. Tylota. Locality. proximum Slightly spined all over ; tapering from near apex, mm. 0-158 by 0-009 With distinct heads, which are distinctly spined. mm. 0-16 by 0-OOS strait of Magellan. (Challenger) Slightly spined all over, more so near the base and apex ; tapering from near apex, 0-17 by 0-007 Heads very slightly de- veloped, very feebly or not at all spined, . 0-19 by 0-005 Tristan da Cunha. nigricans Slightly spined, more so on the base, apex usually smooth ; taper- ing gradually from base to apex, 0-213 by 0-008 Very slender ; heads barely visible or absent, not spined, . 0-195 by 0-003 British Isles. (Challenger) Considerably spined all over ; tapering from near apex. 0-22 by 0-01 Heads distinct and distinctly spined. 0-2 by 0-006 Off south-west coast of Patagonia. hyndmajii Spines mostly near base ; tapering from middle. 0-228 by 0-019 Very slight, faintly spined heads, . 0-196 by 0-004 British Isles. pattcrsoni SUghtly spined all over ; tapering from near middle, . 0-234 by 0-01 Heads scarcely distin- guishable from shaft. 0-25 by 0-006 Britisli Isles. (Challenger) Stronglyspined all over ; tapering from near apex. 0-24 by 0-011 Heads distinct, slightly spined, 0-24 by 0-007 Oif west coast of Patagonia. scandens Very slightly spined all over; tapering from near apex. 0-262 by 0-012 Heads distinct, slightly spined, 0-245 by 0-006 British Isles. Localities. — Station 308, January 5, 1876 ; lat. 50° 8' 30' S., long. 74° 41' 0" W.; off" the west coast of Patagonia ; depth, 175 fathoms; bottom, blue mud. Fragments. Station 311, January 11, 1876 ; lat. 52° 45' 30" S., long. 73° 46' 0" W. ; off the south-west coast of Patagonia ; depth, 245 fathoms ; bottom, blue mud ; bottom temperature, 46°. A number of fragments. Off Nightingale Island, Tristan da Cunha. A number of fragments; Habitat. — The Minch (Bowerbank) ; Strangford Loch (Professor Dickie, given by Bowerbank) ; off Hastings (Bowerbank) ; Shetland (Norman, given by Bowerbank) ; Strait of Magellan (Ridley) ; Tristan da Cunha (Challenger) ; west coast of Patagonia (Challenger), REPORT ON THE MONAXONIDA. 119 lophon chelifer, Ridley and Dendy (PI. XVI. fig. 3 ; PI. XVII. figs. 1, 3, 8). 1886. lophon chelifer, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. sviii. p. 349. Sponge (PI. XVI. fig. 3) amorphous/ massive, honeycombed. Colour in spirit light brown to black. Texture soft, crumljling. Surface smooth. Skeleton. — Composed of an irregular, loose reticulation of spined stylote spicules ; the meshes of the network are more or less rectangular, so that there are two main lines of spicules distinguishable, though only roughly, one of which is approximately perpen- dicular to the surface. Spicules. — (rt) Megasclera; (1) spined styli (PI. XVII. fig. 1, a, h), measuring 0'36 to 0"42 by 0'016 to 0"02 mm.; (2) there are also present a few tylote spicules (PI. XVII. fig. 1, c), apparently confined to the surface, size 0'25 to 0*32 by 0"01 mm.; these spicules have the shaft smooth and the knobbed extremities minutely spined. (b) Microsclera ; very abundant, of two kinds ; (1) anisochelse (PI. XVII. fig. 8), from 0*019 to 0'03 mm. long ; (2) bipocilli (PI. XVII. fig. 3), large (0*019 mm. long) and of very peculiar form ; shaft narrow and strongly bent, small end clawed, with two prongs (whence the specific name), large end bearing two, tliree, or (very rarely) four expanded fiukes, which together form a hollow cup. These spicules occur lining the canals ; the large end is embedded in the wall of the canal, and the small, clawed end projects freely into its lumen.- They also occur scattered abundantly through the soft tissues (perhaps owing to displacement ?). This spicule is the most interesting feature about the species ; it throws considerable light on the relations of the more minute bipocUlate sj^icules of other sponges. We are inclined to regard it as a much modified anisochela; its variability in form (i.e., in the number of flukes forming the larger end of the spicule) is also noteworthy. Three specimens of this interesting species are present, two are fairly large, but broken into fragments, the other is small, and occurs encrusting a branched Polyzoon. The latter is in all probability a young form, and differs in several minor respects from the larger specimens ; in it the anisochelse are mainly arranged in rosettes, with the small ends iu the centre, they are very abundant, while, on the other hand, the bipocil- late spicules are very few and apparently imperfectly developed. The species differs very decidedly from all described forms in the large size and also in the degree of 1 Exact form uncertain, specimen fragmentary. - This arrangement would seem to be connected with the protection of the walls of the canals, probably from animal intruders, rendering the spicules truly defensive, a term which cannot be applied to aU the spicules thus named by Bowerbank. The cases in which a spicule may reasonably be assimied to fulfil this purpose are very few, that of the diancistra or "trenchant bihamates" (see Bk., Mon. Brit. Spong., vol. i. p. 34) being perhaps the best established case hitherto recorded. Cf. also the sigmata in Esperella murrayi, nobis. 120 THE VOYAGE OF H.M.S. CHALLENGER. elaboration of the bipocillate spicules. The other spicules are also larger in almost every case than the corresponding forms in other species of lojihon ; lophon [Alehion) piceum, Vosmaer, approaches it the most nearly in this respect, but even in this species the stylus measures only 0'32 mm. in length. Localities. — Station 142, December 18, 1873 ; lat. 35° 4' S., long. 18° 37' E. ; off the Cape of Good Hope ; depth, 150 fathoms ; bottom, green sand ; bottom temperature, 47°. The young specimen. Station 145a, December 27, 1873 ; lat. 46° 41' S., long. 38° 10' E.; off Prince Edward Island; depth, 310 fathoms ; bottom, volcanic sand. Station 148a, January 3, 1874 ; lat. 46° 53' S., long. 51° 52' E. ; between Prince Edward and Kerguelen Islands ; depth, 550 fathoms ; bottom, hard ground, gravel, shells. lophon cylindricus, Ridley and Dendy (PL XVI. fig. 4 ; PI. XVII. fig. 6). 1886. lophon cylindricus, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 319. Sponge (PI. XVI. fig. 4) erect, cylindrical (whence the specific name), with expanded base encrusting a dead Pecten shell ; showing a slight tendency, to branch at the apex. Height, including base, 44 mm. Average diameter 6 mm. (A second specimen, also growing on a dead shell, appears to represent little more than the base of the sponge.) Colour in sj)irit brown. Texture brittle and crumbling. Dermal memhrane very distinct, transparent', and with numerous round pores. Oscula scattered over the surface (?). Skeleton. — (o) Dermal ; a very well developed network of not very closely aggre- gated tylota. (h) Main ; a reticulation of loose spiculo-fibre, composed of stylote spicules. Spicules. — (o) Megasclera ; slightly curved, smooth styli (PL XVII. fig. 6, a, h), each tapering from near the apex to a very sharp point; size 0"25 to 0"29 by 0*011 mm. {h) Micvosclera ; of two kinds ; (1) palmate anisochelee of the usual lophon tyj)e, measuring 0'025 mm. in length ; (2) bipocilli, numerous and of the usual shape, 0'0127 mm. long. It will be seen that this species approaches Iop>hon laminalis, nobis, in the smoothness of the styli, but the external form is very different. Further, the tylota are little more than half as stout, and the styli only half as stout and not half as long as the corre- sponding spicules in that species. Locality/. — Station 163a, April 4, 1874 ; lat. 36° 59' S., long. 150° 20' E.; north- east of Cape Howe, Australia; depth, 120 fathoms; bottom, green mud. Two specimens. REPORT ON THE MONAXONIDA. 121 lophon laminalis, Eidley and Dendy (PI. XVI. fig. 6 ; PL XVII. figs. 9, 11). 1886. lophon laminalis, Eidley and Dendy, Ann, and Mag. Nat. Hist., ser. 5, vol. xviii. p. 349. Shape doulitful. In its present condition the specimen consists of a number of irregular, flat, or slightly curved, cake-like expansions (PI. XVI. fig. 6), which, taken together, cover an area of about 250 square centimetres. These have no appearance of having lain on the bottom of the sea, and a.s they seem too thin and fragile to have stood erect as a flat lamina, we are inclined to think that, when living, the sponge had a cup-like form, like Alehion {lophon) piceum, Vosmaer,^ from Barents Sea, Thickness of the lamellje 4 to 17 mm. The two sides are slightly difierent in appearance, one being smoother than the other. There is a distinct, thin, dermal membrane, especially visiljle on the smooth side. Colour in spirit dark reddish-brown. Surface uneven, with numerous small oscula scattered over it on both sides of the lamellae. Texture loose, crumbling, slightly stringy. Skeleton. — (a) Dermal; a network (not always easy to make out) of tylota. [h) Main; a loose reticulation, not arranged in definite meshes, of large, subtylostylote spicules arranged longitudinally along certain tracts, so as to form long but loose fibres, which give rise to the somewhat stringy character of the sponge. Spicules. — (a) Megasclera ; (l) large subtylostyli (PI. XVII. fig. 11, h), measuring 0'63 by 0"022 mm. The heads of these spicules are only faintly developed, so that they are very nearly simjjly stylote, and they taper somewhat suddenly, from near the apex, to a sharp point. (2) Tylota (PI. XVII. fig. 11, a), shaft and head usually smooth, but head sometimes minutely spined; size 0'34 by O'OOIS mm. (6) Microsclera ; two kinds are present as usual, (l) Palmate anisochelse, 0'025 mm, long; (2) bipocilli (PI. XVII. fig. 9) 0'013 mm. long The palmate anisochelse seem to be very rare in the one specimen present ; but the bipocilli are extremely abundant and very fine, consisting each of a curved shaft with a large cup-shaped expansion at one end and a small one at the other. Only one specimen, broken into fragments, of this interesting species was obtained. In the fine state of development of its bipocillate microsclera it approaches lophon chelifer, a specimen of which was obtained at the same station ; while in external form it probably comes near to lophon piceum, Vosmaer.^ The species to which it is perhaps most nearly related is, however, lophon cylindricus, which, like it, has the stylote spicule smooth ; but, as will be seen from the description of that species, the projaortions of the spicules in the two differ very widely; and, as in probably no other known lophon, the styli here show a tendency to become tylostylote. ioca/%.— Station 145a, December 27, 1873 ; lat. 46° 41' S., long. 38° 10' E.; off Prince Edward Island; depth, 310 fathoms; bottom, volcanic sand. One specimen. 1 Niederland. Archiv f. ZooL, Suppl. Bd. i., 1881-2, p. 42, pi. i. fig. 19, pi. iii. figs. 75-82. ' Loc. cit., swpra, pi. i. fig. 19. (zooL. CHALU EXP. — PAET Lix. — 1887.) Nnn 16 122 THE VOYAGE OF H.M.S. CHALLENGER. loplion ahnormalis, Eidley and Deudy (PI. XVII. figs. 5, 7). 1886. lophon ahnormalis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviiL p. 350. Sponge represented in the collection by two small, cylindrical fragments, the larger of which is branched, and measures 19 mm. in length by 4 mm. in average diameter. Colour in spirit black. Texture brittle and crumbling. Dermal membrane distinct, transparent, perforated by very distinct, round or oval pores. Skeleton. — (a) Dermal; a network of tylota, loosely placed. (6) Main; a loose reticulation of stylote spicules. Spicules. — (a) Megasclera ; (l) styli (PI. XVII. fig. 5, c), generally spined at the base and also slightly at the apex; size about 0"35 byO'0126 mm. (2) Tylota (PI. XVII. fig. 5, a, b), each with a nearly straight or slightly curved shaft, bearing a distinct spined knob at each end; size about 0"28 by 0 "008 mm. (b) Microsclera ; only the anisochelse appear to be present, but these are of two very distinct sizes. (1) Large palmate aniso- chela3 of the usual lophon type, 0"0378 mm. long, arranged mainly in very fine rosettes (PI. XVII. fig. 7). (2) Smaller anisochelse of about half the size. It is of course possible that the smaller anisochelse are merely young forms of the larger ones, but it is noteworthy that it is the larger ones, and not the smaller, which are arranged in rosettes. It is very abnormal to find an Iop)hon without bipocillate spicules (hence the specific name), and in only one other specimen of lophon have we seen the aniso- chelate spicules arranged in rosettes, viz., in a small specimen from Station 142, which we have referred (though the identification must be considered doubtful) to lophon chelifer, considering it as a young form ; in this also the bipocillate spicules were very rare. Locality. — Ofi" Marion Island, 50 to 75 fathoms. Two fragments. lophon (f) omnivorus, n. sp. (PL XVII. fig. 10). Sponge massive, amorphous. Size of specimen 21 mm. long by 8 mm. broad. Colour in spirit light brown. Texture soft, brittle, very fragile (owing to its sandy skeleton). Surface smootL Dermal membrane very distinct, transparent, perforated by numerous round or oval jjores, about 0"035 mm. in diameter, aggregated in groups over subdermal cavities (PL XVII. fig. 10). Oscula doubtful (? on small, prominent papillae). Skeleton. — Composed chiefly of foreign bodies arranged so as to form a dermal layer (PI. XVII. fig. 10) and a deeper skeleton, in the same way as the skeleton of other species of the genus lophon. The nature of these foreign bodies is very various, sand grains, Foraminifera, Polyzoa, calcareous sponge sj)icules, foreign siliceous spicules, &c., occur in great profusion, and appear to take the place of the proper megasclera. We have, however, in the dermal membrane a good many tylota, some of which may be proper to the sponge ; the heads and shafts of these are quite smooth and they vary much in size, probably many of them are foreign. REPORT ON THE MONAXONTDA. 123 Spicules. — (a) Megasclera; doubtful. {h) Microsdera; two kinds are present. (1) Very numerous palmate anisoclielse of the characteristic lophon type, 0"0145 mm. long ; these sjsicules are frequently arranged around the walls of the canals, with the small end sticking into the wall and the larger end projecting into the cavity. They also occur very abundantly in the dermal layer. (2) A very few small bipocilli (? foreign). It is, of course, very difficult, amongst such a number of foreign spicules of almost all sorts and sizes, to pick out just the ones which are proper to the sponge itself ; hence it is with some hesitation that we refer this species to the genus lojyhon. The most abundant spicules, and those which are most certainly proper to the sponge, are the auisochelse. The sand grains occur chiefly in the deeper parts of the sponge. They appear to be arranged, together with the other foreign bodies, without any definite order, and to take the place of the proper megasclera. The figure (PI. XVII. fig. 10) will give some idea of the arrangement of the foreign bodies — chiefly sponge spicules — in the dermal layer. In its collecting habit this sponge agrees closely with Marshall's Phoriospongia,^ and it seems to us very probable that individual species out of several distinct genera have acquired the same habit of replacing their proper skeleton with a skeleton of foreign bodies. This is sufficient to distinguish the species unmistakably from all others as yet known of the genus lopTion. Locality.— Qtsition 163a, April 4, 1874 ; lat. 36° 59' S., long. 150° 20' E.; ofi"thc south- east coast of Australia ; depth, 150 fathoms; bottom, green mud. One small specimen. Genus Amphilectus, Vosmaer (Pis. XIX., XXV., XLVIL). 1880. Amphilectus, Vosmaer, Notes from the Leyden Museum, vol. ii. p. 109. We make use of this genus in the manner indicated by its founder, namely, as a provisional receptacle for a number of doubtful Desmacidonidse (Esperellinse). As regards the diagnosis and history of the genus we quote Vosmaer's later remarks on the subject.^ " Stabnadeln glatt oder (und) gedornt. Dazu Anker oder (und) Schaufeln mit gleichen oder ungleichen Enden ; eventuell auch Bogen und Haken. Sponginmasse sehr wechselnd. Kosmoj^olitisch;" this, of course, would include jjretty well every species of Desmacidine, and Vosmaer continues: — " Es ist Amphilectus nur ein vorlaufiges Aushilfe-Genus, wo alle diejenigen Desmacidinen untergebracht sind, von welchen keine besondere Merkmale bekannt sind, wodurch sie zu einer anderen Gattung kommen soUten." As such we accept it, but we have not made it quite so comprehensive as Vosmaer has done, and we include in it here only such Esperellinfe as possess isochelate microsdera and (usually) stylote megasclera, to which other forms of spicules 1 Zeitschr.f. wiss. Zool, Ed. xsxv. p. 122. " Bronn's Klass. u. Ordn. des Thierreichi?, Porifera, 1885, p. 353. 124 THE VOYAGE OF H.M.S. CHALLENGER. may be added ; we are thus able to reduce its bulk very considerably, but even thus limited the genus is not a natural one. Esperiopsis, Carter, as emended by us (above, p. 76) forms a natural genus which we have felt able to remove from the general mass of such sponges as Vosmaer unites under the name Amphilectus, and as our knowledge of the Desmacidonidce increases it will no doubt become necessary similarly to detach other groups. Amphilectus apollinis, Ridley and Dendy (PL XIX. figs. 3, 3a, 3i, 3c). 1886. Amphilectus apollinis, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 350. Sponge massive, amorphous, cavernous. The largest specimen is oval, cake-shaped ; about 50 mm. long by 38 mm. wide and 19 mm. thick. Colour in spirit light greyish- yellow. Texture rather soft and spongy. Surface very uneven, deeply folded, slightly glabrous in places where the dermal membrane remains uninjured. Dermal memhrane fairly distinct. Oscula (?).^ Skeleton. — (a) Dermal; in some parts there is a very distinct, but confused, dermal reticulation of slender stylote spicules, while in other parts this appears to be scarcely represented, (b) Main; a confused Halichondrioid reticulation of stout, stylote spicules. Spicules. — (a) Megasclera; of one kind only, but of two distinct sizes. (1) Stout, slightly curved, smooth styli (PI. XIX. fig. 3), size about 0"5 by 0"0168 mm.; making up the main skeleton. (2) Slender, straight styli (PI. XIX. fig. 3a), very sharply and gradually pointed, often very slightly spined at the base and with a slight tendency to become tylostylote; size about 0*315 by 0'0063 mm., almost entirely confined to the dermal layer. (6) Microsclera; of two kinds — (l) small palmate isochelge (PI. XIX. fig. 3c); length about O'OIS mm.; very abundant. (2) Large toxa (PI. XIX. fig. 36), of very beautiful form and with spined ends, size of the full-grown spicule about 0'3 by 0'0045 mm.; very abundant ; the specific name has reference to them. Vosmaer ^ has founded a genus Artem,isina {vide supra) of which the most characteristic spicule is a toxite with spined ends ^ like that which occurs in Amphilectv^ apollinis. Possibly the two species Artemisina suberitoides, Vosmaer, and Amphilectus apollinis, nobis, come near to one another and may even belong to the same genus, but they difi"er very widely in the texture of the sponge, and our present species possesses an additional form of megasclera (viz., the slender stylus with spined base) not present in Artemishia. The presence of spined toxa, taken alone, is certainly not a character of generic importance, for such spicules occur in several species that are widely distinct from one 1 Both specimens are infested by very numerous small Amphipoda which live just below the surface, often causing it to be deeply pitted, but these pits must not be mistaken for oscula. 2 Sponges of the "Willem Barents" Expedition, 1881-82, p. 25. 3 loc. ciL, pi. v. fig. 51. REPORT ON THE MONAXONIDA. 125 another {e.g., Clathria lohata, Vosmaer, Suherites arciger, Sclimidt, Artemisina suheri- toides, Vosmaer, Amphilectus cqwlliiiis, nobis). Locality. — Royal Sound, Kerguelen; depth, 20 to 60 fathoms. Two specimens. Amphilectus ceratosus, Ridley and Dendy (PI. XIX. figs. 10, 10a ; PI. XXV. fig. 2 ; PI. XLVII. fig. 2). 1886. Amphilectus ceratosus, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol xviii. p. 350. Sponge (PI. XXV. fig. 2) massive, lobate, sessile, of very irregular shape. The largest specimen is 44 mm. long and varies much in diameter. Colour in spirit dark reddish-brown. Texture spongy, elastic, but fairly compact. Surface uneven, with small, angular conuli, but glabrous. Dermal membrane rather thick, granular and of a brown colour, peeling off fairly easily. Oscula small and scattered, not noticeably elevated above the general surface of the sponge. Skeleton (PI. XLVII. fig. 2). — (1) Horny, consisting of a well-developed reticulation of ramifying and anastomosing horny fibre, without any spicular core. This horny fibre is doubtless the most important part of the skeleton functionally; it averages about 0*07 mm. in thickness. (2) Spicular — {a) Dermal; consisting of numerous, irregularly scattered tylote spicules, {h) Main; consisting of similar, irregularly scattered tylote spicules ; there is no spiculo-fibre in the deeper parts of the sponge though the scattered spicules are everywhere abundant, but towards the surface the spicules are often collected into loose wisps. Sjncules. — (a) Megasclera; of one kind only; viz., straight, entirely smooth tylota (PL XIX. fig. 10), with a distinct oval head at each end of the slender shaft; size about 0"24 by 0"003 mm. (6) Microsclera; of one kind only ; viz., isochelse (PI. XIX. fig. 10a), profusely scattered both in the dermal membrane and in the deeper parts of the sponge ; length about 0'025 mm. This sponge is extremely interesting on account of the existence in it of a well- developed horny skeleton. We at first thought that this might be due to the joresence of a Euspongia over which the Amphilectus had grown, but there are in the collection three specimens, and there is not the slightest reason to suppose that the horny fibre is not proper to the species. The absence of a spicular core to the fibre must be especially noticed. Coincidently with the development of a horny skeleton we have great reduction in the spicular skeleton, and it is possible that we have in this species an actual transi- tional form between the Siliceous and Keratose sponges. The isochelate spicules, however, are very abundant throughout the sponge, and appear to have suffered no diminution in numbers. It is provisionally only that we include this sponge in the genus Amphilectus, 126 THE VOYAGE OF H.M.S. CHALLENGER. for want of a better one in which to put it ; it is not unlikely that it may form the type of a new genus. Mr. Carter ' has described a sponge which appears to be related to our Amjohi- lectus ceratosus. His Suberites jistidatus (the claims of which to the generic name Suherites are of the very faintest description, being, according to the author, " its cavernous and cork-like consistence ") possesses tylote megasclera with mierospined ends, and palmate, isochelate microsclera ; as regards the presence or absence of horny fibre and the arrangement of the skeleton nothing is said. This sponge is also said to possess "long tubular extensions of different sizes," which "are prolonged from large vents." There can be little doubt that the species are distinct, despite the very similar spiculation. Locality. — Off Port Jackson ; depth, 7 fathoms. Three specimens. Amphilecttis i^Hosus, Ridley and Dendy (PI. XIX. figs. 5, 5', 5a, &c.; PI. XXV. fig- 3). 1886. AmpJiiledus jnlosus, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 350. There are in the collection two specimens which we unite under the above name. As the smaller is a poor, dried-up specimen our description will be taken from the larger, which is in good condition. Sponge (PL XXV. fig. 3) lobate, consisting of a short, stout peduncle, about 13 mm. in diameter, expanding above into a single broad, thick lobe about 63 mm. broad and 19 mm. thick; total height of sponge, 81 mm. Colour in spirit dark chocolate- brown." Texture very coarse and hairy, but rather compact. Surface pilose and shaggy, beset with tufts of large, projecting spicules ; furrowed by deep longitudinal grooves which are not hairy like the rest of the surface. Dermal membrane distinct only in the grooves ; deeply pigmented like all the rest of the sponge. Oscula small, scattered, partly in the grooves. Skeleton. — In the dermal membrane, where this can be distinguished, are found a few irregularly strewn tylota, together with a few large styli, and very long, thin oxea. The main skeleton is irregular in the extreme ; coarse, loose and very iU-defiued bands of spiculo-fibre run towards the surface, where they terminate in the shaggy tufts of spicules ; these represent the primary lines and they are crossed in all directions by loosely scattered spicules. There is very little horny uniting substance. In the peduncle the skeleton becomes much more dense, but is still very indefinite. Spicules. — (a) Megasclera; (1) Tylota, with long, straight, slender shaft and fairly well developed heads, usually flattened and slightly and minutely spined at the ends ; size * Ann. and Mag. Nat. Hist., ser. 5, vol. ix. p. 53. 2 Possibly stained by other specimens in the same bottle. REPORT ON THE MONAXONIDA. 127 about 0'42 by 0"0063 mm., dermal ; (2) smooth styli (PI. XIX. figs. 5, 5'), varying much iu size, up to about 2'0 by 0'02.5 mm.; slightly curved aud sharply aud gradually pointed at the apex. Although styli of this size are abundant it is more common to find them much shorter and stouter, measuring about 0'77 by 0'034 mm. The styli make up the main skeleton. Besides these two forms there is a third large spicule which may be mentioned here, although it is probable that from its peculiar form it should be placed rather in the category of microsclera than that of megasclera {vide infra), only its great size argues against this view. It is a very long and very slender oxeote spicule (PI. XIX. fig. 5a'""), sharply pointed at each end, and commonly with a single rather sharp bend in the centre ; size up to about 2"0 by O'Ol mm. These spicules are abundantly scattered throughout the sponge and sometimes occur in bundles ; their occurrence in the dermal membrane has already been referred to. (6) Microsclera ; ( 1 ) very minute, palmate isochelae, about 0"0065 mm. long, scarce; (2) large, stout toxa (PI. XIX. figs. 5a, bh, &c.), measuring about 0'35 by 0*0063 mm. This is about the average size, but there is very considerable variation in this respect, and indeed it is not difiicult to pick out a connecting series between these toxa and the long oxea above mentioned. Such a series is represented on PI. XIX. figs. 5a-5a'"". This species is very well marked, and may be readily recognised both by its external appearance and its spiculation. AU the spicules, except the minute isochela, which is unusually small, are of exceptionally large size. The toxa are probably the largest known examples of their kind. Some of them were found still enveloped by the mother-cell, as shown on PI. XIX. fig. 56. The most interesting feature of the species is, however, the manner in which the toxa appear to develop into oxea. Localities. — Christmas Harbour, Kerguelen, 70 fathoms. One specimen (the t}^e, from which our description is taken). Off Marion Island, between 50 and 75 fathoms. One specimen. Ainphilectus dnnectens, Eidley and Dendy (PI. XIX. figs. 4, 4o). 1886. Amphiledus annedens, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 351. Sponge massive, lobate, represented only by a single fragment, 25 mm. long by about 13 mm. bi-oad. Colour in spirit pale, greyish-yellow. Texture very soft and spongy. Surface uneven, rugose, hispid in places (perhaps owing to artificial causes). Dermal membrane distinct, thin and transparent. Oscula small and scattered. Skeleton. — {a) Dermal; consisting of tylota loosely scattered through the dermal membrane and also arranged in tufts beneath it. (6) Main; a very loose and irregular reticulation of large styli, in which no distinct fibre is discernible. Spicules. — («) Megasclera; of two kinds — (l) very large, stout, smooth, usually slightly 128 THE VOYAGE OF H.M.S. CHALLENGEE. curved styli (PI. XIX. fig. 4), tapering very gradually to a sharp point at tlie apex and evenly rounded at the base, size about I'O by 0'037 mm.: making up the main skeleton. (2) Tylota (PI. XIX. fig. 4a), with smooth, straight shafts and oval heads, which are usually more strongly developed at one end of the shaft than at the other ; indeed they may be almost absent at one end ; the ends of these spicules are very slightly spined and sometimes flattened^; size about 0"525 by O'Ol mm.; dermal, (b) Microsdera ; small, palmate isochelse, thickly strewn through the dermal membrane and also abundant in the tissues below ; length about 0"02 mm. We have also observed a few large, slender toxa with very faintly spined ends, length up to about 0'2 mm., and a few, usually contort sigmata, length about 0*063 mm. This is an interesting species, and it is a pity that there are not more and better specimens of it. It unites the characters of most species of Amphilectus, viz., the posses- sion of smooth stylote megasclera and of small palmate isochelae, with a feature which is very characteristic of the genus Myxilla, viz., the possession of a tylote dermal spicule. Possibly the new species may form an important connecting link in this very perplexing series of sponges. Zoca%.— Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; ofi" the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°"2. Subfamily 2. Ectyonin^. 1875. Edyonida, Carter, Ann. and Mag. Nat. Hist., ser. 4, yoL xvi., p. 133. 1886. Edyoninx, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 470. Skeleton fibre echinated by laterally projecting spicules. Genus MyxUla, Schmidt (Pis. XXVL, XXVIL, XXX., XLVIL). 1862. Myxilla, Schmidt, Spong. Adriat. Meer., p. 71. 1880. Hastatus, 2}. P; Vosmaer, Notes Leyden Mus., vol. ii. p. 136. Megasclera of at least two kinds : — (1) monactinal : — styli, usually spined, main skeleton spicules ; (2) diactinal : — variously ended ; strongyla, oxystrongyla or tylota, chiefly dermal ; to these are sometimes added (3) small spined styli, which echinate the main skeleton. Usually there is only a very small amount of spongin present. Micro- sdera tridentate isochelae, to which sigmata are often added. The genus Myxilla was founded by Schmidt in 1862 {loc. cit.), and the first species which he places therein is Myxilla rosacea, Liebkhn. sp. The original diagnosis runs as follows : — " Halichondriae polymorphse, moUes et mucosae, fragiles, fere omnes spiculis nodosis insignes," to which the author adds, " Ich gebe gern zu, dass diese Gattung auf 1 It is doubtful whether flattening ever takes place at the larger end of the spicule. REPORT ON THE MONAXONIDA. 129 schwachen Fiissen steht." Vosmaer, in 1885/ gave a better diagnosis: — " Stabnadoln glatt oder gedornt ; letztere vorwiegend. Spongin nicht oder kaum vorhanden. Anker drei-, selten zweizahnig, beide Enden gleich. Oberflacbe am lebenden Objekt scliliipfrig. Canalsystem nacli dem dritten T}^us." This, however, is still very wide and might be made to include a great variety of forms. The Challenger material has enabled us to give a more definite diagnosis, taking as our starting point the original species, Myxilla rosacea, Lbkhn. Still it may be questioned whether we have not made the genus too comprehensive, in that we have admitted certain species possessed of echinating spicules.^ Vosmaer* has already done this in admitting Bowerbank's Hijmeniacidon paupertas into the genus, but without pointing out the great importance of the step which he has taken. If we once admit forms with a distinct, special, echinating spicule, like that of Myxilla paupertas, into the genus Myxilla, then we do away with all family distinction between the Desmacidonidse and Ectyonida. This step appears to us to be very desirable, as the distinction is a very unimportant one, and in the present work we have considered the Ectyoninse (old Ectyonida) as a subfamily of the Desmacidonidse. The genus Myxilla as now constituted falls into two fairly natural groups ; (l) species without definite echinating spicules, such as Myxilla rosacea, Lbkhn., and Myxilla digitata, nobis ; (2) species with definite echinating spicules, such as Myxilla paupertas, Bk., and Myxilla compressa, nobis. The detailed comparison of two such forms as Myxilla rosacea, var. japonica, and Myxilla compressa furnishes a sufficient justification for uniting both groups under the same generic name, at any rate for the present. It wiU thus be readily seen that the genus really occupies an intermediate position between the subfamilies Esperinfe and Ectyoninse. The genus Hastatus of Vosmaer comprises simply Myxillas of thoroughly normal type, the point on which its author lays especial stress, viz., the hastate ends of the dermal spicules, being well shown in the type species of Myxilla {Myxilla 1'osacea) ; and the type species of Hastatus {Hastatus luridus) belongs to the same section of the genus as Myxilla rosacea by virtue of its spined megasclera and the absence of echinating spicules. The other species {Hastatus dichiei) assigned to the genus apparently agrees with Hastatus luridus in all essential details. Our knowledge of the relations of Myxilla to other genera does not rest on a very satisfactory basis (see under Plumohalichondria and Clathria, with which genera, as well as — perhaps more distantly — with Hymedesmia and Hymerrhaphia, it presents considerable affinities). ' Bronn's Klass. n. Ordnung. des Thierreichs, Porifera, p. 349. 2 The " internal defensive spicula " of Bowerbank. ' Notes Leyden Mus., vol. ii. p. 127. (ZOOL. CHALL. EXP. — PART LIX. — 1887.) ^^^ ^^ 130 THE VOYAGE OF H.M.S. CHALLENGER. Myxilla rosacea, Lieberkiihn, sp., var. japonica, nov. (PI. XXVI. fig. 3; PI. XXVII. figs. 8, 8rt, 8h, Sc ; PL XL VII. fig. 3). 1859. Halichondria rosacea, Lieberkiihn, Arcliiv f. Anat. u. Physiol., vol. ix. p. 520, Taf. xi. fig. 2. 1862. Myxilla rosacea, Schmidt, Spong. Adriat. Meer., p. 71. 1862. Myxilla fanciculata, Schmidt {non Lieberkiihn), ' Spong. AJriat. Meer., p. 71. 1864. Myxilla tridens, Schmidt, Spong. Adriat. Meer., Suppl. i. p. 30. 1864. Myxilla Esperii, Schmidt, Spong. Adriat. Itleer., Suppl. i. p. 36, Sponge (PL XXVI. fig. 3) lobate or digitate ; frequently flattened. The largest specimen is of an irregularly lobate form and 56 mm. in length. Colour in spirit very variable ; ranging from very pale, creamy yellow to almost quite black. Texture firm, compact and rather leathery. Surface uneven, corrugated, and sometimes minutely hispid. Dermal membrane very thin and transparent. Pores not very abundant, scattered. Oscula few, small, scattered. Skeleton. — The main skeleton consists of a very compact and symmetrical reticulation ; when seen in thin section (PL XLVII. fig. 3) the network appears to be made up of a number of triangular meshes, the apices of six such triangles meeting in a common point while the bases form a hexagon ; this arrangement is in places very distinctly visible but is, of course, never mathematically exact. The sides of the triangular meshes are of exactly one spicule's length, and are formed of from one to about six stylote spicules lying side by side. There is no proper dermal reticulation, but at a depth of about 0'14 mm. below the dermal membrane the skeleton arrangement just described suddenly gives place to another and totally difi"erent one. This takes the form of a number of brushes of tornote spicules, each brush only one spicule in length, and with the com- ponent spicules diverging upwards towards the dermal membrane, which they support, and through which their points often project for a short distance. Sincules. — («) Megasclera; of two kinds, (l) Entirely spined, short, slightly curved styli (PL XXVII. figs. 8, 8a.) tapering rather suddenly to a sharp point at the apex, size about 0'14 by 0'0126 mm. (the diameter is very variable, that given is about the maxi- mum), making up the main skeleton; (2) smooth, hastately pointed tornota (PL XXVII. figs. 86. 8c) rather thicker in the centre than elsewhere, size about 0"175 by 0"007 mm., occurring in brushes supporting the dermal membrane, {h) Microsclera ; of two kinds — (1) small, tridentate isochelse, with rather strongly curved shaft, which appears to be slightly expanded laterally, especially towards the two ends; length up to abowt 0'03 mm. ; (2) slender sigmata, usually much contort, length up to about 0"045 mm. This sponge difi"ers from Schmidt's types of the species chiefly in the possession of a shorter, stouter, and more strongly spined stylote spicule ; in Schmidt's types, also, the 1 According to Vosmaer, Notes Leyden Mus., vol. ii. p. 123, 1880. REPORT ON THE MONAXONIDA. 131 ends of the tornote spicules may sometimes become flattened and slightly spined. The Challenger variety, again, has a firmer and more regular skeleton. Neither Lieberkiihn nor Schmidt {loc. cit.) mention the chelate spicules, but Vosmaer' records their presence, and we have ourselves found them in one of Schmidt's specimens now in the British Museum. Lieberkiihn and Schmidt obtained their specimens from the Mediterranean, and it is of considerable interest to find an only very slightly modified variety occurring at Japan. Locality. — Kobe, Japan ; depth, 8 to 50 fathoms. Three specimens and some fragments. Habitat. — Mediterranean (Lieberkiihn, Schmidt); Japan (Challenger). Myxilla digitata, Eidley and Dendy (PL XXVL fig. 2 ; PL XXVIL figs. 5, 5a, bh). 1886. Myxilla digitata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 471. Sponge (PL XXVL fig. 2) represented in the collection by a single specimen, about 31 mm. long and 12 mm. broad, cleft at the upper end into five finger-like lobes, whence the specific name. Colour in spirit greyish-yellow. Texture soft and spongy. Surface smooth but uneven and corrugated. Dermal membrane distinct, thin, transparent. Oscula small, scattered, having their margins flush with the surface. Pores scattered. Skeleton. — (a) Dermal ; consisting of tylote spicules loosely scattered in the dermal membrane. (6) Main ; a close but irregular and somewhat Halichondrioid reticulation of spined stylote spicules. Fibres very indefinite. Siyicules. — {a) Megasclera; (1) entirely spined styli (PL XXVIL fig. 5), usually slightly curved and tapering gradually to not very sharp points, with no tendency to the formation of heads and with the base no more spined than the remainder of the spicule; spination well marked ; size about 0"4 by 0"014 mm.; (2) tylota (PL XXVIL fig. 5a), straight and with well-developed heads; size about 0'24 by 0'006 mm. (6) Microsclera ; of one kind only, viz., tridentate isochelse (PL XXVIL fig. 5&), with strongly curved shaft ; length about 0'044 mm. This species is readily distinguished from Myxilla {Hymeniacidon) paupertas, Bk., with which it agrees in several respects, by the absence of the echinating stylote spicules. ioca%.— Station 142, December 18, 1873 ; lat. 35° 4' S., long. 18° 37' E.; south of the Cape of Good Hope; depth, 150 fathoms; bottom, green sand; bottom temperature, 47°'0. One specimen. > Notes Leyden Mus., vol. ii. p. 12J, 1880. 132 THE VOYAGE OF H.M.S. CHALLENGER. Myxilla paucispinata, Eidley and Dendy (PL XXVII. figs. 2, 2a, 2b, 2c ; PI. XXX. fig. 3). 1886. Mijxilla paucispinata, Eidley and Dendy, Anii. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 471. Sponge (PI. XXX. fig. 3) massive, amorphous, cavernous.' Colour m spirit pale yellow. Texture rather soft, very In-ittle and fragile. Surface very uneven and roughly hispid from the presence of numerous large foreign (Hexactinellid) spicules ; with numerous apertures leading into canals ; some of these apertures are undoubtedly oscula, each being surrounded by a fringe of spicules. The surface also presents circular areas, in each of which the parchment-like dermal membrane covers over an underlying cavity ; these are possibly in part pore-areas, but we have not succeeded in demonstrating the fact, and have offered below what seems to be a more likely explanation of their presence. The Oscula are in the centres of similar looking areas and are thickly fringed by radiating brushes of tylote spicules, which occur chiefly in this position and also, in fewer numbers, in the circular areas above mentioned. Pores (?). Skeleton. — (a) Dermal ; a close, irregular reticulation of large stylote spicules similar to those of the main skeleton, with a small admixture of tylota.^ Within a certain radius of the oscula the stylote spicules abruptly and entirely disappear and give place to thick brushes of the tylota ; these converge towards the osculum, the margin of which is fringed by the free ends of the terminal spicules of the brushes. The stylote spicules are also absent from the circular areas above described, their place being taken by a few tylota which sometimes show an incipient radiate arrangement ; from these considerations it seems not improbable that some, at any rate, of the spaces mentioned mark the positions of future oscula. {b) Main ; this consists of an irregular but close reticulation of stylote spicules in which no fibre is developed. Spicules.— {a) Megasclera; of two kinds, (l) Large, stout styli (PI. XXVII. fig. 2) usually curved, with rather blunt apex and not infrequently slightly spined ; size about 07 by 0-031 mm. (2) Tylota (PI. XXVII. fig. 2a) each with a fairly straight shaft bearing a distinct, though only slightly developed, oval head at each end ; entirely smooth ; size about 0-4 by 0-008 mm. [b) Microsclera; of two kinds, (l) Tridentate isochete (PI. XXVII. figs. 2b, 2c), with stout, strongly curved shaft ; length about 0-05 mm. (2) Slender sigmata, simple and contort, varying a good deal in size, up to about 0-056 mm. long. Unfortunately there is only a single piece of this sponge in the collection, and that in very poor condition. It is encrusted in places by a thin, black, coating sponge 1 ^\'■ith embedded Polychjete wormi;, to which the exceedingly cavernous nature may be in part due. One large tunnel penetrates right through the sponge, which is perhaps only a fragment. 2 It is possible that the tylota alone form the true dermal skeleton and that the styli are merely an upper layer of the main skeleton ; this would explain the absence of styli immediately around the oscula. Compare with this the dermal skeleton in Myxilla cribrigcra, nobis. EEPORT ON THE MONAXONIDA. 133 (PI. XXX. fig. 3, e), wliich proves to be a new species, cliaracterised by remarkably fiue microsclera ; it will be found described on p. 84 under the name Esperiopsis pulchella. The spination of the styli appears to be a very inconstant character in this species {vide infra), and it is doubtful whether it is of even specific value ; nevertheless we have made use of the fact that these spicules may be spined in selecting a name for the species. This consideration tends to throw considerable doubt on the correctness of separating genera, e.g., Tedania and Trachytedania, on this character alone. Xoca^%.— Station 192, September 26, 1874; lat. 5° 49' 15" S., long. 132° 14' 15" E.; Little Ki Island, New Guinea; depth, 140 fathoms ; bottom, blue mud. One specimen. A second specimen in the collection is probably a slight variety of the above. It is, however, a mere fragment, but it comes from the same station. The spiculation is as follows: — (a) Megasdera; (l) large, smooth styli, with fairly sharp points (seldom or never showing spination); measuring about 077 hj 0'028 mm.; (2) tylota, with very well developed, oval heads; measuring about 0'42 by 0'0078 mm. {h) Microsclera; (l) tridentate isochelse, with slightly curved shaft; length about 0'063 mm.; (2) large, slender, simple (very rarely contort) sigmata, up to about 0'145 mm. in length. Myxilla mollis, Ridley and Dendy (PI. XXVII. figs. 4, 4«, 45). 1886. Mijxilla mollis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 471. Sponge massive, amorphous. Largest sjaecimen about 62 mm. long by 31 mm. broad and 12 to 19 mm. thick. Colour in spirit pale, creamy yellow. Texture uniform throughout, very soft and spongy. Surface usually very uneven, but smooth. Dermal membrane distinct, thin, transparent. Oscula scattered, having their margins flush with the surface of the sponge. Pores scattered (?). Skeleton. — (a) Dermal ; composed of loosely scattered tylote spicules, together with a certain proportion of styli ; sometimes forming a very irregular reticulation, in which occasionally several spicules are found lying side by side, parallel with one another. (6) Main; as loose and irregular as the dermal layer, consisting of a very indefinite reticulation of stylote or subtylostylote spicules. Spicules. — (a) Megasdera; of two kinds. (1) Smooth sulitylostyli or styli (PI. XXVII. fig. 4); these form the main skeleton and are generally subtylostylote, being more or less enlarged at the base, and tapering more or less gradually to a sharp point at the apex ; size about 0-42 by 0-01 mm.; (2) tylota (PI. XXVII. fig. 4a), each with a straight, slender shaft and a distinct oval head at each end ; entirely smooth ; measuring about 0'22 by 0"006 mm. ; occurring abundantly in the dermal layer of the skeleton, (b) Microsclera; (l) tridentate isochelaa (PI. XXVII. fig. Ab), with slightly curved shaft which, towards the two ends, often appears to be expanded on each side into a slight fimbria ; length 134 THE VOYAGE OF H.M.S. CHALLENGER. about 0"04 mm. ; (2) simple and contort sigmata, measuring up to about 0"063 mm. in length, but varying greatly in this respect. This species resembles HalicJiondria isocUctyalis of Carter,^ but the chelse, sigmata and styli are all smaller in that species than here, and further, the head of the stylus is smaller than the shaft. Locality. — Off the south-west coast of Patagonia (Station 308 or 311, or both). Five specimens. Myxilla spongiosa, Ridley and Dendy (PL XXVII. figs. 3, 3«, Zh, 3c, '^d, 3e, 3f). 1886. Myxilla spongiosa, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 471. Sponge massive, encrusting, growing all round a worm-tube, which it encrusts for a length of 69 mm. and to a thickness of about 12 mm. in the middle. Texture extremely soft and spongy. (The condition of the sponge is so bad that no further account of the external characters is practicable.) Skeleton. — (a) Dermal ; a thick but confused reticulation of scattered tylote spicules. (6) Main ; a very confused reticulation of smooth styli. Spicides. — (a) Megasclera; of two kinds; (1) smooth, stout, slightly curved styli (PI. XXVII. fig. 3), tapering gradually to a sharp point at the apex; size about 07 by 0"02 mm. ; making up the main skeleton. (2) Tylota (PI. XXVII. figs. 3a, Zh), with well-developed oval heads, usually (? invariably) minutely spined at the end ; size up to about 0"4 by O'Ol mm. ; dermal. (6) Microsclera ; of two kinds ; (1) tridentate isochelse (PI. XXVII. figs. 3c, 3(i), with the shaft laterally expanded towards each end ; length about 0"0.5 mm. (2) Rather stout sigmata (PI. XXVII. figs. 3e, 3/"), usually much contort; size about 0'063 by 0'0045 mm. Both kinds of microsclera are abundant. The condition of this specimen unfortunately prevents more detailed description, but the spiculation is suflicient to distinguish it from other species ; it will be seen to approach Myxilla mollis, nobis, in this respect, but difi"ers in details, Besides having the stylus smooth, as in Myxilla mollis, the species further diverges from the more normal character of Myxilla in the large size of the megasclera ai^d chelae. Loca%.— Station 320, February 14, 1876 ; lat. 37° 17' S., long. 53° 52' V^. ; ofi"the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom tempera- ture, 37°"2. One specimen. Myxilla hastata, Ridley and Dendy (PI. XXVII. figs. 1, la, lb, Ic). 1886. Myxilla hastata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii p. 472. Sponge forming a flattened lamella, the shape of which cannot be ascertained, as there are only fragments in the collection. The largest piece measures about 69 mm. by 1 Ann. and Mag. Nat. Hist., ser. 5, vol. ix. p. 285. REPORT ON THE MONAXONIDA. • 135 62 mm. and is about 6 mm. thick ; it is strongly curved. (The colour is now reddish, but this is due to the rusting of the case in which the sponge was packed, which has also affected many other specimens from the same station.) Texture soft and spongy. Surface, on one (the convex) side very rough and uneven, with a worn appearance, on the other slightly rough and longitudinally wrinkled. Pores oval, about 0*03 mm. in long diameter, in irregular groups. Skeleton. — (a) Dermal ; the dermal membrane is supported on the ends of brushes of tomote spicules which radiate outwards from below the surface of the sponge ; these spicules may also occur scattered horizontally through the dermal membrane. (6) Main; the main skeleton is a very loose and confused reticulation of large, smooth, stylote spicules, with a tendency towards the fonnation of incipient fibre. Spicules. — (a) Megasclera; of two kinds, (l) Large, stout, slightly curved, sharply and gradually pointed, smooth styli (PI. XXVII. fig. 1), size about 077 by 0'04 mm. (2) Hastately pointed, smooth tornota (PI. XXVII. figs, la, \b), size about 0'35 by O'Ol mm.; dermal, {b) Microsclera ; of two kinds, (l) Tridentate isochelas (PL XXVII. fig. Ic), with stout, strongly curved shaft; length up to about 0'04 mm., l)ut more usually about 0'025 mm. ; abundant. (2) Very abundant sigmata, rather stout and often much contort, size about 0"07 by 0'004 mm. This species evidently comes very close to Myxillch spongiosa, nobis, from the same station, but is sharply marked oS from it by the form of the dermal spicules (perhaps also by their arrangement), by the greater stoutness of the stylus, and by the diflference in size and shape of the tridentate isochelas, as well, perhaps, as by the general form of the sponge. Zoca%.— Station 320, February 14, 1876 ; lat. 37° 17' S., long. 53° 52' W.; off the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°"2. One specimen. Myxilla crihrigera, Eidley and Dendy (PI. XXVII. figs. 7, 7a, 7b). 1886. Myxilla crihrigera, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 472. Sponge apparently free ; digitate, consisting of a single long, digitate process arising from a shorter, stouter, transverse piece. Total length of sponge 50 mm. ; length of the transverse piece 31 mm.; diameter of main lobe about 12 mm. Colour in spirit dark, yellowish-grey. Texture soft and spongy. Surface minutely aud irregularly hispid. Dermal membrane very distinct, thin, transparent, often readily separating from the underlying tissues. Oscula small, scattered, with their margins flush with the surface (?). Pores arranged in definite, more or less circular pore-areas. The surface of the sponge presents in many places distinct rounded areas, marking the position of sul)dermal cavities covered over only by the thin, transparent dermal membrane, which is in these places penetrated by very large, oval pores. These cribriform areas, and even the pores 136 THE VOYAGE OF H.M.S. CHALLENGER. themselves, are visible to the naked eye. Diameter of pore-areas about 1 mm.; of the pores themselves very variable, averaging about 0"1 mm., but may be much greater. Skeleton. — (a) Dermal ; when pieces of the dermal membrane are peeled off from the surface of the sponge, as is easily done, and examined separately, the only true skeleton they possess is seen to consist of rather sparsely scattered tylote spicules ; some- times, however, the styli belonging to the underlying main skeleton project beyond the surface of the sponge, giving to it its irregularly hispid character. (6) Main; consisting of a very loose and irregular reticulation of styli, often forming loose fibres which run towards the surface and are crossed more or less at right angles by others. Sjncules. — (a) Megasclera ; of two kinds. (1) Smooth, stout styli (PI. XXVII. fig 7), only slightly curved, measuring about 0'65 by 0"025 mm. (2) Tylota (PI. XXVII. fig. *7a), with straight shaft and an oval head at each end ; the head is sometimes very slightly spined at the top, where it may also be slightly flattened. Usually one end of the spicule is a little larger than the other. Size about 0"3 by O'OOS mm. {h) Microsclera; one kind only present ; viz., large, tridentate isochelse (PI. XXVII. fig. 7b), with a slightly curved shaft which is markedly expanded laterally towards each end ; length about 0"08 mm. These are plentiful in the dermal membrane and also occur scattered through the deeper tissues of the sponge. Perhaps the most striking features of this sponge are the great size of the pores and their arrangement in definite pore-areas. These strongly resemble the corresponding structures in Halichondria forcipis, Bk. , as described and figured by Carter.^ The species difi"ers markedly from the two preceding ones in the absence of sigmata. Its spiculation agrees in hind with that of Crihrella hainigera, Schmidt,^ but in that species the styli tend to develop heads, and the sj^icules throughout are merely the miniatures of those of this species. Zoca%.— Station 306a, January 2, 1876 ; lat. 48° 27' S., long. 74° 30' W.; ofi" the south-west coast of Patagonia ; depth, 345 fathoms ; bottom, blue mud ; bottom temperature, 46°. One specimen. Myxillafusca, Eidley and Dendy (PI. XXVII. figs. 11, 11a, llh). 1886. Myxillafusca, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 472. Sponge massive, amorphous, the largest specimen about 19 mm. in greatest diameter. Colour in spirit darkish brown. Texture fairly firm and elastic, but rather coarse. Surface uneven, in places minutely hispid. Dermal membrane adhering firmly to the underlying tissues ; thin, granular and pigmented, brown in colour but trans- lucent. Pores apparently collected into pore-areas, wherein they reduce the dermal 1 Ann and. Mag. Nat. Hist., ser. 4, vol. xiv. p. 246, pi. xiv. fig. 29, a, ' Spong. Adriat. Meer., p. 70, pi. vi. fig. 13. REPORT ON THE MONAXONIDA. 137 membrane to a network, but we have found no perfect examples of this arrangement. Oscula small and scattered, having their margins usually sunk a little below the general surface of the sponge. Skeleton. — (a) Dermal; consisting of a very irregular layer of tylote spicules, scattered, or in loose bundles and brushes which may commence beneath the surface and project outwards, (b) Main ; a very irregular reticulation of entirely spined styli. No very distinct fibres are present, although primary lines of spicules may be recognised, with little difficulty, running vertically towards the surface. Spicules. — («) Megasclera; (1) entirely spined styli (PI, XXVIl. fig. 11), with the spination well marked, usually slightly curved and sharply pointed at the apex, size about 0-52 by 0-034 mm. (2) Tylota (PI. XXVIL fig. 11a), with heads only faintly developed, entirely smooth, with straight shaft and weU-rounded ends, size about 0'42 by O'Ol mm. We have not found any distinct echinating spicules, although sometimes entirely spined styli, much smaller than those described, occur, which may perhaps represent them, {h) Microsclera ; (1) tridentate isochelae (PI. XXVIL fig. 116), with strongly curved shaft and of rather peculiar shape, length about 0'047 mm. (2) Very slender sigmata, usually much contort, length about 0'05 mm. Neither kind of microsclera occurs very abundantly. This sponge, represented in the collection only by two small pieces, closely resembles in external appearance an lophon, and may readily be distinguished from other Myxillse by its dark colour (should this prove to be a constant feature of the species). It is distinguished from Myxilla paupertas, Bk., and Myxilla nobilis, nobis, by the presence of the sigmata, the entire spination of the large stylus, and the absence of the echinating stylus ; and from the latter also by the fact that the tylota are smoothly rounded ofi" at the ends. The sponge contains a considerable amount of foreign matter, chiefly Foraminifera. Xoca%.— Station 150, February 2, 1874; lat. 52° 4' S., long. 71° 22' E.; west of Heard Island, Southern Ocean; depth, 150 fathoms; bottom, coarse gravel; bottom temperature, 35° "2. Two specimens. Myxilla mariana, Ridley and Dendy (PI. XXVIL figs. 12, 12a, 12&, 12c, I2d). 1886. JfyxiWa manana, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 472. Sponge massive ; represented in the collection only by a few fragments, the largest of which is about 25 mm. long by 12 mm. broad. Colour in spirit very pale, yellowish- grey. Texture fairly compact, but soft and rather crumbling. Surface smooth, more or less glabrous, but very uneven and deeply furrowed by wide meandering grooves, so as to give rise to a number of irregular, stumpy lobes. Dermal membrane thin, transparent, adhering rather firmly to the underlying tissues. Pores; these appear to be rather scarce, and we have observed no definite pore-areas. Oscula small and scattered. (ZOOL. CHALL. EXP. PAET LIX. — 1887.) Nmi 18 138 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton. — (a) Dermal ; consisting of a layer of irregularly but very thickly scattered tylota ; these spicules may also be found in irregular, short, thick fibres or brushes, disposed at right angles to the surface immediately beneath the dermal membrane, (b) Main; an abundant but very irregular and ill-defined reticulation of spined styli, sparsely echiuated by smaller spined styli. No very distinct fibres are found, although traces of such occur running vertically towards the surface of the sponge. Spicules. — [a) Megasdera; of three kinds, (l) Entirely spined styli (PI. XXVII. fig. 12), usually slightly curved and tapering gradually to a very sharp point at the apex ; with the spination distinct but not very strongly pronounced and most developed about the base; size about 0"42 by 0"016 mm.; making up the main skeleton. (2) A few, much smaller, entirely spined styli or subtylostyli (PL XXVII. fig. 12a), usually straight and tapering gradually from the base to a sharp point at the apex, size about 0'16 by 0 '0125 mm.; echinating the main skeleton. (3) Tylota (PI. XXVII. figs. 126, 12c), entirely smooth, with straight shaft and not very strongly marked oval heads narrowing towards the end ; size about 0-3 by 0-0094 mm.; the heads of these spicules are often more strongly developed at one end of the shaft than at the other, a character, however, which is by no means peculiar to the species in question. (6) Microsdera ; (1) tridentate isochelse,^ (PI. XXVII. fig. I2d), with only slightly curved shaft; length very variable, up to about 0-04 mm. (2) Slender, usually much contort sigmata, length up to about 0'057 mm. This sponge comes near to Alyxilla fusca, nobis, but is doubtless distinct, as indicated by the great difference in size of the larger spined styli and perhaps by the absence of echinating spicules in that species ; a very important distinction also lies in the forms of the chelas of the two species {cf. figs. 116 and 12d, PI. XXVIL). Locality. — Ofi' Marion Island ; depth, 50 to 75 fathoms. A few fragments. Myxilla mariana, var. massa, Ridley and Dendy (PI. XXVIL figs. 6, 6a, 66, 6c, Qd, 6e, 6/). 1886. Myxilla mariana, var. massa, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 473. Sponge massive, amorphous. The type specimen is irregularly globular in shape and about 37 mm. in diameter. Colour in spirit yellow. Texture soft and spongy. Surface very uneven, but not hispid. Dermal membrane very well developed, thin and transparent. Pores large and exceedingly numerous, plainly visible to the naked eye, collected in large, irregular groups over the subdermal cavities, reducing the dermal membrane in these places to a mere network. Oscula small, scattered, with their margins flush with the surface of the sponge. Skeleton. — (a) Dermal; scattered through the dermal membrane are a great number 1 Occasionally a four-toothed isocliela is found. REPORT ON THE MONAXONIDA. 139 of straight tylote spicules ; these spicules, just beneath the surface of the sponcre, are collected into loose, irregular fibres, which break up into tufts of spicules on approachino- the dermal membrane, (b) Main ; a very irregular, somewhat Halichondrioid reticulation of spined styli, echinated by much smaller spined styli. There is no very distinct fibre, though in places there appears to be a tendency towards the radiate arrangement characteristic of Bowerbank's genus Microciona. Spicules. — (a) Megasclera ; (1) long, slender, straight or slightly curved, entirely spined styli (PI. XXVII. figs. 6, &a), tapering very gradually to a fine point at the apex, measuring about 0"42 by O'Ol mm. (2) Much smaller, usually straight, entirely spined styli (PI. XXVII. figs. Qd, 6e), tapering gradually from base to apex and finely pointed, size usually about 0'14 by 0*0065 mm.; distinguished from the preceding by their position (echinating), their smaller size, and the fact that they are more strongly spined. (3) Straight tylota (PL XXVII. figs. 66, 6c), with usually distinct, elongated, oval heads, often narrowing towards the end; size about 0*28 by 0"0045 mm. (6) Mici'osclera ; (1) tridentate isochelae (PL XXVII. fig. 6f), with slightly curved shaft; length about 0'037 mm. (2) Slender sigmata, usually much contort ; length about 0'056 mm. ((?) (3) A few scattered, long, straight, hair-like, oxeote rhaphides ; length about 0"35 mm. It is possible that these spicules may have been taken in by the sponge as foreign bodies, and that they really belong to some of the Tedanise which occur at the same locality). Eejecting the long hair-like spicules as young forms or foreign, this sponge differs from the type of the species only in the comparative slenderness of the megasclera. The species comes very close in spiculation to Myxilla [Microciona) hihamigera, Waller.^ In the latter, however, the megasclera are all smaller and the diactinal spicules without heads (hastately pointed), while the isochelfe are described as being palmate ; the habit and the arrangement of the main skeleton also differ in the two species. Locality. — Off the south-west coast of Patagonia. One specimen. Myxilla compressa, Ridley and Dendy (PI. XXVII. figs. 9, 9a, 9&, 9c, 9d, 9e). 1886. Myxilla compressa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 473. Sponge massive ("?), flattened. The largest piece in the collection is about 56 mm. long by 25 mm. wide and 4 mm. thick. Colour in spirit yeIlow4sh-grey. Texture soft and spongy. Surface smooth where the dermal membrane is still intact, but the sub- dermal cavities show very distinctly through the transparent membrane and give to the sponge a honeycombed appearance. Dermal membrane distinct, thin and transparent. Oscida small and scattered, having their margins flush with the surface of the sponge. Pores collected in groups — pore-areas — in those portions of the dermal membrane which cover the subdermal cavities. The pores are round or oval and average about 0*2 mm. in diameter. 1 Jmirn. Quekett Micr. Club, vol. v. p. 1, pis. i., ii., 1878-79. 140 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton. — (a) Dermal ; consisting of scattered tylota, sometimes arranged in irregular brushes, (b) Main ; arranged very much as in Myxilla rosacea, var. japonica, but not so firm and dense ; the meshes of the sjsicular network are triangular and the side of each triangle is of one spicule's length, and usually composed of several sf)ined styli lying together parallel with one another, the rudimentary fibre thus formed being echinated, chiefly at the nodes, by smaller spined styli. Spicules. — (a) Megasclera ; of three kinds, (l) Entirely and considerably spined, almost or quite straight, sharply pointed styli (PI. XXVII. figs. 9, 9a), measuring about 0"28 by 0"0155 mm., making up the main skeleton. (2) Much smaller, straight, very sharply pointed and entirely spined styli (PL XXVII. fig. 96), measuring about 0"12 by O'OOS mm., echinating the skeleton fibre. (3) Tornota or tylota (PI. XXVII. fig. 9c), somewhat hastately pointed or with slightly developed, smooth, oval heads, pointed at the ends; size about 0"22 by 0'0063 mm., dermal. (6) Microsclera ; (1) tridentate isochelse (PI. XXVII. figs. 9cZ, 9e) of rather peculiar form, and with very strongly curved shaft ; length about 0"044 mm. Numerous much smaller isochelse also occur which at first sight appear to be different in shape, but they are probably merely young forms of the larger. The isochelse are extremely abundant, especially in the dermal membrane, forming in places an almost continuous layer. (2) Very small and very slender sigmata, usually contort and much bent, length usually about 0'02 mm., very rarely up to 0-063 mm. The external appearance of the sponge and the shape of the isochelse are well- marked characters of this species, and serve to distinguish it from others of the genus. ioca^.— Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; ofi" the mouth of the Eio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°'2. Six or eight pieces, mostly in a rather fragmentary condition. Myxilla noUlis, Eidley and Dendy (PI. XXVII. figs. 15, 15a, Ibh, 15c, 15fZ; PL XXX. figs. 2, 2a). 1886. Myxilla nohilis, Eidley aad Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. \). 473. Sponge (PL XXX. fig. 2) massive or lobate, may be encrusting. The largest specimen consists of two lobes, a large one and a small one, arising from a common base; the base is about 56 mm. long by 25 mm. broad and the larger lobe is 56 mm. in height. Another specimen encrusts the branches of a dead Coral. Colour in spirit greyish-yellow. Texture loose, soft and spongy ; rather cavernous. Surface uneven but not hispid. Dermal membrane very distinct, transparent, glabrous; easily separable from the under- lying tissues. Oscula ; round, scattered openings, having their margins flush with the general surface of the sponge. Pores very distinct, arranged for the most part in very REPORT ON THE MONAXONIDA. 141 definite, rounded or oval pore-areas (PI. XXX. fig. 2a) about 1 mru. in diameter; these pore-areas are often very thickly scattered over the surfiice of the sponge, being separated from one another only by thin strands of tissue. The dermal membrane within the pore-areas is reduced to a fine, lace-like network by the presence of numerous large pores, of which from thirty to forty often occur in a single pore-area. The pores themselves are rounded openings, usually oval, but often more or less polygonal and varying in greatest diameter from 0"07 to 0'35 mm. Skeleton. — (o) Dermal; a loose reticulation of scattered tylota, often arranged in small brushes, in a radiate manner, around the pore-areas, (b) Main ; a very loose and irregular reticulation of styli, united together in parts to form loose fibres ; rather abundantly echinated by smaller, s^^ined styli. Spicules. — (a) Megasclera; of three kinds. (1) Styli (PI. XXVII. fig. 15), entii-ely smooth or slightly spined at the base ; stout, usually slightly curved, and tapering gradually to a very sharp point at the apex; size about 0"52 by 0'03 mm; making up the main skeleton. (2) Much smaller, entirely spined styli (PL XXVII. fig. 15a), usually straight, tapering gradually from the base to a sharp point at the apex; with the spination more jironounced at the base than elsewhere, and often with distinct heads (tylostylote); size about 0"18 by 0 '01 3 mm. ; echinating the loose fibres of the main skeleton. (3) Tylota (PI. XXVII. fig. 156), with straight, smooth shaft and slightly expanded heads, which are very short and abruptly truncated, often slightly spined at the end; size about 0"33 by 0'0063 mm.; dermal, (b) Microsclera; tridentate isochelse (PL XXVII. figs. 15c, Ibd), with strongly curved, stout shaft; variable in size, measuring up to about 0*044 mm. in length; exceedingly abundant, especially in the dermal membrane. Occasionally smooth, simple sigmata are seen, about 0"09 mm. long, but so rarely as to suggest that ■ they probably occur only as foreign bodies. A remarkable feature of this species is the frequent truncation of the heads of the tylota, often accompanied by minute spination. This, however, is not a constant feature and may sometimes be observed in Myxilla rosacea, Schmidt; it may also occur in Myxilla cribrigera, nobis. The two species, Myxilla cribrigera and Myxilla nobilis agree closely in several respects, but are sharply distinguished from one another by the fact that the former has no echinating spined styli such as are present in the latter. Another very interesting point is the transitional character of the larger stylote spicule ; it is apparently normally smooth, but may be slightly spined {cf. Myxilla paucispinata, nobis, p. 132). Localities. — Station 148a, January 3, 1874, lat. 46° 53' S., long. 51° 52' E.; south of the Crozets; depth, 240 to 550 fathoms; bottom, hard ground, gravel, shells. One specimen, apparently a piece broken off" from a larger one. Probably belonging to a slight geographical variety. Station 320, February 14, 1876 ; lat. 37° 17' S., long. 63° 52' W.; off" the mouth of 142 THE VOYAGE OE H.M.S. CHALLENGER. the Rio de la Plata; depth, 600 fathoms; bottom, green sand ; bottom temperature, 37°*2. Three specimens (being the three which we regard as most typical of the species and from which the above description is taken). Myxilla nohilis, var. patagonica, Eidley and Dendy (PL XXVII. figs, 13, 13a, 136). 1886. Myxilla nohilis, var. patagonica, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 473. Sponge massive, amorphous. Size of largest specimen 37 mm. long by 19 mm. broad and 12 mm. thick. Colour in sj^irit light, yellowish-brown. Texture soft and crumbling. Surface uneven, rugose, traversed by meandering furrows, calling to mind the appearance of the Brainstone Coral [Mgeandrina). Dermal membrane distinct, delicate, tran.sparent. Oscula and Pores not observed. Skeleton. — (a) Dermal ; very feebly developed, consisting of scattered tylota. (6) Main; a roughly Halichondrioid reticulation of styli, with irregularly rectangular meshes, about one .sj)icule's length in breadth. The rudimentary fibres, which are almost devoid of horny cementing substance, are echinated, more especially at the nodes, by smaller, spined styli, the bases of which rest upon the filjre. Spicules. — (a) Megasclera; of three kinds, (l) Basally spined styli (PL XXVII. fig. 13), sometimes verging upon the tylostylote form ; usually more or less curved and tapering gradually to a sharp point at the apex; size about 0*42 by 0*02 mm.; making up the main skeleton. (2) Much smaller styK (PL XXVII. fig. 136), often with distinct heads (tylostylote); entirely spined, but more so at the base than elsewhere; usually straight; size up to about 0'175 by O'Ol mm., echinating the skeleton fibre at the nodes. (3) Tylota (PL XXVII. fig. 13a), long and slender, and with but slightly marked oval heads. The heads are sometimes flattened at the ends (and with a very faint trace of spina- tion ?). Size about 0'3 by 0'006 mm. Occurring scattered in the dermal membrane. {h) Microsclera; of one kind only, viz., tridentate isochelse, with strongly curved shaft, length up to about 0"04 mm. i:oca%.— Station 311, January 11, 1876; lat. 52° 45' 30" S., long. 73° 46' 0" W.; ofi' the south-west coast of Patagonia; depth, 245 fathoms; bottom, blue mud; bottom temperature, 46°*0. One specimen (containing a good number of embryos scattered through it) and a few fragments. Myxilla nohilis, var. hacillifera, Ridley and Dendy (PL XXVII. figs. 14, 14«, 146). 1886. Myxilla nohilis, var. bacilli/era, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 473. Sponge massive, amorphous. The largest specimen is irregularly globular in shape and about 25 mm. in diameter. Colour in spirit yellowish-grey. Texture very soft and spongy. Surface uneven, but not hispid. Dermal membrane distinct, very thin and EEPORT ON THE MONAXONIDA. 143 transparent. Oscula few, scattered, rather small. Pores arranged in definite, but rather irregular groups (pore-areas), each group containing about twenty or thirty, fairly closely placed, rounded or oval pores ; diameter of pores about 0"! mm. Skeleton. — Very irregular ; in parts one can distinguish primary lines of loose fibre running vertically to the surface of the sponge, crossed by still looser secondary fibres or by scattered spicules ; in other parts the arrangement is even less regular and may be described as irregularly Halichondrioid. The fibre, such as it is, is composed of basaUy spined styli, and is echinated by much smaller, entirely spined styli (also in places by the larger styli ?). At the surface the main skeleton passes into loose fibres and brushes of tylota. The latter are abundantly present in the dermal membrane, excepting in the pore-areas, which are almost or quite free from spicules and have the tylota around them arranged in a radiate manner. Spicules. — (a) Megasclera; of three kinds. (1) Basally spined styli (PI. XXVII. fig. 14), sharply pointed and usually curved; size about 0 '46 by 0-019 mm.; making up the main skeleton. (2) Much smaller, entirely spined styli (PL XXVII. fig. 14a), usually straight, sharply and very gradually pointed at the apex, sometimes with distinct heads (tylostylote) ; size about 0'175 by 0'0125 mm.; echinating the skeleton fibre. (3) Tylota (PI. XXVII. fig. 1 46), with straight shafts and very slight heads, which terminate very shortly and abruptly and are flattened and slightly spined at the ends ; size about 0'3 by 0'006 mm.; dermal. (6) Microsclera; of one kind only, viz., tridentate isochelse, with very strongly curved shafts ; size variable, up to about 0'034 mm. long. The most characteristic feature of this variety is the extent to which the flattening of the ends of the tylota is carried ; the proj)ortions of the large spined stylus also difier considerably from those of the corresponding spicule in the type specimens, as will be seen by reference to the descriptions. XocaZ%.— Station 307, January 4, 1876 ; lat. 49° 24' 30" S., long. 74° 23' 30" W.; ofi" the south-west coast of Patagonia; depth, 140 fathoms; bottom, blue -mud. Two specimens. The species which we have called Myxilla nohilis, and its varieties, have given us a great deal of trouble in determining their true relations ; they appear to be sufliciently connected inter se to wairant us in considering them all as varieties of one species, and that species perhaps finds its nearest already known ally in Bowerbank's Hymeniacidon {^Myxilla) 2^aupertas ; the two species seem, however, to be distinct. Bowerbank apj^ears to have had only very small and badly preserved specimens, about which nothing is satisfactorily known excepting the spiculation, which, as evidenced by one of his preparations in the British Museum, is as follows: — (l) Basally spined styli, verging upon tylostylote, curved, attenuated and drawn out very gradually into very fine points ; size up to about 0"52 by 0'012 mm. (2) Entirely spined styli (" internal defensive 144 THE VOYAGE OF H.M.S. CHALLENGER. spicula," Bk.), but more spined at the base than elsewhere ; usually curved, finely pointed; verging upon tylostylote ; size up to about 0'2 by 0"0084 mm. (3) Very slender tylota, with very slightly developed heads, not spined nor flattened at the ends ; size up to about 0"32 by 0"0047 mm. (4) Tridentate isochelse,^ with strongly curved shaft and divergent teeth ; length about 0"04 mm. The 'points in which Bowerbank's species difi"ers from Myxilla nohilis will be readily seen by comparison of the descriptions ; they concern more especially the form of the larger stylus and of the tylote spicule.^ Myxilla froridosa, Ridley and Dendy (PI. XXVI figs. 1, la; PI. XXVII. figs. 10, 10a, 106, 10c, lOfZ, lOe, 10/). 1886. Myxilla frondosa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 473. Sponge (PI. XXVI. figs. 1, la) represented by a single broad, flattened frond, of some- what oval shape, about 6 mm. thick, 100 mm. in height and 62 mm. in breadth. It has evidently grown in an erect position. Colour in spirit rather dark, yellowish-grey. Texture tough, fibrous, elastic. Surface, on one side, which is slightly convex, very rough, owing to tlie presence of thickly placed, small conuli, between which are the oscula. The other side is much smoother, the conuli are not so strongly pronounced, and the dermal membrane stretches over them continuously except where it appears to have been rubbed off". Dermal inembrane distinct, thin and transparent. Oscula very numerous, small, round and thickly scattered over the convex surface, to which they appear to be confined ; about 1 mm. in diameter. Pores very numerous, scattered, rounded openings in the dermal membrane on the concave surface, diameter variable, averaging about 0*2 mm. We have seen no pores on the other surface of the sponge. Skeleton. — (a) Dermal; consisting of tylote spicules, which are for the most part irregularly scattered through the dermal membrane but are sometimes found in loose tufts. (6) Main ; a very highly developed and compact reticulation of coarse, branching and sometimes anastomosing spiculo-fibre. The main lines of the skeleton, starting from the base, branch upwards in a dendroid manner, with their ramifications lying mainly in one plane, and giving ofi", approximately at right angles to this plane, very numerous, short branches which end in the conuli on the surface of the sponge. The fibre itself is stout and consists of closely placed styli which have their bases in the centre of the fibre and their apices directed very obliquely outwards and free, point- ing towards the direction in which the fibre is running ; there seems to be very little cementing substance uniting the spicules. S2ncules. — (a) Megasclera ; (l) Entirely, but only slightly spined styli (PI. XXVII. ' The figure of this spicule given by Bowerbank (Mon. Brit. Spong., vol. iii., pi. xxxvii. fig. 8) appears to us to convey a very erroneous idea of its true shape. ^ Cf. Bk., Mon. Brit. Spong., vol. iii., pi. xx,\vii. figs. 5, 6. REPORT ON THE MONAXONIDA, 145 fig. 10), more spined at the base than elsewhere ; slightly curved and tapering gradually to a sharp point at the apex; size about 0'6 by 0'03 mm., forming the main skeleton. It will be seen from their arrangement that all the spicules of the main skeleton are, in a sense, echinating, so that we cannot distinguish between an echinating and a non- echinating stylus as in some other Myxillm. There are, however, a great many stylote spicules (PI. XXVII. figs. 10a, 106) which are much smaller (measuring about 0"28 by 0'013 mm.), and rather more strongly spined than those just described, but otherwise very like them in form, and these are perhaps to be regarded as representing the true echinating spicules ; it may be urged that they are merely young forms of the larger ones, but their abundance and tolerable uniformity in size are against this view. (2) Tylota (PI. XXVII. figs. 10c, lOcZ), usually slightly curved, rather stout, larger at one end than at the other, with oval heads (sometimes not distinguishable), and spined at both ends, size about 0'25 by O'Ol mm. ; dermal, {h) Microsdera; of two kinds — (1) very numerous tridentate isochelae (PL XXVII. fig. lOe 10/), of peculiar form, the most noticeable feature in which is the presence of a slight swelling in the centre of the curved shaft ; length about 0'027 mm. ; (2) sigmata, usually much contort, size about 0"044 by 0-004 mm. The definite form, erect growth and elastic consistency of this sponge call to mind the genus Clathria, and the arrangement of the skeleton recalls PlumohalicJiondria rather than Myxilla, but the spiculation agrees very well with that of the genus to which we have referred it ; in its genus it is further remarkable for the peculiar shape of its isochelate spicule. ZocaZ%.— Station 170, 14 July, 1874 ; lat. 29° 55' S., long. 178° 14' W. ; off Kermadec Islands; depth, 520 fathoms; bottom, volcanic mud; bottom temperature, 43°. One specimen. Myxilla (f) plumosa, Montagu, sp., vav. /usifera, nov. 1818. Spongia plumom, Montagu, Mem. Wem. Soc. Edin., val. ii. p. 116. 1842. Halichondria (?) plumosa, Johnston, British Sponges, p. 103 1866. Mkrociona carnosa, Bowerbank, Mon. Brit. Spong., vol ii. p. 133. 1866. Hymeniaeidon 2}lumosa, Boworbank, Mon. Brit. Spong., vol iL p. 195. 1867. Pronax plumosa, Gray, Proc. Zool. Soc. Lend., p. 536. 1870. Desmai-idon plumosa, Schmidt, Spong. Atlant. Gebiet., p. 76. 1874. Microeiona plumosa, Bowerbank, Mon. Brit. Spong., vol. iii. p. 61, pi xxiv. figs. 7-13. 1880. Myxilla plumosa, Vosmaer, Notes from the Leyden Museum, vol ii. p. 126. 1885. Plumohalieliondria plumosa. Carter, Ann. and jMag. Nat. Hist., ser. 5, vol xvi. p. 355. Sponge irregularly lobate. The largest specimen consists of an elongated base, from which arise two lobes ; length of base not quite 25 mm., height of larger lobe 25 mm., diameter about 8 mm. Coloior in spirit pale, greyish yellow. Texture soft, spongy, elastic, rather tough and fibrous. Surface rather glabrous, but very uneven. Dermal (zool. chall. EXP. — PART Lix. — 1887.) Nnn 19 146 THE VOYAGE OF H.M.S. CHALLENGER. membrane fairly distinct, thin and transparent. Oscula (?). Pores scattered or in irregular groups. Skeleton. — («) Dermal ; composed of thickly strewn, spined styli and fusiform oxea, the former sometimes predominating and the latter sometimes collected into distinct bands of fibre, (b) Main; composed of stout ropes of spined stylote spicules, abun- dantly echinated by numerous other spined styli apparently of the same kind. These fibres are not very regularly disposed and appear to originate from any foreign body which may form a convenient base, from which they radiate towards the surface, often branching in their course. In addition to these fibres, abundant, loosely scattered, spined styli occur. Spicules. — (rt) Megasclera; (1) small, usually curved, sharply pointed, entirely and strongly spined styli, varying in size from O'l by O'OOS to 0'16 by O'OIS mm. We have not been able to satisfy ourselves that there are really two distinct kinds of spined styli, although great variation in size occurs. (2) Fusiform oxea ; slender, and tapering gradual^ to a very sharp point at each end; size about 0"175 by 0"0048 mm.; these spicules are probably homologous with the dermal spicules of tj^iical Myxillie. (b) Microsclera; of one kind only, viz., tridentate isochelas, with very strongly curved shaft, length about 0-015 to 0-02 mm. The Challenger variety difiers from the British form chiefly in the possession of a more tapering oxeote spicule and a stouter isochela. This sjiecies is very aberrant in its skeleton arrangement, resembling Myxilla frondosa, nobis, in this respect, and forming a connecting link between Myxilla and Plumo- Jialichondria. It is only doubtfully that we include it in the former genus. Locality. — Ofi" Bahia, shallow water. One specimen and a fragment. Genus Clathria, Schmidt (Pis. XXVIIL, XXIX., XLVIL). 1862, Clathria, Schmidt, Spong. Adriat. Meer., p. 57. 1870. Teimcia, Schmidt, Spong. Atlant. Gebiet. p. 56. Of various form, generally erect ; possessing usually a well-developed horny fibre cored by stylote spicules and echinated by smaller spined styli. Typical microsclera small palmate isochelse ; no special dermal crust of spicules as in Rhaphidophlus. Schmidt's original diagnosis {loc. cit.) runs as follows : — " Halichondrise maxime ramosse, ramis in modum clathrorum ssepissime inter se connexis. Substantia aciculas involvens subcornea, elastica dum aqua est imljuta, fragilis et fere friabilis, dum spongia est exsiccata. Rete microscopicum spiculorum, ista substantia conjunctorum maxime irregulare." This, though very true so far as it goes, is obviously insufiicient, and Vosmaer ' in > 1 Notes from the Leyden Museum, vol. ii. p. 149. REPORT- ON THE MONAXONIDA. 147 1880 improves upon it with the following: — "Rods smooth or spined, ' echinating.' Anchors minute, equiended ;" this also is very wide and really does not give so good an idea of the genus as Schmidt's original diagnosis. It has seemed to us advisable to restrict the genus by the diagnosis given above. In 1880 {loc. cit.) Vosmaer placed Clathria amongst the Desmacidonidse, saying: — " I have found that the genus Clathria of Schmidt possesses distinct, often very numerous anchors and bows ; so it is to be placed under the Desmacidinse ;" this certainly appears a sensible arrangement, but in 1885 he adopts a different plan and places it apart from the Desmacidines, amongst the Ectyonidse.^ We have endeavoured to show elsewhere that the old Ectyonidse and Desmacidonidae must fall together in one group, being connected through the genus Myxilla (cf. p. 129), in which both species with and species without echinating spicules are found. The two genera Clathria and Myxilla seem to be nearly related, and our Myxilla frondosa forms an interesting connecting link. The original type of the genus is Clathria corall aides, Schmidt {loc. cit, p. 58), an Adriatic species ; but this species is, unfortunately, not a very good example of the genus as now understood; the next one mentioned by Schmidt {loc. cit.), viz., Clathria comi^ressa, appears to be much more typical. Tenacia clathrata, Schmidt,^ is simply a Clathria with very great actual and relative development of the horny fibre. Clathria appears to be essentially a shallow- water genus ; and this fact is no doubt correlated with its horny fibre and fondness for warm seas ; the slight exception presented by Clathria inanchorata {vide infra), from a depth of 120 fathoms, is associated with an aberrant spiculation. The distribution of the genus is wide in tropical waters, but it seems to be best represented in the Indo- Australian area ; as in the case of other horny sponges the genus attains its maximum of development in warmer climates. Clathria aculeata, Ridley. 1884. Clathria aculeata, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 443, pi. xl. fig. i; pi. xlii. fig. K. This species was obtained by the " Alert " in Torres Strait, in which locality the Challenger also found it. It has already been sufficiently described and figured in the " Alert " report {loc. cit.). Locality.— Station 186, September 8, 1874; lat. 10° 30' S., long. 142° 18' E. ; Torres Strait ; depth, 8 fathoms ; bottom. Coral mud. One specimen. Habitat. — Torres Strait, shallow water (" Alert " and Challenger). 1 Bronn's Klass. u. Ordii. des Thierreiohs, Porifera, p. 356. 2 Spong. Atlant. Gebiet., p. 56. 148 THE VOYAGE OF H.M.S. CHALLENGER. Clathria decumhens, Ridley. 1884. Clathria decumhens, Eidley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 612, pL liii. fig. k ; pi. liv. figs, g, g'. With this species we identify a washed out skeleton from Torres Strait ; it differs slightly from the type in external form and in the proportions of the spicules, and the spines at the base of the smooth stjdus are very rarely seen. It much resembles in external appearance specimens of Acarnus ternatus obtained by the " Alert " from the same locality, forming " a clathrous structure of round soft anastomosing trabeculse."^ Locality. — Cape York, Torres Strait; August 7, 1874; depth, 3 to 11 fathoms. One specimen. Habitat. — Boudeuse and Etoile Islands, Amirante Group, 10 to 13 fathoms (Ridley, " Alert") ; Torres Strait (Challenger). Clathria lendenfeldi, Ridley and Dendy (PI. XXVIII. fig. 5 ; PI. XXIX. fig. 6 ; PI. XLVII. fig. 5). 1886. Clathria Lendenfeldi, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 474. Sponge (PI. XXVIII. fig. 5) based upon a slender Gorgonia axis, which it encrusts. From the encrusting portion arise several straight, cylincbical, unbranched processes about 8 mm. in diameter, the longest of which is 94 mm. high. Colour in spirit light yellow. Texture soft, fibrous, elastic. Surface hispid, with a reticulate ajapearance due to the reticulate main skeleton (the specimen appears, however, to be a good deal worn). Oscula small and scattered. Skeleton. — (a) Dermal; consisting of very slender, irregularly scattered, stylote spicules, often found in loose, irregular brushes. (&) Main ; a reticulation of stout, well-developed horny fibre, cored in places (chiefly at the ends of the primary lines) by smooth stylote spicules. These spicules project freely from the ends of the j^rimary fibres and thus give rise to the hispidation of the surface. The fibre (PI. XLVII. fig. 5) abundantly is very echinated by spined stylote spicules, which are most abundant on the secondary fibres. The secondary fibres are, also, almost or entirely without an axial core of spicules. Spicules. — (a) Megasclera ; (1) long, straight, very slender, smooth styli or subtylo- styli, very sharply and gradually pointed at the apex, sometimes faintly and minutely spined at the base, measuring about 0'35 by 0"005 mm.; dermal. (2) Much larger, stouter, smooth, very gradually and sharply pointed styli, slightly curved and measuring about 0"6 by 0*02 mm. ; in the main skeleton. Much smaller styli also occur as constituents » Eidley, Zool. Coll. H.M.S. " Altrt," Brit. Mus., 1884, p. 453. EEPORT ON THE MONAXONIDA. 149 of the main skeleton, the largest being generally found at the ends of the primary fibres. (3) Short, thick, straight, very bluntly pointed styli (PL XXIX. fig. 6), strongly spined all over, size about 0'08 by 0"005 mm. ; echinating the skeleton fibre in great numbers. (b) Microsclera ; of one kind only, viz., minute palmate isochelse about 0'005 mm. long. We have called this species after Dr. von Lendenfeld, who has a much fuller acquaintance with the sponge (viz., in its native haunts) than we have, although he has as yet published no description of it. The strong development of the horny fibre, with the coincident reduction in the spicular element of the skeleton, and the form and great abundance of the echinating spicules, are good characters by which to recognize the species. Locality. — Ofi" Port Jackson. One specimen. Clathria frondifera, Bowerbank, sp. 1875. Halichondria frondifera, Bowerbank, Proc. Zool. Soc. LonJ., April 1875, p. 288. 1880. AmiMlechis frondifer, Vosmaer, Notes from the Leyden Museum, vol. ii. p. 115. 1884. Clathria frondifera, Eidley, Zool. Coll. H.AI.S. "Alert," Brit. Mus., pp. 448, 612, pi. xlii. fig. j : pi. liii. fig. J. This is a widely distributed and very common shallow-water species, and has been already described by two authors (Bowerbank and Eidley, loc. cit), hence it is not necessary to enter into further details regarding it in this place. XocaZ%.— Station 188, September 10, 1874; lat. 9° 59' S., long. 139° 42' E.; west of Torres Strait ; depth, 28 fathoms ; bottom, green mud. One fine specimen. Habitat. — Strait of Malacca ; Gaspar Strait (Bowerbank) ; west of Torres Strait (Challenger) ; Thursday Island ; Prince of Wales Channel, Torres Strait ; Percy Island and Fitzroy Island, Queensland ; Providence Reef and Island, Mascarene Group ; Amirante Islands; SeycheUe Islands (Ridley, "Alert"). Clathria elegantula, Ridley and Dendy (PI. XXVIII. figs. 3, 3a; PI. XXIX. figs. 14, 14a, lib). 1886. Clathria elegantula, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviiL p. 474. Sponge (PI. XXVIII. fig. 3) sessile, the single specimen present consisting of several much flattened, expanded, divided lobes, based upon half of a dead bivalve shell. Height 81 mm., greatest breadth about the same. Thickness of lobes usually about 10 mm. The margins of the lobes are more or less deeply notched and slightly undulating. Colour in spirit pale, brownish-yellow. Texture soft and spongy, elastic and fibrous ; internally rather cavernous. Surface very uneven, beset with numerous projecting conuli, between which is stretched the thin, semi- 150 THE VOYAGE OF H.M.S. CHALLENGER. transparent, parchment-like dermal membrane, covering the large subdermal cavities, and constituting a very characteristic feature of the species (PI. XXVIII. fig. 3a). Oscula smaU and scattered (?) (there are a great number of round openings in the dermal membrane, especially on one side of the sponge, which might readily pass for oscula, but it is impossible to say how far they are due to shrinking back of the dermal membrane from the action of the spirit). Pores ; on one side of the sponge are a number of small oval areas in which the dermal membrane is rediiced to a sieve by numerous small perforations, which are the pores. The pore-areas are well defined and rather sparsely scattered ; it is curious that they should occur on ,one side only, that being the side on which most of the oscula (?) are situated. Diameter of pore-areas and pores variable, of the former averaging about 0'5 mm. of the latter about 0"! mm. Skslet07i.— In balsam preparations the arrangement of the skeleton somewhat resembles that of Plumohalichondria, the chief feature being a series of ascending, plumose columns, each consisting of a core of very slender, subtylostylote spicules echinated by entirely spined styli, the horny matter being almost invisible ; but on examination in alcohol we see that there is a very large proportion of pale homy substance in the skeleton, which is arranged as follows : — There is a well-defined reticulation of horny fibre ; branching primary fibres run upwards in wavy lines, ending at the surface of the sponge, and these alone are cored by long, slender, subtylostylote spicules. Secondary fibres cross the primary lines without much regularity, forming an anastomosing reticulation, and both series of fibres are echinated by spined styli. In the dermal membrane there are numerous subtylostylote spicules, scattered and in brushes. S2ncides. — (a) Megasclera; of two kinds. (1) Very slender, straight, smooth sub- tylostyli (PI. XXIX. fig. 14), measuring about 0"2 by 0'003 mm.; (2) straight, slender, sharp pointed, entirely spined, echinating styli (PI. XXIX. fig. 14a), measuring about 0'07 by 0'0032 mm. (b) Microsclera; of one kind only, viz., palmate isoehelee (PI. XXIX. fig. lib), of rather unusual form, with very slender shaft, rather strongly curved and making an unusually wide angle with the front palm ; length about 0"02 mm. This sponge is most readily recognised by its beautiful and characteristic external form. XocaZ%.— Station 162, April 2, 1874 ; lat. 39° 10' 30" S., long. 146° 37' 0" E. ; ofi" Moncoeur Island, Bass Strait; depth, 38 fathoms; bottom, sand and shells. One fine specimen. Clathria inanchorata, Ridley and Dendy (PI. XXVIII. fig. 4; PI. XXIX. figs. 13, 13a). 1886. Clathria inanchoraia, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 475. Sponge (PI. XXVIII. fig. 4) erect, slender, of very peculiar, rugged external appearance, the best idea of which wiU be obtained by reference to the figure. The REPORT ON THE MONAXONIDA. 151 larger specimen is 200 mm. in height aud has an average diameter of about 12 mm. Colour in spirit dull, yellowish-brown. Texture tough aud fibrous. Surface extremely- uneven and minutely hispid. Oscula numerous, small and scattered. Skeleton. — There is a reticulation of very stout horny fibre, echinated by spined stylote spicules, and having the primary lines cored by smooth styli which may also project freely at the ends of the primary fibre. The secondary fibres appear to be without a spicular core. Spicules. — (a) Megasclera ; (l) smooth styli, very variable in size, measuring up to about 0-54 by 0'024 mm., but often much smaller. These spicules occur in the axis and projecting at the ends of the primary fibre ; those at the ends are of large size. (2) Nearly or quite straight, sharply pointed, entirely spined styli (PL XXIX. fig. 13), echiuating the skeleton fibre ; size about 0'072 by O-QOG mm. (6) Mkrosclera ; of one kind only, viz., smooth toxa (PI. XXIX. fig. 13, a) measuring about 0-029 by 0-0016 mm. The very characteristic external appearance and the fact that there is only one kind of miscrosclera present, and that a toxite, are, perhaps, the most characteristic features of this species, by which it may readily be recognized. Although it possesses no chetse, yet this species agrees so closely with the genus Clathria in other respects that we have deemed it advisable to include it in that genus ; it is, perhaps, a form that once possessed isochelate microsclera and has now lost them. Locality. — Station 163a, April 4, 1874 ; lat. 36° 59' S., long. 150° 20' E.; Bass Strait; depth, 150 fathoms; bottom, green mud. Two specimens. Genus RhaphidoiMus, Ehlers (Pis. XXVIII., XXIX., XL VI.). 1870. Rhaphidophlus, Ehlers, Die Esper'schen Spongien, Erlangen, pp. 19, 31. 1875. Eeldiionema, Carter, Ann. and Mag. Nat. Hist, ser. 4, vol. xvi. p. 185. This genus difi"ers from Clathria only in the possession of a distinct, dense crust of outwardly projecting, spicules, the difi"erence being one of degree rather than of kind. The original diagnosis (loc. cit., p. 31) runs as follows : — " Schwamm aus netzformig vereinigten Balken mit dichter Rindeuschicht aus stumpf-spitzen Nadeln, daruiiter im Gewebe ein Netz von Hornfasern, in welchen und um welche die gleichen Nadeln liegen, ausserdem eingepflanzte gedornte Nadeln, daneben gleichendige Doppelanker und mannigfach gebogene Kieselfaden." This diagnosis is based upon a single species, RhaphidopJilus cratitius, Esper, sp., and is perhaps rather too restricted for generic use. The nature of the " Kieselfaden " is elucidated to some extent by the description of the species (loc. cit., p. 19), which speaks of them as " sehr feine haarformige Kieselfaden, welche ungieich lang und mannigfach gekriimmt sind : einfach spangenformig, oder mit wieder aufgebogenen Enden, oder augh S-fdrmig, doch so, dass die Endtheile nicht in eine Ebene liegen." 152 THE VOYAGE OF H.M.S. CHALLENGER. Although the language is a little ambiguous there is no doubt that sigmata are here referred to. Mr. Carter's genus " Echinonema " must be suppressed, as the species which it includes are not generically separable from the older genus " Rha^jhidophlus," which Mr. Carter appears never to have recognised.^ The genus " Rhaphidophhts" has a further claim to take precedence of " Echinonema " in that its founder gave a generic diagnosis thereof. Rhaphidophlus gracilis, Ridley, sp. 1884. JEchinonema gracilis, B.idlej, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 617, pi. liv. figs. 1, r, 1". One specimen of this species was obtained by the Challenger off Bahia. It agrees very minutely in external form and in spiculation with the type, but instead of being of a " very dark purplish -brown colour " (in spirit), it is greyish -yellow. It is very interesting to find this well-marked species occurring at two so widely separated localities as the Mascarene Islands and Bahia. Locality. — Off Bahia ; dejsth, 7 to 20 fathoms. One specimen. Habitat. — Providence Reef, Mascarene Islands (Ridley, " Alert ") ; off Bahia (Challenger), Rhaphidophlus filifer, Ridley and Dendy (PI. XXVIII. fig. 2 ; PI. XLVI. fig. 9). 1886. Rhaphidophlus filijer, Ridley and Dendy, Ann. and Mag. Nat Hist., ser. 5, vol. xviii. p. 475. Sponge (PI. XXVIII. fig. 2) irregularly ramose ; branches approximately cylindrical, but very uneven and with a gnarled appearance ; the habit of the sponge appears to have been suberect or creeping. Colour in spirit greyish-yellow. Length of specimen 94 mm., average diameter of branches about 6 mm. Texture hard and almost incom- pressible. Surface very rugose and uneven, but with a slightly glabrous appearance. Skeleton (PL XLVI. fig. 9). — [a) Dermal; consisting of a compact crust, about 0'14 mm. in thickness of densely packed, projecting brushes of stylote spicules with outwardly directed points. When viewed in surface section these tufts are seen to be arranged in a reticulate manner, so as to form approximately circular meshes about O'l mm. in diameter. (&) Main ; a well-developed, but rather irregularly disposed reticulation of spiculo-fibre ; one can in places roughly distinguish primar)^ fibres, running towards the surface, from secondary fibres which cross them more or less at right angles and thus give rise to approximately rectangular meshes. The fibre itself is stout and well defined and is 1 Cf. also Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., 1884, p. 61.5. EEPORT ON THE MONAXONIDA. 153 composed of smooth stylote spicules united together by a large proportion of horny matter ; it is echinated by numerous smaller, spined styli, which project from the surface at right angles (PL XLVI. fig. 9). Spicules. — (a) Megasclera ; of three kinds — (l) nearly or quite straight, rather slender, sharply and gradually pointed styli, usually with slightly spined base ; measuring about 0"2 by 0"0065 mm., forming the dermal crust. (2) Smooth, slightly curved, stout styli, measiiring about 0"3 by 0"018 mm., forming the main skeleton. (3) Small, straight, entirely spined styli, measuring about 0"1 by O'Ol mm., echinating the skeleton fibre. (h) Microsclera; of two kinds, (l) Minute, palmate isochelse, about 0'016 mm. long. (2) Very long, slender, hair-hke toxa, usually occurring in very loose bundles, length about 0*16 mm. These are the most characteristic spicules of the species, and it is from their hair-like nature that the specific name has been taken. They may at once be recognised as toxa by the sharp bend in the centre. From the same station as this sponge comes a species of Pachychalina, which so closely resembles the specimen of Rhaphidophlus Jilifer in external appearance as to be very readily mistaken for it (compare PL IV. fig. 1 with PL XXVIII. fig. 2). This close resemblance in external form between two such widely different species is very remarkable. The Pachychalina is a shade lighter in colour than the Rhaphidophlus. This species resembles Mr. Carter's common Australian species Echinonema typicum,^ the latter is, however, of a much more luxuriant growth, and the proportions of the spicules also difier. According to the original description there are no chelae or toxa in Echinonema typicum, and only two forms of megasclera, but an examination of the dried types in the British Museum has given us good reason for asserting our belief in the presence of both forms of microsclera. We have not, however, found sufficient grounds for an identification. A redescription of Echinonema typicum from spirit specimens is much needed. Loco%.— Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E.; Philippine Islands ; depth, 1 8 fathoms ; bottom, blue mud. One specimen. Rhaphidophlus lohatus, Vosmaer, sp., var. horrida,^ nov. (PL XXVIII. fig. 1 ; PL XXIX. figs. 4, 4a, 46, 4c). 1880. Clathria lobata, Vosmaer, Notes from the Leyden Museum, vol. ii. p. 151. Sponge (PL XXVIII. fig. l)stipitate, compressed, palmate, dichotomously branching. The stem, which is about 6 mm. in diameter, springs from a flattened and slightly ex- panded base and divides into two at a height of two inches from the ground, each branch again subdividing. Total height of specimen 106 mm.; breadth at the top (the broadest part) 44 mm. Colour in spirit light, greyish -yellow. Texture fine, fairly 1 Ann. and Mag. Nat. Hist., ser. 5, vol. vii. p. 378. 2 So called from the bristling appearance of the surface when seen under a low pofl'er of the microscope. (ZOOL. CHALL. EXP. PART. LIX. 1887.) Kuu 20 154 THE VOYAGE OF H.M.S. CHALLENGER. firm aud compact. Surface even, very minutely hispid. Dermal membrane thin and difficult to distinguish. Pores scattered (?). Oscula none observed; the sponge appears to be lipostomous. Skeleton. — The main skeleton is a somewhat Isodictyal reticulation of stout stylote spicules ; there is no very distinct fibre (although one can usually distinguish primary skeleton lines running towards the surface and crossed at right angles by secondary lines), but there is a fair amount of horny connecting substance at the nodes of the reticulation ; the whole arrangement is rather vague and confused. In addition to the smooth stylote spicules, which form the chief portion of the main skeleton, there are numerous spined styli, some of which appear to echinate the main skeleton while others are scattered loosely through the soft tissues (some of the latter may have been forced from their proper position in cutting the sections ; there can be little doubt that these spined spicules are really homologous with the echinating spicules of allied species). The primary fibres end on the surface in dense tufts of slender stylote spicules whose points project outwards; mingled with these are more of the echinating spined styli already mentioned, while the centre of each tuft is occupied by one or more large, smooth styli like those of the main skeleton. Spicules. — {a) Megasclera; (l) Smooth, stout, usually slightly curved styli (PI. XXIX. fig. 4), tapering gradually to a sharp point at the apex aud narro-^ving towards the rounded base; size of full grown examples about 0'4 by 0"025 mm.; these form the main skeleton. (2) Much smaller, straight, slender styli or subtylostyli (PI. XXIX. fig. 4c), rounded and usually very faintly spined at the base and tapering gradually to a sharp point at the apex, size about 0"23 by 0"0065 mm.; these form the tufts at the ends of the primary fibres. (3) Small, spined styli (PI. XXIX. fig. 4a), almost or quite straight, very sharply pointed and with large, curved spines, shaped like the thorns on a briar. These spines are very unequally distributed, they are abundant at the base of the spicule, where they are arranged so as to point towards the apex, then they almost cease and at about the centre of the spicule become suddenly abundant again ; in this second position they are very large, stout and curved so as to point towards the base; size of spicule about 0'175 by 0"013 mm; their arrangement has already been indicated. (5) Microsclera ; of two kinds, (l) Very minute, palmate isochelse, about O'Ol mm. long ; abundant. (2) Toxa (PI. XXIX. fig. 4&), with spined ends, size about 0"14 by 0'0024 mm., also abundant. Vosmaer ^ briefly describes under the name Clathria lobata a new species from the Cape of Good Hope ; this sponge certainly has a very remarkable resemblance to the Challenger species under discussion, but the description is so short and imperfect that an absolute identification would be impossible without comparison of specimens ; till such is practicable we shall consider the Challenger sponge as a variety of Vosmaer's sjjecies, 1 Notes from the Leyden Museum, vol. ii. p. 151, 1880. EEPOET ON THE MONAXONIDA. 155 being strengthened in this opinion by the fact that both came from the Cape of Good Hope. The spiculation given by Vosmaer is as follows: — "Spic. tr. ac. f. (tr". ac. f.). tr. ac. sp. (strongly spined). acl (rare), tr^ A (sp.) ^^. anc"." He says also — "The sponge, which in the dried state is pure white, is rather elastic on account of the keratode-fibre. It forms more or less flat, branching lobes. Very characteristic for this species are the bows, the ends of which are spined . . . The anchors are small but rather stout, few in number. The spines of the tr. ac. sp. are very strong and bend towards the blunt end of the spicule. In the formula, I have designated with a --^ the small spicules which are often almost bent into a circle. I could not distinctly see whether the ends are blunt or pointed." We are pretty safe in eliminating the "acl (rare)" either as foreign or abnormal, while the spicules designated with a ^<-^ are almost certainly Diatom rings, which often occur abundantly as foreign bodies in sponge preparations ; the presence of a trl is perhaps a more serious difficulty; the large, smooth styli of the two sponges agree in being fusiform. It is the external appearance and the peculiar and very characteristic form of the sjjiued styli and toxa in both sponges which induce us to put them in the same species, at any rate as a temporary arrangement. It is very unfortunate that Vosmaer has given no spicular measurements and no account of the arrangement of the skeleton ; when these are known it will be much easier to arrive at a definite conclusion. The presence of the distinct tufts of slender stylote spicules at the ends of the primary fibres, forming a dermal crust, obliges us to place this species in the genus Rhaphidophlus rather than in the genus CJathria, though, as we have already had occasion to point out, the distinction is one of degree rather than of kind. Locality. — Simon's Bay, Cape of Good Hope; depth, 10 to 20 fathoms. One specimen. Genus Plumohalichondria, Carter (Pis. XXX., XLVIL). 1876. PlmnoluiKchondria, Carter, Ann. and Mag. Nat. Hist., ser. 4, vol. xviii. p. 236. The skeleton is arranged in plume-like columns (PI. XLVIL fig. 4a). Megasclera oxea and styli ; no special kind of dermal spicule. Microsclera isochelee. In his paper on the Classification of the Spongida^ Mr. Carter founds a group of the "Ectyonida," under the name " Plumohalichondrina," in the following passage: — " Group 2. PlumohalicJiondrina. Here there are two forms of axial spicules, viz. — 1, simple acuate, smooth or spined ; 2, more or less pointed or inflated at the ends, which are often microspined scantily or sparsely. Echinating spicule club-shaped and spined. Flesh-spicule for the most part that termed by Dr. Bowerbank 'angulate equianchorate' (that is, with bow-shaped shaft and alajform arms), sometimes accompanied by a 1 Ann. and Mag. Nat. Hist., ser. 4, vol. xv. p. 144. 156 THE VOYAGE OF H.M.S. CHALLENGER. biliamate or C-sliaped flesh-spicule, sometimes without any flesh-spicule at all. Forms massive, lobo-branched ; branches compressed, dichotomous, separate or anastomotic, flabellate proliferous." On page 195 of the same volume the author gives " Halichondria plumosa, Johnst. { = Hymeniacidon 2^lui^'>-0sa, Bowerbank) " as an example of this group. In voL xviii of the same periodical (1876, p. 236) we first meet with the generic name " Plumohalichondria " in the description of the species named by Mr. Carter " Plumo- halichondria microcionides" ; but we find no generic diagnosis. Under these circum- stances it has appeared to us advisable to retain the genus, which suppUes a great want in the classification of the Mouaxonida, and to give a generic diagnosis as above. Thus constituted, the genus appears to be fairly compact and well characterised; it includes a portion of Bowerbank's \&xj heterogeneous genus " Microciona " ^ and certain other forms. As examples of sponges which belong to the genus " Plumohalichondria," we may give Plumohalichondria microcionides, Carter, and Plumohalichondria mam- millata. Carter." Plumohalichondria mammillata, Carter (PI. XXX. figs. 4, 4a; PI. XLVII. figs. 4, 4«). 1885. Plumohalichondria mammillata, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. xvi. p. 355. Sponge (PI. XXX. fig. 4) massive, with erect, compressed and rounded lobes. Height 94 mm., breadth 137 mm. Colour in spirit pale, yellowish-grey. Texture compact but soft and spongy. Surface glabrous but rather uneven and lumpy. Dermal membrane distinct, much strengthened by the presence of a very well-developed dermal reticulation of spicules. Oscula scattered ; round openings, about 2 mm. in diameter, with their margins sometimes surrounded by a slight, projecting, membranous collar ; sometimes flush with the general surface of the sponge, but usually sunk a little below the surface. Pores very numerous, thickly scattered, rounded openings ; about 0"07 mm. in diameter. Skeleton. — (a) Dermal; a very regular and definite reticulation (PI. XLVII. fig. 4) of small, spined, stylote spicules ; the meshes of the network are roundedly polygonal and only about O'l mm. in diameter, and each side is of about one spicule's length ; the " fibre " (if we may be allowed to use a term which does not seem very applicable to the case in point) is composed of several spicules lying parallel side by side. (&) Main; composed of plumose columns of spiculo-fibre (PI. XLVII. fig. 4a, p.c.) running towards 1 Mon. Brit. Spong., vol. ii. p. 124. - The genus Plumohalichondria, Carter, must not be coufoundeJ with the genus " Aulospongus," Norman, which was founded (Ann. and Mag. Nat. Hist., ser. 5, vol. i. p. 267, footnote) for tlie reception of Bowerbank's " Ealiphipema tubulatum," a remarkable sponge from Ceylon, which has some resemblance to Plumohalichondria but has no chelate nor oxeote spicules and is probably a very different thing. REPORT ON THE MONAXONIDA. 157 the surface of the sponge ; no secondary skeleton lines appear to be developed, but there are a very great number of loosely scattered megasclera. The fibres themselves are composed each of a central axis of smooth oxeote spicules very abundantly echinated by spined styli. Spicules. — (a) Megasclera; (1) smooth, straight, somewhat fusiform, hastately pointed oxea, size about 0'2 by 0"005 mm. (2) Entirely spined, usually slightly curved, sharp pointed styli, which may attain a length of about 0'2 mm., but this is exceptional and they are usually much smaller, measuring about 0*1 by 0*005 mm. Although Carter {loG. cit.) describes two forms of spined stylote spicules, which he distinguishes according to their size, yet, as he himself states, both are echinating, and it appears to us probable that the smaller are simply young forms of the larger, (b) Microsclera ; of one kind, viz., tridentate isochelse, with moderately curved shaft, length up to about 0"02 mm. This is a very fine and well characterised species, and as it is as yet very little known (although possibly common in Southern Australia), and as no figures of it are extant, we have thought it advisable to give a somewhat detailed description of it in this place. Mr. Carter has very kindly sent us a small piece of his sponge for examination ; we find from it that as regards the form and size of the spicules and the arrangement of the main skeleton, the Challenger sponge agrees sufiiciently closely with that from Port Phillip Heads, but in the latter we have been able to discover no definite dermal reticulation such as we have described above, the dermal skeleton being an irregular reticulation of scattered oxeote and stylote spicules. It must, however, be borne in mind that we have only had a small piece to examine, and that this reticulation may possibly be present in other, and perhaps older, parts of the sponge, hence we have not thought fit to separate the two as distinct species. Xoca%.— Station 162, April 2, 1874 ; lat. 39' 10' 30" S., long, 146° 37' 0" E. ; Bass Strait; depth, 38 fathoms ; bottom, sand and shells. One fine specimen. Habitat. — Port Phillip Heads, Southern Australia (Carter); Bass Strait, (Challenger). Genus Plocamia, Schmidt (Pis. XXIX., XXXI. ). 1870. Plocamia, Schmidt, Spong. Atlant. Gebiet., p. 62. 1881. Dirrhopalum, Ridley, Journ. Linn. Soo. Lond. (Zool.), vol. xv. p. 476. Sponge of varying form, may be encrusting or erect and branched. Spicules. — (a) Megasclera ; dumb-bell-shaped or sausage-shaped (the characteristic spicules of the genus), and stylote. (6) Microsclera ; isochelae and (at any rate usually) toxa. For details regarding the history and characters of this genus the reader is referred to Mr. S. 0. Eidley's paper on the genus Dirrhojxdum {loc. cit., s^ipra). The name " Dirrhopalum" was substituted for "Plocamia" on insuflicient grounds, and must be abandoned in favour of the latter. The Challenger only obtained a single specimen of the genus, but this specimen is a very interesting one. 158 THE VOYAGE OF H.M.S. CHALLENGER. Piocamia coriacea, Bowerbank, sp., var. elegans, nov. (PI. XXIX. fig. 9 ; PI. XXXI. fig. 1). 1874. Isodidtja coriacea, Bowerbank, Mon. Brit. Spong., vol. iii. p. 228, pi. Ixxvi. figs. 7-12. 1881. Dirrhopalum coriaeeum, Ridley, Joiu'ii. Linn. Soc. Lond. (Zool.), vol. xv. p. 481 pi. sxix. figs. 3-7. 1886. Piocamia coriacea, var., Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 475. A single, small, dichotoinously branched specimen (PI. XXXI. fig. 1) of the form shown in the figure. Total height 38 mm. ; diameter of branches, which are cylindrical, about 2"5 mm. Colour in spirit pale yellow. Texture soft and spongy. Surface even and minutely hispid. Dermal membrane very thin, delicateand transparent. Oscula and pores unknown ; the sponge is apparently lipostomous. Skeleton. — A close but irregular Isodictyal reticulation of the characteristic sausage- shaped spicules of the genus ; amongst these, towards the surface, one distinguishes radiating lines of smooth stylote spicules, which terminate in free, projecting brushes, causing the hispidity of the surface. Spicules. — (a) Megasclera; of three very distinct kinds, (l) («) Fairly stout, slightly curved, smooth styli, with sharp points and evenly rounded bases (which are rarely or never spined, and difi"er from those of the type in this respect); above the base there is a slight constriction; these spicules vary much in size, averaging about 0"5 by 0"0126 mm.; they occur in the primary , radiating lines, from whose ends they freely project. (/S) Very much slenderer, sharply pointed, straight subtylostyli, measuring about 0'3 by 0-003 mm., abundant in tufts on the surface, scattered through the dermal membrane, &c. (2) Curved, stout, entirely spined, sausage-shaped spicules (PI. XXIX. fig. 9), with a slightly developed head at each end, which bears more numerous but perhaps smaller spines than the shaft; size about O'll by 0*0094 mm. This is the most abundant and characteristic spicule, it forms the Isodictyal reticulation above mentioned. (3) Entirely but slightly spined, slightly curved styli, narrowing towards the base, which bears more spines than the shaft; measuring about 0*18 by 0'0094 mm.; only fairly abundant, echinating the radiating, main filn-es near the surface. (6) Microsclera ; of two kinds. (1) Small, palmate isochelae, length about 0*019 mm., very abundant in the dermal membrane, but not so common in the deeper parts of the sponge ; (2) very numerous, slender toxa, with minutely spined ends, varying considerably in size, full grown examples being about O'l mm. long. Bowerbank's species was, in the first instance, very imperfectly described, but it will be seen by reference to Ridley's description and figures [loc. cit.) that in the arrangement of the skeleton and the form and proportions of the various spicules the Challenger variety comes very near to it. The main diff'erence in spicular measurements concerns the toxa, which seem to attain a greater size in Bowerbank's sponge than they do in ours. EEPORT ON THE MONAXONIDA. 159 The most striking difference, however, lies in the external form, and this is very remarkable, for the British specimen, instead of being of erect growth and definitely- branched, is merely a small crust, and according to Bowerbank " does not exceed one and a half line in thickness" {loc. cit., p. 229), while the colour in spirit (in the case of the Challenger specimen very pale) is stated to be very dark brown. Zoca%.— Station 75, July 2, 1873; lat. 38° 38' 0" N., long. 28° 28' 30" W.; off the Azores ; depth, 450 fathoms ; bottom, volcanic mud. One specimen. Genus Acarnus, Gray. 1867. Acarnus, Gray, Proc. Zool. Soc. Lond., p. 544. Megasclera styli and tylota (? the latter may be absent), and echinating " Grapnel- spicules " (cladotylota), consisting each of a straight shaft with expanded base and a stellate projecting end with recurved teeth ; these are the characteristic spicules of the genus. As microsclera there may be palmate isochelse and toxa.^ Acarnus ternatus, Eidley. 1884. Acar7ms ternatus, Ridley, Zool. CoU. H.M.S. "Alert," Brit. Mus., p. 453, pi. xlii.figs. h, V. With this species we identify a single small specimen growing on a stone. It is in very bad condition and contains a great deal of foreign matter, but the characteristic ternate grapnel- spicules are abundant in it. The proportions of the spicules differ some- what from those of the type, the main skeleton stylus being a good deal longer, but there are no sufiicient grounds for separating the two forms specifically. Locality. — Papiete Harbour, Tahiti ; depth, 20 fathoms. One specimen. Habitat— ^omhixy ? (Coll. Brit. Mus.) ; West Island and Prince of Wales Channel, Torres Strait; and Isle des Neufs, Amu-ante Islands (Ridley, "Alert"); Tahiti (Challenger). Genus Echinoclathria, Carter (Pis. XXIX., XXXI.). 1884. Echinoclathria, Carter, Proc. Acad. Nat. Sci. Philad., 1884, p. 204. External form various, sponge made up of a honeycomb-like mass of anastomosmg, flattened trabecular. Skeleton reticulate, horny, with or without spicules in the fibre. Megasclera smooth, either styli or tylota ; smooth echinating styli commonly present. Microsclera may be present in the form of palmate isochel^. Mr. Carter has given no diagnosis of the genus. Fortunately, however, in addition 1 For further details concerniDg the genus, vide Eidley, ZooL CoU. H.M.S. "Alert," Brit. Mus., 1884, p. 453. 160 THE VOYAGE OF H.M.S. CHALLENGER. to the descriptions published by him of several species, we have been able to refer to specimens in the British Museum named by Mr. Carter himself, and it is from these and from the good series brought home by the Challenger that the above diagnosis has been drawn up. These remarkable honeycombed sponges appear to be very abundant off the south coast of Australia, and the question of their specific differences and their relations to other genera is a very difficult and intricate one. They are probably reduced Clathriids, in which, from the strong development of horny fibre rendering them unnecessary, the spicules are gradually disappearing, or, in other words, they are Clathriids on their way towards becoming horny sponges without spicules. Carter has described no forms with microsclera, but we have shown (see under Echinoclathria favus and Echinoclathria carteri) that palmate isochel^, such as are found in typical Clathrise, may be present. That the genus is closely allied to Claihria there can be no doubt, but it may be distinguished from the latter by the absence of a distinct kind of spined echinating stylus from the spicular complement ; the fibre is, however, echinated by the smooth skeleton stylus or subtylostylus. In accepting Mr. Carter's generic name, Echinoclathria, we hesitated before venturing to give a definite generic diagnosis. The present one must be regarded as preliminary. Before a satisfactory diagnosis becomes practicable a more extensive study of the group is required. The peculiar honeycombed external appearance is certainly very character- istic, but can hardly be regarded as absolutely distinctive ; Clathria frondifera, Bower- bank, makes a near approach to it, and Carter includes in his genus Echinoclathria several forms which do not possess it.^ Echinoclathria favus, Carter (PI. XXXI. figs. 4, 5, 5«). 18S5. Echinoclathria f amis, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. xvi. p. 292. Sponge (PI. XXXI. figs. 4, 5) massive or branched, lobate or digitate, commonly growing over the shells of living Pectens. Honeycombed throughout ; consisting of a close reticulation of anastomosing and interwoven trabeculse, with round or oval meshes between; the meshes on the surface may be either closed in marginally or Mseandriniform. Size very variable, the largest specimen is about 162 mm. high and 75 mm. broad. Colour in spirit greyish-yellow. Texture of trabeculse tough, rather cartilaginous. Surface very minutely hispid. Dermal membrane thin, transparent, very rarely (PI. XXXI. fig. 5a, m) stretching across the meshes at the surface of the sponge. Oscula (PI. XXXI. fig. 5a, o) minute, scattered over the surfaces of the trabeculse. 1 Viz., Echinoclathria tenuis, Echinoclathria nodosa, Echinoclathria suhhispida, Echinoclathria gracilis (Ann. and Mag, Nat. Hist., ser. 5, vol. xvi. pp. 355, .356), but whether these really belong to the same genus as Echinoclathria favus, which is mentioned before any of them {ibid., p. 292) is another question, into which we cannot here enter. EEPORT ON THE MONAXONIDA. 161 Skeleton. — A reticulation of well developed horny fibre, cored and echinated by subtylostylote spicules of one kind only. Spicules. — (a) Megasclera ; (l) smooth, fusiform subtylostyli, sharply pointed and constricted before the head, size about O'l by 0*0048 mm., in the fibre and echinating it and scattered. (2) Very long, thin, hair-like spicules, in which it is extremely difficult to make out the nature of the ends, (1) cylindrical, subtylostylote sometimes ; length about 0'17 mm.; scattered, (b) Microsclera; at first we thought that this species was sharply marked off from EcMnoclathria carteri, nobis, by the absence of microsclera, but since then we have seen a very few, faintly discernible, hair-like isochelse, about 0'0126 mm. long, apparently on the verge of disappearance. This species we have been enabled to identify through our examination of Mr. Carter's dried types in the British Museum.^ His description (loc. cit.), seems to us to be misleading ; it is possible that his account of the spiculation was taken from some other specimen than the two which we have examined (viz., his numbers 554 and 208 bis). The species comes extremely close to our Ecliinoclafhria carteri, but for the sake of convenience we separate them ; they are distinguished by the fact that EcMnoclathria carteri possesses abundant, distinct isochelse,' and well-developed, long, slender subtylo- stylote spicules (scattered) ; the proportions of the spicules also diflfer in the two and the typical external forms of the two species are very distinct [cf. figs. 3, 4 and 5 on PI. XXXI. ). That the two species are connected by a series of intermediate forms there can be little doubt, but for the sake of convenience we at present keep them distinct. The Challenger obtained six specimens of Echinoclatlma favus, all of which are based upon Pecten shells ; the inhabitants of the shells were evidently living when the sponges were dredged. This is a very interesting fact, for possibly we have here a case of commensalism. It is difficult to understand how a Pecten can swim about (in the manner stated of species of this genus by our conchological authorities) with a sponge of the size represented in PI. XXXI. figs. 4, 5, attached to it, but the animal is still in the shell and in good condition; possibly the streams of water engendered by the sponge, bringing constant supplies of food, render motion on the part of the Pecten unnecessary. Xoca%.— Station 162, AprU 2, 1874; lat. 39° 10' 30" S., long. 146° 37' 0" K; off Moncoeur Island, Bass Strait; depth, 38 fathoms; bottom, sand and shells. Six specimens. 1 Mr. Carter, he. cit, says—" There are several specimens of it in the British Museum, mostly under four inches in their greatest diameter, of which Nos. 554 and 555, each registered 59. 10. 7. 106, may be mentioned as massive forms, and No. "208 6is" registered 37. 5. 13. 36. &c., as more or less digitate forms." 2 Mr. Carter mentions no isochelae in EcMnoclathria favus, neither have we been able to find them in his tlried types, but such delicate microsclera may be very readily overlooked in examining dried specimens. (zooL. CHALL. EXP. — PART Lix. — 1887.) Nnn 21 162 THE VOYAGE OF H.M.S. CHALLENGER. EcMnoclathria carteri, Ridley and Dendy (PL XXIX. figs. 12, 12a; PI. XXXI. figs. 3, 3a). 1886. EcMnoclathria Carteri, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 476. Sj)onge (PI. XXXI. fig. 3) cylindrical, ramose, each branch consisting of flat, ribbon- like trabeculse, anastomosing and interwoven so as to form a loose, honeycombed whole, with rounded meshes on the surface about 2 mm. in diameter. The largest .specimen is about 250 mm. long; diameter of branches about G to 15 mm. Colour in spirit pale yellow. Texture (of the individual trabeculse) tough and compact. Surface very minutely hispid. Dermal membrane thin, transparent, very rarely found stretching across the meshes on the surface. Oscula (PL XXXI. fig. 3a, o) minute, scattered over the trabeculse. Skeleton.— k. rather close reticulation of strongly developed horny fibre, cored and echinated by stylote spicules of one kind only, though the spicules within the fibre seem to be usually slenderer than those outside. Spicules. — (a) Megasclera ; (l) smooth, sharply and gradually pointed styli (PI. XXIX. fig. 12a), not markedly constricted above the base; measuring about 0"132 by 0'009 mm.; in the fibre, irregularly echinating the same, and scattered. (2) Long, smooth, very slender subtylostyli (PL XXIX. fig. 12), measuring about 0"16 by 0'002 mm.; irregularly scattered, (b) Microsclera; of one kind only, viz., smaU, palmate isochelse, very abundant in some specimens, about 0"015 mm. long. This is an exceedingly remarkable species. It might be thought that its peculiar external form would be quite sufiicient to distinguish it from all other sponges, but there is a digitate variety of Carter's Echinoclatliria favus (one of the types in the British Museum referred to by him,^ and labelled 37. 5. 13. 36. and 208 bis), also from South Australia, which very closely resembles it. As regards external appearance the two may, however, be distinguished by the fact that in EcMnoclathria carteri, nobis, the anasto- mosing trabeculse usually present a flat surface towards the outside, while in Echino- clathria favus, Carter, they usually have their edges turned outwards. On examination with the microscope the two are more readily distinguished, for Echinoclathria favus possesses few or no cliel^, and there are also other slight diS'erences in spiculation w'hich will be seen by reference to the descriptions. Localities.— ^tut'ion 162, April 2, 1874 ; lat. 39° 10' 30" S., long. 146° 37' 0" E.; Bass Strait ; depth, 38 fathoms ; bottom, sand and shells. One or two specimens. Station 163a, April 4, 1874 ; lat. 36° 59' S., long. 150° 20' E.; south-east Australia; depth, 120 fathoms ; bottom, green mud. Three or four specimens. OS" Port Jackson ; depth, 30 to 35 fathoms. One specimen. 1 Ann. and Mag. Nat. Hist., ser. 5, vol. svi. p. 292. REPORT ON THE MONAXONIDA. 163 Echinoclatliria glabra, Ridley and Dendy (PI. XXIX. figs. 11, 11a ; PI. XXXI. fig. 2). 1886. Echinoclatliria glabra, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii p. 476. Sponge (PI. XXXI. fig. 2) massive but honeycombed throughout, consisting throughout of thin, flattened trabeculse, anastomosing and interwoven with one another, leaving oval or rounded meshes about 4 mm. in diameter. The single specimen in the collection is about 112 mm. high by 75 mm. broad, and 44 mm. thick ; it has evidently been cut off at the base. Colour in spirit yellow. Texture firm and parchment-like. Surface (of the trabeculse) glabrous. Dermal membrane thin, transparent, very rarely stretching across the mouths of the cavities at the surface. Oscula not observed, apparently the sponge is lipostomous. Skeleton. — A reticulation of well-developed horny fibre, sparsely cored by tylota and sparsely echinated by smooth subtylostyli. Spicules. — Megasclera; (1) smooth, fusiform subtylostyli (PI. XXIX. fig. 11a), sharply and gradually pointed at the apex and constricted before the head, which is no wider than the shaft; size about O'll by 0"006.3 mm.; echinating the horny fibre. (2) Long, smooth tylota (PI. XXIX. fig. 11), very slender, with oval heads; measuring about 0*22 by 0'0032 mm., in the fibre and scattered, especially on the surface of the trabeculse. This species is distinguished by its glabrous surface and by the presence of the tylote spicules in the fibre and scattered. No microsclera are present. It is very probable that, owing to its spicular peculiarities, it ought to form the type of a new genus, but we prefer to leave the responsibility of this step to some one who has a larger series of forms for comparison than we have. Zoca%.— Station 162, April 2, 1874 ; lat. 39° 10' 30" S., long. 146° 37' 0" E.; off Moncoeur Island, Bass Strait; depth, 38 fathoms; bottom, sand and shells. One specimen. Genus Agelas, Duchassaing and Michelotti (PI. XXIX.). 1864. AgelaSj'Dvich. and Mich., Naturkund. Verhand. Maatschapp. te Haarlem, ser. 2, vol. xxL p. 76. 1867. Ecfyon, Gray, Proc. Zool. Soe. Lond., p. 515. 1870. Chalinopsis, Schmidt, Spong. Atlant. Gebiet., p. 59. Of various form ; with well developed horny fibre echinated by verticillately spined stylote spicules. No microsclera and no other megasclera. To Carter is due the merit of having pointed out ^ that Ectyon sparsus. Gray, is referable to the older genus Agelas of Duchassaing and Michelotti, nevertheless he retains Gray's generic name, Ectyon; in this we cannot, with due regard to the rules of priority, 1 Ann. and Mag. Nat. Hist, ser. 5, vol. is. p. 281. 164 THE VOYAGE OF H.M.S. CHALLENGER. follow him, since, in spite of the extreme faultiuess of the original descriptions, the genus is, according to Mr. Carter's own showing, recognisable. It is not known as a common type in any seas but those of the West Indies, whence Carter describes several species. The systematic position of this genus is very doubtful. We include it here only on the supposition that it has lost its microsclera. Agelas mauritianus, Carter sp. (PI. XXIX. fig. 10). 1883. Edyon mauritianus, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. xii. p. 310, pi. xii. fig. 3, a, h. This species, so well characterised by its beautiful, mouiliform, echinating spicule (PL XXIX. fig. 10), is represented in the collection by a single specimen of irregular massive form and curiously cavernous structure, but of doubtful locality ; as the species has already been described, and the characteristic spicule figured by Mr. Carter (and now again by us), we need give no further details. A parchment label inside the bottle bears the words, " Station 135 1 60 fms." Habitat. — Mauritius (Carter); off Tristan da Cunha (■?) (Challenger). Genus Echinodictyum, Ridley (PI. XXXII. ). 1881. Echinodictyum, Eidley, Joiirn, Linn. Soc. Lond. (Zool.), vol. xv. p. 493. Skeleton reticulate. Megasclera smooth oxea in the fibre (sometimes accompanied by partially projecting, smooth, slender styli) and spined styli echinating the fibre. No microsclera. The genus Echinodictyum was estabhshed by Eidley {he. cit) for certain species of various external habit, but agreeing in the possession of a reticulate skeleton consisting of smooth oxeote megasclera united together in fibres by spongin, the fibre being echinated by spined spicules projecting from it at right angles. Thus as regards spiculation it is distinguished from Raspailia only by virtue of its oxeote instead of stylote main mega- sclera ; the external form is, however, usually massive and the skeleton reticulate and not radiate as is the case in Rasi^ailia. Subsequently some species were found in which the fibre was accompanied by smooth slender styli sparingly associated with rather than inserted in it ; the spicular distinction thus ceasing to be an entirely absolute one. It is probable, however, that the Ectyonine arrangement, the relations of the fibre and its spicules, and the bulky growth indicate that it forms a natural genus, distinct from Raspailia with its slender cylindrical axes, radiate skeleton arrangement, and almost exclusively stylote spicules. We include this genus amongst the Ectyoninse, like the genus Agelas, on the supposition that it had at one time microsclera which it has now lost. REPORT ON THE MONAXONIDA. IdS JEchinodictyum rugosum, Ridley and Dendy (PI. XXXII. figs. 1, la). 1886. Echinodictijum rugosum, Ridley and Dendy, Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 477. Sponge (PL XXXII. fig. l) stipitate, palmato-digitate, consisting of a sliort cylindrical stem about 19 mm. long, surmounted by a broad, flattened expansion which terminates in a series of flattened, digitate processes. Total height of specimen 187 mm., greatest breadth 131 mm.; thickness only about 4 mm. (the stalk is a little thicker). Colour in spirit greyish-yellow. Texture hard and rather brittle. Surface rugose, thickly beset with small, pointed eminences. Oscula and pores unknown. Skeleton. — A well-developed, compact, but rather iiTegular reticulation of strong spiculo-fibre ; the fibre consisting of a multispicular axis of smooth oxeote spicules, firmly united together, and very abundantly echinated by spiued stylote spicules which project from it approximately at right angles. Spicules. — Megasclera; of two kinds. (1) Smooth oxea, somewhat hastately pointed and usually bent at an angle in the centre, size about 0"3 by 0'015 mm., in the skeleton fibre. (2) Entirely spined styli (subtylostyli), tapering gradually to a fine but not very sharp point and with the spines most abundant on the base; size about 0"13 by 0"012 mm., abundantly echinating the skeleton fibre. This species difi"ers from aU other described members of its genus in its palmate, slightly branched form ; Echinodictyum nervosum, Ridley,^ which also grows in one plane, being ramose from the stem upwards, and Echinodictyum cancellatum, Ridley,^ forming a can- cellate growth in a single plane. Locality.— Station 190, September 12, 1874; lat. 8° 56' S., long. 136' 5' E.; south- west of New Guinea; depth, 49 fathoms; bottom, green mud. One specimen. Echinodictyum asperum, Ridley and Dendy (PL XXXII. fig. 2). 1886. Echinodictijum asperum, Hidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 477. Externally this species (PL XXXII. fig. 2) has very much the appearance of a so-called Keratose sponge. It is bushy and of suberect growth, cavernous and covered with large aculeations ; most of the specimens are attached to fragments of coarse rock or Coral. Height commonly about 50 mm.; breadth about the same or a little more. Colour in spirit rich chocolate brown.^ Texture coarsely fibrous. Swface uneven in the extreme, but glabrous where the dermal membrane is intact. Dermal membrane thin and transparent, containing an enormous quantity of reddish brown pigment granules disposed in small groups. This pigment is still more abundant in the deeper parts of the sponge. 1 Journ. Linn. Soc. Land. (Zool.), vol. xv. p. 496, pi. xxviii. flgs. 7-10. 2 Zool. Coll. H.M.S. "Alert," Brit. Mus., 1884, p. 457, pi. xl. fig. D. ^ Evidently natural, as otlier sponges attached are not coloured like this. 166 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton. — Composed of a very coarse, widely open reticulation of stout, tough spiculo- fibre, brandling and anastomosing, and evidently containing a considerable quantity of horny cementing substance. This fibre is composed of longitudinally placed, very densely packed, slender oxeote spicules, firmly adherent to one another; and is echinated from the surface by spined stylote spicules which project at right angles to the fibre. The fibre itself is commonly about 0"5 mm. in thickness. Spicules. — Megasclera; (1) smooth, slender, slightly curved, gradually and sharply pointed oxea, measuring about 0-35 by 0'0063 mm., forming by far the greater portion of the skeleton. (2) Straight, slender, tapering, bluntly pointed, entirely spined styli ; spines small, abundant near the rounded apex, directed towards the base ; size of spicule about 0"17 by 0'0075 mm.; echinating the skeleton fibre. In its external form and in its wiry, bristle-like fibre this species difiers very widely indeed from the foregoing, though in spiculation the two come very near to one another. Locality. — Papiete Harbour, Tahiti ; depth, 20 fathoms. Seven or eight specimens. Family IV. Axinellid^. Skeleton typically non-reticulate; consisting of ascending axes of fibres from which arise subsidiary fibres radiating to the surface. Fibres typically plumose. Megasclera chiefly styli to which oxea and (or) strongyla may be added. Microsclera rarely present, never chelae. Genus Hymeniacidon, Bowerbank (Pis. XXXIIL, XXXIX., XL., XLV.). 1864. Hymeniacidon, Bowerbank, Mon. Brit. Spong., vol. i. p. 191. Form massive. Skeleton reticulate, composed of ill-defined spiculo-fibre, not plumose. Megasclera all monactinal, styli or subtylostyli. No microsclera. Bowerbank's original diagnosis {loc. cit.) runs — "Skeleton without fibre, spicula without order, imbedded in irregularly disposed membranous structure." The genus certainly is a difficult one to diagnose but we must have something more definite than this to go by. Bowerbank's diagnosis includes far too much for any natural genus and its author makes no less than forty-two species of Hymeniacidon out of a total of one hundred and ninety -three species of British sponges.^ The original type of the genus is Hymeniacidon caruncula^ and, taking this as our starting point, we have endeavoured to construct a more compact genus with the diagnosis as given above. The systematic position of the genus must still be regarded as very doubtful. Having *Mon. Brit. Spong., vol. ii. p. xx. ^ yi^ig M.qti. Brit. Spong., vol. i. p. 191. REPORT ON THE MONAXONIDA. 167 regard to its spiculation we have placed it amongst the Axinellidse, although the arrangement of the skeleton is opposed to this view. Such a skeleton is, however, quite capable of giving rise by further development to a typical Axinellid skeleton, and it is possible, judging from its extreme simplicity, that Hyineniacidon represents the starting point from which the more typical Axinellidas have branched off. Were it not for the form of the spicules, which are monactinal instead of diactinal, we should include the genus amongst the Eenierinaa. It differs from most Axinellidi^ in having the spicules of one form only and all of about the same size. The genus is characteristicall}^ an inhabitant of shallow water. Hymeniacidon caruncula, Bowerbank (PI. XXXIII. fig. 3). 1864. Hymeniacidon caruncula, Bowerbank, Mon. Brit. Spong., vol. i. p. 191, pi. xxxv. fig. 372 ; vol. ii. p. 166; vol. iii. pi. xxxii. figs. 1-4; vol. iv. p. 81. 1867. Reniera caruncula. Gray, Proc. Zool. Soc. Lond., p. 518. 1870. Amorpliina caruncula, Schmidt, Spoiig. Atlaut. Gebiet., p. 76. There are in the collection two specimens of this common British sponge from the Cape Verde Islands ; both are rather small and with uneven (conulose) surface (PI. XXXIII. fig. 3) ; the proportion of horny matter in the skeleton is very con- siderable. As the species is already sufficiently well known we do not propose to give any further description of it in this place. Locality. — St. Vincent, Cape Verde Islands ; shallow water. Two specimens. Habitat. — British Isles ^ (Bowerbank); Cape Verde Islands (Challenger); Port Jackson (Ridley, "Alert"). Hyineniacidon sp. We have to record from Station 313 a single, soft, massive specimen, in bad condition, which seems to be referable to this genus but is not sufficiently well characterised to merit a specific name. The surface, where preserved, is smooth, and the skeleton consists of very sparsely disposed columns of smooth stylote spicules united together by a considerable proportion of horny matter ; the same spicules also occur loosely scattered. Spicides. — Of one form only, viz., smooth, gradually sharp pointed styli, measuring about 0-43 to O'S by 0-0126 mm. Locality.— ^taXion 313, January 20, 1876 ; lat. 52° 20' S., long. 67° 39' W.; south- east of Patagonia; depth, 55 fathoms; bottom, sand; bottom temperature, 47°'8. One specimen. 1 " One of the regular tide-mark sponges of our southern and western coasts," Norman (Mou. Brit. Spong., vol. iv. p. 81). 168 THE VOYAGE OF H.M.S. CHALLENGER. Hymeniacidoyi (?) hyalina, n. sp. (PL XLV. figs. 6, Q>a, 66). Sponge massive, amorphous, encrusted by numerous foreign organisms ; represented in the collection by two or three small, shapeless fragments, remarkable for their transparent, hyaline appearance ; the largest one only about 19 mm. in diameter. Colour in spirit, very pale, yellowish grey, transparent. Texture soft and spongy, internally cavernous. Surface, for the most part encrusted by foreign organisms, where visible fairly smooth. Dermal membrane very distinct, hyaline. Oscula (? rather large and scattered). Skeleton. — (a) Dermal; consisting of tylostylote spicules thickly and quite irregularly strewn through the dermal membrane, (h) Main; diffuse and irregular in the extreme, with only very slight indications of fibres running vertically towards the surface ; composed of loosely and irregularly disposed tylostylote spicules of various sizes. Spicules. — Megasclera; of one form only, viz., smooth, nearly straight tylostyli (PI. XLV. figs. 6, Qa, 66), wdth distinct but not very large, subglobular heads ; sharply and fairly gradually pointed at the apex. Size very variable, but not differing in such a manner as to give rise to outside smaller and inside larger spicules as in the genus Suherites ; the different sizes are throughout mixed up with one another ; the maximum size is about I'l by 0"025 mm. It seemed very doubtful whether it was worth while to describe a new species on such frftgmentary evidence, but it is interesting to come across a form whose spiculation is almost exactly that of the genus Suherites, while the other characters, e.g., the arrange- ment of the skeleton and the nature of the ectosome, are totally different. We doubtfully place the species in the genus Hymeniacidon because we do not know what else to do with it. Locality. — Off the south-west coast of Patagonia. Two or three fragments. Hymeniacidon (?) suhacerata, Eidley and Dendy (PI. XXXIX. fig. 4 ; PI. XL. figs. 5, 5 a). 1886. Hymeniacidon {?) suhaceratu, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 478. Sponge (PI. XXXIX. fig. 4) massive; consisting of irregularly anastomosing trabeculfe, which may be subcylindrical, or angular, or more or less flattened and expanded. The largest specimen, taken as a whole, is about 88 mm. long by 62 mm. broad and 37 mm. thick. Colour in spirit pale yellow ; the sponge is characterised also by a peculiar, waxy, translucent appearance. Texture rather brittle and cavernous. Surface glabrous in appearance, but rather harsh to the touch and very uneven. Oscula of fair size, mostly near the summit of the sponge, having their margins flush with the general surface. REPORT ON THE MONAXONIDA. 169 Skeleton. — The main skeleton consists of a dense but quite irregular reticulation of large stylote spicules, the spicules occasionally lying side by side in loose tracts. At the surface there is a thin crust of similar large styli, also densely and very irregularly reticulate and giving support to numei'ous small stylote or subtylostylote sjaicules, which project more or less vertically outwards ; we have never found these small projecting spicules forming a very dense crust, though in some parts they are very abundant ; it is doubtful whether the layer of large styli on which they rest should be regarded as anything more than the uppermost part of the main skeleton. Spicules. — Megasclera; (l) very large, stout, smooth, fusiform styli (PL XL. figs. 5, 5a), more or less curved or slightly crooked ; tapering fairly gradually to a sharp point at the apex and tapering rather suddenly near the base to a very narrow neck ; the base is evenly rounded off at the extremity, and is often so narrow that the spicule becomes nearly oxeote in appearance, but we have never found the base pointed, always rounded off; size about 1'2 by 0"031 mm. (2) Small, slender, nearly or quite straight, gradually and sharply pointed styli or subtylostyli ; not markedly narrowing towards the base ; size commonly about 0"2 by 0"0063 mm. The mode of occurrence of these two forms of spicules has been noted above. This species would probably have been included by Mr. Carter in his genus Leucophlasus, but we are unable to discern any tangible and constant character wherebj' the latter genus may be separated from Hymeniacidon. One of the most characteristic features of our present species is the shape of the large stylote spicule, which is very peculiar ; it demonstrates that a stylus might readily give rise to an oxeote spicule. The narrowing towards the base of the spicule may also be seen, though not carried to the same extent, in Leucophlceus fenestratus, Ridley,' especially in one of the varieties. The small surface spicule appears to occur in Leuco- jMceus fenestratus also (though not described), but is much stouter there. In Leuco- phloeus proteus,^ Ridley, it appears not to occur, and the large spicule has an ordinary broad, rounded base. The external form of Hymeniacidon (?) subacerata is quite different from that of Leucophlceus fenestratus. Locality.— Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E.; Philippine Islands; depth, 18 fathoms; bottom, blue mud. Three specimens. Genus Phakellia, Bowerbank (Pis. XXXIV., XXXV., XXXVL, XL.). 1864. Phakellia, Bowerbank, Mon. Brit. Spong., vol. i. p. 186, &c. Sponge more or less flabellate or cup-shaped. Skeleton often more or less reticulate. Megasclera styli and often oxea. No microsclera. 1 Zool. Coll. H.M.S. "Alert," Brit. Mus., 1884, p. 4G4, pi. xlii. fii;. s. 2 Op. cit., p. 620, pi. liii. fig. B ; pi. liv. fig. k. (zOOL. CHALL. E.XP. — PART LIX. — 1887.) Nim 22 170 THE VOYAGE OF H.M.S. CHALLENGER. Bowerbank's original diagnosis is, as usual, based entirely on the arrangement of the skeleton. It is unnecessary to quote it at length {vide loc. cit.). The type of the genus is Phakellia ventilahi-um, Johnston, sp. We have thought it desirable in this case to make use of external form as a eeneric character, otherwise we know of no character which would serve to separate the genus Phakellia from the genus Axinella. Phakellia ventilahrurn, Johnston, sp., var. connexiva, nov. (PI. XXXV. figs. 3, 3a). 1842. Halicliondria ventUabrum, Johnston, British Sponges, p. 107. ^ 1864. Phakellia ventUabrum, Bowerbank, Men. Brit. Spong., vol. i. p. 186 ; voL ii. p. 122, vol. iii. pi. xxii. figs. 1-7. 1866. PhaA-ellia rohusta, Bowerbank, Mon. Brit. Spong., vol. ii. p. 120; vol. iii. p. 54. pi. xxi. figs. 13-17. There are from Station 122 a number of fine, froudose specimens which we propose to consider as a variety of the above species. In external form they are distinguished from the tyi^es of the species by their greater size, altogether more robust growth and strong- venation. The specimens are in the form of broad, flattened, erect lamella (PI. XXXV. fig. 3), reaching as much as 200 mm. in height, with very strong, projecting veins caused by the presence of stout bands of spiculo-fibre ramifying and anastomosing from the base upwards. The surface is strongly hispid and in spirit the specimens are of a brownish-yellow colour, in part due to the presence of a considerable amount of sand adhering to them. The spiculation consists of smooth styli and stout "vermicular" spicules, the former measuring about 1-5 to TS by 0'034 to 0-04 mm., and the latter about 0-65 to 072 by 0"035 mm. In spiculation the species Phakellia ventUabrum is practically identical with Axinella erecta, but is separated from the latter by the very different external form ; we also agree with Schmidt '^ in considering Bowerbank's Phakellia rohusta merel}- as a variety of Phakellia ventilahrurn} Localities.— ^t&tion 122c, September 10, 1873; lat. 9" 10' S., long. 34° 49' W.; east of Brazil ; depth, 400 fathoms ; bottom, red mud. A number of large pieces. Station 317, February 8, 1876 ; lat. 48° 37' S., long. 55° 17' W.; north-east of the Falkland Islands ; depth, 1035 fathoms ; bottom, hard ground (gravel) ; bottom temperature, 35° '7. One piece, of much more delicate appearance than the foregoing ; almost white, and translucent, a condition which is, however, largely due to maceration, probably owing to the time occupied in raising the specimen from such a depth. Hahitat. — British seas (Bowerbank, Johnston, &c.) ; Ireland (Johnston); Shetlands (Bowerbank) ; between Scotland and Faroe Islands (Carter) ; south-west coast of ' See also Johnston, loc. cit., for list of synonyms previous to his time. - Spong. Atlant. Gebiet., p. 62. ' Details as to the minute anatomy of Phakellia ventilahrum, var. connexiva, will be found in the Introduction. REPORT ON THE MONAXONIDA. 171 Norway (Schmidt) ; Arctic Ocean, off Norway (Vosmaer) ; Baltic Sea (Fristedt) ; ? Gulf of St. Lawrence (Whiteaves) ; Florida, Gulf of Mexico and Barbados (Schmidt) ; off Brazil and north-east of Falkland Islands (Challenger). The species is typically an inhabitant of deep water, being common in depths over 100 fathoms, seldom occurring in shallower water, and going dowai to 1035 fathoms, as shown by the Challenger dredgings. Phakellia jiahellata, Ridley and Dendy (PI. XXXIV. figs. 2, 3, 3a; PI. XL. figs. 6, 6o). 1886. Phakellia fla J lellata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. .5, vol xviii. p. 478. Sponge (PI. XXXIV. figs. 2, 3) erect, stipitate ; stem rather short, stoutish, cylindri- cal, ending below in a widish, flattened base of attachment, and expanding rather suddenly above into a broad, flattened, vertical lamella with undulating margin. One side of the lameUa is thrown into more or less pronounced, rather irregular, longitudinal furrows and ridges, while the other is comparatively smooth and bears numerous stellately disposed oscula (PL XXXIV. fig. 3a). The finest specimen in the collection is 125 mm. in total height and the same in breadth ; the stem is 25 mm. long and 8'5 mm. in greatest diameter (being not quite round). The lamelliform portion is only about 3 mm. thick (except along the prominent ridges). Colour in spirit greyish-yellow. Texture rather soft and brittle. Dermal membrane thin and transparent. Pores very abundant on the surface which bears no oscula, in small groups over the ends of narrow inhalent canals, where they reduce the dermal membrane to a mere network ; the pores themselves are round or oval openings about 0'07 to O'l mm. in diameter; on the opposite side of the sponge they are scarce or absent. Oscula (PI. XXXIV. fig. 3a) numerous, on one side only of the sponge, in round, stellate groups of eight or ten ; the oscula themselves are minute, round and about 0'5 mm. in diameter; the groups are fairly regularly placed at about equal distances, and are about 4 mm. (or a little less) in diameter. Skeleton. — From the point where the stem enters the sponge stout, branching bands of spiculo-fibre, containing a very large proportion of dark-coloured, horny cementing material, radiate upwards through the soft tissues. In addition to this fibrous skeleton there is a rather irregular, somewhat Halichondrioid reticulation of stout stylote spicules, terminating at the surface in divergent brushes of much smaller, slender styli. Each Inrush surrounds the apex of one of the larger styli, and the apices of its own spicules project for a short distance beyond the surface of the sponge. Spicules. — Megasclera; (l) stout, smooth, slightly curved, sharply and rather abruptly pointed styli (PI. XL. fig. 6); measuring about 0"5 by 0'03 mm.; forming the main skeleton, occurring both in the fibre and in the reticulation outside of it. (2) Much smaller, straight, smooth, slender styli (or subtylostyli) (PI. XL. fig. 6a), sharply and 172 THE VOYAGE OF H.M.S. CHALLENGER. rather abruptly (hastately) pointed, measuring about 0"22 by 0'0063 mm., occurring in the surface brushes. This is a very pretty and well-characterised species, which may be readily recognised by its external form and more especially by the arrangement of the oscula. From the number obtained by the Challenger it would seem to be abundant near Port Jackson. Judging from the description and from a preparation of that species, it would appear to resemble Spongia caliciformis, Lamarck, most nearly of described forms, Ijut that species, as yet only known from the North Sea, is cup-shaped. Locality. — Port Jackson ; depth, 30 to 35 fathoms. Eleven specimens. Phakellia papyracea, Ridley and Dendy (PI. XXXVI. fig. 4). 1886. Phaliellia papyracea, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 478. Sponge (PL XXXVI. fig. 4) very thin, lamelliform. Eepresented in the collection by fragments only, ^jossibly cup-shaped when alive ; perforated all over by very numerous, minute, round (1 oscular) openings. Alike on both surfaces. Thickness of lamellae about 2 mm. Colour in spirit yellow or brown. Texture fragile. Surface fairly even, very minutely hispid. Oscula (?) very abundant on both surfaces, minute. Skeleton. — Reticulate, rather vague and indefinite ; distinct fibres present but very loose, composed of large stylote spicules, with spicules of a smaller size scattered between, or also in loose fibres. There is no special dermal skeleton, l)ut the small stylote spicules are very much more abundant at the surface than elsewhere, especially around the oscidar (?) openings. Spicules. — Mcgasclera; of two kinds; (1) large, stout, smooth styli, rarely subtylo- stylote, usually more or less curved and faii-ly gradually sharp-pointed, size about 07 by 0"02 mm. (2) Much smaller and slenderer styli or subtylostyli, commonly rather crooked; sharply and fairly gradually pointed; size variable, say about 0'35 by 0'0063 mm. but often much smaller. This is a very delicate species, which perhaps comes near to Bowerbank's Isodictya itifmidibidifonnis,^ more especially if it should ultimately prove to be cup-shaped when perfect, but it is distinguished at once and absolutely from that species by the alisence of the oxeote spicules, so that further comparisons are needless. In the absence of the oxeote spicules, however, it agrees with von Marenzeller's Cribrochalina ambigna,^ but diff"ers widely in the size of the spicules, which, in Cribrochalina ambigua " sind Stifte von 0"23 bis 0"34 mm. Liinge," while there do not seem to be two distinct sizes as in our sponge. Localities.— Station 145a, December 27, 1873; lat. 46° 41' S., long. 38° 10' E.; oflp ' Jlon. Brit. Spong., vol. ii. p. 317 ; vol. iii. pi. liv., &c. - Poriferen, Anthozoen, Ctenophoren iind Wiirmer, von Jan Mayen ; Vienna, 1886, p. 1, pi. i. fig. 1. REPORT ON THE MONAXONIDA. 173 Prince Edward Island; depth, 310 fathoms; bottom, volcanic sand. Fairly abundant fragments, of a brown colour. Station 148, Januar}^ 3, 1874; lat. 46° 47' S., long. 51° 37' E.; Possession Island ; depth, 210 fathoms; bottom, hard ground, gravel, shells. Two or three small, yellow pieces. Genus Ciocalypta, Bowerbank (Pis. XXXIII. , XL.). 1864. Ciocalypta, Bowerbank, Mon. Brit. Spong., vol. i. p. 179. Megasclera stylote and sometimes oxeote, forming a more or less dense central skeleton (fibrous or reticulate) from which are given off pillars of spiculo-fibre nearly at right angles. These spread out at their distal ends and support the dermal membrane, with its reticulation of spiculo-fibre, at some distance from the central portion of the sponge, thus leaving huge subdermal spaces into which the water enters through the pores in the dermal membrane. No microsclera. Bowerbank's original diagnosis {loc. cit.) runs as follows : — " Skeleton. Composed of numerous closed columns, each consisting of a central axis of compact, irregularly elongated, reticulated structure, from the surface of which radiate, at about right angles, numerous short simple, cylindrical pedicles, or stout fasciculi of closely packed spicula ; the distal ends of each pedestal separating and radiating in numerous curved lines which spread over the inner surface of the dermal membrane, separating and sustaining it at all parts at a considerable distance from the central axis of the skeleton." In this diagnosis there is not a word as to the form of the spicules. The original type of the genus is Ciocalypta penicillus, Bowerbank, in which the spicules are all stylote. In Ciocalypta tylcri,^ Bowerbank, the spicules are all oxeote, but it seems very doubtful whether this species is referable to the genus Ciocalypta^ at all. Bowerbank's Ciocalypta led,- represented by a single, minute, dry specimen, about 13 mm. in length and only about 3 mm. in greatest duimeter, must be regarded as a young specimen of Ciocalypta penicillux, or at the most as a mere variety, and the same remark applies to the specimen which has been called by Mr. Carter Ciocalypta tuherculata {vide p. 175). There are thus (excluding the doubtful Ciocalypta tyleri) three species of the genus now known, viz., Ciocalypta penicillus, Bowerbank, Ciocalypta hyaloderma, nobis, and CiocalyjJta amoiphosa, nobis ; for -Ciocalypta calva,^ Eidley, is wanting in one of the chief characters of the genus, viz., the characteristic arrangement of the dermal skeleton ; and it also possesses a horny fibre. The genus differs from Axinella most markedly in the possession of a distinct, spiculo- fibrous dermal reticulation, and further in the very large size of the subdermal cavities. • I'roc. Zool. Sue. Land., 1873, p. 21, pi. iv. figs. 9-12. 2 Mon. Brit. Spong., vol. iii. p. 296, pi. Ixxxvi. figs. 1-4. 3 Zool. Coll. H.M.S. "Alert," Proc. Zool. Soc. Load., 1881, p. 115, pi. x. fig. 7. 174 THE VOYAGE OF H.M.S. CHALLENGER. Ciocalypta hyaloderma, Ridley and Deudy (PL XXXIII. figs. 2, 2a). 1886. Ciocalypta hyaloderma, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 479. Sponge (PI. XXXIII. fig. 2) cylindrical, or subcylindiical ; ramose. The largest piece in the collection consists of a very short, stout, basal portion (evidently broken off Ijelow), from which two short branches diverge at a very wide angle ; distance from apex of one branch to that of the other 40 mm. ; diameter of branches from about 8 to 13 mm. A second fragment is cylindrical, measuring 44 mm. in length by 10 mm. in diameter ; it gives off a single branch near the middle, of which the stump only remains. Colour in spirit dirty brown. Texture very delicate and fragile. Surface fairly even. Dermal membrane (PL XXXIII. fig. 2a) very delicate, absolutely transparent ; supported at some distance from the underlying tissues (which can be distinctly seen through it) on slender pillars ; marked -ndth numerous little white stars, caused by the spicules which compose the supporting pillars radiating as they abut against it and forming the principal nodes in a well-developed dermal reticulation of spiculo-fibre. Oscula (PL XXXIII. fig. 2a) scattered over the branches, of fair size, each surrounded by a prominent mem- branous collar; a1)out 2 mm. in diameter. Pores, round openings in the dermal membrane, about 0'05 mm. in diameter ; rather difficult to make out on account of the extreme transparency of the dermal membrane even when examined in spirit. Skeleton. — Consisting, in the solid central portion, of a loose reticulation of bands of spiculo-fibre and detached spicules, with a very marked preponderance of longitudinal fibres. Short pillars of spiculo-fibre, as already stated, connect the central portion of the sponge with the dermal membrane ; the spicules at the distal ends of these pillars expand into divergent Inrushes which form each a " node " in the dermal reticulation. The dermal reticulation is very well developed and very wide meshed, consisting of usually stout bands of spiculo-fibre, radiating, roughly speaking, from the distal ends of the supporting pillars, and branching and anastomosing with one another. Spicules. — Megasclera; of one form only ; viz., smooth, usually slightly curved, fairl}' gradually but not very sharply pointed styli, commonly narrowing slightly towards the base; these are variable in size, in the dermal reticulation they commonly measure about 0"53 by 0'014 mm., while in the central portion of the sponge they measure about TO by 0-037 mm. This species certainly comes very near to the original type of the genus, viz., Ciocalypta penicillus, Bowerbank ; but is distinguished by its different growth and much more delicate structure throughout ; the central axis is not so dense as in that species, and the dermal reticulation is not so confused and has much wider meshes.^ The spicules also attain a considerably larger size than in Ciocalypta ijenicillus. ' In this respect Bowerbank's Ciocalypta led, which is no douht a very young specimen of Ciomlyfta pcnicilluti, resembles the present species. REPORT ON THE MONAXONIDA. 175 There is in the British Museum a single dried specimen ' which in its delicate appearance comes nearer to the present species, but still ought prol)al:>ly to be regarded as a variety of Ciocalypta penicillus ; indeed a careful examination of the types has convinced us that it is so. Mr. Carter' refers to it as follows : — " In the British Museum, among the sj^ecimens dredged up on Ijoard the ' Noma ' on the coast of Portugal, is a sponge of a similar conical form, also grouped, but with a tuberculated surface, each tubercle of which is supported on a bundle of spicules that radiate from a solid, conical, central axis. Here, however, there is only one kind of spicule, viz., acuate, smooth, and sharp pointed ; so that it does not belong to the Ectyonida, liut, belonging to the Axinellida, might be called ' Ciocalypta (Bk.) tiiberculata,' seeing that, like other species of this group al:)out to be mentioned, it will probably have to come under the order Echinonemata." The tuberculated appearance here noticed is caused merely by the sinking in of the dermal membrane between the supporting columns on drying, and is quite as prominent a feature in the dried specimens of Ciocalypta p)enicillus. Locality.— Station 320, February 14, 1876 ; lat 37° 17' S., long. 53° 52' W.; off the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°'2. Four pieces. Ciocalypta amorphosa, Ridley and Dendy (PI. XL. fig. 9). 1886. Ciocalypta amorphosa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 479. Sponge massive, amorphous ; forming shapeless, rounded masses, the largest of which is about 75 mm. long by 50 mm. broad and 38 mm. thick. Colour in spirit grey. Texture very spongy, fibrous, honeycombed ; resembling that of a common bath sponge which has been prepared for use, l^ut not nearly so tough. Surface, in the present condition of the sponge extremely rough and shaggy, beset with numerous fibrous tufts which, in the perfect condition, sup^^ort a very delicate dermal membrane continuously stretched over them at some little distance from the main mass of the sponge. Dermal membrane very thin, delicate and transparent ; only portions of it remain, supported on the outspread ends of the fibrous tufts above mentioned. Oscula (?). Pores extremely numerous, round or oval openings, scattered through the dermal membrane, variable in size. Skeleton. — The main skeleton is a very loose, confused, vaguely fibrous reticulation of very long oxeote and stylote spicules; this skeletal arrangement obtains in the trabeculse of tissue between the numerous wide canals of the sponge. At the surface are given off from the main mass loosely fibrous columns of spicules which spread out at their distal ' Registered 72.5.4.1. - Ann. and May. Nat. Hid., ser. 4, vol. xviii. p. 235. 176 THE VOYAGE OF H.M.S. CHALLENGER. extremities and give support to the dermal membrane. The dermal memljrane is strengthened by a very irregular "dermal reticulation" composed of loose spiculo-fibre. Sjncules. — Megasdera (PL XL. fig. 9); (1) very long, slender, slightly curved, fusi- form oxea; sharply and rather suddenly pointed; size variable, up to about 1"7 by 0"02 mm. (2) Large, usually slightly curved styli, with evenly rounded base ; shorter and stouter than the oxea; size variable, up to about 1'47 by 0'028 mm. The styli are plentiful, Ijut not so common as the oxea ; occasionally also a stout strongylote spicule is seen, evenly rounded off at each end ; these are, at any rate commonly, shorter than either oxea or styli. The full-grown spicule appears to gain in thickness what it loses in length, and this is probably the case in many species, as held by Carter.^ Unfortunately all the specimens of this sponge, which seems to be abundant at Station 320, are much frayed out and injured. It is an interesting species which might easily be mistaken for a Eenierine, but the variety in form of the spicules and the arrangement of the skeleton show its Axinellid relationships. It is at once distinguished both from Ciocaly23ta peniciUns, Bowerbank, and Ciocalypta hyalodevma, nobis, by its external form and also by its spiculation. The manner in which the dermal membrane is supported at a distance from the body of the sponge, on the outspread ends of columns of spiculo-fibre, necessitates the reference of this species to the genus Ciocalyjita. Locality.— Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; off the mouth of the Eio de la Plata; bottom, green sand; bottom temperature, 37°"2. Five or six specimens. Genus Acanthella, Schmidt (PL XXXII.). 1862. Acanthella, Schmidt, Spongien d. Adriat. Meeres, p. 64. 1870. (?) Pandaros, Duchassaing de Fonbressin and Michelotti, Spong. Mer Caraibe, p. 88. Axinellidse of ramose, bushy, or frondose external form ; of cartilaginous consistency, and with glabrous surface beset with ridges and spines ; there is no distinct horny fibre ; smooth linear megasclera {e.g. , styli, strongyla and unequal-ended oxea) are present. No microsclera. Schmidt's diagnosis runs " Halichondrise ramosae et fruticosaj, tanquam spinis obsitas. Cutis Isevis, porosissima, qu« in ramis crassioribus sola pigmento infecta est et verte pellis iustar a parenchymate distinguitur. Parenchyma spisse implctum spiculis simplicibus longioribus, substantia firmiori non inclusis." The genus appears to us to be a good one, although ([ueried by Vosmaer/ and sufficiently distinct from other genera of Axinellidse to deserve separation ; the original t}^e laAcanthella acuta, Schmidt. The cartilaginous consistency is a marked characteristic of the genus, and is evidently due to the peculiar character of the mesoderm. ' Ann. and Mag. Nat. Hist, ser. 5, vul. iii. p. 284. * Broiin's Klass. u. Ordii. il. Thierreichs, Porifera, p. 344. REPOKT ON THE MONAXONIDA. 177 The spiculation is distinguished by its plasticity, the different forms of linear spicules appearing to run into each other, and to have little constancy within the genus ; species are, however, separable by the forms and proportions of their spicules. An elongated flexuous form is fairly constant. Acanthella pulcherrima, Ridley and Dendy (PL XXXII. fig. 3). 1884. Acanthella sp., Eidley, Zool. CoU. H.M.S. "Alert," Brit. Mus., 1884, p. 463. 1886. Acanthellax>ulchernma, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. .5, vol. xviii. p. 479. This species was obtained by the "Alert" and described by Eidley {loc. cit.) without a name and vv^ith no figures. We have decided to quote the original description and to add a figure of the very characteristic external form (PI. XXXII. fig. 3). "Externally resembling Spongia carduus, Lamarck (Ann. Mus. Hist. Nat. xx. p. 381). When guided by the description alone, I had referred the present specimen to this species with more confidence than usual ; but on mounting sections of the probable type specimen at Paris, I saw that it was a different species. The points in which the description does not cjuite suit this form are "pedicule cylindrace, tres-dur," the stem having apparently been flattish, and, though stiff, not inflexible ; and " couleur d'un blanc grisatre," whereas this (in spirit) is flesh-colour. The ridges run longitudinally up and down the sponge, and are 1 to 3 millims. high, and their free edge is beset with sharp (in spirit flexible) points at intervals of one or two millimetres. Texture tough and flexible, substance compact, surface between inequalities glabi'ous. It is a true Acanthella. The spiculation is as follows : — (l) Smooth acuate, slightly curved, tapering gradually to a sharp point, about "4 to '6 millim. by -0095 niillim. (2) Smooth undula- ting cylindrical with rounded ends, length about "7 millim., diameter just '0063 millim. " The species differs from the Adriatic forms, A. acuta and ohtusa, Schmidt, in the broad explanate form and in the smaller size of the spicules, the cylindrical being much shorter and thinner, the acuate much shorter than in those species. The skeleton forms a loose-fibred Axinella-like network of spicules, imbedded in a dense, transparent, almost colourless mass of caoutchouc-like consistency, containing nucleoid bodies about "007 to "008 millim. in diameter. ''■ Hab. Prince of Wales Channel, Torres Straits, 7 fms. A single specimen in spirit, 35 millim. (1| inch) high by 29 millim. across." As regards spiculation we must further add that an unequal-ended oxeote spicule is also common, of about the same size as the stylus. This is a remarkably pretty sponge. The Challenger obtained one specimen (PI. XXXII. fig. 3), measuring about 56 mm. in height and 46 mm. in breadth ; it is based upon an agglomeration of Polyzoa, fragments of barnacles, &c., and is a very much finer example than that obtained by the "Alert." Locality. — Cape York, Torres Strait; depth, 3 to 11 fathoms. Habitat. — Torres Strait (" Aleit," Challenger). (zool. chall. EXT. — PART Lix. — 1887.) Nnn 23 178 THE VOYAGE OF H.M.S. CHALLENGER. Acanthella (?) stipitata, Carter, var. n. 188L Acanthella stipitata, Carter, Ann. and Mag. Nat. Hist., ser. 5, vol. vii. p. 380, pi. xviii. fig. 8. This sponge is represented in the collection by a slight variety from Torres Strait. It differs from the type of the species as described by Mr. Carter (loc. cit.) — (l) in having the stem obsolete and represented only by a constricted base ; (2) in having the stylote megasclera larger, measuring about 0"5 by 0'009 mm. as against 0"347 by 0"009 mm. (25 by | — 1800th inch, as given by Mr. Carter). In addition to the large stylus there is in the Challenger specimen a long, very slender spicule which is extremely abundant and often flexuous, and is very hkely a young form of the other. ]\Ir. Carter further remarks that the styli in his sponge are " abruptly sharp-pointed ;" we cannot say this of our specimen, in which the styli are very gradually sharp-pointed. Mr. Carter's figure of the spicule, however, does not confirm his description. In habit this variety closely approaches Clathria frondifera, which is abundant at the same locality. We are by no means certain that the species is correctly placed in the genus Acanthella ; it seems to us to have close affinities with Echinoclathria, Carter. Locality.— ^tPition 186, September 8, 1874; lat. 10° 30' S., long. 142° 18' E.; off Cape York, Torres Strait ; depth, 8 fathoms ; bottom. Coral mud. One specimen. Habitat. — Bass Strait (Carter) ; Torres Strait (Challenger). Genus Axinella, Schmidt (Pis. XXV., XXXIII., XXXIV., XXXV., XXXVI., XXXVII., XXXVIIL, XL.). 1862. Axinella, Schmidt, Die Spongien des Adriat. Meeres, p. 60. Sponge typically ramose, but may be massive. Skeleton fibre plumose. Megasclera stylote and sometimes oxeote. No microsclera. This is a very critical genus, and it is impossible to give a satisfactory diagnosis of it. It comes very near to Raspailia, but the latter is conveniently kept distinct on account of its very characteristic, whip-like external form. Axinella arhorescens, Ridley and Dendy (PL XXXV. figs. 4, 4a). 1886. Axinella arhorescens, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. .5, vol. xviii. p. 479. Sponge (PL XXXV. fig. 4) erect, branched, stem and branches cylindrical or sometimes more or less flattened. Branches often forming anastomoses. Total height of sponge 212 mm.; length of stem up to first branch 50 mm., diameter of same 15 mm.; diameter of branches about 8 mm. Colour in spirit greyish -yellow. Texture firm REPORT ON THE MONAXONIDA. 179 but rather woolly. Surface even and fairly smooth where intact. Dermal membrane thin and transparent, mostly abraded. Oscula minute, arranged along the branches in small groups, which often have radiating cracks proceeding from them, and thus acquire a stellate appearance. Pores scattered. Skeleton. — Fibre not very distinct ; several main fibres run up the centre of each branch, not very distinct from one another, forming a kind of ill-defined central axis from which primary fibres radiate at more or less acute angles towards the surface, where they terminate in dense brushes of stylote spicules. The fibre itself, though not very weU defined, is clearly arranged on the Axinellid plan, i.e., it consists of stylote spicules whose bases lie together in the centre of the fibre while their apices project outwards and upwards at acute angles. The primary fibres are everywhere linked together by individual spicules, which commonly lie at right angles to them in the horizontal plane. In the centre of the branch there seems to be a fair development of horny cementing substance. Spicules. — Megasclera; these are, almost exclusively, short, stout, slightly curved, rather abruptly pointed styli, very constant in size, measuring about 0"28 by 0'024 mm. Occasionally also one meets with a strongylote spicule of about the same size, l^ut these are of rare occurrence and should probably be regarded merely as abnormal forms of the styli. This is a fine, showy species, with a very characteristic external form. The spiculation is characterised by its simplicity, and by the unusual constancy in size of the styli ; the arrangement of the skeleton, however, in columns radiating from a loose central axis, seems to mark the true position of the species as a member of the genus Axinella. Locality. — Port Jackson ; depth, 30 to 35 fathoms. One specimen. Axinella balfourensis, Ridley and Dendy (PI. XXXIII. fig. 1). 1886. AxineUa balfourensis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 480. Sponge (PI. XXXIII. fig. 1) erect, stipitate, consisting of a spreading, much branched root surmounted by a long, cylincbical stem, which at a height of 112 mm. above the root divides into two branches, each branch again dichotomising ; altogether giving rise to a large head of long, thick, dichotomously branching, finger-like processes, the different In-anches often anastomosing with one another, and terminating in tapering points. Total height of specimen (including root) 350 mm.; diameter of stem a little over 6 mm. Diameter of finger-like processes in head 8 mm. or a little more. Colour in spii-it yellowish-grey. Texture of stem firm and compact ; of finger-like processes in head ex- ceedingly soft and spongy, fragile. Surface very slightly hispid, fairly smooth. Dermal 180 THE VOYAGE OF H.M.S. CHALLENGER. membrane thin and very delicate. Oscrda rather small, scattered up and down the soft, finger-like processes. Pores (? thickly scattered over the finger-like processes). Skeleton. — Loose separate bands of spiculo-fibre run up the centre of each branch and form an ill-defined axis, from which radiate other loose, thin fibres towards the surface, crossed irregularly and sparingly by separate spicules. At the surface are numerous loose, radiating brushes of stylote spicules, whose points project slightly lieyond the dermal membrane. The same arrangement holds good, with very slight modification, in the stem. Spicules. — Mcgasclera ; very slender, smooth, straight or very slightly curved styli, .sharply and gradually pointed at the apex ; size about 0'42 by 0"0075 mm.; of the same shape both in the dermal tufts and in the main skeleton, but in the former of only about half the size. The entire skeleton seems singularly insufficient for the requirements of so large a sponge, and, as a consequence, the sponge is very flexible, soft and fragile. This seems to be a very aberrant species of the genus, as indicated both by its external form and by the extreme sparseness of the skeleton. Locality. — Balfour Bay, Kerguelen ; depth, 20 to 60 fathoms. One specimen. Axinella mariana, Eidley and Dendy (PL XXXIV. fig. 1 ; PL XL. fig. 2). 1886. Axinella mariana, Ridley and Deudy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 480. Sponge (PL XXXIV. fig. 1) of delicate growth, erect, proliferously branched; stem short, rather slender ; branches rather slender, somewhat flattened. Total height 52 mm., breadth about the same. Length of stem 13 mm., diameter 3 mm. Colour iu spirit greyish-yeUow. Texture very soft and friable externally, internally pretty tough. Surface of branches echinated by abundant, very long, projecting spicules. Skeleton. — In the centre of each branch there is a fairly dense core of irregularly arranged, short, bent, stylote spicules. In this confused mass are imbedded the bases of very large, stout styli, whose apices project far beyond the surface of the sponge. Spicxdes. — Megasclera; of tw^o kinds — (l) short, smooth (rarely slightly spincd) styli or subtylostyli, each with a single sharp bend near the base (PL XL. fig. 2, &), finely and gradually pointed at the apex ; size about 0"3 by 0"013 mm. (Imt variable). (2) Very long, smooth styli or tylostyli (PI. XL. fig. 2, a), usually with a slight bend towards the base, very gradually and finely pointed at the apex; size about 2*2 by 0"03 mm.; echinat- ing the surface of the sponge and with their own bases echinated by the smaller styli. This is a pretty little species, distinguished by its external form and by the peculinr shape of the smaller stylote spicule, which seems to be homologous with the " vermi- cular " spicule of Axinella erecta, &c. Locality. — Off Marion Island ; depth, 50 to 75 fathoms. One specimen. REPORT ON THE MONAXONIDA. 181 Axinella profunda, Ridley and Dendy (PL XXXVIII. figs. 2, 3 ; PL XL. figs. 3, 3a). 1886. Axinella profunda, Ridley and Dendy, Ann. and Mag. Nat. Hist., er. 5, vol xviii. p. 480. Sponge (PL XXXVIIL fig. 2) smaH, erect, stipitate, dichotomously branched ; stem and branclies slightly flattened in one and the same plane ; stem expanding below into a small, woody base. Height of specimen 50 mm.; breadth of stem and branches about 4 mm. Colour in spirit yellowish-grey. Texture; there is a tough, woody axis surrounded by a soft, spongy coat. Thin, membranous threads extend and form connections between adjacent branches ; possibly we have here the first steps on the road to complete anastomosis such as not infrequently occurs between adjacent branches of ramose sponges. Surface fairly even where the dermal membrane is intact, hispid. Dermal membrane thin and transparent. Oscula of moderate size, scattered. Skeleton. — There is a dense central axis of more or less longitudinally placed, large stylote spicules, from which similar spicules radiate to the surface in tracts or brushes, projecting beyond it and surrounded by bunches of smaller, slender styli whose apices also project beyond the surface. Spicules. — Megasclera; of one form only, viz., styli (PL XL. figs. 3, 3a), straight, or nearly so, and sharply pointed; ranging in size from about 0'55 by 0'0084 to about 2"0 by 0"037 mm.; the bases of the spicules, with few exceptions, are very minutely spined (PL XL. fig. 3a). This description is taken from one specimen only (that from Station 241). A second (PL XXXVIIL fig. 3), met with since it was written, confirms it in a very satisfactory manner indeed, even down to the spination of the bases of the styli ; but it is of a dark, reddish-brown colour. The most characteristic feature of the species is, perhaps, the incipient sijination of the bases of the stylote spicules ; this, though very minute, is a perfectly definite feature ; it is most noticeable in the smaller spicules. The species is especially interesting on account of the very great depth and the widely distant localities from which it was obtained. It would appear to be a widely spread species in abyssal regions of the Pacific, maintaining its peculiar specific characters with singular constancy in very diff'erent localities. Localities.— Station 241, June 23, 1875; lat. 35° 41' N., long. 157° 42' K; North Pacific Ocean; depth, 2300 fathoms; bottom, red clay; bottom temperature, 35°-l. One specimen. Station 281, October 6, 1875; lat. 22° 21' S., long. 150° 17' W.; South Pacific Ocean; depth, 2385 fathoms; bottom, red clay; bottom temperature, 34°"9. One specimen. 182 THE VOYAGE OF H.M.S. CHALLENGER. Axinella erecta. Carter, sp. (PL XXXV. figs. 1, 2, 2a, 2h, 2c; PI. XXXVI. fig. 2 ; PL XL. figs. 1, la). 18G6. (?) Hijineraphia vermictdata, Bk. {^ part<), Mon. Brit. Spong., vol. i. pi. i. fig. 5; vol. ii. p. 1-11 ; vol. iii. pi. xxvi. figs. 1-3. 1867. (?) Buharis vermiculata, Gray, Proc. Zool. See. Lond., p. 522. 1876. Ihjmeraphia vcrmindata, var. erecta, Carter (pars), Ann. and Mag. Nat. Hist., ser. 4, vol. xviii. p. 307, pi. xii. fig. 4; pi. xv. fig. 26, a, h. This very variable species is rei^resented iu the collection by a large number of specimens from divers localities. The specimens from Inaccessible Island (PL XXXVI. fig. 2) are by far the finest ; erect and massively ramose or subramose, the largest being 100 mm. in height and up to 41 mm. in breadth ; their spicules ai'e a good deal smaller than those of the smaller specimens from other localities (PL XXXV. figs. 2, 2a, 2b, 2c), but the size of the spicules is extremely variable in the same sponge and seems also to vary a good deal in difi"erent individuals of this species. AU the specimens belong to Mr. Carter's var. erecta (loc. cit., supra), which we here consider as the type of the species. The long stylote spicule (PL XL. fig. 1) measures up to nearly 3 mm. in length, with a breadth of about 0"037 mm.; while the "vermicular" (strongylote) spicule (PL XL. fig. lo) measures about 0'2l by 0"037 mm. (these measurements are taken from one of the small specimens from Station 148 a); the "vermicular" spicule is here considerably larger than in Mr. Carter's types, which approach more nearly in spiculation to our specimens from Inaccessible Island. The specimens described by Bowerbank under the name Hymeraphia vermiculata are mere patches of sponge, which might perhaps be young forms of other species as well as of Axinella erecta. Schmidt's Axinella mastophora ^ approaches very near to this species ; it is, however, distinguished by its external form (which may not, perhaps, count for much), consisting of a number of erect, smooth lobes, and by the fact that the large stylote spicule is very con- stantly constricted towards the base. Axinella rugosa, Schmidt," also comes very near to this species in its spiculation and in the possession of a surface covered with rounded rugosities, but is distinguished by the presence of very numerous, minute, angulated oxea. Localities.— Station 145a, December 27, 1873; lat. 46° 41' S., long. 38° 10' E.; off" Marion Island; depth, 310 fathoms ; bottom, volcanic sand. One specimen, oi Clathria- like external appearance, subfrondose and of a deep brown colour. Station 147, December 30, 1873; lat. 46° 16' S., long. 48° 27' E.; west of Crozet Island; depth, 1600 fathoms; bottom, Diatom ooze; bottom temperature, 34°'2. Two specimens. Station 14Sa, January 3, 1874; lat. 46° 53' S., long. 51° 52' E.; oS" Crozet Island; depth, 550 fathoms; bottom, hard ground, gravel, shells. Twenty-six specimens. Inaccessible Island, October 1873 ; depth, 90 fathoms. Four specimens. 1 Spong. Atlant. Guljitt., p. CI, pi. iv. fig. 14. - Loc. cit. REPORT ON THE MONAXONIDA. I S3 Off Tristan da Cunha, October 18, 1873 ; depth, 100 fathoms. One specimen. Off Nightingale Island, Tristan da Cunha; depth, 100 to 150 fathoms. Two specimens, and also a third, which, though it differs rather strikingly in external form from the rest, we have decided to include in this species pending further discoveries. The specimen in question is erect, clathrous, and proliferously branched ; the branches are very short and compressed (PI. XXXV. fig. 1). Habitat. — ? British Isles (Bowerbank); Atlantic Ocean, between the North of Scotland, the Shetland and the Faroe Islands (Carter) ; South Atlantic and Southern Ocean (Challenger). Axinella echidnxa, Ridley, var. n. (PI. XXXVI. fig. 3). 1813. (?) Spongia echidnxa, Lamarck, Ann. Mus. Hist. Nat., vol. xx. p. 448. 1884. Axinella echidnsM,'R\dL\(ij, Zool. Coll. H.M.S. " Alert," Brit. Mus., p. 462, pL xliii. He;, a. There occurs in the collection a single specimen (PL XXXVI. fig. 3), from off Bahia, which we propose to consider as a variety of the above-mentioned species. The various points in which it deviates from the type of the species may be briefly enumerated as follows: — (l) External appearance; the type specimens have, for the most part, a broadly flabellate form, while the Challenger specimen, consisting, however, of only a single branch, is (like one of the types) cylindrical, and its surface is beset with numerous irregular prominences ; these prominences are much larger, broader and further apart than in the tyj^e. (2) Spiculation ; the spicules are, as in the type, smooth oxea and styli, the former being much more abundant than the latter. In the present variety the oxeote spicule measures about 0'5 by 0'022 mm. and the stylus about the same. For the type, however, the spiculation is given as follows: — " (l) Smooth, slightly curved acerate, tapering gradually to sharp points, or more or less blunted at one or both ends; size "3 by "0095 to "44 by '0127 millim. : these forms compose the main bulk of the skeleton. (2) Long smooth acuate, generally slightly curved, tapering gradually to a fine point; size about 1"1 by '0127 millim.: forming part of longitudinal skeleton-lines of surface -tufts." Styli (acuates) as long as those here mentioned appear, however, to be very rare in the type, and considering the well-known irregularity of the Axinellid spiculation in general we feel quite justified in not separating the two forms specifically. Locality. — Off Bahia, shallow water. One specimen. Habitat.— Torres Strait ("Alert"); off Bahia (Challenger). Axinella Jibrosa, Ridley and Dendy (PI. XXXVII. fig. 3). 1886. Axinella jihrosat Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 481. Sponge (PI. XXXVII. 6g. 3) erect, massive, dividing into many lobes at the sum- mit. Height of specimen 162 mm.; breadth 87 mm. Colour in spirit greyish-yellow. 184 THE VOYAGE OF H.M.S. CHALLENGER. Texture soft and spongy, coarsely fibrous ; the fibres becoming readily denuded of the soft tissues so as to project freely beyond the surface. Surface, where intact, subglabrous but conulose, with a minutely reticulate appearance. Dermal memhrane peeling off fairly easily from the underlying tissues, not supported by any special skeleton, hence readily tearing. Oscula small and scattered. Pores scattered through the dermal membrane. Skeleton. — There is no dermal skeleton, but the main skeleton is very strongly developed. It consists of a system of stout fibres, branching and occasionally anasto- mosing, and coming to the surface in tufts. The fibre itself is very characteristic, consisting in the first place of a core of stylote spicules whose bases are, as a rule, in the centre, and whose apices project more or less outwards and forwards; being thus arranged in a fairly typical Axinellid manner. This spicular core is almost completely ensheathed in spongin, beyond which the apices of only a few spicules project. At the surface each branch of the fibre ends in a small, brush-like expansion. The average diameter of the fibre is about 0-4 mm.; it is, of course, slenderer at the surface than in the deeper parts of the sponge. Spicules. — Mcgasclera; of one form only, viz., smooth styli, slightly bent towards the base and usually very gradually and sharply pointed at the apex ; size about 0-63 by 0015 mm. This species is peculiarly interesting on account of the unusually strong development of spongin in the skeleton-fibre. It afibrds an additional good instance of the remarkable fact that horny fibre may be developed in any group of sponges and is hence of very little importance for purposes of classification. The massive form and the putty-like appear- ance of the surface between the fibres are further points distinguishing the species from its allies. Locality.— ^tatiow 313, January 20, 1876; lat. 52' 20' S., long. 67° 39' W.; east of Strait of Magellan; dej^th, 55 fathoms; bottom, sand; bottom temperature, 47°"8. One specimen. Axinella reticulata, Eidley and Dendy (PI. XXXVII. figs. 4, 4a). 1886. Axinella reticulata, Ridley and Dendy, Ann. and Mag. JSfat. Hist., ser. 5, vol. xviii. p. 481. Sponge (PI. XXXVII. fig. 4) massive, sessile, with somewhat constricted base below, and several short, thick-waUed oscular tubes above. Height 38 mm.; greatest breadth 44 mm.; breadth at base 29 mm. Colour in spirit very pale yellow. Texture firm, almost incompressible. Surface very uneven, beset with numerous small conuli which extend over the oscular tubes, but glabrous. Dermal membrane distinct, minutely reticulate (PI. XXXVII. fig. 4a), but almost or quite without spicules. Pores in the meshes of the dermal network; several in each mesh, reducing it to a secondary network. Oscula, one at the summit of each oscular tube, about 4 mm. in diameter. REPORT ON THE MONAXONIDA. 185 Skeleton. — There is an extremely irregular reticulation of scattered spicules, amongst which one can distinguish rather loose, plumose fibres, running more or less vertically towards the surface. Spicules.— Megasclera ; (l) rather short and stout, smooth, slightly curved styli ; usually with a bend towards the base ; sharply and fairly gradually pointed at the apex ; size about 0"45 by 0'02 mm. (2) Smooth, curved oxea, sharply and fairly gradually pointed ; of about the same size as the styli ; scarce. The most characteristic feature of this sponge is the glabrous, reticulate dermal membrane ; the reticulate a^^pearance is not caused by the presence of a reticulate dermal skeleton but by a network of thicker, faintly fibrous-looking tissue in the dermal membrane itself The species is further distinguished within its genus by its large, prominent oscula. Locality. — Bahia ; depth, 7 to 20 fathoms. One specimen. Axinella monticularis, Eidley and Dendy (PI. XXXVIII. fig. 5). 1886. Axinella monticularis^ Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 481. Sponge (PL XXXVIII. fig. 5) massive, subglobular, sessile, but apparently free. The largest specimen measures 44 by 31 by 21 mm. Colour in spirit yellowish -grey. Texture finn and compact, very gritty owing to the presence of a large quantity of calcareous foreign matter {e.g., Foraminifera) within the sponge. Some of the foreign bodies are large and others small, but all become completely enveloped by the sponge. Surface thickly covered with very numerous, sharp, prominent but small conuli, averaging about 1'5 mm. in height, which give to the specimens a very characteristic appearance. Over and between the conuli the surface is glabrous, yet it is in parts very minutely hispid. Skeleton. — There is no dermal reticulation. The numerous foreign bodies in the sponge are echinated by the spicules and many of them also form an origin for stout columns of typical Axinellid spiculo-fibre. The skeleton columns are well developed and run vertically to the surface, ending in the conuli above mentioned. They are composed of stylote spicules of various sizes, some smooth and some (of the smaller ones) spined ; the spicules have their bases towards the centre of the fibre and their apices projecting freely outwards and forwards in a plume-like fashion. Normally all the spicules appear to have some definite base, either in the fibre, where each one is supported by the rest, or on some foreign body ; but a few may be observed loosely scattered between the fibres. Spicules. — Megasclera; styU, of two distinct kinds. (1) Smooth, usually slightly 1 lent towards the base, very gradually and sharply pointed (tapering from base to apex) ; (ZOOL. CHALL EXP. — PART LIS. — 1887.) Nnn 24 186 THE VOYAGE OF H.M.S. CHALLENGER. size about 0"6 by 0"0126 mm. (2) Smaller, entirely spined styli (usually subtylostylote), commonly a little bent towards the base and gradually sharp pointed ; length commonly about O'l mm. but may reach up to 0*25 mm., thickness about 0*0066 mm. This is probably a common shallow-water species in the Cape Verde Islands. It is remarkable on account of the high development of the Axinellid fibre and of the numerous foreign bodies which the sponge contains. The latter character is probably due to the nature of the bottom on which it lives. It is further distinguished by the presence of the entirely spined styli. Localities. — St. Vincent, Cape Verde Islands ; shallow water. Three specimens. Harbour, St. Vincent, July 1873 ; depth, 7 to 20 fathoms. One specimen. Axinella {?) lunsecharta,^ Eidley and Dendy (PI. XXXVII. figs. 1, la, 2). 1886. Axinella (?) lunxcharta, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 481. V Sponge (PI. XXXVII. . figs. 1, 2) massive, sessile, subglobular. Size of largest specimen about 38 by 31 by 25 mm., with a deep concavity on the lower surface. A second specimen is about 21 mm. in diameter and attached to a small pebble. Colour in spirit very pale yellow. Texture fairly firm, but compressible and rather spongy. Surface uneven but glabrous, with numerous small, monticular eminences, amongst which occur fewer but much larger eminences, volcano-like, each with a distinct crateriform depression at the top, which is about 2 "5 mm. in diameter. (The degree of development of these crateriform eminences differs in different specimens.) Pores C?). Oscula minute, in the crateriform depressions (PL XXXVII. fig. la). Skeleton. — There is no special dermal skeleton ; the main skeleton is loosely reticulate, consisting of primary lines of loose spiculo-fibre running vertically to the surface and crossed by still looser and vaguer secondaries composed for the most part of single spicules. The spicules in the main fibres are all, or nearly aU, directed towards the surface of the sponge, but the Axinellid character of the fibre is barely recognisable. Spicules. — Megasclera; (l) smooth styli, more or less bent towards the base and very gradually and sharply pointed; size about 0"4 by 0"014 mm. (2) Smooth oxea, usually gradually sharp pointed at both ends, but very commonly with one end larger and less sharply pointed than the other; size about 0"3.5 by 0'0126 mm. It is very common in Axinellid sponges to find a few oxea along with the styli, but in this species the oxea are abundant, though not so much so as the styli. The not uncommon tendency which they exhibit towards blunting at one end seems to indicate that one form may be derived from the other. 1 So called from the resemblance which the surface of the sponge bears to a map of the moon, owing to the numerous crateriform eminences. REPORT ON THE MONAXONIDA. 187 The species may be readily recognised by its very characteristic external appearance. Localities. — St. Vincent, Cape Verde Islands ; shallow water. Two specimens. Harbour, St. Vincent, July 1873 ; depth, 7 to 30 fathoms. One specimen. Axinella (f) tubulosa, Eidley and Dendy (PI. XXXVIII. fig. 4). 1886. Axinella (?) tubulosa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. x\dii. p. 482. Sponge (PL XXXVIII. fig. 4) erect, tubular ; tubes either open at the top or closed, finger-like, the latter probably young forms of the former. There is only one small specimen in the collection, consisting of one broad, tubular process, 17 mm. wide, and widely open at the summit, and one narrow, closed, digitate tube 6 to 8 mm. wide ; both arising from a common basal portion. Height of specimen 50 mm. Colour in spirit greyish-yellow. Texture fairly firm. Surface uneven, pitted and rather cor- rugated. Oscula small, numerous on the inner surface of the wide, open tube. Skeleton. — Extremely vague and confused, with no distinct fibre, consisting of a loose, irregular reticulation of large, stout, stylote spicules, with small slender styli in loose brushes near the surface and also scattered about. Around the mouth of the wide tube is a narrow, delicate, membranous collar, supported solely by the slender stylote spicules. Spicules. — Megasclera; smooth styli. (l) Large and stout, more or less bent, fairly gradually sharp pointed ; size about 0"87 by 0'03 mm., thickest in the centre. (2) Similar but much shorter and slenderer styli, size about 0'45 by 0'009 mm. The most interesting feature about this species is its tubular form; a character, which, so far as is known to us, is hitherto unparalleled amongst the Axinellida3, except in Schmidt's " Auletta sycinularia;"^ that sponge is, however, regularly ramose instead of merely lobose, like the present species. Locality.— ^tuiion 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W.; oft" the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°'2. One specimen. Axinella (?) paradoxa, Ridley and Dendy (PL XXV. fig. 4). 1886. Axinella {^) paradoxa, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 482. Sponge (PL XXV. fig. 4) sessile, massively lobate ; height of single specimen present about 25 mm. Colour in spirit pale, greyish-yellow. Texture india-rubber-like, with a firm, gelatinous ectosome ; internally fibrous. Surface glabrous, but conulose. The conuli are projections caused by the vertically ascending skeleton columns ; hence they occur chiefly on the summits of the lobes, while the sides are rather ridged or grooved. 1 Spong. Atlant. Gebiet., p. 45, pi. iv. fig. 5. This sponge Ls evidently an A.xiuellid, as is shown by Sclimidl'.i figures of the spicules. 188 THE VOYAGE OF H.M.S. CHALLENGER. Oscula small but very evident, in groups on the tops of the lobes ; diameter of each about 1*5 mm. Skeleton. — There is no dermal skeleton present. The main skeleton is composed of large oxeote spicules wliich, for the most part, form very stout but very loose Axinella-Wkti fibres. The fibres all run more or less vertically upwards, but radiate somewhat from the centre of the base. Numerous spicules also occur irregularly scattered between the fibres, so that the whole skeleton becomes confused. Spicules. — Megasclera; of one kind only ; viz., large, smooth oxea, almost always gradually and finely pointed at both ends, and slightly curved ; size about 0*87 by 0-022 mm. The systematic position of this sponge is very difficult to determine. The form of the spicules would seem to indicate that it belongs to the Eenierinse, but the arrangement of the skeleton places it amongst the Axiuellidse. Probably it belongs rather to the latter group than to the former, and has simply lost the styli which form the most important part of the skeleton in most Axinellids {cf. Axinella (?) lunwcharta, p. 186). The peculiar gelatinous or india-rubber-like ectosome marks it off from most hitherto known species of either group, though it is more nearly paralleled in the Axinellidse than in the Eenierinse {cf. Axinella, Acanthella). This unusual condition is due to an enormous development of gelatinous mesodermal tissues, containing large stellate cells. Locality. — Inaccessible Island; depth, 90 fathoms. One specimen. Genus Ras2Ktilia,l>i urdo (Pis. XXXIX., XL.). 1833. Raspelia, Nardo, Isis, p. 522. 1847. Raspailia, Nardo, Prospetto della fauna marina volgare del veneto estuario, Venice, p. 3. Sponge long and slender, with a dense central axis of spiculo-fibre containing much spongin, from which loose tufts of spicules radiate to the surface. Megasclera stylote or subtylostylote, sometimes strongylote. Spined echinating styli sometimes present. No microsclera. The most characteristic feature of this genus is the external form, taken in connection with the absence of microsclera ; like Axinella, it is a difficult genus to diagnose, and the two genera, as we have already indicated, come very near to one another. Raspailia tenuis, Ridley and Dendy (PI. XXXIX. figs. 2, 2« ; PI. XL. figs, 8, 8«, 86). 1886. Raspailia tenuis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5. vol. xviii. p. 482. Sponge (PI. XXXIX. fig. 2) of more or less erect growth (probably the branches floated in the water), consisting of a very long, slender stem,^ giving ofi" at first ' This description is taken from the larger of the two specimens, the smaller bears only a single branch. REPORT ON THE MONAXONIDA. 189 very long, slender branches, but produced into a terminal portion which bears no branches, and which itself extends further than any of them and is indistinguishable from them excepting in being a direct continuation of the main stem. Length of sponge (=that of the main stem) 475 mm.; diameter of stem and branches about 2 to 2*5 mm. The branches are long, flexuous and string-like, and do not bear secondary branches. Colour in spii-it greyish-yellow. Texture tough, string-like, with a coating of soft, friable consistence. Surface roughened with minute monticular eminences like the surface of a file ; hispid. Oscula minute, scattered. Skeleton. — There is a slender axial core of dense, amber-coloured horny matter containing only a few slender spicules ; this core merges into a thick cylinder composed of obliquely (sublongitudinally) arranged, closely placed spicules, which, when seen in longitudinal section, present an irregular lattice-like arrangement, crossing one another at acute angles. Transverse sections of the stem or branches show that there is also an arrangement of stouter stylote spicules radiating from the central horny axis to the circumference, beyond which they project like the spokes of a wheel. The skeleton is much more strongly developed in the lower part of the stem than in the branches, and a transverse section of this region bears under the microscope a most striking resemblance to the flat bottom of a circular basket, the spicules representing the twigs of which it is composed. In the branches the sheath of spicules surrounding the horny axis is much thinner, and is covered with a thick coating of brownish-yellow, granular choanosome containing the canal system and flagellated chaml^ers. We have already mentioned the large, radiating stylote spicules which project beyond the surface of the sponge ; each of these as it leaves the sponge is suiTounded by a faggot-like bunch of very small, slender styU, which also have their apices projecting outwards. Spicules. — Megasclera ; (l) very long, smooth, slender styli (PI. XL. fig. 8); straight, or nearly so ; usually, but not invariably, tapering very gradually to a very fine point at the apex, size up to about 1'75 by 0"018 mm. (2) Smooth, slender strongyla (PI. XL. fig. Sb), measuring up to about 0"77 by O'Ol mm. (3) Very slender, smooth styli, surrounding the large ones as above mentioned, size about 0'42 by 0"0035 mm. (4) Small, spined styli, of rare occurrence (very rare or absent except near the base of the main stem), found radiately disposed, completely embedded in the densest parts of the spicular skeleton ; spines rather large compared to the size of the spicule, short and usually recurved towards its base ; size of spicule about 0"175 by 0'0125 mm. A few of the slenderer amongst the smooth styli taper so much towards the base as to become almost or sometimes quite oxeote (PI. XL. fig. 8a, h). This species comes very close to Bowerbank's Dictyocylindrus hispidus,^ but difi'ers from it in its mode of branching and in the presence of the smooth, cylindrical spicules (strongyla), which seem to replace the unequal-ended oxea found in the latter species. ' Vide Mon. Brit. Spong., vol. iii. pi. xvii. 190 THE VOYAGE OF H.M.S. CHALLENGER. The rarity of the spined stylote spicules and their occurrence mainly (almost solely) in the oldest part of the sponge suggest the possibility of their being rudimentary and derived from some ancestral form which possessed them in abundance. Localltij. — Off Bahia ; shallow water. One specimen. Off Bahia; depth, 7 to 20 fathoms. One specimen. Raspailia jlagelliformis, Ridley and Dendy (PL XXXIX. fig. 1). 1886. Raspailia Jlagelliformis, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 482. Sponge (PI. XXXIX. fig. 1) erect, stipitate, branched; stem short and stout, rigid; branches cylindrical, very long and slender, flexible and whip-like ; tapering slightly towards the free end and never anastomosing with one another. Total length of sponge about 425 mm.; length of stem 44 mm.; greatest diameter of same 12"5 mm.; diameter of branches about 4 mm. Colour in spirit yellowish-grey. Texture; each branch consists of a dense, tough, horny looking axis, of darkish colour, coated with a thick layer of a soft, rather friable, yellowish substance, which is easily peeled off so as to leave the axis clean. Surface very minutely hispid. Oscula small, scattered. Skeleton. — Containing a very large proportion of horny matter accumulated in a well- defined central axis which is also densely charged with the stylote spicules. In this central axis the spicules are arranged in not very definite tracts, which, instead of being truly longitudinal in direction, are slightly oblique, and cross one another at very acute angles, giving rise to a somewhat lattice-like arrangement ; the spicules are, however, very densely packed throughout the whole axis, so that no large open meshes appear. From the central axis there radiate outwards and obliquely upwards, through the soft tissues towards the surface of the branch, abundant but loose bands of sjDiculo-fibre, formed of spicules like those found in the axis ; these bands terminate at the surface in a dense velvet-like pile, composed of tufts of small, slender, stjdote or subtylostylote spicules whose apices project for a very short distance beyond the surface. Spicules. — Megasclera ; of one form only, viz., smooth, very slender styli, straight or nearly so, very sharply but not very gradually pointed at the apex ; measuring in the surface tufts about 0'3 by 0-0032 mm. (very often shorter), and in the deeper parts of the sponge about 0*45 by 0*009 mm. This sponge has a considerable resemblance both in external form and in its spicula- tion to Rasimilia ausfraliensis, Ridley,^ from Port Darwin, but differs in being branched, in its thicker and more friable cortical layer, and in the smaller size of its spicules ; with the exception of the character of the external form, the same remarks aj)ply to its relations to the reticulate Raspailia clathrata, Ridley '' from Torres Strait. Locality. — Simon's Bay, Cape of Good Hope ; depth, 10 to 20 fathoms. One specimen. ' Zool. CoU. H.M.S. " Alert," Brit. Mus., 1884, p. 460, pi. xlii. figs, m, m'. Op. cit., p. 461. pi. xli. fiy. F. REPORT ON THE MONAXONIDA. 191 Raspailia (?) riyida, Eidley and Deudy (PL XXXIX. Jig. 3). 1886. Raspailia (?) rigida, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. .5, vol. xviii. p. 483. Sponge (PL XXXIX. fig. 3) commencing as a slender stem only 2"5 mm. in diameter and gradually expanding upwards, without any sudden break, to a diameter of a little over 4 mm.; approximately straight; giving off one short, stout, abnormal-looking branch. Height of specimen 121 mm. Colour in spirit yellowish -grey. Texture firm, tough ; the sponge is subrigid, not very elastic nor flexible ; the difference in this respect between Raspailia jiagelliformis, nobis, and the present species being due to the absence in the latter of a stout, dense, horny axis. Surface very distinctly hispid. Skeleton. — There is a dense central axis of closely packed, stout, tylostylote or sub- tylostylote spicules, arranged longitudinally and parallel with one another, and with their apices directed upwards. From this axis radiate towards the surface rather sparse bands composed of similar spicules, terminating in great divergent brushes, which are again composed of the same kind of spicule (and also of some of larger .size) with little more than their bases embedded in the soft tissues and projecting for a long way beyond the surface. There appears to be little or no horny matter present in any part of the skeleton. Spicules. — Megasclera ; of one kind only, viz., large, straight, smooth tylostyli or sub- tylostyli, very finely and gradually pointed and usually with distinct, subglobular heads ; size variable, up to about 2'0 by 0'025 mm.; usually smaller, especially in the dermal brushes, whose component spicules commonly measure up to about 1 '0 mm. in length. The spicules of this species are extremely large for the genus to which it is referred. The peculiarities of its external characters have been already noticed. It is doubtful whether it is a Raspailia at all, or even an Axinellid ; it presents strong resemblances to the crenus Suherites. Zoco%.— Station 142, December 18, 1873; lat. 35° 4' S., long. 18° 37' E.; Agulhas Bank, Cape of Good Hope; depth, 150 fathoms; bottom, green sand; bottom temperature, 47°'0. One specimen. Genus Dendropsis,' Ridley and Dendy (Pis. XXXVIIL, XL., XLVL). 1886. Dendropsis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. .5, vol. xviii. p. 483. Sponge erect, ramose. Skeleton arrangement iPaspcn^m-like ; megasclera stylote (one form of which may be characterised by the presence of two sharp spikes projecting from the base). Microsclera present in the form of minute spined oxea. We have only a single species of this remarkable genus, but the characters are such as to separate it widely from all existing genera. Notwithstanding the presence of the minute spined oxea we are disposed to place the genus amongst the Axinellidse, for the ' So called because of the resemljlance which the type species bears to a small tree. 192 THE VOYAGE OF H.M.S. CHALLENGER. relations of this rare form of spicule in other sponges are so obscure as to afford no ap^^arent clue to its classificatory importance, if it has any. As examples of other genera in which it occurs we may cite Spongilla and Higginsia. Dendropsis hidentifera,^ Eidley and Dendy (PI. XXXVIII. fig. 1 ; PI. XL. figs. 7, 7a, 7b, 7c; PI. XLVI. fig. 8). 1886. Dewlropds hidentifera, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 483. Sponge (PI. XXXVIII. fig. 1) erect, stipitate, dichotomously branched ; stem cylindrical, branches flattened (especially towards their apices), and all approximately in the same plane. The largest specimen has a total height of about 200 mm.; the stem is 75 mm. long and about 8 mm. in diameter ; the branches are at first of nearly the same diameter as the stem, but lessen upwards. Colour in spirit greyish-yellow. Texture tough and hard. Surface rough, with numerous minute prominences, and hispid. Skeleton. — There is a dense, compact, axial core (PI. XLVI. fig. 8, «) of closely in- terlacing, for the most part sublongitudinally disposed, smooth stylote si^icules ; from this central axis very much larger, stout, smooth, stylote spicules radiate to the surface, beyond which the apices of many of them project. Usually these spicules occur in loose bundles or bands of three or four together, the bases of those nearest to the centre being embedded in the axial core ; they are surrounded by dense sheaves (PL XLVI. fig. 8, h) of the characteristic stylote spicules, the apices of which also often project beyond the surface. Spicides. — (a) Megasclera ; (l) smooth, stout, sharply pointed styli (PI. XL. fig. 7, o), commonly with a rather sharp bend towards the base, towards which also the spicule usually narrows ; in the dense central axis these spicules commonly measure about 0'35 by 0'025 mm.; these are, however, very much smaller than those radiating to the surface, which measure about I'l by 0"044 mm. There occur also very long and slender, smooth styli (PL XL. fig. 7, a) ; these are not nearly so abundant as either of the foregoing, but still can hardly be called rare ; they are of variable size, and may measure about 175 by 0'02 mm.; sometimes, through the blunting and rounding off" of the apex, the long slender styli give rise to strongylote spicules (PL XL. fig. 7, h). (2) The " bidentate" styli (PL XL. figs. 7, c, 7a, 7h) ; long, straight, slender, slightly fusiform and hastately pointed, each with two small, sharp spikes projecting from the base, which may be slightly enlarged. In the base the axial thread is also slightly enlarged, and seems to give off" a branch to each spike. These spicules are very constant in shape and size, measuring about 0'56 by 0'0075 mm.; their disposition in the sponge has already been indicated, (b) Microsclera ; small, entirely spined oxca (PL XL. fig. 7c), each one haiang usually a rather sharp bend in the centre ; size about 1 So-called from tlie jiresence of tlie two spikes on the base of the characteristic stylus. REPORT ON THE MONAXONIDA. 193 0'09 by 0"0045 mm.; the spines are sharp, prominent, and very abundant, except in the centre of the shaft, where they are rather scarce. This remarkable sponge is one which may be easily recognised both by its external appearance and by its spiculation, the latter being quite unmistakable. The skeleton is arranged on decidedly an AxineUid type, especially as regards the manner in which the large, radiating, stylote spicules are surrounded by dense sheaves of small, slender styli. The general appearance of the sponge is thus decidedly Axinellid, 1)ut no sponge belonging to this group has hitherto been found with the spined oxeote microsclera, which closely agree in form with the microsclera of Sj^ongiUa lacustris. Locality. — Simon's Bay, Cape of Good Hope; depth, 10 to 20 fathoms. Two specimens. Genus Thrinacophora, Eidley (Pis. XXIII., XXIV., XXXVL, XL.). 1885. TJirinacophora, Eidley, Narr. ChaU. Exp., vol. i. pt. ii. p. 572. 1886. Thrinacophora, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 483. Sponge ramose, with a dense central axis of spiculo-fibre ; megasclera styli and (or) oxea, and (in some species) cladostrongyla. Microsclera, present in the form of tricho- dragmata. This genus was first founded by Ridley {loc. cit.), for the reception of the very remarkable species which we have called Thrinacophora fiiniformis. It was at first thought that the curious furcation of one of the megasclera in this species would prove to be a character of generic importance, but our subsequent examination of the collection ilisclosed another sponge (TJirinacophora cervicornis), which agrees so closely with Thrinacophora funiformis that the two seem to us to be, at any rate for the present, generically inseparable, although the former possesses none of the branching spicules. Hence we have decided that in this case, as in the Desmacidonidse, the microsclera are the most reliable guides to classification, and we consider the presence of the trichodragmata, taken in conjunction with the Axinellid arrangement of the skeleton, as the most important generic character. This one character separates the genus at once and aljsolutely from all other known genera of AxineUidse. Possibly the branched spicule of Thrinacophora funiformis is comparable to the " bidentate " stylus of the only known species of Dendropsis, and it is noteworthy in this connection that both these genera stand apart from the remaining AxineUidse in the possession of microsclera, though the microsclera are of very different form in the two cases. We have already ^ had occasion to point out the special bearing of the presence of a branched spicule in T]irinacop>hora funiformis upon the question of the relationship of the Monaxonida to the Tetractinellida. ' Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 157. (ZOOL. CHALL. EXP. — PART. LIX. — 1887.) Nnn 25 104 THE VOYAGE OF H.M.S. CHALLENGER. Thrinacophora cervicornis, Eidley aud Dendy (PI. XXXVI. fig. 1 ; PI. XL. figs. 4, 4a). 1886. Tlirinacopliora cervicornis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 483. Sponge (PI. XXXVI. fig. 1) erect, branching dichotomously, subcylindrical, flattened towards tlie points of furcation ; altogether much resembling a stag's antler. Total height 175 mm. ; diameter of stem about 6 mm., of branches about the same or a little less. Colour in spirit greyish-yellow. Texture firm and tough, stem and branches flexible and very elastic. Swface beset with minute monticular eminences, so as to resemble the surface of a coarse file ; from each of these eminences there projects, where the surface is uninjured, a single, very long, setiform spicule, extending for .3 or 4 mm. beyond the surface. Skeleton. — When the sponge is microscopically examined it is readily seen that it possesses a thick central axis of horny consistency and appearance, which gives to it its firm and elastic character, and which is surrounded by a comparatively thin external coat of choanosome which readily peels off". After mounting in balsam the horny matter is no longer apparent, and in the place of the central axis we see a dense reticulation of short oxeote spicules, appearing almost uniserial in arrangement, without any distinct fibre. It is this dense reticulation of spicules, all united together by horny cementing material, which gives rise to the central axis. In the soft layer which coats the axis there are numerous long and very slender oxeote and stylote spicules, arranged longitudinally, but in no definite order ; the short oxea also occur scattered in a similar manner outside the axis. We have already mentioned the very large, setiform (stylote) spicules, which project for a long distance beyond the surface of the sponge and give rise to a coarse hispidity ; these have their bases embedded in the central axis, from which they project at right angles ; each one as it leaves the sponge is surrounded by a whorl or tuft of divergent spicules of very much smaller size. These are very slender styli (? occasionally oxea), whose divergent ends project for a short distance beyond the dermal membrane. Sometimes two of the large styli come out together. Spicules. — (a) Megasclera; (l) short, stout oxea (PI. XL. fig. 4, h), rather abruptly and usually very sharply pointed, occasionally unequal- ended, with the large end very blunt; size about 0'28 by 0"018 mm.; occurring mainly in the axis and forming the greater portion of the skeleton. (2) The long, smooth styli projecting from the surface ; these are almost straight and measure about 5*2 by 0'037 mm. (3) The straight, smooth, slender styli (PI. XL. fig. 4, a) projecting in whorls around the last mentioned ; these are usually sharply pointed, but sometimes become blunted at the apex; size about 0"52 by 0'0075 mm. The long, slender oxea, .styli and strongyla, longitudinally disposed in the choanosome, are to be regarded merely as slight varieties of (3) and measure about the same. (6) Microsclera ; trichodragmata (PI. XL. fig. 4a), fairly plentifully REPORT ON THE MONAXONIDA. 195 cattered in the soft tissues, length about 0'0126 mm.; breadth extremely variable ; these bundles appear of a distinctly Ijrown colour, a character which is also noticeable in Thrinacophora funiforniis, nobis. In the arrangement of the skeleton and the general form of the megasclera, and in the possession of trichodragmata, this sponge bears a most marked resemblance to Thrinaco- phora funiformis, nobis, and the two are evidently very nearly related. The charac- teristic forked spicules of the latter are represented in the present species Ijy the slender styli, and have probably been derived from some such form of spicule (or possibly vice versa). We placed this species first in the genus because we considered it to be the least highly modified of the two. Locality.— ^iaXion 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E.; Philippine Islands ; depth, 18 fathoms; bottom, blue mud. Two specimens. Jlirinacophor a funiformis, Ridley and Dendy (PI. XXIII. figs. 1, lo, \h, Ic, Id, le, 1/ Ig; PL XXIV. figs. 1, la, ll). 1886. Thririacophora funiformis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 484. Sponge (PI. XXIV. figs. 1, la, lb) of fairly definite form; always cylindrical and more or less elongated ; may or may not be branched. The largest specimen in the collection is unbranched except close to the base, where there is a slight trace of branching ; it is about 450 mm. long,' flexible and rope-like, and about 6 mm. in average diameter. A second specimen (PI. XXIV. fig. la) evidently younger, is of very diffierent and more definite form. Arising from a flattened and expanded base there is a short, straight peduncle, which, at a height of about 12 mm., divides into two branches, one of which again divides into two. Total height about 81 mm. Sponge composed of a dense, fibrous axis with a soft external coat of a pulpy consistency. Colour in spirit dirty yellow. Surface very uneven, thickly beset with rather stifi", smaU, projecting conuli. Dermal membrane distinct and very granular. Oscula small, scattered. Pores not seen (the specimens are not in very good condition). Skeleton. — There is a very thick and dense central axis of spiculo-fibre from which other fibres radiate to the surface, there entering into the numerous projecting conuli. It is not easy to make out any other definite skeleton, although there are a great number of spicules scattered through the soft tissues of the sponge. Spicules. — (a) Megasclera; (l) long, slender styli (PI. XXIII. figs, la, 1&), straight or curved, with evenly rounded base and tapering very gradually to a very fine point at the apex ; size very variable, up to about 1*8 by 0*025 mm. (2) Long, slender, unequally 1 The end appears to have been broken off; doubtless the specimen when perfect was a little longer, though probably not much. 196 THE VOYAGE OF H.M.S. CHALLENGER. ended oxea (PI. XXIII. fig. 1) ; with the large end slightly uneven near the point and tapering rather suddenly ; at the small end tapering very gradually to a very fine point, or irregular ; size very variable, up to about 17 by 0-023 mm. There can be little doubt that these spicules are merely modifications of (1). (3) Much shorter, equal ended, fusiform oxea (PI. XXIII. figs. Ic, Id) ; usually sharply pointed at each end, but with the ends often rather uneven, usually with a rather sharp bend in the centre ; size variable, up to about 0'6 by 0'023 mm. (4) Spicules of a very peculiar form, cladostrongyla (PI. XXIII. figs, le. If) ; each consisting of a slightly crooked shaft of even diameter all along ; ter- minating at one end in an evenly rounded extremity and at the other dividing into a number (usually about three) of very irregular, short branches, like the fangs of a human tooth. Size fairly uniform, up to about 0*52 by 0'0063 mm. These spicules often occur together in bundles, like faggots of wood, sometimes surrounding the larger megasclera ; they also occur fairly abundantly in the dermal membrane, where they are scattered separately. It is well worthy of note that the axial thread itself forks with the forking of the spicule, (b) Microsdera ; of one kind only, viz., trichodragmata (PI. XXIII. fig. Ig), occurring abundantly scattered through the soft parts of the sponge, especially in the dermal membrane, in parts of which they form an almost continuous layer. They have the form of small, oblong bundles, measuring about O'l by 0"01 mm. Though usualty fairly comjsact these bundles seem readily to break up into a multitude of component spicules which are exceedingly fine and hair-like. This species is at once distinguished by the presence of the peculiar forked megasclera. Locality. — Ofi^ Bahia ; depth, 7 to 20 fathoms. Two good specimens and four fragments. Off Bahia ; shallow water. One specimen. (?) Ofi" Bahia ; shallow water.' One large specimen and two fragments. The large specimen is composed of two long, straight pieces, which have grown up parallel with one another from a common basal portion, and have fused together for a considerable part of their length. Suborder II. CLAVULINA, Vosmaer. 1885. Olavidma, Vosmaer, Bronn's Klass. u. Ordnung. d. Thieneichs, Porifera, p. 328. Sponge t3q3ically corticate. Skeleton usually more or less radiately arranged, with a dermal crust of spicules which may be either of the same form as in the main skeleton or of a special form (spined spirulaj, &c.). Megasclera typically tylostylote. Micro- sdera, when present, belonging almost invariably to the stellate group ; never chelae or sigmata. Spongin usually absent. 1 According to the inside label ; the outside label had — " Off Bermuda." There can, however, be little doubt that Bahia is the correct locality. REPORT ON THE MONAXONIDA. 197 Family I. Suberitid^e. No microsclera present. Genus Suherites, Nardo (Pis. XXV., XLL, XLIIL, XLV.). 1833. Suierite.1, Nardo, Isis, p. 523. 1864. Hymeniaddon, pars, Bowerbank, Mon. Brit. Spong., vol. i. p. 191. Form various, massive to stipitate. Surface devoid of mammiform projections. Skeleton radiately arranged, nearly always with a dermal crust of smaller^ spicules arranged vertically to the surface. Spicules all monactinal, nearly always tylostylote. This genus appears to be the most primitive of the family, giving rise on the one hand to mammae-bearing forms such as Polymastia, &c., and on the other to symmetrically radiate forms like Trichostenima. The most interesting of the new species obtained by the Challenger are several stipitate forms which in varying degrees simulate the genus Stylocordyla in external appearance {e.g., Suherites ramulosus, nobis), always differing, however, from that genus in the presence of monactinal in place of diactinal megasclera. Suherites carnosus, Johnston, sp. 1842. Halichondria carnosa, Johnston, Britisli Sponges, p. 146, pi. xiii. figs. 7-8. 1866. Hymeniaddon carnosa, Bowerbank, Mon. Brit. Spong., vol. ii. p. 203; vol. iii. pi. ,\xxvi. figs. .5-9. 1884. Suherites carnosus, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., 1884, p. 465. The occurrence of this sponge at or near Port Jackson has already been noted by Ridley {loc. cit.) ; it appears to be a very widelj" distributed species, as will be seen from the list of localities in which it has been found, as given below. The Challenger obtained it off the Azores and off Port Jackson, and a variety of the same species from Fernando Noronha. ioca^i^/ei'.— Station 75, July 2, 1873 ; lat. 38° 38' 0" N., long. 28' 28' 30" W.; off the Azores ; depth, 450 fathoms ; bottom, volcanic mud. One specimen. Off Port Jackson ; depth, 6 to 1 5 fathoms. One specimen. Off Fernando Noronha, September 1, 1873. One specimen of a slight variety, in which the arrangement of the skeleton fascicles at the surface of the sponge is more diffuse than usual. The spicules, also, are larger than usual, measuring in the deeper parts of the sponge about 0"56 by 0"008 mm., and the head varies much in size and is often very irregular in shape ; occasionally it may even be absent. Hahitat. — British Isles (Johnston, Bowerbank, &c.) ; Azores (Challenger) ; Fernando Noronha (Challenger); Kerguelen Island (Carter); Vancouver's Island (?) (Carter) ; Port Jackson (" Alert," Challenger). 198 THE VOYAGE OF H.M.S. CHALLENGER. Suberites caminatus,^ Ridley and Dendy (PI. XLI. fig. 2 ; PI. XLV. figs. 5, 5a, 5b, 5c, 5d). 1886. Suhei-ites caminattcs, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 484. Sponge (woodcut, Fig. 7) massive, roughly hemispherical, sessile, corticate, resembling in shape and size Esperella mammiformis, nobis. The single specimen in the collection is attached by a broad base to an empty Brachiopod shell and terminates in a single oscular projection at the apex. The base is not round but oval, and the longer diameter measures 19 mm., and the shorter about 13 mm. Height of sponge about 13 mm. Colour in spirit light, greyish-yellow. Texture very firm, hard and compact. Surface even, but very minutely hispid and rather harsh to the touch ; also marked with shallow meandering and Fig. 7.— Suberites cami- auastomosiug grooves, SO as to bccomc faintly tesselated. Osculum natus ; attached to a o o ■ <■ i Brachiopod shell; », [Ym. 7 o) siuglc, Small, tubular, at the summit oi the sponge. osculum. \ to ' / C5 ' ' ' _ X o Pores diflficult to make out, almost certainly in the shallow grooves on the surface ; seen in section leading into elongated subdermal cavities between the skeleton fascicles of the cortex. Skeleton. — There is a very dense and definite cortex (ectosome), about 0-63 mm. thick, very sharply and suddenly marked ofli" from the underlying tissues and strength- ened by closely placed fascicles of tylostylote spicules which extend right through it, and have their points directed outwards, but projecting only very slightly beyond the surface. The deeper skeleton is rather scanty and is composed of bands of spiculo- fibre (formed of large tylostylote spicules), which run vertically towards the surface of the sponge to join the cortical layer. Sincules. — Megasclera; of one form only, viz., smooth tylostyli (PI. XLV. figs. 5, 5a, 56, 5c, 5d). There are two chief sizes, the smaller ones occurring in the cortex and the larger in the deeper skeleton ; there is great variability in size ; the spicules of the cortex measure about 0-35 by 0-01 mm., they are stouter in proportion to their length than are the spicules of the deeper skeleton, have well developed, roundedly triangular heads and sharp points, and are often slightly bent. The spicules of the deeper skeleton are very long, straight and slender, have rather less developed, pointedly oval heads, and taper very gradually indeed to an extremely fine point at the apex; size about 1*2 by 0'017 mm. Intermediate sizes are abundant. This is a very pretty and interesting little sponge ; it may be recognised by its external form, and more especially by the projecting, well-marked osculum. The form recalls some- what Tentorium semisuberites. This and the further resemblance recorded in the variety ' So called from the chimney-like osculum at the summit of the sponge. REPORT ON THE MONAXONIDA. 199 below, possibly indicate a real alliance between the species. The single specimen is remarkable for the presence of a great number of large, nucleated sperm-cells (?), which are closely packed together, so as almost entirely to fill the soft parts of the sponge. Locality. — Off Marion Island ; depth, 50 to 75 fathoms. One specimen. We have from Station 320 an interesting series of specimens which should perhaps be considered as belonging to a slight variety of the above species ; they do not, however, appear to be distinct enough from the tyjje to justify us in giving a varietal name. They occur, for the most part, encrusting dead branches of a Sporadopora, on which they form colonies (PI. XLI. fig. 2), the different cushion-like individuals being united together by their bases ; there is usually more than one osculum to each individual, situated at or near the summit of the sponge. The growth appears to be altogether rather more robust and the spicules of a rather larger average size than in the type. We noticed in this variety a feature which we had not observed in the type, namely, that towards the base of the sponge the cortical layer of projecting brushes of tylostylote spicules gives place to a confused layer of short, stout, fusiform tylostyli, calling to mind the similar layer in Tentorium semisiiherites. The sponge is further remarkable as forming colonies by continuous gemmation, in a manner very rare in siliceous sponges. Locality.— ^tsLtion 320, February 14, 1876 ; lat. 37° 17' S., long. 53° 52' W.; ofi'the mouth of the Eio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°"2. Several specimens. Suherites microstomus, n. sp.^ (PI. XLI. fig. 3). Sponge (PI. XLI. fig. 3) sessile, hemispherical. The single specimen in the collection is about 25 mm. in diameter and is attached to a piece of black, volcanic rock. Colour in spirit pale yellow. Texture very firm and dense, with a very strongly developed, fibrous cortex (ectosome) about 0'9 mm. thick. Surface fairly even, minutely granulated. Oscula very minute, but distinct, very slightly raised above the general surface of the sponge and with no thin- walled tubular projections as in Suherites caminatus, nobis {cf. woodcut. Fig. 7 and PI. XLI. figs. 2, 3). Pores, small openings between the cortical brushes of spicules, leading into large, elongated, rather lacunar subdermal cavities in the cortex. Skeleton. — The main skeleton is very well developed and composed of very stout bands of spiculo-fibre, which radiate to the surface, expanding at their ends in a brush- like manner and merging in a distinct cortical layer composed of dense brushes of much smaller, radiately disposed spicules, whose apices project for some distance l^eyond the surface of the sponge. It is important to notice that the special cortical layer of spicules This is the sponge referred to in our Preliminary Report (Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 485) as a slight variety of Suherites caminatus, nobis, with very small, non-tubular oscula. 200 THE VOYAGE OF H.M.S. CHALLENGER. occupies only the upper quarter of the cortex, and that the spicules composing it project beyond the surface for nearly half their length. The lower three quarters of the thick ectosome are penetrated and strengthened only by the ends of the main skeleton fibres. (Compare this arrangement with that found in Suherites caminatus, where the cortical layer of spicules occupies nearly or quite the entire thickness of the ectosome.) Spicules. — Megasclera; all tylostylote. (1) Large, stout, straight, gradually and very sharply pointed, with very slightly developed heads, which are a little elongated ; size up to about 1-2 by 0'02 mm. These spicules occur in the fibres of the main skeleton. (2) Much smaller, straight or slightly curved, slender, gradually and sharply pointed, with heads more distinctly marked and usually of the shape termed by Bowerbank "enormi- spinulate," .size variable, say about O'S by 0-006 mm. These spicules form the special cortical skeleton and also occur in bundles or tracts in the deeper parts of the sponge. We at first put this species down as a slight variety of Suherites caminatus, nobis, but our subsequent more careful examination of it has proved it to be really quite distinct, although in many respects the two species resemble one another. The most important points of diflerence concern (1) the oscula, which are minute and non-tubular; (2) the arrangement of the cortical skeleton (w'cZe sttpra) ; (3) the form and size of the cortical spicules, which in the present species are much smaller and slenderer than in Suherites caminatus. In minute histological characters, however, the two species come very close to one another. iocaZ%.— Station 150, February 2, 1874; lat. 52° 4' S., long. 71° 22' K; between Kerguelen and Heard Islands ; depth, 150 fathoms ; bottom, coarse gravel ; bottom temperature, 35°'2. One specimen. Suherites perfectus, Ridley and Dendy (PI. XLI. fig. 9 ; PI. XLV. figs. 3, 3a, Sh). 1886. Suherites perfectus, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, voL xviii. p. 485. Sponge (PI. XLI. fig. 9) erect, lobose. The single specimen in the collection is a simple, upright, subcylindrical lobe, 81 mm. high by about 19 mm. in diameter. Colour in spirit brownish-yellow. Texture hard, firm, very slightly compressible. Surface minutely reticulate, almost glabrous in appearance but in reality very minutely hispid, harsh to the touch, fairly even. Dermal memhrane fairly distinct, but reduced to a network by the very numerous pores. Oscula small, very abundant, irregularly scattered, each on the summit of a small projection. Pores; the surface of the sponge, as already stated, is minutely reticulate ; it is divided into a number of oval meshes by a network of tissue; each mesh is about 0-2 mm. in diameter and is subdivided into smaller meshes by the delicate, cribriform dermal membrane which stretches across it ; there are thus larger and smaller meshes, the larger meshes containing some half dozen of the REPORT ON THE MONAXONIDA. 201 smaller, which are about 0'07 mm. in diameter and are the true pores ; these pores lead into well-developed subdermal cavities between the ends of the radiating skeleton fascicles. Skeleton. — Radiately arranged, composed entirely of tylostylote spicules. There is a dennal crust of small tylostylote spicules arranged in closely placed, but not very tightly packed, divergent brushes, the points of the component spicules projecting freely beyond the dermal membrane and giving rise to the hispidity already mentioned. The main skeleton is composed of radiating fascicles of large tylostylote spicules, running vertically to the surface and there ending in the brushes of small spicules which form the dermal crust ; between these main fascicles are scattered other tylostylote spicules. Spicules. — Megasclera; of one form only, viz., smooth tylostyli (PL XLV. figs. 3, 3a, 36), straight or slightly crooked, slightly fusiform, with well-developed, hemi- spherical or nearly globular heads, and fairly sharply but rather suddenly pointed at the apex. These spicules are very variable in size ; small ones are found in the dermal <;rust and large ones in the deeper parts of the skeleton ; the range in length is about from 0"2 to 1'2 mm. and in diameter about from 0"0078 to 0'03 mm. The spicules of the dermal crust commonly measure about 0'28 by 0*0126 mm.; those of the deeper skeleton about TO by 0'025 mm.; but there is great variation. This is a remarkably fine species and shows the oscula and pores (which are often extremely difiicult to make out in the genus Suberites) very distinctly.^ Its dis- tinguishing characters are the general external ajapearance, the hard texture and the incipient stalk ; in the latter point it approaches Suberites axiatus (below). Locality. — Port Jackson ; depth, 30 to 35 fathoms. One fine specimen. Suberites antarcticus. Carter (PI. XLV. fig. 7, 7a, 7b, 7c, 7d). 1876. Suberites antarcticus, Carter, Ann. and Mag. Nat. Hist., ser. 4, vol. xviii. p. 391. Sponge erect, much branched, bushy ; branches long, cylindrical, may anastomose on coming into contact with one another. Total height of Challenger specimen about 150 mm. ; greatest breadth about 50 mm., diameter of branches 6 to 8 mm. Colour in spirit black (colouring (?) the spirit deep amber). ^ Texture firm, but a little spongy. Surface fairly even, but minutely reticulate and also very minutely hispid ; harsh to the touch. Dermal membrane delicate, transparent, difficult to make out. Oscula numerous, small, distinct, scattered over the branches, sometimes placed serially one above the other, each on the top of a small, low papilla. The excretory canals can be very distinctly seen through the skin, converging towards the oscula in a stellate manner. Pores apparently arranged as in Suberites perfectus, nobis, but not clearly made out. ' Further details with regard to the anatomy and histology will be found in the Introduction. = It is possible that this colour is due to other sponges packed in the same vessel, but Mr. Carter's type specimen in the Britis-h Museum seems to indicate otherwise. (zooL. CHALL. EXP. — PART OX. — 1887.) Nnn 26 202 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton. — In each branch there is a central axis, formed of a very dense reticulation of closely packed, tylostylote spicules. These pass into rather loose and irregular, radiating fascicles of spicules, which terminate in a " dermal crust," composed of divergent brushes of tylostylote spicules whose points project very slightly beyond the surface. Just below the surface there also occur a few more or less horizontally placed spicules, which, together with the radiating brushes, are so arranged as to give to the surface its reticulate appearance. Spicules. — Megasclera ; of one kind only, viz., smooth tylostyli (PI. XLV. figs. 7, 7a, 7b, 7c, 7d), almost or quite straight, with well-developed, nearly globular heads and slightly fusiform shafts, tapering very gradually to a very sharp point at the apex. In size these spicules vary very much, smaller ones being found towards the outside and larger ones in the centre of the sponge. Often in the case of the smaller spicules the head and neck are bent at a slight angle to the remainder of the shaft. The spicules vary in length from about 0'175 to about 0'875 mm., and in diameter from about 0'0094 to about 0'025 mm.; the spicules in the dermal crust being commonly about 0'28 mm. and those of the deeper parts about 0'7 mm. long. Mr. Carter refers twice in the Annals and Magazine of Natural History to his Suherites antarcticus,^ and as his type of the species (dry) is in the British Museum there is no doubt as to what the species really is, although it has not hitherto been fully described. The above description is taken entirely from the Challenger specimen, which is a very fine one, and preserved, of course, in spirit. In Mr. Carter's specimens the branches are much more flattened and very much mpre anastomosing than in the Challenger example. It will be seen that this species comes near to Suherites 'perfectus, nobis, but it differs from it in its manner of growth and in its extraordinary dark colour (?), which may be a very characteristic feature. There are also slight diff"erences between the spicules of the two species, but these do not amount to much. Locality. — Off" Christmas Harbour, Kerguelen ; depth, 70 fathoms. One specimen. Habitat. — Antarctic Seas (Carter) ; Kerguelen (Challenger). ^ Loe. (it. supra and Ann. and. May. Nat. Hid., ser. 5, vol. ix. p. 350. Mr. Carter here says: — "Dredged by Sir James Ross in 74^° S. lat., in 206 i'atlioiiis (Expedition of 1841), undescribed. Character. Stipitate branched; branches digitate, nodose, pollachotomous ; structure subcavernous ; colour dark house-mouse ; spicule pin-like, with large spherical head. Size of specimen about 5| x 3 inches." REPORT ON THE MONAXONIDA. 203 Suberites axiatus, Eidley and Dendy (PI. XLV. figs. 15, 15«, 156, 15c). 1886. Suherites axiatus, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 485. Sponge irregular in shape, lobose or digitate. Unfortunately the species is repre- sented in the collection only by a few fragments in very bad condition ; they are now of a dull, reddish-yellow colour (like many other specimens from the same station), which appears to have been caused by their having been packed in some iron vessel which has rusted and discoloured tlie spirit. Texture soft and spongy, with a thick, dense axis up the centre of the branches. Surface hispid. Skeleton. — There is a dense, stout, central axis of longitudinally placed spicules, which, though closely packed, are not firmly united together, so that the axis is readily broken across, leaving at each broken end a mass of projecting spicules which have been pulled out from amongst their fellows. From this axis radiate bands of spiculo-fibre to the surface of the sponge, where the spicules diverge and give rise to more or less of hispidity. The radiating fibres do not usually come off at right angles from the axis \mi more or less obliquely, so that when a branch is broken across the broken end has the form of a cone on the one hand, and of a funnel, into which the cone fits, on the other. S2'>icules. — Megasclera; of one form only, viz., tylostyli (PI. XLV. figs. 15, 15o, \bh, 15c), with fairly well-developed, subglobular heads and fusiform shafts, tapering gradually to a sharp point at the apex. Size very variable ; the larger spicules, in the interior of the sponge, measure about 175 by 0"031 mm., but towards the outside of the sponge the spicules become very much shorter and slenderer, commonly measuring about 07 bj- 0-0126 mm. The most remarkable point about this species is the very strongly developed axis ; it will be seen to approach closely to Carter's sjiecies Suherites antarcticus, but difi'ers in habit, in colour, and in the much greater size of the spicules. Zoca^.— Station 320, February 14, 1876 ; lat. 37° 17' S., long. 53° 52' W.; off the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 37°'2. Fragments. Suberites durissimus, Eidley and Dendy (PI. XLI. fig. 1 ; PI. XLV. figs. 2, 2a, 2b, 2c, 2d, 2e). 1886. Suherites durissimus, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 485. Sponge (PI. XLL fig. 1) pedunculate, form rather irregular. The single specimen in the collection consists of a basal, rhizome-like portion, from which arise two stalks, each bearing an expanded, oval or subglobular head ; in the case of the smaller one the peduncle is produced upwards beyond the expanded head, while the larger is broadly and evenly rounded at the top. The smaller stalk gives off, near its base, a flattened process 204 THE VOYAGE OF H.M.S. CHALLENGER. ending in a third, small, globular head. Total height of specimen 56 mm.; diameter of largest head 15 mm.; diameter of largest peduncle 6 mm. Colour in spirit light, greyish-yellow. Texture solid throughout, hard, incompressible and woody. Surface even, but covered with a velvet-like pile. Skeleton. — Radiately arranged, very dense, consisting of closely packed fascicles of spicules radiating towards the surface. There is a dermal crust of thickly packed, pro- jecting, tylostylote spicules, and below this come the main skeleton fascicles, composed mainly of stylote spicules ; (we have been unable to determine with certainty whether any of these large styli project beyond the surface of the sponge and take part in the formation of the velvet-like pile, or whether the latter is formed entirely of tylostylote spicules). The dermal crust of spicules is continuous all over the surface, but every alternate fascicle of the main skeleton stops short a little way below the surface and before it reaches the bottom of the dermal crust, and thus arise a number of fairly definite spaces, devoid of spicules, which seem to be comparable to the subdermal cavities of other Monaxonida. Spicules. — Megasclera; (1) tylostyli (PI. XLV. figs. 2a, 26, 2c), straight or nearly so, with well-developed, almost globular heads, fusiform shafts, and sharply and gradually pointed at the apices. These occur mainly in the dermal crust, with their points directed outwards ; a common size for them is about 0*24 by 0"0063 mm., but in this respect they vary extremely, and they pass by a series of transitional stages (both as regards form, size, and position) into (2) the deeper spicules found in the main skeleton fascicles. These spicules are very long, straight, slender, smooth, fusiform styli and subtylostyli (PI. XLV. figs. 2, 2d, 2e), very gradually and sharply pointed at the apices ; size about 1'5 by 0"0157 mm.; having their apices generally (? always) directed outwards. This is one of those species which appear to mark a transition in external form from Suherites to Stylocordyla. Locality. — Ofl:' the south-east coast of Australia. One specimen. Suherites mollis, Ridley and Dendy (PI. XLI. fig. 4 ; PI. XLV. figs. 4, 4a, 46, 4c). 1886. Suherites mollis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 486. Sponge (PI. XLL fig. 4) stipitate, with expanded, lobose head. The short stem is attached by a slightly expanded base to a fragment of stone, and is surmounted by the broad, lobose, slightly compressed, suboval head, bifurcating slightly at the top. Total height of sponge 31 mm., length of stem 13 mm.; diameter of same about 2"5 mm.; greatest breath of head about 13 mm. Colour in spirit pale yellow. Texture (of head) very soft and spongy ; of stem, firm, but brittle, scarcely fibrous. Surface even, but minutely hispid. Dermal merahrane distinct. Pores apparently scattered. Oscula (?). Skeleton. — (a) The skeleton of the stem ; this consists of a dense, well-defined, central REPORT ON THE MONAXONIDA. 205 axis of longitudinally placed spicules, from which radiate bands of very loose spiculo-fibre to the surface, like the spokes of a wheel ; these project more or less beyond the surface and give rise to its hispidity. (b) The skeleton of the head ; this is very diffuse, and consists of loose bands of sjjiculo-fibre which run more or less vertically upwards towards the top of the sponge, radiating very slightly ; as they approach the surface they break up into brushes of smaller tylostylote spicules. Owdng to the direction in which they run it is only at the top of the sponge that the skeleton fibres abut vertically against the surface ; lower down their direction lies more or less parallel with it, and they only run very obliquely into it. Spicules. — Megasclera; of one kind only, viz., smooth tylostyli (PL XLV. figs. 4, 4o, 46, 4c), with straight, or very slightly curved shafts and fairly well-developed, hemi- spherical or subglobular heads ; tapering gradually to a fine point at the apex. Size extremely variable, in the deeper fibres very long and stout, towards the surface usually much smaller. They may attain a length of about 2'0 mm. and a diameter of about 0'03 mm.; but, as already stated, they are very much smaller in the radiating brushes at the surface of the sponge, commonly measuring about 0"4 or 0"5 by about O'Ol mm. The most remarkable features of this sponge are its great softness and looseness of texture, as compared with the more typical species of Suberites, and the reduction of the " dermal crust" of spicules, which no longer forms a distinct cortical layer. Locality. — Station 148, January 3, 1874; south of the Crozets ; liottom, hard ground, gravel, shells ; depth, 240 to 550 fathoms. One specimen. Suberites elongatus, Ridley and Dendy (PI. XLIII. figs. 11, 12). 1886. Suberites elongata, Ridley and Dendy, Ann. and. Mag. Nat. Hist., ser. 5, vol xviii. p. 486. Sponge (PI. XLIII. figs. 11, 12) consisting of a slender, fleshy-looking stalk, expanding below into attaching rootlets ; and an oval, more or less elongated head. The relation, as regards size, between the head and the stalk is a very variable one ; in one specimen the head is about 12 mm. and the stalk about 44 mm. long, while in a second the head is 25 mm. and the stalk only about 12 mm. long. The respective diameters of the head and stalk are, on the contrary, pretty constant, that of the head being about 4 mm. and of the stalk 1 mm. Colour in spirit pale yellow. Texture firm, cork-like. Surface even, on the head minutely hispid, on the stalk stiU more minutely so. Oscula very few, minute, scattered, sometimes one or two, sometimes none, discernible ; usually at or near the top of the head. Skeleton. — The dense central axis of the stem, composed of longitudinally placed, stylote spicules, is continued into the head as a not very definite central core, from which radiate, outwards and upwards, bands of spiculo-fibre to the surface of the sponge ; on approaching the surface these branch more or less and the points of their terminal spicules 206 THE VOYAGE OF H.M.S. CHALLENGER. project freely, and, together with a dense dermal crust of small, projecting, tylostylote spicules, give rise to the hispid character of the head. The interspaces between the radiating bands of spiculo-fibre are occupied by loosely and irregularly scattered tylo- stylote spicules. Spicules. — Megasclera ; (1) very long, straight, slender, smooth styli or subtylostyli, very finely and gradually pointed at the apex ; size about 1'8 by 0'016 mm. (2) Much smaller, straight, slender tylostyli with well-developed, subglobular heads, and very sharply and gradually pointed at the apex ; size rather variable, about 0-35 by 0-0063 mm. As regards spiculation there is but little characteristic about this species ; the great slenderness of the long stylote spicules, and the presence of a fairly well defined dermal crust of smaller tylostyli may be noted. The external form, though variable, is very characteristic ; the elongated oval shape of the head and the relative shortness of the stalk being its prominent features. Zoca/%.— Station 75, July 2, 1873 ; lat. 38° 38' 0" N., long. 28° 28' 30" W.; off the Azores ; depth, 450 fathoms ; bottom, volcanic mud. Eiglit specimens. Suberites sjni'alis, Ridley and Dendy (PI. XXV. fig. 1). 1886. Suherltes spiralis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 486.' Sponge (PI. XXV. fig. 1) erect, stipitate ; cylindrical, with the stalk produced upwards, as an axis, right through the body of the sponge. Height 88 mm.; length of stalk 31 mm.; diameter of stalk 2 mm.; of body 6 mm. Colour in spirit pale yellow. Texture of stalk and axis very dense and firm ; of body fairly firm, but rather open and compressible. Surface very minutely hispid. Oscula scattered over the body of the sponge. Skeleton. — Consisting of a dense, stout, central axis of longitudinally placed spicules, from which, in the bod)^ of the sponge, radiate loose fibres to the surface, arranged in a somewhat spiral manner and terminating in loose, radiating brushes of smaller spicules with outwardly directed apices. Spicules. — Megasclera; (1) long, slender, straight, or nearly straight, sharp-pointed tylostyli ; with well-marked, but not very large, subglobular heads ; size about 1 -0 ):)y 0'013 nmi. ; these constitute the skeleton fibres. (2) Similar, but much smaller spicules, occurring in the surface brushes, length about 0'4 mm,, but not very constant. In the presence of a distinct axis with radiating fibres this sponge resembles the Axiuellidaj. The habit is decidedly that of an Axinellid, and it seems to strengthen the possibility of a close relationship between the Axinellidje and Suberitidae. The peculiar spiral arangement of the skeleton fibres in the body of the sponge is to be found still more distinctly shown in Stylocordyla stijyitata, var. glohosa {vide woodcut, Fig. 9). Locality. — Off the south-west coast of Patagonia. One specimen. REPORT ON THE MONAXONIDA. 207 Suberites ramulosus, Ridley and Dendy. 1886. Suberites mmuloaa, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 487. Sponge stipitate, stalk slender, often crooked, simple or branched ; ending below, in perfect specimens, in spreading rootlets, and expanding above into a pear-shaped head (none of the five specimens present exhibit more than one head, though two show branching of the stem, but there can be little doubt that several heads may be present, as described in the variety below, of which a much more perfect series of specimens is to hand). Size variable ; the largest specimen has a stem 44 mm. long by about r5 mm. thick, and a head about 13 mm. long by 8 mm. broad. Texture rather soft and spongy, surface hispid. Oscula; a single one at the summit of each head, surrounded by a slight spicular fringe. Skeleton. — Very irregular ; a dense central axis of longitudinally placed, large tylostylote spicules in the stem, and in the head longitudinal tracts of similar spicules, with loose brushes of smaller ones whose points project l)eyond the surface and give rise to the hispidity ; (compare the description of the skeleton in the variety given below, where it seems to be much more definitely arranged). Spicules. — Megasclera; (1) large, straight, stout, fusiform tylostyli, with well-developed subglobular heads and (usually) blunt apices ; size about 1'8 by 0"063 mm. (in the stem they may be a little longer and slenderer). (2) Much smaller and slenderer, long, straight, usually sharply pointed tylostyli, with well-marked, nearly globular heads ; these spicules form the brushes at the surface of the sponge ; their size is very variable, they commonly measure about TO by 0028 mm., but it would not be difficult to pick out a series connecting them with the larger kind. In the stem the spicules are frequently very crooked, calling to mind those of some Axinellida^. The chief point of interest in this species, apart from the branching of the stem, lies in its relation to a well-marked variety of the same fr(3m a neighbouring station, which is fully described below ; the comparison of these two forms shows how little reliance can be placed on slight differences in the form of the megasclera, that is, whether "stylote," " strongylote," or "tylostylote," for purposes of classification. We have chosen to consider the form described above as the type, because its spicules, in the presence of well-developed heads, agree with those of more typical species of Suberites, while the larger spicules of the variety differ from them in this respect. The species is further of exceptional interest owing to the manner in which the external form simulates that of Stylocordyla. LocalitT/.— Station 207, January 16, 1875 ; lat. 12° 21' N., long. 122° 15' E.; Philippine Islands; depth, 700 fathoms; bottom, blue mud; bottom temperature, 5V6. Five specimens. 208 THE VOYAGE OF H.M.S. CHALLENGER. Suheiites ranndostis, Y£iT. cylindrifeo-a^ (PI. XXIX. figs. 5, 5a; PI. XLIII. figs.' 13, 14). 1886. Suheriies ramulosa, var. cylindrifera, Kidley and Dendy, Ann. and Mag. Nat.' Hist., ser. 5, vol. xviii. p. 487. Sponge (PI. XLIII. figs. 13, 14) consisting of one or more heads, with long, slender, simple or branched stalks ; when the stalk is simple there is only one head, but when it is branched there is a head at the end of each branch. The stalk is slender, usually more or less crooked, and ends below (in perfect specunens) in branching rootlets ; at its upper end it gradually expands into the elongated, pear-shaped head, which terminates in a single osculum at the summit. When the stem is branched the branches come ofi" from the main stem at various angles and quite irregularly ; when a stem comes into contact with a head or with another stem the parts in contact may anastomose. Size of specimens variable; the finest unbranched specimen has a head which is 31 mm. long by about 13 mm. broad, and a stem about 100 mm. long by about 2*5 mm. thick. Colour in spirit dirty, greyish-yellow. Texture of head rather soft and spongy ; of stem firm and fibrous. Surface strongly hispid. Oscula ; normally there is a single one, sur- rounded by a slight spicular fringe, at the summit of each head. Skeleton. — The skeleton of the stem consists of a dense, central axis of large, blunted stylote spicules, from which project at right angles loose tufts of smaller, subtylostylote spicules whose points project beyond the surface. That of the head consists of ascending lines of very loose spiculo-fibre, composed of the large, blunted styli, running very ol)liquely to the surface, where they end in loose, irregular tufts of smaller subtylostyli whose points project outwards and upwards and give to the surface its uniform hispidity. Spicules. — Megasclera; (l)long, smooth, straight, slightly fusiform (cylindrical) styli (PI. XXIX. figs. 5, 5a), evenly rounded off at the base and tapering gradually to a narrow, but rounded, apex; size about 2*6 by 0"056 mm. (2) Long, straight or sHghtly curved subtylostyli, with slightly developed heads and usually sharply pointed at the apex ; size very variable, passing by occasional transitional forms into the larger spicules, commonly measuring about 1"0 by 0'014 mm. but often smaller. This very interesting variety is well represented in the collection Ijy a large number of fiine specimens ; indeed it is very much better represented than the more typical form of the species. A very remarkable point in connection with it is the high temperature (71°"0) of the water in which it grew ; it was obtained from Station 209 only, while the types of the species were obtained from Station 207. It differs from the types mainly in the absence of heads to the larger megasclera and in their still greater bluntness, and throws considerable doubt on the specific value of these characters ; diflerences also exist in the proportions of the spicules, which will be seen by comparison of the descriptions. As regards external form there is very little difference to note between the specimens ' So-called on account of the strong tendency exhibited by the larger spicules to become cylindrical. REPOKT ON THE MONAXONIDA. '209 from the two stations, save that the variety, in so far as can be judged from the specimens in the collection, seems to attain a greater size and altogether a much more luxuriant growth than does the species ; possibly this is attributable to the higher temperature of the water in which it lives. Zoc«%.— Station 209, January 22, 1875 ; lat. 10° 14' N., long. 123° 54' E.; Philippine Islands; depth, 95 fathoms; bottom, blue mud; bottom temperature, 71°"0. A large number of specimens. Subcrites senilis, Eidley and Dendy (PI. XLV. figs. 1, Iff, lb). 1886. Suherites senilis, Ridley and I>endy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. \k 485. Sponge sessile, hemispherical, attached to a small stone (apparently a manganese nodule) ; covered with very long, delicate, projecting spicules like a coating of grey hair,' with a single, small, oscular (?) projection at the summit. Diameter about 8 mm. (excluding the projecting spicules). Colour in spii'it (after drying) pale, greyish-yellow. Surface very strongly hispid. Osculum (?) single, at the top of a small, chimney-like projection. Skeleton. — Composed of great, radiating, divergent brushes of tylostylote spicules. These brushes arise from various levels in the sponge, some from the base and others from nearer to the surface, and their projecting spicules, many of which extend for a millimetre or more beyond the surface of the sponge, give rise to its haiiy appearance. Within the sponge itself the space (where such exists) between the radiating brushes of spicules is filled up with a confused mass of shorter tylostylote spicules, which ap)pear to be mixed up together without any pretence of arrangement. Spicules. — Megasclera (P\. XLiY . figs. 1, 1«, lb); of one kind only, but varying much in size. The spicules of the radiating brushes are very long, straight and slender, slightly fusiform, with very well developed oval heads and tapering gradually at the apex to a very fine point (usually broken ofi"). These spicules may reach as much as about 3'0 mm. in length with a diameter of about 0"019 mm. The shorter spicules are much more distinctly fusiform than the longer ones, stouter in proportion to their length, wdth narrow, constricted neck and almost globular head ; they are gradually and finely pointed at the apex and often more or less bent near the neck ; length very variable, commonly about 0"5 mm., sometimes less and often more, diameter about 0"015 mm., passing into the larger forms already noticed. The most interesting points about this sponge are the great size of its spicules and the very great depth (2050 fathoms) from which it was obtained. Unfortunately it was put into a very small bottle with a bad stopper, and, though the bottle was bladdered down, was completely dried up. Although it has no well-marked spicular cortex, yet this ' Hence the specific name. (zooL. oHALi.. Exr. — PART Lix. — 1887.) Nnn 27 210 THE VOYAGE OF H.M.S. CHALLENGER. species appears to have a distiuct affinity with Trichostemma ; its small size, general form, and projecting spicules seem to indicate this ; for the same reason it stands well apart from other species in its own genus. Locality.— Station 246, July 2, 1875 ; lat. 36° 10' N., long. 178° 0' E.; North Pacific Ocean; dejith, 2050 fathoms; bottom, Globigerina ooze; bottom temperature, 35°-l. One specimen. Genus Polymastia, Bowerbank (Pis. XLL, XLIL, XLIV.). 1864. Pdhjmastia, Bowerbank, Mon. Brit. Spong., vol. i. p. 177. 1870. {?) Rinalda, Schmidt, Spong. Atlant. Gehiet., p. 51. 1885. Wcherolla, Vosmaer, Sponges of " Willem Barents" Expedition, 1880-81, p. 16. Suberitid^ of massive, sessile form, with more or less numerous mammiform processes on the upper surface, some of which may bear oscula at their summits, but usually without visible openings. Skeleton radiately arranged ; stout main fibres run vertically to the surface, and there is a cortical layer of smaller spicules arranged vertically to the surface, extending both over the body and the mammiform processes. Megasclera tylo- styli or styli. Sponge usually attached and without any supporting fringe of spicules. Vosmaer {loc. cit.) gives the following diagnosis of his genus Weberella : — " Massive, globular, very compact. The surface covered with stout, short papillae. On sections a rind is nearly always distinguished from the mark. Main canals surrounded by compact connective tissue. Canal system of the fourth type. Spicules acuate or pin- shaped." The chief difi'erence from Polymastia appears to lie in the shortness of the mammiform processes, but here, as also in Polymastia, they are not merely slightly elevated oscula but sometimes closed at the extremities, hence we can see no reason for separating the two genera. We have some reasons for believing that the genus Trichostemma will also ultimately have to be united with Polymastia,, but further investigation than the time at our disposal will allow of is necessary before taking this step.^ The genus Polymastia is chiefly confined to deep water, but curiously few repre- sentatives of it were obtained by the Challenger. Polymastia rohusta, Bowerbank (PI. XLL fig. 8). 1861. Eu2^lectella rohusta, Bowerlsank, List of British Marine Invertebrate Fauna, Rep. of Brit. Assoc, for 1860, p. 236. 1866. Polymastia rohusta, Bowerbank, Mon. Brit. Spong., vol. ii. p. 62; see also op. n't, vol. iii., pi. X., p. 23, and vol. iv. p. 31. This species is represented in the collection by a single fine specimen (PI. XLL fig. 8) dredged off the south of Nova Scotia ; in external appearance the specimen has a 1 Cf. footnote, p. 218. EEPOET ON THE MONAXOISTIDA. 211 much closer resemblance to one obtained by the " Porcupine," and now in the British Museum, than to the original type. Locality.— Station 49, May 20, 1873 ; lat. 43° 3' N., long. 63' 39' W.; south of Halifax, Nova Scotia ; depth, 83 fathoms ; bottom, gravel, stones ; bottom temperature, 35°'0. One s]3ecimen. Habitat. — British Seas (Bowerbank) ; off Shetland Islands (" Porcupine ") ; south of Nova Scotia (Challengei"). From the same station (Station 49) there is a second specimen of Polymastia; it is small and attached to a stone. It has three small fistulse and is chiefly remarkable from the presence of a vast number of very long, projecting spicules, which form a deep, velvet-like coat over the body of the sponge. Unfortunately it has been dried up and we cannot be certain of the species to which it belongs ; possibly it is Polymastia mammillaris,^ but probably only a young form. Polymastia corticata, Ridley and Dendy (PI. XLII. figs. 4, 5, 5a, 5h, be ; PI. XLIV. fig. 3). 1886. Polymastia corticata, Eidleyand Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. \i. 487. Sponge (PL XLIV. fig. 3) massive, sessile, cushion-shaped; with strongly convex upper surface, bearing very numerous mammiform processes ; corticate ; consisting of a dense, leathery cortex about 2'5 mm. thick, and an internal, rather friable, dense, amorphous mass of a pale yellow colour, from which the cortex easily separates. The single specimen in the collection is oval in shape, measuring about 87 mm. in length by 62 mm. in width and 37 mm. in thickness. The mammiform processes are of two very difi'erent sizes — (1) small, very abundant (there are considerably over one hundred on the single specimen), hollow, elongated, generally flattened and closed at the ends, which are frequently pointed ; height about 8 mm., breadth at base 3 mm. (2) Very large, flattened and gradually conical tubes, very few in number (only four or five on the specimen), sometimes open fairly widely at the summit, with a fringe of spicules around the opening, sometimes, however, with no trace of an opening ; height about 25 mm., breadth at base a little over 12 mm. In the immediate neighbourhood of the larger processes the smaller ones are scarce or absent. The larger processes, open at the ends, are evidently oscular tubes, each with a single osculum at the summit, while the nature of the small processes is doubtful. The colour of the sponge in spirit is milk- white (this applies to the cortex only, the internal tissues being pale yellow). The surface is smooth but extremely minutely hispid. Pores in scattered groups of two or three, each group over a small round or oval subdermal cavity ; they seem to occur on every part of the sponge, on both kinds of mammiform processes as well as on the general surface. ' Vide Bowerbank, Men. Brit. Sjjong., vol. ii. \>. 71. 212 THE VOYAGE OF H.M.S. CHALLENGEE. Skeleton. — (1) Of the body at large ; the outermost portion of the skeleton consists of a dense layer (PL XLII. fig. 4, a) of thickly set brushes of small, slender, t3'lostylote spicules, whose points project beyond the surface of the sponge ; this layer, the most external layer of the cortical skeleton, is about 0"28 mm. thick;' it is immediately suc- ceeded below by a very much thicker layer (PI. XLII. fig. 4, h) of large stylote orsubtylo- stylote spicules, closely and for the most part more or less vertically placed, and with a few small tylostyli amongst them ; this layer constitutes the remainder, which is by far the greater portion, of the cortical skeleton. Below the cortex the skeleton is no longer definitely arranged, but consists of a confused mass of thickly scattered spicules, often forming rude fibres. (2) In the mammiform projections there is no such thick cortex as on the main l)ody, they are, comparatively speaking, thin-walled. The outermost layer of the cortical skeleton, consisting of closely packed, small tylostylote spicules, is, however, still present, and below this we find definite longitudinal bands of stout spiculo- fibre, composed of the large megasclera, and a loose network of similar spicules irregularly disposed. Spicules. — Megasclera; (1) small, straight or slightly curved, fusiform tylostyli (PI. XLII. fig. 5a), sharply and gradually pointed at the apex and with well developed, pointedly oval heads ; size about 0'28 by 0"008 mm. (2) Large, straight, smooth, fusiform styli or subtylostyli (PL XLII. figs. 5, 5&, 5c), tapering very gradually to a sharp point at the apex and narrowing considerably towards the base ; size about 0'98 by 0-022 mm. The species is distinguished from its congeners by its very pure white colour and the very large size and flattened form of the larger mammiform processes. Zoca%.— Station 125, September 12,1873; lat. 10° 46' S., long. 36° 2'W.; between Pernambuco and Bahia ; depth, 1200 fathoms ; bottom, red mud. One specimen. Polymastia agglutinam, Ridley and Dendy (PL XLI. fig. 6 ; PL XLII, figs. 1, 2, 2a, 2h, 3). 1886. Polymastia agijlutinans, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 488. Sponge (PL XLI. fig. 6) sessile ; encrusting and enveloping pebbles, &c., and collecting and cementing on to its own surface numerous fragments of shells, grains of sand, and other foreign objects (whence the specific name), and giving ofi" long, slender, cylindrical, fistular processes, each with a distinct canal up the centre but closed at the top. These processes are quite clean and free from any coating of foreign objects such as covers the body of the sponge. There are two specimens in the collection, the largest of which has a body of irregularly globular form, about 12 to 18 mm. in diameter, and with a great deal of coarse foreign matter inside and adhering to it, quite disguising it ' Many of the spicules in this layer are drawn out into long, hair-like points, not shown in the figure. REPORT ON THE MONAXONTDA. 213 it bears eight or nine fistulse, varpng from minute buds to processes measuring about 13 mm. in length and about 2 mm. in diameter. Tlie smaller specimen bears only a single long process. Colour in spirit (where the surface is visible) almost white. Texture of the body, firm, very gritty; of the processes, firm, compact, rather stiff. Surface roughened on the body with the foreign objects ; smooth and glabrous-looking on the fistular processes, Ijut in reality very minutely hispid. Oscula and Pores both doubtfully observed, but we have reason to believe, from the examination of our pre- parations, that numerous small openings in the walls of the fistular processes lead from the outside into the central canal. Skeleton. — (a) Of the main Ijody ; as in other species of the genus there are stout columns of spiculo-fibre (PI. XLII. fig. 1, c), running vertically towards the surface, where they expand more or less. The special dermal layer (PI. XLII. fig. 1, a), composed of dense brushes of small stylote spicules, so characteristic of the genus, is almost absent over the body, being replaced functionally by large grains of sand (PI. XLII. fig. \,h), fragments of shell and so forth ; yet it can be found in the crevices between these foreign bodies. The skeleton fibres are composed of large, stout styli or subtylostyli, and in the interspaces between them numerous small tylostyli (PL XLII. fig. 1, d) are scattered, very abundant near the surface, but rare lower down. {Ij) Of the fistular projections ; here the skeleton is very regularly and symmetrically arranged ; beginning at the outside there is a very dense dermal crust (PI. XLII. fig. 3, a) of small tylostylote spicules, arranged vertically to the surface, beyond which their apices project for a very short distance. This crust is not quite continuous all over the surface, but arranged in a reticulate manner, so as to leave interspaces in which probably the pores are situate. Immediately beneath this layer comes a layer (PI. XLII. fig. 3, h) of larger spicules, lying parallel (more or less) to the surface and also arranged in a reticulate manner, so as to leave interspaces corresponding to those in the dermal crust. This layer is not very thick and immediately below it comes a series of stout, longitudinal bands of spiculo-fibre (PI. XLII. fig. 3, c), about ten or twelve in number, and arranged in a circle against its inner surface, like a series of buttresses against the inside of a wall ; then come the soft tissues, almost devoid of spicules excepting immediately around the central canal, where small tylostyli are sometimes abundantly scattered (PI. XLII. fig. 3, d). The structure of the stout fibres in the fistular processes appears to be exactly the same as in the body of the sponge. Spicules. — Megasclera; (1) large, straight styli or subtylostyli (PI. XLII. fig. 2), very gradually and finely pointed at the apex and narrowing somewhat towards the base ; size up to about 1'2 by 0'0157 mm. (2) Very small, usually slightly curved, slender tylostyli (PI. XLII. figs. 2a, 26), with well-developed heads and fairly sharp apices ; size about 0"175 by 0*004 mm. More or less intermediate forms also occur, e.g., in the layer beneath the dermal crust, but the two chief forms are very distinct from one another, 214 THE VOYAGE OF H.M.S. CHALLENGEK. the one being found mainly in the dermal crust and scattered through the soft tissues, and the other in the stout fibres. This is a very pretty and interesting little species ; it seems to be perfectly distinct from all previously known, and its habit of agglutinating comparatively large foreign objects on to the surface of the body is very peculiar, although other species may acquire an accumulation of minute particles of dirt. One function of this crust is possibly to conceal the sponge from its enemies. The skeleton arrangement in the fistular processes calls to mind the corresponding arrangement in Proteleia sollasi, but there is only a single, though more regular circle of fibres {cf. figs. 3 and 7, PI. XLIL). The number of these fibres appears to be inconstant. Zoca%.— Station 75, July 2, 1873; lat. 38° 38' 0" N., long. 28° 28' 30" W.; off the Azores ; depth, 450 fathoms ; bottom, volcanic mud. Two specimens. Genus Proteleia, Dendy and Ridley (Pis. XLIL XLIV.). 1886. Proteleia, Dendy and Kidley, Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 152. Sponge sessUe, corticate ; ujjper surface covered with mammiform processes. Megasclera tylostylote and (or) stylote ; and also spicules with grapnel-like apices pro- jecting freely beyond the surface of the sponge. This is one of the most interesting of the new genera which we have been obliged to found. It comes very close to Polymastia but differs from that genus in the possession of the grapnel-like spicule. This feature seems to approximate it to the Tetractinellida, but this question has been fully discussed by us elsewhere {vide Ann. and Mag. Nat. Hist., loc. cit.). Proteleia sollasi, Dendy and Eidley (PL XLIL figs. 6, 7, 8, 8a, 8b, 8c, 8c/, 8e, 8/ 8g, 8h ; PL XLIV. fig. 2). 1886. Proteleia Sollasi, Dendy and Eidley, Ann. and Mag. Nat. Hist., ser. 5, vol. .wiii. p. 153, pi. V. Sponge (PL XLIV. fig. 2) sessile, apparently coating, consisting of a fiattened, cake-like expansion with slightly convex upper surface, from which arise abruptly numerous short, thick, cylindrical, mammiform projections of various sizes. The single specimen in the collection is about G3 mm. long by 31 mm. broad, and about 13 mm. thick. The mammiform processes vary somewhat in size, being, when full grown, about 8 mm. long by 4 mm. in diameter at the base; they are almost solid and very stiff and firm, contrasting strongly with those of Polymastia rohusta and Poly- mastia mammillaris in this respect, at present they are all closed at the summit, and REPORT ON THE MONAXONIDA. 215 it is doubtful whether or no any opening exists in the living sponge, though traces of what appears to be such can be found. Colour in spirit yellowish-grey. Texture tough and leathery, internally coarsely fibrous ; the cortex is very firmly adherent to the under- lying tissues. Surface between and on the mammiform processes even, seen in sections to be minutely hispid ; the hispidity is more strongly marked over the body than on the mammiform processes ; on the body there is a considerable amount of foreign matter collected, while the mammiform processes are free from it and almost glabrous in appear- ance. Oscula (? minute, at the summits of the papillse). Pores scattered (? singly) over the surface of the body and of the mammiform processes. Skeleton. — (a) Of the main body (1) a thin, very dense and compact external layer (PI. XLII. fig. 6, b), about O'lS mm. thick, composed of vertically placed, tightly packed, smaU, straight and slender tylostylote spicules, with their apices directed outwards and projecting for a short distance beyond the surface of the sponge. (2) Immediately below the above and inseparable from it a similar but very much thicker layer (PL XLII. fig. 6, c) of larger, stout tylostylote spicules, arranged as in the first layer and with their apices embedded in it, thickness about 0-35 mm. These two layers may be considered as cortical. Besides the spicules already mentioned, there are in the cortex spicules of another and very remarkable kind, the grapnel-spicules (PI. XLII. fig. 6, d) to be described later. These have the base and part of the shaft embedded in the cortex while the remainder of the spicule projects freely for a considerable distance beyond the surface of the sponge, and bears at its extremity the grapnel. Immediately below the cortex, as above defined, comes a layer (PI. XLII. fig. 6) about as thick as the second cortical layer, of still larger, stout tylostylote spicules, not vertically disposed Init for the most part horizontally and irregularly, forming a compact mass. Below this layer comes the general parenchyma of the sponge (choanosome), with very numerous scattered tylostylote spicules, and with very well defined, stout fibres (PI. XLII. fig. 6, a), composed of large stylote or sul)tylostylote spicules longitudinally placed and with their apices outwardly directed. These primary fibres run vertically towards the surface of the sponge ; before arriving there they expand into divergent brushes of large spicules whose apices penetrate right into or even through the cortex. Secondary skeleton fibres, if present at aU, are very ill defined, (b) The skeleton of the mammiform jDrocesses (PI. XLII. fig. 7); the cortex and the layer immediately below it are arranged very much as in the main body, except with regard to the grapnel-spicules, which seem to be entirely absent ; then come very definite, stout, longitudinally placed bundles of spiculo- fibre (PI. XLII. fig. 7, a), like those of the main body and arranged mainly, and fairly regularly, in two concentric circles, and with the spaces between them filled with a great number of irregularly but closely arranged tylostylote spicules ; in the centre of the inner circle of fibre bundles is a space almost quite free from spicules and filled with a yellow, granular substance. 21(3 THE VOYAGE OF 11. M.S. UHALLENGEfl. Spicules. — Megasclera; (1) small, slender, very slightly curved, sharply and gradually pointed tylostyli (PI. XLII. figs. 8^, 8/i) with not very well-develoj)ed oval heads, size about 0"16 by 0"0045 mm.; these spicules occur in the outermost layer of the cortex. (2) Much larger, very stout, sharply pointed, fusiform tylostyli (PI. XLII. figs. Be, 8/), with roundish heads ; size variable, about 0"22 by 0"0189 mm.; occurring in the lower cortical layer and passing gradually by spicules intermediate in form and size into (3) the long styli (PI. XLII. fig. 8) of the fibres ; these are smooth, straight, fusiform and sharply and gradually pointed at the apex; size about 1'2 by 0'03 mm. (4) The grapnel spicules (PL XLII. figs. 86, 8c, 8rf); long, very slender, with more or less expanded base and tapering very gradually to hair-like fineness towards the apex, ending finally in a small knob provided with recurved teeth which do not seem to be quite constant in number ; commonly there are about four teeth, but owing to the minute size it is not easy to make out details; sometimes the teeth are absent, leaving only the knob. Length of spicule about 0"52 mm., thickness at thickest part of shaft about 0'0063 mm. Locality. — Simon's Bay, Cape of Good Hope; depth, 10 to 20 fathoms. One specimen. Genus Trichostemma, M. Sars (PI. XLIIL). 1872. Tricliostemiiia, M. Sars, Eemarkable Forms of Animal Life, pt. i. p. 62.' 1870. Radiella, Schmidt, Spong. Atlant. Gebiet., p. 48 (not Radiella, Sollas, Ann. and JNlag. Nat. Hist., ser. 5, vol. ix. p. 162). Sponge free living, with definite, symmetrical, external form ; discoid or hemi- spherical ; with a marginal fringe of long, hair-like spicules serving to maintain the sponge in its position in the mud. Oscula, one or more at the ends of short oscular tubes on the upper surface. Megasclera mainly tylostylote. The original diagnosis {Joe. cit., p. G5) runs : — " Spongia silicea, simplex, libera, in limo demersa et hie fimbria spiculorum setiformium, fiexibiHum, radiantium sustentata, corticc crasso compacto circumdata, iuteriore multo minus compacto, parenchymatoso, lacunis numerosis irregularibus trajecto. Oscula numerosa tubiformia in facie superiore libere prominentia. Sceletum ex spiculis acuformibus fasciculato-radiantibus compositum, aliis brevissimis capite globoso in cortice densissime accumulatis, aliis multo longioribus. fasciculos tenues parenchyma et corticem trajicientes formantibus. " The original type of the genus is Trichostemma hemisphiericum, Sars, which " occurs not rarely at Lofoten .... in a depth of 120-300 fathoms on soft clay bottom." The Challenger adds two new species, both from a very great depth and a bottom of ooze or mud. It is essentially a deep-sea genus, and forms another example of the manner in which deep-sea sponges commonly assume a definite, symmetrical external form ; in this ' The gonus and species are first described in this place as nov. gen. et spec, but the name is used by Sars so far back as 1869 (Forhaiidl. Vitleiisl: Selnk. Cliristiania, 1869, p. 250), without, however, any doscriptionj REPORT ON THE MONAXONIDA. 217 case, however, the object of the flattened form and the long radiating spicules is obvious, namely, as pointed out by Sars, to support the animal in the soft mud on which it lies ; in our new species, Trichostemma sarsii, this arrangement is brought to a much greater degree of perfection than in the original type of the genus [vide PI. XLIII. figs. 1, 1«, 2, 2«, 3, 3a, and woodcut, Fig. 8). The genus has a very wide geographical range, being found in deep water off the north of Scotland (Carter), ofi" the coast of Norway (Sars, Hansen), Gulf of St. Lawrence (Whiteaves), off the Azores (Challenger), ofi" the north-east coast of Australia (Challenger), off the west coast of South America (Challenger), and in the Arctic Sea (Vosmaer). The question as to whether Schmidt's genus Radiella is really the same as Tricho- stemma is a very difficult one. His generic diagnosis is not very recognisable ; it runs as follows: — " Suberiten mit radiarer Schichtung derNadeln; ohne Wurzeln und ohne wahre Einde, d. h. ohne Einde mit Faserelementen. Oscula vorhanden."^ The type of his genus, Radiella sol, is undoubtedly extremely like our Trichostemma sarsii, and the fact of its being also found in deep water (Cuba, 638 fathoms) perhaps strengthens the Fig. 8. — Trichostemma sarsii. Vertical section, showing the arrangement of the skeleton ; /, fringe of long, projecting spicules ; c, cortical layer of vertical spicules ; d, external layer of foreign objects, dirt, &c.; ch, choanosome, containing stellate groups of spicules ; th, " thatch " of spicules on the lower surface of the sponge. x 5. probability of their being closely allied. According to Schmidt's figure- the external form of the sponge is almost identical with that of Trichostemma sarsii, but judging by his description and figure there is no superior cortical layer and no internal confused mass of spicules, the smaller spicules instead running continuously from the lower to the upper surface of the sponge, beyond which they project "uumerklich." The thatch-Hke coat of radiating spicules on the lower surface of the sponge seems to be the same as in Trichostemma sarsii. In form also the spicules of the two species seem to be about the same, but Schmidt gives no measurements. On the other hand Hansen^ states that he has examined an original specimen of Radiella sol and identifies it with Trichostemma hemisphwriciim. Considering, therefore, that the name Trichostemma was used by Sars so far back as 1 Spong. Atlant. Gebiet., p. 48. 3 Norwegian North-Atlantic Expedition ; Spongiadfe, p. 7. (ZOOL. CHALL. EXP. — PART LIX. — 1887.) 2 Loc. cit, pi. iv. fig. 6. Nnn 28 218 THE VOYAGE OF H.M.S. CHALLENGER. 1869 (loc. cit.), though his generic diagnosis does not appear till 1872, and that Schmidt's diagnosis of the genus Radiella is highly unsatisfactory, we think ourselves fully justified in retaining Sars' generic name, although the two seem to be probably synonymous. Upon this latter point, however, we should be very doubtful were it not for Hansen's precise state- ment, for the absence of the upper cortical layer of spicules in Radiella would be a very strong point. Another sponge with almost identical external form is Bowerbank's Halicnemia patera,^ Ijut the genus Halicnemia is at once differentiated from Tricliostemma by the presence of peculiar spicules in the dermal membrane, which Bowerbank tells us are "very numerous, dis- persed, fusiformi-acerate, entirely spin ed, subangulated, and frequently infl ated at the middle. " Probably the peculiar external form of these sponges is comparable to the " Crinorhiza " form found amongst the Desmacidonidse, and may exist in very different genera when living under similar conditions ; another instance being the little known Xenospongia 2oatelliformis, Gray, which only wants the marginal spicular fringe to complete the external resemblance, but has a totally distinct spiculation (including stellate forms).- Tricliostemma sarsii, Eidley and Dendy (PI. XLIII. figs. 1, la, 2, 2a, 3, 3o). 1 886. Tricliostemma Sarsii, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii.p. 488. Sponge (PI. XLIII. figs. 1, la, 2, 2a, 3, 3a,) discoid, much flattened and usually thin. Flattened more especially on one side, which bears an oscular tube in its centre, and must therefore be regarded as the upper one. The lower side appears to be always more or less convex but usually only slightly so ; in the single specimen from Station 184, however, it is strongly convex. The margin of the disk is thin, and provided with a beautiful fringe of very long, silky-looking spicules. The largest specimen is about 13 mm. in diameter (exclusive of the fringe of spicules, which is 3 mm. broad), and only just over 3 mm. in thickness at the centre. The smallest specimen has a diameter of about 5 mm. (exclusive of the fringe of spicules, which is again about 3 mm. broad). The diameter of the Australian specimen is about 8 mm. (excluding fringe), and the thickness in the centre 4 mm. On holding a specimen up to the light and looking at the flat side one sees a number of small, round, translucent patches, about 1 '5 mm. in diameter, and arranged for the most part in a circle, a short distance within the margin of the sponge {cf. PI. XLIII. fig. 2, &c.). The nature of these is very doubtful ; we were at first inclined to regard them as pore-areas, but there is hardly sufficient evidence to 1 Mon. Brit. Spong., vol. i. p. 184, pi. x. figs. 228-233, pi. xxsii. figs. 363, 364 ; vol. ii. p. 96 ; vol. iii., pi. xv. figs. 31, 32 ; vol. iv. p. 45. 2 Concerning the relations of Tricliostemma to Polymastia we prefer at present to keep silence. Vosmaer, however (Sponges of the " Willem Barents " Expedition 1880-81, p. 12), makes the two genera identical, but, though there is much to be said in favour of this view, we are, as already indicated, not yet satisfied about the matter. We cannot, how- ever, at all understand how Vosmaer {loc. cit.) has come to the conclusion " that there is no generic difference between Halicnemia and Polymastia," and also, of course, Tricliostemma, in spite of the difl'erences in spiciilation ; a conclusion at which also von Marenzeller had previously arrived {Denkschr. d. k. Akad. Wiss. JVien, Bd. xxxv. p. 371). EEPORT ON THE MONAXONIDA. 219 support this view. Colour in spirit greyish -yellow. Texture firm. Surface fairly even. Oscula; one {1 sometimes more) at the top of an oscular tube in the centre of the flat surface. Pores (?). Skeleton. — The convex (lower) surface of the sponge is covered with a dense thatch {vide woodcut, Fig. 8, th), of very long tylostylote or subtylostylote spicules, which, radiating outwards and upwards from the centre, form, round the margin of the sponge, the spicular fringe already mentioned. On the flat (upper) surface there is no such thatch, but there is a thick ectosome (woodcut. Fig. 8,c) packed with tylostylote spicules, which are for the most part vertically arranged. On the outside of this ectosome is collected a thick stratum of foreign matter (woodcut, Fig. 8, d), through which the apices of the tylostylote spicules often project for a short distance. Below the ectosome come the softer tissues of the sponge (choanosome, vide woodcut, Fig. 8, ch), which are crowded with vast numbers of closely packed, much shorter, tylostylote spicules ; these have no very definite order except that they exhibit a tendency to an arrangement in thick, radiating groups with the globular heads in the centre. Spicules. — Megasclera ; (l) the exceedingly long, straight, slender, tylostylote or sub- tylostylote spicules forming the thatch and fringe. It is not easy to give the maximum length of these, owing to the manner in which they break on attempting to separate them for purposes of measurement, but we have found examples about 4 '7 mm. long, and this seems to be no uncommon size ; their diameter is commonly about 0'02 mm.; the heads are generally subglobular and not very strongly developed, and at the apex the spicule tapers very gradually to excessive fineness. (2) The spicules of the choanosome ; these are short, straight, stout, fusiform tylostyli, sharply pointed at the apex and with very well developed, globular heads ; size about 0"3 by 0'016 mm. (In the Australian specimen they are shorter than this.) (3) The spicules of the ectosome ; these are also tylostylote or subtylostylote and are intermediate in size between the two forms above described. This species difiers markedly enough in external form alone from the original type of the genus, Trichostemma hemisphsericum, Sars,' to be distinguished at a glance, for Sars' species is very much larger, strongly hemispherical, and has the convex surface uppermost, and provided, in the adult, with many oscula ; while the fringe of spicules slopes down- wards, i.e., away from the osculum -bearing surface. The difi'erences between Triclio- stemma sarsii and Trichostemma irregularis are indicated at the end of the description of the latter species. The geographical distribution of Trichostemma sarsii is a very interesting one ; it seems to be confined to very deep water, and was met with off" the Azores and in Australian seas, thus giving a very wide range. There are, it is true, certain slight differences between the specimens from the two localities, which have been indicated in the description, but there is nothing to justify us in separating them specifically. 1 Remarkable Forms of Animal Life, pt. i. p. 62. 220 THE VOYAGE OF H.M.S. CHALLENGER. Localities.— Station 73, June 30, 1873 ; lat. 38° 30' N., long. 31" 14' W.; off the Azores; depth, 1000 fathoms; bottom, Pteropod ooze; bottom temperature, 39°'4. Five specimens. Station 184, August 29, 1874; lat. 12° 8' S., long. 145° 10' E.; south-east of Cape York; depth, 1400 fathoms; bottom, Globigerina ooze; bottom temperature, 36°"0. One specimen. Trichostemma irregularis, Ridley and Dendy (PI. XLIII. figs. 4, 4a, 5). 1886. Trichostemma irregularis, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 489. There are in the collection two specimens of this species which differ considerably from one another in external appearance. The one (PL XLIII. figs. 4, 4a), which is probably the normal form, has its parts arranged very much as in Trichostemma sarsii. In shape it closely resembles an inverted mushroom without the stalk. The convex, lower surface is protected by a thatch-like covering of very large spicules, which radiate upwards and outwards from its centre, and form a thick, irregular fringe round the free margin. The upper surface is slightly concave, strongly hispid, and bears a considerable accumulation of foreign dirt. It also bears a few, irregularly scattered, small, jDapilliform processes which may be oscular tubes. The specimen is about 1 7 mm. in diameter and 8*5 mm. thick in the centre. (The second specimen (PL XLIII. fig. 5) appears to be abnormal in shape ; it is of about the same size as the first, but the upper surface is strongly convex and the lower slightly concave, and the long, projecting spicules are also very irregularly arranged, projecting almost at right angles all over the lower surface.) Colour in spirit dirty yellow. Texture fairly firm but rather spongy. Loiver surface, in normal specimen, radiately fibrous, upper surface strongly hispid. Oscula (? several, on the upper surface, on the summits of papilliform projections). Skeleton. — Normally arranged as in Trichostemma sarsii; the projecting spicules of the ectosome, on the upper surface, however, project for a very much greater distance than in that species, the portions outside the sponge being long and of hair-like fineness, sometimes even flexuous. Spicules. — Megasclera ; all either tylostylote or subtylostylote, but of most variable length ; the spicules of the ectosome and of the interior of the sponge are much larger and slenderer than in Trichostemma sarsii ; the former often projecting for as much as 1 mm. beyond the surface, whUe the latter commonly measure about 0"5 by 0"012 mm., but are often much larger. The large external spicules seem to be of about the same size as in Trichostemma sarsii ; their heads are very feebly developed. In appearance this species is very much clumsier and coarser than the foregoing, the upper cortical and internal spicules are larger, and there seems to be more than one osculum. REPORT ON THE MONAXONIDA. 221 Locality.— Station 299, December 14, 1875 ; lat. 33° 31' S., long. 74° 43' W.; west of Valparaiso; depth, 2160 fathoms; bottom, blue mud; bottom temjjerature, 35° '2. Two specimens. Genus Tentorium, Vosmaer. 1870. Tliecophora, Schmidt, Spong. Atlant. Gobiet., p. 50. 1885. Tentorium, Vosmaer, Bronn's Klass. v.. Ordn. d. Thierreichs, Porifera, p. 329. Sponge sessile, columnar or conical ; protected by a dense, cylindrical sheath of longitudinally placed spicules, which form a solid, imperforate layer. Over the top there is a proper fibrous cortex containing bundles of smaller spicules arranged vertically to the surface, between which are situate the large, elongated subdermal cavities. Pores on the upper surface only. Oscula tubular, in the centre of the upper surface ; commonly only one. Spicules tylostylote or subtylostylote. Schmidt gives no generic diagnosis, but Vosmaer, in addition to pointing out that the name Thecophora is already occupied, gives the following: — " Cylindrischer mit breiter Basis angewachsener Korper. Oben sind kleiue papillenartige Oscula. Riude besonders deutlich oben. Spic. indie. tr° ox;, f. von verschiedener Grosse." Only a single species of the genus is as yet established. Tentorium semisuherites, Schmidt, sp. 1870. TlieeoiJlwra semisuherites, Schmidt, Spong. Atlant. Gebiet., p. 50, pi. vi. fig. 2. . 1873. Thecophora ihla, Wyville Thomson, Depths of the Sea, p. 147, fig. 24. 1877. Thecophora elongata, von Marenzeller, Die Cffilenterateu, Echinodermeu und Wiirmer der K. K. Osterreichisch-ungarischen Nordpol-Expedition, jj. 12, pi. ii. fig. 4. 1885. Tentorium semisuherites, Vosmaer, Bronn's Klass. u. Ordn. d. Thierreichs, Porifera, p. 329, pL ii. fig. 4; jjl. xxi. fig. 19. This very interesting and well-characterised sponge is represented in the collection by a fine series of ten specimens from Station 49, and a single very fine specimen, attached to a stone, from Station 50. There are also four specimens, much smaller than usual and characterised by correspondingly smaller spicules, from Inaccessible Island, which we propose to consider as belonging to a dwarf variety, although the possibility of their being young forms must be borne in mind. Hansen ^ has already given Thecophora ihla and Thecophora elongata as synonyms of the original Tentorium semisuherites and we are quite prepared to follow him in this. The difference between the three forms depends very largely on the degree of contraction of the specimen ; when the specimen is in an expanded condition, the top appears evenly rounded, as in Schmidt's original figure and 1 The Norwegian North-Atlantic Expedition, 1876-1878, Spongiadse, p. 8. 222 THE VOYAGE OF H.M.S. CHALLENGER. in fig. 23 {Thccophora semisuberites) of the Depths of the Sea; when, on the other hand, the animal is contracted, the top of the sponge, owing to the arrangement of the brushes of spicules immediately below the surface, becomes uneven and " scaly " in appearance as in fig. 24 {Thecophora ihla) of the Depths of the Sea. Evidently, when living, the sponge possesses great power of contractility, a power which would appear to be shared in large measure by the oscular tube. Vosmaer, in his Report on the Sponges of the " Willem Barents " Expedition of 1880 and 1881 (p. 18), has given some account of the minute anatomy of this species, and some observations on this subject are also given in the Introduction to the present work. Zoca/iii'es.— Station 49, May 20, 1873; lat. 43° 3' N., long. 63° 39' W.; south of Nova Scotia; depth, 85 fixthoms ; bottom, gravel, stones; bottom temperature, 35°'0. Ten specimens. Station 50, May 21, 1873; lat. 42° 8' N., long. 63° 39' W.; south of Nova Scotia; depth, 1250 fathoms; bottom, blue mud; bottom temperature, 38°'0. One specimen. Inaccessible Island, October 16, 1873 ; depth, 60 to 90 fathoms. Four small specimens of a dwarf variety. Habitat. — Arctic Seas (Schmidt, Vosmaer, &c.) ; North Atlantic (Carter, &c.) ; Gulf of St. Lawrence (Whiteaves) ; Newfoundland (Smith and Harger) ; off Nova Scotia (Challenger) ; Inaccessible Island (Challenger). Genus Stylocordyla, Wyville Thomson (PI. XLIIL). 1869. Hyalonema (pars), Lov^n, Ofversigt k. Vetensk.-Akad. Fiirhandl., vol. xxv. p. 119. 1873. Stylocordyla, Wyville Thomson, The Depths of the Sea, p. 113. 1876. Polymastia (pars), Carter, Ann. and Mag. Nat. Hist., ser. 4, vol. xviii. p. 393. 1880. Stylorhiza, Schmidt, Spong. Meerb. von Mexico, pt. ii. p. 79 ; pi. x. fig. 5. 1881. (?) Rldzaxinella, Keller, Arch. Micr. Anat., vol. xviii. p. 271. Sponge corticate, differentiated into distinct head and stalk. Skeleton in head radiately arranged, with a cortical layer of smaller spicules set vertically to the surface ; skeleton in stalk consisting chiefly of a dense axis of longitudinally arranged spicules. Spicules oxeote only. This genus still stands aloof from all other Suberitidse in its remarkable spiculatiou, though, as regards external form, the old distinction embodied in the term Stylocordyla no longer holds good, for similar stipitate forms are found to occur in the genus Suherites {e.g., Suherites ramulosus, nobis). Only one species of the genus is as yet established satisfactorily (see sjmonyms and references below). REPOET ON THE MONAXONIDA. 223 Stylocordyla stipitata, Carter, sp. (PI. XLIII. figs. 6, 7, 8, 9). 1873. Stylocordyla horealis} Wyville Thomson, The Depths of the Sea, p. 113. fig. 13. 1876. Polymastia stipitata, Carter, Ann. and Mag. Nat. Hist., November 1876, p. 393. 1880. Stylorhiza stipitata, Schmidt, Spong. Meerb. von Mexico, pt. ii. p. 79, pi. x. fig. 5. (See also Vosmaer, The Sponges of the " Willem Barents" Expedition, 1880 and 1881, p. 11.)^ There are three small specimens (PI. XLIII. figs. 7, 8, 9) of this sponge from Station 147, one (PI. XLIII. fig. 6) from Station 49, and one from off Bahia ; two of the specimens have oval heads only about 4 mm. in longest diameter, while the two largest specimens have heads more of the typical shape, as described by Carter for adult specimens ; even the latter are, however, very small as compared with Carter's types in the British Museum, the largest head (from Station 147) being onlj'' about 11 mm. long by about 4 mm. in diameter at the widest part, where there is a whorl of projecting spicules. The external form of this sponge is very variable. Schulze,^ as well as Carter,* has already noticed that the head in young forms is more or less round ; we have to notice below a variety in which the head is globular even in the adult. We have detected no important difi'erence in spiculation between Mr. Carter's types and the Challenger specimens. It appears to us highly probable that Loven's Hyalonema horeale^ is really the same species as Carter's Polymastia stiintata, in spite of the fact that the larger oxeote spicules in the former sponge are described as having a central inflation, a character which may perhaps be considered as abnormal, for Loven had only two specimens for examination. Still we are not as yet convinced of this identity. The geographical distribution of this sponge is a very wide and remarkable one, as will be seen by the list of localities below, and its bathymetrical range is no less extraordinary. Zoca^{«{es.— Station 49, May 20, 1873; lat. 43° 3' N., long. 63° 39' W.; south of Halifax, Nova Scotia; depth, 85 fathoms; bottom, gravel, stones; bottom temperature, 35°'0. One specimen. Station 147, December 30, 1873; lat. 46° 16' S., long. 48° 27' E.; between Marion and Crozet Islands; depth, 1600 fathoms; bottom. Diatom ooze ; bottom temperature, 34 "•2. Three specimens. Off Bahia ; depth, 7 to 20 fathoms. One very small specimen with nearly globular head. 1 Loven's species {vide infra) with which Sir Wyville Thomson identified Carter's Polymastia stipitata. ^ As we are not convinced of the correctness of Vosniaer's views as to the synonymy of this species as here given, we prefer to make use of the specific name given by Mr. Carter, with whose types we identify our specimens. 3 Exploration of the Faroe Channel, p. 71, Proc. Roy. Sac. Edin., 1881-1882. * Loc. cit., p. 393. ■'' Ofversigt k. Vetensk.-AJcad. Forhandl., 1888 (pub. 1869) p. 105, pi. ii., see also translation of this paper in Ann. and Mag. Nat. Hist., ser. 4, vol. ii. p. 81, pi. vi. 224 THE VOYAGE OF H.M.S. CHALLENGER. Habitat. — Between north of Scotland and Faroe Islands (Carter, Schulze) ; Grenada (Schmidt) ; Southern Ocean (Challenger) ; south of Nova Scotia (Challenger) ; Bahia (Challenger). Stylocordyla stipitata, var. glohosa, Ridley and Dendy (PI. XLIII. fig. 10). 1886. Stylocordyla stipituta, var. glohosa, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 489. We unite under the above name a large series of specimens from Kerguelen Island and three small ones from Station 145. This variety is distinguished from the types of Fig. 9.— Stylocordyla stijntnia, var. glohosa. Sections showing the arrangement of the skeleton. A, longitudinal section of a specimen which exhibits no spiral arrangement of the fibres, x 6. A', transverse section of do., x 4. B, transverse section of a specimen in which the fibres are spirally disposed, x 4 ; o, osculura ; c, cortical spicules ; a, stem ; cm, central mass of spicules derived from the breaking up of the stem ; s, sections of do. ; r/, the radiating skeletal fibres. the species by the almost perfectly globular, bullet-like head of the adult sponge (PI. XLIII. fig. 10), which, though it may become slightly oval, never exhibits the truncated appearance shown by adult specimens of the typical form, and never possesses a whorl of projecting spicules below the osculum, the entire surface of the head being evenly and minutely hispid. The stem also is not nearly so thick and fleshy-looking as in Carter's larger specimens, and has not nearly so large a conical basal expansion ; as in the types, there are a number of fine attaching rootlets. We know of no characters by which it would be possible to distinguish the young forms of the variety from those of the si^ecies proper. The remarkable spiral arrangement of the skeleton in the head of this sponge has REPORT ON THE MONAXONIDA. '225 * already been noted by earlier authors. It is extremely well shown in the present variety, and we have thought it desirable to give a woodcut illustrative of it {vide woodcut, Fig. 9, B). This has been executed for us by Mi-. Highley with his usual skill and accuracy, and renders further description unnecessary. It is, however, important to notice that in one of the specimens examined by us the skeleton was simply radiate, without any spiral twist at all ; this is also shown in the figure (Fig. 9, A, A') contrasting strongly with the normal spiral arrangement. We must also note that this particular specimen contained large quantities of embryos in various stages of development ; it is just possible that these two facts may be correlated, but it is very difiicult to see how. There are in the collection no less than fifty-three specimens which we refer to this variety ; they vary much in size ; the largest has a stem 75 mm. long, surmounted by a head about 17 mm. in diameter. We have never found more than a single osculum, which is situated at the top of the head. The specimens are in an exceUent state of preservation and details as to their minute anatomy and histology will be found in the Introduction. Neither in specimens of the species proper nor yet of the variety have we detected any branching of the stem, and this character, as well as those of the spiculation, &c., distinguishes Stylocordyla stipitata, Carter, from Suberites ramulosa, nobis, which somewhat resembles it in external form. Localities.— Btsition 145, December 27, 1873; lat. 46° 43' 0" S., long. 38" 4' 30" E.; dejjth, 140 fathoms; bottom, volcanic sand. Three small specimens. Off Kerguelen ; depth, 10 to 100 fathoms. Fifty specimens. Genus Quasillina, Norman. 1869. Quasillina, Norman, Last Report on Dredging among the Shetland Isles, Brit. Assoc. Rep. for 1868, p. 329. 1875. Bursalina, Schmidt, Jahresb. d. Comm. wiss. Untersuch. d. deutsch. Meere, Jahrg. ii., iii. p. 116. Sponge corticate, stipitate, with oval body, bearing a single osculum at the summit, and short stalk. In the cortex primary skeleton fibres ascend in parallel lines from the base, crossed at right angles by secondary ones. Spicules, large and small styli. Dr. Norman's diagnosis runs as foUows : — " Sponge consisting of a single elavate hoUow body, widening upwards from the base, and rising at once from the surface of the stone to which it is attached, without any expanded basal mass. Skeleton beautifuU}' reticulate, primary fasciculi ascending in parallel straight lines from the base, and in diverging, radiating lines from a central mammaeform projection at the summit of the sponge ; secondary fasciculi at right angles to the primary ones. Spicula fusiformi-acuate." We cannot agree with Bowerbank in placing the species Quasillina brevis in the genus Polymastia ; it appears to us to difi"er from it very widely, and Dr. Norman has done good service in erecting a new genus for its reception. Schmidt cannot have been aware of (zooL. OHALL. EXP. — PAET Lix. — 1887.) Nnn 29 226 THE VOYAGE OF H.M.S. CHALLENGER. the existence of Norman's genus when, in 1875, he founded his genus Bursalina for a species which appears to be identical with Quasillina hrevis ; he does not venture upon a generic diagnosis, and, unfortunately, gives no spicular measurements. Vosmaer ^ has given a discussion of the genus and species, and has also given some description of the minute anatomy. His specimens seem to have been in much better condition than the Challenger ones. There can be no doubt that the body is not hollow in life, but the soft internal tissues generally shrink up and disappear, or liquefy and run off, after the death of the animal, thus giving „ ,„ ^ ,„, , to the sponge its characteristically hollow form. Fig. 10. — QimsiUtiia brevis, J. o j thl^'SimT^Nat'Tize^ '^^^ commou cxistencc of a single osculum at the summit of the sponge {vide woodcut, Fig. 10, o), a point of considerable interest, is demonstrated by the Challenger specimens. (Vosmaer states that he " never saw an opening on the top larger than those where the sea-water enters," but there can be no doubt that in life there is always an osculum at the summit of the body.) Quasillina hrevis, Bowerbank, sp. 1861. Eiiplectella brevis, Bowerbank, List Brit. Marine Invert. Fauna (Brit. Assoc), p. 71. 1864. 1 Polymastia rohusta, Bowerbank, Men. Brit. Spong., vol i. pp. 178, 28.5, pi. xxix. fig. 358.2 1866. Polymastia hrevis, Bowerbank, Mon. Brit. Spong., vol. ii. p 64 (see also vol. iii. p. 25, pi. xi. figs. 1-9 and vol iv. p. 31). 1869. Quasillina hrevis, Norman, Last Eept. Dredg. Shetland Isles (Brit. Assoc. Rep. for 1868), p. 329. 1875. Bursalina muta, Schmidt, Jahresb. d. Comm. wiss. Untersuch. d. deutsch. Meere, Jahrg. ii., iii. p. 116. The Challenger obtained fourteen specimens of this very remarkable sponge, nearly all attached to pebbles by a stem or peduncle, and all from Station 49. Many of the specimens have a single small osculum at the summit [vide woodcut. Fig. 10, o) ; all appear to be hollow, containing in the large internal cavity more or less of a yellow amorphous substance caked on to the body-walls. The spicules are, as usual, large and small styli, the former measuring about 1-1 by 0*02 mm. and the latter about 0-28 by O'Ol mm.; they agree very fairly with those of Bowerbank's species, as shown by his preparations in the British Museum. It is doubtful whether the sponge described and figured by Schmidt ^ from the Gulf 1 Sponges of the " Willem Barents" Expedition, 1880 and 1881, p. 20. 2 There has evidently been some mistake here, we give this synonym because fig. 358, pi. xxix. vol. i. of Bower- bank's Mon. Brit. Spong. is described in the first volume (pp. 178, 285) as Polymastia rohusta, but is referred to in the third (p. 25) and fourth (p. 31) volumes as Polymastia brevis. ' Spong. Meerb. von Mexico, p. 70, pi. x. fig. 4. REPORT ON THE MONAXONIDA. 227 of Mexico, under the name Bursalina muta, var., is really Quasillina at all ; the external form, as figured by him, is somewhat different, and he has not determined, apparently through fear of injuring the specimen, whether a large central cavity is present or not ; this being, though only a post-mortem condition, very characteristic of the species. The species appears to be widely distributed and not uncommon, especially in northern latitudes. Zoca%.— Station 49, May 20, 1873; lat. 43° 3' N., long. 63° 39' W.; south of Halifax, Nova Scotia ; depth, 85 fathoms ; bottom, gravel, stones ; bottom temperature, 35° '0. Fourteen specimens. Habitat. — Shetland (Bowerbank, &c.) ; south-west coast of Norway (Schmidt) ; south of Nova Scotia (Challenger) ; ? Gulf of Mexico (Schmidt) ; Mediteri'anean, ofi" Benzert, North Africa (specimens in British Museum) ; Arctic Ocean (Vosmaer). Genus Clio^ia, Grant (Pis. XXV., XXIX.). 1826. Cliona, Grant, Edinburgh New Philosophical Journa], vol. i. p. 78. 1833. Vioa, Nardo, Isis, p. 523. 1866. Hymeniacidon (pars), Bowerbank, Mon. Brit. Spong., vol. ii. p. 212. Suberitidfie of boring habit. Spicules tylostyli. Our knowledge of the boring sponges and their classification is perhaps in a more unsatisfactory condition than our knowledge of any other group of like importance amongst the Monaxonida. In the first place it appears to us that far too much stress has been laid on the mere boring habit as a character of classificatory importance, and that it is a mistake to group together all the boring sponges as necessarily belonging to even the same subfamUy, while it is quite unnecessary to erect a special group for their reception, like the " Ecccelonida " of Mr. Carter.' It appears, however, that nearly all the boring sponges as yet known belong to the Clavulina, but some belong to the family Suberitidse, while others belong to the Spirastrellidse. The name Cliona must be confined to the former group, i.e., to the species with no microsclera. This is the only genus here dealt with, and so we shall not in this place discuss the matter further, merely stating that Cliona is a genus of Suberitidse which has taken to boring habits. Cliona dissimilis, Eidley and Dendy (PL XXV. figs. 5, 5a, 5b, 5c, 5d; PI. XXIX. fig. 8). 1886. Cliona dissimilis, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 490. Sponge (PI. XXV. figs. 5, 5a) encrusting and boring into a Madreporarian Coral {Turbinaria sp.). The coraUum of the Turbinaria forms a tabular lamella about 13 mm. thick, now broken into several large pieces. This is coated on both surfaces by a very thin 1 Journ. Roy. Micr. Soc, vol. ii. p. 496. 228 THE VOYAGE OF H.M.S. CHALLENGER. layer (ectosome) of the Cliona, and is also penetrated through and through by the .sponge (choanosome). The tubular canals penetrating the Coral in every direction ramify and anastomose and appear on the surface at fairly regular intervals in the form of circular, vertical pits.' Each of these pits is closed at the top by a thick, cushion -like pad of sponge tissue. These cushions are merely portions of the general crust which the Cliona forms on the surface of the corallum, Ijut whereas the crust is in most places only about 0"2 to 0'25 mm. thick, in the cushion -like areas it attains a thickness of about 0"4 mm. on the one side of the corallum, and 0'8 mm. on the other {vide annexed woodcut). On the one surface, again (PL XXV. fig. 5), the pits are very much more abundant than on the other (PI. XXV. fig. 5a), and average about 2 mm. in diameter, while the distance between two adjacent pits is also about 2 mm. On this surface the cushions are also much thicker than on the other i^vide supra), and they present no appearance of perforation to the naked eye, though there is commonly a slight de- pression in the centre (PI. XXV. fig. 5c). On the opposite surface the pits are fewer and less regularly arranged, and the cushion-like pads of tissue have each a small, but distinct, oscular perforation through the centre (PI. XXV. fig. bh). Colour of the sponge in spirit light brown. Oscula minute (PI. XXV. fig. 56, o and woodcut Fig. 11, o), one in the centre of each cushion-like area on one surface only of the sponge. Pores, narrow, slit-like perforations, between vertical brushes of spicules, many in each cushion-like area on the opposite side of the sponge (woodcut. Fig. 11, ^a). Skeleton. — The ectosome contains numerous, closely packed, tylostylote spicules, many (? most) of which are arranged more or less vertically to the surface, beyond which their apices project for a short distance. In the cushion-like areas on both surfaces the spicules are especially abundant and more regularly placed than in other parts of the ectosome ; around each osculum they form a slightly projecting fringe {vide woodcut, Fig. 11). The skeleton in the main body of the sponge, i.e., in the canals in the corallum, is very loose and scanty, consisting only of a few scattered sjjicules. Spicules. — Megasclera ; of one form only, viz., rather slender tylostyli (PI. XXIX. fig. 8), with very well marked heads ; size about 0'32 by 0"0065 mm. This is an extremely interesting and well characterised species ; the most important points about it are (1) that it completely envelops as well as bores into the organism 1 Originally the calicular cavities of the Coral, of which the sponge has taken advantage. Fig. 11.— Cliona dismnilis. Vertical section of ,i decalcified specimen ; 0,0, oscula; pa, pore-areas, c, ectosome ; ic, inhalent canals ; ec, exhalent canals, x 4. REPOET ON THE MONAXONIDA. 229 which it inhabits ; (2) that the oscula are confined to one surface and the pores to the other, exactly as in so many other sponges which have a lamellar form. The most closely allied species already known appears to be Cliona warreni. Carter,^ but this is obviously distinct, as will be seen by comparison of the description {loc. cit.) with ours. Locality. — Station 188, September 10, 1874 ; lat. 9° 59' S., long. 139° 42' E.; south of New Guinea ; depth, 28 fathoms ; bottom, green mud. Family II. S p i R A s T r e l l T D .e, Ridley and Dendy. 1886. SpirastreUidee, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 490. Microsclera present ; typically forming a more or less continuous dermal crust. Genus Spirastrella, Schmidt (Pis. XLL, XLV.). 1868. Spirastrella, Schmidt, Die Spongien d. Kliste v. Algier., p. 17. Sponge usually massive. Megasclera all monactinal, stylote or tylostylote. Micro- sclera sjiined spirulse, occurring mainly as a dermal crust. Schmidt founded the genus for his species Spirastrella cunctatrix, and gives only the following very brief generic diagnosis. " In der Eindenschicht eine eigenthlimliche Art von strahligen Kieselkorperchen, deren Strahlen spiralig gestellt sind." A good many species have been discovered since and the Challenger adds three new ones. Spirastrella decumbens, Ridley, var. (PI. XLV. figs. 12-12^'). 1884. Spirastrella decumbens, Ridley, Zool. Coll. H.M.S. "Alert," Brit. Mus., p. 470, pi. xliii. fig. c. Represented in the collection by a single fine specimen, consisting of a number of stout, irregular, anastomosing trabeculse, forming together a sessile, cavernous mass, 70 mm. long, by 44 mm. broad. The original tyjje was encrusting and thin, and of small size, and must, therefore, be regarded as only a young specimen. There are also slight difi'erenees in the proportions of the spicules between our specimen and the type, the tylostyli (PI. XLV. fig. 12) and spined spirulse (PI. XLV. figs. 12a-12^) being both of rather more robust growth in the Challenger specimen, the tylostyli measuring 0'4 by 0"015 mm., the spirulse 0'028 by 0*007 mm. (exclusive of the spines) ; but there are no difi'erenees sufficient to justify us in separating the two specifically. 1 Ann. and Mag. Nat. Hist, ser. 5, vol. vii. p. 370. 230 THE VOYAGE OF H.M.S. CHALLENGER. Zoca%.— Station 208, January 17, 1875; lat. 11° 37' N., long. 123° 31' E. Philippine Islands; depth, 18 fathoms; bottom, blue mud. One specimen. Habitat. — Torres Strait (Ridley, "Alert"); Philipjiine Islands (Challenger). Spirastrella massa, Ridley and Dendy (PL XLV. figs. 14-14^). 1886. Spirastrella massa, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 490. Sponge massive, represented in the collection by two large, squarish blocks, which have evidently been cut out of a very large specimen, or possibly from tw^o. Coloui- in spirit pale, greyish-yellow. Texture fairly compact throughout, but rather spongy. Surface almost glabrous in appearance, but rather lumpy and uneven ; very mini;tely hispid. Dermal memhrane thin, transparent. Oscula ; one or two scattered ones, of moderate size, have been observed, hut it is very probable that many more were present in the entire specimen ; possibly they were mainly locahsed in some portion of the sponge which has not been preserved. Pores, small round openings, thickly scattered in some parts, so as almost to reduce the dermal membrane to a network, very rare or almost undisceruible in other parts ; diameter, taken from a part of the sponge where they were abundant, about 0"06 mm. Skeleton. — Very difiuse and with no distinct spiculo-fibre ; consisting of abundant, irregularly scattered stylote and subtylostylote spicules. Immediately beneath the surface these spicules are much more closely packed than elsewhere and are also, for the most part, horizontally disposed. They thus form a fairly thick cortical layer, not, however, very well defined from the deeper skeleton. At the surface itself are irregular tufts of stylote spicules whose apices project for a short distance beyond the dermal membrane. Spicules. — (a) Meyasclera ; slightly curved, rather slender, smooth styli or subtylo- styli (PI. XLV. figs. 14, 14a, 146); the heads, when present, are somewhat irregularly developed ; apex not very constant in form, sometimes very sharply and gradually pointed, at others more blunt, sometimes slightly and irregularly jagged; size about 0"45 by 0'0065 mm.; in the surface brushes a trifle shorter, (b) Microsclera; smaU spined spirulaj (PI. XLV. figs. 14c-14^), the largest of which are slender, have five or six bends, and measure about 0"044 mm. in length. These are, however, not nearly so numerous as the smaller ones, which often consist of only one short length with spined ends; about 0'0095 mm. long. As forms intermediate between the two described can be found, it seems probable that they are all merely different stages of the same spicule. These spicules are nowhere abundant except in the dermal membrane, where they are very plentiful. Perhaps the two most remarkable characteristics of this species are its singularly REPOET ON THE MONAXONIDA. 231 massive external form and the shape of the megaselera. Spirastrella cunctatrix, Schmidt/ approaches it in spiculation, but in that species both megaselera and microsclera are of decidedly more robust growth, and the megaselera arc usually very distinctly tylostylote and quite straight, whUe the form of the spined spii-ulse is also different. Spirastrella massa is also remarkable for the great irregularity in the shape of the megaselera ; these are often very abnormal in appearance and may even (though rarely) become oxeote ; this should perhaps be connected with the apparent interchangeability (in different species) of the stylote and oxeote spicules, and with the instability in the condition of the ends of the oxeote in Latrimculia (?) acerata, nobis. ioca%.— Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' 0" E.; Bass Strait; depth, 38 fathoms; bottom, sand and shells. Two large pieces. Spirastrella solida, Ridley and Dendy (PL XLI. fig. 7; PI. XLV. figs. 13-13e). 1886. Spirastrella solida, Eidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 491. Sponge (PL XLI. fig. 7) erect, sessUe, lobate or digitate ; consisting of a broad base, containing a large amount of embedded foreign matter, from which arise broad, fleshy- looking lobes, the larger of which have each one osculum (1 sometimes more) at the summit. Height of the largest .specimen 88 mm.; breadth at base 54 mm. Colour in spirit light yellow. Texture very firm and hard throughout, excepting immediately around the larger canals, where there is a well defined zone of soft, gelatinous, transjjarent tissue enclosing and following the course of the canal. In the wider parts of the exhalent canals, near the summits of the lobes, the same kind of tissue forms well defined circular diaphragms, each with only a small aperture in the centre. Surface rather lumpy and uneven, almost glabrous in appearance yet harsh to the touch. Dermal membrane difficult to distinguish as such, being very heavily laden with the characteristic microsclera. Oscula few in number, but of fair size, occurring ( ? singly) at the tops of the lobes. Pores irregularly but very plentifully scattered over the general surface of the sponge; they are round openings about 0"044 mm. in diameter. Skeleton. — Consisting throughout of an extraordinarily dense, but quite irregular reticulation of interlacing tylostylote spicules ; no spiculo-fibre. At the surface are loose, radiating brushes of smaller, straight, tylostylote spicules, whose apices j^roject for a short distance beyond the surface. Spicules. — (a) Megaselera; (1) the main spicules, almost straight, fairly stout tylostyU (PL XLV. fig. 13), with well developed subglobular heads and rather blunt apices; size about 07 by 0'19 mm. (2) The spicules of the surface brushes; nearly or quite straight, slender tylostyli with well developed subglobular heads and sharply pointed ^ Spongien d. Kiiste v. Algier., p. 17. 232 THE VOYAGE OF H.M.S. CHALLENGER. apices ; size about 0"31 by 0'0094 mm. (b) Microsclera ; (l) very minute, slender, spined spinilse (PI. XLV. figs. 13a-13(i), with some three or four bends and warty rather than spiny; length about 0"012G mm., diameter about 0*0025 mm. (2) There are also a few much longer and, relatively, much slenderer spined spirulse ; in these it is almost impos- sible to say how many bends there are, they are as a rule simply crooked and rather minutely spined; length about 0*056 mm. These latter are probably merely elongated forms of the smaller kind. The microsclera are, as usual, most abundant in the dermal membrane, where they form a solid crust penetrated by the pores, but they are also fairly plentifully scattered in the deeper tissues of the sponge. The tendency to form lobes, shown in some degree by previously described species of the genus, is here carried to an unusual extent. The texture is unusually dense, forming perhaps the most characteristic feature of the species. i:oca%.— Station 208, January 17, 1875 ; lat. 11° 37' N., long. 123° 31' E.; Philipj)ine Islands; depth, 18 fathoms; bottom, blue mud. Two fine specimens, and a fragment, containing a very large amount of foreign matter, chiefly shells, which seems to belong to the same species. Spirastrella papulosa, Ridley and Dendy (PI. XLI. fig. 5 ; PI. XLV. figs. 11-1 If/). 1886. Spirastrella papillom, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 491. Sponge (PL XLI. fig. 5) massive, erect, sessile; shape conical, broad at the base and tapering gradually to an obtuse apex, where are situated several large oscula. Near the apex the surface is thrown into deep longitudinal wrinkles. The entire surface, except in the immediate neighbourhood of the oscula, is covered with numerous, closely placed papillae of considerable size ; these show a tendency towards arrangement in longitudinal series ; towards the apex of the sponge they decrease in size. Height of specimen about 150 mm.; breadth at base 100 mm. Colour in spirit grey. Texture fairly firm, but rather spongy and, internally, cavernous. Surface uneven, as described above, with a minutely- reticulate appearance, and harsh to the touch. The reticulate appearance is caused, or at any rate largely helped, by the arrangement of the microsclera, which are thickly scattered through the dermal membrane in such a manner as to leave small rounded patches, the meshes of the reticulation, almost or quite free from their presence. In these areas are situated the pores, and the corresponding arrangement of the underlying subdermal cavities also helps to give the reticulate appearance to the surface. Dermal membrane thin, transparent. Oscula large, six or seven in number, grouped at the summit of the sponge, each about 6 mm. in diameter, sometimes a little more. Pores ; very numerous, rounded openings, thickly placed in the meshes of the REPORT ON THE MONAXONIDA. 233 dermal reticulation, where they reduce the dermal membrane to a mere network ; diameter about 0'05 mm. Skeleton. — Very diffuse and with only very slightly developed spiculo-fibre ; con- sisting of thickly but irregularly scattered tylostylote spicules, sometimes arranged in tracts ; towards the surface the skeleton becomes much denser so as to form a kind of cortical layer. This is most strongly pronounced in the mammiform processes, where it consists of a very dense reticulation of interlacing spicules. The cortical layer is, however, much broken up by the large, irregularly shaped, subdermal cavities, and it is not very sharply marked off from the deeper skeleton. At the surface of the sponge are irregular brushes of smaller tylostylote spicules whose apices project for a short distance beyond the dermal membrane. Spicules. — (a) Megasclera; (1) the main spicules; straight or slightly curved, gradually and sharply pointed, fairly stout tylostyli (PL XLV. fig. 11), with well developed, broadly oval heads; size about 0-5 by 0"0157 mm. (2) The spicules of the surface brushes ; smaller, straight, slender tylostyli (PI. XLV. figs, llo, lib), commonly with well developed, broadly oval heads and gradually and finely pointed apices ; size about 0"3 by O'OOS mm. (h) Microsclera ; stout spined spirulge (PL XLV. figs, llc-llg), consisting each of three or four bends and with abundant strong spines ; size, exclusive of spines about 0'5 by 0*009 mm.; smaller ones of rather different appearance are also abundant, but they are probably only young forms of the larger. These S23icules occur in very great numbers in the dermal membrane and for a short distance below it. The jDeculiar warty appearance of the surface seems to be the most characteristic feature of this species, but, owing to information received from Dr. R. von Lendenfeld, we are doubtful whether this will turn out to be a constant character. Locality. — Port Jackson; depth, .30 to 35 fathoms. One specimen. Genus Latrunculia, Bocage (Pis. XXIX., XLIV., XLV.). 1870. Latrunculia, Bocage, Jorn. Acad. real. Sci. Lisboa, vol. ii. p. IGl. 1875. ScepireUa, Hchmidt (jMi-fi), J ahi'dsb. Comm. wiss. Unters. deutscli. Meere, Jahrg. ii.,iii. p. 119. Sponge massive, sessile, usually corticate. Upper surface usually covered with mammiform projections. Megasclera smooth styli (? or oxea). Microsclera discastra, characteristic of the genus and typically forming a dense dermal crust. The numerous small mammiform processes are very characteristic of this genus ; some of these bear oscula while others are raised pore-areas ; we cannot, however, say whether this arrangement obtains in all species. Bocage {loc. cit.) gives no attempt at a generic diagnosis but plunges at once into the description of the type species, Latrunculia cratera. Previous to the appearance of our (ZOOL. CHALL. EXP. — PART LIX. — 1887.) Nlin 30 234 THE VOYAGE OF H.M.S. CHALLENGER. Preliminary Eeport only three sponges had been described under the name Lati'unculia viz. : — Latrunculia cratera, Bocage {loc. cit. supra) ; Latrunculia corticata. Carter ^ and Latrunculia 'purpurea, Carter.^ Both of Mr. Carter's species have, however, oxeote in place of stylote megasclera ; in this they agree with one of the Challenger sjjecies and all are thereby very sharply marked off from the original type and from three new species obtained by the Challenger ; indeed it seems doubtful whether these forms with oxeote megasclera ought to be included in the same genus with the species that have stylote spicules ; this question we must leave open for the present, but for the sake of convenience we shall consider those forms with oxeote spicules as doubtfully belonging to the genus Latrunculia. Schmidt has described two sponges under the generic name Sceptrella; the first of these, viz., Sceptrella regalis^^ is of rather doubtful systematic position, though almost certainly referable to the genus Latrunculia. Sceptrella triloba* the second- species, is undoubtedly a Latrunculia {Latrunculia triloba), as will readily be seen from Schmidt's brief description, which we c[uote in full : — " Bildet Krusten, am Rande mit rohrenformigen Fortsatzen, welche theils eine wirkliche Osculum-Oeffnung haben, theils bloss durch microscopische Poren den Wasserabfluss gestatten. Die st.-sp. Nadeln in der Mitte etwas dicker. Zwischen diesen groberen feinere. Die characterischen Schachfigurenformigen Kieselkorper zeigen die grosste Verwandtschaft mit denen der Sceptr. regalis von Florida, ihre vier Wirtel bestehen aber nur aus drei Strahlen, welche sich ein- bis zweimal dichotomisch gabeln. Diese Korper bilden, indem ihre Axen senkrecht zur Oberfliiche stehen, ein dichtes Pilaster." Podospongia lovenii, Bocage,'' seems to be nearly allied to the genus Latrunculia, but, without entering into the question of spiculation, the presence of a very distinct stalk is sufficient to separate it generically. Latrunculia apicalis, Ridley and Dendy (PI. XLIV. fig. 4 ; PI. XLV. figs. 9-9c). 1886. Latrunculia ajncalis, Kidley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 492. Sponge (PI. XLIV. fig. 4) massive, sessile, with broad base attached to foreign objects, and veiy strongly convex upper surface beset with more or less numerous small, mammi- form processes. The largest specimen (the one from Kerguelen) measures 114 mm. in hmgth by 75 mm. in breadth and 56 mm. in thickness in the middle, the base is oval and the upper surface very strongly convex and beset with an enormous number of small ' Ann. and Mag. Nat Hist., .ser. 5, vol. iii. p. 298, pi. xxvii. figs. 3-4. - Op. cil., ser. 5, vol. vii. p. 380, pi. xviii. fij,'. 5. •' Spong. Atlant. Gebiet., p. 58, pi. v. fig. 24. The systematic position of this sponge is discussed in our Intioduttion. * Jahresb. d. Comm. wiss. Unters. deutsch. Meere, Jahrg, ii., iii., p. 119, pi. i. fig?. 17, 18. '• Jom. Acad, real Sci. Lisboa, vol. ii. p. 159, pi. x. fig. 1. EEPORT ON THE MONAXONIDA. 235 m.'immiform processes ; towards the summit of the sponge many of these processes are hirger than elsewhere, measuring about 6 mm. in height ; these are conical and have each a distinct oscular opening at the top. Much more abundant are the smaller processes, which occur abundantly all over the upper surface of the sponge, excepting where they are replaced by the larger ones ; they are very short, cylindrical, and abiaiptly truncated at the top. The larger processes are undoubtedly cloacal tubes, and we were at first inclined to think that the small ones might be the same in a state of retraction, but we have now very strong evidence in favour of regarding them as raised pore-areas. Canals in the body of the sponge lead up to both kinds ; the smaller ones, however, appear to the naked eye to be almost all closed at the top. This description also applies fairly to the two specimens from Station 320. The larger of the two is more globular and smaller than the Kerguelen sponge ; the difference between the large and small papilli- form processes is not so well marked ; but the sponge is in poor condition for making out anatomical details ; it is even impossible to decide whether there was one or more large oscula at the summit ; that there was at least one is certain. The small processes are more conical in form, and, though plentiful, not nearly so abundant as in the Kerguelen sponge. Colour in spirit (of the specimens from Station 320) yellowish-grey. The specimen from Kerguelen is of a deep chocolate brown colour throughout ; but as all the specimens in the bottle with it, and also the parchment label, were found to be stained of the same colour, we cannot be certain to which sponge the colour may lie proper. Texture fairly compact, but spongy throughout, with tough, parchment-like dermal layer easily separable from the deeper tissues. In the Kerguelen specimen the cortex is a good deal thicker than in the specimens from Station 320, and may be described as tough and leathery. Surface smooth between the mammiform processes, but rather harsh to the touch. Oscula on the tops of mammiform processes. Pores; narrow perforations through the flat tops of the smaller mammiform projections, leading into large inhalent canals. Skeleton.. — There is a dermal crust composed of a single layer of the characteristic Latrunculia " chess-man " spicules (discastra), vertically placed and with outwardly directed, elongated apices. Below these the skeleton is very loose and irregular, com- posed of abundant smooth stylote spicules disposed without any definite order. Imme- diately below the surface these spicules tend to be vertically arranged, but this layer passes very soon into a much denser but utterly irregular reticulation, and below this again, forming the main mass of the skeleton, is a loose, irregular reticulation of the same spicules, often forming incipient fibres but with no distinguishable primary and secondary lines. The discastra also occur scattered promiscuously throughout the sponge. In the Kerguelen specimen the dermal layer of discastra is immediately backed up by a thick, dense layer of closely interwoven stylote spicules. It thus ajipears thnt there is great irregularity in the arrangement of the main skeleton. ^36 THE VOYAGE OF H.M.S. CHALLENGER. Spicules. — (a) Megasclera ; smooth, rather slender styli (PI. XLV. fig. 9), hastately pointed and often slightly crooked; size about O'G by 0*014 mm.; often with slight irregular inflations on the shaft. In the Kerguelen specimen these spicules are a trifle shorter and stouter, and there occur also a very few, apparently abnormal, much larger forms. (6) Microsclera; discastra, the "chess-man" spicules, which in this species resemble in form miniature fir trees (PI. XLV. figs. 9«, 9&, 9c) ; each spicule is composed of an expanded, spinose base, followed immediately by a rather irregular whorl of separate short spines ; then comes a short, stout, smooth, cylindrical shaft which suddenly expands into a broad, flattened, circular whorl with denticulated margin ; three of the indentations are so deep as to cut right down to the shaft, thereby dividing the circle into three distinct segments (PL XLV. fig. %). This is much the largest of the whorls, and is followed by three or four others gradually lessening in size till they end in a few small spines on the shaft. The shaft itself is produced into a long, smooth, terminal portion, gradually tapering to a fine point, and sometimes curved. Total length of spicule about 0"126 mm., of which the apical prolongation may form nearly half; diameter of largest whorl about 0'044 mm. The most characteristic feature about this species, and that from which the specific name has been derived, is the apical prolongation of the discaster, by which it may at once be distinguished. The geographical distribution is very interesting ; we have, during the course of our description, indicated various slight difi'erences between the specimens from the two localities, but there is nothing to justify us in separating them specifically. Details regarding the minute anatomy will be found in the Intro- duction. Localities. — Ofi" Christmas Harbour, Kerguelen, January 29, 1874; depth, 70 fathoms. One specimen. Station 320, February 14, 1876; lat. 37° 17' S., long 53° 52' W.; ofl" the mouth of the Eio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 3 7° '2. Two specimens. Lat7-unculia hrevis, Eidley and Dendy (PI. XLIV. fig. 5 ; PI. XLV. figs, 10, 10a). 1886. Latrunculia hrevis, Eidley and Dendy, Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 492. Sponge (PI. XLIV. fig. 5) massive, sessile, usually attached by the base to some foreign object, and with more or less convex upper surface beset with numerous conical processes. The best specimen in the collection is attached by a narrow base to a fragment of Sporadopora ; it is about 63 mm. in diameter, and the upper surface is rather flattened and thickly covered with the conical osculum-bearing processes, which are aliuut 6 mm. in height. Colour in spirit yellowish-grey. Texture fairly compact REPORT ON THE MONAXONIDA. 237 throughout, but soft and spongy, with easily separable, parchment-like cortex. Surface smooth between the conical processes, but rather harsh to the touch. Oscula on the tops of conical processes. Pores (probably as in Latrunculia apicalis). Skeleton. — There is a dense dermal crust, composed as usual of a single layer of vertically placed discasters ; and below this there is a loose, irregular reticulation of stylotts •spicules, in which one may distinguish poorly developed lines of fibre, some running towards the surface representing the primary skeleton lines. Spicules.— {a) Megasclera ; smooth styli (PI. XLV. fig 10), straight or very slightly crooked, with evenly rounded bases and more or less hastately pointed apices, not very sharp; size about 0"6 by 0'12G mm. (6) Microsclera, discastra; for these we refer to PI. XLV. fig. 10a, and also to the description of the corresponding spicules in Latrunculia aincalis, merely stating that they differ chiefly from the latter in having no apical pro- longations ; the upper whorls are approximated so as to form a thick brush at the top, the Uu-gest occupying the same position as in Latrunculia apicalis; length about 0"05 mm., diameter of largest whorl about 0'044 mm. The form of the " chess-man " spicule is, as usual in the genus Latrunculia, the most characteristic feature of the species. (This remark would scarcely be applicable to Latruncidia cratera, Bocage,^ should the very minute size of the megasclera, viz., 0"I8 mm. long, as ascertained by calculation from the figure, which would at once distinguish it from all the Challenger species, be confirmed.) ^ A slight variety, of which there is one specimen, from the same station as the types of the species, has the " chess-man " spicule a trifle (but very slightly) elongated and the whorls much further apart from one another as compared with the type ; thereby making an approach to the Kerguelen species, Latrunculia hocagei, nobis. It would be curious, if it should prove to be correct, that there should be only one form of mammiform processes in this species, that all should be conical and osculuni- bearing, with no raised pore-areas. AVe are, however, strongly inclined to think, both from the analogy of other species and from the enormous number of the projections, which can scarcely all be cloacal tubes, that the raised pore-areas would be found to exist in more perfectly preserved specimens, although in the Challenger specimens they are indis- tinguishable from the osculum-bearing processes. Locality.— ^tsitww 320, February 14, 1876; lat. 37° 17' S., long. .-53° 52' W.; ofi' the mouth of the Rio de la Plata ; depth, 600 fathoms ; bottom, green sand ; bottom temperature, 3 7° "2. About ten specimens in poor condition, being a good deal crushed. ' Jorn. Acad, real Sci, Lisboa, vol. ii. p. 161, pi. xi. fig. 2. 2>See, however, the remarks after the next specie.s. 238 THE VOYAGE OF H.M.S. CHALLENGER. Latrunculia bocagei, Ridley and Dendy (PI. XLIV. fig. 1 ; PL XLV. figs. 8, 8'^'). 1886. Latrunculia Bocagei, Ridley and Dendy, Ann. and Mag. Nat. Hist, ser. 5, vol. xviii. p. 492. Sponge (PI. XLIV. fig. 1) massive, sessile; the two specimens in the collection are both subglobular and have been attached to foreign oljjects Ijy constricted bases. The strongly convex upper surface' is covered with very numerous papillifomi processes ; a few of these, at the top of the sponge, are rather larger than the rest, in form conical, and each with a small oscular opening at the apex. The remainder are cylindrical rather than conical in form, narrow, somewhat elongated and abruptly truncated at the top, where no distinct opening can be seen ; evidently they are raised pore-areas as in Latrunculia apicalis. There is a distinct, thick, leathery cortex, easily separable from the underlying tissues. The larger of the two specimens measures about 38 mm. in diameter. Colour in spirit very pale yellow. One specimen, which was found in the same bottle with the dark- coloured specimen of Latrunculia apicalis, is, like it, of a dark chocolate colour throughout ; but since the other specimen, which was in another bottle, is of a very pale yellow colour, we must attribute the brown coloration of the former specimen to the action of some staining substance dissolved in the spirit and probal)l}' derived from some other sponge. Texture of the cortex tough and leathery, of the inner parts fairly compact, but rather soft and spongy. Surface fairly smooth between the numerous processes, but rather harsh to the touch and wrinkled in places, especially around the oscula. Oscula distributed singly on the tops of conical eminences. Pores (doubtless arranged as in Latrunculia ajncalis, though this has not been absolutely proved). Skeleton. — Arranged as usual in the genus, with an external layer of the discasters, and an internal, confused mass of smooth styli, with ill developed fibres, making up the main skeleton. In both specimens of this species, however, the layer of " chess-man " spicules is backed up by a thick cortical layer of densely interlacing stylote spicules, as in the Kerguelcn specimen of Latmnculia apicalis, and not by a mere film of interlacing spicules as in Latrunculia hrevis. Spicules. — (a) Megasclera; smooth, slightly crooked, more or less hastately pointed styli (PI. XLV. fig. 8), measuring about 0*6 by 0-018 mm. (6) Microsclera; discastra, differing somewhat inform from those of any other known species ; each spicule (PI. XLV. fig. 8a) has a slightly expanded base armed with two whorls of spines, then comes the smooth, stout shaft bearing three distinct, subequal whorls, well separated from one another, and ending in a crown-like tuft of spines which follows close upon the last whorl. Each of the three disc-like whorls is deeply notched along the rim, but the notches are not equal all the way round. Length of the spicule about 0'07 mm., diameter of whorls about 0'03 mm. REPORT ON THE MONAXONIDA. :*89 In this species, again, the form of the " chess-man " spicule is, as usual, the dis- tinguishing feature. The figures will show that it is really distinct from the corresponding s[)icules in LatruncuUa apicalis and Latrunculia hrevis, although, as we have already had occasion to mention, there is a slight variety of the latter which seems to form a connecting link between the two ; the spicule in Latrunculia hocagei is lai-ger and narrower than in Latrunculia hrevis, and the arrangement and relative sizes of the whorls different. Latrunculia cratera, Socage, the original type of the genus, comes near to our species in this respect, but in that species the whorls are much more minutely and evenly dentate all the way round the rim, and the uppermost one is much smaller than the others. According to Bocage's figures, also, the stylote spicule in Latrunculia cratera is only 0'18 mm. long; but here we feel strongly inclined to think that there has been some error, for we have found that the constancy in size of the stylote spicule is a remarkable feature of the genus ; and a preparation in the British Museum, identified by Schmidt with Latrunculia hocagei, while agreeing with the latter as regards the form of the " chess-man " spicule, has the stylus about 0'5 mm. long. As regards external form it will be seen that Latrunculia hocagei is almost indis- tinguishable from the Kerguelen specimen of Lati'unculia apicalis, and correspondingly different from Latnmculia hrevis ; but in this case we are not inclined to set much value on external form as a specific character, for we have already seen that the specimens of Ijatrunculia apicalis from Kerguelen and from Station 320 respectively, differ in external appearance ; indeed, to judge from the Challenger series of specimens of the genus, it would seem that external appearance depends on the locality and that all the species from the same locality tend to have a similar external form. Locality. — Kerguelen; depth, 10 to 70 fathoms. Two specimens. Latrunculia {f) acerata, Eidley and Dendy (PI. XXIX. figs. 3, 3a, 3&). 1886. Latrunculia (?) acerata, Ridley and Dendy, Ann. and Mag. Nat. Hist., ser. 5, vol. xviii. p. 492. Sponge massive, amorphous, spreading. There is only one specimen in the collection, and that in a very bad state of preservation. It forms a flattened, shapeless mass about .'SO mm. long by 38 mm. broad and 13 mm. thick. Colour in spirit dirty brown. Texture rather firm and compact. Surface very uneven, irregularly grooved ; at present carrying a considerable amount of foreign matter. Dermal membrane distinct, strong, supported by very numerous spicules ; in parts peeling off. Oscula (?). Pores scattered through the dermal membrane (a fragment of the dermal membrane, which was boiled in nitric acid for the sake of separating the spicules, appeared at the end of the process uninjured, nnd showed the pores very distinctly). 240 THE VOYAGE OF H.M.S. CHALLENGER. Skeleton. — {a) Dermal; varying much in its degree of development at diflferent parts of the surface. Typically a more or less closely matted, irregular feltwork of slender strongylote spicules, much smaller than the spicules of the main skeleton, {h) Main ; also varying much in its degree of compactness ; a more or less definite reticulation of large sj)icules, with fibres three or four spicules broad, distinct in parts ; sometimes primary fibres, running towards the surface, can be distinguished from the remainder. Spicules. — (a) Megasclera ; (1) long, smooth, more or less curved oxea (PI. XXIX. fig. 3), very variable at the ends, sometimes gradually and sharply pointed, sometimes blunted at each end ; ends often rather irregular ; size about 0'9 by 0"025 mm., forming the main skeleton. (2) Smooth, rather slender, very slightly curved strongyla (PI. XXIX. fig. 3a), rounded ofi" at each end, measuring about 0'48 by 0"012 mm., occurring chiefly in the dermal skeleton. (6) Microsclera ; very small, slender diseastra (PI. XXIX. fig. 3&) resembling the typical " chess-man " spicules of Latruncnlia, but not nearly so weU developed. Consisting each of a slender shaft, straight or slightly crooked, some- times rather expanded at the base, not distinctly spined but (commonly at any rate) with a roughened appearance ; bearing two saucer-like whorls, a smaller one close to the base and a larger one some distance above it. There are never any indications of more whorls. The smaller one is slightly concave upwards and the larger slightly concave downwards. The margin of each whorl is normally slightly notched all round. The slender shaft is produced for some distance beyond the upj^er whorl, but instead of tapering ofi' to a point it ends abruptly and is often of the same thickness along its entire course. Length of spicule about 0"037 mm. Diameter of larger whorl about 0'0125 mm. These spicules are abundantly scattered through the dermal membrane and are also found occasionally in the subjacent parts. It is extremely unfortunate that the single specimen of this very interesting and important species should be in such poor condition, and also that the locality should Ite uncertain. The label on the bottle bears the inscription "No. 135 ? Date. Oct. 1873. Lat. Long. . Depth, 60 fm's." The species is at once distinguished from any of the three preceding by its spiculation (oxea instead of styli), and by the absence of mammiform processes on the surface. Indeed, as we have already pointed out, it is very doubtful whether it even belongs to the same genus. It is distinguished from Mr. Carter's two species, Latrunculia covticata^ and Latruncnlia purpurea,^ which also have oxeote megasclera, by the form of the discaster, which in our species approaches that of a typical Latruncnlia much more nearly than in either of Mr. Carter's. This forms an argument for including all in the same genus, but it is a very slight one when compared with the arguments against this view. As, however, we have only a single specimen we leave it where it is for the present. ' Loc. cit. su-pra. GEOGRAPHICAL AND BATHYMETRICAL DISTRIBUTION. LIST OF LOCALITIES AT WHICH MONAXONID SPONGES WERE OBTAINED, WITH THE SPECIES OBTAINED AT EACH. Station 24. March 25, 1873 ; lat. 18° 38' 30" N., long. 65° 5' 30" W.; West Indies; depth, 390 fathoms ; bottom, Pteropod ooze. Desmacella annexa. Bermuda, West Indies. Tedania digitata, var. hermudensis. Station 49. May 20, 1873 ; lat. 43° 3' 0" N., long. 63° 39' 0" W.; south of Halifax, Nova Scotia ; depth, 85 fathoms ; bottom, gravel, stones ; bottom temperature, 35°-0. Artemisina suheritoides. Polymastia rohusta-. Tentorium semisuherites. Stylocordyla stipitata. Quasillina hrevis. Station 50. May 21, 1873 ; lat. 42° 8' 0" N., long. 63° 39' 0" W.; south of Nova Scotia; depth, 1250 fathoms; l^ottom, blue mud; bottom temperature, 38°'0. Tentorium semisuherites. Station 73. June 30, 1873 ; lat. 38° 30' 0" N., long. 31° 14' 0" W.; depth, 1000 fathoms ; bottom, Pteropod ooze ; bottom temperature, 39°'4. Rhizochalina fistulosa {?). | Trichostemma sarsii. Station 75. July 2, 1873 ; lat. 38° 38' 0" N., long. 28° 28' 30" W.; off the Azores; depth, 450 fathoms ; bottom, volcanic mud. Reniera implcxa. Gellius angulatus. Plocamia coriacea, var. elegans. Suherites carnosus. Suberites elongatus. Polymastia agglutinans. (ZOOL. CHALL. EXP. — PART LIX. — 1887.) Nuil 31 242 THE VOYAGE OF H.M.S. CHALLENGER. Cape Verde Islands. Reniera tufa. Hymeniacidon caruncula. Axinella monticularis. Axinella {?) lunascharta. Off Fernando Noronha. Suherites carnosus. Station 122c. September 10, 1873; lat. 9° 10' 0" S., long. 34° 49' 0" W.; east of Brazil ; depth, 400 fathoms ; bottom, red mud. Phakellia ventilahrum, var. connexiva. Station 125. September 12, 1873 ; lat. 10° 46' 0" S., long. 36° 2' 0" W.; between Pernambuco and Bahia ; depth, 1200 fathoms; bottom, red mud. Polymastia corticata. Off Bahia. Myxilla j^lumosa, var. fusifera. Rliai^hidophlus gracilis. Plumohalichondria jjlumosa. Axinella ecliidnasa. Axinella reticulata. Pachychalina Jihrosa. Rhizochalina putridosa {?). ? Oceanaina robusta. Toxoclialina robusta. Es^oerella nuda. Esperella fusca. Desmacidon reptans. Raspailia tenuis. Thrinacophora funiformis. Stylocordyla stipitata. Off Tristan da Cunha. Axinella erecta. OS Nightingale Island, Tristan da Cunha. lophon pattersoni. | Axinella erecta. Inaccessible Island, Tristan da Cunha. Axinella erecta. \ Axinella (f) paradoxa. Tentorium semisuberites. Station 135. October 15, 1873; lat. 37° 1' 50" S.; long. 12° 19' 10" W.; Tristan da Cunha ; depth, 360 fathoms ; bottom, volcanic sand. ? Rhizochalina singaporensis. I ? Agelas mauritianus. ? Gelliodes licheniformis. \ ? Latrunculia {?) acerata. REPORT ON THE MONAXONIDA. 243 Simon's Bay, Cape of Good Hope. Esperella simonis. Desmacidon conulosa. Desmacidon {Homceodictya) grandis. Rlmphidophlus lohatus,vav. horrida. Raspailia flagelliformis. Dendropsis bidentifera. Proteleia sollasi. Station 142. December 18, 1873 ; lat. 35° 4' O" S., long. 18° 37' 0" E. ; south of the Cape of Good Hope ; depth, 150 fathoms ; bottom, green sand ; bottom tempera- tm-e, 47°-0. Petrosia similis. Gellius glacialis. Vomerula esperioides. Off Marion Island. Halichondria sp. (&). Gellius carduus. Gellius Jlagellifer. Desmacidon (f ) ramosa. Off Prince Edward Island. Desmacidon {?) ramosa. lophon chelifer. Mijxilla digitata. Raspailia (f ) rigida. lophon ahnormalis. Amphilectus pilosus. Myxilla mariana. Axinella mariana. Suberites caminatus. Gellius carduus. Station 145. December 27, 1873 ; lat. 46° 43' 0" S., long. 38° 4' 30" E.; off Prince Edward Island ; depth, 140 fathoms ; bottom, volcanic sand. Gellius glacialis, var. nivea. \ Stylocordyla stipitata, var. glohosa. Station 145a. December 27, 1873 ; lat. 46° 41' 0" S., long. 38° 10' O" E.; off Prince Edward Island; depth, 310 fathoms; bottom, volcanic sand. Esperiopsis symmetrica. 1 lophon laminalis. lophon chelifer. \ PhaMlia papyracea. Axinella erecta. Station 147. December 30, 1873 ; lat. 46° 16' 0" S., long. 48° 27' 0" E. ; between Prince Edward Island and Crozet Islands ; depth, 1600 fathoms; bottom, Diatom ooze ; bottom temperature, 34°-2. Esperella mammiformis. Esperiop)sis profunda. Cladorhiza tridentata. Meliiderma stipitata. Axinella erecta. Stylocordyla stipitata. 244 THE VOYAGE OF H.M.S. CHALLENGER. Station 148. January 3, 1874; lat. 46' 47' 0" S., long. 51° 37' 0" E.; Possession Island ; depth, 210 fathoms ; bottom, hard ground, gravel, sheUs. Phahellia papyracea. | Suherites mollis. Station 148a. January 3, 1874; lat. 46° 53' 0" S., long. 51° 52' 0" E. ; south of Crozet Islands ; depth, 240 to 550 fathoms ; bottom, hard ground, gravel, shells. Gellius carduus. Tophon chelifev. Kerguelen. Halichondria panicea. Petrosia hisjnda. Pachychalina (?) pedunculata. Desmacidon [Homceodictya) ker- guelenensis. Amphilectus apollinis. Myxilla nohilis. Axinella erecta. Amphilectus pilosus. Axinella balfourensis. Suherites antarcticus. Stylocordyla stipitata, globosa. Latrunculia apicalis. \ax. Latrunculia bocagei. Station 150. February 2, 1874; lat. 52° 4' 0" S., long. 71° 22' 0" E.; west of Heard Island, Southern Ocean; depth, 150 fathoms; bottom, coarse gravel; bottom temperature, 35°"2. Petrosia similis. | Myxilla fusca. Suherites microstomus. Station 157. March 3, 1874; lat. 53° 55' 0" S., long. 108° 35' 0" E.; Southern Ocean, south-west of Australia ; depth, 1950 fathoms ; bottom. Diatom ooze ; bottom temperature, 32°*1. Cladorhiza moridiformis. Off the south-east coast of Australia. Suherites durissimtis. Station 162. April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' 0" E.; Bass Strait; depth, 38 fathoms ; bottom, sand and shells. Pachychalina elongata. Pachychalina (f ) 'pibnctata. Chalina pergamentacea. Chalina sp. (b). Siphonochalina annidata. Rhizochcdina putridosa. Tedania commixta. Esperella arenicola. REPORT ON THE MONAXONIDA. 245 Station 1 6 2 — continued. Esperella sp. Clatliria elegantula. Plumohalichondria mammillata. Echinoclathria favus. Echmoclathria carteri. Echinoclathria glabra. Spirastrella massa. Station 163a. April 4, 1874; lat. 36° 59' 0" S., long. 150° 20' 0" E.; south-east of Australia; depth, 150 fathoms; bottom, greeu mud. Desmacidon fruticosa, var. lophon cylindricus. lophon (f) omnivorus. Clathria inanchorata. Echinocla thria carteri. Off Port Jackson, Australia. Pachijchalina lohata. Sijihonochalina intermedia. Rhizochalina putridosa. Gelliodes pocidum. Tedania digitata. Tedania digitata, var. fibrosa. Esperella murrayi. Esperella porosa. Esperiopsis cylindrica. Amphilectus ceratosus. Clathria lendenfeldi. Echinoclathria carteri. Phahellia Jlabellata. Axinella arhorescens. Suherites carnosus. Suherites perfectus. Spirastrella pjapillosa. Station 163d. June 12, 1874 ; lat. 33° 57' 30" S., long. 151° 39' 15" E. ; depth, 120 fathoms ; bottom, green sand. Pachychalina megalorrhaphis. | Tedania massa {?). Station 170. July 14, 1874 ; lat. 29° 55' 0" S., long. 178° 14' 0" W. ; off Kermadec Islands ; depth, 520 fathoms ; bottom, volcanic mud ; bottom temperature, 43°"0. Myxilla frondosa. Station 174. August 3, 1874 ; lat. 19° 6' 0" S., long. 178° 14' 20" E. ; Kandavu, Fiji Islands ; depth, 140 fathoms ; bottom, coral mud. Chondrocladia, clavata. Api, New Hebrides. Halichondria solida, var. rugosa. Rhizochalina, pedunculata. Station 184. August 29, 1874 ; lat. 12° 8' 0" S., long. 145° 10' 0" E. ; south-east of Cape York ; depth, 1400 fathoms ; bottom, Globigerina ooze ; bottom tempera- ture, 36°-0. Trichostemma sarsii. 246 THE VOYAGE OF H.M.S. CHALLENGEE. Torres Strait. Chalina sp. (a). Gelliodes Jihulata. Esperella gelatinosa. Clathria decumbens. Acanthella pulcherrima. Station 186. September 8, 1874 ; lat. 10° 30' 0" S., long. 142° 18' 0" E. ; off Cape York, Torres Strait ; depth, 8 fathoms ; bottom, coral mud. CJudina palmata. | Clathria aculeata. Acanthella (?) stipitata, var. Station 188. September 10, 1874; lat. 9° 59' 0" S., long. 139° 42' 0" E. ; west of Torres Strait ; depth, 28 fathoms ; bottom, green mud. Rhizochalina Jistulosa {?). Sideroderma navicelligei'um. Clathria frondifera. Cliona dissimilis. Station 190. September 12, 1874 ; lat. 8° 56' 0" S., long. 136° 5' O" E. ; south-west of New Guinea ; depth, 49 fathoms ; bottom, green mud. Echinodictyum rugosum. Station 192. September 26, 1874 ; lat. 5° 49' 15" S., long. 132° 14' 15" E. ; off Little Ki Island, south of New Guinea ; depth, 140 fathoms ; bottom, blue mud. Esperiopsis pulchella. \ Myxilla paucispinata. Amboina. Halichondria pelliculata. Station 196. October 13, 1874 ; lat. 0° 48' 30" S., long. 126° 58' 30" E. ; east of Celebes Island ; depth, 825 fathoms ; bottom, hard ground ; bottom temperature, 3 6° -9. Esperiopsis challengeri. Station 203. October 31, 1874 ; lat. 11° 6' 0" N., long. 123° 9' 0" E. ; Philippine Islands ; depth, 20 fathoms ; bottom, mud. Halichondria sp. (c). Station 207. January 16, 1875 ; lat. 12° 21' 0" N., long. 122° 15' 0" E. ; Philippine Islands ; depth, 700 fathoms ; bottom, blue mud ; bottom temperature, 51° -6. Suherites ramulosus. REPORT ON THE MONAXONIDA. 247 Station 208. January 17, 1875 ; lat. 11° 37' 0" N.,long. 123° 31' 0" E. ; Philippine Islands; depth, 18 fathoms ; bottom, blue mud. Petrosia variabilis. Petrosia similis, var. compact a. Petrosia truncata. Pachychalina fragilis. Pachychalina melior. Pachychalina fibrosa, var. Chalina rectangularis. Gellius varius. Esperella parishii. Rhaphidophlus filifer. Hymeniacidon (f ) subacerata. Thrinacophora cervicornis. Spirastrella decumbens. Spirastrella solida. Station 209. January 22, 1875 ; lat. 10° 14' 0" N., long. 123° 54' 0" E. ; Philippine Islands; depth, 95 fathoms; bottom, blue mud; bottom temperature, 7l°"0. Reniera cinerea. Suberites ramulosus, var. cylin- drifera. Station 214. February 10, 1875 ; lat. 4° 33' 0" N., long. 127° 6' 0" E. ; east of Meangis Island, south of Philippine Islands ; depth, 500 fathoms ; bottom, blue mud ; bottom temperature, 41°'8. Esperiopsis challengeri, var. meangensis. Station 216a. February 16, 1875 ; lat. 2° 56' 0" N., long. 134° 11' 0" K; north of New Guinea ; depth, 2000 fathoms ; bottom, Globigerina ooze ; bottom tem- perature, 35°*4. Chondrocladia crinita. Japan. Halichondria panicea. | Myxilla rosacea, var. japonica. Station 241. June 23, 1875 ; lat. 35° 41' 0" N., long. 157° 42' 0" K; North Pacific Ocean ; depth, 2300 fathoms ; bottom, red clay ; bottom temperature, 35 "1. Axinella profunda. Station 246. July 2, 1875 ; lat. 36° 10' 0" N., long. 178° 0' 0" E.; North Pacific Ocean ; depth, 2050 fathoms; bottom, Globigerina ooze; bottom temperature, 35°'l. Suberites senilis. Station 248. July 5, 1875 ; lat. 37° 41' 0" N., long. 177° 4' 0" W.; North Pacific Ocean; depth, 2900 fathoms ; bottom, red clay; bottom temperature, 35 '1. Chondrocladia concrescens (f). 248 THE "VOYAGE OF H.M.S. CHALLENGER. Honolulu. Esperiopsis anomala. Station 264. August 23, 1875 ; lat. 14° 19' 0" N., long. 152° 37' 0" W.; North Pacific Ocean ; depth, 3000 fathoms ; bottom, red clay ; bottom temperature, 35°'2. Cladorhiza longipinna. Station 274. September 11, 1875; lat. 7° 25' 0" S., long. 152° 15' 0" W.; Mid Pacific Ocean ; depth, 2750 fathoms ; bottom, Radiolarian ooze ; bottom tem- perature, 35°"1. Cladorhiza abyssicola, var. rectangularis. Tahiti. Halichondria solida. \ Acarnus ternatus. Echinodictyum asperum. Station 281. October G, 1875 ; lat. 22° 21' 0" S., long. 150° 17' 0" W.; South Pacific Ocean ; depth, 2385 fathoms ; bottom, red clay ; bottom temperature, 34°"9. Esperella hiserialis. Cladorhiza ahyssicola,ya.v. linearis. Cladorhiza similis. Axinella profunda. Station 291. October 27, 1875; lat. 39° 13' 0" S., long. 118° 49' 0" W.; South Pacific Ocean ; depth, 2250 fathoms ; bottom, red clay ; bottom temperature, 34°-6. Esperella hiserialis. | Axoniderma mirabile. Station 299. December 14, 1875; lat. 33° 31' 0" S., long. 74° 43' 0" W.; west of Valparaiso; depth, 2160 fathoms; bottom, blue mud; bottom temperature, 35°-2. Tedania actiniiformis. | Trichostemma irregularis. Station 306a. January 2, 1876 ; lat. 48° 27' 0" S., long. 74° 30' 0" W.; off the south- west coast of Patagonia ; depth, 345 fathoms ; bottom, l)lue mud; bottom tempera- ture, 46°-0. Myxilla crihrigera. Station 307. January 4, 1876; lat. 49° 24' 30" S., long. 74° 23' 30" W.; off the south- west coast of Patagonia; depth, 140 fathoms ; bottom, blue mud. Reniera sxibglohosa. | Myxilla nobilis, var. bacilli/era. REPORT ON THE MONAXONTDA. 249 Station 308. January 5, 1876 ; lat. 50° 8' 30" S., long. 74° 41' 0" W.; off the west coast of Patagonia ; depth, 175 fathoms ; bottom, blue mud. Tedania tenuicapitata. | Trachytedania patagonica. lophon pattersoni. Station 311. January 11, 1876; lat. 52° 45' 30" S., long. 73° 46' 0" W.; off the south-west coast of Patagonia ; depth, 245 fathoms ; bottom, blue mud ; bottom temperature, 46°'0. Gellius carduus, var. magellanica. Tedania tem(,icap)itata. lophon pattersoni. Myxilla nobilis, var. ijatagonica. Off the south-west coast of Patagonia. Halichondria caduca. Reniera aquseductus, var. infun- dihukms. Tedania tenuicapitata. Tedania infundihuliformis. Trachytedania patagonica. Myxilla mollis. Myxilla mariana, var. massu. Hymeniacidon (f) hyalina. Siiherites spiralis. Station 313. January 20, 1876 ; lat. 52° 20' 0" S., long. 67° 39' 0" W. ; east of the Strait of Magellan ; depth, 55 fathoms ; bottom, sand ; bottom temperature, 47°-8. Tedaiiia massa. Esperella magellanica. Esperiopsis edivardii, var. americana. Hymeniacidon sp. Axinella fibrosa. Station 314. January 21, 1876 ; lat. 51° 35' 0" S., long. 65° 39' 0" W. ; between the Strait of Magellan and the Falkland Islands ; depth, 70 fathoms ; bottom, sand ; bottom temperature, 46°'0. Petrosia similis, var. massa. I Tedania tenuicapitata. Station 317. February 8, 1876 ; lat. 48° 37' 0" S., long. 55° 17' 0" W.; north-east of the Falkland Islands; depth, 1035 fathoms; bottom, hard ground (gravel); bottom temperature, 35°'7. Phahellia ventilabrum, var. connexiva. (ZOOL. CHALL. EXP. — PART LIX. — 1887.) Niin 32 250 THE VOYAGE OF H.M.S. CHALLENGER. Station 320. February 14, 1876 ; lat. 37° 17' 0" the Eio de ]a Plata ; depth, 600 fathoms ; ture, 37°-2. Halichondria latrunculioides. Hcdichondria sp. («). Gellius Iwvis. Gellius ccdijx. Gellius Jlahelliformis. Gellius sp. Tedania massa. Vomerida esperioides. Esperella lapidiformis. Phelloderma radiatum. Amphilecttis annectens. S., loug. 53° 52' 0" W. ; off the mouth of bottom, green sand ; bottom tempera- lyxiiia sp)ongiosa. Myxilla hastata. Myxilla comp>ressa. Myxilla nobilis. Cioccdypta hyaloderma. Ciocalypta amorp>hosa. Axinella (?) tubulosa. Suherites caminatus, var. Suherites axiatus. Latruncidia apicalis. Latruncidia brevis. Station 332. March 10, 1876 ; kt. 37° 29' 0" S., loug. 27° 31' 0" W. ; South Atlantic ; depth, 2200 fathoms ; bottom, Globigerina ooze ; bottom temperature, 34°'0. Cladorhiza inversa. REPOET ON THE MONAXONIDA. 261 Table of Species obtained by the Challenger, showing their Geographical and Bathymetrical Range.^ < Coo Halichondria panicea, „ caduca, , , solida, n. sp. , . „ „ var. rugosa, aov., „ pelliculata, n. sp., ,, latriinculioideSy n. sp., sp. (a), sp. (b), . ., (?)sp.(c), Petrosia similis, n. sp., ,, ,, var. massa^ nov., . ,, ,, var. co7npacta, nov., ,, truncata, n. sp., . ,, variabilis, var., ,, hispida, n. sp., Reniera cinerea, ,, implexa, var., ,, aqiuxductus, var. infmidibularis, „ siiiglobosa, n. sp. , . ,, tufa, n. sp., Pachychalina fraijilis, u. sp., meiior, n. sp., . fibrosa, n. sp., , lobata, . megalorrhaphis, n. sp., elongata, n. sp,, (?) punctata, n. sp., (?)^ed!»ic!(tote, n. sp. , Chalina palmata, „ pergavientacea, ,, rectangularis,ri. up., sp. («), — « = 1" o K C CI Is ii.2 S'e SC- IO- TO ■100 1 This table comprises information derived distribution of previously known forms the reader from the Challenger material only; for fuller information as is referred to the Description of Genera and Species. to the 252 THE VOYAGE OF H.M.S. CHALLENGER. Table of Geographical and Bathymetrical Range — continued. Siphoiiochalina hitermedia, n. sp., „ annulata, n. sp., Rhizochalina fisUdosa (?) „ putridosa, „ singaporensis, „ pedunculata, n. sp., Oceanapia robusta, . Gellius varius, carduus, n. sp., ,, var. marjellanica, Icevis, n. sp., . glacudis, n. sp., „ var. nivea, nov., flagellifer,B.SYt., calyx, n. sp., angulatiis, fldbellifonnis, n. sp. , sp., . GeUiodes Jiiulala., lichenifomiis, pocidum, n. sp., Toxochaliiia robusta, . Tedania digiiata, ,, var. ^6tos«, nov., ,, var. bcrmudensis, nov., tenuicapitata, commixta, n. sp., . massa, n. sp., infimdibulifonnis, u. sp., actiuiiformis, n. .sp., Trachytedaaia patagonica, n. sp., . Desinacella annexa, . Vomerida esperioidcs, n. sp., EspereUa marmiifonnis, n. sp., ,, lapidiforinis, n. sp. , < Co 13 2 ""^ 4 REPORT ON THE MONAXONIDA. 253 Table of Geographical and Bathymetrical Range — continued. Esperella parishii, „ ijelatinosa, ,, nuigellanica, „ murrayi, n. sp., ,, porosa, n. sp., ,, nuda, n. sp., ,, fusca, n. sp., „ arenicola, n. sp., . ,, mnonis, n. sp., . „ biserialis, n. sp., . sp., Esperiopsis symmetrica, n. sp., ,, edwardii, var. americana, nov, ,, cylindrica, n. sp., ,, challengeri, n. sp., ,, ,, var. meangensis, nov., „ profunda, n. sp. , ,, anomala, u. sp., ,, pulchella, n. sp., Cladorhiza cCbyssicola, var. rectangularis, nov., „ „ var. linearis, nov., ,, inoruliformis, n. sp., ,, tonjrJpj'jMia, n. sp. , „ similis, n. sp., . ,, inversa, n. sp., ,, (.') tridentata, n. sp., Axoniderma mirabile, n. gen. et sp., Chondrocladia concrescens (?), ,, clavata, n. sp., ,, crinita, n. sp., Meliiderma sUpitata, n. gen. et sp. , Desmacidon. fruticosa, vsii. , . ,, reptans, n. sp., ,, conulosa, n. sp., (.!) ranwsa, n. sp., asx. ^^ *5 ft •25'? '33 5^§ ^'& — ^ ** = =3-3 -1. is 254 THE VOYAGE OF H.M.S. CHALLENGER. Table of Geographical and Bathymetrical Eange — continued. Desmacidaii [Hoiriceodiciya] kergudenemis, n. sp. , „ „ grandis, n. sp., a Artemisina suberiioides, Phelloderma radiatum, n. gen. et sp. Sideroderma navioelligeram, n. gen. et sp, lophon paitersoni, „ chelifer, n. sp., „ cylhidricus, n. sp., . „ laminaliSj n. sp., „ abnormalis, n. sp., . „ (>) omnivorus, n. sp., . Amphilectus apullmis, n. sj). , ,, ceratosus, n. sp., ,, pilo^us, u. sp., ,, annecUns, u. sp., Myxilla rosacea, var. jcqjonica, nov, „ digitala, n. sp., ,, pcmcispinata, n. sp., ,, mollis, n. ,sp., ,, spo^igiosa, n. sp., . ,, hastata, n. sji., ,, cribrigera, n. sp., . ,, Jusca, u. sp., „ mariana, n. sp., ,, ,, var. massa, nov. „ compressa, n. sp., ,, nohilis, n. sp., ,, ,, Ya.T. patagonica, nov., „ ,, var. bacilli/era, nov., ,, frondosa, n. sp., . ,, plwmosa,'v&r.fusifera,no\., Clathria acuUata, „ decumbens, . „ hndenfeldi, n. sp., ,, frondifera, . E " W i (i) rigida, 11. s])., . . . . . X X Dendropsis buleniifera, n. gen. et sp., X X Tlirinacophora cermcornis, n. gen. et sp. , X X „ funiformis, n. sp., X X Suberites eamosus, ...... X X X X X ,, caminatiis, n. sp., X X „ micro.Homus, n. sp., .... X X ,, perfectioSj n. sp., X X ,, aniarcdcus, ..... X X ,, axiatus, u. sp., ..... X X „ durissimus, n. sp., X ,, 7nollis, u. sp. , X X „ elongattis, u. sp., ..... X X „ spiralis, n. sp., . X ,, ramulosits, u. sp. , X X ,, „ var. cyliiidn/era, nov., X X ,, senilis, n. sp., ..... X X Polymastia rohusta, ...... X X ,, corticata, n. .... X X „ agglutinans, n. sp., .... X X Proteleia sollasi, n. gen. et sp., . X X Trichostemma sarsii, n. sp., .... X X X X ,, irregwlare, n. s-p. , . . . . X X Tentorium scmisvherites, ..... X X X X Slylocordyla stipitata, ..... X X X X X X „ ,, var. glohosa, nov., . X 10- -100 Quasillina brevis, ...... X X Clirnia dissimilis, n. sp., ..... X X Spvrastrella, decicmOens, var. , . . . . X X ,, massa, n. sp., ..... X X ,, solida, n. sp., . X X REPORT ON THE MONAXONIDA. 257 Tablk of Geographical and Bathymetrical Range — continued. o' s;^ «D 1 It Is 1 s2 Is2 CO 1 1 x: a 1 5 S IS'2 ■^3 II Ml II 3 i i Spirastrella papulosa, n. sp., X X Latrunculia apicalis, n. sp., X X X X „ bmis, n. sp., . X X „ bocagei, n. sp., . X 10- -70 „ (?) acerata, n. sp., .... 1 ? DISCUSSION OF THE GEOGRAPHICAL DISTRIBUTION. In glancing over the list of localities at which Monaxouid sponges were obtained, one can scarcely fail to be struck with their small number as compared with the total number of dredgings made by the Challenger. Out of a total of 277 distinct stations we find only 50 represented in the collection, supplemented by 20 "localities" to which no station number is attached. This is a very remarkable circumstance ; it is probably to be accounted for partly by the fact already pointed out in our Introduction (p. v), viz., that the Monaxonida are, as a rule, very insignificant and uninteresting in external appearance, and hence extremely likely to be overlooked amongst the rubbish in sorting out the contents of the trawls and dredges, unless a specialist be on the spot to look out for them, and also partly by the fact that the sponges in question are usually very soft and fragile, and hence peculiarly liable to destruction by the rough treatment to which they are necessarily exposed in trawling, and, more especially, dredging operations. These explanations are, however, hardly sufiicient to account entirely for the absence of Monaxonida from the great majority of localities examined by the Challenger, and we are forced to conclude that the Monaxonida are not, on the whole, a predominant group in deep water, although, as we shall subsequently show, individual genera range down to very great depths, and are occasionally, but very rarely, found in great abundance in deep water {e.g., Station 320). Thus, as the Challenger confined its operations mainly to deep water, we have an explanation of the facts before us ; an explanation precisely similar in kind, though not so far reaching in degree, as that arrived at by PolejaefF in the analogous case of the Keratosa. Conversely, we find that in those cases where explorations were made by the Challenger in shallow or comparatively shallow water, rich harvests of Monaxonid (ZOOL. CHALL. EXP. — PAHT LIX. — 1887.) Nnn 33 258 THE VOYAGE OF H.M.S. CHALLENGER. sponges were commonly obtained. We shall, however, return to tke question of bathymetrical distribution later on, and refer to it in this place only in explanation of a very striking fact which meets us at the outset. Comparatively little is as yet known of the geographical distribution of sponges at large, and still less, of course, of the distribution of the Monaxonida. Vosmaer,^ it is true, has dealt with the subject in some detail, and constructed tables both of geographical and bathymetrical range, but his results are far from being satisfactory. His list shows that sponges are most abundant in the Mediterranean and Atlantic Oceans, but, as he himself points out, this is only because the Mediterranean and Atlantic have been more or less thoroughly explored — witness Schmidt's works on the Atlantic, Adriatic and Algerian sponges, and Carter's researches on the Atlantic sponges obtained by the " Porcupine," — while the sponge faun^ of other seas and oceans is or has hitherto been almost entirely unknown. Hence it follows that the results of the Challenger Expedition taken alone are likely to give more trustworthy information on this head than those of all previous workers, simply because the observations were ex- tended more impartially over an enormous area. Vosmaer, then, makes six geographical areas, which, he tells us, are "voUkommen kiinstlich ; " viz., Mediterranean, Atlantic Ocean, Pacific Ocean, Indian Ocean, Arctic Ocean, and Antarctic Ocean. Such an arrangement may be very suitable for general purposes, but is obviously inapplicable to the present case, as will be seen by a glance at the chart which illustrates our remarks on this head. The areas which we distinguish (vide Chart) are necessarily made to suit our own requirements. This is unfortunate, as it makes comparison a somewhat more difficult matter, but it cannot be avoided. One or two peculiarities strike us in examining Vosmaer's table. Leaving out of account the Spongillids, with which we are not concerned in this place, we find here set forth the geographical distribution of 74 genera, or so-called genera, of Monaxonid sponges. Of these, two, viz., Amjihilectus and Clathria, are stated to be cosmopolitan. The first, inasmuch as it is, as is stated^ by Vosmaer himself, "nur ein vorliiufiges Aushilfe- Genus, wo alle diejenigen Desmacidinen untergebraclit sind, von welch en keine besondere Merkmale bekannt sind, wodurch sie zu einer anderen Gattung kommen sollten," might naturally be expected to be more or less cosmopolitan, while we are inclined altogether to deny the cosmopolitan character of the second {Clathria), at any rate until more evidence is forthcoming. It is not a little surprising to find that, according to Vosmaer, Monaxonid sponges^ have been hitherto altogether imknown from the Antarctic Ocean, and that only nine genera are recorded from the Indian Ocean. Of course, in the present state of our knowledge, the only definite conclusion to be drawn from this is that ' Bronn's Klass. u. Ordnung. d. Thierrichs, Porifera, p. 447. ^ Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 353. ' Perhaps we ought to except the genera Amjjhilectus and Clathria, though the Antarctic Ocean is about the last place in which we should expect to find the latter. EEPORT ON THE MONAXONIDA. 259 these two areas liaA'e been less explored than elsewhere, and are especially deserving of attention at the hands of collectors, but Dr. Vosmaer must surely have overlooked Mr. Carter's Suberites antarcticus (cf. p. 201). Unfortunately the Challenger adds no information concerning the Indian Ocean, and this little known field will probably yield a rich harvest to whoever has the good luck to thoroughly investigate it.^ If we arrange our own seven geographical areas according to the number of species * obtained in each, we arrive at the following result : — Indo-Australian (74 species), South Atlantic (52 species). Southern Ocean (36 species), Patagonian (21 species). North Atlantic (19 species), South Pacific (16 species), North Pacific (7 species).' Thus of the seven areas the Indo-Australian has proved to be the most prolific and the North Pacific the least so, and with this we must correlate the fact that the waters in the former area are, comj)ared to those of the North Pacific, very shallow. Of individual stations, however. Station 330 (off" the mouth of the Rio de la Plata, 600 fathoms) has yielded by far the richest harvest. No less than 22 species were obtained here, presumably at a single haul of the trawl. These species were all of them new to science and most of them (e.g., Halichondria latrunculioides, Gellius calyx, Gellius fiabelliformis, Tedania massa, Esperella lapidiformis, Phelloderma radiatum, Ciocalypta hycdoderma, Latruncidia apicalis and Latrunculia hrevis) of great and exceptional interest. This locality, indeed, seems to be a great focus of Monaxonid sponge life, where the conditions must be exceedingly favourable, and where not only is the diversity of species very great, but the individual species attain a high degree of development and a luxuriant growth, some of them (e.g., Tedania massa) reaching huge proportions, and being represented in the collection by considerable numbers of specimens. Simon's Bay, Cape of Good Hope, also proved to be of exceptional interest for the remarkable new species obtained there, for two of which we have been obliged to establish new genera {Dendrop>sis and Proteleia), and the seas in the neighbourhood of Port Jackson also yielded a very rich harvest, including, amongst a total of 17 species, such remarkable forms as Siplionochalina intermedia, Esperella murrayi, Plutkelliajlahellata, Spirastrella p)apillosa, and others. In these two cases, however, we do not know how much time and trouble was spent upon the collection of the material, so that it is hardly fair to compare them with individual stations where (presumably) only a single haul was taken. Very interesting and novel results were also obtained at stations in the Pacific, Atlantic, and Southern Oceans where the depth exceeded 1000 fathoms, but in these cases the number of species obtained at each locality was always very small, the greatest 1 Cf., however, the Report ou the Zoological Collections of H3I.S "Alert," Brit. Mus., 1884; and also Mr. Carter's papers on Sponges from the Gulf of Manaar, Ann. and Mag. Nat. Hist., ser. 5, vol. vi. p. 35 and ser. 5, vol. vii. p. 361. 2 For the sake of simplicity we here include varieties as distinct species. 3 For the localities included by us in each area, vide Chart and Table of Geographical and Bathymetrical Range. 260 THE VOYAGE OF H.M.S. CHALLENGER. being 6 (Station 147, IGOO fathoms) ; but this question will be more fully dealt with in discussing the bathymetrical distribution. The geographical range of individual species^ appears, with few exceptions, to be somewhat restricted. As exceptions we may mention Halichondria panicea, Reniera cinerea, Chalina i^crgamentacea, Tedania digitata, lophon jxittersoni, Axinella jJrofiaida, Suberites carnosus, Tentorium semisuhei'ites and Latruncidia apicalis. Five of these (Nos. 1, 2, 4, 5, 7) are very generalised types, with few definite specific characters to go by, and might consequently be expected to have a very wide range. One, Axinella profunda, comes from very deep water (2300 and 2385 fathoms) at widely remote localities in the North and South Pacific respectively (cf. p. 181), but apparently under exactly similar conditions in the two cases ; while the distribution of the remaining three is very hard to understand. The genera and families, on the other hand, have mostly a very wide range, although some appear to be characteristic of special localities and to be comparatively rare elsewhere. Thus the Chalininse and. Eetyoninse are highly characteristic of the Indo-Australian area, and the Tedaniinse more or less so of the Patagonian area. DISCUSSION OF THE BATHYMETRICAL DISTRIBUTION. We have already had occasion to point out that the Monaxonida are not a predominant group in very deep water. If we analyse our table of bathymetrical distribution as we did that of geographical distribution we shall find this conclusion fully confirmed, especially if we bear in mind the large proportion of Challenger dredgings which were taken in deep water as compared with those taken in shallow water. Our four bathy- metrical areas may then be arranged as follows, according to the number of species^ obtained in each : — 0-.50 fathoms (85 species); 50-200 fathoms (55 species); 200-1000 fathoms (46 species); 1000-3000 fathoms (24 species). Here then we have the very definite result that the number of species varies in inverse proportion to the depth, or in other words that the number of species gradually diminishes as we go downwards. According to Vosmaer •' (and we have no reason to doubt the correctness of his statement) the greatest depth at which a Monaxonid sponge has hitherto been obtained is 860 fathoms (recorded for Cladorhiza). This depth is now extended to 3000 fathoms (for Cladorhiza longipinna, nobis) by the results of the Challenger investigations, which 1 For details as to the geographical distribution of individual species the reader is referred to the Table and to the Description of Genera and Species. 2 For the purpose of simplifying the discussion, varieties are here again treated as distinct species ; which these are will be seen by reference to the Table of Geographical and Bathymetrical Range. 3 Bronn's Klass. u. Ordnung. d. Thierreichs, Porifera, p. 455. REPORT ON THE MONAXONIDA. 261 have brought to light no less than 24 species of Monaxonid sponges (belonging to 14 genera) which live below the thousand fathom line, amongst which 13 live at depths of 2000 fathoms or more, as will be seen from the following list : — List of Species which live below the 1000 fathom line. Species. Depth in fathoms. 1. Tedania actiniiformis, • 2160 2. Esperella mammiformis, . 1600 3. „ hiserialis, . . 2250 and 2385 4. Esperiopsis profunda, . 1600 5. Cladorhiza ahyssicola, var. t ^ectangidaris, , 2750 6. „ „ var. / Hnearis, 2385 7. ,, mondiformis, . . • 1950 8. „ longipinna, . 3000 9. „ similis, . . 2385 10. „ inversa, . , . 2200 11. „ (1) tridentata, , , 1600 12. Axoniderma mirahile. , , 2250 13. Chondrocladia concrescens, , , 2900 14. ,, crinita, • . 2000 15. Meliiderina stipitatum, . • 1600 16. Phakellia ventilahrum, var. connexiva, 1035 17. Axinella profunda, . . . 2300 and 2385 18. „ erecta, , , 1600 19. Suherites senilis. . 2050 20. Polymastia corticata, 1200 21. Trichostemma sarsii. . . . . 1000 and 1400 22. ,, irregidare, , 2160 23. Tentoritim semisidterites. , 1250 24. Stylocordyla stipitata, . 1600 Of the 14 genera thus represented, 9 {Cladorhiza, Axoniderma, Chondrocladia, Meliiderma, Phakellia, Polymastia, Trichostemma , Tentorium and Stylocordyla) must be regarded as characteristically belonging to the deep-sea fauna, being rarely or never met with in shallow water, while the remaining j&ve are on the contrary typically more or less shallow-water forms, and are rarely met with at very great depths. The scarcity of Monaxonid sponges at very great depths is fully compensated for by the unusual interest attaching to the few species which do occur. In examining the twenty- 262 THE VOYAGE OF H.M.S. CHALLENGETl. four species just enumerated, and comparing them with their shallow-water allies, we cannot fail to be struck with one very remarkable fact, and this is that while the shallow water species are characteristically more or less amorphous in external form, or, at the most, simply digitate or ramose, those from below the thousand fathom line have, almost without exception, beautifully symmetrical and definite shapes. This interesting circum- stance is prolialjly to be accounted for by the fact that the deep-sea forms are exposed to precisely similar external conditions for very long periods of time, and variations in any one particular direction (which prove to be advantageous to the species under its particular external conditions) have time to develop into fixed and definite characters ; while in the shallow-water forms the ever changing external conditions necessitate corresponding changes in the sponge, and no external forms can become fixed and permanent, excepting such as, from their very general and unpronounced character, are suited to the ever varying conditions under which they are placed. We have already had occasion to show, in discussing the value of external form as a guide to classification, that it is of all characters perhaps the most variable and, consequently, the least trustworthy. The particular external form which a deep-sea sponge assumes may not infrequently be explained by reference to the conditions under which it lives, and in this fact lies strong confirmation of the views to w'hich we have just given expression. Thus in Tedania actiniiformis, a sponge whose shallow-water congeners are charac- teristically shapeless, the peculiar external appearance (PL XL fig. 2), and chiefly the arrangement of the pores and oscula, are explained Ijy the fact that the sponge lives on a bottom of mud in which it is nearly buried. Many of the most remarkable external forms which we know are due to the necessity of obtaining some means of support to prevent the sponge from sinking bodily into the soft mud or ooze which is so prevalent at great depths. Thus we can easUy account for the " Crinorhiza " forms already discussed,^ and for the analogous " Tricho- stemma " forms.^ Nor need we be surprised that species of distinct (though allied) genera {e.g., Cladorhiza, Axoniderma, and Chondrocladia), all living under precisely similar conditions, arrive at a precisely similar solution of this difficult problem of support, viz., the " Crinorhiza" form. Other perfectly definite and characteristic external forms found in deep-sea sponges are, however, as yet unaccounted for. Amongst these we may mention those of Stylo- coi'dyla and Tentorium; but by far the most extraordinary and beautiful is that exhibited by Esperiopsis challengeri (PL XVIIL) which has been fully described elsewhere," and which comes from a depth of 825 fathoms (hence it is not included in the foregoing list, which contains only species from a depth of over 1000 fathoms, though of course a deep-sea form). ' Vide p. 87. ^ yide p. 216 £< seq. ' 3 Vide p. 80. REPORT ON THE MONAXONIDA. 263 We may briefly sum ujd our conclusions with regard to the geographical and bathymetrical distribution as follows : — 1. The Monaxonida as a group are cosmopolitan. 2. They are most abundant in the Indo-Australian area. 3. Individual species have not usually a very wide range, and rarely or never except in the case of generalised types, or where the conditions of life are very similar, as in the case of Axinella lyrofunda. 4. They are most abundant in shallow water and gradually decrease in numbers as we go downwards. The limit of depth, so far as we at present know, is 3000 fathoms. 5. The shallow- water species are characteristically without any definite, sym- metrical external form ; but in the abyssal species a perfectly definite and usually symmetrical external form is almost invariably present. APPENDIX. SYNONYMY OF THE SPICULAR TERMINOLOGY. The synonymy here given does not pretend to be complete. It is intended for the use of those who are already acquainted with the principal terms used by otlier authors. For further synonyms of continental authors vide Vosmaer's article on the Porifera (p. 151) in Bronn's Klassen und Ordnungen des Thierreichs. The symbol = must here be interpreted in a wide sense. The terms used by us include but are not necessarily co-extensive with those of other authors given. 1. Megasdera Oxea Tornota Strongyla Tylota Slyli Tylostyli SuhtyJostyli Cladostrongyla Cladotylota 2. Microsdera Rhaphides Trichodragmata Toxa Toxodragmafa Sigmata Sigmadragmaia Diandstra Chelx Isodielx Anisodielx BijMdlli Spirulx Discastra Ampldastra skeleton spicules (audorum). acerates (Bowerbank); ac.-, ac. ac. (Vosmaer). hastately pointed acerates. cylindricals (Bowerbank, &c.); tr.- (Vosmaer). biclavated cylindricals (Bowerbank); bicapitate cylindricals; tibieUs (Carter); tr"- (Vosmaer). acuates (Bowerbank); tr. ac. (Vosmaer). spinulates (Bowerbank); tr°ac (Vosmaer). subspinulates (audorum). (new type). spinulo-recurvo-quaternates (? &c.) (Bowerbank). flesh spicules {audorum). trichites {audiyrum). trichite sheaves, or bundles {audonun). tricurvates {audorum) ; tricurvato-acerates (Bowerbank) ; /\ (Vosmaer). tricurvate sheaves. bihamates (Bowerbank); CV3 (Vosmaer). bihamate sheaves. trenchant bihamates (Bowerbank) ; ^i^ (Vosmaer). anchorates (Bowerbank). equianchorates (Bowerbank); anc- (Vosmaer); rut." (Vosmaer). inequianchorates (Bowerbank); anc. anc. (Vosmaer); rut. rut. (Vosmaer). bipociUate bihamates (Bowerbank) ; grotesque spicules (Carter), multiangulated cylindricals, spinulo-multiangulated cylindricals, elongo-attenu- ato-stellates (Bowerbank); spini-spiruli'e (Carter), sceptrellaj (Carter), birotulates (Bowerbank); amphidisks {audorum). INDEX. Note. — The numbers of the pages on which the species, &c., are most fully dealt with are printed in darker type. abnormalis (loplion), . abyssieola (Cladorliiza), abyssicola (Cladorliiza), ■ var. corticocancel lata, abyssicola (Cladorhiza), var. linearis, abyssicola (Cladorhiza) var. rectangularis, Acanthella, . Acanthospongia, . Acarnus, acerata (Latrtmcvdia?) Acervochalina, . Acervocbalinida, Actinia, actiniiformis (Tedania), Plate Page XVII. 122, 243, 254 j 86, 88,90,91,248, \ 253, 261 91, 98 XX., XXI. 90, 248, 253, 261 XX. 86, 88, 253, 261 XXSII. aculeata (Clathria), acuta (Acanthella), adriatica (Euspongia ) officinalis, var.), . i segagropila (Esperia), aerophoba (Dendrilla), affinis (Microciona), Agelas, agglutinans (Polynias tia), Alcyoncellida;, Alcyoncellum, Alcyonium, Alebion, AlgiL' (parasitic, he). anibigua (Ciibrochalina), anibigua (Reniera), anibigua (Tedania), americana (Esperiopsis edwardii, var.), Amorpliina, (ZOOL. CHAI.L. EXP. PART LIX. 1887. i xxii, xxiii, lis, \ 176, 188, 246, 255 viii xvii, lix, Ix, 148, 159, 248, 255 XXIX. 231, 239, 242, 257 26 ix 50 xli, 50, 55, 248, 2, 261, 262 147, 246, 254 176, 177 XI., XXIX /xli, ■ \ 252, lii Hi xxiv 110 XXIX. lix, 163, 242, 255 XLii /''-''■^i"' l-'fi. 213, '1241,256 viii liv 33, 34 116, 117 3, 67, 72 172 51 51 XIX. 78, 249, 253 1, 2, 167 Plate XXV., XLVII. amorphosa (Ciocalypta), aniphactis (Crinorhiza), Amphilectus, Amphipoda (infesting \ sponge), . . . } Ancorinidse, angulata (Halicliondria), angulatus (Desmacodes), angulatus (Gellius), annectens (Amphilectus), annexa (Desraacella), . annexa (Desmacella 1 vagabunda, var.), . J annexa (Desmacodes 1 vagabundus, var. ), . / annulata (Siphonocha- 1 liua), . . .J anomala (Esperiopsis), antarcticus (Suberites), Aphrocallistidse, . Aphrocerasidse, . apioalis (Latrunculia), XLiv., XL v., Li. Aplysinida, apollinis (Amphilectus), xix appeudiculatuui (Histo- } derma), . . . i aqujeductus (Keniera), I., Ii. arborescens (Axiuella), xxxv. arciger (Suberites), arcoferus (Gellius), arenicola (Esperella), . xv., xvi. Arenospongia, Page 173, 175, 250, 255 88 xviii,lix,65,78,85, 123,149,243,244, 245, 250, 254, 258 124 viii, ix 44 44 38, 44, 46, 241, 252 127, 250, 254 59, 241, 252 59 59 30, 31, 244, 252 84, 248, 253 /201, 203, 244, ■ I 256, 259 viii viii r xxxii, XXX viii, xxxix, xliii, xliv, xlv, xlvi, xlvii, ' xlviii, 7, 234, 237, 238, 239, 244, 250, 257, 259, 260 ix 124, 244, 254 115 16, 249, 251 178, 245, 255 125 38, 46 Hxi, 65, 72, 244, I 253 viii Nnn 34 266 THE VOYAGE OF H.M.S. CHALLENGER. Plate armigei'a (Cladoclialiua), Artemisina, Page 27 lix, 112, 124, 125, Askoncmatids, asperum (Echino dictyum), Asychis, Aulena, Auletta, aulopora (Schmidtia), Aulos2iongus, au-stralieusis (Raspailia), axiatus (Suberites). Axidie, ;} Axinella, Axinellida, . Axinellidse, Axonidei-ma, Axos, .... bacillifera (Myxilla \ nobilis, var. ), . . J balfourensis (Axinella), bermudensis (Tedania \ digitato, var.), . . i Bibulida, bidentifera (Dendrop sis), . Bicmna, bihamigera (Myxilla ( (Microcioiia)), . . ) biserialis (Esperella), bocagei (Latrunoulia), borassus (Esperia), boreale (Hyalonema), borcalis (Stylocordyla) bowerbanki (Phakellia Brachiopod shell (en riisted by sponge), brevis (Euplectella), brevis (Latrunculia), brevis (Polymastia), brevis (Quasillina), Bubaris, fXXV., XXXIII., I XXXIV., XXXV., XXXVI.,XXXVII., XXX VIII., XL., XL IX. XX., XXI.. ( lix, 112, 12^ 1241, 254 165, 248, 255 37 xxiv 187 10, 13 156 190 fxxvi, 201, 203, ■ I 250, 256 viii r xxii, xxiii, xxx, xxxiv, xliv, xlv, xlvi, Iv, lix, Ixii, 170, 173, 178, 242, 243,244,245,247, 248, 249, 250, 255, 1-256, 260, 261, 263 ix , XXV, xxvi, xlii, I liv, Iv, lix, Ix, Ixii, I Ixiii, Ixvii, 166, 1.207 ,xxi, xxvii, xxviii, lliii, lix, 87, 88, 93, 96, 102, 248, ^253, 261, 262 47 XXVII. XXXIII. \XXXVIII., XL., I XLVI. XLIV., XLV. XLIV., XLV, 142, 248, 254 179, 244, 255 51, 241, 252 ix / .xvii, 192, 243, 1256 32, 36 139 rl, li, 62, 75, 90, \ 248, 253, 261 (■237, 238, 244, I 257 78 223 223 xlvi 198 226 / 236, 238, 239, t 250, 257, 259 226 225, 226, 241, 256 182 bullata (Tuba), . bursa (Weberella = Polymastia), Bursalina, . Cacospongia, caduca (Haliclioudria), Calcarea, Calcispongia, Calcispongise, caliciformi.s (Spongia), calva (Ciocalypta), calyx (Gellius), . camiuatus (Subei'ites), cancellatum (Echino dictyum), carduus (Gellius), carduus (Spongia), Carnosa, oarnosa (Halichondria), camosa (Hymeniacidon), carnosa (Microciona), . carnosus (Suberites), . Carteriadje, . carteri (Echinoclathria), Plate Page 49 xlvi, xlvii 225, 226, 227 xxxiii II. 3, 249, 251 vi, vii, viii, x, 88 xi vi, ix 172 173 vin.,xin.p8' 43, 46,250, (.252,269 r xxx, xliii, xlv, XLL, XLV.j 198,200,243,250, 1 256 165 vin.,xiii.P«'39,40,41,243, 1 244, 249, 252 177 vi, ix 197 197 145 (■197,241,242,245, I 256, 260 XXIX., XXX,.. I IS"' 161' 162, 1 245, 255 caruncula (Amorphina), 167 caruncula (Hymenia- " cidon), . ■ 1 XXXIII ■ xlv, 166, 167, 242, .255 caruncula (Reniera), 167 Casuladse, . viii caulil'era (Pachychalina), 25 Cavochalinida, ix Ceraospongiw, vi, ix, bd Ceratellada:, vii Ceratina, vi, ix, X Ceratosa, . xi ceratosus (Amphilectus), XIX., XXV., XLVII. 125, 245, 254 cervicornis (Thrinaco- | phora), . . . i XXXVI., XL 193, 194, 247, 256 (-vii, XXV, Iviii, Chalina, v., XLVI. Ixiii, 15, 25, 108, 109, 244, 246, 247, ''251, 260 Chalinese, . ... ix, 18 Chalinida, . • •• ix Chalinida", . vii, viii Chalinina, . 18 I liv, Iv, Ivi, Iviii, Chalinina;, . ... < Ix, Ixi, Ixiii, Ixvii, ' 18, 49, 260 Chalinopsidinw, . ix Chalinopsis, 163 Chaliuula,. Hi, 26 REPORT ON THE MONAXONIDA. 267 hallengeri (Amphi- \ lectus), . . . ) chall^ngeri (Esperi- \ opsis), . . . / chelifer (lopbon), Chondrilladfe, Oliondrocladia, . Plate XVIll., XIX. XVI., XVII. Page 80 xxii, xxiii, xli, xliii, xlv, xlviii, I 80, 246, 247, 253, 262 xxi, xxviii, 119, 243, 244, 254 viii ■ xix, xxvii, xlix, 1, I liii, lix, 86, 88, 94, I 97, 99, 102, 245, ' 247, 253, 261, 262 Chondrosiosa, Choristida, . cinerea (Halichondria), cinerea (Isodictya), cinerea (Reniera), cinerea (Spongia), Ciocalypta, . Cladochalina, Cladorliiza, clathrata (Raspailia), . clathrata (Tenacia), XXXIII., XL, {xlii, liv, 250, 255, 15 15 15, 247, 251, 260 15 !ix, 173, 255, 259 22, 26 rxx, xxi, xxvii, I xxxiv, xlix, liii, I lix, 62, 76, 78, xx.,xx[., XLIX. -J 86, 97, 98, 99, I 102, 243, 244, 248, I 250, 253, 260, * 261, 262 190 147 Iv, lix, 9, 129, 145, Clathria, ' XXVIII., XXIX., XLVII. XXV, 125, 146, 155, 245, clavata (Chondrocladia), ClavuUna, Cliona, ClioniadiB, . coloneusis (Forcepia), . commixta (Tedania), , compacta (Petrosia i sirailis, var. ), . . f compacta (Reniera), compressa (Clatliria), . conipressa (Desmacidon ), compressa (Myxilla), . concrescens (Chondro cladia), ir,3, 154, 160, 178, 246, 254, 255, 258 XX., XXI., 100, 245, 253 fx, XXV, xxxiv, xxxvi, xlii, xliii, xlv, xlvi, xlvii, xlix, Iv, Ivi, lix, be, Ixi, Ixii, Ixvii, I Ixviii, 196 ( xlii, lix, 227, 246, XXIX. I 256 vii, viii, x, Ixviii xviii 58, 244, xxiu. C'^'' I 252 :> II., III. 12, 247, 251 17 147 106, 108 fxl, 129, 139, ^'^^"- \ 250, 254 (•xix, 99, 247, 253, X., XXI.Ugi Plate concrescens (Cladorhiza), :ellia \ Page connexiva (Phakellia ventilabrum, var, •■} ":) conulosa (Desmacidon), Conalliospongia, . coralloides (Clatliria), . Corbitellidai, coriacea (Isodictya), coriacea (Plocaraia), coriacea (Plocamia, var. elegans), coriacea (Siphonochalina), coriaceum (Dirrhopalum), CoruacuspongiiB, corticata (Latrunculia), corticata (Polymastia), corticocanccllata (Clad- orhiza), . cratera (Latrunculia), . crateriformis (Reniera), cratitius (Rhaphi- ) dophlus), . . / Crella, cretacea (Cribrochalina), Cribrella, cribrigera (Myxilla), Cribrochalina, crinita (Chondrocladia), Crinorhiza, . Crinorhiza form.s. Crustaceans (infesting \ sponges), . . . i cunctatrix (Spirastrella), cuuniughami (Espcria), cylindrica (Esperiopsis), cylindricus (lophon), . cylindrifera (Suberites \ ramulosus, var.), • I Daetylocalycidie, Dasyehalina, decumbens (Clathria), . decumbens (Spirastrella), Demospongite, Deudrilla, . Dendrupsis, Dcsmacella, Desmacellina, Desmacellinse, Desmacidinse, 99 /•xxx, xl, xli, .xlv, XXXV.. XLIX. i -^'^' ^1^'"' 170, 242, 249, 255, ^261 xxin., XXIV. j 106, 109, 112. 1 243, 253 viii 147 viii 158 158, 241, 255 XXIX., XXXI. 158 29 158 X 234, 240 XLII., XLFV. 211, 242, 256, 261 89, 91 2.33, 234, 237, 239 13 151 115 18 136 xxvn.j^l'132, 135, 141, 1.248, 254 XX., XXI. I ' \25c 18, 172 ,,, /I., 94, 101, 247, :53, 261 { 1, liii, 87. 92, 93, 94, 96, 97, 99, 101, 218, 262 xxviii, 8 XVI., XV .VII.j"'' 1245, XXIX 229, 231 67 XIX. 79, 245, 253 117, 120, 121, 254 X., xLni.|'=^^"> 208, 247, I 256 viii 19 148, 246, 254 XLV. 229, 247, 256 xi .xxiii, sxiv ( XXXVIII., XL., j xvii, lix, 191, 243, I XLVI. 1 256, 259 J Iviii, Ixv, 43, 58. (241,252 58 Iviii, Ixiv, Ixv, 58 ix 268 THE VOYAGE OF H.M.S. CHALLENGEE. Desmacidou, Desmacidonidae, Desmacodes, dickiei (Hastatus), Dictyocylindrus, . Dictyonina, diffusa (Cladochalina), diffusa (Pachychalina), digitata (Myxilla), digitata (Reuiera), digitata (Spongia), digitata (Tedania), Dirrhopalum, dissimilis (Cliona), domunciila (Suberites), Donatiadie, dura (PetrosiaV dura (Reuiera), durissimus (Suberites), Dysidea, Dysideidse, . Ecccelonida, echidnsea (Axinella), eoliidnaea (Spongia), Eohinoclatliria, . Echinoderms, Echinodictyum, . Echinonema, Echinonemata, Ecionemia, Ectyon, Ectyonida, . Eotyonidse, . Ectyoninffi, . edwardi (Amphilectus), edwardii (Esperiopsis), edwardii (Esperiopsis, \ var. americana), ■ I edwardii (Isodictya) . elegans (Plocamia 1 coriacea, var.), . I elegantula (Clatliria), . elliptica (Melonanchora), elongata (Pachychalina), elongata (Suberites), . elougata (Thecophora), "Ms Plate Page rxx, xxviii, xxxvii, XXII., XXIII., I lix, l.xiv, Ixvi, 32, XXIV., XXIX., \ 74, 103, 108, 145, XXX., XLVII. i 242, 243, 244, 245, I253, 254 /-viii, X, xviii, I xxxiii, Iv, Ivi, Ivii, -{ lix, Ix, Ixi, Ixiii, I Ixiv, Ixv, Ixvi, 62, 1. 193, 218 37, 38, 44, 58 129 liv, 189 xi 22 22 XXVI., xxvii. 129, 131, 243, 254 51 109 /51, 241, 245, 252, .260 157, 158 xxv., XXIX. xlii, 227, 246, 256 ,xlv viii 12 17 XLI., XLV. 203, 244, 256 l.xi vii 227 xxxvi. 183, 242, 255 183 XXIX., XXXI. I ^=^^'1^'!'^' 159, 1 178, 245, 255 96 xxxii.j"^' 164, 246, I 248, 255 151, 152, 153 vi, ix, Ixvii liv 163, 164 ix, 128 X, Ixvi /'XXV, liv, lix, Ixvi, } Ixvii, Ixviii, 128, I 260 78 XIX. 77, 78, 249, 253 78 78 XXIX., XXXI. 158, 241, 255 XXVIII., XXIX. 9, 149, 245, 255 XX VI. 23, 244, 251 205 221 elongatus (Suberites), . electa (Axinella), . . 4 erecta (Hymeraphia j vermiculata, var.), . 1 Esperella, Esperellinie, Esperia, Esperiada;, . esperii (Myxilla), Esperina, esperioides (Vomerula), Esperiopsis, Euplectella, Euplectelladse, Euspongia, . Farreadfe, . fasciculata (Myxilla), . favus (Echinoclathria), fenestratus (Leucophlceus), fertilis (Chalinula), fibrosa (Axinella), fibrosa (Dasychalina), . fibrosa (Pachychalina), Plate Page XLiii. xxvi,205,241, 256 XXXV., XXXVI. ■) 170, ISO, 182, 242 XL. / 243, 244, 255, 261 182 r xviii, XX, xxv, xxviii, xxix, xxx, xxxi, xxxiii,xxxiv, xxxvii, xxxviii, xxxix, xliii, xliv, xlv, xlvi, xlvii, 'xiii., XIV., XV., J xlviii, 1, li, lii, Iv, XVI., XLViu. Ivi, lix, Ixv, Ixvi, 40, 56, 61, 62, 77, 78, 90, 96, 119, 198, 242, 243, 244, 245, 246, 247, 248, 249, 250, 252, 253, U69, 261 / xxiv, lix, Ixii, t Ixvi, Ixvii, 62, 129 xlix, lii, 62, 64, 65, 66, 67, 77, 78, 117 viii 130 62 r xxviii, xxxvii, XII., xvii.J 60, 74, 243, 250, V262 fxxii, xxiii, xxxiii, xli, xliii, xlv, xlviii, lix, 76, 124, 133, 243, 245, 246, 247, 248, 249, 253, ^261, 262 210, 226 {i xvin., XIX. XXVI., XXX., - XL VI VIU lii, Ixi, 125 viii 130 fibrosa (Tedania digi- tata, var.), Fibularia, . fibulata (Axos), . fibulata (Gelliodes), fibulata (Rhizochalina), fibulatus (Desmacodes), Fibulia, filifer (Rhaphidophlus), ;} XXXI / \ xxv, 160, 162, 245, 255 169 lii XXXVII Iv, 183, 249, 255 21 IV. t xlviii, 21, 242, 247, 251 51, 245, 252 37, 47 47 XII. 47, 246, 25i: 37 38 37 II., XLVI. 152 247, 255 REPORT ON THE MONAXONIDA. •269 Plate Page filigrana (Eeniera), 1, Hi finitima (Chalina), 28, 29 fistulatus (Suberites), . 126 fistulosa (Desmacidon), 32 fistulosa (Rhizochalina), VIII., IX. 6, 32,34, 35,241, iTl < (? o r .T '} flabellata (Pliakellia), . flabelliformis (Gellius), flagellifer (Gellius), flagelliformis (Raspailia). Foliolina, . Foraminifera (in sponge), . Forcepia, forcipis (Halicliondria), fiagilis (Dasychalina), . fragilis (Pachyclialina), frondifer (Amphilectus), frondifer (Clathria), frondifera (Halichondria), frondosa (ilj^itla,), fruticosa (Desmacidon), fruticosa (Halichondria), fruticosa (Spongia), funiformis (Thrinaco phora), fusca (Esperella), fusca (Myxilla), fusifera (Myxilla (? plumosa, Tar.), gelatinosa (EsperoUa), gelatinosa.(Esperia), Gelliada;, Gelliina, Gelliinffi, GelUodes, Gellius, Geodia, Geodiadfe, . Geodinidse, . glabra (Echinoclathria), glacialis (Gellius), globosa (Stylocordyla 1 stipitata, var.), . / I 245, 255, 259 XXVI. I ^^'' 38,45,250, I 252, 259 XIII. 42, 243, 252 XXXIX. 190, 191, 243, 256 82 137, 185 xviii xviii, 136 20 IV. 20, 21, 247, 251 149 fl49, 160, 178, 1246,254 149 f xxxvii, xli, Ixviii, XXYI., XXVII. -^ 144, 146, 147, U45, 254 XXIII., XXX. 104, 105, 245, 253 104 104 ^ fxvii, 193, 195, '1242, 256 XIV., XV. xlv, 72, 242, 253 XXVII. 136, 13S, 244, 254 145, 242, 254 fxxx, xlv, 66, 246, XVI. } ' ' < ' I 253 viii, 37 37 Iviii, Ixiv, Ixv, 37 ("Iv, Iviii, Ixi, Ixiii, 47, 242, 245, 246, 1 252 /-xli, xlii, Iv, Iviii, ^jjj I Ixiv, 36, 37, 58, ^^;;^;24i, 243, 244, XXIII., XXIV. X., xir., XLvii. VIII., I 247, 249, 250, 1 252, 259 liv viii ix XXIX., XXXI. 163, 245, 255 VIII., XIII. 41, 42, 243, 252 fxxvi, XXX, xxxi, 1 xxxii, xxxvii, xliii, XLIII. , L. "I xlv, xlvi, xlvii, I 206, 224, 243, I244, 256 gracilis (Amphilectus), gracilis (Echinoclathria), gracilis (Echinonema), gracilis (Isodictya), gracilis (Rhaphidophlus), grandis (Desmacidon \ (Homceodictya)), . / Grantiadue, Gumminese, Gumminida, Halicnemia, Halichondria, Halichondriadae, Halichondridfe, Halichondi'ina, . Halicnemia, Haliphysema, Halisarca, . Halisarcida, Hali.sarcinae, Halisarcosa, Hamacantha, Hamacanthidse, HamacanthiniE, hamigera (Cribrella), Hamispongia, hastata (Myxilla), Hastatus, . hemisphoericum (Ti'ich ostemma), Heterociela, Heterorrhaphidae, Hexactinellida, Hexactinellidse, Higginsia, . Hirciniadfe, Hircinida, . hispida (Petrosia), Plate 152, 242, XXII., XXIX. If™' 106' 1111,243,254 78 160 152 78 255 109, viii ix Iz liv I., II., XLVI. ■{ / xviii, XXV, xxxiii, 1 xxxviii, xl, xliii, I xliv, xlv, xlviii, ! Hi, Ivi, Iviii, 1, 38, I 39, 44, 59, 103, i 108,117,134,136, j 145, 149, 156, 197, 243, 244, 245, 246, I 247, 248, 249, 250, L251, 259, 260 rx. Vll, VUl x-viv, XXV, XXXlll, xxxiv. ! xxxn, xlii, xHii, ■ xliv, xlv, xlvii, xlix. Hi, liv, Iv, Iviii, Ix, Ixi, Ixii, ^Ixviii, 1 218 liv, 156 1 i.x ix xi / xxviii, Lxvi, 58, I 59, 60, 61, 74 viii, 59 /xviii, Iviii, Ixiv, 1 Ixv, 59 1.36 viii 134, 250, 254 128, 129 xlv, 216, 217, 219 r xxiv, liv, Ivi, -] Iviii, Ixi, Ixiii, V Lxiv, lxvi, 31 vi, ix, X, xi vi, ix 192 vii ix {xxxvii, -xlv, xlviii, 14, 244, 251 270 THE VOYAGE OF H.M.S. CHALLENGER. Plate Pago hispijus (Diotyocylin- dius), Histoderma, Holoraphidota, . Homocasla, . Homceodictya, 189 ■( XXII., XXIV. , XXIII., XXIX. Homorrliaphid;i',. horrida (Khapliido- 1 , , ,, / ^ l-xxviii. pnlus) lobatus, var., . J hyalina (Hymeniaci- 1 dou?), . . . J hyaloderma(Ciocalypta), Hyaloiiema, Hyalo.spongi;c, . . HyalothaumadiE, Hymedesmia, . . \l XXIX. 115 vi, ix X xxxvii, xlviii, lix, Ixiii, 27, 104, 106, 108, 243, 244, 254 xxiv, liv, Ivi, Iviii, xi, Ixiii, IxiT, Ixv, 153, 243, 255 Hymeniacidon, . • Hymeraphia, Hymerrhaphia, . hyiidmani (Alebion), . hynJmani (E.speria), . hyndmani (Halichon- 1 dria), . . -J ibia (Thecopbora), implexa (Reniera), iuanchorata (Clatbria), infmidibulifoi'mis (Iso- dictya), . infuudibularis (Reniera 1 aqiuuductus, var.), infuiidibuliformis (Ted ania), intermedia (Sipbono- chalina), . inversa (Cladorbiza), XLV. 168, 249, 255 xxx.ii.-f 173, 174, 176, (. 250, 255, 259 222, 223 X viii liv. 59, 61, 129 ^xlv, lix, 129, 131, XXXIII., XXXIX., , 143, 145, 156, 166, XL., XLV. I 197, 227, 242, 247, "249, 255 liv, 182 129 117 117 117 XXVIII., X.XIX. 221, 222 15, 241, 251 147, 150, 245, 255 172 )., 11. 16, 249, 251 XI., XXIX. XX., XX,. [9^ l2e lopbon, irregulare temma), Isodictya, (Trichos :} isodiotyalis (Haliebon- 1 dria), . . . i Isopod (infesting sponge), japnniea (Myxilla acea, var.). xlviii, 50, 54, 56, 249, 262 XLVi /30, 31, 245, 252, I 259 93, 94, 250, 253, 261 , xxi, xxviii, lix, \ Ixi, Ixv, 116, 137, 1 242, 243, 244, 245, ' 249, 254, 260 f219, 220, 248, 1256, 261. j-lxiv, 14, 36, 38, A 39, 78, 108, 109, U58, 172 XVI., XVII. 134 40 roK-l XXVI., x.xvii., (■-"^'^".^ .[ XLV„.il29, ) 1247,2, XXVI, xxvii, xxxvii, 130, 1 40, 54 jeffreysii (Biemna), jeffreysii (Desmacidon), jeffreysii (Oceanapia), johnsoni (Hamacantha), johnsoni (Hymedesmia), Keratosa, Keratospougia, . kerguelenensis (Des^ macidon (Homceo dictya)), . lacustris (Spongilla), Iffivis (Gellius), , laminalis (lophon), Lanuginellidse, lanugo (Esperia), lapidiformis (Esperella), j Plate Page 3C Ixiv, 32, 36 36 xxviii, 60, 61, 74 59, 61 ( vii, viii, xi, Iv, t Ivi, Ix, 257 vii _ r xxxvii, 109, 110, XXIII., XXIV 112, 244, 254 Latrunculia, latrunculioides (Hali- cbondria), leei (Ciocalypta), Leiospongia, lendenfeldi (Clatbria), Leucophloeus, licbeniformis (Gel- liodes), licbeniformis (Spongia), limbata (Chalina), linearis (Cladorbiza abyssicola, var.), lingua (Esperella), Litbistida, . Litbistidfe, . Litbistina, . litoralis (Reniera), lobata (Clatbria), lobata (Pachycbalina), y lobata (Spongia), lobatus (Rbapbidopb- } lus), var. horrida, . 1 lobularis (Halisarca), . lobularis (Oscarella), . longipinna (Cladorbiza), Lophurellida;, lovenii (Podospongia), . Luffarida, , xvii, 193 xrii. 40, 250, 252 XVI., xvii. 120, 121, 243, 254 viii 65 (XXX, xxxi, xxxvii, xxxix, XIV, xlviii, I, lii, 64, 250, 252, 259 fXxi, xxvii, xxviii, xxxii, xxxviii, xxxix, xliii, xliv, I XXIX., xLiv., ^ xlv, xlvi, xlvii, xlviii, lix, Ixii, Ixviii, 7, 231,233, 242, 244, 250, 257, '^ 259, 260 xxxviii, xl, xlviii, 6, 250, 251, 259 173, 174 vii xxv, 148, 245, .254 169 1., II., XLVl. \ XXVIII., f XXV I XXIX., XLVII. 1254 XII., xi.vii. 47, 48, 242, 252 48 28 90, 248, 253 xxxix xi vi, ix x 18 153, 154 V XLVI -f-^''''' xxxviii, 22, ■ I 245, 251 22 xxviii., XXIX. 153, 243, 255 1 1 vv vv, /92, 98, 248, 253, ■ I 260, 261 vm 234 ix REPORT ON THE MONAXONIDA. 271 lunsecharta (Axinella ?), luridus (Hastatus), Lyssakiiia, . MacandrewiadiE, . niagellanica (Esperella), magellanica (Esperia), magellanica (Gellius\ carduus, var.), • . J mammiformis (Esperella), mammillaris (Poly- \ mastia), . J mamniillata (Plumo- 1 halichoudria), . . J mariana (Axinella), mariana (Myxilla), massa (Fibularia), massa (Myxilla mari- 1 ana, var. ), . . J massa (Petrosia) similis, 1 var., . . . / massa (Spirastrella), massa (Tedania), . mastophora (Axinella), mauritianus (Agelas), . mauritianus (Ectyon), . meangensis (Esperiop- \ sis challenger!, var.), ) megalirrliapUis (Amor- \ phina), . . . I megalorliapbis (Amor- 1 phina), . . . i megalorrhaphis (Amor- 1 phina), . . . J megalorrhaphis (Pachy- 1 chalina), . . . f Meliiderma, melior (Dasychalina), . melior (Pachychalina), Melonanchora, Microciona, microcionides (Plumo- \ halichondria), . . J microstomus (Suberites), mirabilc (Axoniderma), mirabile (Trochodernia), mollis (Myxilla), . mollis (Suberites), Monactinellidse, . Monaxona, . Monaxonia, . Moxanonida, Monaxonidse, >late Page XXXVII. 186, 188, 242, 255 129 xi viii 67, 249, 253 67 XIII. 40, 249, 252 XIV., XV. X.XX., XLVIl. XXXIV., XL. XXVII. ( XX, xxxvii, xxxix, 1 1, li, 63, 96, 198, ( 243, 252, 261 211, 214 ^xxvii, 156, 245, ^ 255 180, 243, 255 ^xl, Ixiii, 137, 243, . 249, 254 47 XXVII. xl, 138, 249, 254 II., III. XLV. XI., XXIII. 11, 249, 251 XXX, xlv, 230, 245, 256 /53, 55, 56, 245, t 249, 250, 252, 259 182 164, 242, 255 164 82, 247, 253 2 2 monticularis (Axinella), moruliformis (Clador- 1 hiza), . . .J rauggiana (Keniera), murrayi (Esperella), niuta (Bursalina), V 23, 245, 251 nobilis (Myxilla), var • i ■ XX, xxvii, 97, 102, .243, 253, 261 patagonica, . X-K., XXI. nodosa (Echinoclathria , 21 Non-calcarea, IV 20, 247, 251 norniani (Isodictya), XX Duda (Esperella), "liv, no, 139, 145, .156 obtusa (Acanthella), Oceauapia, . 156 oculata (Spongia), XLI 199, 244, 256 officinalis (Euspongia), XX 'xxi, 87, 93, 97, ' 248, 253, 261 oleracea (Rhizochaliua) Oligosilicina, 97 omnivorus (loplion ?), . XXVII 133, 134, 249, 254 XLI., XLV. 204, 244, 256 OphistospongiadiE, vi, ix, 1 Oplilita-spongia, . xi Oscarella, . xi OscillatorianAIga(para- • 1, vi, xi, Iviii, 1 sitic). 1 Pachastrellida, . Myxilla, Myxospongia;, navioelligera (C'rella), . navicelligerum (Sidero- I derma), . . . / nervosum (Echinodic- 1 tyum), . . . / nigricans (Esperia), nigricans (Halichondria), nigricans (lophon), nivea (Gellius glacialis, ) var.), . . . I nobilis (My.xilla), nobilis (Myxilla), var. ) bacillifera, . . J Plate Page xxxviii, 185, 242, 255 xx.,xxl|87>90'2«, 253, I 261 51 xxviii, xxix, xxx, xxxi, xxxiv, xxxviii, xxxix, 'XIII., XIV., J xliii, xliv, xlv, xlvi, xlvii, xlviii, 1, 56, 63, 65, 67, 119, 245, 263, L259 226, 227 r x.xvi, xxvii, xxxvii, xl, xli, xlii, xliii, xliv, xlv, li, Iv, lix, Ixiii, Ixv, XXVI., xxvn., . XXX., XLVII. -J Ixvii, Ixviii, 84, ' 128, 147, 242, 243, 244, 245, 246, 247, 248, 249, 250, 254 vi, xi 115 VIII., IX. xix, 115, 246, 254 165 117 117 117 VIII. XIII. 42, 243, 252 xl, xliii, xlv, li, xxvn., XXX. ^ 137, 140, 143, 244, 250, 254 142, 248 XXVII. XXVII. 142, 249 160 X 109 XV., XVI. 65,70, 242, 253 177 1^ / Iviii, Ixiv, 32, 36, ' 1 242, 252 26 lii Ixiv, 36 .Kvn.-f'-^"'' 1^^' 254, ■ 1 254 vu, vm vii 1 3 ix 272 THE VOYAGE OF H.M.S. CHALLENGER. Plate Page (XXX, XXXVlll, Pachychalina, , . . ■ Pachymatisiiia, . Pacliytragida, palmata (Clialina), palmata (Desmacidon 1 (Homceodictya)), . / palmata (Halichondria), palmata (Homceodictya 1 (Clialina)), . . / palmata (Isodictya), . palmata (Spongia), Pandaros, panicea (Amorpliina), . panicea (Halichondria), papillata (Hamacantha), Papillina, . papillosa (Spirastrella), papyracea (Phakellia), . paradoxa (Axinella ?), . parasitica (Esperella), . parishi (Esperia), pai'ishii (Ampliilectus), parisliii (Esperella), parisliii (Rapliiodesma), patagouica (Myxilla ) nobilis, van), . . ) patagonica (Trachyte dania), patelliformis (Xeno- \ spongia), . . . ) patera (Halicnemia), pattersoni (Esperia), pattersoni (Halichon- i dria), . . . ( 7., v., vi.jjxlviii, Iviii, 19, XLVI. j 153, 242, 244, 245, 1 247, 251 Hv ix V. 26, 246, 251 109, no. 111 lOS, 109 108 108, 109 26, 27, 108 176 2 ' xliii, xlv, 2, 5, 15, 244, 247, 251, 260 61 58 jxlviii, 232, 245, XLi., xi,v.|257_ 259 XXXVI. 172, 243, 244, 255 f xxii, xxiii, xxx, XXV., XLix. -! xxxiv, xlv, xlvi, 1 187, 242, 256 Ixi 65 65 65, 247, 253 65 .xl uA 15 I Of: XXVII. XXIII. 142, 254 57, 249, 252 218 218 117 117 pattersoni (lophon), Patuloscula, I 116, 117, 242, 249, 254, 260 29 paucispinata (Myxilla), xxvii., xxx. | ^^' ^^^' ^^^' -^^• 78 129, 131, 143 129, 131, 137, 143 27, 66, 161 paupera (Isodictya), paupertas (Hjnnenia- cidon), paupertas (Mjrxilla), . Pecten (encrusted by sponges), . pedunculata (Pachy- ) chalina ?), . . I pedunculata (Rhizo ehalina), pelliculata(Halicliondria). Pellina, V. 25, 244, 251 VIII, IX. " 34, 35, 245, 252 I., n. 5, 246, 251 15 peltata (Foliolina), penicillus (Ciocalypta), pennatula (Cladorhiza), perfectus (Suberites), . pergamentaoea (Chalina), perganientacea (Clado- 1 chalina), . . .J Petrosia, Plate XLI., XLV., L. II., III. Phakellia, Phakelliadre, Phelloderma, 'XXXIV., XXXV., XXXVI., XL., XLIX. XXIII., XLVII, Phloeodictyina, . Phlffiodictyinae, . Phlojodictyon, Phoriospongia, Phoriospongiffi, . . Phyllospongia, piceum(Iophon(Alebion)), pilosus (Amphilectus), xix., xxv. PlacospongiadEe, . Plethospongife, . Plocamia, . Page 82 173, 174, 175, 176 76, 86, 89 f xxii, xxiii, xxiv, J. xxxi, 200, 202, 1 245, 256 /26, 27, 244, 251, 1260 27 / xxxeii, xlv, xlviii, \ liii, Iviii, 4, 6, 9, j 20, 243, 244, 247, ' 249, 251 xxx, xxxiii, xl,xli, xlii, xliii, xlv, xlvi, xlviii, liv, Ivi, lix, 54, 169, 242, 243, 244, 245, 249, 255, U59, 261 vii, viii /•xxxiii, Iv, lix, Ix, J 85, 113, 250, 254, 1 259 Ixiv, 31 Ixiv, {Iviii, 31 Ixv, 32, 34 Ixi, 73, 123 xxviii xlviii 117, 118, 120, 121 1 xviii, 126, 243, I 244, 254 Plumohalichondria, Plumohalichondrina, . plumosa (Desmacidon), plumosa (Halichondria), plumosa (Halichondria 1), plumosa (Hymenia- 1 cidon), . . . ) plumosa (Microciona), . plumosa (Myxilla), plumosa (MyxUla ?), \ var. fusifera, . . J plumosa (Plumoliali- ) chondria), . . i plumosa (Pronax), plumosa (Spongia), poculum (Gelliodes), . Podospongia, Polychaete worms l (living in sponge), . I XXIX., XXXI. XXX., XLVII XI 241, flix, 157, 1 255 /■xxvii, lix, 129, J. 145,146,150,155, 1 242, 245, 255 155 145 156 145 145, 156 145 145, 242, 254 145 145, 242 145 145 47, 48, 245, 252 234 132 REPORT ON THE MONAXONIDA. 273 Polymastia, Polyiuastiadse, Polymastidffi, Polyzoa, Porifera, Poripliora, porosa (Esperella), Potamospongia, . Potamospongida, profunda (Axinella), profunda (Esperiopsis), Pronax, Protcleia, proteus (Leucophlceus), proximum (Alebion), Psammoneraata, . Pseudoceratida, . Pscudochalinida, . Pseudoliirciuida, . Pseudotetraxonia, PteronemadEe, pulchella (Esperiopsis), pulcherrima (Acanthella), pumilio (Desmaoella), . punctata (Pachychalina?), purpurea (Latrunculia), putridosa (Rhizochalina), putridosum (Alcyonium), QuasiUina, . radiatum (Phelloderma), Eadiella, Plate Page 'xxriii, xlv, xlvi, xlvii, liv, lix, Ixi, I "XT, IV XXXVIII., XL. 222, 223, 225, 226, L2il, 242, 256, 261 vii, viii X, Ixviii 88 vi, vii, X, xi vii 69, 71, 245, 253 viii ix /181, 247, 248,255, t 260, 261, 263 ^. /78, 83, 243, 253, 1261 145 XLii.,XL,v.|l-'^-'214, 243, y 256, 259 169 117 vi, ix ix ix ix X viii XIX., XXX. 84, 133, 246, 253 , xx: 255 xlviii, 24, 244 xxxii.|^^"' '^^i"' 177, I 246, 58, 59 251 234, 240 vii..,ix.j32. 33, 242,244, I 245, 252 33, 34 flix, 225, 241, 1256 xxni.,XLVi..|85."3.250,254, I 259 216, 217, 218 ramosa (Desmacidon?), j^xiir., xxiv., \ 107,243,253 I XLVII. I ramulosa (Suberites), ramulosus (Suberites), Bapbidonemata, Rapbioderma, Raphiodesma, Rapbiophoridoe, Easpailia, . Kaspelia, rectangularis (Chalina), 207 i-xxxii, 197, 207, XXIX., XLiii. J 222, 225, 246, 247, 1256 .. vi, ix 62 62, 65 vii, viii (x.xii, xxiii, xxxiii, xlv, lix, 80, 107, 164, 178, 188, 242, 243, 256 188 v., XLVL 26, 27, 247, 251 rectangularis (Clador- ) hiza abyssicola, var. ), i regalis (Sceptrella), Reniera, Renierida, . Renierina, . Renierinas, . repeus (Thalysias), leptans (Desmacidon),. reticulata (Axinella), . Rbaphidophlus, . . -j Rhizaxinella, Rhizochalina, rigida (RaspaUia ?), Rinalda, robusta (Euplectella), . robusta (Isodictya), robusta (Oceanapia), . robusta (Phakellia), robusta (Polymastia), . robusta (Toxochalina), . robustus (Gellius), rosacea (Halichoudria), rosacea (Myxilla), rosea (Dendrilla), rotalis (Desmacidon), . rotalis (Esperella), rubens (Spongia), rubispina (Spongia), rugosa (Axinella), rugosa (Hahchondria •» solida, var.), . . j rugosum (Echino- -j dictyum), . . / rustica (Pacbycbalina), SarcobexactinelUda, Sarcovitreohexactinellida, sarsii (Tricliostemma), ■ scandens (Esperia), scandens (Halichondria), scandens (lophou), Sceptrella, . Schmidtia, . Plate Page 86, 88, 248, 253 Ixii, Ixvi, 234 XXV, xxxvii, xliv, xlv,xli.x,l,lii, Iviii, 1, 9, 13, 14, 50, 51, 167, 241, 242, 247,248,249,251, 260 LX 1 ix, liv, Ivi, Iviii. Ix, Ixiii, Ixv, 1, 20, 25, 167 10 I xxviii, 105, 242, xxiii.|253 xxxvii. 184, 242, 255 xxvin., XXIX., /lix, 146, 151, 242, XLvr. I 243, 247, 255 222 I Iviii, Ixiv, 6, 32, VIII., IX. j 241, 242, 244,245, (246, 252 XXXIX. 191, 243, 256 210 210 Ixiv, 36 Ixiv, 32, 36, 242, 1 Ixiv, (252 r XXVI., xxviL, XLVII. 170' fSlO, 214, 226, ^^^- 1 241, 256 50, 242, 252 36 130 XX vi,xxvii, xxxvii, 128, 129, 130, 140, 141, 247, 254 xxiv 74 74 19 47 182 4, 245, 251 165, 246, 255 19 ix ix ' 217, 218, 241, '• I 245, 256, (ZOOL. CHALL. EXP. — PAET LIX. 1887.) 261 117 117 117 Ixii, Lwi, 233, 234 9, 10, 12 Nun 35 274 THE VOYAGE OF H.M.S. CHALLENGER. Plate semisuberites (Tento- j- rium), semisuberites (Theco pliora), semitubulosa (Reniera), senilis (Suberites), Sideroderma, Silicea, similis (Cladorhiia), aimilis (Petrosla), simonis (Esperella), simulans (Halichondria), singaporense (Plilceo- 1 dictyon), . . .J siiigaporcnsis (Rbizo- 1 chalina), . Siphonochalina, . sol (Radiella), soliJa (Halichondria), solida (Spirastrella), sollasi (Proteleia), sparsus (Ectyon), Spbferospongia, . Spiculispongic'e, . spinata (Tracbytedania), Spinosella, . spiralis (Suberites), Spirastrella, Spirastrellidse, SpongiadiB, . Spongia, Spongia;, Spougida, . Spongida;, . Spongilla, . Spongilladse, SpongillidK, spongiosa (Myxilla), Page xxix, xxxi, xxxviii, xl, xliii, xliv, xlv, 198, 199, 221, 241, 242, L256, 260, 261 221, 222 I., IX. \ xlix, 15 XLV. 209, 247, 256, 261 I xix, lix, 114, 246 ' 254 vii, viii, liv r93, 95, 98, 248, XX., XXI. 1 ' ' 1253,261 11 III P'"' ^' 243. 244, ■' '1247, 249, 251 r xxviii, xxxvii, XV. -! Ixv, Ixvi, 73, 243, 1 253 Hi 34 34, 36, 242, 252 /- xlviii, Iviii, 29, XL VI. J 49, 244, 245, 252, 1 259 217 4, 245, 248, I 251 XLI. XLII., XLV. / xlvii, xlviii, Ixi, 1231, 247, 256 XLiv. 214, 243, 256 163 viii X 57, 58 20, 29 XXV. 206, 249, 256 f xxi, xxvii, xxviii, XXX, xxxiii, xlv, xlvii, xlviii, Ivi, lix, Ixi, lx\Tii, 229, 245, 247, >-256, 257, 259 f!ix, Ix, Ixi, Ixii, llxviii, 229 vii rl5, 19, 22, 26,30, i 47, 48, 108, 109, U45, 172, 177,183 vi ix 1 xvii, 192, 193 viii X, Ixiii XXVII. j 134 135, 250, I 254 Plate Sporadopora (encrusted ) by sponge), . . J stipitata ( Acantbella ?), stipitata (Chondi-o- \ cladia), , . .J stipitata (Pol}Tnastia), stipitata(Stylocordyla), stipitata (Stylorhiza), . stipitata (Meliiderma), stolonifera (Esperia), . XLIII., L. XX., XXI. Stylocordyla, Stylorhiza, . subacerata (Hymenia- 1 cidon ?), . . .1 suberca (Papillina), subcreus (Desmacodes), Suberispongia, Suberites, . Suberitida, . Suberitidae, Suberitidinifi, suberitoides (Axtemisina), subglobosa (Reniera), . subbispida (Echino clathria, ) . subtriangularis (Petrosia), subtriangularis (Thaly- "1 sias), . . .J suctoria (Tedania), sycinularia (Auletta), . symmetrica (Esperiopsis), XLIII., L. -i XXXIX., XL. Page 113, 199 178, 246, 255 102 223 fxxvi, XXX, xxxi, xxxii, xxxvii, xliii, xlv, xlvi, xlvii, 206, 223, 241,242,24.3,244, L256, 261 223 XX, 102, 243, 253, 261 64 'xxvi, xxvii, XXX, xxxi, xxxii, xxxiii, xxxvii, xliii, xlv, xlvi, xlvii, xlviii, Ivi, lix, Ixviii, 197, 204, 206, 207, 222, 241, 242, 243, 244, 266, 261, 262 222, 223 168, 247, 255 XXV., XXIX., 58 vii f xxii, xxiii, xxiv, xxvi, xxvii, XXX, xxxi, xxxii, xliii, xlv, lix, Ixii, Ixviii, XLI., XLIII., -! 125, 126, 168, 191, XLV., L. I 197,222,225,241, 242,243,244,245, 246, 247, 249, 250, ,256, 259, 260, 261 ix fvii, viii, x, xxvi, I xxix, xxxiii, xliii, -j xliv, xlv, liv, lix, I Ix, Ixi, Ixii, bivii, Llxviii, 197 ix (■112,124,125,241, 1254 xxxvii, 17, 248, 251 160 11 9, 10 52 187 77, 243, 253 [ XIX., XXVI., L XLVI. REPORT ON THE MONAXONIDA. 275 Plate Page Plate Page 'xvii, xxvii, xli. triloba (Sceptrella), 234 xlviii, Iviii, Ixi, Trochoderma, 96 • I 'ti/1 n m 1 XI., xxni., 50, 57, 133, 139, truncata (Petrosia), n., ni. 4, 12, 247, 251 1 6(mill(i, ■ > • XXIX. 241, 244, 245, 248, Tuba, ... 29, 49 249, 250, 252, 259, tuberculata (Ciocalypta), 173, 175 -260, 261, 262 tubulatum (Hal iphysema), 156 Tedaniina, . 50 tubulosa (Axiuella ?), XXX VIII. 187, 250, 256 Tedaniinse, . Iviii, Ixiv, Ixv, 50 tubulosa (SiphonochaUna), 31 Tenacia, 146, 147 tubulosa (Spongia), 30 tenda (Vomemla), 59 tufa (Reniera), I., II. 18, 242, 251 xxix, xxxi, xxxii, tunicatum (Desmacidon), Ixvi xxxviii, xl, xliii, Turbinaria, 227 Tentorium, . L. J xliv, xlv, lix, 198, 199,221,241,242, ,256, 260, 261, 262 tyleri (Ciocalypta), typicum (Echinonema), vagabunda (Desmacella), ... 173 153 58, 59 tenuicapitata (Tedania), XI. 52, 53, 54, 249, 252 vagabundus (Desma- ' codes), var. annexa, . 59 tenuis (Ecliinoclathria), 160 vagabundus (Gellius), . 43 tenuis (Raspailia), ' XXXIX. I XL.ifxxii, xxiii, xlv, XLix. 1188, 242, 256 varia (Halichoudria), . variabilis (Petrosia), . II. 38,39 13, 247, 251 tematus (Acamus), 148, 159, 248, 255 variabilis (Scbmidtia), . 12, 13 Tetliea, liv varius (Gellius), . VIII. 38, 247, 252 Tethyadje, . viii, xxi, Ix, Ixviii velutata (Espcrella), . 70 Tetractina, . X ventilabrum (Halichon- " dria). 170 Tetractinellida, . 'vi, xi, xxi, xxix, . Ix, Ixviii, 19-3 f XXX, xl, xli, xlv. Tetractinellidce, . vi, be ventilabruni (Pbakellia), XXXV., XLIX. xlviii, 54, 170, Tetraxonia, . xi ( 242, 249, 255, 261 Thalassospongia, . vii vermiculata (Bubaris), . ... 182 Thalysias, . 9, 10 vermiculata (Hyme- \ raphia), . . -J 182 Thecophora, 221, 222 Theneadse, . viii vesiculifera (Cacospongia), xxxiii Tlirinacopliora, . t XXIII., XXXVI. xxiv.,/xvii, lix, Ix, 193, , XL. 1 242, 247, 256 villosa (Aulena), villosa (Esperia), xxiv 77 tibicen (Vomerula), Ixvi viUosa (Esperiopsis), . 77 titubans (Desmacidon), XX Vioa, .... 227 Toxocbalina. 'Iv, Iviii, Ixi, Ixiii, ' . 47, 49, 242, 252 virgata (Cbondrocladia), 86, 99 A- \^.^\^ \^ ^^**^-t ■ *^*J • virgata (Isodictya), 38, 39 Trachytedania, . XXIII. (I"".'--' 50, 57, 1 133, 249, 252 Vitreohexactinellida, . — ix f xxviii, xxxvii. /-xlv, lix, 197, 210, Vomerula, . XII., XVII. 1 Iviii, Ixvi, 58, 59, Trichostemma, . XLiii.J 216, 241, 245, 1 248, 256, 261, 262 Vomerulinse, 1 74, 243, 250, 252 59 tridens (Myxilla), 130 warreui (Cliona), 229 tridentata (Cladorhiza ?) XX., XXI XLIX -f""^' ^'' ^^"' ' "195,243,253,261 Weberella, . Xenospongia, xlvi, xlvii, 210 218 triloba (Latrunculia), . 234 Xeuospongiadse, . — viii PLATE I. (zooi. CHALL. EXP.— PART Lix.— 1 887.)— Knn. PLATE I. PAGE Fig. 1. Halichondria peUicidata, seen horn in iront ; o, oscula ; natural size, . 5 Fig. la. Seen from behind ; natural size. Fig. 2. Reniera aquxductus, Y&x. infundihularis ; natural size, . . .16 Fig. 3. Reniera subglobosa : natural size, . . . • .17 Fig. 3a. View of portion of the inside ; natural size. Fig. 4. Reniera implexa, var.; natural size, . . . . . 15 Fig. 5. Halichondria latrunculioides ; o, oscula; p.a, raised pore-areas; natural size, .....••• 6 Fig. 5a. One of the raised pore-areas ; x 1 2. Fig. 6. Reniera tu/h ; natural size, . . . ■ • .18 Fig. 6a. Vertical section ; natural size. The Voyage of H.M.SrChallengep Monaxonida. PI I . Bcrjeaii k HigUey del el l.lV. HALICHONDRIA, RENIERA "Mirxten-i Bros . imp PLATE II. PLATE II. Fig. 1 Fig. 2 Fig. 3 Fig. 4 Fig. 5 Fig. 6 Fig. 7 Fig. 8 Fig. 9. Fig- 10 Fig. 11 Fig. 12 Fig- 13. Fig. 14 Fig. 15 Fig. 16. Halichondria latruncuUoides, oxea ; x 122, IIcdicho7idria jianicea {Kerg\\e\en) , oxen ; x 122, HaUchondria panicea (Jsi^nn), oxed; x 122, . Halicho7idria (?) s]). (c), oxea ; x 122, HaJichondi'ia solida, strongylsi ; x 122, Fig. 5a. Three ends of spicules, showing irregularity in form ; x 284. Reniera svhglohosa , oxe& ; x 122, Reniera tufa , oxea ; x 122, Reniera aquwductus, vm: infnndibularis, oxea ; x 122, Halichondria peUicuJata, oxea ; x 122, Petrosia mnilis, oxea; x 122, . Petrosia siim,ilis, war. massa, oxea,; x 122, Petrosia variabilis, var., oxea; x 122, . Petrosia similis, wax. compacta, oxea; x 122, . Petrosia truncata, strongyla ; x 122, . Halichondria cadnca, oxea ; x 122, Petrosia hispida, oxea. ; x 122, PAGE 7 • 3 3 8 4 284. 17 18 ) 16 6 10 11 14 12 13 3 14 ,e of H.M.Sr Challenger; Moiiaxomda . PI. II 1 X lEE. Artluir Dendy <3el R Morgan hti\ . 7 X izz 10 X 122. 3 X 12£ S X 122 . 11 X IZ?. 4 X 122 . 9 X 122 . 15 X 122 . 13 X 112. 5 X 122 5a X 284'. Mint-erri Broi; imp HALICHONDRIA, PETROSIA , RENIERA, PLATE III. (ZOOL. OnAfX. RXP. PART ux. — 18^<7.)— Nnii PLATE HI. Fig. 1. Petrosia trimcata ; natural size, . Fig. 2. Petrosia hispida ; natural size, . Fig. 3. Petrosia similis, repent form ; natural size, Fig. 4. Petrosia similis, erect, lobose form ; natural size, Fig. 5. Petrosia similis, var. compacta ; natural size, Fig. 6. Petrosia similis, var. massa ; natural size, PAGE 12 14 9 9 12 11 The Voyage of H.M.S."Challengep'.' Monaxonida. PI. IJI. A . GawQTi del et. IiLK - PETROSIA Mir»ter-n Bros . imp . PLATE IV. PLATE IV. PAGE Fig. 1. Pachychalina melior ; natural size, ..... 20 Fig. 2. Pachychalina /ragilis ; natural size, • .... 20 Fig. 2a. PachychaKna fragilis ; natural size. Fig. 3. Pachychalina fibrosa (Bahia (? or Bermuda)); natural size, . . 21 Fig. 4. Pachychalina ^brosa, var. (Station 208); natural size, . . .22 The Voyage of H.M.SrChaller-.ger'; MorLaxonidsL , PI , IV. Barjeaoj A; Hi^Kley del etlitli PACHYCHALINA Mirttem Bro3 . imp- PLATE V. (ZOOL. CIIALL. KXP. PATIT I.IX. 1 SST.)^?^"""- PLATE V. Fig. 1. Pachychalina lohata ; natural size, . . . . . Fig. 2. Challiia pergmnentacea ; a, Hat side ; l>, edge, sliowing oscula ; natural size, .....■•• Fig. 3. Chalina rectangularis ; natural size, .... Fig. 4. Chalina palmata ; natural size, ..... Fig. 5. Pachychalina (f) pedunculata ; natural size, Fig. 6. Pachychalina megalorrhaphis ; natural size, PAGE 22 27 27 26 25 23 The Voyage of H,M.S"CKanenger" Monaxonida . PI . V. K f .• if '• u-'V^^ A . Ga-wan. del et litK . \ ^::r: ■& .;/, ^ ..rT« \^m"^'^^' Mmtem Bros . imp PACHYCHALINA, CHALINA PLATE VI. PLATE VI. PAGE Fig. 1. Pachychaliva elongata ; natural size, . . . . .23 Fig. 2. Pachychalina (?*) 'punctata, osculum-bearing surface ; natural size, . 24 Fig. '2a. Upper portion of surface without oscula ; natural size. Fig. '2h. Porlion of osculum-bearing surface ; o, oscula ; p, pores (?) ; x 4. Fig. 2c. Portion of surface without oscula ; p, pores {Vj ; x 4. The Voyage of H. M.S." Challenger" Monaxonida.Pl VI, Berjeau &. HigViley del et lith PACHYCHALI NA Mirvtem Bros, imp. PLATE VI] . (ZOOL. CUALT.. ICXl'. PAU'I' LIX. — 1887.) — JS'r PLATE VII. PAGE Fig. 1. Siphonochalina intermedia; natural size, . . . .30 Fig. 2. Sijohonochalina ayinulata ; natural size, . . . . .31 TKe Voyage nf H.M.S/Chd.llengep'r MoiidxonidH. . PI . V[l. R Min-lfei-n. del el IiLL Miiiterrv Bros . imip SIPHONOCHALINA^ PLATE VIII. PLATE VIII. Fig. 1. Rhizochalina pedunculata ; natural size, .... Fig. 2. Rhizochalina fistulosa (Station 188); natural size, Fig. 2a. Section; x 2. Fig. 3. Gelliics carchms (Prince Edward Island); natural size, Fig. 4. Gellius varins ; natural size. Fig. 5. Rhizochalina putridosa{8t-dtioii 162); x |, . Fig. 5a. Fistula of same ; x 2. Fig. 6. Gelliu.s calyx; natural size, Fig. 6a. Osculum of same ; natural size. Fig. 7. (jpJlm.'i glaciali.s ; natural size, .... Fig. 8. Gellim glacialis, vav. nivea ; natural size, Fig. 9. Sideroderma navicelligerum ; a, parasitic Alcyonarians ; natural size, PAGE 35 32 39 38 33 43 41 42 115 The Voyage of H,M.S "Challenger'' Monaxonida.Pl.Vm. BeijeaM & Higbley del et litK. MirLtexTv Bros - in^ RHIZOCHALINA, GE LLI US, SIDE RO DERMA. PLATE IX. (zooL. ciiALL. EXP.— PART Lix. — 1887.) — Nnn. PLATE IX. PAGE Fig. 1. Rhizochalina 2yii'tridosa, oxen fvom the inteviov oi theaijouge ; x 325, . 34 Fig. 2. Rhizochalina pedunculata, strongyla from tlie dermal layer ; a, d, adult ; b, c, probably young forms, showing central inflation ; x 325, . 35 Fig. 3. Oceanapia rohusta ; a, 5, c, uxca ; ri, sigmata ; x 325, . . .37 Fig. 4. Rhizochalina Jistulosa, oxea ; a, h, adult; c, young; x 325, . . 32 Fig. 5. Siderodcnaa navicelligerum ; a, h, tylota ; c, trichodragma ; d, larger sigmata ; e, smaller sigmata ; /j minute isuchela? ; x 175, . 115 Fig. G. Rhizochalina pedunculata, oxea from the interior of the sponge ; x 325, 36 Fig. 7. Rhizochalina piutridosa, portion of the end of a fistula, showing the arrange- ment of the spicules ; a, fibres of the " bast " layer ; h, tufts of spicules echinating them, ....... 34 Fig. 8. Sideroderma navicelligerum, minute isochclse, side and front views ; X 1400, ........ 116 Fig. 9. Sideroderma navicelligerum, portion of vertical section of rind, showing the alternating layers of spicules ; a, cut ends of spicules ; h, longi- tudinally placed spicules ; c, outermost layer without spicules, . 115 The Voyage of H.M.S.'Challenger Monaxonida. PI. IX 4 " 325 5 >« 175 d ^w C .. 6- &?. ; : .,^, "^\ vth.\xv Be-ndy, del. M'FATlaiif ?/.Kr«lvnio. I,iL>i''*' Kdn R|-il70CH ALI NA. OCEANAPIA, SIDERODFRMA. PLATE X. PLATE X. PAGE Gelliodes poculum ; natural size, . . . , . . .48 m o sx) r; cu ( — I Id P 3 o o a. en UJ Q O UJ •00 £ PLATE XL (zooL. cHAi.i,. F,xp. — PART Lix. — 1887.) — N'nn. PLATE XL PAGE Fig. L Tedania infundihuliformis ; natural size, . . . .54 Fig. 2. Tedania actiniiformis ; o, oscula ; p. z, pore-ztme : x 2, . . . 55 Fig. 2a. Portion of pore-zone, showing the pores ; x 1 0. Fig. 3. Tedania digitata, cyliiirli'ical specimen (restored); natural size, . . 51 Fig. 4. Tedania massa (Station 320) ; 0, oscula ; natural size. . . .53 Fig. 4a. Portion of surface, showing groups of pores ; x 5. Fig. 5. Tedania tenuicapitata (Station 314) ; natural size, . . .52 The Voyage of H.M.S ''Challenger" Monaxonida PI . H . '.?^- 9 9~ ~S> 'S © ©©©*£© e c ® *. • ^ ^ © <^«? e®e © ® »» - &- HigKLey del et ttti TEDANIA Mmtern lir^s itrip- PLATE XIL PLATE XII. PAGE Fig. 1. Vomerula esperioides ; natural size, . . . . .GO Fig. 2. Gelliodes fihidata ; natural size, ...... 47 Fig. 3. Gelliodes licheniformis ; natural size, . . . . .48 The Voyage of H.M-S,"C>u3.11enger! MonsLXonida .PI. HI. 3. „#' -^'^ ^* «^ RMordaj^ deletliLh. Minlern. Btos . imp. GELLIODES,V0MERULA. PLATE XIII. (ZOOL. CHALL. EXP. PART LIX. — 1887.) Nlin. PLATE XIII. Fig. 1. Fig. 2. Fig. 3. Fig. 4. Fig. 5. Fio-. 6. Fio-. 7. Fig. 8. Fio. 9. Fig. 10. Fig. 11. Fio-. 12. Gellius glacialis, oxeote spicule from dermal layer ; x 180, Gellius calyx, oxeote spicule from dermal layer ; x 180, 6*e^/ms sp., strongylote spicule ; x 180, Gellius glacialis, YSiX. nivea, oxaotti siAgvXq; x 180 (re^^ws ^agre^Z^/er, oxeote spicule ; x 1 80, Gellius carduus, var. magellanica, oxeote spicule ; x 180, Gellius carduus, strongylote spicule ; x 180. (re^Zms Z«ms, stroiigylote spicule ; x 180, Gellius calyx, sigmata ; x 320, Gellius fiagellifer, sigma ; x 320. Esperella murrayi, sigmata ; x 330, . Gellius glacialis, var. nivea, sigmata ; x 320, Fig. 16. Fig. 13. Esperella murrayi, diagrammatic vertical section across one of the crack- like pore-areas, showing the arrangement of the skeleton ; a, pore- area ; b, dermal skeleton ; c, primary fibres ; d, secondary fibres, Fig. 14. Esjierella murrayi, tnchodrsigmsita ; x 260, . Fig. 15. Gellius glacialis, sigmnta ; x 320, . . . . . EsjiereUa murrayi, portion of a pore-area, seen from above, showing the arrangement of the surrounding skeleton, slightly diagrammatic ; a, brushes of spicules fringing the margins of the crack ; h, transverse bands of fibres ; c, pores ; d, spicules of the dermal skeleton. (The microsclera are not represented.) x 40, . Esperella murrayi, palmate anisochela from a rosette ; x 500, EsjJerella murrayi, stylus, ; x 180, . Gellius glacialis, portion of dermal layer, showing the arrangement of the spicules ; x 62, ....••• Fig. 17. Fig. 18. Fig. 19. PAGE 41 44 46 42 43 40 39 40 44 43 69 42 68 69 41 68 69 69 41 The Voyage of H M S"aiallenger" MonoLXoruda. . PI . XIIT. ,3 X leo 1.x 180 4~^ J 4. X180. IE X 3S0. 15 . X320 =^?**4/ \ ^ 1} i II, IS. A 40 1 7 X 500 6 X180. S X 320 18 X 180 19 X £2 MjnLern Bro» . mvp. Ailhuj- Dendy del R. Morgan litK . GELLIUS, ESPERELLA. PLATE XIV. PLATE XIV. Fig. 6. Esperella mammiformis, a specimen attached to a stone ; o, osculum : 3 2 PAGE Fig. 1. Esperella murrayi ; o, oscula ; />.«, pore-areas ; natural size, . . G7 Fig. la. Summit of out of the lobes, showing the o.scula ; natural size. Fig. 2. Esperella hiserialis, & specimen with most of the soft tissues still attached to the axis ; a, spicular axis, seen where the soft tissues have been removed ; x 2, . . . . . . .75 Fig. 3. Esjjerelia hiserialis, a specimen from whicli the soft tissues have been removed ; e, embryos within the spicular axis ; x 2, . . . 75 Fig. 4. Esperella fusca, ; natural .size, ...... 72 Fig. 5. Esperella mammiformis, a.n unattached specimen, with " pinched-up " base ; o, osculum ; x f , . . . . . 64 X f, . . , . . . . . .63 The Voyage of H.M,S."Cl-ian;er,ger" Monaxonida . PI . XIV- e 4 -v^- K v.- N Mf I \ ^ f\ « ^ I Berjea-u &. HigUey d<=l el lith ESPERELLA.. Mint-em. Bros . iTtip PLATE XV. (zoOL. CHALL. KXP. — PART Lix. — 1887.) — Nnn PLATE XV. PAGE Fig. 1. Esperella mammiformis, suhtjlostylwii : x 180, . . . 63 Fig. 2. Esperella lapidiformis, stylus ; x 75. . . . . .65 Figs. 3, 3a. Esperella fusca, tylostjli ; x 180, . . . . .72 Figs. 4, 4«. Esperella arewcWa, tylostyli : x 175, . . . . 73 Fig. 5. Esperella mtda, tjlostyiua ; x 175, . . . . .71 Fig. 6. Esperella porosa, tylostyluH ; x 180, . . . . .70 Figs. 7,7a. Esperella simonis, suhtylostyli ; x 180, . . . . 74 Fig. 8. Espei'ella biserialis, small tylostylus ; x 180, . . . . 75 Figs. 8a, 8b. Large styli from the stem ; x 78. Fig. 9. Esperella porosa, aigmat-A ; xl80, . . . . . 70 Fig. 10. Esperella lapidiforrnis, Tpalmate eLmsochela, ivont view ; x 360, . 65 Fig. 10a. Palmate anisochela, side view ; x 360. Fig. 11. ^s^^ereZ/a nwrfa, palmate anisochelse, front and side views ; x 860, . 71 Figs. 12, 12a. Esperella simonis, toxn; x 180, . . . . .74 Fig. 12a. Shows the remains of the cell in whieb the spicule developed still adhering to it. Fig. 13. Esperella simonis, sigma. ; x 180, . . . . .74 Fig. 14. Esperella 7mda, nigmat-d : x 180, . . . . .71 Fig. 15. As^ereiZa /i. 2a X 2. — k3T^e J^ W)} \^^y,. *£;■.- k. ■^4? ^_ ^'Af I I -^. ■^^^5^-* -t.'-i^i^^idf ''^^es-f a ,^ M « 8 ,** ^ ^ Bei^enu &. Hi^Kley del e,L litK MiiiLei-n. Bios , itup. ESPERELLA, lOPHON PLATE XVII. (ZOOL. CHALL. EXP. — I'AUT LIX. 18a7.) Nuil. PLATE XVll. I'AGK Fi>--. 1. /opho 1 1, chel if ('!■; a, h, tiynued Htjli] c, tylote ii])imlii; x 180. . . 119 Fig. 2. Vomenda esperioldes ; a, b, two views of the diancistrou, x 180; c, sioma, X 475 ; d, young diancistron, showing the hooks still united to the shaft by a thin web, x 475, . . . • .61 Fiff. 3. /o2:»/icm c/ie/, styli ; c, tylote spicule : x 180, . . 120 Fig. 7. lophon abnonualis, n«ette of anisochelse ; x 540, . . . 122 Fig. 8. iop/io/t cAt;/i/('r, auisochela, side view ; x 1200, . . .119 Fig. 9. /o^j/ion ^amwafc; a, 6, <■, three views of the bipocilli; x 1350, . 121 Fig. 10. lophon ('?*) omnivorus, portion of dermal layer, showing dermal skeleton of foreign l)odies, and pores ; x 175, . . . .122 Fig. 11. lophon laminalis ; a, tylote spicule ; t, stylus ; x 180, . . • 121 Fig. 12. Vomenda espcrioidea, portion of dermal layer, showing pores and dermal skeleton ; x 63, . . • • • • .60 The Voyage oF H. M. S."Challenger' d. ,V5 2. O. ' yc 8 " 1200 d :-/ 11 - .,o Monaxonida, PI. XVII. 4., \\ t 6 - 180 9 « 1350 'n -SISpE •X G^ ^ *3;^"' rO^^,. ~jo o /' I 10 .. 12 - 6i. Atrtliur l! endy, del OPHON, VOMERULA. PLATE XVIII. PLATE XVIII. Fig. 1. Esperiopsis challengeri, half front view of au adult specimen ; natural size. PAGE 80 80 Fig. 2. Esperiopsis challengeri, half back view of an adult .specimen; natural size, Fig. 3. Esperiopsis challengeri, side view of a young specimen with only two lobes ; a, the commencing bnd of a third lobe : natural size, . . 81 Fi . 4. Esperiopsis challengeri, a portion of the concave surface of one of the lobes, showing the pores and dermal skeleton ; x .50, . . 81 The Voyage of H.M S "Challenger"' MoT--a:-orLida , ?1 . XvTU . 1 t .50 J J Wild del RMci-^an l.th Lern oros . imp- ESPERIOPSIS CHALLENGERI PLATE XIX. (ZOOL. UllALL. EXP. — PART LIX — 1887. ) NllU. PLATE XIX. PAGE 80 Fig. 1. Esperiopsis profunda, suhtylotityhxs ; x 190, . . ■ .83 Fig. Ire. Palmate isochela, side view ; x 284. Fig. li. Palmate isochela, front view ; x 284. Fig. 2. Esperiopsis cylindrica, stout stylus projectiug from the surface amongst a number of slender styli ; x 190, Fig. ia. Slender .stylus from one of the surface brushes ; x 190 Fig. '2b. Isochela, side view ; x 500. Fig. 3. Ariip)hilectus a pollinis,\diVg& atyivis; x 284, . . . .124 Fig. 'ia. Small stylus with minutely spined base ; x 284. Fig. 'ib. Toxite ; x 284. Fig. 3c. Isochela; x 284. Fig. 4. Amphilectu.s annectens, stylus of the main skeleton ; x 122, . . 128 Fig. 4a. Tylote spicule from the dermal layer ; x 190. Figs. 5, 5'. Arapkik'ctus pilosus, atyli ', x 66, . . . ■ .127 Figs. 5a, 5a', 5a", 5a'", 5a"", 5a'"". A series of spicules showing the tiaasition from small toxa (5a) to long, slender oxea (5a"'"); x 122. Fig. 5b. Toxite with the remains of the mother-cell still adherent ; x 190. Figs. G, 6'. Esperiopsis symmetrica ; styli; x 122, . . • .78 Fig. 6a. Portion of the membrane from between two radiating spicular processes, loaded with difi'erent kinds of microsclera ; x 512. Fig. 6/a Palmate isochela, front view ; x 512. Fig. 6r. Palmate isochela, side view : x 512. Figs. 7, 7'. Esperiopsis edwardii, var. americana, sty|i ; x 284, . . 79 Fig. 7a. Isochela, side view ; x 284. Fig. 8. Esperiopsis challengeri, styius ', x 284, . • • .82 Fig. 8a. Isochela, side view ; x 284. Fig. 8/'. Isochela, front view ; x 284 Figs. 9, 9'. Esperiopsis pidchella, polytylote megasclera ; x 284, . . 85 Fig. 9a. Palmate isochela, front view ; x 284. . Fig. 9/'. Palmate isochela, side view ; x 284. Fig. 9'-. Young isochela ; x 284. Fig. 9'/. StiU younger isochela ; x 284 Fig. 10. Amphilectus ceratosus, tylote .spicule ; x .500, . . . .125 Fig. 10a. Isochela, side view; x 500. The Voyage of H.M.SXKallenger" Monaxonida PI XIX la X 284- n lb X 2S1-- db X 2S1- 2b :'. bOO l2 a X190- 4a, X 130. 3a. X ZS4. l' b / Z34 9a X 2841 -K. ^r fib X 512 , 1 £c X MZ 7a. X 284- S X 122. S X 122 1 8 x as* t:ii. X Z84. 8b X 28* I ^Sa. X 1?2 5a' X IK:-; 4 ^ a?. 7 X 264 5'xsa 5 X (se 9x zs*. 3' X 284 Arritur Den^ dal , R Morgsiv hth Mitxtem Bros. imp. ESPERIOPSIS. AMPHILECTUS. PLATE XX. PLATE XX. Figs. Fig. Fio. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. PAGE 1. lo. Chondrocladuiclavata, two STpecimena ; x 4. . . .100 2. Cladorhiza longipinna, x 2, . . . . . .92 3. Cladorhiza moruliformis. x 2, . . . . . .90 4. Chondrodadia crinita \ natural size, . . . . .101 5. Axoniderma mirabiJc. x 2, . . . . . .97 (i. Cladorhiza abyssicola, var. linearis, two .specimens united to one another by interlocking of the spicular tufts; the soft parts have Ijeen washed off; x 2, . . . . . . .90 7. Cladorhiza simMis ;x2, . . . . . .93 8. Cladorhiza, inversa; x 2, . . . . . .94 9. Cladorhiza (?) tridentata, seen from above ; x 2, . . .95 Fig. 9a. Side view; x 2. 10. Cladorhiza abyssicola. var. reetangularis ; x 2, . . .88 11. Meliiderma stipitata , natural size, . . . . .102 12. Chondrodadia concrescrrts {?), t'vugmv^^t\ natural size, . . .99 The Voyage of H M STKalle-ngep" Monaxonida, - PL XX. BerjesLO iK hi^l.Iey del et litK Wir\lei-rL Broi-. . -.rrp. CLADORRHIZA, AXONIQERMA, CH ON DROCLADI A , MELIIDERMA. PLATE XXI. (ZOOL. CHALL. EXr.— PART LIX. 1887.) NllU. PLATE XXL PAGE Fig. L Cladoi'hiza moruliformis, .stylus ; x 44, . . . , yi Fig. 2. Cladorldza abyssicola, Y'dvAlnearis, »YiX'A\\ >ity\vi'!i-, x 190, . . 90 Fig. 2a. Large stylus; x 44. Fig. 3. Chondi'ocladia crinita, stylus, from one of the supporting processes : X 44, . . . . . . . .101 Fig. 4. Cladorhiza lonyipinna, small tylostylus : x 190, . . .92 Figs. 5, 5t(, 56. Cladorhiza similis. small tylostyli ; x 180, . . . 93 Figs. 6, Qa, 6b, 6c. 3Ieluderma .stipitata, tylostyli from the stem ; x 44, . 102 Fig. 7. Chondrodadia concrescens (f), atylua ivom iixis ; x 44, . .100 Fig. 7a. Base aud ai)e.\ of auother stylus from axis; x 190. Fig. 8. Axoniderma inirabi/e, anisochelae ; a, front view ; b, side view ; x 430. 98 Fig. 9. Axoniderma mirabile ; a, b, amphiastra, seen from the side ; c, d, end views of same : x 290, ...... 98 Fig. 10. Axoniderma mirabile, sigmat-d ; x 290, . . . .98 Fig. 11. Chond rocladia clavata, isoclielaj ; a, front view; b, side view: X 290, ........ 100 Fig. 12. Chortdrocladia concrc^cens (?), isochelje ; a, siile view; b, c, end views; c?, front view ; x 290, . . . . . .100 Fig. 13. Cladorhiza inversa, anisochelae, showing the nucleus of tlie mother- cell (n), the cell itself has shrunk up on the spicule ; a, front view ; b. side view : x 900, ....... 94 Fig. 14. Meliide'i'ma stipitata, isochelse ; a, side view; b, c, end views; d, back view; x 290, ....... 103 Fig. 15. Cladorhiza mortdiforini'i, aignvd ; x 190. . . . 91 Fig. 16. Cladorhiza (f) tridentata, 6igmata : x 190, . . . .95 Fig. 17. Chondrocladia. crinita, isochelse; a, side view; 6, end view; c, front view; x 290, . . . . . . .101 Fig. 18. Cladorhiza tiimilis, anisochelsid ; a, front view ; 6, side view ; x 900, . 93 Fig. 19. Cladorhiza moruliformis, anisochelae; a, front view; b, side view; X 360, ........ 91 Fig. 20. Cladorhiza (f ) tridentata, ani.sochelse ; a, front view ; b, side view ; c, end view of larger end ; x 360, ...... 95 Fig. 21. Cladorhiza longipinna, smaller anisochela? ; a, front view ; b, side view ; x 675, ....... 92 The Voyage of H . M. S " Challenger" Monaxonida, , Pi XXI. '?.Q.:^ 4-4 'KX 130 5 >; lEO P/ 19 « 360.1 20 X .'iCO. 18 X 900, y ■■^ p fc ArtKitr Deridy del. R Morgan, latih. Ijlmtem-Bros , imp CLADORRHIZA, AXONIDERMA, CHON DKOCLADIA, ME^llDERMA. PLATE XXII. PLATE XXII. PAGE Fig. 1. Desmacidon (Homoeodictya) grandis ; x f, . . . .111 Fig. la. Portion of the surface without oscnla; x 3. Fig. Ih. Portion of the oscuhnn-bearing surface, showing the stellately arranged oscula (o); X 3. r; o X ai P O C4 >< ^i§ ^^R J^5^c|i:^^*f ?-^^rf;-.-'-' ,3, P*',*:S^?,V^^,>' -V •# ■(- (fe ■|^: #^^ f .<- l*i?*t*»^"' t? ^ dW;3&«^ ^ ;'-.# Q z < (J) < Q o UJ o )! ^ o X ■00 c; CO 5^ o OJ ■00 OJ '^, «?; PLATE mil, (ZOflL. CHALL. KXl'. — I'AliT UX.— 1887.)— NllH. PLATE XXIII. PAGE Fig. 1. Thrinacophorafuniforniia, large, unequal ended, oxeote spicule, x 122, 195 Figs, la, lb. StyH; x 122. Figs. Ic, Id. Smaller oxea; x 190. Figs, le, If. Cladostrongyla ; x 190. Fig. Ig. Trichodragmata ; x 190. Fig. 2. Teda7iia 7nassa, stylna ; x 190, . . . . .54 Figs. 2a, lb. Tornota; x 190. Figs. 3, 3c(. Desmacidon (Homoeodictya) kerguelenensis, oxeSi ; x 190. . . 110 Fig. 35. Isochela, side view; x 512. Fig. 3c. Isochela, front view; x 512. Figs. 4, 4a. Desmacidon {f) ramosa, oxea. ; x 190, .... 107 Fig. ib. Isochela, side view; x 512. Fig. 4c. Isochela, front view : x 512 Figs. 5, 5rt. Desmacidon conulosa, oxea; xl90, . . . . 106 Fig. 5&. Isochela, side view; x 512. Fig. 5f. Lsochela, front view ; x 512. Figs. 6, 6a, 66. Trachytedania patagonica, s^inQdiStyli; x 190, . . 57 Fig. 6c. Oxeote spicule ; x 190. Figs. 7, 7a. Desnyxcidoii reptans, oxen; xl90,. . . . . 105 Fig. lb. Isochela, side view; x 512. Figs. 7c, Id. Sigmata; x 512. Figs. 8,8a. Phelloderma radiatuin, aijYi {i)o\yty\oie); x 190, . . .114 Fig. 85. Isochela, side view; x 512. Fig. 8c. Isochela, front view, the two front teeth united ; x 512. Fig. 8(7. Isochela, front view, the two front teeth separate; x 512. Figs. 9, 9a. Tedania commixta, suhtY\os.ty\i ; x 190, . . .53 Figs. 96, 9c. Tylota; x 190. Figs. 10, 10«. Desmacidon Jruticosa, vav., oxen; x 512, . . . 104 Fig. 105. Isochela, side view; x 512. Fig. 10c. Isochela, front view ; x 512. Fig. lOrf. Oxeote spicule ; x 1 90. The Vq>-age of H-M-SrChallenger"^ ltX.122 IcXlSO. m ill Sa.XlSO. Id X 130, lie X190 l,f. X 130 8t X 612 MoTiaxorada. PI, XXIU Z Xiao, 3x130, 3aX190, 4a 120, 4a,X190, 8 a X S12 7 X ISO, 7b X sia 6c. X 130 8 c X 612 7a. X 190, 6 X ISO. S'b X130 5 X130, 2b X ISO, 3t X 51?,, 3c, X 512. 413 X 512 , 4c X 512 7d X az , 10(Ul90 ■> S'b XS12, 5c X 512 \ lO'bX.'jlE, ItV. X512, 3X190, 9a X 130. 3b X 190 3c X lao. ixiso 8a X 190, 10 X 512, 10a X 512 Arthur Derv^ del R. Morgan, ith.. I Mintem- Bros vrnp TEDANIA, TRACHYTEDANIA, DESMACIDON, H OMOEODI CTYA . PHELLODERMA . TH Rl N ACOPHORA . PLATE XXIV. PLATE XXIV. PAGE Figs. 1, la, \h. Thrinaco'phorafuniformis; three sppoimcns. showing the variation in external appearance ; natural size, . • .195 Fie. 2. Desmacidon conulosa ; natural size, . . . • . 1 OG Fig. 3. Desmacidon (HomoBodictya) kerguelenensis ; natural size, . . 110 Fig. 4. Desmacidon (?) ramosa ; natural size, . . . . .107 The Voyage of H.M.Sl'Challenger'': MonoLXonida , PI XXI\" Rerjeaii & Hlghley del et "l^tK THRINACOPHORA, DESMACIDON.HOIVIOEODICTYA TAitTl.an\ Bros . mip . PLATE XXV (ZOOL. CHALL. EXP.— TART UX,— 1887.) — Kllll. PLATE XXV. Fig. 1. Suberites spiralis ; natural size, .... Fig. 2. Amphilectus ceratosus ; natural size, Fig. 3. Amphilectus pilosus {K.evg\xe\Q\i) ; natural size, . Fig. 4. Axiyiella (f) paradoxa ; o, oscula ; natural size, . Fig. 5. Cliona dissvmilis, pore-bearing side ; natural size, Fig. 5a. Osculum-bearing side ; natural size. Fig. 56. Osciila {u) ; x 3. Fig. oc. Poie-areas {p.'i.); x 3. Fig. 5r/. Vertical section of the oorallum, showing the spaces occupied by the sponge ; natural size. PAGE 206 125 126 187 227 The Voyage of H.MS "Challenger:' Monaxo-nida . PI . XXV. /'^ ,p Cl> .X r« .^'^Z.-f^—-.' 5T-3r?...^ ad « \ 3i 3. If' *s xf'^i? '-: ■■ifet' ^ » » "'^ ^'?^, ^u/% e. * <1m.a2 rV^-^ 3. % ., •' ^ s 5a ^4- A-Gnwari ddelSlK, GELLiUS, E5PERI0PS1S, MYXILLA "MinLem Bros . imp- PLATE XXVII. (ZOOL. CHALL. EXP. PART LIX.— 1^87.) — Xllli. 1:^4 PLATE XXVII. PAGE Fig. 1. MyxiUa liastata, aiylus ; x 190, . . . . j3g Figs, la, 16. Tornota; x 190. Fig. Ic. Isoehela, side view ; x SliJ. Fig. 2. MyxUla pauciifpiiiata, styhie, ; x 190, . . . y^.-, Fig. 2a. Tylote spicule ; x 190. Fig. 26. Isoehela, side view; x 512. Fig. ic,. Isoehela, front view: x 512. Fig. 3. Myxilla spongiosa, stylus ; x 1 90, ^ . Figs. 3a, 3b. Tylota; x 190. Fig. 3c. Isoehela, front view ; x 512. Fig. 3d. Isoehela, side view; x 512. Figs. 3c, 3/ Sigmata ; x 512. Fig. 4. Myxilla mollis, stjlua ; x 190, . . . . lo.i Fig. 4a. Tylote spicule ; x 190. Fig. 46. Isoehela, side view ; x 512. Fig. .5. Myxilla digitata, spined stylus ; x 190, ..... j^i Fig. 5«. Tylote spicule ; x 190. Fig. 56. Isoehela, side view ; x 512. Figs. 6, 6a. Myxilla nuiriana, vm: inaf * V , « « / -* ./ %V -(«s^-*S f- '^ ? .■.;H v ^^ '^i v,%A ,»» .i;'-tl '3^^ f^^ 4a X 150 2a z e 4. ■^ -,*^ Rohe-rl. Morgn-.i del et. Iith ESPERIOPSIS, DESMACIDON, MYXILLA, PLUMOHALICHONDRIA. Mintern Bros imp - PLATE XXXI. (SiOOL. CHALL. EXP. — PAKT LIX. — 1887.) — NuU. PLATE XXXI. PAGE Fig. 1. Plocamia coriacea, Mir. elegant; x 2, . . . . 158 Fig. 2. Echinuclathria glabra ; natural size. . . . . 163 Fig. 3. Echinoclathria carteri; natural size. . . . 162 Fig. 3a. End of a Inaiicli, enlarged ; ", o.scula : x lU. Fig. 4. Echinoclathria fttvus, ^roy;in^ uyuii -d I\-cte'ti ; nntural size, . 160 Fig. 5. Echinoclathria favus, another specimen, also growing u])Oii ii Pccfeii ; natural size. . 1 60 Fig. 5(1. Portion of fig. •'>, enlarged, showing the oscula {o), and. at iii, a delicate meiubrani tynipanizing one of tlie larj,'e openings ; x .5. 1h£ Voyage =f H M S " Chailenger" Monaxomda. PI. XXXI. y<)j^ r- ^. ff. BtijM. & H>^^ i^ a },^ 5^ ^9fe* "ife^ ,rg|f ^%m plocami'. '"^OCLATH Mn^l«^T. Broi "f- RlA PLATE LIXII. PLATE XXXn. PAGE F\(f. 1. Echinodictyxtm rngosnm : iintura! size, . . . , . Ifif) Fig. la. P'lifl of a branch : x 3 Fig. 2. Echinodictyum asperum, attached to a stone; natural size. . . Ifi5 Fig. 3. Acanthella pulcherrima ; natural size, . . . . .177 Tlie Voyage of H.MSrOia.llenger-"? Monaxonida. , PI XOdl . ■wji'f #/'li |*-A A Go-wari. del et htk , ECH INODICTYUM, ACANTHELLA. MiTitem Pros . imp . PLATE Xmil. (zooL. CHALL. Rxp. — PART Lix. — 1887.) — Nnn. PLATE XXXII 1. PAGE Fig. 1. AxineUa halfourensis ; natural .size, . . . . .179 Fig. 2. Ciocalypta hyaloderma ; natural size, . . . . .174 Fig. •2a. Portion of surface, with oscular projection ; x 3. Fig. 3. Hi/menicicidon caruncula; natural size, . . . . .167 The Voyage of H.M.SrChaJlenger" Monaxorada, . PI ^ XXXIU. A. Gawan. del et IHK Minterrv Bros . imp . HYMENIACIDON , CIOCALYPTA, AX I N ELLA. PLATE XXXIV. PLATE XXXIV. PAGE Fig. 1. Axinella mariana; natural size, . . . . .180 Fig. 2. P/iftM^m^a&e^^ato, surface without oscula ; natural size, . . 171 Fig. .3. Phakellia fiahellata, another specimen, surface without oscula ; natural size, ••■..... 171 Fig. ?,a. Portion of osoulum-bearing surface, showing the stellate oscula ; x 2. The Voyage of H,M S 'Challenger? Monajconida, PI XXXIV • ;M Hersohell del.etlith. PHAKELLIA, AXINELLA West. Newman i. Co imp PLATE XXXV. (ZOOL. CHALL. EXP. — PART LIX. — 1887.) — Nnu. PLATE XXXV. PAGE Fig. 1. Axinella erecta (off Nightingale Island, Tristan da Cunha) ; natural size, ........ 183 Figs. 2, 2a, 2h, 2c. Axinella erecta, a series of small specimens (Station 148a); natural size, . . . . . . .182 Fig. 3. Phakellia ventilabrum, ya,i: C07inexiva ; x §, . . .170 Fig. 3a. Portion of osculum-bearing surface of same specimen, to show oscuJa ; x 2. Fig. 4. Axinella arborescens; x f, . . . . . .178 Fig. 4a. Portion of a branch ; x 2. The Voyage of KM- S. " Challenger:' Monaxonida, PL XXXV. GM-HerBoheU- deletlitii PHAKELLIA, AXINELLA. West, Newman &■ Co rmp PLATE XXXVI. PLATE XXXVI. Fig. 1. Thrinacophora cervicornis ; natural size, . . . . .194 Fig. 2. Axinella erecta (Inaccessible Island) ; a, parasitic Alcyonarians ; natural size, . . . • • • • .182 Fig. 3. Axinella echidnie.a ; natural size. . . ■ • .183 Fig. 4. Phakellia papyracea, portion only of specimen (Station 145a) ; natural size, . . ■ • • • • .172 .e Voyage ot H.M.SrCinsller.ger. Monaxonida. . PI . XXM. Mintann Eros. imp. A Ge-wan. del ex. bth - AXINELLA, PHAKELLIA, TH Rl NACOPH ORA . PLATE XXXVll (zooL. CHALL. EXP. — PART Lix. — 1887.) — Nnn. PLATE XXXVII. PAGE Fig. 1. Axindla kmaecharta ; natural size, .. . . . .186 Fig. la. One of the crateriform projections, showing a small osculum; x 3. Fio-. 2. Axinella lunaecharta, ?i\\othQr nYiQaimeii; natural size, . . .186 Fig. 3. Axindla fibrosa ; natural size. . . . . . .183 Fig. 4. Axindla reticidata ; natural size, . . . . .184 Fig. 4ffl. Portion of snrface ; x 4. The Voyage of H.M.3 "CKaller.ger" Monaxomda. . PI - XXXVU / -■^ ^I^r^^- -^ r/ la %« W-nff'' 4 a » *. A Gswart del et litK Mmtem Bros . imp. AXINELLA. PLATE XXXVIII. PLATE XXXVIII. Fig. 1. Dendropsis bidentifera ; natural size, . ■ Fig. 2. Axinella ■profunda (Station 241); natural size, Fig. 3. Axinella pro funda (Station 281) ; natural size, Fig. 4. Axinella (f) tubulosa; natural size, Fig. 5. Axinella monticidaris ; natural size, PAGE 192 181 181 187 185 The Voyage of H M S^^>_allenge^ .'' MoTiSLXorada PI XX.WIl] (»w.ar, del it lit-K DEriDROPSlS , AXINELLA Miriiern Bros imp PLATE mix. (ZOOL. CHALL. EXP. PART LIX. — 1887.) Nuil. PLATE XXXIX. '€!'• 1 n •!■ t PAGE rig. 1. HasjMiliaJiageUiformis; natmed size, . . . .190 Fig. 2. Easpailia temds ; natural size, . . . . _ jgg Fig. 2a. Portion of branch, enlarged, showing oscula ; x 3. 'Pig. 3. Maspailia (?) rigida ; natmal size, . . . . .191 Fig. 4. Jl!fmeniacidon{f)subacerata; natural size, . . . .168 The Voyage of H,M S-'OiaJlen^er" M„_, ,,,,,.-]- n, yvvT BASPAILIA "'"'ENIACIDON f.,,. PLATE XL. PLATE XL. PAGE Fig. I. Axinella erecta [Station 14:8a), s.tj\i; x 44, . . . . 182 Fig. la. Crooked strongy la ("vermicular " spicules) ; x 44. Fig. 2. Axinella mariana ; a, a, large subtylostylus and stylus ; b, small styli ; X 44, 180 Fig. 3. Axinella p7vfunda, suhtylofityli ; x 66, . . . . .181 Fig. 3a. Bases of suMylostyli .; x 284. Fig. 4. Thrinacophora cervicornis ; a, stylus, from one of the surface In-ushes ; ?>, oxeote spicule ; xl90,. . . • • .194 Fig. 4a. Trichodragmata ; x 190. Fig. 5. Hymeniacidov (f) sitbacerata, stj]i ; x 66, . . . .169 Fig. ^a. Base and apex of stylu.s ; x 190. Fig. 6. PhaMlia jflahellata, large stjli ; x 190, . . . .171 Fig. 6a, Small styli from surface tuft ; x 190. Fig. 7. Dendropsis Udentifera ; a, styli ; h, strongylote spicule ; c, stylus with spined base ; x 66, . . . • • • 192 Fig. 7a. Two basally spined styli ; x 190. Fig. 7h. Bases of three of the hasaUy spined styli ; x 284. Fig. 7'-. Spined oxea ; x 284. Fig. 8. Raspailia tenuis, two oi the large sty]! ; x 66, . . • ■ 189 Fig. 8a. a, styli; h, oxeote .spicule ; x 190. Fig. 8?/. Strongyla; x 190, Fig. 9. Ciocalypta amorphosa ; a, oxea; ?>, styli ; x 66, . • .176 The Voyage of H MS." Challenger.' /I lax 44. 1 X 4^!-. '^' Ml 7a X 190 7x 66. Arthur Dervdy ad. R Moi-^sn li ih . a. ja- ax**. 7c X 284- s\ 0/ :/a n 3x ss 8x ss Monaxordda. . PI . XL. t3 3a. X 284 h I 1 r^ .5 x6S. 5axl90- \^\ 6a X 190. I 8b X 130 e X 130. 3x 66 8axi30. iJliiiter^r. Bros utip AXINELLIDAE PLATE XLI. (ZOOL. CHALL. EXP.— PAHT LIX. — 1887.)— NllD. PLATE XLI. Fig. 1. Suberites durissimus ; natural size, ..... Fig. 2. Suberites caminatus, group of specimens encrusting a fragment of Sporadopora (Station 320) ; natural size, Fig. 3 Suberites microstomus ; natural size, Fig. 4. Suberites mollis ; x ^, . Fig. 5. Spirastrella pa2nllosa ; x |, Fig. 6. Polymastia agglutinans ; x 2, Fig. 7. Spirastrella solida ; natural size. Fig. 8. Polymastia robusta ; natural size, Fig. 9. Subetites perfectus ; natural size, PAGE 203 199 199 204 232 212 231 210 200 The Voyage of H.M.S:'Cnal]er.ger " Monaxorada . PI XL! 1 SUBERITES, POLYMASTIA, 3 PI RAST?,ELLA lointem Fires imp . PLATE XLII. PLATE XLII. PAGE Fig. 1. Polymastia agglutinans, vertical section, showing the arrangement of the skeleton ; a, cortical layer of spicules ; b, grains of sand ; c, main skeleton fibres ; cZ, small, scattered spicules ; x 45, . . . 213 Fig. 2. Polymastia agglutinans, large stylus from the hody of the sponge; x 120, 213 Figs. •2a, 2b. Small tylostyli, found scattered in the soft tissues ; x 284. Fig. 3. Polymastia agglutinans, transverse section of one of the long, mammi- form projections, showing the arrangement of t]\e skeleton ; a, cortical layer of spicules ; b, subjacent layer of spicules ; c, longitudinal bands of spiculo-fibre, seen in section ; d, zone of small spicules scattered around e, the large central canal ; x 40, . . . . 213 Fig. 4. Polymastia corticata, vertical section, showing the arrangement of the skeleton; a, brushes of spicules in the outermost layer of the cortex ; b, large, longitudinally placed spicules in the lower laver of the cortex; c, choanosome ; x 45, . . . . 212 Fig. 5. Polymastia coj-ticata, large stylus ; x 66, . . . .212 Fig. 5a. Tylostylus from outermost layer of cortex ; x 284. Figs. 5b, 5c. Bases of two spicules from the inner layer of the cortex ; x 284. Fig. 6. Proteleia sollasi, vertical section of body, showing the arrangement of the skeleton ; a, primary fibres ; b, outermost layer of cortical spicules ; c, subjacent layer of cortical spicules ; d, grapnel spicules ; e, accumula- tion of foreign matter on the surface ; /, choanosome, with scattered spicules; x 44, . . . . . . _ 215 Fig. 7. Proteleia sollasi, transverse section of one of the mammiform processes, showing the arrangement of the skeleton; a, longitudinal bands of spiculo-fibre, seen in section ; b, outermost layer of cortical spicules ; c, ■subjacent layer of cortical spicules; x 12, . . . .215 Fig. 8. Proteleia sollasi, large stylus ; x 120, . . . . .216 Fig. 8a. Smaller stylus ; x 190. Fig. 8&. Grapnel .spicule ; x 190. Fig. 8c. Grapnel spicule, with terminal expansion but no teeth ; x 190. Fig. 8d. Free end of grapnel spicule ; x .900. Figs. 8e, 8/. Large tylostyli ; x 190. Figs. Sjr, 87/. Small tylcstyli; x 190. TKe Voyage of H.M.S"Challenger'! LV.Or.aJ-. PI XL 11. ''''l^^*(ifc-«- ';<->^>y, j< 1 ^ 45. V „ljU, l..\ \ l^/j . ^ /If 1 tl tH:< II 'mw^^R !/ i;<' I II l*'M\ y ^ -i- ■■■ 'I'i AriKur DenBy dd . R Morgu.n lith A v.4\ 1 8b r. 130 ^. 2^ X 2P- f?^ Se X130 8 d X E>ou u POLY MAST _TELEIA. Vlintem Broa . imp. PLATE XLIII. (ZOOL. CHALL. EXl'. — PART LIX, 1887.) — NllU. PLATE XLIII. PAQK Fig. 1. Trichostemma sarsii (Station 73), seen IVom nliove; x 2, . . 218 Fig. la. Sfimo spoeiinen, side view : x 2. Fig. 2. Trichostemnm .•^arsri, another specimen (Station 73), seen from above ; X 2, . . . . . . .218 Fig. 2a. Same .specimen, .side view : x 2. Fig. ;'>. Trichostemma A'arsM (Station 184), seen from above ; x 2' . . 218 Fig. 3a. Same specimen, .side view ; x 2. Fig. 4. Trichostemma irregularis, seen from above ; x 2, . . . 220 Fig. \a. Same specimen, .side view ; x 2. Fig. .5. TV'ze/iostemma 7rre9?/iaris, aiiotber specimen, side view ; x 2, . . 220 Fig. fi. Sfylocordyla stipitata (Station 49) : x 2, . . . 223 Figs. 7, 8, 9. Stylocordyla stipitata {Station 147); x 2. . . .223 Fig. 10. Stylocordyla stipitata. vav. g/ohosa (Ker^ne\en) ; natural size, . . 224 Figs. 11, 12. Suberites elongatus ; x 2. . . . . . . 205 Figs. 13, 14. Suberites ramv/osiis, var. eyliiidri/evd : natural size, . . 208 The Voyaj5e of H.M.S "CKallenger" Monaxorada. PI XLIII. Bcrjeau &- HigKley del eL litir.. Mir\tem Bros - imp. SUBERITES, TRICHOSTEMMA, STYLO CORDYLA . PLATE XLIV. PLATE XLIV. PAGE Fig. 1. Latrunculia bocagei ; natural size, . . • • 238 Fig. 2. Proteleia sollasi ; natural size, . . . • .214 Fig. .3. Polymastia corticata ; natural size, . . . • .211 Fig. 4. Latrunculia apicalis (Kergueleu) ; p.a., raised pore-areas; o, oscula ; natural size, ...■•••• 234 Fig. 5. Latrunculia brevis, growing upon a fragment of Sporadopora; natural size, 236 The Voyage of H.M.S'.'Challengep" Motiaxonida , PI. XLT/, 1\ ^^^ r:*' ti»-~- .^1^^ -<^>j ^N,^ ^SdKift-i >*;>;- ,.,^ ^li \^aJ- .-!* I."* '^-<; r* ^-.^ ^^^^ ^W^Si^ftf *■# w A &awan- del etlitK. POLYMASTIA, PROTELEIA, LATRUNCULIA MiTilem Bros rmp PLATE XLY. (ZOOL. CHALL. BXP. PART LIS. 1887.) NnU. PLATE XLV. PAGE Fig. 1. Suberites senilis, \eivge tylostyius ; x 66, . . , . 209 Fig. la. Smaller tylostylus ; x 190. Fig. 1&. Part of large tylostylus ; x 190. Fig. 2. Suberites durissimus, large sub tylostylus ; x 66, . . . 204 Figs. 2a, 26, 2c. Smaller tylostyU; x 190. Figs. 2(^, 2e. Bases of larger spicules ; x 190. Fig. 3. Suberites per fectus, large tY\os.tj\vi&; x 66, . . , .201 Fig. 3a. Smaller tylostylus ; x 190. Fig. 36. Base of larger tylostylus ; x 190. Figs. 4, 4a. Suberites mollis, large tylostyli (fig. 4a, a portion only); x 66, . 205 Fig. 46. Base of large tylostylus ; x 190. Fig. 4f. Smaller tylostylus ; x 190. Fig. 5. Suberites caminatus, Isxge iy\o?,tYhxii ; x 66, . . . . 198 Figs. 5a, 56. Smaller tylostyli ; x 190. Figs. 5c, 5rf. Bases of large tylostyli ; x 190. Fig. 6. iy^HiemaaVZon (?) %a^ma, large tylostylus ; y. QQ, . . .168 Fig. 6a. Base of large tylostylus ; x 190. Fig. 66. Smaller tylostylus ; x 190. Fig. 7. Suberites antarcticus, large tylostylus ; x 66, . . . . 202 Figs. 7a, 76. Bases of large tylostyli; x 190. Figs. 7c, 7(i. Smaller tylostyli ; x 190. Fig. 8. Latrunculia bocagei, stylus ; x 190, . . - . . . 238 Fig. 8a. Discaster; x 280. Fig. 9. Latrunculia apicalis {Station 320), stylns ; x 190, . . . 235 Fig. 9a. Discaster; x 280. Fig. 96. Large whorl of discaster, fuU view; x 280. Fig. 9c. Discaster (Kerguelen) ; x 280. Fig. 10. Latrunculia brevis, stylus ; xl90, . . . . .237 Fig. 10a. Discaster; x 280. Fig. 11. Spirastrella pajnllosa, larger tylostylus ; x 190, . . . 233 Figs. 11a, 116. Smaller tylostylus and subtylostylus, from surface ; x 190. Figs, llc-11^. Spined spirulte ; x 500. Fig. 12. Spirastrella decumbens, tylostylus ; x 190, . . . . 229 Figs. 12a-12r;. Spined spirulse ; x 500. Fig. 13. Sjnrastrella solidu, tylostylus ; x 190,. .... 231 Figs. 13a-13e. Spined spirulse ; x 500. Figs. 14, 14a, 146. Spirastrella massa, subtylostyli and stylus; x 190, . 230 Figs. lie-lAk. Spined spirulse ; x 500. Fig. 15. Suberites axiatus, large tylostylus ; x 66, . . . . 203 Fig. 15a. Smaller tylostylus ; x 190. Figs. 156, 15c. Bases of two of the larger tylostyli; x 190. The Voyage of' J-LMS/'Challenger'- Monaxonida^. PI. XLV. 2 aim la X 130. 2a X ISO 0> &c X 190. 0 "tt X 190. 5a. X ISO 2 d X ISO 2 b xwo n i I 2eX130 u 2c X ISO lb X 130 7a X 130 VW /.'1 7c X ISO 7 x GS 7b X 190 7 d X 130 11a X 130 11 d X 500 ^IiC X 500. ' ^. llg X soos II b X 1»0 11 X 190. i lie X bOO 11 f X 500 ft M e t X 130 bb X 190. G X 66 6a X i.'^o. 10X190 Jil4gx.W0 14 e X SCO 14 K X SCO iXWkxsoo ^ ^-v MinJterti Bros. imp. SUBERITES, SPIRASTRELLA, LATRU NCU LI A, &c. PLATE XLYI. PLATE XLVI. Fig. 1. Pachychalina {?) punctata, dermal skeleton reticulation surrounding jo, one of the pores (?); pr, projection into the pore, . . . .24 Fig. 2. Pac%cAaZma(?')pMnctoto, skeleton arrangement, as seen in vertical section, 24 Fig. 3. Siphonochalina intermedia, skeleton arrangement, as seen in vertical section ; a, primary fibres ; b, secondary fibres, . . .30 Fig. 4. Pachychalina lohata, surface section, showing the dermal skeleton reticu- lation and the pores; d.s., dermal skeleton; d.m., dermal membrane; p, pores, . . . . . . . . 22 Fig. 5. Halichondria latrunculioides, dermal skeleton, . . . . g Fig. 6. Chalina rectangularis, skeleton arrangement, as seen in vertical section, . 28 Fig. 7. Esperiopsis symmetrica, skeleton arrangement, as seen in longitudinal vertical section ; a-a, imaginary long axis of sponge ; b, a band of spiculo-fibre of the inner series ; c, a band of spiculo-fibre of the outer series ; d, radiating fibres (part only shown); e, embryos, . . 77 Fig. 8. Dendropsis bidentifera, skeleton arrangement, as seen in transverse sec- tion ; a, central axis ; b, tufts of small spicules surrounding the laro-e projecting spicules; c, small, spined oxea, . . . .192 Fig. 9. Rhaphidophlus filifer, skeleton arrangement, as seen in vertical section ; (i.e., the characteristic dermal crust of spicules, . . .152 The Voyage of H.M.S'.'Challenger" Monaxonida.PLXLVI. 6 5. TYPES OF SKELETON ARRANGEMENT, Vjnl.eiTi ihos - mtT' PLATE XLVII. (Z0C)[ .. uHALL. isxp.— PART Lix.— 1887.) — Nnn. PLATE XLVII. PAGE Fig. 1. Gelliodes licheniformis, surface section, as seen in alcohol, showing the horny fibre {h.f.), ....... 48 Fig. la. Surface section, as seen in Canada balsam, showing the spicules {s), while the horny fibre Qi.f.) is barely visible. Fig. 2. Amjohilectus ceratosns, vertical section, showing the horny fibre {h.f.), and the arrangement of the spicules, . . . .125 Fig. 3. Myxilla rosacea, var. japonica, vertical section, showing the arrangement of the skeleton ; d.s., dermal brushes of spicules ; m.s., main skeleton, 130 Fig. 4. Plumohalichondria mammillata, surface section, showing the arrangement of the dermal skeleton, . . . . . .156 Fig. 4a. Vertical section, showing the plumose skeleton fibres {p.c). Fig. 5. Clathria lendenfeldi, section parallel with and near the surface, showing the skeleton arrangement ; h.f., the stout horny fibre echinated by spined styU, . . . . . . .148 Fig. 6. Desmacidon (?) ramosa, surface section, showing the radiating tufts of oxea, 107 Fig. 7. Phelloderma radiatum, vertical section, showing the arrangement of the skeleton; c, cortex; s.f., skeleton fibres; f.b., foreign bodies; ^, isochelse, . . . . . . . .114 TKe Voyage of H M,S" Challenger" MoTToUfonida, PhXLVn. H- :^1 V/^^; /../ ;-^.' ^=^L; mi: 2. ^^^;'.'i^I^4'i^>'-'"- R "Mor^ax-^ del et UtK. MinbeTi\ Broa ■ imp TYPES OF SKELETON ARRANGEMENT. PLATE XLYIII. (ZOOI,. OHALL. EXP. — PART LIX. — 1887.) — l^nil. PLATE XLVIII. PAGE Fig. 1. Es^wrella lapidiformis, portion of section of embryo ; a., portion of the main mass; c, cap of large, nucleated cells; n., nuclei; m., investing membrane ; x 190, . . . . . lii Fig. 2. Esperella murrayi, portion of horizontal section, taken at right angles to the surface; ect., ectosome ; ch., choanosome ; p.a., crack -like j^ore- area, seen in transverse section ; s. , elongated subdermal cavity, seen in transverse section ; i.e., inhalent canal, seen in longitudinal section; e.c, exhalent canal, seen in transverse section; f.c, flagellated chambers, lying in the trabeculse of tissue between the inhalent and exhalent lacunae {I.); d.s., dermal skeleton; /., fibres of the main skeleton ; e, embryo, in the act of escaping through an exhalent canal ; x 13, . . . xxx, xxxix, xlvi, &c. Fig. 2a. Portion of the surface of the sponge, showing the crack-like pore-areas, p.a. ; the pore-bearing dermal membrane has been removed at p.a'., showing the openings of the inhalent canals (i.e.) in the floor of the elongated subdermal cavities (s.c.) ; x 2. Fig. 2b. Portion of the dermal membrane from a pore-area ; p., the pores ; /., the bands of fibrous (muscular) tissue; »., the nuclei of the epidermic cells; /*'., the nuclei of the fibrous cells ; x 180. Fig. 2«. Portion of a section like that represented in Fig. 2, more highly magnified ; comprising only the ectosome surrounding an inhalent canal (i.e.); c, darkly staining, amffiboid (1) cells, lying in the gelatinous ground-tissue ; t., tricho- di'agmata; w., walls of the inhalent canal; s., sigmata, projecting into the lumen of the inhalent canal; x 190. Fig. 'M. Portion of section of the choanosome ; f.c, flagellated chambers ; c, mesodermal cells; e., epithelial cells lining the canals ; x 190. The spicules are printed in blue. The Voyaje rf HMS. "Challeiiger " MonajcoTiida.. PL XLVin. ArfhuT Dendy iel. BcrjeauX-fiigUey sc "Mmtem. Bros imp ANATOMY AND HISTOLOGY, ESPtRELUA. PLATE XLIX. PLATE XLIX. PAGE Fig. 1. Baspailia tenuis, transverse section of axis (a) and tissues immediately surrounding it; g., granular cells, seen in transverse section; c, globular cells of unknown function ; x 190, . . . xxii Fig. la. Portion of longitudinal section in the same region ; a., axis ; r/., granular cells seen in longitudinal section ; x 190. Fig. 2. AxineUa (f) paradoxa, portion of section of choanosome ; i.e., inhalent channels ; y^c, flagellated chambers; e.c, exhalent channels ; m.c, mesodermal cells ; x 290, .... xxii, xlvi Fig. 2a. Portion of a band of elongated mesodermal cells (;/.) found accompanying a skeleton fibre ; x 290. Fig. 3. Phakellia ventilahrum, var. connexiva, section through a frond, taken at right angles to the surface; o., oscula; d.m., dermal membrane; p., pores; i.e., large inhalent canals, or subdermal cavities; e.c, exhalent canals ; f.c., flagellated chambers ; x 18, xxx, xli, xliii, xlv Fig. 4. Cladorhiza (f) tridentata, radial, vertical section through one of the marginal, globular bodies ig.h.); gl., glandular organs ; s., jelly-like secretion (?); c/l, choanosome ; x 44, .... xxxiv Fig. 4a. Small portion of section through the interior of a globular body, showing the irregularly shaped cells {c), lying in the jelly-like matrix ; x 284. Fig. 4/). Section through one of the cup-shaped, glandular organs ; g.c, the individual cells (or nuclei of the cells) of which it is composed ; x 284. The spieulea are 'printed in blue. •!Tie Voyage of tl MS. "Challenger" Manajconida PI. XLIX. -dijn. --r^ _. - 1 ■»!'■' c AninirDeTilyli Jjtilr :-.c ¥-.' ANATOMY AND HISTOLOGY. CLADORH IZA(?1,PHAKELUA, AXINELLA '?J, t^ASPAlLIA, PLATE L. (ZOOL. CFIALL. EXP. PART UX. 1887.) Nllll. PLATE L. PAGE Fig. I. Stylocordyla stijntata, var. glohosa, segment of horizontal section; ect., ectosome ; ch., choanosome ; p., pore; i.e., inhalent canal; e.c, small exlialeut canal opening into large exhalent canal ; /., fibrous wall of large exhalent canal ; x 44, . . xxxii, xliii, xlvi, &c. Fig. 1«. Portion of similar section at the junction of ectosome and clioano.some ; y.c, V glandular [1) cells of the ectosome ; f.t., fibrou.s ti.ssue of the ectosome ; i.e., inhalent canals; f.c, flagellated chambers; e.o., opening of a flagellated chamber into an exhalent canaliculus ; x 284. Fig. 16. Small portion of section through choanosome, showing the flagellated chambers. Fig. 2. Suberites perfectus, section at right angles to the surface; ect., ectosome; ch., choanosome; p., pores; s.c, subdermal caYvties; f.t., fibrous tissue ; x 66, ....... xxiii Fig. 2a. Fibrous tissue, from a spot where the granular bodies {g.) are much more developed than usual ; x 284. Fig. 21. Fibrous tissue, from a spot where the granular bodies (g.) are developed to about the usual extent ; x 284. Fig. 3. Tentorium semisuherites, external view of the entire sponge, showing o, the osculum, and p.a., the single, large ^iore-area ; natural size, ...... xxxi, xl, xliii Fig. 3«. Vertical section through the pore-area; ed., ectosome; ch., choanosome; a., outer layer of ectosome ; b., middle layer of ectosome ; r., inner layer of ectosome; ^j., pores; s.c, vertically elongated subdermal cavities ; p'., narrow canaliculus leading from subdermal cavity into i.e., inhalent canal ; x 44. TTie spicules are printed in blue. TheVoyB.geaf HM.S/'aiaiier.gs ".'lonaxoTdda. PI L. ?=- O % ) I ^7 h > f I . li^ W1 ?."¥» 1 '' o%« '-^ /.^ t> P///^f4 iVy^^^Oo' t?j ll' =.>& ^riy-i-^'f" f fc 71 Ck:. \i^.rS'--'< CK- 9 ■ pa >i\^ ^ 1 * 7 ' i|. Arlhur DeEii^Sel EerjeaiS^Bi^ffiff^ I'tTitem Eros iTiro ANATOMY AND HISTOLOGY,' SUBERITES, TENTORIUM, STYLOCORDYLA. PJ.ATE LI. PLATE LI. PAGE Fig. 1. LatnmcuUa apicalis from Kerguelen, section taken at right angles to the surface, through a raised pore-area and an oscular projection; ect., ectosome ; ch., choanosome ; p., pores, on the top of a raised pore- area; i.e., large inhalent canal; f.c, flagellated chambers; e.c, large exhalent canal, or oscular tube; o., osculum ; d., dermal skeleton of discastra ; g.c, groups of glandular (?) cells in the ectosome; e., embryos; x 15, . . xxxii, xxxix, xliv, &c. Fig. la. Small portion of section of choanosome, more highly magnified; i.e., inhalent lacunae; f.c, flagellated chambers; e.c, exhalent lacunaa; e.o., opening.s of flagellated chambers into exhalent lacunaj ; x 284. Fig. Vi. Small portion of section of ectosome, more highly magnified; d., discastra of the dermal skeleton; f.t., fibrous tissue in which the bases of the discastra are imbedded; v.c, vesicular cells; g.c, glandular (?) cells; x 284. The .•spicules are 'printed in blue. TKe Voyage of H.MS Cliallen«,er' Monaxoniia- PI. LI. ch! , r-^^A/^;#^,;i| ^- ^/ • e e /iTti^iUT D'ni'iy dd 5eijeau i-Hi^KIey sc Mintem Bros. imp. ANATOMY AND HISTOLOGY. LATRUNCULIA. THE VOYAGE OF H.M.S. CHALLENOEK. ZOOLOGY, EEPORT on the Myzostomida collected during the Voyage of H.M.S. Challenger during the years 1873-76. By Dr. L. von Graff, Professor of Zoology and Comparative Anatomy in the University of Graz, Austria. SUPPLEMENT. Since the publication of my Report on the Myzostomida collected during the voyage of H.M.S. Challenger, I have received from my friend Dr. P. Herbert Carpenter, F.R.S., a number of specimens, of which the following pages contain a Supplementary Account. The collection includes seven new forms, besides fourteen species previously described. I shall follow the enumeration given in my Report, and append the new species to the previous list ; in the case of each species references are also given to the literature which has appeared since the publication of my Report. I must refrain from entering upon any discussion of the numerous interesting questions which have been raised by recent researches on the Myzostomida,' and this I do the more willingly, since my friend and pupil Dr. von Wagner has undertaken the task of working out, on the basis of thorough anatomical investigation, the generic classification of the species, which, including those aliout to be discussed, now numl^er seventy-eight. Of the forms included in Carpenter's collection, three are in such an imperfect state of preservation that no adequate diagnosis can be given. One of these was dredged by Dr. J. Anderson, F.R.S., at Padau Bay in the Mergui Archipelago, on Antedon milberti, Valenc. sp. The second was found in the same locality, either on 'A complete bibliography will be found in the large memoir by Fridtjof Nansen, Bidrag til MyzostomemeB Anatomi og Histologi, Bergen, 1885, 4to, 80 pp., 9 pis. In addition to this must be mentioned the contemporary memoir by F. voii Wagner, Das Nervensystem von Myzostoma (F. S. Leuckart), Graz, 1886, 8vo, 52 pp., 1 pi. (kOOL. CHALL. EXP. — PART LXI. 1887.) Ppp 1 2 THE VOYAGE OF H.M.S. CHALLENGER. Antedon milberti or Aritedon comata, P. H. C. On neither of these Crinoids have Myzostomida been hitherto described. The third specimen was obtained from Antedon incisa, P. H. C. (Station 170, ChalL Exp.), and presents some resemblance in form, thickness, and translucent marginal zone to Myzostoma hrevipes (Report, No. 16). Its diameter measures 2*2 mm.; the ten marginal cirri are about 0-14 mm. in length ; the colour of the body is dirty yellow. The back bears five pairs of low but sharply angled radial ridges. All the other specimens admitted of identification. 2. Myzostoma longipes, Graff. Myzostoma longipes, GraflF, Bull. Mus. Comp. Zocil. Cambridge, vol. xi. p. 126, 1883. 6. Myzostoma alatum, Graff. Myzostoma alatum, Graff, Proc, Eoy. Soc. Edin., vol. xii. p. 379, 1884. This Myzostoma, found on Antedon phalangium, Miill. sp., off Carthagena in 80 fathoms of water (by H.M.S. "Porcupine," 1870), is a species new to the Mediterranean fiiuna. In size (3-5 mm. in longitudinal diameter), colour, and form, it exactly resembles the specimen figured in the Report (pi. i.), except that on the back the insertions of the parapodia appear as sharply projecting bosses, on the apex of which the base of the hook could be seen as a small tubercle. The thin marginal portion reveals without any treatment, but better still after saturating with glycerine, the very abundant, wide, and thickly crowded intestinal branches, which like the ova extend close to the margin of the disc. The animal was situated between the mouth and anal-tube of its host, in part astride of the anal tube, and with its pharynx turned towards the oral margin of the Antedon. It was very firmly attached.^ 10. Myzostoma glahrum, F. S. Leuckart. Myzostoma glabrum, Beard, Mittheil. Zool. Stat. Neapel, Leipzig, vol. v., 1885. „ „ Nansen, Bidrag til Myzostomernes Anatorai og Histologi, Bergen, 1885. „ „ "Wagner, Das Nervensystem von Myzostoma, Graz, 1886. A new host for this species is Antedon petasus (Diib. and Kor.), from which Carpenter sent me two specimens (0'6 and 1"5 mm. in diameter). 11. Myzostoma gigas, Llitken (PL II. fig. 4). Myzostoma gigas, Graff, Proc. Eoy. Soc. Edin., vol. xii. p. 378, 1884. „ ,, Nansen, Op. cit. Of two specimens from Antedon eschrichti, MiiU. sp., from Station 48 of the Challenger Expedition, the one measured 175 mm., the other 2 mm. in longitudinal diameter, and 1 Station 13 of the "Porcupine" Expedition of 1870, where this species was also obtained, is in the Atlantic, ofi' Mondego on the Portuguese Coast, and not off Duncansby Head as given on p. 32 of the Eeport. REPORT ON THE MYZOSTOMIDA. S" both exactly corresponded in form and colour with figs. 7 and 8 on pi. ii. of the Keport, except that there was here no trace of the dorsal median depression. The larger of the two individuals was doubly deformed ; for in the first place the first left parapodium was reduced to a very small prominence, and secondly it bore, between the seventh and eighth left cirri, a small supernumerary cirrus (fig. 4, c). 12. Myzostoma testudo, Grafi". Myzostoma testudo, GrafE, Bull. Mus. Comp. Zool. Cambridge, vol. xi. p. 128, 1883. 15. Myzostoma marginatum, Graff. Myzostoma mwginatum, Graff, Ihid. 16. Myzostoma brevipes, Grafi". Myzostoma brevipes, Graff, Ihid., p. 127. 17. Myzostoma carpenteri, Graff. Myzostoma carpenteri, Gratf, Proc. Roy. Soc. Edin., vol. xii. p. 380, 1884. „ „ Nansen, Op. cit. Since the Report was published Dr. Carpenter has discovered that the host of this species, Antedon sarsii of the Scandinavian naturalists ( = Alectro dentata, Say), is identical with the Asterias tenella of Retzius ; and it must therefore be known in future as Antedon teyiella, Retzius sp. 18. Myzostoma areolatum, Graff. Myzostoma areolatum, Graff, P)iill. Mus. Comp. Zobl. Cambridge, vol. xi. p. 127, 1883. 20. Myzostoma coriaceum, Graff. In the collection of the K. zool. Genootschap in Amsterdam, I found on a black Actinometra (n. sp.), from the Moluccas, a specimen of the above. It measured 8 mm. in diameter, and agreed in every respect with the specimen described in the Report (pi. xi.). The latter was found, however, on an Antedon^ {Antedon insignis), the 1 The generic difference in the liost of Myzostoma coriaceimi is a fm-ther proof of the correctnes,s of a criticism made by Dr. P. Herbert Carpenter, who write.s to me as follows : — "I notice that you say on page 21 (Report) ' where one species infests more tliat one host, tlie latter are always closely allied.' This does not always hold good, as tlie genus is sometimes different, e.g., Myn. echinus, elecjans, carinatum, inflator, and costatum, while Ant. eschrichtii and Ant. carinata, on which Myz. gigas is found, are very different indeed ; so also Ant. hageni and Ant. spinifera (hosts of Myz. agassizi). I fancy that in many cases, e.g., Myz. tenuispinain, it is rather a question of locality." 4 THE VOYAGE OF H.M.S. CHALLENGED. name of which has been meanwhile altered by Bell in the " Alert" Report' to Antedon loveni. 22. Myzostoma pulvinar, Graff. Myzostoma pulvinar, Graff, Proc. Roy. Soc. Ediu., vol. xii. p. 379, 1884. 26. Myzostoma pictum, Graff. Myzostoma pic f till I , Graff, Bull. Mus. Comp. Zobl. Cambridge, vol. xi. p. 128, 1883. 28. Myzostoma cirriferum, F. S. Leuckart. Myzostoma ciirifenmi, Graff, Proc. Roy. Soc. Edin., vol. xii. pp. 378, 380, 1884. „ „ Beard, Mittheil. Zool. Stat. Keapel, Leipzig, voL v., 1885. „ ,, Nansen, Op. cit. ,, „ Wagner, Op. cit. P. H. Carpenter's collection includes numerous specimens of this species, from 0"2 mm. in length to fully grown forms. They were obtained partly from Antedon petasus, D. and K. sp. (Cattegat), partly from Antedon rosacea (from Torquay, Arran, Oban, Isle of Man, Keumaes Bay, north coast of Anglesea). Those from the last two localities were dredged by the Liverpool Marine Biological Committee, and among them there is one form with a supernumerary cirrus between the seventh and eighth left cirri, and another which is remarkable because of its all but perfect circular contour (longitudinal diameter 1'52 mm., transverse 1'5 mm.). 29. Myzostoma crenatum, Graff, Myzostoma crenatum, Graff, Bull. Mus. Comp. Zool. Cambridge, vol. xi. p. 129, 1883. 31. Myzostoma vastum, Graff. Myzostoma vastum, Graff, Ibid. 32. Myzostoma agassizii, Graff. Myzostoma agassizii, GraS, Ibid. 35. Myzostoma elongatum, Graff. A badly preserved specimen, 2-5 mm. in length, lying loose in the spirit with Antedon occulta, Antedon similis, and Antedon tuberculata, all new species, from Station 174, Challenger Expedition, to the south-west of Fiji. 1 Report on the Zoological Collections made in the Indo-Pacific Ocean during the Voyage of H.M.S. "Alert," 1881-82, p. 158. London, 1884. REPORT ON THE MYZOSTOMIDA. 37. Myzostoma clentatvm, Graff. A specimen measuring 27 mm. in length, lying loose in a vessel along witli Actino- metra Solaris, Actinometra bennetti, and five species of Antedon, from the Molluccas (K. zool. Genootschap, Amsterdam). 38. Myzostoma jimhriatum, Graff. Two specimens found on Antedon quadrata, P. H. C, from Station 48, Challenger Expedition. Both were almost circidar, the one measuring 2 mm., the other 2*3 mm. in diameter. The cirri of the latter measured 0'17 mm. in length, so that the length given in the Report (p. 49) must have referred to much contracted cirri {cf. also the figure). The habitat of the above specimens has removed any doubt as to whether the host was Antedon eschrichtii or Antedon quadrata. 39. Myzostoma excisum, Graff. Myzostoma excisum, Graff, Bull. Mus. Comp. Zool. Cambridge, vol. xi. p. 128, 188.3. 40. Myzostoma irregulars, Graff. Myzostoma irreyulare, Graff, Ihid., p. 130. 41. Myzostoma caribheanum, Graff. Myzostoma carihheanum, Graff, Ihid., p. 130. 42. Myzostoma rotundum, Graff. Myzostoma rotundum, Graff, Ildd., p. 130. 43. Myzostoma oblongum, Graff. Myzostoma ohhmgum, Graff, Ihid., p. 130. 44. Myzostoma abundans, Graff. Myzostoma ahundans, Graff, Ihid., p. 131. 45. Myzostoma elegans, (^raff. Myzostoma eleyans, Graff, Ihid., p. 130. 6 THE VOYAGE OF H.M.S. CHALLENGER. 47. Myzostoma cornutum, Graff. On Antedon alternata, P. H. C, from Station 236, Challenger Expedition, there was found a dirty yellow translucent specimen of a small Myzostoma with damaged posterior extremity, but agreeing in the character of its cirri, suckers, and parapodia with Myzo- stoma cornutum (Das Genus Myzostoma, Taf. x.). It was apj)arently a very young form, only 1 "2 mm. in length, and the difference in age may also account for the almost circular outline of the body. Antedon alternata is a new host for this species. 49. Myzostoma Jissum, Graff (PL I. figs. 5, 6). From two additional specimens of the above species I am now able to amplify my previous description, which was based on a much contracted form. Both these new specimens measured, including the caudal appendages, about 4 '5 mm. in length, and were of a yellowish-brown colour. On the one (figs. 5, 6) the first and second pair of caudal appendages (I, II) were considerably shorter than the third, and the dorsal ridge exhibited a sharp edge, and a less wavy outline than the specimen figured in the Pieport (pi. iv.). The other specimen occupied, as regards its caudal appendages, a median position between the two forms above mentioned, and the dorsal ridge was only slightly defined. In both, however, it was distinctly seen that the figure of the ridges given in the Report was doubly at fault. In the first place the ridges have a radial course on the disc, the anterior running forwards, the posterior backwards to the margin ; and secondly, similar ribs extend also from the body disc to the dorsal surface of the caudal appendages. The figures show the ventral surface and exhibit the position of the mouth (m), the strong parapodia (p), and the very slightly developed suctorial pits {s). These specimens were found lying loose in spirit with Antedon occulta, Antedon similis, and Antedon tuhercidata, all new species from Station 174, Challenger Expedition, to the south-west of Fiji. 51. Myzostoma quadrijilum, Graff. This specimen was found on Antedon microdiscus. Bell, from Station 186, Challenger Expedition, off Cape York. It was dark brown in colour, and measured, without caudal appendages, O'S mm. in length and 0*64 in breadth, like the form represented in the Report on pi. iv. fig. 3. There is, however, this difference between the two, that in this form the external caudal appendages are much more strongly developed than the inner paii'. They are as long as the body ; the thick basal portion measuring O'S mm., and the terminal REPORT ON THE MYZOSTOMIDA. 7 filament 0"3 mm. in length. Here again the cirri of the Jirst pair are the longest (0'2 mm.), the others measuring only about 0'12 mm. Since the species exhibits manifold variations (see Report, pp. 56, 57), I do not regard these differences as suflBcient for the establish- ment of a new species. Antedon microdiscus is a new host for this Myzostoma. Another specimen, from Antedon hidentata, P. H. C, from Station 186 (Cape York), Challenger Expedition, exactly resembled in form that figured in the Eeport, pi. iv. fig. 4. The body was 0"8 mm. in length, and outside the base of the third pair of .feet the slightly raised male genital papilla was very distinctly recognisable. 52. Myzostoma qiiadricaudatum, Graft'. A specimen from Antedon Jluctuans, P. H. C, dredged in the Torres Strait from a depth of 10 fathoms by H.M.S. "Alert." The body measured 0"83 mm. in length up to the bifurcation, the maximum breadth was 078 mm., and the length of the largest caudal appendage was 0*44 mm., the two posterior being slightly unequal. 54. Myzostoma hicaudatum, Graff. Myxostoma hieaudahim, Graff, Bull. JIus. Comp. Zodl. Cambridge, vol xi. p. 1.31, 1883. 55. Myzostoma filicauda, Graff. Myxostoma filiemida, Graff, Ihvl,, p. 132. 57. Myzostoma carinatum, Graff. Myzostoma carinatum, Graff, Ibid., p. 131, In a vessel containiug Actinometra Solaris, Actinometra hennetti and five sjsecies of Antedon from the Moluccas (collection of the K. zool. Genootschap, Amsterdam), there was a Myzostoma, 2 '8 mm. in length, with lateral margins much folded towards the ventral surface, but closely resembling Myzostoma carinatum (Eeport, pi. ii.) both in colour and in the dorsal markings. The cirri, which occur very abundantly on the margin, are relatively shorter and thicker, and have a maximum length of 0'2 mm. The median dorsal ridge is especially prominent, and the paired laterals are less wavy than in the above form, and start with their full breadth from the median pad instead of beginning as fine ridges and becoming gradually broader. These differences are not however sufficient for specific distinction. 8 THE VOYAGE OF H.M.S. CHALLENGEK. 63. Myzostoma cysticolum, Graif. Myzostoma cysticolum, Graff, Bull. Mus. Comp. Zool. Cambridge, vol. xi. p. 132, 1883. 65. Mijzostoma willemoesii, Graff (PI. III. figs. 5, 6). The collection before me includes a deformed fragment of arm, and a deformed pinmde of Antedon Jlexilis, P. H. C, from Station 192, Challenger Expedition, in the Arafura Sea. The former contains a hermaphrodite Myzostoma which I regard as new, and which is described below as Myzostotna heardi (No. 77); while the pinnule deformity arises from the presence of Myzostoma ivillemoesii. The deformity has a close resemblance to those represented on pi. xiv. of the Eeport. The segments are broadened, shaped like a roof, and united into a capsule which encloses the roof-shaped widened ambulacral groove. But while in the former the whole pinnule shares in the deformity, the point of the pinnule (a) remains in this case unchanged and free. And while formerly I always found (Report, p. 71) in each deformity a pair of Myzo- stomata — a larger female and a dwarf malC' — the present deformity contains only a single male. This measures 1'4 mm, in length, and 1 mm. in maximum breadth. It corresponds with that figured in the Report, pi. xiv. fig. 5, except that the suckers are less con- spicuous, being only partially recognisable with the lens, and that the ten cirri are very irregularly developed. In the former the cirri were not indeed of equal length, but in the specimen before me the differences are much more striking, and the length varies from 0'03 to 0'13 mm., as can be distinctly seen by an inspection of fig. 6 where the- cirri (III-VII) of the left margin are represented. After saturation with glycerine it was seen that the restricted stomach branches leave a free marginal zone, O'l mm in breadth. 66. Myzostoma inflato7\ Graff. Myzostoma infiator, GrafF, Bull. Mus. Comp. Zoiil. Cambridge, vol xi. p. 133, 1883. 67. Myzostoma murrayi, Graff. Myzostoma murrayi, Graff, Fhid. 68. Stelechopus hyocrini, Graff. The host of this species is Hyocrinus bethellianus, Wyv. Thompson. REPORT ON THE MYZOSTOMIDA. 9 69. Myzostoma cirripedium, Graff. Myzostamacirripedium, Graflf, Trans. Linn. Soc. Lond. (Zool.), ser. 2, vol. L pp. 444-445, fig. 19, 1885. This species, found on Metacrinus roUmdus, P. H. C, is closely allied to Myzostoma wyville-ihomsoni (No. 30), but differs from all forms hitherto described in the possession of parapodial cirri. 70. Myzostoma giganteum, Nansen. Myzostoma giganieum, Nansen, Bidrag til Myzostomernes Anatomi og Histologi, Bergen, 1885. Very like Myzostoma gigas (No. 11), found on Antedon prolixa, Sladen.^ 71. Myzostoma graffi., Nansen. Myzostoma graffi, Nansen, Op. eif. Resembling Myzostoma marginatum (No. 15), found on Antedon prolixa, Sladen. 72. Myzostoma nanseni, n. sp. (PL II. figs. 2, 3). This beautiful Myzostoma, which I have named in honour of JVIr. Fridtjof Nansen, has a close resemblance to the Myzostoma folium described in the Report. The single specimen before me measures 8 mm. in length, and in its uncontracted state must have been about 3 mm. in breadth, so that it is not so elongated as Myzostoma folium. It differs from the latter, further, at first sight, in the absence of a distinctly defined marginal zone. The body disc, which is about 1 mm. in thickness, seems to thin off imperceptibly into the finely notched margin. The marginal notches are besides more individualised and pointed, not blunt as in the species above referred to. A flat longitudinal groove marks the middle line of the ventral surface, while the median dorsal line projects conspicuously only in the region of the very strongly developed pharynx (p/i). The parapodial insertions are marked on the back by five pairs of lateral elevations. The colour is a uniform dirty brown with a slight reddish tinge. The mouth is situated not far from the anterior end, but the cloacal papilla (cZ) lies at the commencement of the last quarter of the body. The parapodia (p) are very strong, and disposed at equal intervals in two longitudinal rows between the mouth and cloacal papilla. The round suckers (s) are very slightly developed, and hardly recognisable with a lens. Found lying loose in spirit with a black Actinometra (n. sp.) and several other Comatulse from Moluccas (K. zool. Genootschap, Amsterdam). 1 This species was referred to by Nansen as the Antedon celtica of Marenzeller and of Sladen ; but Dr. P. H. Carpenter, who has received specimens of it from Nansen, identifies it with Antedon prolixa, Sladen, (zool. chall. EXP. — PAET LXI.— 1887.) Ppp 2 10 THE VOYAGE OF H.M.S. CHALLENGER. 73. Myzostoma longicirrum, n. sp. (PI. I. fig. 4). In the circular form of the body, in the crenate character of the sharply defined and hyaline marginal fringe, as also in the topographic distribution of the organs, this new species resembles Mysostoma crenatum (Report, pi. vii.). The well-developed parapodia (p) lie halfway between the margin and the middle of the body, the mouth (m) is almost in the same line, and the comparatively small round suckers (s) are situated halfway between the insertions of the parapodia and the margin of the body. The most striking feature in this form, the disc of which has a diameter of about 1 '25 mm., is the development of the cirri, of which the first to third and eighth to tenth pairs appear extraordinarily elongated, but not uniformly, while the shortened fourth to seventh pairs are of equal length, as the following measurements show : — I. pair of cirri measures 1 0-86 mm. IL 0-84 „ III. 0-49 „ IV.- -VII. 0-23 „ VII I. 0-57 „ IX. 0-78 „ X. 0-73 „ There are furthermore slight difierences between the cirri of the same pair, especially in the tenth. The single specimen was much damaged, and the figure given is thus to a large extent a restoration. It was found on Actinometra parvicirra, Mtill. sp., from Samboangan (Challenger Expedition). 74. Myzostoma amhiguum, n. sp. (PL IL fig. 1). This form recalls at first sight Myzostoma antennatum (Report, pi. viii.), but difiers from it not inconspicuously in tliis, that both posteriorly and anteriorly there are two pairs of elongated cirri. The anterior and posterior ends of the body are so like one another, that it is necessary to examine it by transmitted light, in order to distinguish the two extremities. It measures 1-4 mm. in length, and the maximum breadth is 1-1 mm. The diniensions of the cirri can be given only approximately, since many of them are torn away (as indicated in the restoration by the lighter lines), while others are too much twisted to admit of exact measurement. The short cirri of the third to eighth pairs have a maximum length of O'S mm., the first pair 0*54 mm., the second 07 mm., the ninth 0-8 mm., the tenth 1 mm. The colour is a bright yellowish-brown. The stomach branches {i) and ovaries leave only a narrow marginal fringe free. The mouth (m) is situated on the border of the latter, and behind it is seen the muscular bulb of the REPORT ON THE MYZOSTOMIDA. 11 pharynx (p/i) measuring 0'3 mm. in length. The somewhat weakly developed parapodia (p) are rather nearer the mai'gin than the centre of the body; the oval suckers (s) lie halfway between the margin and the parapodia They measure 0"06 mm. in their longer diameter. One specimen found Ijring loose in spirit with Actinometra solans, Actinometra bennetti, and five species of Antedon from the Moluccas (K. zool. Genootschap, Amsterdam). 75. Myzostoma furcatum, n. sj). (PI. II. figs. 5—7). Eecalling Myzostoma lobatum, Graff" (Das Genus Myzostoma, Taf. ii.) in its four caudal appendages, into which the hind end of the body divides, as also in the unequal notching of the margin. The single specimen before me shelves off from the middle line towards the sides, and is bent upwards both anteriorly and posteriorly, so that no exact figure of the contour can be given. The lateral portions of the body are protruded anteriorly, so that in its extended state the middle of the anterior margin must appear to be turned in. The maximum length in the lateral portions, including the caudal appendages is nearly 8 mm. The inner pair of caudal appendages (cfXj) is a little longer than the outer pair (c«i), and measures 1*5 mm. The latter have a conical form and an almost perfectly smooth surface without thinned lateral fringes. Nor are the marginal notches of the disc markedly thinner than the latter, and look more like warts compressed from above downwards. Along the middle line of the back there extends a narrow, but sharply defined ridge (fig. 5, aa), and there are similar radial ridges, which extend on either side from beside the median pad on to the marginal notches. Each of the caudal appendages bears a similar ridge. The colour is a dark warm sepia-brown, both above and below. The very strongly developed parapodia (fig. 6, p) lie in two almost parallel rows nearer the (depressed) middle line of the ventral surface than the margin, and almost halfway between them and the margin are the round and markedly projecting suckers (s). The pharynx (j^jJi) is in this form stretched far forwards. On one side the blunt male genital papilla ($) can be recognised. Found lying loose in spirit with a black Actinometra (n. sp.), and several other Comatulse from the Moluccas (K. zool. Genootschap, Amsterdam). 76. Myzostoma memhranaceum, n. sp. (PL I. figs. 1-3). Of this species I have two specimens before me, of which the larger is figured. It measures 7*5 mm. in length, and when the ventrally incurved sides were unfolded 12 THE VOYAGE OF H.M.S. CHALLENGER. must have had a maximum breadth of about 4 mm. The whole of the body-disc has an almost uniform thickness of about 0'25 to 0"33 mm., and is slightly translucent through- out, with the exception of the parapodial insertions. The margin is very finely notched ; the colour uniformly greyish -yellow. The back bears a gently ridged longitudinal swelling (fig. 3), and at its sides five pairs of oval prominences, which represent the parapodial insertions on the ventral surface. No trace of suckers could be detected. The slightly developed parapodia (fig. 2, p) are disposed in two longitudinal rows, somewhat nearer the median line than the external margin, and the first pair is at an unusually great distance from the anterior end. The mouth (ph.) and cloacal papilla {cl) are also very small, and are situated at the extremities of a slight longitudinal ventral eleva- tion representing the dorsal median ridge, and much nearer the margin of the body than the parapodia. The smaller specimen is about 5 mm. in length, and is so completely bent in ventrally, that it has the form of a boat. In the form of the body and in the absence of suckers this species seems closely related to Myzostoma folium (Report, pi. iii). Found on Antedon marginata, P. H. C, from Station 208 of the Challenger Expedition. 77. Myzostoma heardi, n. sp. (PI. III. figs. 1-4). I have named this species in honour of the successful investigator of the develop- ment of the Myzostomida, Dr. John Beard. In every respect this form very closely resembles Myzostoma j^sfi'tao^ii'd (Report, No. 61), with which I at first regarded it as identical. It causes on Antedon flexilis, P. H. C, the same deformity as the above species on the arms of Pentacrinus alter nicirr us, P. H. C. The single specimen before me was obtained from the same Antedon fiexilis (Statioii 192, Challenger Expedition, Arafura Sea) as the above described deformed pinnule with Myzostoma willemoesii. The arm in question (PI. III. figs. 1, 2) has been modified in the neighbourhood of a syzygy. The enlargement is most marked in the joint just below the syzygial one ; and it extends upwards, not only on to the hypozygal (a), but also on to the epizygal joint, all three being somewhat expanded on the right side. The expanded portions are tubercled, and the most swollen joint, the one before the syzygy, has on the side a smaU funnel-shaped aperture (fig. 2), which leads into the cavity of the cyst. A second much larger opening is found, however, on the ambulacral side (fig. 3). In this inconspicuous cyst, which would quite have escaped the notice of a less accurate observer than my friend Carpenter, lay the brown Myzostoma represented in fig. 4 from the ventral surface. There was certainly no second individual within the cyst. REPORT ON THE MYZOSTOMIDA. 13 The almost circular body has a diameter of 2*2 mm., and bears on its margin twenty cirri up to 0"17 mm. in length. So far there is no difference between this form and Myzostoma pentacrini, unless indeed in the fact that the body is somewhat less thick, and seems to be flattened out, so that only posteriorly and on the left side is there any curvature towards the dorsal surface. The mouth occupies, however, a subterminal position on the ventral surface, from which the short, broad, cloacal tube (fig. 4, cl) also arises. The parapodia (p) are furthermore considerably stronger than in the former, the male genital papilla ( ^ ) is distinctly demonstrable on both sides, and the suckers are represented by shallow grooves with radial folds (s). In addition to these points of difference, the intestine does not exhibit that extra- ordinarily abundant branching which is seen in Myzostoma pentacrini. The branches extend, however, close to the margin of the body, and in consequence of the marked expansion of the csecal terminations appear to be closely packed. 78. Myzostoma platypus, n. sp. (PI. III. figs. 7-12). P. H. Carpenter sent me a Myzostoma cyst with the accompanying note: — " Among six individuals of Actinometra nohilis from Samboangan, I find one with some struc- tures on the disk that I now recognise as Myzostoma cysts, though I did not know them in 1878. They seem to have no plating whatever, which is curious. I send you one with the Myzostoma inside which I cut from a corner where the disk was torn ; and I shall have another part of the disk with a group of cysts drawn by my best artist.^ They are always close to the ambulacral grooves and most frequently open into them, I suppose in order that the Myzostoma may get hold of some of the food particles which are travelling towards the mouth." The cyst is figured in figs. 7 and 8, from the ambulacral surface, and from the side. The wall really exhibits no calcification and consists exclusively of a thin, dehcate, readily phable skin. It is transversely expanded below the ambulacral groove, and exhibits, close beside the latter, a slit-like aperture, through which the Myzostoma can be seen {a). The animal lies within the cyst transversely to the ambulacral groove, and the portion (a) which is seen in fig. 7 is the anterior end of the dorsal surface. The body is bent dorsally both in front and behind, so that the ventral surface has a markedly convex curvature. The latter lies close to the external convex wall of the cyst, and is attached to the same by means of the parapodia. The character of the ventral surface in this Myzostoma is very remarkable. In the first place, in the median line there is a conspicuous row of five raised, longitudinal ridges (fig. 9, 1 to 5). Their free surface is somewhat broadened and separated by sharp margins from the side, which slopes down to the surface of the body. These ridges, and also the parapodia I See the forthcoming 'Report on the Crinoidea, Part II., pi. Ixv. (Zool. ChaU, Exp., part Ix.). 14 THE VOYAGE OF H.M.S. CHALLENGER, (j)), and the suckers (s), appear like raised plates fastened to the ventral surface. The contour of the parapodial plates is heart-shaped, and from their externally directed notch springs the cyhndrieal, terminal joint of the parapodium. Its origin is marked by a small swelling in the middle of the plate. The suckers are remarkably large. They also look like round, raised plates between the parapodia ; but instead of the usual hole in the centre, there is only a slight, annular depression surrounding a small, central boss. The size and form of these suckers suggest that a careful investigation would yield some answer to the cpestions raised by Nansen in regard to the function and morphological import of these organs. The mouth is situated below the anterior margin of the first ventral ridge, and in the specimen before me the anterior end of the pharynx (ph) is adorned with a circlet of delicate papillae, 0"15 mm. in length. The cloaca seems to open close beside the posterior end. Outside the third pair of parapodia lie the large male genital papillse ( ? ). Thej' are more strongly developed than in any other species, and exhibit a retort-Like form. Their free tubular termination seems to be movable, since it is bent on either side in different directions, on the right side backwards, on the left forwards. The ventral surface unoccupied by the above organs appears to be finely canaliculated, which is probably an expression of the closely packed, radiating, intestinal branches. The dorsal surface examined under the lens does not exhibit any particular sculpturing. The thickness of the body-disc is very considerable, and in the middle must measure almost 1 mm., gradually decreasing, however, towards the blunt margin, but without exhibiting any trace of a marginal fringe. On the other hand, the margin exhibits ten pairs of fine cirri, of which the first pair, distinctly represented in fig. 1 1 (c), attain a length of 0'58 mm. The total length of the body in its elongated state must have been almost 6 mm. ; the maximum breadth of the single specimen before me is 5 mm. The colour, above and below, is a uniform greyish -yellow. REPORT ON THE MYZOSTOMIDA. 15 THE CYSTS OF ANTEDON ROSACEA (PI. IV.). In 1885 P. H. Carpenter^ directed attention to peculiar sweUiugs which he very frequently obsei-ved on the pinnules, and less frequently on the arms, of the Antedon rosacea of European seas. From the results of the study of deformities produced by Myzostomida on Crinoids, Carpenter could not but regard it as probable that 'the malformation was in this case also referable to the same cause. From the aljundant material entrusted to me, I have in figs. 1 to 3 represented some of these deformities, magnified five times. The pinnule malformations are extra- ordinarily numerous, and occur on Aiitedgn rosacea from the most diverse localities, as Carpenter has noted. Sometimes they are hardly noticeable thickenings of a single joint, or of two, less frequently of three adjacent joints (figs. 3, a and 2, h), sometimes larger spherical dilatations towards the adambulacral side (fig. 1, a). Somewhat rarely dilata- tions occur like that represented in fig 2, a, which is really only a pit surrounded by a thickened wall. On the other hand there are not unfrequently two or three such swellings on one and the same pinnule, and on a single arm of an Antedon rosacea from Milford I counted seven of the malformations, and in all fourteen characteristic swellings on one individual. Less numerous are the arm-swellings, such as that represented in fig 2, c. On that specimen (collected by the " Vettor Pisani " near Gibraltar), three distinct deformities occur in close approximation. The very slight arm-swelling extends over two joints, and is tolerably uniform on either side. I have observed, however, one-sided conical swellings of the arm-joints, sometimes combined with an enlargement of the basal joint of the attached pinnules. I have opened fourteen of these malformations of Antedon rosacea under a lens, and, after decalcifying them, have cut longitudinal sections, but in no case have I found a Myzostoma or any other encysted organism. On the contrary, both in the various pinnule deformities and in the arm-swellings, I found a roundish brown foreign body, which was apparently the cause of the deformity, though it was not possible for me to determine anything definitely as to itg nature or origin. In fig. 4 one of these bodies is represented, which I removed from the pinnule swelling of an Antedon from Cumbrae. The diameter of the sphere measured 0-3 mm., and the substance was distinctly divisible into a strongly refracting cortical layer and an opaque granular internal mass. The former was radially divided into cell -like portions, and had quite the appearance of cylinder epithelium both in optical section (a) and on surface view (c). There was not however any sharp boundary between the outer layer and the central mass. The whole had thus the appearance of a superficially segmented ovum. ' Nature, 1885, vol. xxxii. p. 391 ; voL xxxiii. p. 8. 16 THE VOYAGE OF H.M.S. CHALLENGEE, These bodies sometimes lie in the centre of the swollen joint in the calcareous substance {cf. the longitudinal section, fig. 6), or near the integument (fig. 5), or even embedded in the latter as it was the case in the walled pit of fig. 2, a. In sections, one always finds that this egg-like body is surrounded by a non-calcified tissue (m), uniformly stained by carmine. When the body lies close below the integument (fig. 5) the tissue is connected with the latter. The sections shed no light upon the nature of the contained body. In fig. 5, which represents a body with a diameter of 0*26 mm., the more darkly stained cortical substance appears in part roughly divided, and in the central mass also there are fine Lines which suggest a similar division ; when the contents are compressed they fall into clumps like yolk segments. ■ In fig. 6 the smaller swelling («) contains a single body with a diameter of 0-058 mm., in which no structure could be detected, and the distinction between cortical and central mass was less obvious. On the other hand, in the three bodies (Jj, c d), which are con- tained in the larger sweUiag of the same puinule, and which measure from O'l to 0-12 mm. in diameter, the structure described in fig. 4 is already indicated. That these structures are not ova seems to me to foUow from the fact that none of the stained preparations exhibited any trace of nuclei. Furthur investigation is therefore necessary in order to elucidate the nature of the cyst contents. PLATE I. (ZOOL. CHALL. BXP.^PART LSI. — 1887.)^Ppp. PLATE I. The following letters have the same signification throughout, d, Cloacal papilla or opening. i. Intestine. '/«. Mouth. p. Parapodium. ph. Pharynx. s. Sucker. (5 . Male genital papilla or opening. Figs. 1-3. Myzostoma membranaceum, n. sp. Figs. 1, 2. The animal, from the ventral side ; magnified 5^ diameters. Fig. 3. The same, from the dorsal side. Fig. 4. Myzostoma longicirrum, n. sp. Fig. 4. Seen from the ventral side; magnified 108 diameters; Nos. I-X, the number of the cirri. Figs. 5, 6. Myzostoma Jissum, Grafi". Figs. 5, 6. Seen from the ventral side ; magnified 5^ diameters ; Nos. I-III, the three pairs of caudal appendages. The Voyage of H.MS "Challenger:' Myzastomida.( Supplement). PI I. L.v. Graff fee F Hut)., Lift' Earn' 1-3, MYZOSTOMA IM E M B R AN AC E U M , 4 M . LON G I C 1 R R U M , 5-6. M _ FISS U M . PLATE 11. PLATE II. The following letters have the same signification throughout. cl. Cloacal papilla or opeuing. i. Intestine. m. Mouth. p. Parapodium. ph. Pharynx, s. Sucker. <5 . Male genital papilla or opening. Fig. 1. Myzostoma ambiguum, n. sp. Fig. 1. The animal, seen from the ventral side; magnified 75 diameters; Nos. I-X, the numbers of the cirri. Figs. 2, 3. Myzostoma nanseni, n. sp. Figs. 2, 3. Seen from the ventral side ; magnified 5^ diameters. Fig. 4. Myzostoma gigas, Liitken. Fig. 4. The left margin with the supernumerary cirrus (c) between the seventh and eighth normal cirrus. Figs. 5-7. Myzostoma furcatum, n. sp. Fig. 5. The animal, seen from the right dorsal side ; magnified 5^ diameters ; with the median dorsal elevation (a-a). Figs. 6, 7. The same, from the ventral side. m„ the outer, ca„ the inner caudal appendage. The Voyage of H.M.S'Xhallengef: Fz^.3. HI A IV "«K. Fij.l I m // uc^-^A: Ji \ ,v^ .fc V,. MyzostoTnida.(Supplement].Pl.lI. cl Fiq. k m VI VII /Jy.J. vm // t^ i ^^ ■^ .... ^ r- ./^.^ pK caz Na.7. A Lv.Qrafffec. P Hnft, Htl>' Ea«' l.MYZOSTOMA AMBIGUUM, 2 -3 . M . N A N S E N 1 , 4.M.GIGAS, 5-7. M FURCATUM. PLATE III. (ZOOIi. CHALL. EXr.— PART LXI. -1887.) — Ppp. PLATE III. The following letters have the same signification throughout. cl. Cloacal papilla. p. Parapodium. ph. Pharynx. s. Sucker. (5 . Male genital papilla Figs. 1-4. Myzostoma heardi, n. sp. Figs. 1-3. Three aspects of the arm swelling, a, the hypozygal of the syzygy. Fig. 4 The Myzostoma, seen from the ventral side ; magnified 38 diameters. Figs. 5, 6. Myzostoma willemoesii, Graff. Fig. 5. The deformed pinnule, with the enlarged ambulacral furrow (*) and its free (not deformed) end (a); magnified 5^ diameters. Fig. 6. The left margin of the body with (III-VII) cirri. Figs. 7-12. Myzostoma platypus, n. sp. Figs. 7, 8. Two views of the cyst, a, the Myzostoma visible at the opening of the cyst ; magnified 5^ diameters. Figs. 9, 10. The Myzostoma, seen from the ventral side, with its median elevations (1-5) ; magnified 5-g diameters. Figs. 11, 12. The same, seen from the anterior end, with the first pair of cirri (c) ; magnified 5g diameters. The Voja^e of H.M.S.'Xlialkii^er" Mjzostomida (Supplement) PLllL Fuf. 1. Fi• I ►^f/ N^. ^. pk. Ftg. J. V> /. ,00^ © z-^..?. ^^Qr-::^- ' p ...J ,^--( ^.i \^ cL F'ff-7- <4j, /ZJ7-?. ^ __ ..-^-; ^ 11 i t /><7.J. ^ ' 'J IH ■,\ W Fig. 6. IV /iy./Z ,«^, pfb Fi^. IZ. -v-4 / U.1 1#* ^*^ L V Graff, f ec- F Hulh, Lift' Edin' 1-4 MYZOSTOMA BEARDI. 5-6 M WILLEMOESII, 7-IZ.M PLATYPUS. PLATE TY PLATE IV. All the Figures refer to Antedon rosacea. Figs. 1 and 3. The pinnule cysts (a) from an Antedon, dredged by the Liverpool Marine Biological Committee, oflf Keumaes Bay, north coast of Anglesea. Fig. 2. Three deformities from an Antedon, dredged by the steamer " Vettor Pisani" at Gibraltar, a, a groove-like deformity ; h, a pinnule-cyst ; c, an arm- swelling. Fig. 4. The egg-like content of a pinnule-cyst from an Antedon, from Cumbrae. The °right side (c) is drawn in surface- view ; the left side in optical section, with the cortical layer (a) and the internal mass (b). Fig. 5. Longitudinal section through a pinnule deformity, showing the content (a), enclosed in an uncalcified tissue (m). From the same locality as fig. 1. Fig. 6. Longitudinal section through a doubly deformed pinnule of an Antedon, from Milford Haven, with four contents (or-d). (Figs. 1-3 are enlarged 5 times, figs. 4-6, 75 times.) Tie Voyage of H.M.S"Challeiiger" fujJ. SIS5- _ Mjzostomida. (Supplement), PI IV Fiff. Z. f^' Ficf. ^, Fiff.5. ■ I if^^ ■■ A Fi-q. ff. L V Graff fee F.HuA.Lift'Edm' 1-6. CYSTS FROM ANTEDON ROSACEA. THE VOYAGE OF H.M.S. CHALLENGED. ZOOLOGY. KEPORT on Cephalodiscus dodecalophus, M'Intosh, a new type of the PoLYZoA, procin-ed on the Voyage of H.M.S. Challenger during the Years 1873-76. By William C. M'Intosh, M.D., LL.D., F.R.S., &c., Professor of Natural History in the University of St. Andrews. Class POLYZOA, J. V. Thompson. Section Aspidophoea, G. J. AUmau. Genus Cephalodiscus, M'Intosh. CeioJialodiscus dodecaloj^hus, M'Intosh. Cephalodiscus dodecalophus, M'Intosh, Ann. and Mag. Nat. Hist., ser. 5, vol. x., 1882, p. 337, > et seq. Among the collections made by the Challenger Expedition in the Strait of Magellan, there was a structure which, from its external appearance, was in the first instance placed among the Compound Ascidians. When it was found not to belong to this group Mr. John Murray forwarded the specimen to various authorities for examination, and sub- sequently my attention was drawn to the anomalous organism in the hope that it might be found to have affinities with the Annelida. An examination of this remarkaljle type proved that amongst its other relations it was an ally of Rhabdoiileura, a new type of the Polyzoa which had been ably described and figured by Professor AUman in 18G9' ' Qimrt. Jovrn. Micr. Sci., vol. ix., N.S., 1869, pp. 57-63, pi. viii. (ZOOL. CHALL. EXP. — PART LXII.— 18S7.) Qqq 1 2 THE VOYAGE OF H.M.S. CHALLENGER. from specimens procured at a depth of 90 fathoms in the Zetlandic Seas by Drs. Gwyn Jeflfreys and Merle Norman ; and which had received further elucidation at the skilled hands of Professor G. 0. Sars,^ as an inhabitant of the still waters in the deeps off the Lofoten Islands; Though it thus fell within the department of Professor AUman, or that of the late lamented Professor Busk (each of whom had arrived at a similar conclusion in regard to its systematic position), yet both most disinterestedly desired that its descrip- tion should remain in my hands. A preliminary account accordingly appeared in the Annals and Magazine of Natural History for November 1882,^ having been previously communicated to the Southampton meeting of the British Association. The specimens of this remarkable form were trawled at Station 311 (in the Strait of Magellan), January 11, 1876; lat. 52° 45' 30" S., long. 73° 46' 0" W.; at a depth of 245 fathoms ; bottom, blue mud ; temperature at the bottom 46°-0, surface 50°-0 ; specific gravity at the bottom 1-02454, surface 1-01904. The bag of the trawl in this region was filled, Mr. Murray tells me, with a vast mass of Hemiasters, numerous examples of a Veyius, and multitudes of Compound Ascidians, four species of which have been described by Professor Herdman, who also noticed the distinction between Cephalodiscus and the Ascidians. Further, in connection with the habitat of the new form, it is interesting that several peculiar moUuscoid rarities had previously been found in the Strait of Magellan by Professor E. 0. Cunningham, naturalist on board H.M.S. " Nassau," such as his Goodsiria coccinea, a long, lobed, rooted fibro-gelatinous mass of a vivid scarlet colour, with the minute flask-shaped animals in circumferential cells, and the equally curious Pyura molinse, of Blainville. Thus if the Strait be not the head- quarters of pecuhar MoUuscoida, it is certainly one of the centres round which many are grouped, including the present new type — perhaps the most remarkable of them all. Amongst the branches of the coenoecium of Cephalodiscus were a few minute Arachnida, sessile-eyed Crustaceans, fragments of Annelids and sponges, besides many Foraminifera of the Eotulate type, which were chiefly studded on the spines (or filaments) and other parts. I On some Remarkable Forms of Animal Life, &c., vol. i., 1872, pp. 1-18, Tab. i., ii. , - Ser. 5, vol. x. p. 337. REPOET ON CEPHALODISCUS DODECALOPHUS. DESCRIPTION. The description of this peculiar type may be taken under the following heads : — I. The House or Ccenceeium ; II. The Poljrpides, including — a. Buccal disk. h. Branchial plumes. c. Digestive system. Mouth. Pharynx. GEsophagus. Stomach. Pyloric chamber. Intestine. Food. d. Body-wall and body-cavities. c. Muscular system and pedicle. • f. Nervous system. g. Eeproductive organs. h. Buddins;. III. General Remarks and Homologies. I. CCENCECIUM. At first sight the flexible coenoecium might be mistaken for a sea-weed, since it is composed of a much branched fucoid tissue, tinged of a pale brownish hue, and semi- translucent. Moreover the whole surface of the thickish stems is hispid with long tapering spinous processes or fimbriae of the same tough secretion, and perforated here and there by somewhat large apertures with smoothly rounded edges. A more minute inspection reveals various opaque bodies in groups in the interior of the branches ; and the appearance of these suggested the relationship of the structure to the Polyzoa or Ascidians. When first seen in the Strait of Magellan it was supposed, Mr. Murray observes, to be a Compound Ascidian. After having subjected it to more careful examina- tion on the completion of the voyage, the late lamented Professor Busk, however, pointed out its distinction from the ordinary Polyzoa ; while Professor Herdman, as already mentioned, was satisfied it did not fall under the Ascidians. The house or coenoecium of Cejjhalodiscus dodecalophus^ (PI. I. fig. 1) consists of ' Kt(pa'Kvi, head ; o/axo?, disk ; tuiixa, twelve ; Xoipoj, plume. ^ THE VOYAGE OF H.M.S. CHALLENGER. thickish, irregularly rounded or flattened stems of the consistency of soft sea-weed, and having a slight lustre like the semitransparent tubes of many Annehds. The stems have a diameter varying from four or five millimetres to double or treble that breadth in the flattened expansions ; but the general size of many of the branches is nearly uniform. They cover a considerable area with their network, the extreme length in one example being about 9 inches, and the breadth 5 or 6 inches. The main trunks appear to have sprung from submarine objects, such as stones or sponges, but instead of standing erect as in a soft Gorgonian, to which the inosculations of the branches give it some resemblance, they seem to have been more or less horizontal, since pillars of the Fig. 1. — Fragment of Polyzoarium of CejiJutlodiscus dodecalophiis, M'liitosh. coenoecium occasionally pass, like aerial roots, from the underside to the plane of attach- ment. Various foreign bodies, such as tubes of Serpulse and portions of sponge, are, moreover, occasionally enveloped by the coenoecium, the originally soft secretion having insinuated itself into all the irregularities of their surfaces, and extended around and beyond them. The surface of the branches is everywhere studded with elevations and ridges, which terminate in long spines of the same tissue — simple, bifid, trifid or multifid — and here and there bending downward to join the main stem, so as to form loops and arches, or inosculating with adjacent spines or fimbriae (woodcut, fig. l). Some of the REPORT ON CEPHALODISCUS DODECALOPHUS. 5 spinous processes arc very large, and project far beyond the others, while occasionally they occur in groups. They generally taper a little towards the tip, which is often attenuate, and of a deeper brownish hue than the rest of the coencecium (PI. VII. fig. 1). The free tips of the branches frequently show a somewhat palmate arrangement, with longer spines variously divided. The irregularity in regard to the distril^ution of the spines recalls the processes on the peculiar sponge Chondrocladia, though this feature is much more marked than in the latter. All the spines are hollow, and in connection with the canals and cavities of the coencecium. The surface of the coencecium, moreover, is dotted, especially at the bases of the spines, with large rounded apertures, which lead into the interior of the stem, the latter being honeycombed from end to end by an irregular system of wide canals and somewhat rounded cavities, intersected by bridles and arches, which thus provide for the constant ingress and egress of sea-water throughout the entire system. The inner wall of these canals and chamlDcrs is as smooth and glistening as the outer surface of the coencecium, the secretion being perfectly homogeneous. It cuts with great readiness, and as cleanly as a soft Fucus ; while it is much less tough than the glistening tubes of the Annelids. Microscopically it is composed of numerous layers of a translucent and very fine mem- branous secretion, so that in the preparations there are endless lines and folds, while the sheen or lustre is doubtless due to the same arrangement.^ The whole disposition of the tissue clearly indicates that it is the work of the polypides, just as much as the tube of an Annelid or Phoronis, the more regular and less bulky tube of Rhahdopleura, and in some respects the shell of a MoUusk. Like the Annelidan tubes it most approaches, it is little affected at first either by nitric acid or caustic potash, though the former after a time somewhat softens and bleaches it. This secretion of Cephalodiscus is paralleled by the curious investment or " house " of Appendicularia, which by some has been held to be the homologue of the Ascidian test, and which fills the tow-net with a semi-solid mass when the animals are abundant. It difi"ers considerably from the branched system of annulated tubes formed Ijy Rhah- dopleura, each of these corresponding to a single polypide, while the rings of which it is composed are successively produced at the termination of the tube by the secreting powers of the great buccal shield or prae-oral disk. Professor Lankester, moreover, has very clearly explained^ that the differences of the rings in the attached or recumbent part of the tube and those of the erect portion — dififerences first pointed out and figured by Professor Allman ^ — are due to the changes in the buccal disk which secretes them, this disk being characteristically bifid in the young specimens which form the recumbent portion of the tube. This symmetry and regularity are absent in the house of Ccphcdo- 1 All these features are well seen in a series of sections mounted by the dextrous hands of the late Professor Busk. 2 Quart. Journ. Micr. Sci., vol. xxiv., N. S., p. 625. ^Journ. Roy. Micr. Soc, vol. ix. pp. 61-02, pi. viii. figs. 4, 6, &c., 1869. 6, THE VOYAGE OF H.M.S. CHALLENGER. discus, apparently because the polypides are not restricted and regulated in their labours by the contractile stalk, but are free to wander throughout the coencecium, and to add layer upon layer to strengthen their protective investment. The buccal shield is in all probability the chief secreting organ, the great sheets of membrane, consisting of the secretion hardened in sea- water, being formed by its agency, but the shape of the spinous processes or fimbrise suggest some other assistance, such as might be obtained from the enlarged and. glandular tips of the plumose arms. While there can be little question that the protective house of Cephalodiscus differs materially from the thickened cuticle of the posterior region of the polypides in the ordinary Polyzoa, and which collectively is termed zocecium, yet it seems unnecessary to complicate the subject by the introduction of new nomenclature.^ The term coencecium, instituted by Professor AUman, which was used in the preliminary description, points to an obvious character, and gives that amount of significance which it is always well to preserve if possible in scientific terms. The term tubarium, proposed by my friend Professor Lankester, is very appropriate in the case of Rhahdopleura, but does not apply to the condition in CephalodisciLS. in which the common abode of the polypides is more aptly indicated by the already existent term coencecium. It is not a matter for surprise that creatures so minute should secrete so conspicuous a home for themselves, or that it should assume the algoid or zoophytic outline, especially when the productions of sponges and other forms are remembered, or when we reflect that even a transparent structureless fluid inside a smooth capsule (as in the Nemertean stylet-pouch) can produce, in countless examples of each species, precisely the same form of solid crystalline stylet. The enlistment of numbers in the present ease supplies any deficiency likely to arise from minute size. The secretions, indeed, both of this form and Rhahdopleura, are most interesting, and indicate a degree of skill and persistence of pattern quite as marked as in much more elevated types. The condition in Cephalodiscus is perhaps the more striking of the two, on account of the perfect freedom of the polypides, the spinous processes or fimbriae of the surface, and the numerous anastomoses of the coencecium. The peculiar shape of the latter, moreover, has probably been found to be that best adapted for the preservation of the animals, by its resemblance to seaweeds or allied structures in the neighbourhood, on the one hand, and on the other, by its afi'ording complete aeration, abundant supply of food, and security to the little architects and their delicate plumes. II. Polypides. The rounded cavities and canals of the semitransparent coencecium contain numerous opaque masses (the polypides) and large ova slightly attached by their peduncles. The former often occur in groups, each individual, however, except in the case of buds, ' E... Edicularium. EEPORT ON CEPHALODISCUS DODECALOPHUS. being perfectly free, and at liberty to wander anywhere along the chambers or externally through the apertures. In some cases they are packed closely together in the cavity, probably from external causes acting after immersion in spuit ; for thin partitions, bridles, and pillars of the semitransparent coencecium often separate the individuals. The cavities are generally clean, though occasionally a little mud containing sponge and other spicules, including peculiar reticulated fragments apparently of Eadiolarians, occurs. This would seem to show that currents of sea-water sweep through these chaml^ers very freely, probably assisted by the active movements of the cilia covering the tentacular plumes. Moreover, in dissecting out the latter, an operation performed with ease, owing to the friability of the ccenoecium, at first sight it may almost be supposed than an ovigerous envelope containing embryos is before us, so remarkable is the profusion of eggs and animals, and apparently so active is the reproductive function. The aspect of the adults and their terminal buds, the proportion- ally large size of the ova, and other features, however, negative such a supposition. Each adult polypide (and they are some- what uniform in size) measures, from the extre- mity of the cephalic plumes to the tip of the pedicle, about two millimetres (woodcut, fig. 2); and of this length the body -proper — that is from the buccal disk to the posterior bulbous region above the pedicle, — is rather more than one millimetre. The body in most is bean- or kidney-shaped (PI. II. fig. 1), generally more rounded and bulbous posteriorly, since there is a tendency to a forward curve behind the pedicle. The dorsal surface is smooth and convex, a distinct constriction, however, being usually evident just behind the anterior region bearing the brownish-red pigment-spots. The latter region is generally bulbous and prominent, and in many a slightly elevated median ridge leading to the anus is present. So far as the spirit-preparations go, therefore, the external difi"erentiation of the anterior region, called " thoracic " by Lankester in Ehahdoj^leuiu, is indistinct in Cephalodiscus, but internally the collar body-cavities are diagnostic. As the pedicle is often curved forward or projected outward at a small angle to the body, the ventral surface is thus rendered comparatively short (PL III. fig. 2) ; indeed, in those which are much bent, the base of the pedicle touches the buccal disk. This contour of the body is interesting in relation to the oblique direction of the cup-like body of Loxosoma. When vvT.C.M Fig. 2. — Ventral view of Cephalodiscus dodecalophus, M'lutosh. 8 THE VOYAGE OF H.M.S. CHALLENGER. the pedicle is extended (PI. III. fig. 1) the ventral surface is nearly straight and con- tinuous with the pedicle, which leaves the body posteriorly at the ventral edge, while the kidney-shaped mass of the body projects dorsally. The pedicle in contraction is shorter than the body, is nearly cylindrical, and terminates in either a rounded or somewhat flattened end ; but in extreme elongation (PL III. fig. l) its appearance is much more attenuate. It is marked ventrally by various longitudinal strise from the muscular bands. The anterior region of the body curves somewhat suddenly downwards and back- wards, and forms a flattened surface on which the great buccal disk or prse-oral lobe rests. Buccal Disk. The great buccal disk (PL II.; PL VI., fig. 2, hs) forms a thin plate with two slight and generally bilaterally arranged elevations in the centre anteriorly, and is divided into two regions by a notch at each side, the anterior moiety being the larger and thicker. The surface of the latter is marked by an arch of brownish pigment-grains, which are densest in the centre of the curve, and shade off gradually on each side ; while a very conspicuous and well-defined deep brownish-red band commences in the posterior division at the notch, and runs with a backward curve to the opposite side. Between this and the posterior margin a brownish pigment-belt — less developed than in front — occurs. The two bands just mentioned form, when completed, a somewhat flattened ring. In many specimens, however, the brownish pigment has been entirely removed by the spirit, leaving only the well-defined reddish band posteriorly. In intimate structure (PL VI. figs. 2, 3, bs) ' the disk is found to consist of a ventral plate, the superficial characters of which have been described, and a j)edicle. In section the former presents the features of a great hypodermic shield somewhat similar in structure to the Nemertean or Annelidan skin, and the surface of which in life is probably clothed with cilia, as indeed Professor G. 0. Sars found in Rhabdopleura, though no distinct cuticular layer is visible in the preparations. This hypodermic tissue is marked in section by vertical strise, which in the thick anterior central region assume a somewhat radiate aspect. The free parts of the shield present dorsally a firm basement- tissue under the cuticle, while ventrally, that is, on the secreting surface, the granular glandular tissue terminates in a translucent smooth edge. As the pedicle is approached, a narrow reticulated region appears within the basement-tissue strengthening the dorsal region of the disk, and the inner as well as the outer wall of this region assumes a firm structure — so as to resemble the basemeut-tissue. In the centre of the pedicle this region is traversed by a radiate series of muscular fibres, which spring from the firm tissue, constituting a kind of skeleton strengthening the basal part with which the plumes are 1 I am much indebted to my assistant, Mr. John Wilson, B.Sc, for making a series of finely stained sections of Ceplmloducus. These have enabled me to determine features not fully seen in unstained sections made in cork by the hand. KEPORT ON CEPHALODISCUS DODECALOPHUS. 9 connected. From this insertion they radiate into the thick central mass of the shield, and some appear to reach the pale ventral region of the hypoderm. Somewhat behind the former region the fibres also arise from the basement-tissue lining the h}^odermic investment of the jaedicle. This fan-like arrangement of the muscular fibres must confer great mobility on the disk, so that its broad scale-like surface can be applied either as a sucker or in an undulatory or partial manner, indeed enabling it to act as a useful locomotive organ. In this connection also it is probable that the basement-tissue may be highly elastic, especially in the absence of any signs of horizontal or transverse muscular fibres, and in connection with the entrance or exit of fluid by the proboscis pores. The glandular nature of the disk, again, shows that it is a structure with secerning powers of great activity, and in close relation with the remarkable ccencecium. Superiorly the pedicle of the buccal disk runs into the region at the base of the arms, and in sections a fine layer of longitudinal and oblique muscular fibres occurs on the inner surface of the dorsal wall of the pedicle, though such have not yet been seen any- where on the inner surface of the shield. In the centre of the buccal shield is a large mesoblastic cavity through which the radiate muscular fibres before mentioned pass, and which communicates with the exterior by two well-marked pores situated dorsally on each side of the middle line at the great central nervous system. In oblique sections {e.g. PI. VI. fig. 3, h])) these pores lie close together in their progress inwards. What relation the ciliated " sense " organ of Rhahdo'pleura, as described by Sars, and also figured by Lankester, may have to the proboscis-pores of Cephalodiscus is a feature of moment for futiire consideration. In Cephalodisctis these pores seem to be formed by invaginations of the hypoderm of the region, but their function is as uncertain as the single pore perforating the nervous system in the proboscis of Balanoglossus. The proboscis in the latter is much more muscular and has an evident proboscis-gland. While therefore the buccal disk of Cephalodiscus is in all probability the main organ of locomotion, just as in Rhahdopleura, which was seen by Professor G. 0. Sars drawing itself up to the aperture of its tube, it diSers from the shield of the latter by its much greater size. In the form just mentioned the organ somewhat resembles the truncated and thickened opercular process of certain Annelids, while in Cephalodiscus it overlaps the neighbouring parts to a great extent. The intimate structure of the shield in Rhahdopleura has only been alluded to by Sars, and he does not appear to have clearly made it out. He says — " On examining more closely this buccal shield we observe in the middle of it an opaque part which seems to contain an interior glandular organ. Continuing the investigation and slightly pressing the animal, we notice, however, that this opaque appearance is not produced by any such internal organ, but by a peculiar and seemingly muscular structure of the shield itself It exhibits, seen from below, in (ZOOL. CHALL. EXP. — PART LXII. — 1887.) Qqq 2 10 THE VOYAGE OF H.M.S. CHALLENGER. the middle numerous small bubbles situated rather far from each other, or somewhat irregularly formed small cells, which, however, when more closely examined (and this is particularly evident in those which lie nearer to the periphery of the disc), show themselves to be external rounded extremities of small inwardly prolonged cylinders, which together appear to form a thick fascicle of incompletely differenced muscular fibres penetrating into the stalk of the buccal shield." He thus does not refer to the structure of the hypoderm of the disk, and yet this is one of its most important features, especially in connection with its functions. In all probability, however, it closely agrees with the hypoderm in the disk of Cephalodiscus, and the muscular fibres described by Sars no doubt arise from the basement-tissue connected with the lophophoral arms, and radiate in a fan-like manner into the hypoderm of the prse-oral shield. Further examination of Rhabdopleura would also appear to be necessary in regard to the presence or absence of a pr£e-oral cavity and proboscidian pores as seen in Cephalodiscus and Balanoglossus. The buccal disk is apparently the homologue of the epistome in the ordinary Polyzoa. It is also interesting to find paired spaces in the epistome of Loxosoma} The hypoderm of the buccal disk folds evenly over anteriorly to pass backwards and upwards to the pedicle, and as the latter is connected with the basal framework of the arms, this region forms the common ground for the origin of the twelve plumes.^ On removing the disk, some of the plumes often remain attached to the pedicle, while others in the lateral regions are fixed to the basal tissue in front of the broad apron-like post-oral lamella on each side. In transverse section the centre of the disk, even in early buds, presents a large median chamber, traversed by the radiate fibres of its muscular system, and communi- cating with the exterior through the paii' of pores occurring in the region of the nervous system. It would thus seem that sea-water could be admitted into the interior, though, perhaps, this is by no means indispensable for the performance of its functions. Mr. S. F. Harmer has drawn my attention to the great similarity between the proboscis of Balano- glossus and this organ (buccal shield) in Cephalodiscus, though, it is true, only a single proboscis-pore leading through the nerve-ring of the stalk exists in the former, which lil\;ewise has a proboscis-gland and the so-called " heart." The functions of the organs difi"er very considerably, but there can be little doubt as to the nature of the interesting homologies between the two forms. Branchial Plumes. From the dorsal edge of the basal region just described, twelve plumes arise almost in linear series, six on the one side of the median line and six on the other. A thickened hypodermic area, with a median fissure, is visible in sections, and the pale region underlying ' S. F. Haruier, Quart. Jouvn. Micr. Sci, vol. xxv., N.S., pi. xx. fig. 18. ^ Hence the name dodecalophus. REPORT ON CEPHALODISCUS DODECALOPHUS. 11 this indicates the central nervous system. The hjqDoderm of the basal region contains numerous granular masses (gland-cells and pigment) which have a brownish hue by trans- mitted light. The plumes (PI. II.) are nearly of uniform size, and consist of a thickish central stem, occasionally slightly crenate, and furnished with a series of longitudinal fibres ; while distally each is terminated by a peculiar bulbous enlargement, which at first sight resembles the tip of certain hydroid tentacles {e.g., Coryne or Syncoryne) bristling with dart-ceUs and pigment. The rugose appearance, however, is due to large gland-cells containing granules and globules (PI. V. fig. l), which are arranged in a somewhat regular manner round a central cavity, and which present a deep yellowish tint in the preparations. This structure may perhaps be a further and special development of the somewhat large hypodermic granules of the tips of the pinnae. The appearance of these bulbous enlarge- ments in section is shown in PL IV. fig. 3, part of the upper wall of the stem in this case being formed of the ordinary hypoderm below the tip. When the latter is cut longitudinally, the space in the centre of the bulbous extremity is found to be continuous with a similar space at the end of the arm. Very soon, however, transverse bridles and fibres occupy the central region of the latter, so that a kind of meshwork takes the place of a canal. In transverse section the terminal region of the stem is formed of a thick coating of hypoderm (probably in life covered by a ciliated cuticular layer) somewhat regularly marked (PI. IV. fig. 4) by strise so that the cell-like divisions are frequently wedge-shaped. The hypoderm abuts on a basement-tissue, apparently continuous with that which belongs to the basal apparatus next the disk, and which is in relation laterally with the axes of the pinnae on each side. The wall of the canal of the arm, even in this region, presents a series of fibres which render it hirsute in section, but they do not in every case meet across the lumen. As we proceed downward, however, the sections of the arms are flattened and the margins prominent, so as to form ventral grooves, and the two sides are bound together by transverse fibres, a median junction especially being conspicuous. This meshwork of fibres is better seen in good horizontal and longitudinal sections of the plumes, in which the transverse fibres pass from side to side in almost parallel series, minute nuclei or corpuscles being everywhere abundant, apparently adhering to the fibres, or perhaps indicating their origin from cells of the connective tissue. As in the basal region, therefore, the centre of the arm is composed of a series of reticulations or meshes. The hypoderm also of the arm below the terminal region is considerably thinner, showing that this system of lacunae reaches its culminating point in the termin.al enlargement. The longitudinal fibres inside the basement-tissue are probably those observed in the external views of the arms. The sides of the stem (PI. IV. fig. 1) are rendered plumose by a large number of long slender filaments having rounded or slightly bulbous extremities, which show a linear streak from base to apex, from the presence of the axis or skeleton. The latter was first clearly discriminated as a " skeleton " in Cephalodiscus by Professor Eay Lankester, for 12 THE VOYAGE OF H.M.S. CHALLENGER. it had only been indicated as a septum in my preliminary account in the Annals of Natural History. The examination of fresh specimens of Rhabdopleura off the Norwegian coast had enabled him to detect the existence of a " consistent mesoblastic skeleton " in the lophophoral arms, as in Phor-onis, and thus forewarned he had comparatively little difficulty in making out " a precisely similar skeleton " in Cex>halodiscus. In the latter the so-called " skeleton " of the arms is fixed to the basal apjiaratus formerly described, and seems to consist of a somewhat firm basement-tissue with longitudinal fibres and reticulations in certain parts. It differs considerably from the condition as figured in Rhahdopleura, in which twisted filaments and particles are described by Lankester. The pinnte which pass out from the main stem do not taper, and are composed for the most part of granular hypoderm with a few brownish pigment-cells, and the central axis or skeleton. The pigment gives in some a light pinkish or pale violet blush to the feathery plumes, which in life must have been finely tinted ; and it is further interesting that the same pigment occurs in the lophophore of Rhabdopleura, as shown in Professor Lankester's excellent figures.^ The skeleton (PI. IV. figs. 1, 2a) runs from base to apex and terminates within the cellular tip. It is somewhat dilated where it joins the main stem (PI. V. fig. 2, sJc), and the exact mode of its junction with the axial channel of the latter is difficult to trace, so gradually is it merged into the tissues of the region. No definite ending of these axial structures occurs as in Rhabdopleura, where Professor Lankester figures them as if articulated to the skeleton of the arm, the base of the pinnules dilating, and the central region abruptly terminating, as it reaches the main stem. The dilated bases of the skeletal rods of the filaments in Cephalodiscus join the sides of the reticulated main channel, but no evidence of a continuous central lumen is observed in transverse sections of the free portions, though the double outline, and the appearance of sections of their bases (PI. V. fig. 2), would indicate the possibility of such. Endosmosis at least would thus readily occur. In the transverse sections of the bases of the processes just alluded to a series of apertures appears in the tissue of the arm. The condition as described in Rhabdopleura therefore differs from that in Cephalodiscus, especially in regard to the skeleton of the arm, though the general plan of structure is similar. It would also appear to be more readily made out in the former than in the latter, though perhaps this may be partly owing to the examination of fresh examj^les. Professor Lankester describes the skeleton in Rhabdopleura as cartilaginous, but so far as appearances go in Cephalodiscus it more resembles a structureless translucent basement- substance, probably a modification of connective tissue. It was best followed in the preparations immersed in a weak solution of caustic potash. The skeleton of the arms and their pinnules gives a definite character to the processes, as observed in the sketches. Though perfectly mobile, the pinnules stand out from the stem somewhat stiffly, the curves being for the most part terminal, and thus they ' Quart. Journ. Micr. Sci., vol. xsIf., N.S., p. C21. REPORT ON CEPHALODISCUS DODECALOPHUS. 13 do not mix with each other in an inextricable manner. In the same way the beautiful plumes of the Sabellidse and Phoronis have a certain amount of rigidity from their internal skeleton, while their graceful motions and their branchial functions are in no way interfered with. No special muscular apparatus can be made out in the pinnules, the covering of the central axis consisting of hy[3odermic cells and granules. Nor would such be necessary in regard to the physiology of the organs, the elasticity of the skeletal axis and its con- nection with that of the main stem being sufficient to keep the parts in a position suitable for their functions without any effort on the part of the animal. The main stem has a series of longitudinal fibres, but their muscularity is doubtful. In any case the motions of the disk would influence that of the entire lophophoral apparatus, especially as its great fan-like muscles arise from the skeleton of the basal apparatus of the arms. The bases of the arms are hollow and in communication with the two great cavities (one on each side) of the region (PI. VII. fig. 3, cv) which Mr. Harmer, on good grounds, identifies with the coUar-spaces of Balanoglossus. In section the basal spaces are generally filled with fibres detached from the walls, but in some views definite corpuscles in groups are visible. The latter consist of minute rounded bodies with a central nucleus. From the structure of the parts it will thus be apparent that though probably pervious none of the arms show a clear median channel except at base and apex. The paired cavities connected with the lophophoral apparatus communicate with the exterior by a weU-marked and comparatively large pore on each side in front of the gill-slits. These pores present a radiate arrangement of the hypodermic wall in transverse section and thus are readily recognised ; while in certain longitudinal sections a more or less urceolate aspect is produced. The tentacles of Loxosoma are stated by Professor Vogt and others to be devoid of a central chamber, and the central axis of the same organs in Pedicellina^ is only cellular (and translucent). Even in Rhabdopleura careful examination under most favourable circumstances by Professor Lankester gave no indication of a median canal, even in the main stems. He was unable to detect any definite cell-structure in the skeletal tissue, but observed that it had a refringency indicating a certain density, and presented small twisted filaments and particles within its substance at intervals. The relation of the twisted filaments to the fibres described in the main stem of each plume in Cephalodiscus is a subject that requires further investigation, and the same may be said of the "particles" which occurred at intervals — in relation to the nuclei already described. Whether Rhabdopleura shows any indication of the lacunge at the base of the lophophore is a question also requiring determination, though if such had existed it could hardly have escaped, in the living animal, two observers of such experience as Sars and Lankester. 1 Vid£ Nitsclie, Zeitschr. f. wi»s. Zool., Ed. xx. p. 22, Taf. iii. figs, 1, 2. 14 THE VOYAGE OF H.M.S. CHALLENGER. The arrangement of these numerous tentacular plumes differs considerably from that in Rhahdopleura, in which only two symmetrical tentacular arms with their pinnae occur. In certain young buds, however, in which the first pair of plumes far surpass the others in length, a striking resemblance is temporarily produced to the condition in Rhabdo- pleura. The latter and Cephalodiscus diverge from the ordinary Polyzoa in this respect, both having mobile plumes that curve gracefully in various directions, instead of the somewhat stiffish corona and straight tentacles of the other forms. In Cephalodiscus, besides in all probability branchial functions, they are apparently of great tactile service, and if covered with cilia, as in all likelihood they are, they must aid in providing currents in the cavities of the coenoecium, and, as Sars and Lankester suggest, may indirectly cause food-currents, that is, bring the minute particles which constitute the nourishment of the species within reach of the currents between the buccal shield and the post-oral collar. The efferent currents again would readily find exit by the gill-slits behind the latter lamella. Both Rhabdo^yleura and Cephalodiscus difi'er from the ordinary Polyzoa in the absence of the tentacular web at the base. Both have very long pinnae ; but Cei^halo- discus excels the other in this respect, and is further characterised by the remarkable glandular tips to the arms. The plumes are wholly absent as such in Balanoglossus, and this constitutes a marked distinction externally. As formerly stated in regard to Phoronis,^ however, there are certain evident homologies between the several forms. Post-oral Lamella. In Rhahdopleura Sars described "a strongly projecting, nearly semilunar border of skin, cihated on its edges," and extending from the base of the tentacular arms down- wards on each side, thus forming with the buccal shield a narrow half-tube or channel leading to the mouth, through which the nourishment is probably conveyed to the mouth hy the ciliated tentacles. The condition in Cephalodiscus, however, considerably diverges, since the post-oral lameUa (PI. II. fig. 1) forms a flattened apron-like j^rocess, fixed anteriorly to the ventral surface behind the oviducts, and sloping along this margin backwards to the mouth, the surface gently merging into the mucous membrane of the oral cavity. Moreover, a central space — more or less distinct according to the line of section — occurs between its layers. Laterally and posteriorly it forms a somewhat free lamella. In minute structure this lamella presents two layers of hypoderm, each with a fine basement-layer, and having intermediate fibres, chiefly muscular. A strong series of these passes out from the basement-tissue of the post-oral mucous membrane, and radiates to the outer (ventral) layer of the lamella, for the posterior or dorsal has merged into the mucous membrane at the sides. The hypoderm of the two surfaces just mentioned offers certain differences, especially in the free part of the lamella (PI. VI. 1 Proc. Roy. Soc. Edin., 1880-81, vol. xi. p. 217. REPORT ON GEPHALODISCUS DODECALOPHUS. 15 fig. 2, pi ; PI. VII. fig. 5, pi), that covering the ventral surface being denser and more finely granular, and with a more definite margin, which is probably richly ciliated. This denser and somewhat regularly streaked hypoderm (which also stains more readily) shows several prominent frills or rugas where it joins the oral region, and it just turns the outer edge of the lamella all round and then ceases. The dorsal layer of hypoderm on the other hand is more lax, and is thrown into a series of frills or crenations in the preparations, the streaks in it being more conspicuous than the granules. It resembles, indeed, the somewhat lax hypoderm observed on the pedicle, and like the latter contains numerous pigment-corpuscles which do not readily stain. So far as the structure can form a guide, the ventral surhice would seem to be more important functionally than the dorsal. The oral region therefore has a different environment from that in Rliabdopleura, though the plan of structure follows parallel lines. Thus in the excellent figures of Lankester,^ a well-marked plate situated behind the mouth, and running into the buccal disk in front of it, is apparently the homologue of this lamella. When the disk is folded backwards {op. cit, fig. 2) the two surfaces come more or less into contact, and would thus send currents more surely into the mouth. Lankester does not allude to this region, which lies just in front of his thoracic division in Rhahdopleura. In the flattened surface of the post-oral lamella the buccal shield is closely applied in the preparations, though in life they can of course be separated at will, thus permitting the currents caused by the cilia of the opposed surfaces to reach the oral aperture. As its posterior face has perhaps only to perform the function of separating the currents connected with the mouth from those of the gill-slits, the difi"erences in structure are thus explained. The post-oral lamella may have some relation to the Molluscan foot, and also to the post-oral ring of cUia in Polygordius, especially as a ciliated groove in the latter runs between it and the mouth. Harmer's view that it is homologous with the oper- culum of Balanoglossus, as described in Bateson's valuable and suggestive papers ^ on this form, appears to be well founded. Digestive. System. Mouth. — The margin of the oral lamella leads on each side (PI. III. fig. 3 ; PI. VI. fig. 2, m) into the mouth, and in some ventral views it passes straight inwards to the sides of the latter, and forms a transverse margin anteriorly. The edges of the mouth are slightly raised or frilled laterally and posteriorly, the latter often being spout-shaped. 1 Op cit., pi. xxxviii. figs. 1, 2. 2 Quart. Journ. Micr. Sci., April 1884 ; also Studies from the Jlorph. Lab. Univ. Camb., vol. ii. part i., and vol. iu. part i. The author has overlooked some remarks pre\'iously published iu this country on Balanoglossus. Vide Nemerteans, Bay. Soc, 1873, p. 144. 16 THE VOYAGE OF H.M.S. CHALLENGER. Anteriorly it is devoid of any well-defined boundary other than the attachment of the pillar of the great buccal shield, and leads directly upwards into the alimentary canal (PL III. fig. 1, m). Moreover, as Mr. Harmer first noticed in his sections, a sohd diverticulum proceeds upwards dorsally into the stalk of the buccal shield, and this may fairly be held to be the homologue of the notochord of Balanoglossus. In transverse section it is nearly circular and presents a somewhat regular arrangement of its cells, so that a concentric appearance is frequently present. In longitudinal section, on the other hand, the process, which is small, has a slightly bent clavate outline, a dotted axis indicating the lumen continued from the alimentary canal. The mucous membrane of the buccal chamber and gullet are thrown into many prominent rugse, from the dense glandular nature of the tissue. The latter is especially thick on the ventral side of the mouth (the region lying in front of the post-oral lamella), and a strong layer of muscular fibres passes to the basement-tissue of this region, which must thus possess considerable mobility. The surface of the mucous membrane is apparently richly cUiated, the cylindrical epithelium of which it is composed being so closely arranged as to give a finely striated character to the tissue. In some preparations a thin film occurs on the surface of this glandular layer, but this is evidently due to mucus, and not to the separation of a superficial or cuticular coat. Pharyngeal Region. — Beneath the post-oral lamella and immediately behind the pores of the second region are a pair of gill-slits, which were first clearly recognised as such by Mr. Harmer, who has kindly interested himself in the structure of this form, and whose very thin sections enabled him to unravel certain points wliich would otherwise have been obscure. The folds leading to these in transverse section are shown in PI. VI. fig. 2, gs. Immediately behind the collar-pores the ordinary hypodermic coatino- of the body becomes continuous with the translucent wall of the slits, which seems to be a modified continuation of the pharyngeal mucous membrane. The granules are finer, and the whole tissue is more translucent. It also does not stain so well as either the collar-pores or the pharyngeal lining proper. In connection with this structure it is interesting to note that Bateson^ mentions that the gill-slits in Balanoglossus arise as dorso-lateral evagiuations. As soon as the posterior boundary of the mouth is completed, and this is easily recognised in the preparations by the appearance of the pigment-ceUs in the dorsal layer of the hypoderm of the post-oral lamella as it now stretches right across the ventral surface, the spacious pharynx presents a thickly folded wall of the same kind of minutely glandular tissue. The projection of some of these thick folds of glandular tissue under the post-oral lamella, sometimes causes peculiar appearances in transverse section, as if special diverticula existed. Bounding this thick glandular wall externally is a firm basement-layer, probably of a highly elastic nature, and it is to this coat that the muscular fibres formerly alluded to are attached. ^ Ox'. cit., p. 14. REPORT ON CEPHALODISCTJS DODECALOPHUS. 17 (Esophagus. — The firm and finely glandular oesophageal region (PI. III. fig. 3, a little behind ce), which is characterised by the paucity of its folds, is comparatively short, for it merges into the gastric chamber a little behind the termination of the buccal shield (PI. III. fig. 3). Stomach. — The stomach forms a large cavity, which in some instances fills the whole of the body-space, with the exception of the dorsal area occupied by the intestine. In certain longitudinal sections the oesophagus forms a comparatively limited tube, while the stomach fills the entire body-cavity (PI. III. fig. 3), liut dorso-ventral flattening of the former may have occurred in such cases. The wall of this region is also somewhat thick, and when fully formed, is distinguished from the oesophageal region by its more numerous and often symmetrical folds, and sometimes by a differentiation into three pseudo-strata, viz., a deeply stained, granular, epithelial, superficial layer marked by close parallel striae, a pale intermediate region with granular glands, and externally another deeply stained stratum of granular glands. It is not implied that there is separation in continuity between the three regions indicated, but rather that either from preparation or otherwise such a condition appears in the stomach. The colour of this region in life is probably as characteristic as in Rhabdopleura, where it is yellowish, but this feature cannot be made out in the spirit-specimens, for all are bleached. In the preparations it not unfrequently happens that rupture of the alimentary wall occurs along the middle region, so that the complexity of the folds in the body-cavity is increased. Externally the whole organ is surrounded by a firm basement-layer continued from the oesophageal region, and from its elasticity this probably subserves certain of the functions of a more elaborate contractile apparatus, for muscular fibres have not been clearly determined. The chamber narrows posteriorly towards the pedicle, and terminates in the intestine, the glandular wall, however, undergoing no change of note. In Rhabdopleura, Sars ^ mentions that the stomach has tolerably thin walls, but in all probability he speaks comparatively, as from the nature of the glandular tissue such organs have proportionally thick walls. Intestine. — As in Rhahdopleura, the stomach terminates at the posterior end of the body-cavity at the base of the pedicle by a wide aperture in the intestine. In favourable sections (PL III. fig. 3) the intestine is observed to leave the ventral side of the fundus of the stomach, and passing under it, curve forwards along the dorsal wall. The glandular lining of the ventral wall of the stomach passes evenly into the intestine and gradually diminishes in thickness, whereas the lining of the posterior wall shows a characteristic bend at the pylorus, and again a fold in the anterior wall of the intestine behind the fundus (PI. III. fig. 3, near vtd). This peculiar fold in the wall of the canal probably indicates a tendency to the formation of a second or pyloric stomach, as in Phoronis, and is therefore of considerable morphological significance. The intestine ' Op. cit., p. 5. ^OOIi. CHALL, EXP. — PART LXII. — 1887.) Qqq 3 18 THE VOYAGE OF H.M.S. CHALLENGER. acquires its narrow firm texture even before taking the forward curve along the dorsal wall. The gut forms a somewhat large canal, which proceeds along the dorsal wall above the stomach to terminate on the anterior prominence of the body in the anus, which lies considerably above the region of the plumes, and, indeed, the area in which the large pigment-spots are situated intervenes. There is thus a decided difference when the anal region is contrasted with that in Rhabdoiileura, in which the anus is situated close to the base of the tentacular arms on the dorsal side of the animal, that is, on what Sars terms the posterior region. Lankester, however, in his figure ^ shows the anus elevated on a rectal cone, with a depression between it and the base of the lophophoral region ; and, moreover, on the lateral and ventral faces of this cone isolated blackish pigment-corpuscles are present. No rectal cone is present in Cephalodiscus, for the prominent anterior end of the body carries the anus on its summit. It has the form of a more or less elliptical aperture, often of considerable size. The minute structure of the wall of the intestine differs considerably from that of either gullet or stomach by its well-defined boundary- line — -both externally and internally. The external consists of the firm basement- layer, which ventrally invests the stomach, and which at each side of the usually elliptical or transversely elongated gut (in section) runs into a thinner basement- layer bounding the canal dorsally within the proper wall of the body. The granular glandular coat which follows is narrow, and is limited internally by a remarkably definite margin in section, so that the canal is at once distinguished in the prepara- tions. In Professor Lankester's section^ of the intestine in Ehabdoplcura, no such compact and definite wall is observable, the gut apparently being enclosed by a somewhat moniliform layer of cells. This divergence in structure doubtless indicates difference in function, probably in relation to the free and the fixed conditions of the respective animals. The terminal region of the gut (or rectum) frequently shows considerable dilatation, the indigestible debris being probably sent out at intervals in a stream, and it is this deposit which is occasionally found in certain crevices of the ccenoecium. Food. — A survey of the fine muddy debris found in the alimentary canal, and especially in the intestine, shows that the currents — set up in the surrounding water by the plumes, and conveyed towards the oral aperture by the ciliated surface of the post- oral lamella and the great buccal disk — carry inward, amongst indigestible sponge- spicules and sand-particles, many Diatoms, bodies resembling minute Thalassicollidse and other Eadiolarians, as well as organic particles of various kinds. When large forms like the Ascidians flourish on a diet composed largely of Diatoms, it is evident that this minute type is amply cared for in this respect. The honeycombed condition of the ccenoecium, ' Op. cit., pi. xxxviii. fig. 2, h, and pi. xl. fig. 11, e. ' Op. cU., pi. xli. fig. 13, h. REPORT ON CEPHALODISCUS DODECALOPHUS. 19 and the multitude of external apertures, thus suffice to place the little ciliated animals in favourable circumstances as regards food, especially when the nature of their surroundings is taken into account. Body-Wall In Rhabdopleura two layers of the body-wall were clearly distinguished by Professor AUman, who had not the material aid which sections give the younger inquirers. He called them ectocyst and endocyst, the latter " a very delicate membrane." Professor Sars, subsequently, with fresh specimens at his disposal, denied that there was any endocyst " (unless we consider the glassy skin, which closely surrounds the digestive apparatus, to be an endocyst), consequently also no perigastric fluid." Professor Eay Lankester, again, from the examination of living examples, recently observes of Rhabdo- pleura that " the tissue which bounds the body-cavity consists of fusiform cells tapering into fine fibres, sometimes branched." He further figures the structure of the body-wall in optical and in transverse section, the coat formerly mentioned having within it apparently a basement-membrane with ciliated enteric cells projecting from its inner surface. In whatever way this form is considered, the structure of its body-wall very much difi"ers from that of Cephalodiscus. I am unable from the mode of preparation of the examples (in spirit) to say much about the pigment of the surface of the skin, but in some numerous specks of a reddish- brown colour are still visible over the entire surface (PI. II. fig. 1) ; while, as already mentioned, the buccal disk almost always presents the dull reddish band. In all probability it is brightly tinted in life. In this respect it approaches the condition in Balanoglosus ; thus Balanoglossus howalevslii has a white proboscis, a brilliant red-orange collar with a whitish line round the operculum, while the rest of the body is orange-yellow. Externally the surface is probably covered in life by a delicate ciliated cuticle, but this cannot be diff"erentiated in the preparations. The same difficulty is met with in the cuticular tissues of the Nemerteans. A decided difference is thus apparent between Cephalodiscus and Loxosoma, in which the cuticle is considerably developed. What remains is a well-marked layer of hypoderm (PI. VI. fig. 2, /i^) of the usual granular, glandular structure. The coat just mentioned attains its greatest thickness at the base and on the pedicle, but this may be partially due to corrugation from contraction. In this layer are the numerous pigment-corpuscles and gland-cells, which latter do not readily stain with carmine. It is bounded internally by a basement-layer, which is thin dorsally, but better marked ventrally, especially behind the mouth, for the layer of longitudinal muscular fibres now forms an additional coat in this region, and rests against the base- ment-layer. Like the hypoderm the latter passes over the pedicle at the posterior end of the body, and both are often thrown into wrinkles from contraction. The body-wall 20 THE VOYAGE OF H.M.S. CHALLENGER. is further strengthened ventrally by a muscuhxr layer, but this will be described under its special head The preparations showed traces of what might be an epithelial layer on the inner surface of the before-mentioned basement-tissue, but such were far from being distinct. In the living Rhabdopleura, ou the other hand, such an epithelial layer is described by Professor Lankester, under the name of " enteric epithelium;" and its distinctness in this form suggested its presence in Cej^halodiscus. Body-Cavity. — The foregoing layers enclose the body-cavity {co in sections of buds), which is generally filled more or less completely by the alimentary canal. In the preliminary account ' it was pointed out that this investment was probably homologous with the " thin glassy skin " of Sars surrounding the digestive canal in Rhahdopleura, and that the preparations gave no evidence of perigastric fluid. Though the existence of a body-cavity was not specially noticed, the preparations did not warrant a denial of its presence in Cephalodiscus, as Professor Lankester states in a recent paper," for thus the hypoderm and basement-tissue must have been amalgamated with the coat of the alimentary canal, which was not the case. This statement does not in any way detract from the credit which Lankester has in clearly describing for the first time the chamber in the living Rhahdopleura. Small nucleated corpuscles were occasionally seen in groups in the cavity of Cephalodiscus in the sections, but they may have been intro- duced from other sources. Neither Sars nor Lankester observed such in the living Rhahdopleura. In sections the continuation of the body-ca-vdties in front are seen a little behind the paired cavities connected with the lophophoral apparatus, and are likewise surrounded by basement-tissue. Muscular System and Pedicle. As previously mentioned, the short ventral surface of the body is continued into the cylindrical pedicle, which is invested by the hypoderm and basement-tissue, the former being thrown into numerous and rather regular transverse wrinkles in contraction, and being thicker dorsally than ventrally. At the terminal region of the foot (PI. VI. fig. 1, hps) the hypoderm is much increased in thickness, but has the same structure. It is free from the wrinkles which characterise other parts of the region ; and appears indeed in favourable preparations to form a flattened sucker-like disk. The basement-layer within the terminal hypoderm is thick, and has attached to it the longitudinal muscular bands, so that it is possible it may be occasionally used as a sucker like that of Loxosoma, or like the larval organ in Balanoglossus. The entire pedicle within the basement-tissue is filled with the longi- tudinal fibres, which arise on the ventral wall of the body in the region of the mouth, where they present the form of a thinner lateral region and a denser central, the latter in ' Op cit, p. 344. 2 Polyzoa, Ency. Brit., v6l. xix. p. 436. REPORT ON CEPHALODISCUS DODECALOPHUS. 21 the sections being somewhat semicircular and of considerable thickness. These fibres pass backwards (causing the ridge on the ventral surface in contraction) towards the pedicle, which they enter, filling the central space. In transverse section the appearance of the latter varies, but in the best preparations a certain uniformity is observal)le (PI. IV. fig. 5), viz., beneath the smoothly rounded hypoderm a median fold occurs in the basement-tissue dorsally, while a much larger and wider one occurs ventrally, and at each side of this an inner longer and an outer shorter process exist. The whole has a symmetrical appearance. Such an outline would indicate that the basement-tissue was elastic and that no circular muscular fibres existed, and indeed from the descriptions of Vogt and other authors similar elastic tissues subserve the function of circular fibres in Loxosoma. In some sections of unstained examples the thick basement-tissue had a somewhat diiferent aspect (PI. VII. fig. 2), probably from the condition after immersion in spirit ; this, however, was exceptional. A layer of fine longitudinal fibres lies on the inner surface of this coat, and from it numerous transparent and somewhat gelatinous fibres of connective-tissue pass to the central area, which contains large structures deeply stained and almost resembling gland-cells, but which appear to be sections of the long mobile muscular fibres continued from the fundus of the body into the pedicle. The close approach made by these to the "muscular band of closely-set fusiform cells " described by Lankester in the stalk of Rhahdopleiira is noteworthy ; they probably represent a further development of that tissue — in which the cells have disappeared. A similar series of muscle-cells in the stalk is described by Harmer in his valuable paper on Loxosoma} In contraction the transverse wrinkles of the basement- tissue are so close that they resemble a circular muscular coat. No distinct canal is thus observed in the centre of the pedicle throughout its length. Towards the tip, however, certain spaces containing what appears to be a coagulable fluid occur, and also globules and granules, as explained in connection with the buds. The pedicle in Cephalodiscus is thus evidently a development of the body-cavity, which in the young bud freely opens into its upper region. It difi'ers, therefore, from the soft stalk of Ehahdo2)leura, first so clearly described by Professor AUman, for that has no connection with the body-cavity in the adult ; yet as the development of both forms is unknown a little reservation is necessary, especially as it also contains muscular fibres in Rhahdopleura, and gives rise to the buds and branches. The fibres of the stalk in Rhahdopleura, according to Professor Sars, proceed " rather high up on the ventral side," and " over the skin which encloses the digestive apparatus. Its ventral fibrous part may still be traced (see fig, 15) a considerable distance forward in the form of a rather wide, clear, skin-border which gradually disappears in front of the cardia. In this skin-border the fine longitudinal fibres may still be distinctly observed diverging like radii, but I was not able to trace their course further."^ The arrangement of the parts 1 Op. cit, p. 19. 2 Op. cit., p. 10. 22 THE VOYAGE OF H.M.S. CHALLENGER. iu Rhahdojileura has also been carefully figured on a larger scale from life by Professor Ijankester.^ There are thus interesting points in analogy between the two forms in regard to the polypide-stalk, but at the same time important structural diflferences. Professor Lankester again holds that the chitinous covering (his Caulotheca) of certain parts of this region is the true homologue of the ccenoecium of an ordinary Polyzoon, and there is something to be said in favour of such a view. As already explained, however, I prefer to adhere to the term already in use, especially in the present uncertainty in regard to the development of the types under consideration. Nor do I fully share my friend's views concerning the "serious error" of confounding the pedicle in the forms just mentioned with the funiculus of the Eupolyzoa. The structural relations of the organ in the several forms no doubt differ, but the remarkable analogy in regard to the budding shows that from the " vagrant protean funiculus " of the Gymnolsematous Ectoprocta to the pedicle and soft stalk of Cephalodiscus and Rhabdopleura there is at least one striking function carried out often on very similar lines. Variations, it is true, occur, in which the endocyst is associated with the funiculus in producing the buds in the marine Ectoprocta, but this does not affect the main point at issue, but rather brings the analogy closer -with such forms as CejyJialodiscus. The pedicle on the other hand nearly resembles the stalk in Loxosoma. There is no pedal gland, however, in Cejjhalodiscus. One of the most remarkable points of resemblance between Cephalodiscus and Balanoglossus is the occurrence of a pedicellate structure in the young of the latter [Balanoglossiis kowcdevskii) as described by Mr. Bateson.^ This organ presents itself on the disappearance of the cilia as a small papilla, and is situated at the central part of the posterior surface. Moreover, it is directed ventrally, just as the pedicle of Cephalodiscus is, and, indeed, the general contour of the young form at this stage simulates the condition in Cephalodiscus. This conical process serves as a sucker by which "the animal can attach itself to foreign bodies sufficiently firmly to prevent being washed off by a stream of water from a pipette. The anterior surface of the proboscis is also slightly suctorial, and by thus fixing itself posteriorly, and extending the proboscis, it is ahle to creep slowly about, somewhat in the manner of a leech." The organ " subsequently attains a considerable size and is traversed by several wrinkles. It afterwards entirely disappears, but as to its mode of disappearance I have no certain observations. It would appear to occur very suddenly at the stage when the animal possesses seven to eight gill- slits. I have found animals with eight gill-slits which possess this sucker, and also animals of apparently the same age without it ; hence it may be inferred that it under- goes a rapid atrophy at this point." Mr. Bateson further observes that similar suckers occur as larval organs in Tunicata, Ganoids and Amphibia, but these fall far short of the 1 Op. cit, pi. xl. fig. 12. 2 Quart. Journ. Micr. Sci.; and Stud. Blorpli. Lab. Univ. Camb., vol. iii. part i. p. 3, pi. i. figs. 1-4, 1886. REPORT ON CEPHALODISCUS DODECALOPHUS. 23 interest connected with the condition of the permanent pedicle in Cephalodiscus. This is a truly homologous process of the posterior body-cavities, and in sections at the base, as already explained, the septum is visible {vide PI. V. fig. 4). It is further placed ventrally in regard to the intestine, and its tip, so far as structure shows, is in all probability also used as a sucker ; indeed, the buccal shield and the tip of the pedicle form far more efficient and permanent sucking disks than ever occur at any stage in Balanoglossus. It is the disappearance of the " tail " in the latter which alters the character of its reproduction, and confines it so far as known to the sexual process. The presence of this organ in Cephalodiscus, again, and its striking activity in bud-formation, open up new fields in homology and assist in bridging over the gaps which formerly existed between apparently isolated groups. . Nervous System. In the preliminary account of Ccphcdodiscus^ no distinct nervous system was observed. In Mhabdopleura all that the careful observations of Sars could distinguish in regard to a nervous system was that " immediately behind the anal aperture, between the terminal part of the intestine and the dorsal wall of the gullet, which here forms a little concavity, there appeared a clear cellular body (fig. 1 5, r) in which several evident nuclei were visible. I cannot, however, pronounce any decided opinion as to the significance of this object ; it can scarcely be a nervous ganglion, as it does not lie in the substance of the body itself, but only in the thin external skin which encloses the body." Lankester, again, considers that it is probable that the "clear cellular body" just mentioned by Sars was a gonad. It is doubtful, however, whether the latter interpre- tation can be maintained, especially when the condition now known in Cephalodiscus is considered. At the base of the plumes in the latter, and situated over the median space existing there — above the mouth — is in section a region (PL VI. fig. 3, nc ; PI. VII. fig. 3, tic), which appears to me to contain the central nervous system of the animal. The area is of considerable proportional size, and is minutely cellular and granular, while fibrous bands stretching from it occur in other views. Its inner face rests on the thick basement-layer bounding the collar-spaces at the base of the arms, and to the opposite wall of which the muscular fibres of the great buccal disk are attached ; while its outer covering is formed by a thick layer of hypoderm. . This region forms an elevation dorsally between the bases of the arms (PI. VI. fig. 3, nc), and thus is in close communication with these organs, while it is also within a short distance of the pigment of the oviducts, though no branches have been traced to these organs. It extends a considerable distance laterally on each side along the basal region, whence the plumes spring, and for some distance on the dorsal surface of the buccal disk. The relations of the nervous system to ' Ann. and Mag. Nat. Hist, ser. 5, vol. s. p. 337, 1882. 24 THE VOYAGE OF H.M.S. CHALLENGER. the rudimentary notochord will be specially referred to by Mr. Harmer in a note appended to this paper. The view of Professor Sars that this structure cannot be a nervous ganglion, because it does not lie in the substance of the body, would not seem to merit the importance he attaches to it, when the condition of the great nerve-cords and ganglia of Annelids, for instance, are considered. These are purely hypodermic in position, lying between the latter and the basement-layer beneath. In Phoronis, again, a somewhat similar condition to that in Cephalodi!^cus is present, the nervous concentration taking the form of a ring round the mouth at the bases of the tentacles, and this, like the cord running along the foot, is epidermic (hypodermic) in position.^ The same position (hypodermic) of the nervous system is found in Balano- glossus, so that the relations of the nerve-centre are by no means exceptional. The position of this nervous centre would not appear to correspond with the larval brain of Loxosoma as described by Mr. S. F. Harmer. In the stalked or adult Loxosoma, again, the ganglion is subcesophageal, and is therefore not the homologue of the larval brain. A very different condition, however, occurs in Cephalodiscus, in which the young buds soon present this and all the other organs of the adult, although it is true the develojjment and perhaps metamorphoses of the species from the egg are unknown. It has also to be borne in mind that certain parts of the central nervous system may have been suj)pressed, and that we may have only a much modified peripheral system remaining. Sense-Organs. In the preliminary account of Cephalodiscus, the close relation of the so-called " eye- spots" to the ovaries was duly pointed out, and recently the examination of more satisfactory sections made with a microtome demonstrated at once their true nature'; viz., that they are oviducts with thick pigmented walls. The resemblance of these structures, both externally and in section, to a modified organ of sight, is one of the most remarkable features in the animal. Their description will be given subsequently. In Rhahdopleura, Lankester mentions the occurrence of five spherical pigment- corpuscles at the superior dorsal margin of the buccal disk, and regards them as rudi- mentary sense-organs for the perception of light. The position of these is certainly peculiar for organs of vision, but if Rhabdop)leura has a trace of the central nervous organ observed in Cephalodiscus, these would readily be within reach of its communications. It is interesting that eyes occur in most larval forms of Loxosoma; indeed, they make their appearance when the larva is still in the egg. They are situated under the hypoderm, and resemble pigment-masses. In Loxosoma leptoclini, for instance, Harmer describes a pigment-spot on each side of the larval brain (even when the lumen is stifl present after involution). They consist of creseentric reddish-brown masses of pigment, with a 1 Caldwell, Proc. Roy. Soc, vol. xxxiv. p. 372. EEPORT ON CEPHALODISCUS DODECALOPHUS. 25 prominent lens, but whether the latter is a solid body, a cuticular formation, or only fluid in the centre of the mass, is not explained. Mr. Harmer bases one of his arguments for the interpretation he has given of the " dorsal organ " (brain) on the constant relation to each other of ganglion-cells, a fibrous layer and the eyes, and he seems to have facts in his favour. In Bhabdopleura a ciliated tubercle (considered by Lankester and others to be a sense-organ) was discovered by Sars on the dorsal surface at the base of each arm of the lophophore. In Loxosoma Vogt ^ also describes a tactile papilla on each side near the arms, and Salensky ' traced nerves from the ganglion to these posterior sense-organs in Loxosoma crassicauda. No such organs, however, can be observed in the preparations of Ce^ihalodiscus, though it is possible the search in the living animal may be more successful. The oral folds of glandular tissue and the parts connected with the gill- fissures are probably highly sensitive and ciliated, but no other sense-organs could be observed. Reproductive Organs. No diflferentiation was noticed in regard to the sexes, and no dimorphism of the zoids, as in Professor Ehler's remarkable burrowing form {Hypophorella expansa^). Nothing is more striking, however, than the profusion of buds and the abundance of ova, apparently one of the chief ends of the species being propagation. The chambers of the co3noecium present many of the large ova, and they are occasionally found amongst the plumes, as in Phoronis, with its swarms of minute eggs ; but such in the former is probably accidental. Almost every adult again bears one or more buds attached near the tip of the pedicle. Ovary. — In most specimens a pair of large ova are observed projecting anteriorly (PL III. fig. 2, ov), so that their pure white colour is recognised through the attenuated integument above and behind the eye-like oviducts, which, as it were, mark the anterior boundary of the ovary. In section it is found that a septum passes from the median wall of the rectum to the opposite wall of the body, thus dividing the body-cavity in that region into two spaces, in which are the ova supported on a pair of lateral mesenteries. As soon as the glandular tissue forming the dorsal wall of the buccal cavity appears, the mesenterial septum just indicated is attached to its basement-layer externally, and also more or less in the median line. The septum disapjDears on the approach of the stomach, or about the posterior termination of the great buccal disk, and any product remaining towards the end of the latter is generally pushed to one side. As a 1 Sur les Loxosoma des Phascolosomes (sep. copy), p. 8, pi. xii. fig. 1. ' Ann. d. Sci. Nat. (Zool.), ser. 6, t. v. p. 12 (art. 3), pi. xii. figs. 2, 3. ' Abhandl. d. k. Oesellsch. d. wiss. Gottingcn, 1876. (zool. chall. bxp. — PART LXii. — 1887.) Qqq 4 26 THE VOYAGE OF H.M.S. CHALLENGER. rule, one of the spaces anteriorly contains a group of small ova (PL VI. fig. 2, ov), with nucleus and nucleolus. In the smallest ova observed the nucleus is very distinct, and of a more or less rounded form — with a large nucleolus. The yolk outside the former is minutely granular, and sometimes, to judge from the preparations, does not c^uite fill the egg-capsule. As the eggs increase in size the yolk-granules become somewhat more distinct, the nucleus being large and rounded, and the nucleolus a highly refracting circular body of considerable size. Moreover, as they increase they tend to move away from the smaller ova, and indeed bulge outwards on the opposite side. The largest ova, of which, as a rule, there is but one, are much more coarsely granular, and present only a nucleus with minutely granular contents, the nucleolus having disappeared or having become very indistinct. The capsule surrounding each of the eggs is definite and tough, and the eggs are further enveloped by a common layer, which in all probability is allied to the follicular layer in fishes and other types, and this dips between each, apparently enclosing the ova in separate chambers. The ova are probably extruded through the oviducts with the pigment in their walls, and it is noteworthy that, as Mr. Harmer has specially pointed out to me, no pigment is visible in young individuals in which the ovaries are not yet functional. There is thus an approach to the condition in Loxosoma in this respect, for the latter has oviducts for conveying the ova into the vestibule. The same may be said with regard to Fhoronis, though in this the so-called nephridia lie on the other side of the intestine. The comparatively large size of these ova recalls the condition in the Arctisca, in which the great ova produce embryos about a third the size of the adult. Oviducts and Pigment. — On the ventral surface are two large and conspicuous pigment-spots, which as already mentioned closely resemble eyes (PI. III. fig. 1) : they indeed give a most peculiar and characteristic aspect to the animal when viewed from the ventral surface. These are placed a little in front of the anterior margin of the post-oral lamella, and in ordinary preparations are more or less covered by the buccal disk. They are circular or irregularly rounded, and of various shades of brown or reddish-brown, with occasionally a tinge of violet. In ordinary views from the exterior these organs show a pale centre surrounded by a broad margin of pigment, so that the former assumes the aspect of a lens (PI. VI. fig. 2, od). In section the hypoderm in the central region is hollowed more or less deeply, and its wall presents a finer, columnar arrangement of its cells, and, in addition, it is almost, if not quite, transparent. Moreover, an aperture exists in the centre, as already mentioned. In shape the ducts have the form of a blunt cone, the apex of which abuts on the modified hjqjoderm, while its base rests on the small anterior ova. The pigment-cells form a thick layer ; their inner pale portions projecting internally, so that their resemblance at first sight to an optical apparatus is remarkable •; yet Dr. Marcus Gunu, who kindly examined them in a special manner, could discover no refractive mechanism. REPORT ON CEPHALODISCUS DODECALOPHUS. 27 Nothing intervenes between their cavity and the ovigerous envelope, and indeed, as formerly stated, they are so closely related that they are generally removed together in dissection. No nerve-fibres could be traced from the central apparatus to these organs, though appearances were favourable to such a view. Mr. Harmer, who independently arrived at a similar conclusion with regard to the function of these organs, thinks the pigment of an excretory nature. When the life-history of this peculiar form is more completely marked out, it is possible that these oviducts may be found to be light- producing organs. On extrusion the ova (PI. V. figs. 5, 6) are pure white, and either pyriform or rounded in shape. Each is provided with a well-formed pedicle of the transparent investment, truncated at tlie extremity for attachment. The diameter of the circular kind is about 0'55 mm., or including the short stalk 0-63 mm. In the more pyriform or longer forms the total length (including the stalk) is about 07 mm., and the transverse about 0*5 mm. The diameter of the contained yolk is about 0"36 mm. in the former kind, and in the latter is nearly of the same proportional size, though more ovoid in outline. A large space (perivitelline) existed around these eggs, but whether endosmose had occurred after deposition, or otherwise, is at present unknown. The capsule is hyaline and structure- less, presenting only a few wrinkles at the base of the stalk, which is hollow and truncated at the tip. The opaque white central region is coarsely granular, as in the intra-ovarian eggs. Each is attached to the wall of the chamber of the ccencecium by the pedicle, though as now seen, that is after the action of spirit, many have become detached. The products of the foregoing ova are yet undiscovered, though in all probability they are motive embryos which carry the species to fresh sites on which to construct the coenoecium. Such embryos had all been swept out of the chambers either before or after capture, as no trace of them could be found in the preparations. The reproductive organs seem to be more largely developed in this genus than in BJutbdo pleura, none indeed having been found in the latter by its discoverer. Professor AUman, or subsequently in the living examples by Professor Sars. Professor Lankester, however, was successful in finding a testis in " the form of a much elongated sac ending blindly at one end and opening by the other to the exterior by a special pore." The latter occurs near the anus. The position of the aperture of the male generative system thus probably indicates what will be found in the living Cephcdodiscus, and it is possible that the male organs are developed at a different period from the female in the same animals, or that the males exist in separate colonies. In some of the sections of the reproductive organs minutely granular masses like sperm sacs were occasionally seen, but further examination gave no grounds for supposing that they were anything but im- perfectly preserved contents of the ovaries. 28 THE VOYAGE OF H.M.S, CHALLENGER. Budding. No feature is more striking in this species than the great abundance of buds throughout the entire series of adults inhabiting the ccenoecium. Very few exist without them, most presenting from one to three or more buds at various stages, and in all cases these proceed from the terminal region of the pedicle. The early buds consist of minute and, in the preparations, somewhat pale clavate or pyriform bodies attached by the narrow end to the pedicle. They are situated round or near the tip (which has the thickened investment), and are observed as little processes j)rojecting from the hypoderm. They appear to rise close above the terminal disk, and, in their earlier stages at least, seem to have a channel of communication with the pedicle of the adult (PI. VI. fig. 1, ch). This channel runs upwards along the stalk of the bud as far as the downward bend of the hypoderm (shield) where it is lost. In the earlier buds longitudinal sections, for an examination of the best of which I am indebted to Mr. Harmer, show very clearly the three regions of the body, viz., shield, nuchal region and body-proper, and each has its special spaces. Thus the shield presents its single cavity, the nuchal or collar-region its paired spaces, and the third region contains the body- cavity divided by the mesenteries (dorsal and ventral). Very soon the pedicle is differentiated from the disk ; the young animal consisting of a large, flat, and somewhat thick disk, and a short, broad, and somewhat crenate pedicle attached to the pai-ent. Two ovoid opaque thickenings are observed about this stage dorsally (PI. V. fig. 3, Ip), and these are the earliest indications of the lophophoral plumes. In cross section from above downwards the shield consists of a thick layer of hypoderm which attains its greatest development ventraUy, for this coat is in the form of a flattened ring round the central chamber. The latter is bounded by a firm investment of basement- tissue having fibres internally. It is this chamber in the younger forms that communicates with the pedicle of the adult. The two dorsal bosses are simply hypodermic growths of the upper layer of the shield, and present an elevation between them. At a somewhat lower level the elongated cavity in the centre of the shield gives place to a rounded median chamber (PL V. fig. 3, vc), with the dorsal hypodermic mass above it ; while the great bosses of the plumes lie on each side. The latter still present the same structure as the hypoderm of the buccal disk, though there is a faint indication of a linear streak in the centre. In the next slice the two nuchal or collar-spaces appear, with the alimentary canal in the centre. In the septum, between the collar-spaces, the tip of the notochord is observed, and it appears to be larger proportionally in the young than in the adult. The basement-tissue bounding the nervous centre is now defined, and the shield and bosses are diminishing. Behind this the folds of the alimentary canal fill the central region of the body, the intestine being especially distinct as a median canal, the longest axis of which is directed ventrally instead of being transverse REPORT ON CEPHALODISCUS DODECALOPHUS. 29 as in the adult. Posteriorly the body presents, firstly, two lateral chambers with fibres internal to the basement-layer, and secondly, a median (dorsal), with a process of basement-tissue running to the ventral line, and forming the septum between the two former. The pedicle seems to be filled with muscle-cells. In the terminal region of the pedicle of the adult certain areas containing a coagulable fluid with globules and granules are present, and occasionally in the elevation caused by a developing bud one or two of them are observed ; and they are also seen in the buds here and there in the course of the central muscles of the pedicle. Some appear to be nucleated. The origin of these bodies is unknown, but they may be connected with the mesoblastic or hypoblastic elements for the buds, though this is only a conjecture. In regard to the presence of the three primary embryonic elements, as a rule, in such buds, the remarks of Professor Haddon ' seem to me to be very interesting, but they have yet to be proved. The doubt remaining in the present case, for instance, relates to the precise nature of the hypoblastic elements. In connection with this subject it is well to state that some observers, as Mr. Harmer, demur to the connection of the latter layer with the origin of the buds. As development proceeds, the anterior or disk-bearing region of the body increases much more in proportion than the posterior or pedicellate part. The buccal disk is rapidly enlarged, and shows traces of the broad arch of pigment anteriorly and the reddish band posteriorly, as well as the two median elevations on the surface. The posterior moiety of the disk is especially large. Moreover, the body begins to project outward superiorly, and the papillge of the plumes increase in number. These papUlse form a slightly curved row in front of the dorsal projection of the body (woodcut, fig. 2). In the next stage the disk is almost completely formed, though of smaller size and more massive than in the adult. It is thickest anteriorly, much thinner posteriorly. In the former region it presents in transverse section the elongated central chamber, bounded ventrally by the massive hypoderm of the shield, which has the two median prominences observed in the adult. This hypoderm folds over at each side, and is continued as a thinner stratum dorsally, with a basement-layer next the chamber. Two additional structures have now appeared, viz. , a dense (narrow) layer outside the basement- tissue just alluded to, and fan-like fibres from the middle of the dorsal wall of the chamber. Two of the most ventral plumes (the first to appear) have now attained some size, the tip being furnished with the radiate terminal glands having the central chamber, and the sides with short papillse representing the filaments. Both the latter and the axes of the plumes are composed of hypodermic tissue, and are apparently solid. Longi- tudinal striae are visible along the centre of the axis, and they run into the lining of the cavity in the terminal enlargement. Then the central chamber appears dorsally 1 Quart. Journ. Micr. ScL, N.S., vol. xxiii. p. 34, 1883. 30 THE VOi'AGE OF H.M.S. CHALLENGER. in the middle liue (PL V. fig. 3, vc), and the fibres to the ventral wall of the shield radiate from the basement-tissue. Dorsally several less developed plumes also make their appearance (PI. VI. fig. 4), and the pale (nervous) area lies under the hypoderm. Next the two nuchal chambers take the place of the former, and the hypoderm over the median nerve-area increases much in thickness, the whole forming a somewhat triangular region between the bases of the plumes, which are thus carried outwards. None of these plumes are so well developed as the first pair. While the nerve-area retains a large size, with the two chambers and the fan -like fibres ventrally (PI. VI. fig. 5), the knife now severs the anus, which at this stage lies close behind the massive dorsal hypoderm of the nerve-area. Then the glandular wall of the buccal chamber and the post-oral lamella appear, while the body-cavity on one side of the rectum presents a granular mass (PI. VII. fig. 4, ov), the rudiment of the ovary. The great buccal disk becomes much broader as well as thinner, and is solid, while a central chamber appears in the post-oral lamella. The body-cavity is much better marked in the young forms than in the adults, and just behind the mouth (PI. VII. fig. 5, r, vt, oe) has in section a symmetrical arrangement of the gullet, stomach and intestine, the body-wall showing the basement-layer beneath the hypoderm. The pharjTix, gullet and stomach are rounded or ovoid in transverse section, but the intestine is triangular. The longitudinal muscular fibres appear along the ventral wall immediately behind the mouth, and soon form the marked pattern so characteristic of the region (PI. V. fig. 4, co). The body-cavity thus shows from above downwards the comj)aratively small intestine, now rounded, and the massive glandular wall of the stomach in the middle, both being surrounded by a firm investment which leaves what may be called the keel of the stomach to be attached to a pointed incurvation of the ventral wall. On each side of the latter is the thick central mass of the longi- tudinal muscles, which externally also present another increase before being lost on the body- wall. A considerable perivisceral cavity exists on each side of the digestive organs. Behind the foregoing the body-wall becomes thicker and the central chamber less, especially as the stomach ends in the intestine. The ventral muscles have considerably increased in bulk, and the double inflection for the longitudinal muscles on each side of the median line more marked. At the curvature of the alimentary system in the Ijody- cavity, the pedicle proper commences, the external wall having a proportionally thicker coating of hypoderm than the body-proper, and supported internally by the basement- layer, which forms the W-shaped 2:)attern inferiorly with a secondar}^ curve on each outer leg. The ventral half of the space is filled with the muscular fibres, while the rest is split into two divisions by the median septum continued from the alimentary canal. The sides of the septum and the inner surface of the wall are covered with fibres. The buccal shield extends backwards to this region in the form of a broad, thin lamella, having a, median fine of basement-tissue separating the ventral and dorsal layers of hypoderm, and so little has the pedicle increased in length, that in some cases the free posterior REPORT ON CEPHALODISCUS DODECALOPHUS. 31 margin of the disk almost touches the pedicle of the parent to which the bud is attached. The pigment in the disk now approaches the condition in the adult. The body-cavity of the buds at this stage thus glides insensibly into the first part of the pedicle, but the brevity of the latter organ gives rise to rapid changes in section. The next slice, indeed, shows the muscular fibres filling up the entire central area of the pedicle, while the septum proceeds only a short distance inwards from the dorsal wall. A somewhat radiate arrangement of the fibres also takes place, and is best marked ventrally in transverse section. After a short course the fibres terminate on the hypo- dermic covering of the eijd of the pedicle, which at this stage presents proportionally great development. The immature pedicle of the bud thus differs from that of the adult in the outline of the basement-layer, which is simply reniform in transverse section, whereas that of the adult presents two mamillse on each side ventrally, and a median dorsal fold. Externally, as indicated, the short pedicle almost immediately follows the anterior dorsal projection of the body ; and as a peculiar curvature of the latter has now taken place, the tip of the pedicle scarcely projects beyond the margin of the disk. Shortly after reaching the stage just described, and while the symmetrical series of filaments on each side of the plumes is quite small, the bud separates from the parent. The pedicle and its sucker-like hypodermic termination are fairly developed ; and as soon as it is detached (and sometimes before) a little bud appears near the tip. Thus the increase of the species by budding alone must be considerable, even comparatively young forms giving rise to a succession of buds. One or two buds are most frequently seen on the pedicle of the adult, though occasionally three or more exist. Shortly before obtaining freedom the buds in some cases present a striking resemblance to certain stages in the buds of RhaMopleura, as shown by Professor Allman. Thus in examples in which the first pair of plumes are very long and the succeeding short, while the stalk is in a state of extension, a condition closely approaching Professor Allman's figure 8 ^ is produced. The buds thus difi"er from those in Rhabdopleura in attaining freedom on reaching a certain stage, but both probably arise in a similar way, two at least of the embryonic layers taking part in their production. The first layer is represented by the dermal layers, nervous centre, buccal region, and the rectum, and the second by the skeleto- genous tissue and longitudinal muscles. The presence of the third layer, as already mentioned, is more doubtful, though it is possible that the cells and globules observed towards the end of the pedicle may be of hypoblastic origin, being derived from the central region of the alimentary canal. The sections of the younger buds present, in the arrangement of the alimentary canal, a close resemblance to the young Pedicellina, as shown by the careful researches of Dr. Barrois and Mr. Harmer. The position of the ' Quart. Journ. Micr. Set., 1869, vol. ix., pi. viii. 32 THE VOYAGE OF H.M.S. CHALLENGER. ganglion in the diagram ^ of the latter would nearly correspond with the nerve-centre in Cephalodiscus. In Loxosoma the position of the buds is very different, viz., in the region of the stomach, and there are also often more than two. Homologies. Cephalodiscus approaches Rhahdopleura very nearly in many structural features, and it is probable, when more complete investigation of both is carried out, these resemblances will be increased rather than diminished. Thus the Cocncecium in both is largely developed and wholly independent of the polypides, while it is mainly secreted by the buccal shield or disk. The regularly ringed cylindrical ccencecium of Rhahdopleura is, however, very different in form from the irregular, much branched and hispid ccsncecium of Cephalodiscus. Moreover, in the latter it is the secretion of the adults, whereas in RhabdopAeura much of it would appear to be the product of the younger buds. Phoronis, again, secretes its simple gelatinous investment in the sand, or in the form of tubes attached to stones or other foreign bodies, while an Australian species betakes itself to the gelatinous case of Cerianthus.^ There is thus comparatively little method in the formation of its isolated dwelling. Balanoglossus, on the other hand, has only a mucous lining to its perforation in the sand, though the secretion of this form is also very abundant. Further, Bateson describes a peculiar odour in the living animals, and the spirit-preparations of Cep)halodiscus also give evidence of a characteristic odour, though it may differ from that of the former. The general form of the polypides of Cephalodiscus and Rhahdopleura diverges very considerably, the former being free, while the latter is fixed by the axial stem. Both, however, are small, while the size attained by Phoronis is a distinctive feature, as also is the absence of a pedicle from its cylindrical body. The Buccal Shield is much larger in Cephalodiscus than in Rhahdopleura, and its secreting powers more active. The buccal shield is absent in Phoronis as such, but is represented by the epistome. As will be pointed out by Mr. Harmer, the proboscis of Balanoglossus appears to be the homologue of the disk, though only one proboscis-pore is usually present, while two exist in Cephalodiscus. Further examination is necessary in regard to these organs in Rhdbdop)leura. The Branchial Plumes have a kind of skeletal system or basement-tissue in both Cephalodiscus and Rhahdopleura, but they are much fewer in the latter than in the former, which, moreover, has a bulbous and glandular tip to the main axis, thus simulating such organs as the large eye at the tip of the branchise of Branchiomma. The large ' Quart. Journ. Micr. Sci., October 1886, vol. xxvii., pi. xxii. fig. 19. 2 Vide Dr. Haswell, Proc. Linn. Soc. N. S. JV., vol. vii. p. 607. REPORT ON CEPHALODISCUS DODECALOPHUS. 33 size and firm nature of the simple branchial filaments of Phoronis are sufficiently diagnostic, as also is the fine, double, convoluted arrangement seen in the Philippine and Australian forms. The skeleton and circulatory system of these organs is much more highly developed than in either Cephalodiscus or Rhabdopleura. In Balano- glossus, again, considerable divergence has happened, for the branchise are now arranged in lateral series along the second region of the body, and are supported by an elaborate skeleton of chitinous elements ^ and furnished with numerous gill-slits. Such a modification, however, does not seem very far fetched when a section of the bases of the filaments after entering the axis of the plumes is made in Cephalodiscus. In connection with the arrangement of the plumes it is also interesting that in the Eupolyzoa {e.g., as described by AUman^ in Paludicella, Nitsche^ in Flustra mem- branacea, and Haddon * in Flustra carbasea) the growing tentacles in the bud present bilateral symmetry. The Circulatory System is evidenced only by the lacunse (nuchal or collar-spaces) and their connections with the bases of the plumes in Cephalodiscus, but it would appear to be more largely developed than in Rhabdopleura, for its presence has not yet been indicated in that form. The circulatory system in Phoronis attains a much higher degree of complexity, since its large vascular ramifications with the well-marked nucleated corpuscles have no parallel in either. Cephalodiscus, however, agrees in that its collar- spaces are in connection with the reticulated or lattice-like centre of the main stem in each plume. The circulatory system in Balanoglossus, again, is also largely developed, especially in connection with its branchial system, though the contents of the vessels are less conspicuous. On this head all the foregoing widely diverge from the ordinary Polyzoa. The Digestive Apparatus in both Cep>halodiscus and Rhabdopleura closely agrees with the type in the Polyzoa, all being characterised by the flexure which causes the close proximity of mouth and anus. Moreover, the pyloric difi"erentiation indicated in Cephalodiscus is prevalent in the Polyzoa and also in Phoronis, and though it has not been described in Rhabdopleura, traces of it may yet be found. The environment of the mouth in the latter and Cephalodiscus is related, but while the post-oral lamella is connected with the buccal shield in Rhabdopleura, it forms a special structure in Cephalodiscus. In Phoronis the general plan of the digestive system is the same, though the pyloric region of the stomach attains much greater size. In regard to this system all the foregoing closely approach the Eupolyzoa, the oesophagus, stomach, pyloric vestibule, intestine and rectum showing a similar arrangement. When Balanoglossus, 1 Vide Monograph on the Brit. Nemeiteans, Ray Society, 1872-73, p. 146. * Fresh-water Polyzoa, p. 36. 3 Zeitschr. f. wiss. ZooL, 1871, Bd. xxi. p. 457. * Quart. Journ. Micr. Set., 1883, vol. xxiii. p. 518. (ZOOL. CHALL. EXP. — PART LXII. — 1887.) Qqq 5 34 THE VOYAGE OF H.M.S. CHALLENGER. however, is contrasted with them very considerable differences are encountered, one of the most decided being the straightness of the canal and its terminal anus. The Nervous System is prse-oral in Cephalodiscus and it has a somewhat peculiar structure. It is situated between the hypoderm externally and the basement-tissue internally. None has yet been described in Ehabdojjleura, and thus comparison at present must remain in abeyance. In Phoronis the nervous elements, which lie along the base of the branchial processes, are similar to those in Cephalodiscus, and there is also a central area between mouth and anus. The nervous system is subhypodermic, and is chiefly concentrated in the corresponding region to that in Cephalodiscus. So far as known, therefore, all have something in common under this head. Unless the oviducts in Cephalodiscus are to be credited with remarkable functions, sense-organs, so far as can be noticed in the spirit-preparations, are absent. Further inquiry is necessary on this head, but I am inclined to consider with Dr. Marcus Gunn, on whose special experience and caution I place reliance, that there is no refractive mechanism. This localized pigment perhaps indicates either phosphorescent organs or local heat-producers. In Rhabdopleura a pair of ciliated pads or papillae occur at the bases of the plumes, the minute structure of which, however, is in need of careful revision. Phoronis, again, presents only the ciliated furrows, which are external to the anus, and have a nervous expansion beneath the hypoderm. In Loxosoma papillae similar to those just described occur on the dorsal aspect, and the suboesophageal ganglion is well developed, as shown in Harmer's beautiful researches on Loxosoma crassicauda. The structure of the Body-ivall in Cep)halodiscus considerably diverges from that in Rhabdopleura, the definite layers of hypoderm and basement-tissue in this form contrasting with the cuticular epithelium and underlying connective-tissue cells in the latter. Something like basement-tissue seems to be indicated in Lankester's pi. xl. fig. 12 (though no mention of it is made), and he describes and figures the enteric epithelium lining the body-cavity, the cells being connected with the wall of the stomach by processes. The body-waU in Phoronis again deviates from that in either of the foregoing, since, besides cuticle, hypoderm and basement-tissue, it shows a circular and a longitudinal muscular coat, the latter being chiefly grouped in longitudinal bands which in transverse section show a somewhat pennate arrangement. There is little in common, therefore, beneath the basement-tissue, and the absence of the pedicle in Phoronis is a marked feature of divergence. The structure of the body-wall of Cephalodiscus most nearly approaches that of Balanoglossus (a phosphorescent form), though there is a wide gap in this respect as there also is between it and Phoronis. In Loxosoma a transparent cuticle and a hypodermic layer, " associated at certain points with muscular fibres more or less pronounced," according to Vogt are present. Cephalodiscus and Rhabdopleura agree in certain respects in regard to the Body- cavity, but the former has the pedicle as an appendix. Phoronis on the other hand has REPORT ON CEPHALODISCUS DODECALOPHTJS. 35 a much more spacious chamber divided by various mesenteries. In Balanoglossus this chamber posteriorly is inconspicuous in the adult. The condition again in the Ento- proctous Polyzoa {e.g., Loxosoma) considerably diverges, for no body-cavity exists; while in the Phylactolsemata it is present in the adult, and is lined by ciliated epithelium, such not being the case in the body-cavity of the Gymnoltemata. It is an interesting fact that the Muscular System both in Cephalodiscus and Rhahdo- 'pleura is connected with the pedicle, if we may for the moment so term the soft contractile stalk of the latter. In the former, however, it is much more largely developed and is continued directly from the body-cavity ; whereas in Rhabdo2ileura it is wholly external to that chamber, and is less distinctly differentiated on the surface of the axial skeleton, which forms another feature of distinction in this form. If the funiculus of one of the Eupolyzoa be disconnected from the digestive system and formed into an external process in the line of the ordinary communication-plate, something similar in structure and function to the pedicle in the Aspidophora will be made. As a rule in the same group the retractor muscles of the body and lophophore arise from the peritoneal lining. The funiculus of the Eupolyzoa, according to Haddon, is probably derived from the irregular strands of funicular tissue which occur in the parent zooecium. It appears as a thickish cord stretching from the fundus of the developing polypide 'to the base of the zooecium. It is in direct communication with the brown body, directing " the developing alimentary tract to that nutritive mass, thereby ensuring the better nutrition of the growing bud." The bud is thus developed at a distance from the brown body, but approaches it and extracts nutriment from it. This has been noted by other authors. In Loxosoma the stem quite differs, since there is no communication with the body- cavity. The present condition of our knowledge of the Reproductive Organs, both in Cephalo- discus and Rhahdopleura, is incomplete, so that a satisfactory comparison cannot be made. The ova in the former are very large, but no male elements have been seen. In Rhahdopleura the testis occurs as a long sac adjoining the intestine and even projecting beyond the abdomen. It opens near the anus, and thus agrees with the condition in Cephalodiscus and Phoronis, as well as offers certain resemblances to the condition in the Entoproctous Polyzoa. The reproductive organs in Phoronis are posterior in position, and both male and female elements are usually conspicuous. In Balanoglossus these elements occur between the liver and the anterior part of the body. The early appearance of the ova in the young buds of the Eupolyzoa, for instance Bugida Jlabellata, as noticed by Haddon, is worthy of mention. These ova are in close relation to the wall of the digestive tract. The distinction of the Aspidophora from those of the Eupolyzoa in which the ova of the parent pass ready-formed into buds is marked. 36 THE VOYAGE OE H.M.S. CHALLENGER. Budding. — The fact that Cephalodiscus is free while Iihahdo2)leura is fixed causes considerable divergence in regard to the buds ; and, moreover, the functions performed by the bud in the latter species, while yet incompletely developed, and with a bifid buccal shield (viz., the secretion of the ccenoecium or tubarium) is an important difference. Another essential divergence is the occurrence of the buds in a regular series on Rhahdo- •pleuva, the youngest nearest the terminal polypide, the oldest next the distal. The confinement of the buds in Cephcdodiscus to the region just within the terminal hypodermic plate is peculiar, and makes it difficult to institute anything like parallelism between them in this respect. Further, Laukester is inclined to think that after the complete development of the i^olypide in Rhahdopleura, there is no evidence that it takes upon itself bud-production ; that is to say, the buds are given ofi" at an early period of its growth. It is not quite clear, however, that the budding of this form is in the same category as that of Cephalodiscus, in which the stalk is a process of the body-cavity, whereas the soft stalk of Rliahdo2^leura, if the descriptions are understood correctly, has not yet been shown to be so, though at first sight it might be interpreted otherwise. Nothing like the arrested buds of this form is known in Cejyhcdodiscus. The source of the hypoblastic elements, if these are present, in the bud of Rhahdopleura is thus in obscurity. In PJioronis no bud is known, while the small ova are extremely numerous, and the embryos (having the form of the well known Actinotrocha) pelagic. In Balanoglossus likewise no bud occurs, the ova are numerous and small, and the embryo free-swimming (Tornaria). On taking a general survey of the subject, then, it occurs to me that in the present state of our knowledge, and while fully admitting the remarkable resemblances between it and certain hitherto isolated types such as Balanoglossus (which I have for the most part left in the able hands of Mr. Harmer), it will lead to no disadvantage if Ceplialo- discus be left as formerly near the Polyzoa ; and, further, though the divergences between it and Rkabdojileura are noteworthy, in the same group as formerly, viz., the Aspido- phora of Professor AUman. It is well to exhaust the structural, developmental, and other features in the various forms reviewed in the preceding paragraphs before changes in classification are promulgated. Cephalodiscus approaches the Polyzoa in regard to its ccenoecium, its digestive system, and its buds, and it is peculiar that in these points there is a lack of conformity in Balanoglossus, and to some extent in Phoronis. Viewed as a whole, the several systems mentioned agree most with the type of the Polyzoa. Though Phoronis forms a tube, and Balanoglossus secretes very abundant mucus, a feature common to many diverse groups, such as the Nemerteans, Discophora, and MoUusca, nothing like the regular ccenoecium of Rhahdojjleura or Cephalodiscus is constructed. While again the digestive system of Phoronis resembles that in the Polyzoa, the same system in Balano- glossus is very difi'erent, for the straight alimentary canal with its terminal anus has no REPORT ON CEPHALODISCUS DODECALOPHUS. 37 parallel in the life-history of Cephalodiscus, not even an embryonic approach occurring in either form. The anus in Balanoglossus is posterior and terminal at all stages. The relations of the dorsal and ventral surfaces in the two forms are also at variance. Buds again, are unknown in either Phoronis or Balanoglossus, and in both the eggs are very numerous and small, whereas in Cephalodiscus they are few and very large. The main resemblances between Cephalodiscus and Bcdanoglossus lie in the structure of the skin, the presence of three body-cavities (disk, collar, and body proper), the proboscis or disk-pores, collar-pores, gill-slits, and rudimentary notochordal structure, and they are of a most interesting and suggestive character ; and as ]\Ir. Harmer, whose valuable researches on the Entoproctous Polyzoa are well known, has most ably studied these features and formed independent conclusions, I have thought it best to give his views in his own words as an Appendix. These will show how difficult it is in some cases to draw clear lines of distinction — so intimately are the several characters, in apparently diverse groups, blended. In a former paper ^ I had observed with regard to Phoronis and Balanoglossus — " If indeed the branchial skeleton supporting the vessels (of Phoronis) were thrown in, and arranged at the sides of the anterior region of the body, so that the water would enter by lateral slits to aerate the circulating fluid, and the digestive canal enlarged and attached as a single tube to the body-wall, a form resembling Balanoglossus would be indicated." It has to be borne in mind also that Alexander Agassiz thought that the latter resembled the Tunicates from the nature of the gills and their mode of formation, in opposition to the views of Kowalevsky and others, who placed its affinities with the Annelids proper. Perhaps Balanoglossus may at present be ranged near the Aspidophorous group of the Polyzoa, for though Metschnikoff's view that it approaches the Echinoderms rests on the remarkable fact that in Tornaria the original evagination from the gut is on the left side, just as in Asteroid larvae the water-vessel is developed from the left primitive diverticulum (Bateson), yet there are stronger reasons for associating it with other groups as above mentioned. 1 Proc. Boy. Soc. Edin., Session 1880-81, vol. xi. p. 217. APPENDIX. By Sidney F. Harmer, B.A., B.Sc, Fellow of King's College, Cambridge, and of University College, London. The following observations on Cephalodiscus were made in consequence of a letter received from Professor M'Intosh, calling my attention to certain remarkable features in tlie anatomy of the genus. I am very greatly indebted to Professor M'Intosh for his courtesy in giving me specimens of Ce2)halodiscus, and most of all for his kmdly expressed destre that I should publish my conclusions as an appendix to his own mono- graph. Time has not permitted of my seeing the proofs of this monograph, and I must therefore claim indulgence for any descriptions which would otherwise appear unnecessary repetitions of the results of Professor M'Intosh himself. In examining sections of Cephalodiscus, I have been struck with the existence of various organs which appear to me to point to the conclusion that this remarkable genus is a near ally of Balanoglossus. This very unexpected result will be understood by comparing the following woodcuts (with the remarks which accompany them) with Bateson's papers on the anatomy and development of Balanoglossus} Fio-. 1 represents a longitudinal, right and left, section of a young bud of Cephalodiscus. The re- semblance between this section and Bateson's diagrams of the larvae of Balanoglossus^ is, in all essential details, exact. The body of the young Cephalodiscus is divided, by means of two transverse grooves, into three well- marked regions. Of these the anterior (p.) may be compared to the proboscis of Balanoglossus; the middle division (c.) to the collar of the same animal, and the posterior division (tr.) to the trunk or body. It will further be noted that the proboscis is pro- 1 Qmrt. Journ. Micr. Sci., vols, xxiv., xxv., xxvi. 2 Quart. Journ. Micr. Sci., vol. .\xvi., pi. xxxiii. fig. 7. Fig. 1.— p., r™l"'scis ; c, collar; tr., trunk; ai., alimentary canal ; nch., notochord ; 6.c.i, body-cavity of proboscis ; b.c", body-cavity of collar ; b.c.^, body-cavity of trunk. 40 THE VOYAGE OF H.M.S. CHALLENGER. Tided with an anterior, undivided body-cavity, the collar and the trunk each containing a body-cavity which is composed of two halves separated by means of dorsal and ventral mesenteries. These five spaces are completely separated from one another, and their arrangement is in exact accordance with that of the corresponding sections of the TT>- UL Fig. 2. — Longitudinal section of an adult Cephalodiscus, supposed to be taken sufficiently on one side of the middle line to allow of the representation of one of the ovaries, and of one of the prohoscis-pores. j)., proboscis ; stk., stalli ; »«., mouth ; ph., pharynx or branchial region of gut ; ces., oesophagus ; st, stomach ; int., intestine ; a., anus ; nch., cotochord (really visible only in a median section) ; m., ovary ; Vitk Proc. Eoy. Soc, vol. xlii. p. 146, 1887. 46 THE VOYAGE OF H.M.S. CHALLENGER. manner, and that the alimentary canal has acquired its adult relations by its partial passage into the ventral protuberance of the body. Whilst it appears to me easy to compare Cephalodiscus and Balanoglossus, it does not seem to me impossible that the former may have affinities in other directions as well. For, imagine that the ventral elongation of the body would be conveniently post- poned until after the end of a free larval life ; the stalk might then be invaginated into the body of the larva in preparation for its evagination when metamorphosis should take place, as actually occurs in Actinotrocha. After the metamorphosis of the latter, the alimentary canal has the same dorsal flexure as in Cephalodiscus, and this explanation of the metamorphosis of Phoronis is in accordance with the suggestions of previous observers. The following considerations may perhaps indicate some affinity between Cephalodiscus and Phoro7iis ^ : — 1. The archenteron of Phoronis is developed by a well-marked invagination, whilst part of the mesoblast {vide Caldwell) is formed by a process of (modified) archenteric pouching (as in Balanoglossus). 2. The prae-oral lobe is large in Actinotrocha, and is provided with a body-cavity which is completely shut off, by means of a septum, from the body-cavity of the trunk. The post-oral region is prolonged into tentacles, which, although differing in a striking manner from the tentacles of Cej^hcdodiscus, may still have some connection with these structures, or with the operculum of the same genus. 3. The "foot" of Phoronis has precisely the same relations as the stalk of Cephalo- discus. 4. The nervous system of Phoronis occurs outside the basement-membrane. The ganglion of the prse-oral lobe of Actinotrocha is comparable with a portion of the nervous system of the Hemichordata, whilst the post-oral nerve-ring of Actinotrocha (following the line of the bases of the tentacles) may not impossibly be the homologue of the nerve- ring which passes round the posterior border of the collar in Balanoglossus. If this were the case, the lophophores of Actinotrocha and of Phoronis might be regarded as develop- ments of the collar-region. 5. Phoronis possesses a complete ventral mesentery, the dorsal mesentery, however (persistent in Cephalodiscus), having disappeared in the adult animal. The ovaries and oviducts of Cep>halodiscus are supported by lateral mesenteries which are apparently arranged in the same manner as the lateral mesenteries in Caldwell's diagram " B." ^ The oviducts of Cephalodiscus do not, however, open into the body-cavity, and it is possible that the collar-pores, rather than the oviducts, may be the homologues of the nephridia of Phoronis. ' Cf. W. H. Caldwell, Prelim. Note on the Structure, Development, and Affinities of Phoronis, Proc. Roy. Soc, No. 222, 1882 ; and Blastopore, Mesoderm, and Metameric Segmentation, Quart. Journ. Micr. Sci., vol. xxv., 1885. 2 Proc. Roy. Soc, 1882. REPORT ON CEPHALODISCTJS DODECALOPHUS. 47 6. Previous observers (M'Intosh, Lankester, &c.) have been led to assume the affinity of Phoronis to Cephalodiscus and Rhabdopleura, this conclusion being based on such features as the relations of the adult lophophore to the mouth and anus. It must be noted, on the contrary, that Phoronis is not known to possess any repre- sentatives of the notochord, gill-slits, collar-pores, and proboscis-pores of Cephalodiscus, whilst there is no evidence of the existence of a collar body-cavity in the former. It appears to me that a renewed consideration of Phoronis, anatomically and development- ally, can alone settle the question of the possibility of an affinity between it and Cephalo- discus. The remarkable larva of Balanoglossus described by Weldon {loc. cit, fig. 3) is in some of its features by no means unlike Actinotrocha. Such features are the general form of the prse-oral lobe and trunk, the absence of the notochord and gill-slits, and the existence of only three divisions of the body-cavity. These are (l) the unpaired cavity of the prse-oral lobe, and (2) the two cavities of the trunk-region. In the absence of these cavities and of the notochord and gill-slits Actinotrocha differs from the larval Balanoglossus described by Bateson. It cannot, however, be denied that the dif- ference between the tentacles of Weldon's larva and those of Actinotrocha is very considerable, if not fundamental. The relation between Cephalodiscus and Rhabdopleura is in need of further elucida- tion. In spite of the great resemblance between the lophophores and epistomes of the two genera, many of the most important structures found in Cei^halodiscus are not known to exist in Rhahdopleura, and there does not at present appear sufficient justification f«r the removal of Rhahdojjleura to the Hemichordata, although the balance of evidence might perhaps be in favour of so doing. I do not think that the above considerations are in any way calculated to strengthen the view that Phoronis and the Polyzoa are nearly related. The result of the examina- tion of Cephalodisctis appears to me to show that this genus (and Rhabdopleura also ?) must be entirely removed from the Polyzoa. If this is the case, it is obvious that any affinity which may be shown to exist between Cei^halodiscus and Phoronis can in no way affect the question of the relationship of the latter to the Polyzoa. CONTENTS. Intkoddction, . Description, . Ccencecium, Polypides, Buccal Disk, Branchial Plumes, Post-oral Lamella, Digestive System, Body-wall, Muscular System and Pedicle, Nervous System, Sense-Organs, . Eeproductive Organs, . Budding, Homologies, . Appendix, by Sidney F. Harmer, B.A., B.Sc, PAGE 1 3 3 6 8 10 14 15 19 20 23 24 25 28 32 39 PLATE [. (zooL. ciikhh. EXP.— PART Lxii.— 1887.;— Qqq. PLATE I. Cceuoecium of Cephalodiscus dodecalophas, about the natural size ; a few of the liraiiches terminate in somewhat flattened tips. At ((, one of the pillars, which like aerial roots pass downwards to the stones and sponges on which it grows, is represented, its tip being expanded into a broad surface. Numerous inosculations take place between the various branches. I have to th;iuk Mr. E. E. Prince for most kindly aiding nic in drawing certain figures, as indicated on the Plates. The Voyage of H.M.S. "Challenger iilwVi W>1 w}. \! M. \i\ 'or di;' 'X Cephalodiscus. Pl.I. f/ <^ A^'f.. >A v.j X /Nil I ^ y ■■;^ W C.M i E Prince Se! FHulh.Litli'Eilra' CdNECIUM OF C E PH A LO D I SC U S D 0 D E C A L 0 P H U S, M ■= (uncoloured ) PLATE II. PLATE II. Poly|3ide of Cephalodiscus dodecalop/ms, seen from the ventral surface, and enlarged under a lens, the actual length of the entire animal being about 2 mm. Only a few of the tentacular plumes are visible, the rest being concealed by the dense mass of filaments. The tip of the pedicle bears an early bud, the pedicle of which is somewhat elongated. This example presented numerous minute pigment- spots of a brownish hue all over its surface. The Voyage of H.M.S"Ckllen^er." Cephalodiscus.Pi.n. F.HutKLith'Edin' C E P H A LO D 1 5C U S D 0 D E C A L 0 P H U S , M' PLATE III. (ZOOL. CHALL. EXP. — PART LXII. 1S87.) — Qqq. PLATE III. Fio-. 1. Polypide of Cephalodiscus dodecaloj)hus, obliquely seen from the ventral surface, much enlarged, and having the buccal shield removed. The pedicle is in a state of extension, with a fairly advanced bud at the tip; an, anus; p?., post-oral lamella ; m, mouth. Fig. 2. Lateral view of a polypide, also deprived of the buccal shield, and much enlarged. The pedicle is contracted and bent forward, with a pair of buds at the tip. The same letters are used as in fig. 1 ; ov, the position of the large ovum, which distends the lateral region just behind the pigment-spots (od); hp, torn tissues at the base of the plumes, which have been removed. Fig. 3. Lateral view of an entire specimen, with the pedicle as well as the buccal disk thrust outwards, or ventrally. The buccal shield has fallen from the larger of the two buds at the tip of the pedicle. The alimentary canal has been somewhat diagrammatically outlined in this example ; ce, above the line is the pharynx, below is the CBSophagus ; vt, stomach, largely distended ; vtd, peculiar region following the stomach, and presenting a thick glandular coat with a distinct fold beyond the point touched by the dotted hne, and which probably is homologous with the second stomachal dilatation in Phoroms ; r, rectum, usually much dilated ; hs, buccal shield ; p/., post-oral lamella. Fig. 4. An abnormal specimen, similarly magnified and viewed laterally. The pedicle is absent, a prominence only {ped) indicating its position. Instead of the smoothly bulbous posterior extremity (almost like the bowl of a retort), a somewhat sharp angle occurs ventrally. The anus is elevated on a rectal cone. The Voyage of H.M.S "Challenger" ^ # \ W. CM del /^ / Cephalodiscus. PI 111. >>, bp pi- J ped. FHufh.Lith'Eam' C E PH ALO DISCU S D 0 D E C A L 0 P H U S , M = PLATE IV. PLATE IV. Fig. 1. Tentacular plume of Cephalodiscus, seen as a semitranslucent object. The line ill tlie middle of the filaments indicates the axis or skeleton ; x 90. Fig. 2. Portions of the pinnse or filaments acted on by dilute potash so as to exhiliit the central axis or skeleton, which at a has been exposed. The cellulo- granular coating of the organs is hypodermic ; x 350. Fig. 3. Slightly oblique section of the bullions tip of one of the plumes, showing the large globules and gland-cells. The hypoderm has intruded on the left, showing that the slice has been made close to the base of the process ; x 470. Fig. 4. Transverse section of the tip of a plume just behind the foregoing, and while the central lumen remains very distinct. The hypoderm shows somewhat regular wedge-shaped divisions, as if composed of a single layer of large cells. The wall of the central cavity presents numerous fibres projecting all round, as if the reticulations were already commencing ; x 470. Fig. 5. Transverse section of the pedicle, showing the remarkable form usually assumed by the elastic basement-tissue (bt) in contraction. The hypoderm (hp) occurs externally ; and within the basement-tissue the large muscular fibres and somewhat gelatinous connective-tissue fill up the central region. The small dorsal and the large ventral incurvations are readily recognised, as well as the two lateral ventral projections of the basement-tissue. The hypoderm on the ventral surface has been somewhat stretched ; x 350. The Voyage of H.M S"Cha!lenger.' Ceplialodiscus. PI. IV. W C M del Fij S.EPrmce- FHuth.LitKEdin' CEPH ALODISCU S D 0 D E C A LO P H U S , IV1'= I. PLATE V. (ZOOL. CHALL. EXP. PART LXII. — 1887.) — Qqq. PLATE V. Fio-. 1. Tip of oue of the plumes of the polypide of Cephalodiscus, showing the glandular uaturc of the enlarged region. The filaments (/) have been turned to the left ; ca, central axis of main stem ; hp, hypodcrm. The longitudinal fibres running alonsr the axis are observed to the left as well as over the axis ; x 210. Fig. 2. Portion of the middle of a plume with the bases of the filaments {/), viewed laterally, after the action of a dilute solution of potash. The skeletal axis is observed in the centre of each filament, and as this skeletal axis {sk) widens out at its base, the double outline at each side is well marked ; ca, central axis of main stem, with its longitudinal fibres ; x 350. Fig. 3. Transverse section of a young bud on the appearance of the first two lophophoral processes or plumes (Zp) as two rounded bosses composed of tissue resembling hypoderm. The buccal disk [hd) is cut at its anterior region, but the central space {he) of the organ is well developed. The collar space {yc) is com- paratively large at this stage, and lies close beneath the nerve-centre ; x 350. Fig. 4. Transverse section of the terminal region of the body-cavity of an older bud than the foregoing, the tip of the alimentary canal {al) being left as a thin plate in the centre, and bound dorsally and ventrally l^y the median mesentery (wis). The great longitudinal muscle is cut near the commence- ment of the pedicle, and already shows the double ventral curvatures so characteristic of the latter ; ht, basement-tissue ; co, the body-cavity ; x 350. Fig. 5. Ovum on its escape from the adult ; ec, egg-capsule ; ov, ovum proper ; st, stalk; x90. Fig. 6. Outline of another ovum in which the yolk is ovoid, instead of circular as in the former case ; x 90. The Voyage of H.M. S/'Chailen^ef.' Cephalodiscus. Pl.V. \ f. ■ hp td. 3. &!?!^''?®Si^ .-yagsx-iw'^jassasi, be -1? '"%;- .s^' ov IP al Fijf 1,2,5,6, WC M. Fi^= 3,4,E Prmce del tt FHulh.Lith-Eain CEPHALODISCUS D 0 D EC ALOPH U S , M= I. PLATE VI. PLATE VI. Fig. 1. Longitudinal section of the tip of the pedicle of an adult Cephalodiscus, showing the attachment of a young bud just above the terminal surface (which is somewhat sucker-like) ; Im, strands of longitudinal muscles ; hps, modified hj^poderm forming the terminal surface of the j^edicle ; gm, section of a young bud, showing the central chamber (ch) continuous with the axis of the pedicle of the adult ; x 850. Fig. 2. Transverse section of an adult Cephalodiscris through the region of the mouth (m) and ovaries, one pigment-mass (od), however, only having been cut. The buccal shield (bs) on the right is seen folding over dorsally {hs}:)), so as to I'un into the tissues at the bases of the plumes (hi-) ; g, grooves at the bases of tlie lophophoral arms. The notochordal region lies between this letter and m, though the notochord is not seen in this section. The mouth is observed at (m) and the pharyngeal region with its numerous folds and a gill slit (gs), while on the left is a section of the j)ost-oral lamella (j^l) with its central cavity or chamber ; 7', greatly distended rectum ; od, pigment of oviduct ; oi\ large, and ov', small ova, with nuclei and nucleoli ; x 90. Fig. 3. Transverse section of the shield and dorsal region in the long axis of the central region of the nervous system (wc). The thick coating of hypoderm (hp) bounds the latter externally, while internally it rests on the basement-tissue (ht.); hp, sections apparently of the shield-pores in their progress inwards ; hsm, radial muscles of the buccal disk or shield; hs, hypoderm of shield; od, position of the pigment of the oviducts ; x 210. Fig. 4. Transverse section of the same bud as in PI. V. fig. 4, in the anterior region, so as to strike the commencement of the nerve-centre («c), with the central cavity (c). The young plumes are irregularly cut from their variable position, and the sides of the disk fold over into the basal region of the plumes. The large size of the terminal process of the plumes is well seen in the larger organ ; x 210. Fig. 5. Section a little behind the foregoing, giving the nerve-centre {nc) in full develop- ment, under the thick hypodermic coat Qip), while the tip of the anus [an) has also been included ; cva, collar space going to plumes ; hsm, radial muscles of buccal disk ; c, region of the notochordal. The plumes are more widely separated in this region, and the young filaments have the form of hypodermic papillse ; X 210. Fig. 6. A somewhat oblicpie section through the buccal shield {hs) and the post-oral lamella {pT) with its central cavity, to which the dotted line goes. Beneath is a collar pore (cp), which is, however, indifferently shown in this section. The nerve-centre {nc) and a pore {hp>) leading into the cavity of the shield are also cut, the two 1)ody-cavities of the region lying above the former {nc) in this view, that is ventrally in nature. The other parts are an ovary {ov) and a portion of the pigment-mass of an oviduct {od) ; x 90. TheVoya Ghanenrfer" Ic- hp. si' hps.- bp. =*y'v •^ bs^ ^^^PS, .^1'""'>, 'v^ 4 ov; 1 "br. hp. Cephalodiscus.Pl.Vl. j-^crcwcjyc;^. br ""^^bsm-.- hp bp nc. bsp hp %^.'5| bs br^ hs. bi hs W. C M & E Prince del T.Hurti.lith'EdW CEPHALODISCUS DODECALOPHUS, W"=l. PLATE VII. (ZOOL. CUALL. EXP.— PART LXII.— 1887.)— Qqq. PLATE VII. Fig. 1. Portion of the coencecium of Cephalodiscus, about natural size, tinted as in .the spirit-preparations. Fig. 2. Transverse section of a pedicle of an unstained specimen, showing a somewhat different arrangement of the folds of the basement-tissue and the central muscular fibres ; letters as before ; x 350. Fig. 3. Oblique section of the chief area of the nerve-tissue (hc), which lies under the thick hypoderm (lip) between the plumes {hr). The coUar spaces {cv) lie next the basement-tissue, bounding the nerve-centre internally, and a trace apparently of the notochord is observed at c. Some of the folds {mf) of the mouth have been included in the section, and the radiate muscles [hsni) of the buccal shield are observed on the left ; x 350. Fig. 4. Transverse section of the same bud as in PI. VI. fig. 5, somewhat behind the former. The body-cavity {co) now shows the rudimentary generative organs {ov), the rectum [r), and the oesophagus {ce) in section. The collar spaces {cv) are distinct, and the post-oral lamella {pi) is seen on the right. The position of the buccal shield is indicated by the letters hs; x 210. Fig. 5. Transverse section of the same bud posterior to the former, and the three regions of the alimentary canal now occupy the body-cavity, viz., r, the intestine ; vt, the stomach, while what is probably the pharynx {ce) lies ventrally. A collar pore is observed at cp; pi, post-oral lamella ; other letters as before. The Voyage of H M SXhailen^er" Cephalodiscus.Pl.VIL \i -■ hp U- i^. 1^-4 w » hp /^^ 1\ 'fe^. iP« ft 4. ' A,,.- CO - T - CO - ' . ,,^^w^ pi .<^/->, _ -:i OS ■-•'- •br. -' c-n --■ - bs. \^''' .^; . cv cv. bs. W, CM JE Prince del F Huth, Lith' HJra' C E PH A LO D I S C U S D 0 D E C A LO P H U S , M< -^ ;< ' ')i} '■-' 'i,.- ■ . :r, '*<. fit -<.r I . ' > k^fm'''--'- 1 I.' «>, •' . . 5,.=*^i^^>V-..-: . ■