y: OF THE UNIVER' . OF FORMS OF ANIMAL LIFE G. ROLLESTON AND W. H. JACKSON JTonboit HENRY FROWDE OXFORD UNIVERSITY PRESS WAREHOUSE AMEN CORNER, E.G. FORMS OF ANIMAL LIFE A MANUAL OF COMPARATIVE ANATOMY WITH DESCRIPTIONS OF SELECTED TYPES BY THE LATE GEORGE ROLLESTON, D.M., RR.S. it LINACRE PROFESSOR OF ANATOMY AND PHYSIOLOGY IN THE UNIVERSITY OF OXFORD SECOND EDITION REVISED AND ENLARGED BY W. HATCHETT JACKSON, M.A. NEW COLLEGE F.L.S. ; NATURAL SCIENCE LECTURER, ST. JOHN'S COLLEGE OF THE UNIVEKS1TY AT THE CLARENDON PRESS M.DCCC.LXXXVIII [All rights reserved} Ai6 8ei /XT) 8vo-)(epaii>€ii> 7rai8iK<3s rrjv Traort yap rots $vcriKOts e^ecrrt ri Oav^CKTTov ..... KCU Trpos T?)Z; cS^ ^Iwa)^ Trpoo-teVat 8et JUT) 8txra)7rov//,ez;oz/ ws ev airaorLV OVTOS KaXoi).— ARISTOTLE, De Part. Anim. i. 5. Trpocrfleu'ai ro eXXetTToi^. — ARISTOTLE, Ethic. Nicom. i. 7. MUSEUM, OXFORD: March 5, 1870. PREFACE TO SECOND EDITION. THE present edition of ' Forms of Animal Life ' was taken in hand by the late Professor Rolleston in the Long Vacation of 1879. The work was carried on with prolonged interruptions, incident to a life of many and varied engagements, until he left England in December 1880. By that time he had completed the descriptions of Preparations 1-9 ; three new Plates (Pis. IV, VII, IX) had been engraved under his direction, and he had compiled notes upon them, which have been employed in the descrip- tions printed in this volume. Soon after beginning his work, the Professor asked me to undertake a joint authorship of the book. The part then assigned to me was to rewrite the descriptions of a certain number of the Preparations, the general accounts of Urochorda, Arthropoda, parasitic Vermes, Coelenterata and Protozoa, as well as of several minor classes. He read my account of Protozoa, and settled that it Should form a model for the accounts of all other groups. As now printed it has been so far modified as to accord with the recent progress of knowledge. When Professor Rolleston went abroad he put me in possession of his plans for the rest of the work, handed his papers to me, and expressed a hope that, if he were disabled from completing the new edition, I might be the person to do it in his stead. It is almost needless for me to add that in fulfilling this sacred trust I have endeavoured to carry out his wishes, which were mainly three : (i) to enlarge the descriptions of the Preparations and accounts of the various classes of animals, and bring them to the standard of contemporary knowledge ; (2) to add to each class or group a brief classification ; and (3) to give as full a bibliography as space would permit. The method I have adopted to meet the last requisition is to cite the most important and recent authorities which, when consulted, will in most cases give the names of all other accounts worth reading, so as to form vi PREFACE TO SECOND EDITION. a really very complete index to the state of present knowledge. These authorities write chiefly in foreign languages, and I need scarcely remark that every modern anatomist must also be a modern linguist. Literature, though not everything in Science, is yet indispensable, and as Professor Rolleston observed in the preface to the first edition of his book (pp. viii- ix):- ' In some cases even the beginner will find it necessary to consult some of the many works referred to in the descriptions of the Preparations and in the descrip- tions of the Plates ; but the bibliographical references have been added with a view rather to the wants indicated in the words " Fur Akademische Vorlesungen und zum Selbststudium," so often prefixed to German works on Science, than to those of the commencing student.' The debateable points of Phylogeny are not treated at any length. Professor Rolleston, having tried one experiment, particularly desired that they should be omitted on account of the great space their adequate dis- cussion must needs occupy. One alteration in the arrangement of the volume had been con- templated by the Professor, but left unsettled, and has now been carried out in consequence of the opinion of Professor Huxley in its favour. The descriptions of the Preparations, those of the Plates, and the general account of the Animal Kingdom, have changed places. The two former stood last in the first edition, but take precedence in this. The new arrangement tallies better with the order in which Professor Rolleston wished the several parts to be studied, as stated, loc. cit. pp. vii-viii : — * It is recommended that in all cases the study of the described Preparation or specimen should precede that of the accounts in the Introduction (i. e. General account of the Animal Kingdom, Ed.} of the Class and Sub-kingdom (i.e. Phylum, Ed.} to which it belongs, and that the study of the Plates should be taken up only after the attainment of a considerable familiarity with actual specimens by the practice of dissection.' The Plates, however, illustrate the Preparations, and are therefore placed as the second section of the book 1. 1 A few changes in the Preparations have been made. Some have been added, e. g. those relating to the Rabbit ; the Privet Hawk Moth has been substituted for the Death's Head which is difficult to procure ; similarly, the Dog's Tapeworm and its Cysticercus replace the Bladderworm of the Sheep (Coenurus}. The skeleton of the Common Fowl, two dissections of a Caterpillar, the angular Sea Cucumber, and the Bugle Coralline, have been omitted. The Preparations were made by Charles Robertson, Esq., Demonstrator of Anatomy in the University Museum ; the greater part PREFACE TO SECOND EDITION. v[( Although the Professor contemplated the above-mentioned changes, he desired to retain the ' distinctive character 5 of the book. This character, as he himself said, loc. cit. pp. v-vi : — ' Consists in its attempting so to combine the concrete facts of Zootomy with the outlines of systematic Classification as to enable the student to put them for himself into their natural relations of foundation and superstructure. The founda- tion may be made wider, and the superstructure may have its outlines not only filled up, but even considerably altered by subsequent and more extensive labours ; but the mutual relations of the one as foundation, and of the other as superstructure, which this book particularly aims at illustrating, must always remain the same.' Another observation may be quoted, loc. cit. p. vi : — * It is hoped that this work, though written with a view chiefly to the needs of University students of Comparative Anatomy, and with special reference to the application of that branch of science as an engine of instruction, may in some measure meet the requirements of the now not inconsiderable number of persons who are attracted to the study by seeing the important bearings which it has upon questions not only of theoretical and philosophical, but also of practical interest.' It would have been more agreeable to my own feelings if this second edition had been issued at an earlier date. But the great length of time which has elapsed since the publication of the first — full seventeen years — has brought with it so many and such vast changes in Comparative Anatomy that great labour and consequent delay became inevitable. I may mention that scientific periodicals on the general subject and its branches have since 1870 been almost doubled, not only in number, but also in bulk ; and one whole science — that of Comparative Embryology — has been formulated and now constitutes the foundation of all Anatomy. Any worker placed single-handed under such conditions, is at a great dis- advantage even with all the modern paraphernalia of abstracts. And I have to add a lesson learnt by personal experience, that in most cases the best abstract available cannot by any means stand in the place of the original paper. Among other causes of delay my own employment as a teacher must be taken into account, and there have also occurred of them were exhibited by him as a ' Zoological series with Dissections in illustration,' in the Educational Department of the Great Exhibition of 1862. Thirteen woodcuts have been added in the text, and three new Plates (Pis. IV, VII, IX). Of the old plates, one (PI. IX of the first edition) has been cancelled. The woodcuts 1-5 and the three new plates have been drawn by Julian Drummond, Esq., the present Radcliffe Artist ; the plates of the first edition were drawn by his predecessor, George Crozier, Esq. viii PREFACE TO SECOND EDITION. many unforeseen interruptions. Under these circumstances it is my pleasant duty to thank the Delegates of the University Press for their kind forbearance on the question of time. My warmest thanks are due to Sir H. W. Acland, K.C.B., F.R.S., for the use of a room belonging to his own suite in the University Museum, and for a free and extensive command of all requisite literature in that rich storehouse of scientific books, the Radcliffe Library, without which my task could never have been accomplished. To Professor Moseley, F.R.S., Dr. Rolleston's successor in the Linacre chair, I render my best thanks for an unlimited employment of the anatomical collections under his charge, and the loan of his own MS. notes on the Anthozoa Zoantharia. I have also to record a debt of gratitude to my early and constant friend, Professor Westwood, who has given me much assistance by way of access to specimens and pamphlets on the difficult phylum Arthropoda. In kind compliance with my request, Professor Ray Lankester, F.R.S., liberally allowed me to copy two figures of his own construction (Woodcut 13, A, B, and T, 2, 3, 4) ; and Professor Kitchin Parker, F.R.S., furnished valuable information, at the time unpublished, relative to points in the development of the Vertebrate skull, and also granted permission for the use of two figures (Woodcuts, 6 and 7) illustrating the skull of the common Frog. I have to thank Mr. C. Robertson, who was conversant with Professor Rolleston's wishes, for assistance on various points, and Mr. G. C. Bourne for several suggestions and for other help ; nor must I by any means omit Professor W. B. Spencer of Melbourne University, and Mr. G. H. Fowler, to whose most timely and friendly care I am entirely indebted for the addition of the Index. The compilation of this book, though laborious in the extreme, has been attended by its pleasures. But the crowning pleasure of all can only befall me if its publication gains the sympathy of those who are com- petent to judge the nature of the task, and the book itself proves a real aid to students in this most fascinating science of Comparative Anatomy. WM. HATCHETT JACKSON. MUSEUM, OXFORD : September 24, 1887. TABLE OF CONTENTS. PAGE List of abbreviated Titles -, * '• \ " , * , «•• ' , •, ••'•'••.' xv Additions and Corrections * . . . , / ," . • " . ' * \ xvi Errata , » < / . • -. . , . » « , , xviii General Introduction . t . . , . xix DESCRIPTION OF PREPARED TYPES. MAMMALIA. 1. Common Rat, Mus decumanus » ' . fc » •* ". » ' i 2. Skeleton of same . , v » . . T-JI . »v - •> 4 3. Skeleton of Wild Rabbit, Lepus cuniculus^ van fefa . • *.- , - 9 4. Vertebrae of same . . , » » > , » , 14 5. Upper half of same . . . , » . v », . 17 6. Duodenum, pancreas, etc, of same * .. • '* .« v- . - . 24 7. Caecum of same , , ,. . , , , » . . 27 8. Bladder and male genitalia of same ,. « » , . . . 30 9. Female genitalia of same « -•' k ' .. ";. , , ••, , . 34 AVES, 10. Common Pigeon, Columba livia » » ;» , » , \ , 46 11. Skeleton of same . . . « • » , » .' . . 58 REPTILIA. 12. Common Ringed Snake, Tropidonotus natrix , . " , .' 67 13. Vertebrae of Python sp. ? . , ^ ' . . . 72 AMPHIBIA. 14. Common Frog, Rana temporaria . .74 15. Skeleton of same ... 79 PISCES. 1 6. Common Perch, Perca fluviatitis » , , , , ,'-, 83 17. Skeleton of same ... . . 90 1 8. Vertebrae of Cod, Gadus morrhua » . . . , 99 UROCHORDA. 19. Ascidian, Ascidia affinis .... 102 GASTROPODA. 20. Shell of Edible Snail, Helix pomatia . . .107 21. Heart and respiratory chamber of same . 109 22. Digestive and reproductive organs of same . .113 23. Nervous system of same . . . . . . .119 X TABLE OF CONTENTS. PAGE LAMELLIBRANCHIATA. 24. Shell of Fresh-water Mussel, Anodonta cygnea . .» . . 124 25. General external features of same . . . . . . . 128 26. Viscera in situ of same . . . . . . . .132 27. Nervous system of same . . . , , ' . 135 INSECTA. 28. Common Cockroach, Periplaneta orientalis . . . . .138 29. Larva of Privet Hawk Moth, Sphinx Ligustri -.' ''* . . 147 30. Pupa of same . . . ...... . . 152 31. Imago, male and female, of same . ..... ... . . 156 CRUSTACEA. 32. Common Crayfish, Astacus fluvia tills, female . . . . 162 33. The same bisected, male . . . . • . . .. 177 34. Heart and vessels of same . . . . . » ..179 35. Digestive, reproductive, and respiratory organs of same . . .. 181 36. Nervous system of same . . . . . . .-- 187 ASTEROIDEA. 37. Common Starfish, Asterias rubens, dried • . . » .'- . .• 190 38. Digestive and motor organs of same . « » . , — . r- .. 195 CHAETOPODA. 39. Common Earthworm, Liimbricus terrestris , . , » 196 40. Digestive, reproductive, and circulatory organs of same . . . 201 41. Nervous system of same . . » * '. * « . 209 HIRUDINEA. 42. Medicinal Leech, Hirudo medidnalis , , . . . . 212 43. Digestive organs of same . . . '« . . .216 44. Nervous system of same .. . . . . . . 219 45. Reproductive and excretory organs of same . . . . 221 CESTODA. 46. Tapeworm, Taenia serrata, and cysts of Cysticercus pisiformis . 224 47. Cysticercus pisiformis, mounted for microscope . ., . . 229 POLYZOA. 48. Broad-leafed Hornwrack, Flustra foliacea . . . . . 234 ANTHOZOA. 49. Sea Anemone, Tealia crassicornis . . , , . : , ' . 239 HYDROZOA. 50. Sea-Fir, Sertularia abietina ; Medusa . . . . 245 PORIFERA. 51. Fresh-water Sponge, Spongilla lacustris • Euspongia officinalis . 249 INFUSORIA AND AMOEBINA. 52. Infusorian, Paramecium Aurelia and Amoeba Proteus . . . r 254 TABLE OF CONTENTS. DESCRIPTION OF PLATES. PAGE Plate I. Common Rat, Mus decumanus . - . • . - . • r .• 26i „ II. Common Pigeon, Columba livia . . , , . , 7 4 . 265 „ III. Common Frog, Rana temporaria . .. «•• / :^ . . 269 „ IV. Skate, Raja batis\ heart, brain, Elasmobranch embryo . . 273 „ V. Cellar Slug, Limax flavus s. variegatus . « . -. . 281 „ VI. Fresh-water Mussel, Anodonta cygnea . ' . . . . . 285 „ VII. Lamellibranchiata : i, e. Ostrea edulis, its digestive tract, heart of Area Noe, diagram of a left nephridium, Veliger of Car- dium pygmaeum . . . » . ." . . * 289 „ VIII. Common Cockroach, Periplaneta orientalis . . . .. ,. 295 „ IX. Arthropoda, i. e. Machilis polypoda, mouth-parts of Cidndela littoralis, same of Apis mellifica, mouth-parts and ambulatory limb of Scolopendra morsitans, Epeira fasciata, digestive organs of Tegenaria domestica, Nauplius of Lepas fasci- cularis . . . .- • , • * • • • 299 „ X. Common Crayfish, Astacus ftuviatilis, male-- . . . . 307 ,, XL Common Starfish, Asterias rubens , . , ...... . 311 ,, XII. Earthworm, Lumbricus terrestris . . . . „ .--. ,*• »' . 315 ,, XIII. Medicinal Leech, Hirudo medidnalis . . ;. . . 319 „ XIV. Cestoda, i;e. Tapeworm, unripe proglottis, uterus of ripe proglottis, proscolex, cystic Coenurus cerebralis . *». . 323 Hydroidea, i. e. Hydra viridis, tentacle of Eucopella, sense-cell, supporting-cell, cnidoblast, ganglion-cell of a guard polype of Plumularia, two nematocysts of Millepora, one discharged, the other not . . . . . . • *••. . . 326 xii TABLE OF CONTENTS. THE ANIMAL KINGDOM. A. METAZOA, p. 333, (a) Coelomata, p. 333. PAGE I. CHORDATA . . • 333 *Vertebrata . ..... 335 **Amniota . . . . . 359 (i) Mammalia . . 359 Sauropsida .......... 372 (ii) Aves . . 373 (iii) Reptilia ... . . . . . . .381 **Anamniota s. Ichthyopsida 393 (iv) Amphibia ......... 394 (v) Pisces . 409 (vi) Cyclostomi ...... ,432 *Cephalochorda s. Pharyngobranchii . . • 437 *Urochorda s. Tunicata .441 II. MOLLUSCA . __L^._ 448 *Glossophora . . g. . f; s," 456 (i) Cephalopoda . . . . 7^ . « . 456 (ii) Scaphopoda .... . . . . . 466 (iii) Pteropoda . --\.^ t , « , . . . 467 (iv) Gastropoda « , . . « . . 469 (a) Isopleura . . . , . . . . 470 (b) Anisopleura . . * . . . . 472 *Lipocephala ......;.... 484 (v) Lamellibranchiata . , , . . . . 484 III. ARTHROPODA . .4 * . , 490 (i) Insecta . ... . . . . 497 (ii) Myriapoda , , . , . . 514 (iii) Protracheata " ^ .- . . , . . . 519 (iv) Arachnida IT . . . , . I . 522 (v) Crustacea . . , » , . " , . 531 Pycnogonidae . . . . . . . .542 IV. ECHINODERMATA 543 (i) Holothurioidea 549 Echinozoa . . . . - . . . . . 554 (ii) Echinoidea . . . . . . . 554 (iii) Asteroidea » . • . . . " . . . 563 (iv) Ophiuroidea . . . . . . . .567 Pelmatozoa . , \ . . . . . . . . 570 (v) Crinoidea \. . . . . . . , 570 (vi) Cystoidea . /.. .- ... . . . 577 (vii) Blastoidea f . , . . . . . . 577 TABLE OF CONTENTS, xiii PAGE VERMES 578 Dinophilus, p. 586; Gasterotricha, p. 587; Echinoderidae, p. 587 ; Desmoscoleridae, p. 588 ; Chaetosomidae, p. 588. (i) Enteropneusta . . . .- . . V. . 589 (ii) Chaetopoda . . . . . , , . 593 Myzostomidae, Archi- Chaetopoda, p. 609. (iii) Archi- Annelida . . . • . ; . - . • . . 613 (iv) Gephyrea . *. . . • . - . . . . 616 (v) Hirudinea . .- v . » . . . 627 (vi) Rotifera . . . . . . . . - . 632 (vii) Nemertea . . . .... , . - . 635 (viii) Trematoda . . . . . . . . . 642 (ix) Cestoda . ' . . . . . . . ... 655 (x) Turbellaria .' . . ... . . 666 (xi) Chaetognatha . , . . . . . .674 (xii) Nematoda . • . . ... . . 676 (xiii) Acanthocephala . . .: ^^ % . . 689 BRACHIOPODA, VERMIFORMIA, POLYZOA, PTEROBRANCHIA . . ., . 691 Brachiopoda . . . . . . . . « 692 Vermiformia . . . . . ... . 703 Polyzoa . . . . , . , . . . 704 Pterobranchia . . .;..,... . , . 711 (b) Coelenterata, p. 712. (i) Ctenophora . . . . . . . 716 (ii) Anthozoa . . v . .^.. '. ' . . . 724 (a) Alcyonaria s. Octactiniae . . . . . 726 (b) Zoantharia s. Hexacgralla . . . , .733 (iii) Hydrozoa . . . . •''•-• • • • 745 (a) Craspedota . . k . . " . . -- . 748 Trachy medusae, p. 749; Hydroidea, p. 755 ; Siphonophora, p. 770. (b) Acraspeda . . . . * 780 Ephyroniae, p. 780 ; Tesseroniae, p. 786. (iv) Porifera .- . . . . . . . 790 MESOZOA, p. 813. B. MESOZOA, p. 813. Orthonectidae, p. 813; Rhombozoa, p. 814. C. PROTOZOA, p. 8 1 8. *Plegepoda . . . . . ... .824 (i) Acinetaria . . . . . . ... . 824 (ii) Infusoria < . . . . . . . 830 (iii) Mastigophora 840 Flagellata, p. 841 ; Choanoflagellata, p. 847 ; Dinoflagellata, p. 848; Cystoflagellata, p. 852. xiv TABLE OF CONTENTS. PAGE *Endoparasita . . . . . . . .857 (iv) Sporozoa . . • . . . " . . . . 857 Gregarinida, p. 857; Amoebosporidia, p. 862 ; Sarcosporidia, p. 862 ; Myxosporidia, p. 863. *Rhizopoda . . . . . . •-. . . 866 (v) Heliozoa . . . . • . . . . . 866 (vi) Radiolaria . , . . . . . .874 (vii) Foraminifera . , . . . . 884 (viii) Amoebina . . . . . , . . 897 (ix) Mycetozoa . . . . . '. • , . . 908 (x) Labyrinthulidea . . . • .. , , . 913 (xi) Proteomyxa ' . . . . . . . . 915 INDEX . . .923 XV LIST OF ABBREVIATED TITLES. The following works which are continually quoted are referred to in the text by letters only. It is hoped all other abbreviations will explain themselves. It maybe well to mention that 'SB.' and 'Dk.' stand respectively for'Sitzungsberichte* and * Denkschriften.' A. M. A. = Archiv fur Microskopische Anatomic, Bonn. A. N. = Archiv fiir Naturgeschichte, Berlin ; often cited in literature as Wiegmann's or Troschel's Archiv. A. N. H. = Annals and Magazine of Natural History (various series), London. A. Sc. N. = Annales des Sciences Naturelles, Zoological section (various series), Paris. A. Z. Expt. = Archives de Zoologie ExpeVimentale et Ge'ne'rale (two series), Paris. C. R. = Comptes Rendus des Stances hebdomadaires de 1'Acade'mie des Sciences, Paris. J. L. S. • = Journal of Linnean Society, Zoological section, London. J. Z. = Jenaische Zeitschrift fiir Naturwissenschaft, herausgegeben von der Med. Nat. Gesellschaft zu Jena. M. J. = Morphologisches Jahrbuch, eine Zeitschrift fiir Anatomie und Entwickelungsgeschichte, Leipzig. Ph. Tr. = Philosophical Transactions of the Royal Society, London. P. R. S. = Proceedings of Royal Society, London. P. Z. S. = Proceedings of Zoological Society, London. Q. J. M. = Quarterly Journal of Microscopical Science, London. Tr. L. S. = Transactions of Linnean Society (two series), London. Tr. Z. S. = Transactions of Zoological Society, London. Z. A. = Zoologischer Anzeiger, Leipzig. Z. W. Z. = Zeitschrift fiir Wissenschaftliche Zoologie, Leipzig ; often cited in literature as Siebold and Kolliker's Zeitschrift. ADDITIONS AND CORRECTIONS. Page 131. Labial tentacles, Thiele, Z. W. Z. xliv. 1886. Pericardial gland (also in Gastropoda), Grobben, Z. A. ix. 1886; x. 1887. pp. 144-5. There are ten stigmata, the abdominal being eight, not seven as stated. The testes and their ducts are paired. p. 146. Studies in Comparative Anatomy, iii ; the Structure and Life-history of the Cockroach, &c., Miall and Denny, London and Leeds, 1886. Translucent white spots on head, Carriere, Z. A. ix. 1886, p. 496. p. 177. Development of Astacus, Reichenbach, Abhandl. Senck. Ges., xiv. (i), 1886. p. 186. Green gland, Grobben, A. M. A., xxx. (2), 1887. p. 206, 1. 17. There are several species of Monocystis inhabiting the Earth- worm. Bergh mentions (Z. W. Z. xliv. p. 308, note 2), in his paper on the genitalia and their development in Lumbricus, that an ovarium or a pair of ovaries may be present in somite xiv in addition to the ordinary pair in somite xiii. Two pairs of ovaries are present in other Earthworms alsoj see Beddard, P. Z. S. 1887, pp. 388-9. p. 212. Giant fibres or neurochord, Leydig, Z. A. ix. 1886. p. 226, 8 lines from bottom : add that, the three lateral nerves spring from the principal ganglion on each side. p. 250, 12 lines from bottom: the cells collect round not in the canals and ampullae. The error is due to a mistake in an abstract, the only account accessible at the time. p. 258, ad fin. The Amoebulae of Pelomyxa probably belong to a parasite; cf. p. 900, note i. p. 302. Tongue of Bee, Breithaupt, A. N. 52, ( ), 1886 ; Graber, Z. A. x. 1887. p. 343 and note. Pineal eye of Lizards, Spencer, Q. J. M. xxvii. 1887; of Fishes (Cyclostomt)> Beard, Nature, xxxvi. 1887, p. 246. p. 347. Ear-ossicles of Mammalia. For other views, see a paper by Dollo, in Q. J. M. xxiii. p. 579, and Baur, Q. J. M. xxviii. p. 169. It seems to me that embryological evidence is conclusive in favour of the view in the text ; see the plates to W. K. Parker's paper on the Skull of Mammalia, Ph. Tr. 176, 1885. p. 349, 10 lines from top. There are five visceral clefts in some Lacertilia and Ophidia, as in Chelonia (Rathke'), according to van Bemmelen, Z. A. ix. p. 352, 10 lines from top. The pulmonary artery of Lacerta, Tropidonotus, and the Hen is the Vlth aortic arch as in Amphibia, the Vth aborting. Id. ibid. PP- 37°> 37i. Selenka, Didelphys Virginiana, Studien iiber Entwickelungs- geschichte der Thiere, i., Wiesbaden, 1886; Caldwell, Embryology of Monotremata and Marsupialia, P. R. S. xlii. 1887. p. 446. On the Salpa chain, see Brooks, Studies Biol. Lab. Johns Hopkins Univ., iii. pt. 8, 1887. ADDITIONS AND CORRECTIONS. xvii p. 483. On the asymmetry of Gastropoda, Biitschli, M. J. xii. 1887. p. 509. Add the following definitions of Thysanura and Collembola : — 1. Thysanura. Rings of the thorax similar; abdomen with ten somites, elongate, terminated by 2-3 appendages usually jointed; rudimentary abdominal limbs sometimes present ; antennae long, many jointed ; maxillary palpi y-jointed, labial /{.-jointed; e.g. Camp odea, Lepisma, Machilis. 2. Collembola. Prothorax small, meso- and meta-thorax not clearly segmented, or the pro- concealed by the meso-thorax; abdomen with six somites, very often globular, with a ventral adhesive tube and a terminal forked springing organ ; antennae 4-8 jointed, palpi absent ; e. g. Smynthurus, Podura. p. 519. Peripatus Capensis. Sedgwick has shown (Q. J. M. xxvii. 1887) that the coelomic cavities of the somites persist as small vesicles in the appendages, that the nephridia are the developed tubular portions of these cavities which obtain external openings, and that the nephridia of the 3rd somite give origin to the salivary glands. The coelome, so-called, of the adult, is a system of vascular spaces (=meta- coele, p. xxix, post) developed independently of the cavities of the somites. The generative cells are derived from the endoderm in the 6th-2oth somites, pass into the dorsal parts of the true coelomic cavities of the same somites, which unite to form the generative glands ; the generative ducts are developed from the coelomic cavities of the 2ist somite; their openings are therefore nephridial. There are no cilia in this species. p. 526. The Malpighian tubes of spiders (Araneidae) have been shown to be, as in some Amphipoda, appended to the mesenteron, not to the proctodaeum ; see Loman, Tijdschrift der Nederl. Dierk. Vereen. (2), i. p. 109. p. 562. Hamann, Beitrage zur Histologie der Echinodermen, pt. 3, Anat, etc., der Echiniden und Spatangiden, Jena, 1887. Doroddaris, etc., Prouho, A. Z, Expt. (2), v. 1^887. p. 576. Me'moire sur 1'Organisation, &c., Antedon rosacea, Perrier, Nouvelles Archives du Musee, Paris (2), ix. 1886. p. 578. Catalogue of Blastoidea, Geol. Department Brit. Mus., Etheridge and Carpenter, 1886. p. 588. For Echinoderes=Kinorhyncha, see Reinhard, Z. W. Z. xlv. 1887. p. 612. Nephridia etc. of Polychaeta, J. T. Cunningham, Q. J. M. xxviii. (2), 1887. pp. 621, 622. The so-called sexual organs of Priapulidae, are both nephridial and sexual, the two parts opening on opposite sides into the main canal. The sexual part is developed last. Schauinsland, Z. A. ix. p. 574. p. 694. Sollas has shown that the coecal processes of the shell in Waldheimia cranium lodge a sensory cell beneath the periostracum, which is connected by a fibril to a nerve-ganglion cell in the mantle. Sc. Proc. Roy. Dublin Soc. v. 1886, p. 318. p. 711. Cephalodiscus proves to be an Enteropneustan, a near ally of Balanoglossus. The epistome or buccal-shield is the proboscis, and contains a coelomic cavity opening by two dorsal pores. It is followed by a collar with a right and left coelomic cavity, opening each by a pore, and extending into the six arms of each side. The body has also a right and left coelomic cavity. The nervous system is ectodermal, and principally massed on the dorsal aspect of the collar, but, extends on to the proboscis and the arms. There is a pair of gill-slits overhung by an b xviii ADDITIONS, ETC. ERRATA. opercular fold, and a diverticulum homologous with the * notochord ' of Balano- glossus. See Harmer, Appendix to M'Intosh on Cephalodiscus, Challenger Reports, xx. 1887. p. 716. Polyparium (not as printed, Polypodium) ambulans; See Korotneff, Z. W. Z. xlv. 1887 (transl. A. N. H. (5), xx) ; Ehlers, ibid. p. 732. Classification of Alcyonaria ; see Studer, A. N. 53 (i), 1887. p. 8 1 1. Hyalospongiae. F. E. Schulze, ' HexactinellidaeJ Challenger Reports, xxi. 1887. ERRATA. Page 53, line 12 from bottom, for and read end. p. 55, line 2 from bottom, for interclvaicular read interclavicular. p. 59, line 2v,for vertebrae read vertebra. p. 61, line 20 from bottom, for long-oblique read long, oblique. p. 91, line 4, for (S. O.) read (£. O.). p. 112, line 24, for 1871 read 1817. p. 113, line 21 from bottom, for ganglia read ganglion. p. 1 68, line 6 from bottom, forgive read fine. p. 1 99, line 1 1 from bottom, for Acanthrodrilus read Acanthodrilns. p. 232, line 20, for Staphylocystes read Staphylocystis. p. 243, lines 20, 21, for Ophridium read Ophrydmm. p. 249, line 15 from bottom,/^ Abth. ii. read Abth. i. p. 250, line 12 from bottom, for in the read round the. p. 253, line 20 from bottom, should read A. N. H. (5) : vii. 1881 ; gemmule of Carterella, ix. 1882 ; x. 1882 ; fossil spicules. p. 273, line 1 6, / for Tyrogiyphidae read Tyroglyphidae. P- 53^) lme 5 from bottom, dele commas after Phyllopoda and Branchiopoda. p. 584, note, line 8 from bottom, /?r arms ra27/ anus. p. 647, line 7 from bottom, for peristallic raw? peristaltic, p. 658, line 1 6 from bottom, for viscual read visual. p. 716, line 17, for Polypodium read Polyparium. GENERAL INTRODUCTION. THERE are two kingdoms, an Animal and a Vegetable, to one of which everything that lives may be assigned with more or less certainty. The contrast between the higher or multicellular animals and plants is too great and constant both in an anatomical and physiological sense to leave room for doubt. An animal possesses the power of locomotion ; it has a compact form, a special digestive organ for the reception of solid food, and it is unable to utilise Carbon dioxide as a source of carbon for the production of carbohydrates ; it has organs of special sense with a nervous system, specialised contractile cells or muscular tissue, complex excretory products, and in the majority special organs of excretion, lymph or blood, with circulatory organs, &c. Its tissue cells, with few excep- tions, are not isolated or all but isolated by closed and firm cell-envelopes. Supposing that it is fixed, its other characteristics remain ; if it takes on a branched or plant-like mode of growth, examination shows that it con- sists of a connected multitude of typical animals, each one perfect in itself. A few instances are known in which chlorophyl bodies are present ; and it appears probable that they enable the organism to utilise Carbon dioxide for the preparation of starch under the influence of light. Whether, however, the chlorophyl bodies are in all these instances intrinsic parts of the organism is a matter of dispute (pp. 242-5). A typical multicellular plant, on the other hand, is either branched, and it then consists of a root with a stem, bearing a number of repeated organs, the leaves, which are subject to modification, or it is compact and its cells very similar one to another. It is fixed ; it has chlorophyl bodies, which under the influence of light enable it to utilise Carbon dioxide as a source of carbon ; it is able to build up protoplasm and therefore derive tissue elements from the Carbon dioxide of the air, ammonia, nitrates and mineral con- stituents of the soil ; it is devoid of digestive organs, of special sense organs, nervous system, excretory organs, special contractile tissue. Its tissue cells become isolated completely or all but completely by closed and firm cell-envelopes. It may require for food partly elaborated material or fairly complex organic compounds in solution, and may then b 2 xx GENERAL INTRODUCTION. be devoid of chlorophyl bodies, but its other characters remain unchanged. If it has a motile initial stage the course of development proves its plant- like nature. In those rare instances, i. e. in Insectivorous plants where solid food is digested and the products of digestion utilised, the process of digestion is carried on externally to the organism, and absorption takes place by the outer surface. What is true of multicellular animals and plants is true, within limi- tations, of unicellular. There are of course in a unicellular animal no specialised systems of organs such as the digestive, for example, but power of locomotion remains, and the natural irritability, automatism and con- tractility of the protoplasm are very strongly developed. Solid organic food is ingulfed within the protoplasm and is broken down, giving rise to fat, albumen, glycogen or other starchy bodies as in higher animals ; it also leaves generally a faecal residue, and there is reason to think that complex and sometimes crystalline excretory products are formed. But the organism may in some cases utilise organic food in solution, in other words it is saprophytic, e.g. some Flagellata, and probably the My- cetozoa ; in other cases, e. g. the Flagellate Euglena, owing to the presence of chlorophyl bodies, nutrition becomes holophytic or completely plant- Jike. In these instances recourse can be had only to considerations of structure, life-history, comparison with other forms, or the behaviour of the doubtful organism under altered conditions of life. Good examples of these considerations may be drawn from Flagellata and Mycetozoa. The position however of some few forms, e. g. the Volvocina, remains a matter of doubt, and they are claimed by botanists and zoologists alike. Their nutrition is holophytic, and their structure is paralleled in undoubted vegetable organisms x. However complex in structure a multicellular animal or plant may be, it can be traced without exception to an origin from a single cell. Many animals, the whole group known as Protozoa, and many plants never attain a higher degree of morphological complexity than a single cell. But in some Protozoa, at any rate, that cell possesses highly de- veloped vital energies and a corresponding specialisation of parts. In its simplest aspect a cell may be defined as a mass of protoplasm (cytoplasm) containing one or more nuclei. It has been shown that non-nucleated masses of protoplasm, derived from nucleated, are in the Protozoa capable of growth in size, but they have no power of reproduction. On the other hand, there are a few Protozoa (certain Proteomyxa) which appear to 1 See Maupas, C. R. 88, 1879, p. 1274. GENERAL INTRODUCTION. xxi consist really of non-nucleated protoplasm, capable not only of growth but reproduction. They may be distinguished as * cytods ' from the cell to which a nucleus is essential. It is possible however that the elements of the nucleus are in these cases disseminate. Protoplasm, or ' the physical basis of life/ is a substance of complex chemical composition containing Nitrogen, Carbon, Oxygen and Hydrogen, with Sulphur, Phosphorus, Sodium and Potassium. From the physical point of view it is viscid, of variable refrangibility, more or less doubly refractile, colourless, hyaline in its purest condition. It appears sometimes to be structureless, but as a rule it is more or less vesicular, consisting of a denser substance (mitome) enclosing droplets of a more fluid character (enchylema, para- mitome), and it is endowed with certain physiological properties, the sum of which constitute life. It is contractile, irritable, possessed of auto- matism, able to convert other protoplasm or less complex compounds, sometimes organic only, sometimes only inorganic, into its own substance. And this nutrition not only maintains the status quo, but if over sufficient for that purpose leads first to the storage of superfluous material in the shape of fat, albumen and starchy bodies ; and secondly, causes a positive increase of bulk, with which is connected the power of repro- duction in its most primitive form — a division of the mass into two similar parts. But all these powers are exercised at the cost of a chemical transformation or degradation of the protoplasm itself, in part respiratory, i. e. oxydative. The products of this degradation, Carbon dioxide and various nitrogenous compounds, are useless to the organism, and are excreted. The whole of the vital properties enumerated can be exercised only while the protoplasm is saturated with water. One of the con- sequences of the vital energies of protoplasm is that as a substance it can never be obtained in a chemically pure condition, which is only approached when it is starved. Otherwise it is laden with the products of progressive and regressive metamorphosis. It may be added that the protoplasm of a cell often gives origin by conversion to an external or internal cell-skeleton, the characters of which, both chemical and physical, are extremely variable l. The nucleus of a cell is a structure sharply marked off from the protoplasm. In its simplest state it is homogeneous and more or less 1 See the introductory chapter in Foster's ' Textbook of Physiology,' and on the movements, &c. of protoplasm, Engelmann, ' Die Protoplasma- und Flimmerbewegung,' in Hermann's Handbuch der Physiologic, i. 1879, P- 343 et seqq., the first part of which is translated in Q. J. M. xxiv. 1884. On the structure and physiology of ciliated cells, see also Engelmann, Pfliiger's Archiv fiir Physiologic, xxiii. 1880. xxii GENERAL INTRODUCTION. rounded, but capable sometimes of changes of shape. It consists prin- cipally of a substance termed nuclein from a chemical point of view, or chromatin from its marked physical peculiarity, that of readily absorbing, and to a much greater degree than the rest of the cell, various kinds of stains such as carmine. But this simple structure is rarely retained. The nucleus is limited externally by a nuclear membrane, its chromatin is disposed in very various ways, as a reticulum, a much coiled thread, in fragments as a lining to the nuclear membrane and one or more central spots. To nodal thickenings of the thread, &c., or to the fragments of chromatin, the term nucleolus is generally applied. The intervals between the chromatin elements are occupied by a nuclear fluid, composed of a protoplasm (caryoplasma), which may be resolved like the protoplasm of the cell (of which it is a part) into a denser and more fluid portion. As this protoplasm takes up stains but slightly, it is frequently designated achromatin. The structure of the nucleus is not always the same at all periods of its life ; it may be at first homo- geneous, but as a rule assumes one of the more complex forms. The process of division of the cell may be comparatively simple or complicated. The division of the protoplasm is preceded, or accompanied, by division of the nucleus. The latter process may be direct or amitotic, the nucleus simply elongating, and being split by a constriction. Or it may be indirect or mitotic, the achromatin being disposed in lines parallel to the long axis of the nucleus, making the figure of a spindle, and the chromatin grouped at the centre of the spindle, dividing into parts which move in opposite directions to either pole of the spindle, whilst a con- striction splits the nucleus in two. It is rare for the chromatin to be grouped in two masses on the equator and the split of the nucleus to take place through its poles. The figures seen in the process are spoken of as karyokinetic. It has been found that the typical mitotic and amitotic modes of divisions are connected by intermediate phases, at least in some tissue cells. The nuclear membrane is dissolved in mitosis and reconstituted round the new nuclei1. 1 The denser mitome of the nucleus and the body of the cell may give rise to an equatorial plate, or the former may do so, and not the latter. This plate, which is common in plants, but has only been detected in certain tissue-cells of Arthropoda, may evanesce, or fission may take place through its median plane. It may be noted that the mitome of the cell-body is frequently arranged in radii during the nuclear changes, and that a clear spot, the polar spot or corpuscle, may appear at each pole of the spindle. For the structure of the cell, see Carnoy, ' La Biologic cellulaire,' Lierre, Fasc. i. pt. 2, 1884 ; for cell-division (cytodieresis) in Arthropoda, Id. ' La Cellule/ Lierre, i. 1885-6, and the summary of both papers by A. Bolles Lee in Q. J. M. xxvi. 1886, p. 481 ; cf. also Flemming and Carnoy in Z. A. ix. 1886. For the fission of the giant-cells, &c. in the medulla of bone, see Denys, < La Cellule,' ii. (2), 1887. GENERAL INTRODUCTION. xxiii The single cell from which a multicellular animal is developed is known as an ovum. It may be derived from an epithelium or sub- epithelium, ectodermic or endodermic (Coelenteratd) ; or from a special organ, the ovary, furnished with a duct and developed usually in the mesoblast, sometimes from special cells set apart at a very early stage of development, e. g. in some Insecta, perhaps in rare instances from the endoderm (some Turbellaria). It may be naked, or provided with one or more envelopes, derived from itself, from surrounding cells, or special glands1. It may be hyaline, or it may be rilled to a greater or less extent with nutrient material, derived by its own vital energies from the lymph-plasma of the body, from the products produced by the regressive changes of surrounding cells (granulosa cells) of the ovary, rarely from other cells. This nutrient reserve-material may be distinguished as food- yolk or deutoplasm from the protoplasm with which it is mixed2. Or the nutrient material may be derived from a special gland, the vitellarium, and be inclosed with the ovum in the egg-shell to be utilised as the ovum segments (some Turbellaria^ Trematoda, Cestoda)*. In one phase of the life-history of the digenetic Trematoda, the Sporocyst or Redia, the re- productive cell is one of a number of cells filling the central part of the body and lining the body walls. These cells may perhaps be regarded as collectively making up an undifferentiated ovary, i.e. as cells from which, in another phase, the immature Fluke, the reproductive organs are derived. As soon as the ovum has attained its definitive size, it very generally, probably universally, gives origin to two polar bodies, or globules, or directive vesicles. The ovular nucleus (germinal vesicle or vesicle of Purkinje with nucleolus or germinal spot) approaches the surface, under- goes karyokinetic changes, and finally one moiety is extruded with a very small amount of protoplasm. After a brief period of rest the phenomenon is repeated. The polar globules may themselves divide again, and the 1 An egg-shell must be carefully distinguished from structures inclosing a number of ova like the cocoons of the Leech and Earthworm, which are secreted by the surface of the body. 2 An ovary in which every ovarian cell becomes an egg, may be termed panoistic ; one in which some only become eggs, others giving origin to secondary yolk -or an egg-membrane, meroistic. The terms are Brandt's, and were originally applied by him to Insectan ovaries. 3 There can be no doubt that a vitellarium is essentially a part of an ovarium. Certain Rhabdocoela prove this point remarkably well ; see von Graff, Monographic der Turbellarien, i. Rhabdocoelida, Leipzig, 1882, p. 138, on the ' Keimdotterstock.' Granulosa cells, yolk cells, epithelium cells connected with the egg, have very generally a similar origin to the egg itself. See A. Thomson, 'Recent researches on Oogenesis,' Q. J. M. xxvi. 1886, p. 602, with lit. given p. 606, and a paper by Korschelt, 'Uber die Entstehung, etc. der versch. Zellenelemente des Insekten- ovariums/ Z. W. Z. xliii. 1886. xxiv GENERAL INTRODUCTION. r nucleus or nuclear moiety they contain pass through mitotic changes1. In some Rotifera, Crustacea, and Insecta one polar body only is formed, and the ovum then proceeds to segment. Such ova are termed partheno- genetic, and the process parthenogenesis. But in the vast majority of multi- cellular animals it is necessary for the ovum to be impregnated, i.e. it must fuse or conjugate with another cell, the spermatozoon, just as in some unicellular animals a temporary or permanent conjugation between two individuals is requisite from time to time to perpetuate the race. The spermatozoon is typically a flagellate cell, with or without the addition of a vibratile membrane ; it is rarely amoeboid ; sometimes of very various shapes even in the same class (Turbellarid) ; motile, except in Crustacea and a few other Arthropods2. It is produced by the repeated fission of a cell, or of part of a cell, belonging to a testis, an organ homologous with the ovary. During its evolution, a process analogous to the formation of polar bodies in the ovum, or homologous with it, is supposed to occur very generally3. The testis may co-exist with the ovary in the same 1 The formation of polar globules is certainly due to cell-fission ; the fact that the nucleus under- goes mitotic changes during their appearance is sufficient to prove the point. They may themselves divide again, and their nuclear fragment show mitosis : see especially Trinchesi, ' Evoluzione nei Molluschi,' Atti Acad. Lyncei, (3), vii. 1879, P^5- J anc^ 8. But they are commonly degenerate in structure. For recent observations see A. Thomson, 'Recent Researches on Oogenesis,' Q. J. M. xxvi. 1886, p. 591, with lit. p. 605. Their significance is a difficult question. If a spermatozoon enters the ovum, e. g. in an Asterias before they are formed, no union between the male and female pronuclei takes place until the process is completed. A single polar body is found in partheno- genetic ova (Weismann, SB. Natf. Ges. Freiburg, i. B., iii. (i), 1887), a fact that disproves the view that the bodies are a male element which must be got rid of before impregnation, or a safeguard against self-fertilisation. See the views of Weismann, Nature, xxxvi., p. 607, and Minot, American Naturalist, xiv. 1880, p. 106 ; cf. Balfour, Comp. Embryology, i. pp. 61-4, and E. van Beneden, 'Recherches sur la fecondation,' Arch, de Biologic, iv. 1883, especially pp. 482, 527, 603 et seqq. Compare with the account given by the last-named, Carnoy, on the germinal vesicle and polar globules in Ascaris megalocephala, ' La Cellule,' Lierre,ii. (i), 1886 ; in sundry Nemat odes, Id. op. cit. iii. 1887, and ibid, in the appendix to the 'conference.' For views connected with the import of the nucleus, &c. to the doctrine of Heredity, see Weismann, ' Die Continuitat des Keimplasma's,' Jena, 1885, summarised by Moseley, Nature, xxxiii. 1885-6, p. 154; Id. 'Die Bedeutung der sexuellen Fortpflanzung,' &c., Jena, 1886, similarly summarised, Nature, xxxiv. 1886, p. 629 ; also Kolliker, ' Die Bedeutung der Zellkerne,' &c., Z. W. Z. xlii. 1885, summarised in the American Naturalist, xix. 1885, p. 1222; Id. 'Das Karyoplasma,' &c., Z. WT. Z. xliv. 1886; O. Hertwig, 'Das Problem der Befruchtung,' &c., J. Z. xviii. 1885, and the chapters on the reproduction of plants in Vines, Phy- siology of Plants, Cambridge, 1886, or the corresponding lectures in Sachs, 'Physiology of Plants,' transl. by Marshall Ward, Clarendon Press, 1887; also Geddes, 'Theory of Growth,' &c., Proc. Roy. Soc. Edinburgh, 1886. 2 The immobility of the spermatozoa of Crustacea and some other Arthropoda is somewhat doubtful. The spermatozoon of Cypris becomes active when transferred to the female ; and that of the Cladoceran Polyphemus shows amoeboid motion (Zaccharias, Z. W. Z. xli. 1885). 3 For the accessory globule of the spermatozoa, see E. van Beneden and Julin, Bull. Acad. Roy. Belg. (3), vii. 1884, p. 322 ; Brown, Q. J. M. xxv. 1885, pp. 350-1, 357 ; and A. Thomson, 'Recent Researches on Oogenesis,' Q. J. M. xxvi. 1886, pp. 596-8. The accessory globule has been supposed to get rid of a female element in the developing spermatozoon and thus to be homologous with a GENERAL INTRODUCTION. xxv a animal, which is then said to be monoecious or hermaphrodite ; or it may be lodged in another animal of the same species, in that case said to be dioecious or of separate sexes. In hermaphrodite animals the testis may ripen at a different time to the ovary, a phenomenon known as successive hermaphroditism, and in most instances certainly a safe- guard against self-impregnation, e. g. in the hermaphrodite Gastropoda. Some hermaphrodites, however, are self-impregnating, such as Cestoda, some Trematoda. The Nematode genus Angiostomum is a unique example of an organism, which though anatomically a female, is yet a self- impregnating hermaphrodite. The actual process of impregnation is, briefly stated, first the penetration of the spermatozoon into the ovum, either through its envelope at any spot, or by a special aperture, the micropyle, secondly the fusion of the protoplasm of the two cells, which is perhaps an unessential feature, followed thirdly by fusion of the nuclei, often termed the male (spermatozoal) and female (ovular) pronuclei. The two pronuclei approach each other, and the granules of the surrounding protoplasm are arranged round each of them, so as to form a star or aster with a pronucleus as a centre. This aster is most pronounced on the aspects of the pronuclei turned to one another. The ovum has now become an oosperm, and it speedily undergoes fission or segmentation and gastrulation. What is generally considered, but perhaps wrongly, to be the most primitive mode of segmentation is seen in an oosperm, which is alecithal, i. e. devoid, or nearly so, of food-yolk. The nucleus divides with mitosis, and a constriction splits the oosperm into two equal or sub-equal halves or blastomeres. Each half then divides again into two, and so on. The two first divisions take place in a vertical plane, the third in a horizontal, the fourth in a vertical, and the fifth in a horizontal, and then regularity is lost. In many instances, however, a regular sequence of stages is not recognisable. The result of segmentation is the formation of a hollow sphere, the blastula or blastosphere, the cells or blastomeres being disposed in a single layer round a central cavity filled with an albuminous liquid, the blastocoele or segmentation cavity. In many instances the blasto- coele is absent, or nearly so, and the term morula is then used instead polar globule. Such a value can hardly be assigned to the various forms of blastophores, nucleated and non-nucleated, seen in spermatogenesis. The male cell is incapable of further development, i. e. of parthenogenesis in animals where it is highly specialised ; so, too, in the higher plants, but in certain of the lower male parthenogenesis appears to occur, see Vines, ' Physiology of Plants,' p. 674 ; cf. Weismann, ' Continuitat des Keimplasma's,' Jena, 1885, cap. ii. p. 70, on the significance of polar bodies, and cap. iii. p. 88, on the essential character of parthenogenesis ; Id. Nature, xxxvi. 1887. p. 607. xxvi GENERAL INTRODUCTION. « of blastosphere1. The cells of the blastosphere are frequently (? always) dissimilar at opposite poles, one set typically smaller and more clear, the other larger and more granular. The latter undergo invagination or embole, that is to say they sink inwards, obliterating more or less completely the blastocoele. The result is an invaginate, or embolic Gastrula, an ovate or spherical body composed of a double layer of cells, an outer epiblast ( = ectoderm), an inner hypoblast ( = endoderm), separated or not by a space, the remnant of the blastocoele, and continuous at the blastopore or Gastrula mouth. The central cavity into which the blastopore leads is the archenteron2. Instances of typical or equal segmentation are met with in most groups of multicellular animals, but the process is commonly modified by the accumulation of food-yolk ; the blastocoele may be absent or only slightly indicated, and transformation into a Gastrula is sometimes carried out by an invagination, sometimes by a modified form of invagination known as overgrowth or epibole, that is to say the epiblast grows round the hypoblast or yolk. The food-yolk may accumulate at one extremity of the oosperm, or in its centre ; to these two types the terms telolecithal and centrolecithal are respectively applied. A telolecithal ovum may segment completely but unequally, the hypoblastic cells being larger and dividing more slowly, or its segmentation is partial and confined to a disc at one pole, a large amount of yolk remaining unsegmented ; nuclei appear in it, however, at the pole of segmentation, a certain amount of protoplasm becomes segregated round them, and cells are thus added to the blastoderm or segmented area. The extremes are connected by transitional forms. Centrolecithal ova are confined to Arthropoda. The central aggregation of the yolk may be present from the first, or take place during segmentation. The yolk is in the latter case always massed at the central ends of the blastomeres, which may or may not fuse, whilst in the former case the furrows between the blastomeres are superficial, i. e. do not penetrate to the centre of the oosperm. The central mass of yolk thus left either does not segment at all or does so at a late period, and the masses to which it gives origin are non-nucleate. The blastomeres may be equal or unequal, their formation may be simul- taneous or successive, and is very often preceded by a multiplication of 1 The term ' morula ' is also applied to solid masses of cells produced by segmentation and not yet definitively arranged in Gastrula-fashion. 2 The blastocoele is sometimes open by one or more pores to the exterior, e. g. in the amphi- blastula of Sycon (Sycandra}. It would have been better if the term blastopore had been restricted by usage to such openings, and some such term as ' gastropore ' applied to the Gastrula mouth. GENERAL INTRODUCTION. xxvil nuclei. Indeed in Peripatus cells are never distinctly delimited, the result being a syncytium1. The Gastrula is derived either by invagination or by differentiation of the yolk-cells. In some Coelenterata the Gastrula stage is attained either by im- migration of cells from one or various points of the blastosphere into the blastocoele, with subsequent differentiation of the immigrant cells, by delamination of the inner ends of the blastospheral cells, or by a mixture of the two processes. See p. 746, note i, pp. 752, 764, 800. The varia- tions observable in closely allied genera, make it probable that the phe- nomena as observed in these cases are of secondary origin, due perhaps to a precocious formation of the endoderm. The name ' parenchymula ' or ' parenchymella ' has been applied to the form where the central cavity is filled with cells. There can be no doubt that segmentation and gastrulation, processes which take place in every life-history or ontogeny, represent ancestral stages in the evolution or phylogeny of multicellular animals. But at the present time there is no such thing known as a sphere or blastosphere leading an independent life and reproducing its kind, its component cells united by a bond indissoluble without entailing death on each cell. The same statement is true of the Morula and of the Gastrula. The claims of the Mesozoa to represent a Gastrula are excessively doubtful ; those of Haeckel's Gastreadae^ at present inadmissible. See pp. 817-8. Putting, the Mesozoa aside, the vast majority of multicellular animals may be classified as Metazoa. The growth of the individual is compli- cated by the formation of tissues and systems of organs. . Sensory and nervous tissue, contractile tissue, supporting or connective tissue, localised reproductive tissue, are differentiated in connection either with the epi- blast s. ectoderm, and hyboblast s. endoderm of the Gastrula, or in part independently of them. Two main divisions of Metazoa are recognisable, the Coelenterata and Coelomata. The typical characters of the Coelenterata are as follows. The funda- mental symmetry of the Gastrula is as a rule retained ; the vertical axis passing through the blastopore persists ; if an anterior and posterior extremity are distinguishable they are equal ; and the same is true of a right and left side. The blastopore may close, or not be developed ; the archenteron, except in Porifera, opens by a single principal aperture, which is either a perforation of the two embryonic layers, or a distinct involution of the epiblast known as stomodaeum, which may assume a 1 This may be a very primitive condition ; cf. Sedgwick, Q. J. M. xxvii. 1886, pp. 515-30. xxviii GENERAL INTRODUCTION. digestive function (Ctenophora). There appears between the ectoderm and endoderm a gelatinous lamella— the mesoglaea, which may be structure- less, partly fibrillate, or invaded by cells derived from one of the two epithelia. Sensory and nervous cells are epithelial or sub-epithelial ; the same is true of the contractile cells, which may, however, become im- bedded in the mesoglaea. The generative products are sub-epithelial and localised. But a differentiation of the mesoglaeal cells may occur, princi- pally in Porifera. Reproduction by division of the organism is rare ; by buds or outgrowths of the ecto- and endoderm jointly, common ; by the division of a single mesoglaeal cell, or the growth of mesoglaeal cells plus endoderm cells confined to Porifera. See pp. 713-6 ; 804. The typical characters of the Coelomata contrast with those of the Coelenterata as follows. The fundamental symmetry of the Gastrula and the vertical axis passing through the blastopore do not persist. As a rule equal right and left sides are distinguishable ; but the anterior and posterior parts of the body, if the permanent mouth, as is most natural, is taken as a point of reference, are not equal, the former being relatively small, and constituting a more or less distinct head. The archenteron, the mesenteron of the adult, communicates with ,the exterior by a mouth, and as a rule by an anus. The blastopore may become obliterated in its centre, and its two ends may coincide with the position of the future mouth and anus as in Peripatus ; it may close from behind forwards, or vice versa, and then the mouth or anus correspond respectively to the part left open ; it may close and leave no trace ; or it may never be found at all as in Insecta. The permanent mouth and anus of the adult are gene- rally, perhaps always, formed by a more or less pronounced ingrowth of ectoderm, either at the open part of the blastopore, or independent of it. In the latter case it is a question how far the ingrowth coincides with the obliterated part of the blastopore, or the spot where it might be expected to be. To the oral and anal ingrowths, the terms stomodaeum and proc- todaeum are applied. Instead of a mesoglaea, there is a cellular meso- blast or mesoderm. Its cells in all cases lie or come to lie between the epi- and hypoblast. They are formed at an early period in the ontogeny, and as a matter of fact are derived in several ways the mutual connection of which is disputed. They may have a single or a double source. As to the former, the cells arise (i) as immigrants (mesenchyme cells), from the walls of the blastosphere, or from its endodermal pole as in Nemertea ; (2) from the walls of the archenteron close to the blastopore, e. g. many Crustacea ; (3) from cells specialised at an early period at the blastopore, GENERAL INTRODUCTION. xxix e. g. the pole-cells of Chaetopoda ; (4) from the primitive streak behind the blastopore in Peripatus, the same streak in Insecta or Spiders, which may be partial or complete homologues of the streak in Peripatus l ; (5) from the walls of diverticula of the archehteron or enterocoelic pouches as in Balanoglossus, Sagitta and Brachiopoda. When it has a double source it may be (i), supra, combined with enterocoelic pouches as in Echinoder- mata, or with pole-cells as in Thalassema ; (3) S2ipra, combined with entero- coelic pouches as in Amphioxus ; (2) or (4), supra, combined with cells derived from the hypoblast as in most Vertebrata. Whatever significance these facts may have, the mesoblast gives rise to the muscular, connective and skeletal, blood and lymph tissues, and very generally to the genital and excretory cells. In Vertebrata, Cephalochorda, Arthropoda, some Vermes, e. g. Chaetopoda, it is broken up into a series of paired segments, sometimes, as in many Crustacea, obscurely marked, giving rise in the adult to a metamerism or serial segmentation, the primitive origin of which is a matter of doubt. In connection with the mesoblast or its segments is the cavity, or series of cavities, known as body cavities or coelome, which are not homologous throughout the Coelomata. In some Vermes a coelome is absent, or represented by irregular spaces or gaps in the mesoblastic tissues, as in Turbellaria, Trematoda, and Cestoda, but in other Coelomata it is probable that it falls under one of the following heads, (i) It is an archicoele, or remnant of the blastocoele as in the vascular system of Nemertea, the head cavities of some segmented Vermes, e. g. Polygordius, some Chaetopoda, the body cavity of Rotifera, and Dinophilus (?). (2) It is a system of channels and spaces excavated in tKe mesoblast secondarily, e. g. the principal portions of the coelome, or the vascular spaces in Peripatus, and perhaps in Mollusca, or the whole of it in Arthropoda in general. To this type the name metacoele might be applied 2. (3) It is an enter ocoele, i. e. the persistent cavity of diverticula of the archenteron (supra], as in Amphioxiis, Balanoglossus, Sagitta, Brachiopoda, and Echino- dermata. There remain the coelomic cavities of Vertebrata, the isolated nephridial pouches and genital ducts of Peripatus, the series of cavities in segmented Vermes, the pericardium, nephridial, and perhaps genital cavities of Mollusca. In these instances they are now usually regarded as enter o- 1 See on the primitive streak, Sedgwick, Q. J. M. xxiv. p. 79, and xxvii. p. 530; Haddon, Introduction to the Study of Embryology, 1887, p. 41 ; and the various references given in the Index vol. ii. of Balfour's Comparative Embryology. 2 Sedgwick uses the term pseudocode, but it would apply equally well to an archicoele, or to a schizocoele, if there is such a thing, supposing his definition of a coelome to be accepted (Q. J. M. xxvii. p. 533). xxx GENERAL INTRODUCTION. coeles abbreviated in development, a view which has most probability so far as the Vertebrate are concerned, and if a distinctive name is applied to them, crypt-enterocoele might be suggested ; or they may be simply splits in the mesoblast, not derived from enterocoeles, and the well-known term schizccocle may be retained for them. Whatever value is attached to the coelome, the result of its presence is in most instances a division of the mesoblast into two portions, one applied to the body-wall, the other to the mesenteron, for which when the separation takes place in the embryo, the names somato- and splanchno-pleure are used. It remains only to state that the epiblast gives origin in the adult to the epi- or hypo-dermis, to exo-skeletal structures, to glands, to organs of special sense and the nervous system, to the stomo- and procto-daeum and organs derived from them ; the hypoblast to the intestinal or mesenteric epithelium, to the epithelia of glands or other outgrowths derived from the mesenteron, such as the lungs, thyroid, thymus, and the notochord of the Chordata. Reproduction by fission, or by gemmation in which the three layers are' always (?) implicated, is not common ; but the power of reproducing lost parts is met with in Coelcmata as high in the scale as Lacertilia *. The connection between the Coelenterata and Coelomata is probably only that of descent from a common form of ancestor, unless it be sup- posed that the larval Coelomates with enterocoelic diverticula have sprung from Gastrulae common to them and the Anthozoa. The classes of Coelenterata as they now exist are specialised ; it is possible that two phyla may be indicated, one which has given origin to the Porifera, the other to the three remaining classes. Among the Coelo- mata are found groups, the relations of which are absolutely uncertain, e. g. Brachiopoda, Polyzoa. Certain phyla, or lines of common descent, may be indicated in other cases with confidence. These are (i) the Chordata with which Balanoglossus and Cephalodiscus are allied, if they are not actually to be considered as Chordates ; and (2) the Echino- dermata, both of which are related to ancestors with enterocoelic pouches ; (3) the Mollusca^ descended from a trochosphere-ancestor common to them and most Vermes ; (4) the Arthropoda, segmented animals which 1 For views on the various points touched on in the foregoing account, see Sedgwick, ' On the origin of Metameric segmentation,' &c., Q. J. M. xxiv. 1884; Id. 'The Development of the Cape Species of Peripattis] op. cit. xxvii. 1886, pp. 515-40; Caldwell, 'Blastopore, Mesoderm, Meta- meric Segmentation,' Q. J. M. xxv. 1885 ; Hubrecht, ' The relation of the Nemertea to the Verte- brata,' Q. J. M. xxvii. 1886. On the Mesoblast see also, Kleinenberg, ' Die Entstehung des Annelids aus der Larve der Lopadorhynchiis, Z. W. Z. xliv. 1886 ; Hubrecht, on Lineus, Q. J. M. xxvi. 1886 ; Salensky, on Pilidium, Z. W. Z. xliii. 1886; Metschnikoff, on the wandering cells of Asterids and Echinids, Z. W. Z. xlii. 1885, p. 656. GENERAL INTRODUCTION. xxxi probably have two separate lines of descent, one including Arachnida and Crustacea, with an ancestor represented possibly by a Naufttms-form ; the other Insecta, and Myriapoda with Peripatus, the latter indicating perhaps an ancestry related to the segmented Vermes. There remain (5) the majority of Vermes, an assemblage of apparently very diverse forms ; the remarks on pp. 583-4, and the enumeration of classes, pp. 585-6, may suggest possible connections. The task of unravelling the phylogeny of the subdivisions of Metazoa is one beset with extreme difficulty. The records of Geology have estab- lished such points as the pedigree of the Horse, the derivation of Birds from extinct Reptilians ; they show the extreme antiquity of some living types, e. g. Insectans, Scorpions, Elasmobranchs in Silurian strata, the great prevalence in ages past of forms now extinct or almost extinct, their replacement in other instances by derived .types. Other phenomena speak to vast changes: the present geographical distribution of many terrestrial and aquatic animals ; mimicry, migration, social habits ; the degeneration which has so evidently befallen certain types, due to a sedentary mode of existence, to minute size, to adaptation for a parasitic life, i. e. one de- pendent on the living tissues or vital processes of another animal or plant for sustained nurture, whether it be external or internal, ecto- or endo- parasitism ; Alternation of Generations, first discovered by Chamisso in Salpa, and by Steenstrup in Hydroids, Trematodes, &c., whether in the form known as metagenesis, i. e. the alternation of asexual and sexual individuals, or as heterogamy, i. e. the alternation of parthenogenetic and sexual races, or in one instance (Angiostommn) of an hermaphrodite and self-impregnating individual with bisexual individuals ; the occurrence of prolonged metamorphoses, such as are seen in many Arthropoda, and the shortened metamorphoses of the early ontogeny of most animals ; the degradation of an individual into an organ — a rare occurrence — exemplified in the avicularia and vibracula of Polyzoa, or the converse phenomenon of parts of an individual becoming elevated into the semblance of a number of individuals, the most probable interpretation to be put on the strobila of Cestoda. Nor has natural selection left untouched the record written on the pages of the life-history of any animal ; it has falsified it in various ways at every stage — the ovum, its segmentation, the embryo. Special embryonic organs may attain a great prominence ; normal embryonic phases may be slurred over, or perhaps extinguished. It is often hard to say what is ancestral, what acquired, to distinguish between structures which may be inherited or independently evolved. xxxii GENERAL INTRODUCTION. General works. H. Milne Edwards, Lesons sur la physiologic et Fanatomie comparee de 1'homme et des animaux, Paris, 14 vols. 1857-81. Gegenbaur, Grundriss der Vergleich. Anatomie, Leipzig, 1878, transl. by F. J. Bell, revised by E. Ray Lankester as 'Elements of Comparative Anatomy,' London, 1878. Glaus, Grundziige der Zoologie, ed. 4, Marburg, 1880-2. Traite de Zoologie, Glaus, transl. by Moquin-Tandon, ed. 2, Paris, 1884. Textbook of Zoology, Glaus, transl. by Sedgwick, London, 2 vols. 1884-5. Hayek, Handbuch der Zoologie, Vienna, 3 vols. 1885-87. Schmarda, Zoologie, Vienna, 2 vols. 1877-8. Pagenstecher, Allgemeine Zoologie, Berlin, 4 parts, 1875-81. The Various 'Suites a Buffon.' Embryology. F. M. Balfour, Comparative Embryology, 2 vols., London, 1 880-8 1 . A. C. Haddon, An Introduction to the Study of Embryology, London, 1887. Distribution. A. R. Wallace, Geographical Distribution, 2 vols., London, 1876. Id., Island Life, London, 1880. Practical Manuals. A. M. Marshall and C. H. Hurst, A Junior Course of Prac- tical Zoology, London, 1887. T. J. Parker, A Course of Instruction in Zootomy (Vertebrata), London, 1884. Carl Vogt and Emile Yung, Trait£ d'anatomie com- pare'e pratique, Paris, 1882 (incomplete). W. K. Brooks, Handbook of Invertebrate Zoology for Laboratories and Sea-side Work, Boston, 1882. T. H. Huxley and H. B. Martin, A course of Elementary Instruction in Practical Biology, ed. 4, London, 1877. G. B. Howes, Atlas of Practical Elementary Biology, London. 1885. Bibliography. Zoological Record, London, from i. 1864, onwards. Zoologischer Jahresbericht, edited from the Zoological Station at Naples, Berlin, 1879 and on- wards. Jahresberichte iiber die Fortschrifte der Anatomie und Physiologic, Hoffman and Schwalbe, Leipzig, i. 1873 and onwards. Lists in the Zoologischer Anzeiger, i. 1878, and onwards. The Berichte in vol. ii. of the yearly issues of the Archiv fur Naturgeschichte, commencing with the issue for 1836. Royal Society's Catalogue of Scientific Papers, 1800-1863, London, 6 vols., 1867-72 ; second series, for period 1864-73, 2 vols., London, 1877-9. L. Agassiz, Bibliographia Zoologiae et Geologiae, edited by Strickland and Jardine, 4 vols., London, Ray Society, 1848-54. Biblio- theca Zoologica, Carus and Engelmann, works from 1846-60, 2 vols., Leipzig, 1861 ; Continuation under same title, works from 1 861-80, O. Taschenberg, vol. i. issued. D'Arcy Thompson, Bibliography of Protozoa, Sponges, Coelenterata, Worms, Polyzoa, Brachiopoda, and Tunicata, 1861-83, Cambridge, 1885. The various catalogues of the Natural History Department of the British Museum. For Insecta, see Hagen, Bibliotheca Entomologica, 2 vols., Leipzig, 1862-3. Palaeontology. Zittel, Handbuch der Palaeontologie, Abth. i, Palaeozoologie, vols. i. and ii. on Non- Vertebrates, complete; vol. iii. on Vertebrata, commenced, Miinchen and Leipzig. Quenstedt, Handbuch der Petrefactenkunde, ed. 3, Tubingen, 1885. Caps, vi, vii, viii of Wallace's 'Distribution/ supra, for distribution of extinct animals. Fossils in relation to strata, Part II of Phillip's Manual of Geology, ed. by Seeley and Etheridge, London, 1885. Bibliography of Palaeontology. Geological Record, Whitaker and Dalton, 1874-78; the lists and summaries in the Zoological Record, the Zoologischer Jahresbericht (supra}, and the Neues Jahrbuch fur Mineralogie, Geologic, and Palaeontologie, Stuttgart. K UNIVEKSITY %g4L DESCRIPTIONS OF PREPARATIONS. i. COMMON RAT (Mus decumanus], Dissected so as to show its craniospinal nervous axis in its entire length as well as portions of most of the organs of vegetative life. A RED injection has been thrown into the veins ; and the left halves of the walls of the craniospinal, thoracic, abdominal, and pelvic cavities, as well as the greater part of the integument in the facial region and the greater part of the left lung, have been removed so as to show in situ the organs previously concealed by these structures. Of the encephalic nerve-centres we see most anteriorly the olfactory lobes : next to them the cerebral, separated from each other by the longi- tudinal fissure in which is lodged the longitudinal sinus : next the cere- bellum bounded off anteriorly from the posterior border of the cerebral lobes by the diverging lateral sinuses, into which the longitudinal sinus divides. The presence of the lateral sinuses prevents us from seeing the corpora quadrigemina which would otherwise be visible in the middle line, owing to the divergence there from each other of the cerebral lobes. The medulla oblongata, which is, like the cerebellum, of considerable width, comes into view between the two occipital condyles, from which point down to the second dorsal vertebra, recognizable by its long spine carrying an ossicle articulated to its apex, the medulla spinalis is of much greater thickness than it attains posteriorly. It is seen in the lumbar region to break up into the cauda equina. In the dorsal region, a black bristle has been passed under the aorta where it underlies the bodies of the vertebrae, and this position relatively to the craniospinal canal superiorly, as also to the digestive tract next in- feriorly, and the heart most inferiorly, is held by the aorta in all vertebrata. The singleness of the aortic trunk in the adult state is characteristic of all warm-blooded animals ; but mammals, as is seen here, differ from birds in having the single trunk arching from the heart over the left and not over the right lung's root. Behind and to the right of this black bristle from before backwards are to be seen, firstly, the fourth lobe of the right lung in its pleural cavity resting on the diaphragm below, and in relation above with the heart, and on the left with the phrenic nerve ; secondly, the oeso- phagus, a lowly vascular tube the small calibre of which is correlated with the working of the dental apparatus in these creatures ; thirdly, the third B 2 DESCRIPTIONS OF PREPARATIONS. lobe of the right lung placed far back and to the right, and, like the lungs of all mammals, freely suspended in its pleural cavity and bearing no im- pressions on its exterior from the different bony constituents of the thoracic cavity ; fourthly, the vena azygos of the left side between the aorta and the vertebral column, passing up to arch over the root of the left lung, and join the vena cava descendens of that side ; and fifthly, the spinal cord. The com- plete diaphragm, forming a dome-shaped floor, with the heart and lungs in relation with its convex, and the liver, stomach, spleen, and kidney in relation with its concave surface, is eminently characteristic of Mammalia, that of the Crocodilina alone approaching this grade of development. The upper part of the pericardial sac has been removed, and the two ventricles (less distinctly separated from each other than in many mammals) and the left auricle are brought into view. The anterior surface of the heart is more equally shared in by the two ventricles than is the case in many mammals, in which the right ventricle forms nearly the entire anterior aspect of the organ. The left vena cava descendens, a trunk which is found in most Rodents, except the Guinea Pig and Agouti, is seen to pass in front of the root of the left lung in company with the phrenic nerve round to the back of the heart to end in the right auricle. The vena azygos of the left side is seen to join it just above the root of the left lung, and at a point some way above this, the vein from the fore-leg, which is in relation with the nerves going to that limb, is seen passing up to join another vein, which, from its being placed superficially to the sternomastoid muscle, we know to be the homologue of the external jugular of anthropotomy. The external jugular is the main trunk by which the blood from the interior of the skull returns to the heart in the Rodents and many of the lower Mammalia, and by its confluence with the vein from the anterior limb the vena cava descendens is constituted. Internally to the external jugular, just above its confluence with the subclavian vein, is seen a part of the hibernating gland ; externally to it lies the submaxillary ; above this again we see the parotid with its duct ; and above the parotid, the facial portion of the lacrymal gland sending up a duct, under which a piece of blue paper is placed, to enter the orbit and join there with the duct of a second portion of the lacrymal gland, which is placed within the orbit, and anteriorly to the duct of the extra- orbitally-placed portion. Within the orbit we see the Harderian gland in relation with the third eyelid l. 1 For a fuller description of these glands, see Description of Plate I, which represents a dissection somewhat different from that which we have of these organs in this preparation. The lacrymal gland is somewhat similarly bilobed in the human subject, consisting of a palpebral and an orbital part. See Hirschfeld et Leveille", Neurologic, 1853, PI. 76, fig. 4. In man however there are between twelve and fourteen minute lacrymal ducts instead of a single one as here. But the macro- scopic Harderian gland and duct of the Rodentia and mammals lower than Primates except the Chiroptera do not similarly represent the minute Meibomian glands with separate ducts on the free edge of the eyelids ; for both sets of glands coexist in Rodentia. COMMON RAT. In the middle line of the body inferiorly to the heart we see surfaces of the six sternal bones, and in the angle intercepted between the lowermost of these and the diaphragm, we see some lobules of fatty tissue set in the process of serous membrane which connects the apex of the pericardium with the sternal bones and with the diaphragm. From these structures a vein passes back along the pericardium to end in the vena cava descendens of the left side. In the angle between the Inferior surface of the diaphragm and the lumbar muscles, the two psoas muscles and the quadratus lumborum of the left side, we see the smooth-surfaced kidney, which by this external character, as also by the internal one, of the separation of its cortical or secretory from its medullary or excretory parts, characterizes the class Mammalia. The spleen is in relation with it on the right ; to the right of the spleen we have the left end of the stomach, which is less vascular and glandular than the pyloric half, here concealed and overlapped by the left lobe of the liver. From the inferior or convex margin of the stomach the curtain-like omentum^ a process of peritoneum found, thus developed, only in mammals, hangs down over the left cornu of the uterus, which is dis- tended with embryos, and over portions of the intestines. Immediately below the kidney and the spleen, the left ovary and Fallopian tube and the upper end of the left cornu uteri are situated. A fibrous band, under which a black bristle is placed, and which is the remnant of the ligament by which the Wolfftan body in the foetus was kept in relation with the diaphragm, attaches the ovary and tube to the peritoneal covering of that muscle. Below the upper end of the left cornu uteri is seen the caecum, which is of less size and complexity than that of Rodents with rootless molars and less varied and nutritious food than these omnivorous members of the order, or than that of those, such as the Squirrels, which live on seeds and have, like most Mnridae, rooted molars. It tapers off superiorly into the large intes- tine, which however in many Rodents is not, when compared with the small intestine, as much inferior in length and larger in calibre and thicker in its walls as its name and the homology of anthropotomy might lead us to expect. Below the caecum we see the cut ends of the veins from the hind-limb, and lower still we see a bristle passed underneath the ureter as it passes forwards to enter the base of the conically contracted bladder. The vagina, rectum, and bladder have, each of them, separate and indepen- dent outlets ; into those from the two latter organs black bristles have been passed. The flat nail on the rudimentary thumb ; the presence of tactile vibrissae above the eyes as well as upon the snout ; and of hairs of great coarseness along the mesial dorsal region ; the absence of hair from a small area, bifid, as usual in Rodents, in which are the orifices of the nostrils, and which is called the ' muffle ; ' and its presence between the annulate scales on the tail ; are points worthy of notice. B 2 4 DESCRIPTIONS OF PREPARATIONS. For the relations held by the cerebrum and cerebellum to each other and to the tentorium, see Turner, Proceedings Royal Society of Edinburgh, March 3, 1862. For the various arrangements observable in the system of the vena azygos, see Milne-Edwards, Lesons sur la Physiologic, vol. iii. p. 598, ibique ritata. For the histology of the Hibernating Gland, see Hirzel and Frey, Z. W. Z. xii. 1862. For that of the Harderian, in Mammals, see Wendt's Monograph, Die Hardersche Druse, 1877; and in Birds, see MacLeod, Bulletin Acad. Royale Sci. Belgique, 1879, pp. 797-810. For a figure and account of the ligamentum diaphragmaticum in the foetal state, see Kolliker's Entwickelungsgeschichte, p. 961, Fig. 587, 1879; and Tr. Z. S. vol. v. p. 286, 1863. For an account of the perforation of the clitoris by the urethra in the Cape Mole, see Hunterian Catalogue of the Physiological Series contained in the Royal College of Surgeons, vol. iv. p. 2745 ; for a similar arrangement in Talpa and Stenops and Lemur, see loc. rit. 2810, 2811, 2812. For the use of the word ' muffle,' see Waterhouse's Nat. Hist. Mammalia, vol. i. p. 50; vol. ii. pp. 7, 8; Sundevall's Linne's Pecora, Germ. Transl. 1848, pp. 41-43. 2. SKELETON OF COMMON RAT (Mus decumanus). THE skeletons of many of the lower Mammalia bear a general resem- blance to those of certain quadrupeds lower in the scale of life in such points as the nearness of the level at which their trunk is carried by their limbs to that of the ground on which they move ; and in the maintenance by the long axis of their head, of much the same direction as that of the long axis of their entire trunk. But they invariably present the following distinctive characters, which are as peculiar to the Mammalian class as any of the points furnished by the soft parts, such as the blood-cells, the hairy integument, or the mammary glands. In every Mammalian skeleton each half of the lower jaw is made up of a single mandibular bone on each side, which at birth at least, if not, as it is here, throughout life, is distinct from its fellow of the opposite side, and articulates by a convex facet with the squamosal element of the cranial wall ; and the vertebrae in the trunk always differ from those of the different lower Vertebrata in one or more or all of the following points : either in the anchylosis of their several elements, or in the size of their neural canal, or in the shape of the articular ends of their centra, or in the means whereby in the recent state these articular ends are brought into relation with each other. In the vertebra of a young mammal the neural arch may not have anchylosed with its centrum ; but in all such cases two discoid epiphyses belonging to the articular ends of the centrum would also remain unanchylosed, as they fuse SKELETON OF COMMON RAT. 5 with it at a later period than the neural arch, and they furnish a mark as distinctive of the Mammalian class as any other connected with the verte- brae. Some mammals have an opisthocoelian ball and socket articulation between the centre of their vertebrae ; and the crocodiles resemble the mammals in having interarticular fibrocartilaginous discs to connect their ball and socket centre-joints instead of synovial joints ; but in such cases the greater size of the neural canal or the absence of neurocentral sutures, or the absence of sutures between the body and the lateral processes, would enable us, without having recourse to a microscopic examination of the bony tissue, to identify a vertebra as having belonged to a mammal. In all mammals, except the Cetacea, the maximum number of phalanges in any one digit is limited to three ; in nearly all, the number of cervical vertebrae is neither more nor less than seven ; and the number of the lumbar vertebrae is never less than two. There are very rarely any verte- brae with unanchylosed ribs anteriorly to the first dorsal vertebrae. The jaws are ordinarily dentigerous, but teeth are never found elsewhere than upon the mandibular, maxillary, and intermaxillary bones ; the grinding teeth very frequently have more than a single root or fang, a method of implantation never observed in any other class. The most distinctive character of the Rodent order is the possession of the pairs of scalpriform incisors in the upper and lower jaws, from which their name is taken. There is a single pair of incisors in the upper jaw in all Rodents, except the Leporidae, which are hence called ' Duplicidentati] as having two pairs placed one behind the other, the hinder pair being the smaller. In the lower jaw there is a single pair only in all living Rodents, without exception. The upper incisors form a larger segment of a smaller circle, the lower a smaller segment of a larger circle. The peculiarities of their growth, which goes on uninterruptedly during the life of the creature from a persistent pulp, and of their functions, entail changes of great im- portance in the conformation both of the skull and of particular bones. The intermaxillaries, in relation with which the upper incisors are first developed, and which form a great part of their permanent sockets, are larger in relation to the rest of the skull and of the animal than in perhaps any other mammals ; — they form the whole, or nearly the whole, of the sides and under surface of the bony snout, and in all living Rodents, as in the Elephants, they interpose between the nasals and the maxillaries, whilst failing themselves to reach the lacrymals. The maxillary bone gives origin on the concave surface of its malar process to a large part of the masseter muscle, but a more deeply placed part of the muscle passes behind or inside of that process and takes origin from the sides of the snout. The presence of this deep head to the masseter is peculiar to but by no means constant in Rodents, varying with the infra- or ant-orbitally placed canal through which it passes. It co-operates, by passing round the back of the malar 6 DESCRIPTIONS OF PREPARATIONS. process of the maxillary as round a pulley to an insertion just below the socket of the mandibular premolar, very strongly with the temporal muscle in moving the lower jaw in a vertical direction, and bringing its incisors into play upon those of the upper jaw ; whence probably the inverse ratio which has been observed to obtain between the temporal and the antorbital fossae is to be accounted for. The masseter muscle arises from nearly the whole length of the malar arch, which is made up ordinarily of the malar process of the maxillary, of the malar bone, and of the malar process of the squamosal, and sometimes of the lacrymal also. It is by the contraction of those of its fibres which pass backwards on to the posterior edge of the lower jaw, aided by that of the pterygoids, that the anteroposterior move- ment of the lower jaw with its molar series upon that of the upper jaw is effected. The glenoid cavity has, to allow of this movement, its long axis running anteroposteriorly as in all Rodents except Leporidae, and as in the Mesotheriidae ; whilst the unbroken molar series and the absence of canines are characteristic of the entire order without any exception. Some involu- tion of the angle of the lower jaw, which resembles that observable in the Marsupialia, and the considerable size of this portion of the bone are points worthy of note as being present in many Rodents. Though the malar arch has a downward, rather than, as in Carnivora, an outward curve, still the interzygomatic diameter is in all Rodents the widest transverse cranial diameter. The temporal is never separated from the orbital fossa ; the cranial cavity is always much compressed from side to side on a level with the optic foramina, so as frequently to leave an interorbital fenestra by the fusion of the two foramina into one, at a point a little behind that at which the olfactory chamber succeeds the cerebral internally. The length of the tail and the number of the caudal vertebrae vary much within the limits of this order, just as the external concha of the ear and the characters of the integumentary system do. But, in spite of the very various special habits of the animals belonging to this order, the two pairs of limbs almost invariably present the same ratio of development inter se, the hind limbs being the stronger and longer pair. The tibia and fibula are anchylosed here and in Leporidae, but not in Sciuromorphi nor Hystri- comorphi. There is, however, little tendency to anchylosis in the skeleton of the Rodents ; in this specimen the posterior pair of sacral vertebrae are not anchylosed with the anterior, with which the ilia articulate, and the mandibular bones never throughout the order anchylose, as they do in Pro- boscidae, Suidae, and Perissodactyla, at the symphysis, in spite of the great afflux of blood which their permanently growing incisors bring into them. In the trunk we observe that the spines of the dorsal vertebrae, from the largely developed spine of the second dorsal to that of the tenth inclusively, point backwards, whilst those of the six lumbar vertebrae and of the two last, the thirteenth and the twelfth, dorsal, point forward towards SKELETON OF COMMON RAT. 7 the vertical spine of the eleventh dorsal, which has been called in conse- quence the 'anticlinal' vertebra. The anterior dorsal vertebrae diminish progressively in size as they are placed nearer to this vertebra, whilst the vertebrae placed posteriorly to it, and markedly the transverse processes of the lumbar vertebrae, increase in size as we pass backwards from it towards the sacrum. Well-marked and distinct anapophyses and metapophyses are developed on the anticlinal vertebra, and are to be seen on the succeeding vertebrae nearly or quite up to the sacrum. The direction of its spine relatively to those of the other vertebrae in front of and behind it, causes it to be the point of greatest mobility in the trunk. Points of less striking proportions, but more or less distinctive of, and universal in, the order are presented in the skull by the presence of an interparietal bone ; by a vacuity in the skull walls for the blood to pass out from the lateral sinus, either as here by a conjugate foramen between the squamosal and the periotic, or by a foramen in the squamosal itself, the so-called ' canalis temporalis ;' by the development of the post-auditory process of the squamosal into a lamina of bone, which may reach as far back as the occipital, but serves always to keep the tympano-periotic, with which it never anchyloses, in place ; and, finally, by the smallness of the angle formed by a line drawn from the posterior edge of the supraoccipital on to the basicranial line. The depth of the symphysis pubis, and the oblique forward direction of the transverse processes in the lumbar region, are points probably correlated functionally with the strength of the hind limbs. The large size of the abdominal rela- tively to the thoracic cavity may be connected with the multiparous character of the order generally. The spine of the second dorsal vertebra has a small ossicle articulated to its apex, and pointing forward, much as in the long- necked grazing mammals the ligamentum nuchae is placed along the dorsal and cervical regions. The two first cervical vertebrae are, as is usual in mammals, much the largest in the series, and they contrast, as in all pla- cental mammals, with the other cervical and also with all the other moveable vertebrae, in having, when adult, the centre of the first fused with that of the second, and in being connected with each other and the skull by cartilages and synovial membranes without fibro-cartilaginous discs. The first rib has its head articulated to the bodies, and its tubercle to the trans- verse processes of both the last cervical and the first dorsal vertebra. There are two lateral episternal bones between the first of the six sternal bones, the so-called * manubrium/ and the clavicle, one on each side, but there is no central cervical prolongation of the sternum as in Lepus. In the carpus there is the same number of bones as in that of man, for though the scaphoid and lunar are fused into one bone, the scapho-lunar, as they are also in Carnivora and Chiroptera, a bone, the os centrale, exists between it and the os trapezium, os trapezoides, and os magnum in the second row of carpals, which is not represented by a distinct bone in the 8 DESCRIPTIONS OF PREPARATIONS. human carpus, nor in those of Ungulata, Cetacea, Chiroptera, Edentata, Marsupialia, and Monotremata, but only in those of Rodentia, Insectivora, and Simiadae exclusively of the Chimpanzees. As in all mammals, though in no amphibian a single bone, the os unciforme, supports the two outer metacarpals. In this enumeration the ulnar sesamoid bone, or 'os pisiforme,' is not reckoned as a carpal bone, nor any bone of similar function in con- nection with the tendons on the volar side of the hand. In Rodentia we find two more bones in the tarsus than in the human subject, the os scaphoid es being double, and an accessory bone present on the inner side of the inner os cuneiforme. For general anatomy, Krause, Die Anatomic des Kaninchens, Leipzig, 2nd ed., 1884; and T. J. Parker, Zootomy, London, 1884, p. 301. For an account of the position of the anticlinal vertebra in the order Rodentia and elsewhere, see Giebel, Beitrage zur Osteologie der Nagethiere, 1857, p. 35, or Abhandl., Nat. Verein fur Sachsen und Thuringen, i. p. 223, ibique ritata ; or Die Saugethiere, 2nd ed. 1859, p. 6. For the general characteristics of Mammalian vertebrae, see Professor Owen, Descriptive Catalogue of the Osteological Series of the Royal College of Surgeons, vol. i. pp. 7, 8, 1853. For the nomenclature of the several elements of a vertebra, ibid. p. xliv. For the Pro-atlas of Amniota, see Albrecht. Z. A. iii. 1880 ; and Bull. Mus. Roy. d'Hist. Nat. Belg. ii. 1883. For the Osteology of the Rodentia, see Cuvier's Ossemens Fossiles, 2nd ed., 1823, vol. v. pt. i. pp. 4, 14, 44. Waterhouse, Mag. Nat. Hist., N. S., vol. iii. 1839, A. N. H. 1841, 1842, takes the conformation of the lower jaw and of the anterior portion of the zygomatic arch as furnishing a basis for classifying the order in the three sections, Murina, Hystricina, and Leporina. In Johnston's Physical Atlas, 1856, ad¥\. 28, he divides the order into four families, Muridae, Sciuridae, Hystricidae, and Leporidae. See H. N. Turner, P. Z. S. 1848, p. 63. J. F. Brandt, Untersuchungen iiber d. craniologisch. Entwickelungstufen der Nager d. Jetztzeit, M&noires de 1' Academic Imp. des Sciences de Saint Petersbourg, ser. vi. torn. vii. pp. 127-336, 1855, laying weight (p. 141) on the form and general contour of the brain-case, the characters of the base of the cranium, the direction of the pterygoid processes, the conformation of the palate with the foramina ma'siva, and of the ossa lacrymalia, divides the order Rodents s. Glires into four suborders, Myomorphi, Sciuromorphi, Hystricomorphi, and Lagomorphi. For later views, see E. R. Alston, P. Z. S. 1875, 1876, and infra, pp. 43-45. For the resemblances between the skeleton of the Rodents and that of the Elephant, see Cuvier, /. c. i. pp. 10-12. For the differences between the skeleton of the Rodents and that of the Aye Aye, Chiromys Madagascar iensis, see De Blainville, Osteographie, Fasc. iii. 1841 ; Professor Owen, Tr. Z. S. v. 1863, pp. 79-83; Professor Peters, Berlin Abhandl. for 1865, pp. 89-92. For the microscopic characters of the teeth of the Rodents, and their classi- ficatory value, see J. Tomes, Ph. Tr. 1850, pp. 553-561, and C. S. Tomes, Manual of Dental Anatomy, pp. 332 and 339. For the resemblance of the SKELETON OF WILD RABBIT. 9 microscopic character of the molar enamel of all Rodents (except Leporidae and Hystricidae) to that of the Proboscideans, see Trans. Odont. Soc. vol. iii. p. 239, 1871. For the classificatory value of the ossicula auditus, see A. Doran, Tr. L. S. 1878, p. 418. For other characters of the order Rodentia, see Waterhouse, Natural History of the Mammalia, vol. ii. pp. 1-9, 1848; De Quatrefages, Considerations sur les caracteres zoologiques des Rongeurs, Paris, 1840 ; Milne-Edwards, Recherches sur les Mammiferes, i. 1868, pp. 29, 30. For those of the Myomorphi, see Osteological Catalogue, Royal College of Surgeons, vol. ii. Preparations 2223-2245 ; Waterhouse, Mag. Nat. Hist., /. c. p. 92; Brandt, /. c. pp. 152, 156, 300; Peters, Monatsber. Ak. Berlin, 1867. For the Carpus and Tarsus and Shoulder-girdle, see Gegenbaur's Untersuch- ungen zur Vergleichenden Anatomic, Hft. i. ii. 1864, 1865 ; for the shoulder-girdle and sternum, see also Gotte, A. M. A. xiv. 1877 ; Ruge, M. J. vi. 1880; Hoffman, Niederland. Archiv fiir Zool. v. 1879-82. For the carpus, see Leboucq, Arch, de Biol. v. 1884; carpus and tarsus, Baur, Z. A. viii. 1885, For the ' Canalis Temporalis,' or ' Foramen jugulare spurium,' see Otto, Nova Acta, xiii. pt. i. p. 27 ; Luschka, Dk. Wien. Akad. xx. 1862, p. 204; and Kolliker, Entwickelungsgeschichte, p. 929, 1879. For the means whereby the vertebral centra are articulated in the different classes of vertebrata, see Rathke, Entwickelungsgeschichte der Wirbelthiere, mit einem Vorwort von A. Kolliker, 1861, p. 130. 3. SKELETON OF WILD RABBIT (Lepus cuniculus, var.fera). THE skeleton of the Rabbit differs from that of the Rat and many though not all other Glires Myomorphi, not merely in such points as its larger absolute size, the incompleteness x of its clavicles, the absence save in rudiment of a hallux, the unguiculate character of its pollex, the number and rootlessness of its molars, and the smaller number of its caudal verte- brae, but in many points of greater morphological importance than any of these. Some of these latter points show that the suborder Lagomorphi is more closely allied than the Myomorphi to certain lower forms of Verte- brata ; others indicate more clearly than is seen in the Myomorphi that a certain affinity exists between the Rodentia s. Glires and the large Ungu- 1 The older zoologists (e.g. Fischer, Synopsis Mammalium, 1829, pp. 286, 366; Catalogue of the Royal College of Surgeons, Part iii. 1831, pp. 79, 87) divided the order into the two sections of Glires claviciilis completis, saepe validissimis , and Glires daviculis nullis aut imperfectis. The inadequacy of this basis of classification may be judged of by the fact that the tail-less Hares (Lagomyes\ which form a subfamily, as shown by Pallas, Glires, p. 28, closely allied to the true Hares, have complete clavicles. In the Rabbit no trace of the clavicle is visible at birth (see Parker, Shoulder-Girdle, PL xxv. Figs, i, 2, pp. 207-210; Flower, Osteology of Mammalia, p. 229), though it becomes developed before adult life. In the human subject, on the other hand, the clavicle ossifies before any other bone in the developing foetus. 10 DESCRIPTIONS OF PREPARATIONS. late Mammalia ; whilst others may have their connecting character expressed by saying that the distinctive peculiarities of the Rodent type are not so sharply pronounced in this as in the other suborder already mentioned, or indeed in either of the two other suborders of Glires, the Hystricomorphi and the Sciuromorphi. Among the last of these three sets of peculiarities may be mentioned the shape of the articular surface furnished by the squamous bone for the lower jaw. This surface is transversely, not, as usually in Rodents, antero- posteriorly elongated ; and it permits consequently of a much greater lateral movement of the jaw, correlated with which we find the molars above and below not with horizontal but with concave and alternately sloping grinding surfaces. The presence of six incisors in the upper jaw of young, and of four in that of adult Lagomorphi^ is a third, the smaller size of the sockets for the lower incisors a fourth, point indicating less specialization in this suborder. In the relatively small extent to which the temporal muscle is de- veloped, in the great extent to which the lower jaw is developed behind the plane of its articular process, in the presence of a diastema between the anterior scalpriform teeth and the molar series, and in the keel-shaped prae- sternum prolonged into the cervical region as a ' proosteon,' the Lagomorphi and most if not all other Rodents resemble many Ungulata, both Artio- dactyle (such as Sus) and Perissodactyle (Equus and Tapirns). In the reduction of the independence and importance of the fibula the Lagomorphi and the Myomorphi resemble each other and many or most Ungulata, and differ from all other Rodents, with some apparent exceptions, e. g. Pteromys and Castor. In the length and slenderness of a process given off by the squamosal posteriorly to the articular surface furnished by it to the lower jaw, which process not being anchylosed, as in many other mammals, to the tympano-periotic, nevertheless clamps it into fixed relations with the other skull-bones adjacent to it, the skulls of the Lagomorphi and many other Glires resemble those of some Perissodactyle Ungulata, whilst the presence of a third femoral condyle, and of an internal alisphenoid canal for the external carotid artery, are points in which they strikingly resemble all living Perissodactyla. On the other hand, a curious illustration of the com- bination in these Rodents of peculiarities which become separated in other divisions of the class Mammalia is furnished to us by the ischium of the Lagomorphi) which closely resembles the ischium both of the Ruminant and non-Ruminant Artiodactyles in what is considered to be a distinctive pecu- liarity of at least the latter of these two divisions of animals, viz. in the presence on the outer side of the bone a little way in front of its upper and posterior angle of a well-marked outstanding forwardly -curving process of bone. The exposure in the dry skull of the turbinated bones in the nasal cavity by the deficient ossification of the lateral walls of that chamber is SKELETON OF WILD RABBIT. II another point on which weight may be laid as connecting the Lagomorphi with some at least of the true Ruminants, e. g. Capra and Cervus, As points of degradation in the Lagomorphi as compared with higher mammals we may note in the Rabbit and Hare the absence or great retardation of any anchylosis to each other of the basicranial bones, the sutures between the basioccipital and basisphenoid and between the basi- sphenoid and presphenoid remaining open not only when the occipital and interparietal bones are fused, but even after these bones have become abundantly fenestrated by senile absorption l ; the vertical and transverse perforations in the basisphenoid communicating with the pituitary fossa ; the small antero-posterior length of the palatal plates of the palatine and maxillary bones leaving the stalked leaf-shaped end of the vomer exposed behind them, and the anterior end of the same bone exposed in front of them, when the dry skull is looked at along its base-line ; the development of the ' foramina incisiva ' into wide fissures continuous with the latter of the two sets of vacuities just spoken of ; the persistence of open fontanelles in the occipital bone, in the interspace between that bone, the squamous, and the tympano-periotic, in the space, that is, which corresponds to the ' asterion ' of Professor Broca, and in the interspace between the two last- named bones below the backwardly-running bar of the squamous ; and probably also the singular fenestration or vacuolation of the anterior and upper part of the maxillaries. To these points, dependent upon a defi- ciency of ossification, may be added the involution of the angle of the lower jaw, which represents, though but rudimentarily, the inversion of that part of the jaw in the Marsupials ; and the fusion of the optic foramina into a single mesial foramen bounded inferiorly by the presphenoid much as in Birds. Other points worthy of note in the Rabbit's skeleton, either as com- pared with those of most Rodents of other suborders, or as compared with those of other mammals, are presented to us in the imperfect differentiation of the coronoid from the ascending ramus of the lower jaw; in the approach to horizontality in the symphysis of that bone ; in the large size and back- ward direction of the tympanic process of the tympano-periotic ; in the small size of the infra-orbital canal and of the anterior part of the malar process of the maxillary, and the large size of the free backwardly-project- ing process of the malar bone proper; in the presence of large supra-orbital processes attached in the middle and projecting freely at either end of their length ; in the fixed attachment of the upper lamellae of the ethmoid to 1 The persistence of patency in the sutures of the basis cranii appears to possess considerable morphological value, but this cannot be said of the sutures of the roof of the skull. For example, the interparietal anchyloses very early in the Subungulate Hystricomorphi, but it does the like also in Sciurus, whilst it remains distinct for a long while in Myoxus, Castor, and the Murini, as well as in the Lagomorphi. I a DESCRIPTIONS OF PREPARATIONS. the nasal, constituting ' naso-turbinal ' bones ; the loose often lost attach- ment of the much convoluted inferior turbinals to the maxillaries l ; in the formation by the posteriorly expanded vomer of a floor to the true olfactory portion and of a roof to the lower narial or respiratory portion of the nasal cavity ; in the wide interval between the pterygoids and the tympanic bulla, and in the presence in this last bone of a canalis caroticus. The pitted appearance of the interparietal bone, of the upper part of the occipital bone in apposition or fused with the interparietal, and of the upper arch of the first cervical vertebra, is worthy of note as suggesting a comparison with Lophiomys, and with the rugose 2 zygomatic and frontal bones in the Paca (Coelogenys pacd). In aged specimens this pitting almost amounts to fene- stration. The anchylosed tibia and fibula are specially noteworthy. The imperfection of the clavicles in the Rabbit and Hare prepares us for their entire absence as reported to exist in some few of the Subungulate Hystricomorphiy and contrasts with their complete development in some families of the same suborder (Chinchilloides, Spalacopodoides), as also in the entire suborders Myomorphi less Lophiomys ', and Sciuromorphi, and even in the subgenus of the suborder Lagomorphi represented by the tail-less Hares, Lagomys. The presence of a backwardly and downwardly projecting pro- cess of the acromion is similarly a peculiarity observable in the Lagomorphi and certain Hystricomorphi, whilst it is absent in many Rodents, though present in other orders of Mammalia, and notably in the Elephant. The non-development of fangs confers the same privilege of perpetual growth on the molars of the Lagomorphi, the true Cavies, and the Chinchillas, which is enjoyed by the incisors of all Rodents. The white colour and the shortness of the incisors again are points of similarity between the Hares and the Cavies. The number of molar teeth is greater in the Hare and Rabbit than in any other Rodents, being $ as against -£ in the allied sub- family Lagomys, against % in Hystricomorphi and Sciuromorphi, f in Myo- morphi, and f in Hydromys. The vacuity in the lower jaw, posteriorly to the socket for the last molar tooth, and the vertical upgrowths from the tubercles of the second to the eighth pair of ribs, are peculiarities in the skeleton of the Lagomorphi. In the ossa ilii the glutaeal surfaces are much more extensive than the iliac, a line drawn forwards from the tubercle for the short head of the 1 The maxillo-turbinals are more complexly and finely convoluted in the Rabbit than in the Hare, the subterranean habits of the former of these animals creating a greater need for warming the inspired air. Similarly, as remarked by Professor Flower (Osteology of Mammalia, p. 183, second edition, 1876), in the Elephant, where the inspired air is sufficiently warmed by having to pass along the elongated proboscis, the maxillo-turbinals are wholly aborted. 2 See Flower Osteology, p. 156; and Waterhouse, History of Mammalia, ii. p. 369, who suggests that periodical deposition analogous to that of the horns of deer causes this. Compare the strange account of Lepores cornuti given by Schreber, Saugethiere,i. Taf. cclxxxiii. B; Pallas, Novae species Glirium, p. 14, ibique citata. VERTEBRAE OF RABBIT. 13 rectus just above the acetabulum along a faintly-marked and rounded ridge represents the acetabular border of other Rodents, such as the Beaver, and shows the limit of the respective surfaces. The incisura acetabuli through which the blood-vessels and nerves enter for the supply of the hip-joint is reduced in size, and the rim of the acetabulum is interrupted only by a linear fissure. The symphysis of the pubis is deep. The skull of the tame Rabbit differs from that of the wild in having the roof of its brain-containing portion much flatter as measured either from before backwards or from side to side than is the case in the very distinctly arched calvaria of the wild variety. The lateral boundaries of the same cavity as constituted by the squamous are much more wall-sided than in the wild race, and instead of curving gradually into a vaulted vertex they are defined or delimited off from it by largely-developed anteroposte- riorly running ridges. The height of the occipital foramen is less relatively to its breadth, its upper and lower borders not being emarginated into secondary curves as in the wild variety. The length of the entire skull is considerably greater relatively to its breadth, though not relatively to the size of the entire body. This may be double that of a wild specimen, whilst the absolute breadth of the skull may be identical in the two subjects of comparison, and the absolute length may be less than 30 per cent, greater in the tame than in the wild variety. The lines and processes of the cranium and lower jaw are less sharply defined and sculptured than in the wild variety, and the surface of the cranial bones generally is inferior in gloss and polish. The same applies to the bones of the trunk and limbs in many domestic animals as compared with animals of the same species in a wild state, and indeed is usually more clearly appreciable than in the case of the two varieties here compared with each other. For the possession of rootless molars by other Rodents (Octodon, Capromys, most Arvicolae] • for that of molars with short roots or with roots incomplete or late to be developed, by the Agouti, by the Paca s. Spotted Cavy, by the Beaver and the Porcupine ; for that of rooted molars by the true Mice and the Squirrels, see Owen, Odontography, p. 401 ; and by Leporidae in their milk dentition, see Hilgendorf, Monatsber. Ak. Wiss., Berlin, 1876, p. 673. For the presence in Leporidae of a perfect investment instead of, as in all other living Rodents, merely an anteriorly placed plate of enamel on the incisors, see Hilgendorf, /. c. But preparations made by Mr. C. S. Tomes suggest that this perfect investment exists only in the enamel membrane of the developing tooth. For numerous other points of similarity between the Hares and the Cavies, see Waterhouse, History of the Mammalia, ii. p. 156, 208; Buffon, cit. Pallas, I.e. p. 29. For the Shoulder-Girdle of the Rodents, see Parker, Shoulder-Girdle, 1868, pp. 207-210; and for the mesial prolongation of the praesternum in Lepus and 14 DESCRIPTIONS OF PREPARATIONS. Cavia into the cervical region, see Plates xxiv and xxv ; and for the same pro- longation in Ungulata, see PI. xxix. /. c. For a detailed comparison of the osteological differences between tame and wild Rabbits, see Darwin, History of Plants and Animals under Domestication, i. pp. 120-136, 2nd ed. 1875. 4. CERVICAL, DORSAL, LUMBAR, SACRAL, AND CAUDAL VERTEBRAE OF RABBIT (Lepus cuniculus), GREAT mobility is secured by the particular arrangements observable in the region where the two upper cervical vertebrae articulate with each other and with the skull, and in the region of the lower dorsal and upper lumbar vertebrae. On the other hand, the transverse processes of the lower cervical vertebrae and the imbricated neural spines of the upper dorsal ver- tebrae prevent the possibility of any great range of movement between any two of the constituent segments of those portions of the spinal column. The cervical vertebrae are seven in number, as almost invariably in the Mammalian class ; the numbers of the dorsal and lumbar series are variable, but twelve and seven, the numbers of the dorsal and lumbar ver- tebrae respectively in the Rabbit, are very common numbers for those series throughout the class. The number of the caudal vertebrae is the most variable, that of the lumbar next, that of the dorsal less than that of the lumbar, that of the cervical the least variable of these four sets of ver- tebrae. As the number of the cervical vertebrae is all but invariable, the variability of the length of the cervical region depends upon variations in the length of the bodies of the seven vertebrae. The first cervical vertebra or * atlas ' is the widest from side to side of all the neck vertebrae ; it has a low but broad neural arch, and superadded to it in front a smaller arch which is in the perfect condition of the parts made into a ring for the reception of the * odontoid process ' of the next vertebra by a transverse ligament. Its neural arch is overhung by the spine of that vertebra, and it does not give any point of attachment to the ligamentum nuchae. It con- tains two more or less separated canals for segments of the vertebral artery ; one of them pierces the base of its broad ' transverse process ' from behind forwards, the other turns more or less horizontally from without, inwards, behind and below the articular processes. This latter canal may be repre- sented merely by a groove in the Rat, and ordinarily has this imperfect character in the human subject. The former has generally a short horizontal canal leading forward from it and opening on the anterior surface of the transverse process ; it is however absent in the Leporidae, though present in the Rat and many or most other Rodents. The second cervical or ' axis ' vertebra has its spine greatly developed, both anteroposteriorly and verti- cally, giving attachment by it both to the muscles which move, and the VERTEBRAE OF RABBIT. 15 elastic ligamentum nuchae which supports, the head. It has no anterior articulating processes upon its neural arch in mammals, but it comes into ar- ticular relation with the atlas by means of two oblique zygapophysial surfaces developed on either side of the base and a third on the front of its odontoid process, which is the backwardly displaced and anchylosed centrum of that vertebra. It is the deepest from above downwards, and the longest from before backwards, but also the narrowest from side to side of the cervical series. The first two cervical vertebrae articulate with each other and with the occiput by means of synovial joints as the neurapophysial processes are articulated to each other throughout the rest of the trunk, where however the centra are connected by interarticular fibrocartilaginous discs containing in their central pulp remnants of the primitive chorda dorsalis. The neural spines of the third and fourth cervical vertebrae are low but long, corre- sponding with the long neural roof which these two vertebrae possess ; the spines of the shorter neural arches of the fifth, sixth, and seventh vertebrae have more of the shape which their name implies. The lateral processes or < cervical ribs ' of these vertebrae are greatly developed ; those of the atlas more or less obliquely outwards, those of the axis backwards ; those of the third, fourth, fifth, and sixth, both anteriorly and posteriorly, and those of the seventh outwardly. The fourth, fifth, sixth, and seventh have prominent upgrowths developed on this process or rib which are homologous apparently with the prominent tubercles of the ribs of these creatures, or, possibly, with the metapophyses of the trunk vertebrae. This process makes up by itself almost the whole of the transverse process of the seventh cervical vertebra, the inferior, antero-posteriorly-produced, process, which is much larger in the preceding vertebrae and largest of all in the one immediately preceding, being lost in this, the last of the series. These inferior elements of the transverse processes, by bending inwards form with the vertebral bodies furrows, in which the long anterior neck muscles are lodged, a central slightly-raised line marking the line of separation of these muscles and representing the homologously-placed hypapophyses of certain lower verte- brata. The segment of bone which completes the ring of the atlas anteriorly is homologous with these hypapophysial downgrowths. The last cervical vertebra in the Rabbit has not, as it has in the Rat, any connection with the tubercle of the first dorsal rib. Eight of the dorsal vertebrae, from the second to the ninth inclusively, have, each, two half facets on their centra, the first has one whole facet anteriorly and a half facet posteriorly, and the tenth, eleventh, and twelfth have, each of them, single whole facets placed on the anterior superior angle of the lateral aspect of their centra, for articulation with the heads of the ribs. The neural spines of the dorsal vertebrae are largely developed, their apices from the second to the ninth showing a tendency to become bifid antero- posteriorly. The tenth is the anticlinal vertebra (for which, see p. 8, supra}, 1 6 DESCRIPTIONS OF PREPARATIONS. and upon it and upon each succeeding vertebra down to the sacrum a large, as upon the ninth and eighth dorsal vertebra a small, metapophysis is developed. A small anapophysis is also seen to take origin from the base of its neural arch, and to be possessed by each succeeding vertebra up to the antepenultimate lumbar. Several of the anterior, as also of the pos- terior dorsal vertebrae, have low hypapophysial ridges developed subcen- trally ; and longer ones possessing the character of spines are developed on the three anterior lumbar vertebrae, in relation in the living animal with the crura of the diaphragm. The Hedgehog, Erinaceus europaeus, and the Mole, Talpa europea, have paired unanchylosed ossicles developed intervertebrally in the same region, like caudal chevron-bones. The lumbar vertebrae as wholes, and also most of their processes, increase in size from before back- wards as far as the penultimate one ; the transverse processes point obliquely forward, but form a more open angle with the long axis of the column than they do in the Rat. Behind the lumbar vertebrae we have, though not invariably, four vertebrae united to each other by anchylosis of their centra, their transverse, and their articular processes ; and united to a fifth vertebra by anchylosis of the lateral processes. These five vertebrae may be taken as corresponding to the os sacrum of anthropotomy. The two most anteriorly placed of these five vertebrae form by their transverse processes a pouched-shaped or auricular articular surface for the ilium, the posteriorly placed convex end of which is constituted by the transverse process of the second and the two sides by that of the first. In the Beaver the second vertebra contributes a relatively much smaller proportion to this articular surface, and in the Rat and many other Rodents it scarcely con- tributes anything. The third and fourth of these post-lumbar vertebrae do not in any Rodent furnish any articular surface to the ilium. The four or five anterior caudal vertebrae have largely developed sub- quadrate transverse processes, with their free angles, both anterior and posterior, somewhat produced. From eight to ten more rudimentary vertebrae follow upon these, the most posteriorly placed being merely bars of bone, with dilated ends corresponding to the articular aspects of the centra of other vertebrae. The caudal vertebrae of the Rabbit have no chevron-bones as have those of the the long-tailed Rodents, and of many other such animals from the Ichthyosauri to the Primates, with the excep- tion of the Ungulata and the Proboscidea, which are allied in so many other points to Rodents. UPPER HALF OF RABBIT. 17 5. UPPER HALF OF RABBIT (Lepus cuniculus), Dissected so as to show some of the muscles of the head, neck, shoulder, and fore-leg *. The letters and the description correspond with those on the figure annexed. With Figure I . THE skin has been removed from the front of the lower jaw backwards, the sheet of cutaneous muscle, x, covering the region of the neck and known in anthropotomy as the panniculus carnosus or platysma myoides,\\aj$> been entirely removed on the right side, as also from the region of the thorax on both sides ; the incomplete clavicle with the muscles in con- nection with it has been separated from its ligamentous union with the sternum and displaced to the right so as to show the subjacent nerves, phrenic and brachial, on the same side ; below the level of the clavicle, portions of the pectoral, c, d, and the sterno-clavicular muscles, g, have been dissected and turned back. On the left side, the anterior part of the panniculus carnosus has been left with its two fixed insertions, x and xf , into the lower jaw, and the other, x"' ', into the sternum, intact ; the clavicle is left in situ, but the muscles passing down to it from the head have been cut away behind the line of the cutaneous muscle to show the sterno-scapular muscles, h and /, in connection with the clavicle, at j ; and the serratus magnus, o. The deep cervical fascia (j3) is seen to the left of the middle line of the upper half of the neck and of the intermandibular space, at a lower level than the panniculus carnosus. It has been removed anteriorly to show the insertion of the digastric muscle. On a level with the hyoid the deep cervical fascia is raised into a convexity (a') by the submaxillary gland underlying it, and it furnishes, externally and posteriorly to this area, a capsule to the parotid. This gland, however, being less convex than the submaxillary, does not cause the sheet of fascia to bulge upwards. On the opposite side this fascia has been removed in the intermandibular space ; and the submaxillary gland being raised from its bed is seen to send a duct in towards the mouth in an interspace between the internal pterygoid and digastric muscles. This latter muscle is in the Rabbit, as in many other mammals, monogastric, its posterior belly being represented by a tendon, which however pierces the stylo-hyoid just as in man. Its muscular portion is inserted into the lower jaw on either side of the symphysis, and from the same portion of the jaw a depressor ; w, of the lip passes forward. On the 1 It is for various reasons advisable that the student should proceed to the preparations illustrating the splanchnology of the Rabbit before addressing himself to this somewhat complex dissection. And it will be found advantageous to immerse the upper half of the body of the animal, the heart and lungs having been removed, in spirit 10° over proof (sp. gr.° 910) for three or four days before following out the detailed anatomy of the muscles here described and figured. For making special dissections of the nerves it will be found useful to acidulate the spirit with dilute nitric acid in the proportion of one part in forty. This treatment facilitates the mechanical process of dissection in several ways, and makes it less easy to overlook the more delicate structures concerned. C 1 8 DESCRIPTIONS OF PREPARATIONS. outer side of each of these depressor muscles the panniculus carnosus, x, x , x", takes a fixed insertion by a band of muscular fibres, which on the right side are cut short, and on the left are seen to be continuous with the rest of the muscle. In the interval between this strip of muscle and another which passes from the platysma to distribute itself in the region of the mouth, the buccinator muscle, j/, comes into view. The lobes of a somewhat variable buccal gland have been removed to show these three muscles in this region. The tendon of the stylo-hyoid forms an oblique angle with its mus- cular belly, s, being connected with the greater cornu of the hyoid, along which it passes to be attached to the body of the bone, /. To the muscular part of the stylo-hyoid two nerves, from the portio dura of the seventh pair with which the glosso-pharyngeal portion of the eighth pair of cranial nerves is connected, may be seen in dissection under a lens, though not in this figure, to distribute themselves. A semilunar space has been formed in the right axilla, between 8 and y, by separating the latis- simus dorsi, y, as it narrows up to its insertion on the inner side of the humerus, from the similarly narrowing panniculus carnosus, 8, of the regions of the flanks and back, the ' dermo-humerien ' of Cuvier (Anat. Comp. iii. p. 597), the ' costo-alaris ' of Humphry (Observations in My- ology, p. 131), which passes in front of it to be inserted into the humerus together with the tendon of the pectoralis major. The insertion of the homologous muscle in Birds is shown in the figure of the Dissection of the Pigeon in the shape of a tendinous slip attached to the tendon of the great depressor pectoral muscle, x. From the tendon * of the latissimus dor si of the left side a slip of muscle, e, the ' latissimo-condyloidetis ' of Bischoff1, the ' dorso-epitrochlien ' of Duvernoy (Archiv. du Museum, viii. p. 80), is seen to take origin and pass down at right angles to that tendon to an insertion on the inner side of the olecranic process of the ulna. On either side and behind the latissimo-condyloideus are to be seen the three heads of the triceps, with which muscle the latissimo-condyloidetts is frequently fused. Externally and anteriorly is seen the biceps, 0, here, like the similarly misnamed digastric, but a monogastric muscle. From its anterior surface, one band of fascia passes off to connect itself with the fascia enveloping the muscles of the fore-arm, another leaves the tendon of the muscle at a lower level and connects itself with the radius and the tendon of the pronator radii teres, thereby setting up a secondary connection with the radius with which it is principally connected in man, whilst its principal insertion is made into a well-marked depression just above the inner and inferior edge of the olecranic process of the ulna by a broad and flat tendon. The 1 Bischoff gave this muscle the convenient name of latissimo-condyloideus, and was followed by Dr. H. C. Chapman, in his Memoir on the Structure of the Gorilla, Proc. Acad. Sci. Phila- delphia, 1878. UPPER HALF OF RABBIT. 19 muscular bellies of the pronator and the flexor muscles arising from the ulna and internal condyle of the humerus have Been cut through and reflected to show this insertion of the biceps. This dissection contrasts with a similar dissection of the human subject in the imperfection of the clavicles ; in the absence or rudimentary con- dition of the omo-hyoid', in the presence in the neck of two additional muscles, the acromio-basilar, n, and the cleido-occipital, m ; in the formation of a compound * cephalo-humeral ' muscle, as in many other mammalia, by the physiological combination of the cleido-mastoid, k, the cleido-occipital r, m, and the acromio-basilar, n, with the deltoid, f\ in the prolongation of the sterno-clavicidar, g, and ster no -scapular muscles, h and i, the homo- logues of the subclavius, on to the spine of the scapula at /, whence they act in the way of slinging up the horizontally-carried trunk ; and, finally, in the greater size of the cervical platysma, x, and the development of a cuta- neous muscle, 8, the ' dermohumdrien* s, ' costoalaris] s, ' brachiolateraV, in the regions of the back and flanks. The cleido-occipital, k, the acromio- basilar •, n, and the ' latissimo-condyloideus', e, are more or less frequently represented by muscular varieties occurring in the human subject, and the absence of any scalenus anticus in the Rabbit is paralleled in man by the occasional perforation of the fibres of that muscle by the upper factors of the brachial plexus, one of which, the fifth cervical, has been observed to pass entirely in front of the muscle. The digastric, M, the biceps, /, and the pectoralis minor, e, present points of difference, stated above, in which the Leporidae and very many other mammals coincide with each other and differ from man. The rectus abdominis is very usually prolonged in mam- malia lower than man up to the second or first rib, and in the rabbit up even to the base of the manubrium ; when thus prolonged, it is known as the ' rectus thoracis! It is frequently crossed at its upper end, as here, by a muscle known as the ' sterno-costalis', passing downwards and covering the front of the rectus just as the external oblique does that of the rectus abdominis. It may be better however to speak of the sterno-costalis as being a lateral efflorescence of the rectus \ On the other hand, the arrangement and relations of the various struc- tures of the lowly organised mammal here figured are sufficiently similar to those of man to cast considerable light upon some even of the more intri- cate points of anthropotomy. Among them we may specify the occurrence of certain varieties in muscles, the relations of the deep cervical fascia, /3, and the insertions of the pectoral muscle, c and d. 1 For a discussion of the homologies of the rectus thoracis and the sterno-costalis, see Professor Turner, Journal of Anat. and Physiol. May, 1867, pp. 247-253; May, 1868, pp. 392-394; Wood, Ph. Tr. for 1870, pp. 110-112, vol. 160. These two muscles are not lettered in this figure, but are seen to form a triangle with part of the first rib for its base and with its apex covered by the sterno-clavicular muscle g. For a figure showing the rectus abdominis giving off a part of the pectoral, see Ecker, Anatomic des Frosches, 1864, p. 95. C 2 DESCRIPTIONS OF PREPARATIONS. FIG. i. — DISSECTION OF SUPERFICIAL MUSCLES IN ANTERIOR REGIONS OF HEAD, NECK, AND THORAX IN RABBIT (Lefus cuniculus] ; NEARLY NATURAL SIZE. UPPER HALF OF RABBIT. 21 a. Bifid upper lip concealing muffle. See p. 26, y', infra. b. Sterno-mastoid muscle, arising from the prolongation of the manubrium into the neck and in- serted into the mastoid. c. Greater pectoral muscle of left side, arising along nearly the entire length of the sternum from the manubrium anteriorly down nearly to the leaf-shaped cartilage ending it posteriorly. c '. Part of the origin of this muscle on the right side reflected. c". Pouch-like insertion of it into the humerus, the fibres from the lower part of the sternum forming the posterior, those from the upper the anterior portion of the pouch ; the former passing to the inner ; the latter to the outer tuberosity of the humerus, and the outwardly- looking pectoro- deltoid ridge. d. Smaller pectoral muscle arising from the outer side of the keel-shaped manubrium superficially to the upper fibres of the greater pectoral, inserted together with fibres of that muscle and of the clavicular portion of the deltoid into the fascia covering the biceps, and reaching by tendon nearly to the lower end of the humerus at d" \ d'. Tendon of origin of smaller pectoral of right side cut short. d". Tendon of insertion of smaller pectoral of same side. e. Third pectoral muscle, corresponding to the pectoralis minor of anthropotomy, as being similarly innervated, but inserted not into the coracoid, but into the head of the humerus, and only by a few fibres into the costo-coracoid membrane. / Deltoid, a bicipital muscle, its anterior head, supplied by the circumflex nerve, arising from the outer end of the clavicle and the fibrous tissues external to that insertion covering the shoulder- joint, and the posterior head arising from the apex of the acromion and the anterior edge of the metacromial process of the scapula. The external tuberosity of the humerus appears in the interval between the two muscular bellies. g. Origin of sterno-clavicular muscle, from the sternum down to the level of the sixth rib, from the cartilage of which it receives some fibres, at a deeper level than that of the origins of the three pectorals, but superficially to the plane of the rectus thoracis and of the sterno-costal muscle, which are seen between the origin of the sterno-clavicular muscle and the plexus of brachial nerves to form, with the first rib, a triangle with its apex pointing downwards and inwards. g'. Insertion of the sterno-clavicular muscle into the clavicle. The under fibres pass under the clavicle without being attached to it, to be inserted, together with fibres from the two muscles next to be named, into the spine of the scapula and the fascia covering the supraspinatus. h. Upper sterno-scapular muscle, continuous at its origin from the manubrium sterni with the sterno- clavicular muscle, which may therefore be considered a prolongation of this muscle. i. Lower sterno-scapular muscle, arising from cartilage of first rib, innervated from the same source as the two muscles, g and h, last mentioned, partially inserted like the sterno-clavicular, g, into the clavicle, but passing on with it and the upper sterno-scapular, h, to be inserted into the spine of the scapula and the fascia covering the infraspinatus muscle posteriorly. j. Prolongation of the three muscles just named onwards on the left side from beneath the clavicle to the insertion just specified ; the three next to be named, k, m, and n, having been removed to allow of this being seen. k. Cleido-mastoid muscle, arising from occipital bone externally to sterao-mastoid, b, and inserted into the greater part of the length of the bony clavicle. It is the homologue of the cleido- mastoid portion of the human sterno-cleido-mastoid, holding the same relation to the cervical plexus. There is no sterno-maxillary as in the Horse, and in the Hyrax. m. Cleido-occipital muscle, arising from the basioccipital just externally to the acromio-basilar, inserted into the outer end of the clavicle, and the head of the humerus, and becoming continuous between these two bony attachments with the fibres of the anterior portion of the deltoid. n. Acromio-basilar muscle of right side, arising from the basi-occipital and inserted into the meta- cromial process of the scapula just anteriorly to the insertion of the anterior part of the trapezius. In most of the lower mammals this muscle arises from the atlas ; in some from the axis also, thus coming to represent the upper digitations of the levator anguli scapulae in Primates. In Rodents and some Ungulata, ' by becoming amalgamated by longitudinal and lateral fusion with the recti capitis it may be attached to the lateral or basilar process of the occipital bone.' This is the muscle called 'levator claviculae' by Wood, pp. 88 and 94, 'acromio-basilar' by Vicq. d'Azyr, and ' acromio-trachelien ' by Cuvier, in whose Le9ons d' Anatomic Comparee, i. p. 271, ed. 2de, we find it thus spoken of : 'On le trouve dans tons les mammiferes, 1'homme excepte, ce 22 DESCRIPTIONS OF PREPARATIONS. qui semblerait prouver qu'ils est une des conditions de la station quadrupede.' This would apply more correctly to the prolongation of the muscles, g , h, i, on to the scapula, as shown on the left side at/ 5 ri. Metacromial insertions of acromio-basilar and trapezius. o. Anterior or cervical portion of serratus magnus of left side, exposed by the removal of the three last-named muscles. A muscular fascicle, which not rarely arises between the upper part of the origin, here seen, of the serratus and those of the scaleni, is not shown in this figure. It is a long and slender slip, and passes down vertically to join the thoracic portion of the serratus, and to be inserted with it into the posterior and inferior angles of the scapula, and appears, when com- pared with the omohyoid of the horse, to represent that muscle. /. Sterno-hyoid muscle, arising from sternum and cartilage of first rib. q. Sterno-thyroid, fused posteriorly with dorsal surface of preceding muscle. r. Thyro-hyoid. s. Stylo-hyoid, with its tendon running along the thyro-hyal portion of the hyoid arch at an oblique angle to its muscular belly. Within this angle lies the ninth nerve. The trunks and branches of the pneumogastric and sympathetic nerves, as also of the carotid artery, have been removed from the triangular space bounded by the sterno- and thyro-hyoids mesially, by the sterno- mastoid externally, and by the muscular part of the stylo-hyoid superiorly, and in the space thus bounded we see the upper part of the cleido-mastoid muscle, m, externally, and a part of the rectus capitis anticus major internally passing up to take origin in company with each other from the basi-occipital. /. Body of hyoid bone ; the longer cornu, or thyro-hyal, passes backwards in connection with the tendon of the stylo-hyoid, the anterior cornu and arch are concealed from view. «. Digastric muscle, here represented by a single muscular belly, placed anteriorly and inserted into the symphysis of the lower jaw, and by a tendon taking origin from the paroccipital process and representing the posterior muscular belly of anthropotomy and the single muscular belly of the Carnivora. v. Mylo-hyoid muscle. •w. Depressores labii inferioris. x. Insertion of platysma myoides into lower jaw on left side below buccinator, anteriorly to masse ter muscle. x'. Fibres passing off from this cutaneous muscle to end in the moveable tissues round the mouth. x" ' . Insertion of platysma myoides into lower jaw of right side cut short. x'" . Insertion of platysma myoides into manubrium sterni. x"". Delamination of platysma myoides into two layers. y. Buccinator muscle. y1. Masseter muscle, with much less obliquity in its fibres than is usual in Rodents, as necessitated by the relations of the mandible and malar arch. It consists here of two strata as in the Horse, but has of course no antorbital factor as have so many Myomorphous and Hystricomorphous Rodents. It is bounded anteriorly by a stout tendinous band, which prevents the lower jaw to which it is affixed below from being separated beyond a certain distance from the jugal arch, to the freely projecting anterior angle of which it is affixed above. The mobile bifid lip compensates some- what for this restriction on the opening of the jaws. 2.' Internal pterygoid muscle. (a). Submaxillary gland and duct. (a'). Submaxillary gland of left side, covered by deep cervical fascia. (/3). Deep cervical fascia, forming sheaths for the muscles and capsules for the glandular structures in this region. A circular bulging indicates the area where it is underlaid by the Submaxillary gland. (7). Latissimus dorsr of right side, passing beyond the tendons of the great pectoral and the ' dermo- humerien ' cutaneous muscle to be inserted, together with the teres major, and underneath the tendon of the coracobrachialis, upon a well-marked facet below the inner tuberosity of the humerus. (S). Tendon of ' dermo-humerien ' muscle displaced outwards so as to leave a half-moon-shaped space between it and the tendon of the latissimus dorsi. The ' dermo-humerien ' muscle joins the posterior part of the pectoralis major, c, and first gains a fixed attachment to the pectoro- deltoid ridge, and then, by arching over the biceps, to the inner tuberosity of the humerus. (c). 'Latissimo-condyloideus' muscle, passing down from the tendon of the latissimus dorsi to be UPPER HALF OF RABBIT. 33 inserted into the olecranic process of the ulna on a semilunar raised line a little way above its posterior angle. It here joins the triceps, to which in lower mammals it usually gives an addi- tional head, only through fibrous expansions connecting it with the scapular head £. See Nat. Hist. Rev. 1 86 1, p. 512. (C). Scapular head of triceps, concealing external humeral head. (77). Internal humeral head of triceps, exposed by removal of brachial vessels and nerves. (0). Biceps flexor brachii, which gives off a band of fascia from the anterior surface of its muscular belly, passing on to the fascia enveloping the muscles of the fore- arm ; which, secondly, gives off a narrow tendinous slip from its broad principal tendon, which connects itself with the radius and the tendon of the pronator radii teres ; and which, thirdly, takes insertion by its broad principal tendon into a well-marked pit just above the inner and lower border of the ulna and below the anterior horn of its sigmoid articular surface. To show this insertion the pronator radii teres and the flexor muscles of the fore-arm have been divided and turned aside. («). Brachial plexus, seen, in the absence of any fibres of the posterior belly of the omohyoid or of the anterior scalene muscle, to be crossed vertically by phrenic nerve, and to give nerve-supply to the sterno-scapular and sterno-clavicular muscles by a slender nerve arising by three roots ; and to distribute other branches to the pectorals and shoulder and arm muscles. The phrenic nerve has one principal root in the neck above the level at which the formation of the brachial plexus begins ; it is connected, however, very usually with the factors of this plexus by more than one nervous filament. The nerve passing to the sterno-scapular and sterno-clavicular muscles is the homologue of the nerve given to the subclavius in man. For this portion of the nervous system of the cervical region the figure and description given by Hirschfeld and Leveille in their Neurologic, PI. 40, Fig. I, 1853, and reproduced in Quain's Anatomy, ed. 1882, i. p. 604, Fig. 338, may be compared. For other portions of the nerves of the cervical region, see Ludwig and Cyon, in Ludwig's Arbeiten for 1866, p. 148, or as reproduced in Cyon's Atlas zur Methodik der Physiologischen Experimenten, Taf. xvi. 1876, or with some modifications in Handbook for Physiological Laboratory, PI. xciii, or in Foster's Physiology, 4th ed. 1883, p. 190 (ed. 3, p. 176). For the nerves in the upper part of the neck in the Rabbit, see Loven, Ludwig's Arbeiten, /. c. Taf. i, and for the phrenic, Budge, Physiologic, 1862, p. 76. For the cervical region in the Dog, see Schmiedeberg, ibid. 1871, p. 56, reproduced locc. citt. The literature of Comparative Myology is very extensive. Amongst the older memoirs upon this subject may be specified Douglas, Myographiae Comparatae Spe- cimen, 1707, and the authorities cited by Otto in his Compendium of Human and Comparative Pathological Anatomy, translated by South, 1831, p. .245. Meckel, in 1 828, devoted one volume of his System derVergleichenden Anatomic to this subject. It also occupies a great part of the first volume of the second edition of Cuvier's Lemons d' Anatomic Compared, published in 1835, and is treated of in certain special departments in the third and fourth volumes also. In the Vergleichende Anatomic der Myxinoiden, S. A. pp. 216-247, 1835, S. A. pp. 109-111, 1841, of Johannes Miiller, valuable views as to the general homologies of muscles are to be found. Memoirs with similar scope but differing in results were written by Professor Goodsir in 1857-1858 (see Anatom. Memoirs, i. p. 451, 1856), and the subject has subs'e- quently been treated as a whole, and also in many specialised memoirs, by Professor Humphry in successive issues of the Cambridge and Edinburgh Journal of Anatomy and Physiology (see his Observations in Myology, 1872, and especially pp. 105-188). Cuvier's plates of the Myology of Mammals were published by M. Laurillard in 1849. The following memoirs may be mentioned as treating of the Comparative Myology of the Primates : — Ilg, Anatomic der Sehnenrollen, 1824 ; Burdach, Beitrag zur Vergleichenden Anatomic der Affen, 1838; Vrolik, Recherches d' Anatomic Compare'e sur le Chimpanze, 1841 ; Duvernoy, Archives du Museum d'Histoire Naturelle, viii, 1855-1856; Les Grands Singes pseudanthropomorphes ; Owen, P. Z. S. i. pp. 28-67; Church, Nat. Hist. Rev. 1861-1862, Myology of Orang; 24 DESCRIPTIONS OF PREPARATIONS. Burt Wilder, Boston Journal of Natural History, vol. vii. 1862, Myology of Chimpanzee; Prof. Flower and Dr. Murie on the Dissection of a Bushman, Journal of Anat. and Physiol. vol. i. 1867, pp. 196-205; Pagenstecher, Mensch und Affe, Zoologische Garten, April, 1867 ; Bischoff, Anatomic des Hylobates leuciscus, S. A. 1870, pp. 7-70; Abhandl. k. Bayer. Akad. Wiss. Matth.-phys. Cl. Bd. x. Abth. 3, 1870, pp. 203-266 ; Champneys, Muscles and Nerves of a Chim- panzee, Journal of Anat. and Physiol. vol. vi. pp. 176-211. Professor John Wood's papers in the Philosophical Transactions for 1869 on the Varieties of the Human Shoulder Muscles and their homologies in the Mammalia should be read in connection with the above description of those muscles in the Rabbit. Many other memoirs on Myology in its various aspects, morpho- logical and physiological, have appeared in this country from the pens of Professors Turner, Haughton, Macalister, Mivart, and Drs. Murie, J. D. Cunningham, A. H. Young, and others in the Proceedings of the Zoological Society, the Journal of Anatomy and Physiology, and elsewhere; and abroad from those of Professors Gruber, Gegenbaur, Fiirbringer, Riidinger, and MM. J. C. G. Liicae and Paul Albrecht. The muscles of the Rabbit are treated of in Professor Krause's mono- graph, Die Anatomic des Kaninchens, pp. 136-138, ed. 2, 1884, and those of the other domestic animals in Chauveau, Traite* d' Anatomic Compare'e des Animaux domestiques, 2nde £d. ; 1871, pp. 200-347, Franck. Anatomic der Hansthiere, 1871, pp. 343-478 ; Gurlt's Handbuch der Vergleichenden Anatomic der Haussaugethiere, ed. Leisering and Miiller, ed. 5, 1873, pp. 206-329. For the masseter muscle in Rodents and its relation to the antorbital fissure, see Waterhouse, History of Mammalia, ii. p. 151, PI. 6 a, 1848; Brandt, Mem. Acad. Imp. Sci. St. Pe'ters- bourg, Ser. 6, Sc. Nat. Tom. vii. 1855, p. 153 ; Giebel, Zeitschrift, Ges. Wiss. 1865, p. 427; A. Milne Edwards, Nouv. Archiv. Muse'um, iii, p. 92, 1867, Lophiomys; and Cuvier, Le9ons, ed. 2, 1835, iv. pt. i, p. 67, where the maxillary portion of the masseter is called ' musculus mandibulo-maxillaris.' 6. DUODENUM, PANCREAS, SPLEEN, AND LEFT KIDNEY, TOGETHER WITH PORTIONS OF STOMACH, OF THE JEJUNUM, OF THE LARGE INTESTINE, AND OF THE MESENTERY OF RABBIT (Lepus cuni- Culm) IN THEIR MUTUAL RELATIONS. With Figure 2. THE stomach, which in these animals, like the paunch of Ruminants, is never found empty after they begin to eat solid food, has been removed, with the exception of a little more than an inch of its pyloric end, a. As in nearly all Vertebrata, with the exception of the Ophidians, the terminal segment of the digestive tract, h to k, comes into close juxtaposition with this portion of the stomach and the duodenum 1. As noteworthy points 1 See Cuvier, Lemons d' Anatomic Comparee, ed. 2, 1835, torn. iv. pt. 2, p. 657: 'Dans toutes les classes des vertebres, 1'ordre des ophidiens seul excepte, le canal alimentaire a toujours une portion qui repond au gros intestin plus ou moins rapprochee de 1'estomac ou du commencement du canal intestinal.' And compare the plates in Rathke's memoir, Ueber den Darmkanal des Fische, 1824, which show that this arrangement exists in most orders of Fishes as well as in air-breathing Vertebrata. DUODENUM, PANCREAS, ETC., OF RABBIT. 25 more or less distinctive of the Rabbit may be enumerated, firstly, the great distance, a foot or even as much as twenty inches from the opening, c, of the bile-duct, at which the pancreatic duct, e, opens into the duodenum FIG. 2. — (One-half less than natural size.) DUODENUM, PANCREAS, AND SPLEEN, WITH PORTIONS OF STOMACH AND LARGE INTESTINE OF RABBIT (Lepus cimiculus) IN THEIR MUTUAL RELATIONS. a. Pyloric portion of stomach. b. Dilated commencement of duodenum, receiving c. The bile-duct. d. Concavity of duodenal loop, within which is contained a large part of e. The pancreas, to the duct of which, opening in the lower portion of the ascending limb of the loop, this line is drawn. f. Portion of pancreas in relation with the spleen and corresponding to the ' tail ' of the organ of anthropotomy. g. Spleen, with two accessory splenculi. a6 DESCRIPTIONS OF PREPARATIONS. h. Coil of colon supported by same lamina of peritoneum which attaches the pancreas and duodenum to each other and to the rectum. The letter points to the proximal end of the last coil described by the colon before it joins the rectum. i. Peyer's patch marking point where duodenum passes into jejunum in a plane anterior to that occupied by the rectum. j. Left kidney. k. Rectum containing scybala. secondly, the arrangement of the pancreas in widely-scattered loosely-com- pacted lobules, spreading all the way from the spleen at f down nearly to the concavity of the loop of duodenum ; thirdly, the great length of this loop reminding us of the similarly long duodenum of birds ; fourthly, the dilatation, £, at the commencement of the duodenum, an enlargement observable in many phytophagous Rodents, as Lagostomus, and also in the Hyrax, the Llama, and the Bottle-nose whale, Dephinus Dalei1. The descending colon, k> and the loop of duodenum are connected together by a continuous sheet of mesentery, the name therefore of intestinum mesenteriale cannot be confined in these animals, as it has been in others, to the jejunum and ileum ; in other words, the colon and duodenum have much greater freedom of movement allowed them by the greater extent of their mesen- tery than in certain other Mammals. Between the portion of the colon, shown here in section at ^, and the segment next the caecum, shown similarly in Preparation 7 and in Fig. 3 at f, certain coils not shown in either figure intervene. A part of these coils corresponds to the spirally-coiled portion 2 of the colon of the Artio- dactyla, but this correspondence is more plainly demonstrable in the Guinea Pig than in the Rabbit. Professor Claude Bernard has, in the Supplement aux Comptes Rendus, torn, i. PI. 3-4, Fig. 5, 1856, figured -a pancreas of the Rabbit with a second, which is a much smaller, duct opening into the bile-duct just before its entrance into the intestine. On PI. 7-8 /. c. he has figured the same organ from a Rabbit in which only the single duct figured here was present, and in which oily matter had been mixed with the animal's food. In this case it is only distally to the point of entrance of the duct that the lacteals are seen to be filled with white fluid and to have absorbed the oily matter ingested. This fact has been explained by Bidder and Schmidt, Die Verdauungssafte und der Stoffwechsel, p. 256, 1852, as being due to the oily matter having been absorbed in the proximal segments of the duodenum, and having been also passed through the lacteal vessels in connection with them, and so having disappeared from view in the period of from five to six hours which they suppose to have been allowed to elapse between the ingestion of the oily matter and the examination of the duodenum. Bernard's views were controverted by other physiologists (see Schiff, Moleschott's Untersuchungen, ii. p. 345, 1857, 1 See Owen, P. Z. S. 1839, P- I76' and Hunterian Catalogue, vol. i. 566 B. 2 For a figure showing the spirally-coiled portion of the colon, the caecum, and the small intestine of an Artiodactyle, see Dr. Cobbold, Cyclopaedia of Anatomy and Physiology, vol. v. art. ' Ruminantia/ Fig. 361, p. 538. For several showing similar arrangements in Rodentia, see Pallas, Novae Species Glirium, PI. xvii. 1778. CAECUM OF RABBIT. 27 and Krause, /. c. p. 163), but in answer, Bernard appealed to the existence firstly of a second duct in the situation above specified, and secondly of certain accessory pancreatic glands either sessile upon or with ducts opening into the bile-duct (see figures, pp. 350-351, Legons sur les proprie'te's physiologiques des Liquides de Porganisme, 1857). The second duct, however, of the pancreas in the Rabbit, though such an accessory duct does exist in several Ungulata, several Carnivora, the Elephant, and the Beaver, has been allowed by Bernard himself not to be constantly present (see Comptes Rendus, /. c. p. 390, and Le9ons de Physiologic Expe'rimentale, 1856, ii. p. 271, in which last work it is best described as 'petit conduit pancre'atique exceptionnel venant s'ouvrir dans le canal chole'doque ') ; and Krause, /. <:., suggests that a branch of the arteria gastro-epiploica dextra has been taken in an uninjected preparation for a duct. As regards the smaller accessory pancreatic glandules in connection with the bile-duct, Bonders and Kuhne have remarked that they would be quantitatively inadequate to account for the emulsification of fatty matters which has sometimes been observed to have taken place in the duodenum above the opening of the pancreatic duct. On the whole, however, it seems that though some emulsification is effected independently of the pancreas, the appearances figured by Bernard, Comptes Rendus, /. c., are so constant as to show that the process is very greatly helped by the secretion in question. For good summaries of the whole question, see Kuhne, Lehrbuch der Physiologischen Chemie, 1868, pp. 131-133, and Foster's Text-Book of Physiology, 4th ed. 1883, p. 295 (ed.- 3, p. 279). The existence in certain animals of more than a single duct to the pancreas was first pointed out by De Graaf, in his Tractatus Anatomico-medicus de succi pancreatici natura et usu, pp. 16-17, 1671. The duplex and triplex ductus, how- ever, are illustrated by him only by instances from Birds ; amongst Mammals he only records the presence of a second smaller duct as having been occasionally found in Man and the Dog. See also Nuhn, Lehrbuch der Vergleichenden Ana- tomic, pp. 50-51, 1878. 7. CAECUM AND PARTS OF LARGE AND OF SMALL INTESTINE OF RABBIT (Lep^cs cuniculus). With Figure 3. THE lymphatic (or lacteal) sinuses * surrounding the Peyerian follicles in the walls of the vermiform appendix, #, in the dilated termination of the small intestine, c, and in a saucer-shaped patch of glands, d, in the large intestine, have been injected with Berlin blue, so as to show the relation of the lacteal vessels to the Peyerian follicles in the walls of the intestine, as also to certain mesenteric glands left in relation with it. The mesentery connecting the caecum and vermiform appendix with the segments of the small and large intestine has been cut through, and the caecum disposed so 1 The lymphatic and lacteal vessels of many organs may be readily injected by the simple 'Einstichsverfahren' of Hyrtl and Teichmann. It is especially easy to obtain successful results in the case of the Peyerian glands of the Rabbit where, as in Bos and Oms, the base of the follicles is surrounded by lymphatic sinuses, and not as in man and in the Carnivora by a reticulation of lymphatic vessels. 28 DESCRIPTIONS OF PREPARATIONS. as to form the periphery of the preparation. The centre-point of the preparation and of the figure is occupied by a globular sac, c, the ' saccuius rotundusj the thick walls of which consist of Peyerian follicles, and represent the histologically somewhat similar termination of the small intestine in man. The small intestine dilates to form this sac after de- scribing a siphon-shaped course internal to one formed by the commence- ment of the colon. A saucer-shaped patch, d, similar in structure to the saccuius rotundus, lies immediately to the right of it in the wall of the large FIG. 3. — (Half natural size.) CAECUM WITH VERMIFORM APPENDIX AND PARTS OF LARGE AND SMALL INTESTINE OF RABBIT (LepUS CUniculus]. a. Vermiform appendix. b. Thin-walled portion of caecum, spirally constricted in correspondence with an internally situated spiral valve. c. The saccuius rotundiis, or dilated globular end of small intestine, with thick walls consisting of Peyerian follicles. d. Saucer-shaped thickening of wall of large intestine, on the internal periphery of which, not shown in this figure, the ileo-caecal valve opens. e. Distal end, i. e. segment next caecum, of small intestine in section. f. Proximal end, i. e. segment next caecum, of large intestine in section. Attached to the spirally-constricted part of the caecum on its inner edge, just before it passes into the vermiform appendix, we see certain mesenteric glands into which an injection of Berlin Blue has passed, having been introduced by ' Einstichung ' or puncture with a fine syringe into the lacunar spaces surrounding the Peyerian follicles of the vermiform appendix. UNVEBSITi CAECUM OF RABBIT. intestine, and overlies the opening of the ileocaecal valve. The caecum consists of two parts : the first of them, b, 14-5" in length, is wide in calibre, diminishing somewhat towards its termination, thin in its walls, and spirally constricted externally in correspondence with the spiral valve developed internally ; the second, a, about 5-5" long and corresponding to the vermi- form appendix of anthropotomy, is of much smaller calibre, but of much thicker walls, consisting mainly of Peyerian follicles ; it has no internally projecting fold nor externally impressed furrows, but is mapped into well- defined spaces by the injection occupying the lacunar spaces of the Peyerian patches. This injection may be seen to have passed into some mesenteric glands attached to the spirally-constricted portion of the caecum close to its junction with the appendix. What may be called the caput coli lies in this preparation between the concavity of the caecum and the two Peyerian agglomerations of glands already mentioned. Its external surface is not puckered or corrugated, but from it the three longitudinal muscular bands, the so-called taeniae coli, take origin and give a sacculated appearance to the commencement of the colon proper. The mucous membrane of the colon is seen to be beset with villi of a granulation-like appearance, an unusual appearance in Mammalia. The inwardly projecting folds of the colon are, the spiral coils of the caecum are not, obliterated by distension. From this Preparation and Figure it may be seen that the Peyerian agminate glands may take either the form of a caecal cylinder with thick walls as in the vermiform appendix, a, here and in man ; or that of a globular sac such as that developed here at c immediately proximally to the ileocaecal valve ; or that of a saucer-shaped thickening of the walls of the large intestine immediately beyond that valve. In like manner the follicles, which are the essentially distinctive characteristic of the Peyerian ' patches/ are themselves found, when examined under low powers of the microscope, to be of very various shapes ; those, for example, from the vermiform appendix of the Rabbit presenting the outlines of the sole of a shoe, whilst those of the Guinea-pig and of Man are subspheroidal in shape. For a description of the caecum and the sacculus rotundus, see F. Bohm, De Glandularum intestinalium structura penitiori, Berolin. 1835, Diss. inaug., cited Frey, Z. W. Z. xiii. p. 28, 1863. For the histology of the Peyerian glands generally, see Frey, /. r., and Hand- buch der Histologie et Histochemie des Menschen, ed. 5, 1876, p. 525 (lit. p. 529) ; His, Z. W. Z. x. p. 333 ; xi. p. 416; xiii. p. 455. For the histology of a Payer's patch in small intestine of Rabbit, see Verson in Strieker's Histology, Eng. Trans, vol. i. p. 566, fig. 108. For that of the sacculus rotundus, see His, Z. W. Z. xi. p. 426, Taf. xxxv. Fig. 6 ; Frey, /. c. p. 65. For that of the vermiform appendix, see Frey, Z. W. Z. xiii. p. 64, Taf. iii. Fig. i. 30 DESCRIPTIONS OF PREPARATIONS. For that of the Peyerian patch on the colon, where the follicles are larger than anywhere else, see Frey, ibid. p. 68. For the method of injection by puncture and the use of Berlin blue, see Frey, Z. W. Z. /. c. p. 52, and Mikroskopische Technik, ed. 7, 1881, pp. 128, 132, 293, 295; and Schafer, Practical Histology, pp. 144 and 157. For the presence of villi in the large intestine, see Cuvier, Legons, ed. 2, 1835, iv. pt. 2, pp. 243 and 274. 8. BLADDER, UTERUS MASCULINUS, UROGENITAL CANAL, AND RECTUM, WITH GLANDS IN RELATION WITH THEM RESPECTIVELY, OF MALE RABBIT (Lepus cuniculus). With Figure 4. THE upper half of the bladder has been removed, and the cavity of the lower half, a, laid open from behind, as have been also the cavities of the Uterus masculinus, s. ' Organ of Weber,' s. Vesicula prostatica, b, and of the urogenital canal at d. The right cms penis has been cut through at /, as also the left, not seen in the figure of this Preparation ; and the parts removed from their connection with the pelvis. A wire has been passed from the cavity of the bladder along the urethra behind the uterus masculi- nus, d, into the urogenital canal, d, and along this canal to its outlet in the glans penis. The rectum, c, has been displaced to the right from the mesial plane which it naturally occupies. The walls of the bladder have been turned outwards, and the orifices of the ureters appear consequently to lie upon its anterior instead of upon its under or posterior surface as in nature. The utems masculinus, b, is of colossal proportions in the Leporidae, attaining in some cases an absolute length of as much as three inches. With the exception of the Koala, it has not been observed in any marsupial, the generative organs of which animals, however, resemble those of the Rabbit in some points, as, for example, in the absence of seminal vesicles. It bears here a striking resemblance to the upper part of the vagina of the female Rabbit, as shown at b in Fig. 5, p. 37, having a similar rudimentary septum developed upon its anterior wall. But the presence, not merely of a considerable interlacement of muscular fibres, but also of glandular structures in its walls, as also some other reasons, would seem to make it safer to consider this organ to represent at once both uterus and vagina. Its upper angles are slightly produced and project between the upper boun- daries of the prostatic glands, h, and the cut short ends of the vasa de- ferentia. These projections, like the somewhat similarly drawn-out upper angles of the true human uterus, must be taken to represent the cornua uteri of the uterus duplex and uterus bicornis. The organ is flanked on BLADDER, ETC., OF MALE RABBIT. 31 either side by three packets of tubular glands, two lettered h and z, and a third which, being placed between them and somewhat anteriorly, is not seen clearly in this figure. This last consists of three simple unbranched tubules, which, though said by Weber to be variable in presence, are yet usually to be found in the interval between the lower ends of the two packets, h and z, and open by three distinct orifices into the urethra, as stated by Martin St. Ange, £tude sur 1'appareil reproducteur, p. n, who calls them ' accessory prostates.' The larger of the two other packets of glands, h, is readily separable from its fellow of the opposite side, poste- riorly as well as anteriorly ; the smaller, z, interdigitates more or less intimately with its fellow on the posterior surface of this portion of the urethra, but, as in the human subject, the glandular elements of the pro- state do not cross the middle line anteriorly. The characteristic concentri- cally laminated prostatic concretions are found in the tubuli of the smaller packet, though not in the larger ; but it is in accordance with general usage to speak of both of these sets of glands as ' prostates,' the smaller as the ' anterior ' and the larger as the * posterior prostates.' The anterior prostate is divaricable into two lobes, each with a separate duct : the posterior has but a single duct : with, however, the three ducts from the accessory pro- states there are six ducts opening into this portion of the urethra on either side. In the adult Rabbit the vasa deferentia open by an unusual arrange- ment into the literus masculinus about two lines from its orifice ; in the embryo the normal arrangement exists whereby the uterus masculinus opens a little way above these ducts. The testes have been removed in this preparation, together with the greater part of the vasa deferentia, the cut ends of which are seen passing behind the upper angles of the uterus mas- culinus. The portion of the urogenital canal which is laid open at d corre- sponds to the portion known in anthropotomy as the * membranous,' ' muscular,' or ' interfascial ' portion, and as the * isthmus urethrae.' Though longer relatively than in our own species, it has by no means the same relative preponderance which the homologous portion of the canal has in some other Rodents, as e.g. Arctomys, and in some Carnivora, e.g. Canis. The wire which has been passed down the urogenital canal from the interior of the bladder projects from the orifice of the urethra, which in these animals, retromingent like other Rodents, has the shape of a vertical backwardly-looking slit. The Cowper's glands, /, organs found in all mammals except, so far as is known, Mustela and Delphinus, with capsules of voluntary muscular fibre and ducts leading into the commencement of the spongy part of the urethra, are seen on either side of the sinus urogeni- talis. On the left side we see at /, in the plane in which this canal passed under the pubic arch, one of the crura penis with the pubo-cavernosus and ischio-cavernosus muscles cut through and detached from that arch, and on the opposite side we see, at k, two ano-preputial glands, one much larger 33 DESCRIPTIONS OF PREPARATIONS. than the other, and a hairless area of integument, g, on to which the ducts of these glands open, and which is prolonged into the fold covering the glans penis. A similar arrangement to this exists in the Tenrec. Just FIG. 4. — BLADDER, UTERUS MASCULINUS, AND UROGENITAL CANAL FROM ITS COMMENCE- MENT TO ITS TERMINATION AT THE ORIFICE OF THE PENIS, WITH THE PROSTATIC AND COWPERIAN GLANDS OPENING INTO IT, TOGETHER WITH THE RECTUM AND ANO- PREPUTIAL GLANDS OF RABBIT (Lepus cuniculus) SEEN FROM BEHIND. One-third larger than the parts figured are in the period of sexual activity of an adult male. a. Urinary bladder. The upper half has been removed and the cavity laid open, and the walls re- flected so as to show the openings of the two ureters towards its lower end. b. Uterus masculinus (in section), with prostatic glands on either side. c. Rectum, with two of the cystic stages (Cysticercus pisiformis} of Taenia serrata attached to the external surface of the longitudinal muscular coat of the intestine, which has been reflected to show their connection with it. d. Urogenital canal. A wire passes along it to its termination at the external orifice of the penis, BLADDER, ETC. OF MALE RABBIT. 33 having been introduced into it from the bladder at a. The penial urethra conveys both urinary and seminal products in all mammals above the Monotremata, in which group it conveys seminal products only, reversing the functions of the homologous canal of those female mam- mals, such as the Rat and the Mole, in which there is a closed clitorid urethra. e. Anus. f. Penis, with vertically elongated backward-looking orifice. g. Hairless patch on to which the ano-preputial glands open. h. Upper prostatic glands. i. Lower prostatic glands. j. Cowper's glands. k. Larger and smaller ano-preputial glands, rudimentary in man. /. Corpus cavernosum of right side in section, with parts of the pubo-cavernosus and ischio-cavernosus muscles lying upon it. below this area the anus opens at e. The rectum, cy has been displaced outwards and to the right so as to give an unimpeded view from behind of the organs just described. At its upper part it carries, attached to its longitudinal muscular coat, two hydatids, the cystic stage (Cysticercus pisiformis] of Taenia serrata, which takes on its cestoid or ' strobile ' form in the intestines of the Dog or Fox. A comparison of this preparation with such figures as those of the homologous organs in the human subject at pp. 419, 428, figs. 304, 310 of Quain's Elements of Anatomy, 8th ed. 1878, will show clearly the different proportions of the organs in the two subjects respectively, whilst a com- parison of it with the similarly dissected and displayed preparation of the female organs of the Rabbit (see infra, Prep. 9, Fig. 5, p. 37, and Descrip- tion, pp. 34-38) will show the close correspondence which exists between the male and female organs in this animal from the commencement of the urogenital canal outwards. The structure of the crura clitoridis and that of the crura penis are strikingly similar, especially in the very considerable thickness of their external fibrous sheath. The median septum which these sheaths form by their apposition remains distinct throughout the length of the compound organ they make up, as is the case in many mammals of the Rodent and other orders which possess a penial ossicle, whilst in the Ungulata, such as Cervus, Sus, Tapirus, Eqims, and the Cetacea there is neither median septum nor penial ossicle. An excellent article on the male generative apparatus and the anal glands of Mammals by Prof. Leydig is to be found in the Z. W. Z. ii. pp. 1-58, 1850. One of similar merit on the Uterus masculinus, s. Vesicula prostatica, by Professor R. Leuckart, is contained in the Cyclopaedia of Anatomy and Physiology, sub voc. 'Vesicula prostatica,' 1852. See also E. H. Weber, Abhandlungen Jablonowskischen Gesellschaft, 1846, pp. 382-385, 396, 405, Taf. v. figg. 1-3, where the generative organs of the Rabbit are figured ; as also one by Van Been, Z. W. Z. i. Taf. xx. 1849, and the work by Martin-Saint- Ange, L'Appareil reproducteur des Animaux Vertdbr^s, PI. i. and ii. 1854. See also Huschke in Soemmering's Anatomic, 1844, and a good article, translated by D 34 DESCRIPTIONS OF PREPARATIONS. Prof. Peters, in Miiller's Archiv, 1849, from the Swedish of F. Wahlgren. A uterus masculinus of somewhat similar proportions is figured by Pallas, ' Novae species Glirium,' 1778, from the Arctomys bobac, p. 117, Taf. ix. fig. 5, but in the genus Lagomys, judging from his figures 9 and 15, Taf. iv. B, we must suppose the two vasa deferentia to be, by a very unusual arrangement, fused into a single tube but a very short distance above the bulb of the urethra, and the uterus masculinus to be absent contrary to what Leuckart has (I.e. p. 1419) suggested. The organ is said to be somewhat smaller in the Hare than in the Rabbit. It is much smaller relatively in the human subject than in most Mammalia in which it has been seen ; for figures, however, of a large specimen from a human embryo of 32 weeks, see Betz in Miiller's Archiv, 1850, Taf. ii. For the General Morphology of the male and female generative organs in Mammals, see Allen Thomson in Quaint Elements of Anatomy, vol. ii. 1882, p. 911, and Banks * On the Wolffian Bodies,' 1864, ibi citat.; Leuckart, Zur Morphologic und Anatomie der Geschlechtorgane> aus Gottingen Studien, 1847, and /. c. supra ; Meckel, Zur 'Morphologic der Harn- und Geschlechtswerkzeuge, 1848; Fredrik Wahlgren in Miiller's Archiv, 1849, pp. 686-713; Rathke, Vortrage zur Vergleichenden Anatomie, 1862, pp. 135-170; Gegenbaur, Grundriss der Vergleichenden Anatomie, 1878, p. 645, where the uterus masculinus is taken as the homologue rather of the vagina than of the uterus. See also Watson, P. Z. S. 1878, p. 424; Young, Journal of Anat. and Physiol. 1879, p. 315; and for opening of vasa deferentia into uterus masculinus, see same two authors, on Elk, J. L. S. 1878, p. 375; on Hyaena crocuta, P. Z. S. /.<:.; and Dr. Young on Koala, /. c. For Taenia serrata and Cysticercus pisiformis, see Preps. 46 and 4 7 post. For figures, see P. J. van Beneden, Me'moire sur les Vers Intestinaux, Paris, 1858, p. 148, PI. xx. And for the fact that this particular parasite affects the locality h as a * point de predilection' (much, perhaps, in the way that the stress of certain infectious diseases has preferential sites), see Martin St. Ange, /. c. p. 7. For the migration of these and other Cysticerd from their first site in the liver, and for the recovery of the liver from the injury thus inflicted, see Leuckart, Die menschlichen Parasiten, 1879, pp. 92, 93, and 174. Flukes will similarly migrate from the liver into the peritoneal cavity. 9. GENERATIVE ORGANS, TOGETHER WITH THE BLADDER AND RECTUM, AND THE GLANDS IN RELATION WITH THEM, OF FE- MALE RABBIT (Lepus cuniciilus}. With Figure 5. THE dissection has been made upon the same plan as the preceding Preparation of the urinary and generative organs of the male Rabbit, and it shows very clearly the exactness of the homology existing between the parts from the commencement of the urogenital canals, c and d, outwards. The two organs, b and b, in the two preparations are very closely similar, and this similarity has led to the belief that the uterus masculinus of the GENERATIVE ORGANS OF FEMALE RABBIT. 35 male Rabbit should be held to represent merely the vagina of the female. It is a sounder view probably to consider it, as stated above, p. 30, to represent both vagina and uterus. The extent to which the vagina projects beyond the summit or ' superior fundus ' of the bladder is a remarkable point of contrast in the adult female Rabbit as compared with the adult females of most other species of mammals, placental and other ; in the newly-born female however (in which it may be remarked the parts in question are curiously similar to the analogous and homologous ones of the male at the same age), these pro- portions are exactly reversed, whilst in well-injected adult specimens the superior vesical arteries are sufficiently obvious to remind us of their functions in foetal life. The junction of the uteri, a, to the Fallopian tubes, at e, is marked by the difference in calibre of the two parts of the continuous and tortuous cylinders which they make up, and by the attachment to it of the ' liga- mentum rotundum' which passes down to be inserted into the pubic eminence, as also of the ' ligamentum ovarii ' and the * broad ligament V The two uteri are entirely distinct from their points of junction with the Fallopian, tubes up to their openings by separate or a tincae, one on either side of the rudimentary median septum on the anterior wall of the vagina, b. In the Bizcacha, Lagostomus trichodactyhis, a South American Rodent with many points of affinity to the Marsupials (see Darwin, Origin of Species, p. 379, 6th ed. 1872), what is a rudimentary septum in the Rabbit forms a perfect division in the vagina for a distance of as much as one inch beyond the ora tincae, constituting thus a transition towards the arrangements characteristic of Marsupialia. (See Owen, P. Z. S. Part vii. 1839, p. 77.) The uteri are similarly distinct, forming a ' uterus duplex ' in Sciunis^ Arctomys^ Spalax, Bathyergus^ Echimys, Erethizon, Hydrochaerus amongst Rodents, and in Orycteropus amongst Bruta ; they fuse into a ' corpus uteri ' with cornua uteri superadded to it at a greater or less dis- tance from the commencement of the vagina, forming thus a uterus bicornis, in Mus, Cavia, Caelogenys, Dasyprocta amongst Rodents, as in most other placental Mammals ; whilst even in the uterus simplex of Dasypus among the Bruta and in that of the Primates, more or less of the embryonal bifidity is retained by the production upwards and outwards of the angles of the fundus uteri towards the Fallopian tubes, much as are the angles of the uterus masculinus of the Rabbit shown in Fig. 4, b. In the lower part of the vagina, a short distance above its junction with 1 It is instructive to compare such a preparation as this of these ligaments with such a figure as that given by Dr. A. Farre (after Richard) in the Cyclopaedia of Anatomy and Physiology, Supple- ment, art. < Uterus,' p. 598, fig. 404; or that given by Prof. Allen Thomson in Quain and Sharpey's Elements of Anatomy, ii. 1882, p. 707, fig. 608 ; or those by Henle, Handbuch der Menschlich. Anat. ii. ed. 2, 1873, figs. 364, 374, pp. 474, 487* D 2 36 DESCRIPTIONS OF PREPARATIONS. the urethra to form the urogenital canal (V), two valvular involutions with their concavities looking upwards are developed, and have been taken to represent the somewhat similar projections in the double vaginae of Marsu- pials. Similar folds exist also in the lower portions of the vagina of the Rhinoceros and some other female Mammalia ; and in the upper of the Cetacea, the Suidae, and the insectivorous Tenrec. The urogenital canal, c, is laid open in part of its course, from the point where it commences at the junction of urethra and vagina, and is seen to be of considerable length as compared with the homologous segment in most Mammalia. This canal appears to be of the greatest relative length in the Monotremata, where the two uteri open into it separately and immediately above the openings of the ureters and below the downward opening of the bladder, that is to say, without the interposition of any vagina proper ; and of the least in the Primates, where it corresponds simply to the Vestibulum vaginae. But the morphological value of these facts is somewhat diminished when we find that in an order with tolerably uniform internal anatomical arrangements such as the Carnivora (to say nothing of an order such as the Bruta, compre- hending such widely different forms as Dasypus sexcinctus zx\&peba, Brady pus tridactylus and Cholaepus didactylus}, the length of this canal may vary as much as it does in species as closely allied as Hyaena brunnea (Murie, Tr. Z. S. for 1867, p. 504) and H. crocuta (Watson, P. Z. S. 1877, p. 376). And in orders such as that of the Rodents as illustrated by the Rabbit here, by Hystrix and by Dasyprocta, as that of the Insectivora as illustrated by Erinaceus, and as that of the Prosimii as illustrated by one at least of the Lemurida and by the Aye-aye, in which there is a urogenital canal of greater or less length, we may find precisely the opposite condition set up, that namely of entire separation of the genital .and urinal canals, by en- closure of the urethra within the lower part or corpus spongiosum of the clitoris ; reversing thus the arrangement which is usual in female mammals and is homologous to the condition known as ' hypospadias ' in males. Thus in the females of the Norway Rat, Mus decumanus, as also in Lago- stomus, Arvicola, Bathyergus, and Myodes amongst Rodents, Sorex and Talpa amongst Insectivora, and Stenops, Tarsius, and some but not all Lemurs amongst Prosimii, the clitoris is perforated by the urethra, and the urinary and female genital canals are entirely distinct, which is not the case here. One crus of the clitoris is seen here with its accompanying muscle as cut away from the os pubis ; the bilobed termination of the organ lies concealed within the vulva, with the inner -walls of which it is continuous. The ano-preputial glands, z, like the other organs from the commencement of the urogenital canal outwards, correspond very exactly with those of the male already described ; the glands representing the Cowperian glands of the male, and known in the female as the glands of Bartholini or Duverney, are seen between the ano-preputial glands and the GENERATIVE ORGANS OF FEMALE RABBIT. 37 crus of the clitoris, and in this specimen are smaller in size than the homo- logous glands in the other sex. The ovaries, as in animals still in the breeding period of life, are tuber- culated or nodular from the presence on their exterior of Graafian follicles FIG. 5.— OVARIES, FALLOPIAN TUBES, UTERI, VAGINA, UROGENITAL CANAL, RECTUM, AND ORGANS IN RELATION WITH THE EXTERNAL OPENINGS OF THE TWO LATTER CANALS FROM FEMALE RABBIT (Lepus cuniculus). (One-half less than natural size.) a. The two uteri opening by distinct ora tincae into the single vagina. b. Vagina laid open from behind in two parts of its length and showing a rudimentary septum on its anterior wall in its upper segment, and two semi-lunar folds in its lower and narrower segment. c. Junction of urethra with vagina to form the urogenital canal, the upper half of which is laid open. d. Entrance of ureters into bladder ; the points of entrance are drawn a little higher up than they ought to be. e. Junction of Fallopian tube with uterus of either side. From this point on the right a flat band, 38 DESCRIPTIONS OF PREPARATIONS. the homologue of the * ligamentum rotundum ' of anthropotomy and of the gubernaculum testis in the male, passes downwards and inwards, whilst from it and the Fallopian tube beyond it, that part of the ' broad ligament ' known as < ala vespertilionis ' passes to the funnel-shaped opening of that tube. f, Fimbrjated and funnel-shaped opening of the Fallopian tube ; the so-called ' Pavilion ' or ' Infun- dibulnm,' the mucous orifice looking into the peritoneal cavity inwards and backwards from the posterior aspect of the broad ligament. g. Ovary, connected on the right side with the pavilion by the tubo-ovarian or ' infundibulo-ovarian' ligament, formed by the drawing out of the periphery of the pavilion, and on the left with the uterus of that side by the ovarian ligament. There are no fimbriae on the infundibulum of the Monotremata, and its outer periphery is almost entire in the placental genus Sus, but they are present, so far as is known, in all other Mammals. h. Rectum, with one hydatid affixed to it. i. Ano-preputial glands. Between the smaller of these glands and the rectum is seen part of the gland homologous with Cowper's gland and known as the gland of Bartholini. J. One crus of the clitoris, with muscle in relation with it. k. Rectum a«d- external outlet of urogenital canal. with ova nearly mature ; they are seen to be attached to the infundibular fimbrjate opening of the Fallopian tubes on either side by one of the fimbriae of that opening, the so-called tubo-ovarian ligament, which secures that an ovum on escaping from its Graafian follicle shall readily find its way into the oviduct. The ligaments and peritoneal laminae passing from the opening of the oviduct and from the ovary to the proxi- mal end of the uterus cause the oviduct to take a very contorted course. In some mammals, e.g. Talpa^ Canis, Ursus, Meles, the peritoneal laminae, which here form only a widely open capsule for the ovary and infundibulum, coalesce and form an all but completely closed pouch in which ovary and opening of oviduct are both enclosed, so that an escape of an ovum into the peritoneal cavity becomes nearly impossible except through the small orifice near the uterine cornu where the interior of the pouch communicates with the peritoneal cavity. In the two carnivora last specified, even the small orifice thus left is frequently filled up by a hernia-like protrusion into it of one of the fimbriae of the infundibulum ; and the ovary, a small portion of which is usually to be seen at that orifice, is thus cut off from view and from access to the peritoneal cavity. In some other Carnivora (Lutra and Mustela), at least in the young state, this orifice appears to be entirely closed by an overgrowth of peritoneal membrane. For descriptions and figures of the organs of the human subject corresponding with those described here, see Farre, Cyclopaedia of Anatomy and Physiology, vol. v. article ' Uterus and its Appendages.' For the general homologies of the female generative organs, see Leuckart, Rathke, and Allen Thomson, as referred to above, p. 34. For the vagina and uterus duplex as an abnormality in anthropotomy, see Dr. J. Matthews Duncan, Journal of Anat. and Physiol. May, 1867, p. 269. The Palaeontological history of the order Rodentia may seem to put the AFFINITIES OF RODENTS. 39 affinity which has so often been alluded to as existing between them and the Ungulata into a clearer light than even the most detailed account of the anatomical resemblances which exist between the living representatives of the two groups under comparison. The most obvious, though perhaps also essentially the least important difficulty besetting such a comparison, is that which is based on the difference in size ; but this is done away with by the discovery in American Miocene deposits of ' various small ruminant-like animals, some not larger than a Squirrel in size, to which the names Leptomeryx, Hypisodus, Hypertragulus, have been applied.' See Flower, Proc. Royal Institution, March 10, 1876. On the other hand, the discovery in the South American Pliocene deposit of the animal known as Mesotherium, which was a little larger than the Capybara, and has been supposed to link the Rodents and notably the Leporidae to the Perissodactyle type by many connecting peculiarities, whilst retaining itself so many of the character- istics of the Rodent order as to have induced Mr. Alston, P. Z. S. 1876, pp. 73, 74, to create for its reception a third suborder, that of Glires hebetidentati (see p. 44), equivalent to each of the two other suborders, those of Glires dupliddentati and Glires simpliridentati, into which Rodents may be divided, might seem to give as positive an illustration of the absence of any sharp circumscription in the delimita- tion of this group as can be asked for. The order, Tillodontia, established by Professor O. C. Marsh, would seem to bear more striking evidence still in the same direction, coming as it does from the earliest geological formation, the lower Eocene, in which Rodents are found ; and combining some of the characters of Carnivores with those of both Ungulates and Rodents, as though they were survivors of some still earlier type unrepresented, as indeed are all Mammals, in any as yet explored strata of the Cretaceous age. Similar affinity has often been spoken of as existing between the Toxodontia of the Post-Tertiary South American deposits and the Rodents; but as this much later date might have led us to expect, the similarities between the two orders are by no means so clearly made out. As regards existing Ungulata, the affinities of the Rodentia are distributed pretty evenly between the two divisions of Ungulata Artiodactyla, the non-Ruminantia s. Bunodonta and the Ruminantia s. Selenodonta (for which see W. Kowalewsky, Ph. Tr. vol. 163, pp. 69-74, 1873) and the Perissodactyla ; but it must be borne in mind that though the differences between the now existing Hoofed animals thus named are very sharply defined, geological researches, especially in the American Tertiaries, have revealed to us forms in which these lines of demarcation tend to become obscured, or at least approximated. It would be easy to say, for example, that the imperfection of the orbital ring was a point of considerable importance, and united the Rodentia with the Perissodactyla; but Professor O. C. Marsh (see Introduction and Succession of Vertebrate Life in America, 1879, p. 161) states that several species of Cervidae from the lower Pliocene of the west of America fail to have the orbit closed behind. A simple way of expressing the known facts may perhaps be furnished by saying that though the Rodentia with reference merely to their claws would be ranged with the Unguiculata rather than with the Ungulata of Linnaeus (see Sy sterna Naturae, ed. xiii. vol. i. p. 17), they nevertheless present both in their skeleton and in their internal organs more marked points of essential affinity to 40 DESCRIPTIONS OF PREPARATIONS. the latter than to the former of these two sets of Mammalia, and that even as regards the particular difference, such as it is, which exists between claws and hoofs, the subungulate character of certain South American Rodents (Dasyprocta, Caelogenys, Dinomys, Cavia, Dolichotis, and, notably, Hydrochaerus) very much reduces its value. The fact pointed out in the Description of Preparation, p. 27, note, to the effect that the microscopic character of the ultimate radicles or spon- gioles of the lacteal system in the Rabbit resembles that of the Ruminants rather than that of the Unguiculata, illustrates this position. On the other hand, however, Krause, Hdbk. der Menschlichen Anatomic, p. 234, 1876, has pointed out that the Rabbit, unlike the Ungulata, Horse, Pig, Ox, and Sheep, has capillary veins in the spleen instead of wide funnel-shaped orifices for the commencement of those vessels. The semi-pedunculate fashion in which the deciduous serotina is in the Guinea Pig, Agouti, and some other Rodents, attached to the wall of the uterus, and the fact that in the former at least of the animals named this structure is sometimes retained within the uterus at parturition, may appear to point to the existence of some approximation to the character of the non-deciduate cotyledonary placentation of typical Ruminants. But the springing up in later foetal life of a zone of villi, supplied by omphalo-mesenteric vessels around and exteriorly to the placental area in Rodents, though of physiological interest, as was pointed out in the Zoological Transactions for 1863, vol. v, still does not constitute, as seems to have been more recently stated, any but a physiological approximation of the Rodent to the Ruminant type of placentation. The peculiarity of the placentae of the Guinea Pig and Agouti just alluded to, the comparative simplicity of their caecum, and the great development of the spirally-coiled portion of their colon, and the presence in them o£ but a single superior cava, the small number of their mammae and of their young, and the early attainment by them of the faculty of self-help, are points in which they resemble the Artiodactyla and differ from the LagomOrphi • but, as the Descriptions of the Preparations (Nos. 3 and 4) show, Brandt is entirely justified in pointing, I.e. p. 291, to numerous connecting links, characteribus nonnullis generis Leporini cum Ruminantium ordine. The apparent paradox of an affinity connecting the Proboscidea (the Elephant and Mastodon) with the living Rodents, has already been alluded to, p. 8. One of the most striking of these resemblances lies in the facts that in both sets of animals the intermaxillary interposes itself between the maxillary and the nasal bones and joins the frontal without however touching upon the lacrymal bone, relations which do not exist either in the Aye-aye, or in the Hyrax, or in the Dugong, animals with incisors of somewhat similar relative proportions. But this remarkable peculiarity had not been attained to by the Pliocene genus and sub- order Glires hebetidentati of Alston, represented by the Mesotherium, in which animal the intermaxillary's nasal process is too short to prevent the maxillary from abutting upon the nasal, and fails, as in Hyrax alone of the animals just mentioned, to reach the frontal. The absence of enamel from the incisor teeth or tusks of the adult Elephant, otherwise so exactly homologous with the incisors of the Rodents, might appear to constitute a difference between them and those of the Rodents, did we not know that these teeth when first cut in the Elephant have a cap of enamel, and that a AFFINITIES OF RODENTS. serially homologous fact is presented to us in the possession by the American Pliocene Mastodons of a band of enamel on their tusks (see Marsh, /. c. p. 41). The naked-eye structure of the molars of many Rodents, such as the Capybara, is strikingly like that of the Elephants, and the microscopic arrangement of the enamel in the same teeth in the Elephant and the Mastodon has been shown to be of the same type as that observable in all Rodents except the Leporidae and the Hystricidae. The coronoid processes of the lower jaw are small, and the relations of the malars to the malar processes of the maxillaries and of the squamous bones are the same in both orders, both the latter bones entering into the com- position of the arch. The Elephant might have been expected to have had a single superior cava, as have some of the largest existing Rodents, the South American Subungulata, and all other existing large-sized mammals. But it has, like all the Old World and all Nearctic Rodents, two. And it appears that there is some reason for holding that the Proboscidea, in contradistinction to the living Perissodactyla, and to the Artiodactylous Camel, Pig, and Deer, are an Old World type, and to be expected therefore to follow that type in such particulars as the one specified. The Elephant however has the symphysis of the lower jaw perfectly anchy- losed ; it is more entirely testicondous ; its brain is richly convoluted, as indeed are the brains of most existing mammals of large size except the Manatee, and its placenta is zonary. The zonary form of the placenta similarly distinguishes the Hyracoidea from the Rodentia whilst uniting them more or less with the Proboscidea. In the structural composition of the malar arch Hyrax agrees with the two orders just mentioned, and differs from the Ungulata, but on an estimate such as is given by Brandt (Mem. Acad. Sci. St. Pe"tersbourg, 1869, Tom. xiv) of the sum total of the affinities of this animal, it should be ranked as an Ungulate with Rodent ward affinities rather than as a Rodent. Brandt's own words run thus, p. 119 : * Es werden daher die Hyracen im Allgemeinen wohl am passendsten als Ungulata gliriformia oder glirioidea bezeichnet und als eine, jedoch weit mehr den Huft- thieren ahnliche, Mittelform zwischen den genannten Thieren und den Glires angesehen werden konnen.' On another page, p. 116, Brandt suggests that Hyrax may connect the ' Pachyderms,' by which he in this connection means the Perisso- dactyla, with the Sciuromorphous Rodents specially, and also with the Lagomor- phous, whilst the Mesotherium would stand similarly in relation to the Lagomorphi and the larger animals in question. Andreas Wagner indeed had expressed himself in 1844 in opposition to Cuvier, and to the same effect as Brandt, by saying that a separate family should be created for Hyrax amongst the Pachyderms, and that it should be considered as forming a transition towards the Rodents. Cuvier, as is well known, in opposition to the view which trivial names, such as that of lMarmotte bdtardj given to Hyrax, embodied, went so far as to speak of it as being a dwarf Rhinoceros, and in his Ossemens Fossiles, ii. p. 127 seqq. 1822, he enumerates certain points of resemblance between the two animals. These points are such as the number, 20-21, of the ribs, as the transverse direction of the condyle of the lower jaw, as the absence of canines, as the shape of the nails, and as the presence of but three toes on the hind feet, and are inadequate to the support of such a view as the epigrammatically stated one just quoted. His 42 DESCRIPTIONS OF PREPARATIONS. suggestion as to placing Hyrax between Rhinoceros and Tapir is less open to objection. It may be remarked that some of the peculiarities of Hyrax, such as the bisacculate character of its stomach, the presence of a sterno-maxillary muscle, the dilatations of its Eustachian tubes, the flatness of the roof of its skull, the perfect orbital ring, the absence of a clavicle, and of an acromion on the scapula, and the presence of a diastema between the incisors and the canines, are points which are at least as indicative of equine affinities as of connection either with the Rhinoceros or with the Rodent stock. And if on the other hand the presence of a languette on the dorsal, and ®i papillae foliatae on the lateral surface of the tongue in Hyrax, are points curiously reproduced in the Rabbit and the Rat, the important point of the absence of a second superior vena cava distinguishes Hyrax from all rodents except the Caviidae, and the zonular character of its placenta distinguishes it essentially from all known Rodents whatever. Few points of real affinity connect the Rodentia with the Insectivora in addition to those more superficial peculiarities in general appearance, size, and habits, which have caused the two orders to be connected in common language. And these points are mainly such as may be characterized as indications of comparatively low organization in the scale of Mammalian life ; and they are rarely constant in all the members of the two orders. Among them we may specify the not infrequent vacuolation or fenestration of the bony roof of the palate, the imperfect condition of the bony support of the tympanum, and the retention of the primitive jugular foramen. Similarly in the soft parts of both orders we find usually, if not always, cerebral hemispheres devoid of convolutions, and two superior cavae, as we do in all known Sauropsida. But irrespective of the differences in dentition, which may appear to lose some of their importance since we have been acquainted with the existence of Apatemys from the Middle Eocene of America, which had (see Marsh, /. c. p. 43) gliriform incisors combined with insectivorous molars, the digestive organs in the two orders are strikingly different, the intestinal tract being provided with a caecum in all Rodents except the Myoxini, whilst it is absent in all Insectivora except Macroscelis, Rhynchocyon, and the Tupaiidae. Both orders alike, it is true, have a discoid deciduate placenta, but in the Rodents the omphalo-mesenteric vessels take a large share in the nutrition of the foetus up to the end of pregnancy, the umbilical vesicle lines the whole of the chorion which is not occupied by the disk of the placenta proper; and this disk is never attached except to the mesometrial border of the uterine cornua ; whilst in the Insectivora as in the Chiroptera the umbilical vesicle is attached, with the exception of Sorex, only over a limited circular area of the chorion opposite that occupied by the true placenta, the functional importance of its vessels is less, and the site of the allantoid placenta may be on any part of the uterine walls. There can scarcely be any doubt that the Insectivora must be considered to be as ancient a form of mammalian life as the Rodentia ; indisputable remains however of the order have not as yet been found lower than the Middle Eocene, whilst remains of Rodents have been identified in the lowest Eocene, the Cory- phodon beds of North America. This absence of Insectivora however must be ascribed to the present imperfection of the Geological record ; and it should be AFFINITIES OF RODENTS. 43 added that a jaw of a small Mammal (Dromotherium) from the Trias of North Carolina, though supposed to have belonged to some Entomophagous Marsupial allied to Myrmecobius^ is not definitely proved not to have belonged to a true Insectivora, and indeed that the likeness between these two sets of animals is a very strong argument for the antiquity of the Insectivora, as also for their inferiority when compared with the Rodentia, at least as regards the lowest members of each order. This inferiority is manifested in the Geographical Distribution of the two orders. Though both alike are favoured as regards spreading over the world by the smallness of their size and their faculty of hibernation, and, in the case of certain Insectivora, of aestivation also, the Insectivora are obviously a ' failing ' order, being, though represented in all the Zoo-geographical regions except the South American and Australian, still poor in numbers both of individuals and of species. Many points, however, bear evidence to the antiquity of the Rodent type of Mammalian life, and to its alliance with still lower forms, such as the Marsupials and the Sauropsida, which is correlated with that antiquity. The imperfect ossification which leaves perforations or fenestrations in many cranial and facial bones, and allows sutures such as those of the basicranial bones and the symphysis of the jaw to remain unanchylosed, while in higher Mammals we find them con- tinuously ossified ; the retention both of bones such as the presphenoid and of processes such as the basipterygoid in a distinctness and independence which is lost in higher forms ; the small size of the coronoid process and the tendency to inversion of the angle of the mandible, the occasional persistence of the vomer in two distinct moieties, and the constant imperfection of the orbital ring, are some of such points furnished by the skeleton. Great as are the variations observable within the limits of the order Rodentia, all living Rodents agree in the following particulars : they have the homologues of the two central incisors in both jaws furnished with permanently growing pulps, and with chisel-shaped cutting edges ; these teeth are so largely developed as to have caused all the other incisors to abort either wholly, as in the Glires simplitidentati, or all but wholly, as in Glires dupliddentati or Lagomorphi; and the canines to abort invariably, leaving thus a diastema between the incisive and the molar series of teeth. The symphysis of the lower jaw is never anchylosed perfectly. The greatly developed incisive bones always separate the maxillaries from the nasals, but are themselves separated by processes of the maxillaries from the lacrymals. The lacrymal canal opens inside the orbit. The orbital ring is never perfect. The maxillary bone always forms a part of the jugal arch together with the malar and the squamous. The omphalo-mesenteric vessels contribute importantly to the nourishment of the foetus during the whole of intra- uterine life ; the allantoid or true placenta is attached to the mesometrial part of the circumference of the uterus. Till the discovery of the singularly aberrant Lophiomys Imhausii, the absence of the opposable hallux so commonly observable in Marsupials was supposed to be characteristic of all Rodents. This Rodent, in which the malars-and parietals extend over the temporal fossa, as is the case in Chelonia, but in no other known Mammal, is also the known Rodent which is pedimanous. See A. Milne Edwards, 44 DESCRIPTIONS OF PREPARATIONS. Nouv. Arch, du Museum, 1867, p. 114. The name ' Prensiculantia ' or 'Pfotler' was suggested by Illiger, Prodromus, p. 81, i8n5 after Buttmann, as a substitute for that of Glires upon the following grounds : — ' Nota : nomen hujus ordinis a celeb. Buttmann excogitatum est ut maniculorum instar manuum usus his Mam- malibus familiaris indicetur, aliis e motus instrumento desumtis notis toti ordini communibus simulque characteristicis, deficientibus.' The differences between the Glires duplicidentati, s. Lagomorphi, represented by the Rabbits, the ordinary Hares, and the Tailless Hares, and all the other living Rodents, are so great, and those which separate the three other subdivisions of the order, the Myomorphi, represented by the Mice, the Sciuromorphi, represented by the Squirrels, and the Hystricomorphi, represented by the Porcupines and Cavies, are so much smaller as to make it convenient to divide the order into two main divisions or Suborders : viz. the Simplicidentati, comprehending the three sections just mentioned, and the Duplicidentati, comprehending the Hares and Rabbits. Of these last the following propositions may be made in contradistinction to the Simpliddentati. They have, as adults, two small incisors placed behind the two large ones in the upper jaw, these two pairs of teeth representing the anterior and the posterior of the three pairs of early life. They have a larger number of teeth, f in the molar series, than any other Rodents. The incisors are surrounded by a perfect zone, if not of enamel in the adult, at least of enamel membrane in the developing tooth. The enamel of the incisors is not divisible into two layers. The incisive and the optic foramina are, severally, confluent, and the bony palate greatly reduced. The glenoid fossa and the articular condyle of the lower jaw are less specialized to antero-posterior movement than is the case in other Rodents. The coronoid process and the sockets for the incisors in the same bone are also less specialized. Ossification is less perfect, as for example in the facial part of the maxillary and in the basicranial bones. They have a canalis caroticus in the tympanic, but no true alisphenoid canal. The fibula is anchylosed to the tibia, but articulates with the os calcis. The placenta is not disc-shaped, but consists of two or more lobes sessile on the chorion and clamped together by a saddle-shaped decidua serotina. These differences appear to place the Lagomorphi at a much greater distance from the group made up of the Hystricomorphi, the Myomorphi, and the Sciuromorphi, than any one of those sections is from any other. Of them the Hystricomorphi, especially as represented by the Cavies, come nearest to the Lagomorphi '; the Myomorphi should be placed centrally, and the Sciuromorphi highest in position. Mr. E. R. Alston, P. Z. S. Jan. 1876, pp. 73, 74, has proposed the establish- ment of a third Suborder, that of Glires hebetidentati, for the reception of the genus Mesotherium found in South American Pliocene deposits, and represented by an animal a little larger than the Capybara. This animal had the same number of incisor teeth placed in the same way, viz. two above and four below, as the Hyrax ; they were surrounded, not merely faced, with enamel, and were not brought to a sharp cutting edge by wear, but came to present a transversely hollow blunt surface, whence the name hebetidentati. The mandibular condyle and the glenoid fossa were not specialized to perform the antero-posterior movements so characteristic of the Simplicidentati, and the three anterior and upper molar teeth were convex out- AFFINITIES OF RODENTS. 45 wards as in Toxodon, not inwards as in typical Rodents. The massive heavy ridges on the skull give it an appearance not unlike that of the Sirenian skull when looked at from above ; in some particulars, such as the great development of its paroccipitals and its palate, it resembles at once the Capybara, an exclusively South American type, and the Artiodactyla, but notably differs from both, and resembles the Sciuromorphi or most specialized Rodents, in possessing perfect clavicles. Professor Gervais, who has figured its skull and skeletal bones in his Zoologie et Palaeontologie Ge'ne'rales, pp. 134-139, 145, and PI. xxii-xxv, considers its affinities to be Lagomorphous ; and it resembles the Hares and differs from the Cavina in having its fibula in articular relation with the calcaneum. But it differs from all existing Rodents in having its intermaxillaries so short as to allow the maxillaries to abut upon the nasals ; in having the cutting surfaces of its incisors pit-shaped like those of the Horse ; and- in having, as have Perissodactyla, Suidae, and Pro- boscidea, the two halves of the mandible anchylosed, not merely suturally joined at the symphysis. And with its other characteristics, not found to coexist in living animals, it combines the peculiarity, observable in the South American Bruta, except Cydothurus didactylus, of having the ischium articulated with the caudal vertebrae. In face of these anatomical facts recovered from the fossil remains, and bearing in mind that with them may have been combined in the living animal differences in the soft parts which would have definitely prevented us from ranking it with the Rodentia, we may hesitate to accept Mesotherium as constituting a third suborder of that division of Mammals. Its various affinities appear to be fairly stated by M. Serres in his fifth Memoir relating to it, Comptes Rendus, Ixv. 1867, p. 599, in a passage which, for this reason, as also because it shows on how many sides orders, which may at first sight appear to be sharply circumscribed, may come into relation with each other, it may be well to give in extenso : — ' Quelque am- biguitd que nous offre, en effet, ce singulier animal, ressemblant i°. aux Rongeurs par la disposition de ces incisives, du me*sodonte (p. 145^ et par les dents uni- radiculees ; 2°. aux jeunes Pachydermes par la forme ge'ne'rale, et le rudiment des fossettes des incisives et des molaires ; 3°. aux Edent^s, ses contemporains, par la masse, la lourdain de sa tete et de ses membres, ainsi que par la bifurcation de la derniere phalange, enfin 4°. aux Cetacds, d'une part, par 1'enfoncement de 1'occiput, 1'affaissement de la voute du crane et la petitesse de 1'encephale, qui en est la con- dition premiere, et d'autre part, par le nez large et court, un peu ouvert en dessus ce qui concorde avec Pide'e de M. le Dr. Sene'chal qui pense que le Mesotherium e*tait, peut-etre, un animal aquatique ; ne"anmoins, au milieu de ces conformity's si diverses, celles qui le rapprochent plus particulierement des Rongeurs et des Pachydermes dominent tellement les autres, que c'est entre ces deux ordres de Mammiferes que nous croyons qu'il doit etre place* comme un anneau interme'diaire qui les relie. Get anneau serait-il, selon la pense'e de Blainville, un des chamons perdus de la seVie animale?' 46 DESCRIPTIONS OF PREPARATIONS. 10. COMMON PIGEON (Cohimba livia\ Showing nervous, digestive, circulatory, and parts of respiratory and renal systems. THE brain has been exposed in situ by the removal of the roof of the cranium ; the integument has been removed from the right side of the front of the cervical region, as have also most of the feathers from the entire body ; an opening has been made into the right side of the crop, which has been distended ; the larger part of the right half of the body walls has been removed, together with the muscles and the limbs which it supported, and a red injection has been thrown into and filled the venous system. The surface of the cerebral hemispheres is smooth ; the proportion of the encephalic nervous mass to the intraspinal is much greater than in the cold-blooded Vertebrata. The backward projection of the cerebellum is very considerable. The eyes are large. The vertical third eyelid is drawn forward. The nostrils open externally as long slits overhung by a soft, bare, tumid membrane1; the external auditory meatus, which has no concha, has the feathers arranged round it like a circlet of tentacles. The great pectoral muscle, the main depressor of the humerus and the wing, is seen in section along its origin from the lower portion of the keel of the sternum, and from the furculum, the outer and lateral portions of the sternum, from which it also took origin, having been removed. Placed dorsally with reference to this muscle we see the second pectoral, the main elevator of the humerus and the wing, arising from a larger portion both of the keel and of the lateral parts of the sternum than the pectoralis major, and passing internally to the coracoid to enter the pulley-like canal, the foramen triosseum, formed by the clavicle or furcula, the coracoid, and the scapula. This muscle is supplied by nerves which pass in front of, whilst the great pectoral is supplied by nerves which pass below the coracoid, the first being homologous with the subclavian, the second with the anterior thoracic nerves of anthropotomy. Whilst the tendon of the second pectoral or great levator humeri muscle, which is cut short, is seen issuing from its canal on the further side of the glenoid articular surface; on the proximal side, the humerus having been removed, we see the tendon of the biceps, homologous with the { short head ' of anthropotomy. The cut-short triangular end of the pectoralis major is seen to become partially bifid 1 This membrane is sometimes called a ' cere,' but it is better to restrict the term to the denser structure similarly placed in and similarly distinctive of the Aetomorphae and Psittacomorphae. Some of the Charadrii-morphae (Plovers), which on account of a peculiarity in the nasal bones (in which they resemble the Pigeons) were placed with them in a separate order, the Charadriiformes, s. Schizorhinae, resemble them also in the conformation of this membrane. See Coues, Key to American Birds, 1872, p. 26; Strickland and Melville, The Dodo and its Kindred, p. 46 ; Garrod, P. Z. S. 1873, p. 33, 1874, p. 100. COMMON PIGEON. 47 toward its apex • in the perfect condition of the parts the smaller inferiorly-placed division of the muscle gave off two tendons1, one to the long and the other to the short extensor plicae alaris anterioris ; the larger division passed over a smooth facet on the humerus and over the coracoid head of the biceps to be inserted upon the inwardly-looking surface of the great triangular tuberosity of the humerus. Dorsally to the apex of the great pectoral we see a thin stratum of muscle in relation internally with the crop and homologous with the deltoid of anthropotomy. This muscle is divided into two strata by delamination. The superficial layer consists of three parts, of which the first and most internally-placed joins the long extensor of the anterior fold of the alar membrane; the second and mesially-placed joins the short extensor, whilst the third and dorsally-placed portion is inserted into the outer aspect of the humerus from its middle down to a nodule at its lower fourth marking the origin of the long radial extensor of the carpus. The deep layer consists of one short muscle2 innervated by the circumflex, arising from the portions of the coracoid and scapula and of the ligaments in relation with the shoulder-joint, and inserted into the upper surface of the humerus along a line reaching from the apex of the triangular tuberosity receiving the tendon of the great depressor humeri to the facet receiving the tendon of the great long levator. Overlapped by this muscle, which acts as a levator humeri, and wedged between it and the great pectoral depressor, is a second short levator humeri, innervated as is the coracobrachialis, not as is the deltoid, arising from the coracoid and passing down on the outer side of the tendon of the biceps to be inserted under the upper portion of the tendon of the great pectoral. This muscle therefore should be con- sidered to be a coracobrachialis. We have thus three muscles — the ' pectoralis secundus,' or long levator ; and two shorter muscles, the former of which may be called ' deltoides externus,' and the latter 'coracobrachialis brevis' — entrusted with the work of raising the humerus, but each with a distinct innervation. In some birds, e.g. Anser^ the deltoides externus passes into and takes an enlarged origin from the walls of the foramen triosseum, and gains some mechanical advantage by availing itself of its pulley-like outlet. In relation with the lower portion of the right coracoid may be seen two other coracobrachiales (cut short in this preparation, but shown in situ at v and u, in 1 The alar extensor muscles and many of the other muscles of the wing in Birds will be found well figured and described in a monograph by Schoepss in Meckel's Archiv, 1829. Those of the Pigeon are similarly figured and described by Macgillivray in his History of British Birds, i. pp. 34- 42, Plate iii, a work with which I was not acquainted when the first edition of this book was pub- lished. Mr. Macgillivray remarks, p. 38, that the small muscular mass called by him retractor plicae ', and figured here, PI. ii. w', infra, had not been met with by him in any other birds except Pigeons. A doubling back of the tendon of the short alar extensor on to the nodule of origin of the long radial extensor of the carpus has been shown by Professor Garrod to be characteristic of the true Passeres. See P. Z. S. 1876, p. 509 ; Prof. Bell's translation of Miiller's Vocal Organs of the Pas- seres, Appendix by Garrod, p. 64, Oxford, 1878. 2 This muscle corresponds with No. 19, the Deltoides externus of Schoepss, as given in his monograph on the Muscles of Flight in Birds in Meckel's Archiv for 1829, and is called 'levator humeri ' by Tiedemann, and ' le petit releveur de 1'humerus ' by Vicq d'Azyr. It appears to have been often confounded with the muscle next to be spoken of and lying close to it, which, though similar in function and size, is differently innervated and quite separate from it. The latter muscle is correctly described by Schoepss, I.e. p. 122, and named (No. 20) ' Deltoides inferior.' 48 DESCRIPTIONS OF PREPARATIONS. Plate ii, infra), the pectoralis tertius audorum, s. coracobrachialis inferior, No. 1 6, Schoepss ; and the coracobrachialis superior. The ' coracobrachial muscular ap- paratus ' has been shown by Professor Wood, Journal of Anatomy and Physiology, i. 1866, p. 55, to be somewhat similarly multifid in many mammals normally and in man occasionally. The crop forms a sac with bilateral glandular1 pouches at the lower end of the distended oesophagus. As in the oesophagus of all Sauropsida, the muscular coat may be seen to consist of two layers, the outer one of which lies transversely2, and the inner one parallel to the long axis of the digestive tube, whilst both alike consist of unstriped fibres, and thus contrast in two points with the muscular coat of both Mammals and Fishes. It rests on either side upon the furcula and the muscles arising from it. In the cavity of the thorax a black bristle has been passed between the proventriculus and the aorta as this vessel arches over from the left to the right. The right side of the heart rests upon the right lobe of the liver from which the vena cava inferior is seen to pass up into the right auricle, entering it at a point a little superiorly as well as posteriorly placed to that at which the vena cava superior of the right side opens into it. The left lobe of the liver is, like the right, deeply excavated on its inner aspect for the reception of the heart ; and it is still more extensively excavated on its under surface for the reception of the powerful gizzard, so as to be less than one-half the size of the other main lobe of the gland. The veins from the upper extremity and shoulder are cut short at their point of junction with the jugular to form the vena cava superior. The pneumogastric nerve is seen in relation superiorly with the jugular vein ; superiorly again, and internally to the nerve, we see the proventriculus ; and superiorly again to it, the longi colli muscles arising from the vertebral hypapophyses. Tracing the aorta backwards towards the heart from the point where it arched over the right bronchus, which, together with the pulmonary artery placed before it and the pulmonary vein placed behind it, has been removed in this dissection, we see it pass behind the vena cava superior dextra, and give off the two arteriae innominatae, one for either side of the body, very close to the base of the heart. These two arteries together with the pulmonary artery give a characteristic appearance to the region at the base of the heart in all Birds. Each arteria innominata divides into a common carotid3 and a subclavian trunk. The subclavian, after giving off a small branch homologous with the internal mammary artery of 1 This is the usually accepted statement: but see p. 53, infra. 2 Gulliver, P. Z. S. 1842, 1869, 1870; Leydig, Fische und Reptilien, 1853, p. 41 ; Histologie, 1857, p. 324- 3 There are two carotids in all Columbae and in the majority of non-passerine birds, but in no true Passerines. COMMON PIGEON. 49 anthropotomy, divides into an axillary trunk, which passes into the wing together with the brachial nerves, and into the much larger arteria thoracica externa which supplies the great pectoral muscles. The gizzard is concealed from view by the right lobe of the liver and the posterior or xiphisternal end of the sternum which supports and protects both these viscera. The duodenal loop containing the pancreas, and the segment beyond that portion readily recognisable as belonging to the duodenum by its large calibre, form a rudimentary spiral coil. The distal end of the loop and of the pancreas inside it are bent upon the proximal segment next to the gizzard, and are in relation with the lower lobe of the kidney on the right as the gizzard is with the lower lobe of the kidney on the left side. The distal segment of the duodenum bends up at some distance from its pancreatic loop and comes into relation with the right lobe of the liver which is excavated to receive it. In the interval between these coils of the duodenum portions of the two other convolutions characteristic of Columbae show themselves. Of these the distal one is the smaller in calibre ; it shows some Peyerian glands, and is connected with the proximal portion of the duodenum by a lamina of mesentery much as is the colon in the Rabbit (see Fig. 2, supra}. The coil interposed between the duodenal and the distal coil is much the longest and most distinctively spiral of the three, but being placed dorsally to them is not seen in its full extent till they are displaced1. The lung, which occupies a much smaller space in the dorso-sternal plane than in mammals, and in an ordinary dissection of a bird's viscera scarcely comes into view until either the ribs are displaced a little outwards or the lobes of the liver a little inwards, reaches backwards so far as to interpose itself for some distance between the anterior lobe of the kidney and the -os ilii. Anteriorly one of the musculo-tendinous languettes which in the Bird represent the diaphragm of mammals passes inwards from the rib to the covering of the lung and interposes itself between the region of the anterior kidney-lobe and that of the lung. The Bird's differs further from the Mammalian lung in being lodged conformably to the intercostal spaces, and being indented by the six unanchylosed ribs, instead of being freely suspended, as is invariably the case in mammals, and divided into lobes, as is very ordinarily the case in those animals 2. Another and most important point of 1 Upon the differences observable in the number and character of these coils Dr. Hans Gadow has based a classification of Birds in two memoirs containing much valuable information, J. Z. xiii. 1879. 2 That however different at first sight the topographical relations of the lungs, liver, and heart may seem to be in Birds and Mammals respectively, they nevertheless are not essentially dissimilar, may be seen from the fact that the technical works on the physical examination of the heart in man speak of the ' difficulty of separating the adjacent edges of the heart and liver ' by percussion, and of the dulness produced by the apex of the heart being indistinguishable from that produced by the convex surface of the liver below it. See e.g. French's Diseases of the Liver, New Sydenham Soc. Trans. 1860, p. 30 ; Walshe, Diseases of the Heart, 1862, p. 42. E 50 DESCRIPTIONS OF PREPARATIONS. difference is furnished by the prolongation of the lung, by means of its bronchial stem and branches, into air-cells permeating a very large part of the entire body. The largest of these receptacles are the infra-renally placed 'abdominal air-sacs,' the right one of which extends from the posterior border of the lung above and behind the liver, so as, firstly, to interpose itself between the inferior surface of the kidney and the intestines, and, secondly, to stretch beyond the region of the kidney into that of the rectum. The kidney, like the lung of Birds, is shaped conformably with the bones supporting it ; and it is divisible here into three lobes, increasing in size from before backwards in correspondence with the iliac and pelvic surfaces in relation with them. The division of the kidney into three lobes is better marked when seen as here from the side than from in front. Even from in front however the anterior lobe may be seen to be more or less limited off from the middle lobe by the great vein from the lower extremity which corresponds to the external iliac of mammals ; and the middle lobe in its turn to be limited off from the posterior by the chief artery of the lower extremity, which is in most Birds, as here, the sciatic, not the external iliac artery. The sciatic artery gives off branches to the two posterior lobes of the kidney, and an arteria renalis superior arising from the aorta mainly supplies the anterior lobe. The veins from the lower limbs are supposed, and with considerable probability, to act as a renal-portal or inferent system, as in the cold-blooded Sauropsida (see infra, p. 56). The (rec 'trices) feathers having been removed from the caudal tract, the anal oil-gland (glandula uropygii) is brought into view1. Its duct projects freely, and is apically biperforate and tongue-shaped. It has no circlet of feathers, differing herein from that of the Fowls, the Diurnal Raptores, and all Aquatic Birds, and resembling that of the Nocturnal Raptores and all Passerines. Its outlines pass gradually into those of the bilobed gland mass, and with them make up a heart-shaped contour, the transverse axis of which is somewhat shorter than its anteroposterior. Though the Pigeon resembles the Passeres in the absence of feathers round this duct and upon the skin covering the oil-gland, and herein as in some other particulars comes nearer to that order than do the Gallinaceous Birds, the oil-gland and duct of the Passeres are nevertheless sufficiently distinctive to enable us to distinguish a specimen of the order from one of any other, even in the absence, not merely of the head, but also of the feathers of the whole body. These distinctive characters are the predominance of the transverse Over the anteroposterior diameter of the gland, the shortness and apical bluntness of its efferent duct, the thinness of its walls, and the distinctness 1 For a full account of the uropygial gland, see Nitzsch's Pterylography, Ray Soc. Trans. 1867, pp. 38-42. For the correlations of this gland with certain other structural peculiarities, see Garrod, P. Z. S. 1874, p. 118. COMMON PIGEON. 51 of its contour lines from those of the gland itself. The oil-gland of the Lamellirostres figured by C. G. Carus, Tab. Anat. Comp. Illust. Pars vii. Tab. 7. fig. 5, furnishes us with a sharp contrast to that of the Pigeon and the Passeres in being very deeply bilobed and in having its anteroposterior axis much longer than its transverse. The Ostrich, Emeu, Cassowary, and Apteryx agree with the cursorial Bustard in lacking this gland. Its presence is not constant in all the species of either Columbae or Psittacidae. It is larger in size in Birds of aquatic than in those of other habits. In all Birds, and in no other class of animals, will the same description as that given here apply to the nerve-system, to the relations of the muscles of the anterior limb, and to the relations of the aorta to the right bronchus. The peculiarities of the pancreas and duodenum are probably nearly equally distinctive. The crop and the uropygial gland are peculiar to, though not universally found in Birds ; but the presence or absence of these two latter structures is explicable probably by reference to the special habits or special needs of the species possessing or lacking them, and is therefore of physiological rather than of morphological importance. The epidermic skeleton consists, as in all birds, of the horny covering of the bill, of claws to the toes, and scales covering the metatarsalia and toes, and of feathers. Of the latter, the Pigeon possesses two kinds, contour feathers or pennae, and filo- plumes — down-feathers or plumulae being absent. Every penna consists of the following parts : a central axis shaft or scapus divisible into a proximal hollow tube or calamus, and a distal solid white shaft or rachis : of barbs borne upon the rachis and bearing in their turn barbules. Rachis and barbs together make up the vane or vexillum. The calamus is implanted in a follicle of the skin to which small muscular bundles are connected. It has^-a proximal aperture or inferior umbilicus, and a distal, the superior umbilicus, at its junction with the rachis. The barbules are implanted on the proximal and distal surfaces of the barb forming two series of process pointing obliquely towards the edge of the vane. The distal series of barbules of one barb overlaps the proximal series of the barb beyond, i. e. nearer the tip of the feather. The distal barbules bear on their under surfaces microscopic booklets, each one of which catches hold of an underlying proximal barbula. The vane thus acquires great solidity. The filo- plumes are closely associated with the pennae from which they differ in having a slender shaft with but a trace of the tube and a rudimentary vane composed of a few barbs bearing simple and disunited barbules. In many birds an after- shaft or hyporachis arises close to the superior umbilicus, and resembles when well developed, e. g. in Ratitae or gallinaceous birds, a second feather. It varies much in size and in the character of its vane, and is sometimes absent as in the Pigeon. The distal barbules sometimes carry barbicets, structures which resemble the hook- lets minus the terminal hook, Other varieties of feathers not found in the Pigeon are (i) the down-feathers or plumulae, which lie beneath and between the contour-feathers, and haye either a simple soft rachis bearing soft barbules or a tube with a crown of soft bar- E 3 52 DESCRIPTIONS OF PREPARATIONS. bules ; and (2) semiplumes (penno-plumae] which have a stiff rachis and soft barbs and lie at the outer margin of the pterylae and are covered by the contour feathers. In a few birds down-feathers exist which grow persistently and break off at the apex; whilst a powder or dust is poured out of the follicle lodging the tube. They are known as powder-down feathers, and either occur scattered all over the body — e. g. in some Parrots, or restricted to limited tracts — e.g. in Ardea. The tuft of feathers springing from the pollex constitutes the bastard wing. The row of large wing-feathers is termed remiges, and is divisible into a set of primaries and of secondaries attached, the former along the manus, the latter along the ulna. They are covered above and below by the upper and lower wing coverts. The large tail feathers are known as rectrices, and they are covered above and below by the tail coverts. The feathers are not implanted irregularly into the body but along certain tracts or pterylae between which are bare spaces or apteria. The following special points should be noted in the internal anatomy. In the central nervous system the small olfactory lobes; the cerebral hemispheres pointed in front, broad behind, showing in horizontal section a huge corpus stri- atum, a lateral ventricle reduced to a narrow chink and thin internal and pos- terior walls : a small pineal gland, reverted, with walls composed chiefly of fibrous tissue, and its extremity attached to the dura mater : the solid optic lobes widely separate in the middle line where the cerebellum touches the cerebral hemispheres : the cerebellum composed of a large median lobe and two small lateral floccular lobes, the median lobe showing in longitudinal section an arbor vitae as in Mammalia : the well-marked angle between the medulla oblongata and spinal cord. The latter has a large lumbar swelling, in which a mass of neuroglia or substantia reticularis lies immediately dorsal to the central canal, the posterior fissure is widely open and the gap filled by a gelatinous tissue derived from the pia mater. This lumbar swelling was of immense size in the extinct Stegosaurus (JDeinosauria). The spinal cord ends with a filum terminale and the posterior nerves form a cauda equina. The sympathetic system is double in the neck : one part accompanies the vertebral artery and vein and is lodged in the vertebrarterial canal : the other accompanies the carotid arteries on the ventral aspect of the neck. The two parts are connected. \ The structure of the eye is peculiar in some points. The sclerotic coat has an anterior conical portion containing a ring of bones, and a posterior spheroidal portion. A pigmented vascular fold of membrane — the pecten — runs obliquely forwards from the entrance of the optic nerve and projects into the vitreous humour. The line of attachment marks the position of the embryonic choroidal fissure. Its capillaries are continuous with those of the optic nerve, and not of the choroid, and are contained within lymphatic sheaths. The nictitating mem- brane is moved by two special muscles — a quadratus or bursalis, and a pyrami- dalis, which lie at .the back of the eye and take origin from the sclerotic. The former is a square muscle ending in a tendinous border, but the tendon is tubular. Through the tube runs the cord-like tendon of the pyramidalis which is inserted into the lower angle or edge of the nictitating membrane. When the pyramidalis contracts, its tendon is prevented from pressing on the optic nerve over which it runs by the simultaneous contraction of the quadratus. There is a well-developed Harderian gland for the third eye-lid lying below the eye-ball. COMMON PIGEON. 53 The tongue is of fair size. It has been discovered by Fraisse (Z. A. iv. 1881, p. 310), that in the embryo Duck there are embryonic feathers developed on the tongue which are arrested in development. The crop in the Pigeon is remark- able for its large size and bilateral symmetry. Gadow has distinguished between a true crop with glandular walls and a * Haut ' or ' Schlund ' crop with non-glan- dular walls. The former exercises a chemical action on the food and occurs in the Fowl and Pigeon and in their congeners ; whilst the latter exercises no such action and is simply a storehouse for food swallowed, e. g. in many Ducks, Cassowary, &c. But the researches of Hasse (Zeitschrift fur Rationelle Medizin, xxiii. 1865) proved long ago that the upper part of the oesophagus and the crop itself are non- glandular in the Pigeon, whereas the portion of the oesophagus below the crop, like the proventriculus, is provided with glands. The crop, and the upper as well as the lower part of the oesophagus, are lined by a many-layered epithelium — the lower cells of which are granular and plump, but as they pass to the surface become flattened out yet not cornified. If the surface of the crop is scraped a small amount of a whitish liquid can be collected at all times. The amount is greatly increased in both sexes for about the first eight days after the hatching of the young, which are fed with the so-called ' pigeon's milk ' regurgitated by the parent bird into the mouth of the young. It is a milky liquid containing cheese-like solid morsels. Hasse found that at this time the epithelium of the upper parts of the oesophagus and of the crop, but especially its side parts was much thickened, the bloodvessels dilated and full of blood. He also found that the cells of the epithelium undergo rapid division : are granular, and contain abundant fatty gran- ules : that they are set free in masses which break down partially. The rem- nants form the cheese-like morsels, whilst the fatty cells set free give the liquid a milky look. The cells in the masses retain their nuclei, those set free have either lost them or show them undergoing fatty degeneration. The ' milk ' collects within the crop whence it is expelled by the action of two muscles which spring from the upper part of the clavicles and are inserted into the skin ven- trally. The physiological properties of the fluid do not appear to have been fully investigated. It is doubtful, perhaps, whether the small amount of it present at times other than the breeding season has any chemical effect on the food. The swelling of grain, peas, &c., in the crop may be due only to the action of moisture and warmth, and is therefore a physical effect. It is stated by Hasse that a similar milky secretion occurs in some species of Parrots. The lower oesophagus has an epithelium similar to that of the upper find of the crop, but there are a small number of glands in the mucous membrane with an alkaline or neutral secretion. In the proventriculus the epithelium is reduced to a single layer of columnar cells. The glands of this region secrete the acid gastric juice. In the Pigeon they are small and simple : in the Fowl and Goose they have lateral loculi. But their size and character vary a good deal in different birds. They are largest in Rhea and the Ostrich. The gizzard is well developed in the Pigeon as in the Fowl and the Lamel- lirostres s. Chenomorphae (Ducks, &c.). A short tube, which is always pale as com- pared with the vascular proventriculus, connects that organ to the gizzard. The pylorus is placed on the right side and close to the entrance of the proventri- culus. The walls of the gizzard show two tendinous spots which lie one on the 54 DESCRIPTIONS OF PREPARATIONS. right, the other on the left side in the natural position of the organ. The two tendinous spots are the centres whence radiate the musculi laterales which make up the bulk of the organ. They are composed of smooth muscle fibres which in transverse section appear to lie in columns, the fibres in each column being connected to the fibres in the adjoining columns by short tendinous fibres. Two softer muscular bundles lie one close to the entrance of the proventriculus, the other at the opposite pole. These are the musculi intermedii. The mucous mem- brane is glandular ; and the glands secrete the horny internal lining. This lining is discoloured by the food, and it is continually worn away by the attrition of stones, &c., swallowed with the food ; and it is continually formed anew by the action of the glands. If it is stripped off by force, the attached surface appears as if covered by very fine short villi, or processes which have been pulled out of the gland-tubes. In sections of the gizzard these processes can be readily made out dipping into the gland-tubes ; they are conical, more transparent, and ap- parently softer than the superficial layers. Vertical lines, apparently formed either by irregularities of structure or by imbedded cells, are traceable nearly through the thickness of the horny layer. In many birds, e. g. flesh-eating birds, the muscular walls of the gizzard are thin and its secreted lining soft and tenacious. The degree of development of the muscles and the lining is closely connected with the character of the food — as was shown by Hunter's experiment of feeding a Sea-gull with barley. The muscles then became at least double the thickness of those in a Gull which had lived on fish. Cf. Catalogue of Physiological Series, Royal College of Surgeons' Museum, i. p. 49, preps. 522 D, and 523. For the bile and pancreatic ducts, see description of Plate II. The two caeca appended to the commencement of the large intestine are very small in the Pigeon — a contrast to the long caeca of the common Fowl, Pheasant, Grouse, &c. The large intestine is short and straight, as in all birds except the Ostrich. The rectal aperture lies at the apex of a cloaca common to it and the urogenital ducts. The rectal region of the cloaca is large and is separated by an annular ridge, which in some birds is but feebly indicated, from a small middle or urogenital chamber into which open the ureters and genital ducts on the dorsal wall, the genital apertures externally to those of the ureters. An annular fold always present separates the urogenital chamber from the third, outer or pos- terior chamber, the external opening of which is guarded by a strong sphincter muscle. An aperture on the dorsal wall of this outer chamber leads in young Pigeons into the Bursa Fabricii — an ovoid sac with a narrow neck lying dorsally to the cloaca. In the Ratitae (? Apteryx) the urogenital chamber opens into the Bursa owing to the fact that the neck of the latter is not constricted, and its aperture is commensurate with the dorsal aspect of the outer chamber of the cloaca. In Plotus anhinga (Darter), a Carinate, Garrod found a large aperture to the Bursa, and Forbes has confirmed the fact. The Bursa commences to atrophy in the Pigeon at the sixth, in the Fowl at the eighth, month according to Martin Saint-Ange. There seems to be much variety in this respect among birds, and it is possible that it occasionally persists. As a rule, however, its aperture closes, its cavity is obliterated and its walls atrophied — a more or less fibrous remnant persisting. It is large in the embryo. Its function and homology are unknown : its cavity contains only remnants of faeces or concretions, the origin of which is COMMON PIGEON. 55 not certain. It appears early in development as a solid outgrowth of cells from the dorsal wall of the proctodaeum ( = cloaca) before the rectal aperture opens into it. The central cells atrophy and thus form the cavity of the organ. In the adult, according to Stieda, the walls consist of a fibrous outer coat with internal prolongations which form the axes of the primary and secondary longitudinal folds which project internally ; and an internal epithelium which has several layers of cells, the superficial columnar, the deep angular, separated from the fibrous coat by a membrana propria. In addition there are the ' follicles ' so-called, which are im- bedded in the longitudinal folds. Each follicle consists of a central mass of minute rounded nucleated cells, continuous at the apex of the follicle with the deep layer of the epithelium, and probably never cut off from it ; of a membrana propria continuous with that of the epithelium ; and of a surrounding investment of adenoid or reticular tissue with numerous capillaries developed from the same embryonic cells as the fibrous coat. The respiratory system has the usual structure seen in birds. The syrinx or lower larynx is simple. The last 3-4 rings of the trachea send towards one another on the ventral surface processes which do not fuse. The two last rings are widely separated dorsally, but they are joined inter se by a cartilaginous rod. The bronchial rings are only half-rings, i. e. are incomplete on the inner surface of the bronchus. The free ends of the two first rings meet dorsally and ventrally, and together with the modified tracheal rings form the tympanum. A thin membrane — the membrana tympaniformis interna — forms the inner wall of the anterior part of each bronchus where the ends of the half-rings are wide apart. Inside the tympanum the mucous membrane is thickened, and there are two cushion-like projections, one on each side. A fold, the membrana semilunaris, projects forwards between the cushions from the point of bifurcation of the trachea. It is supported by the pessulus, a rod of cartilage, sometimes ossified. The Pigeon has but one pair of syringeal muscles — the tracheales laterales. The syrinx is least developed in the Cathartidae^ or American Vultures. It is a distinctive avian structure, and is present in the Ratitae, where it is often said to be absent, and it is best developed in Rhea. The trachea is tied to the sternum by a pair of muscles — the sterno-tracheales. The structure of the lungs, bronchi and air-sacs, is very similar in all birds. The main bronchus enters the lung, gives off above the spot where the pulmonary artery crosses it four eparterial bronchi and below that spot nine hyparterial. The ultimate branches of the bronchi are long tubes closely packed. Their inner walls carry projecting circular fibrous septa connected by longitudinal septa. The respiratory capillaries are distributed on these septa. An epithelium lining these tubes has not been detected. Other structures to be carefully noted are the costo- pleural muscles, small muscular bundles which spring from the junction of the vertebral and sternal ribs and spread out like a fan in an aponeurosis (i. e. ten- dinous expansion) which lies on the ventral aspect of the lung between the pleura and the air-sacs. The air-sacs appended to certain bronchi are nine in number and are as follows, — an interclavicular sac lying in front of the trachea and formed by the fusion of two sub-bronchial sacs, the bronchial apertures of which lie just anterior to the spot where the bronchi enter the lungs ; two prae-bronchial sacs lying one dorsal to each lobe of the interclfcficular sac, with apertures at the an- terior border of the lung : two anterior intermediate or thoracic sacs with aper- 56 DESCRIPTIONS OF PREPARATIONS. tures close behind the point of exit of the pulmonary veins : two posterior inter- mediate sacs lying one on each side in front of the corresponding posterior or abdominal sacs, with apertures at the postero-external angles of the lungs. The walls of the sacs are thin, and lined by a pavement epithelium, the ciliated epi- thelium of the bronchi ending at the spot where they open into the sacs. The following additional points should be noted in the heart. A thin muscular Eustachian valve protects the entrance of the vena cava inferior. The right auriculo- ventricular valve consists of two muscular flaps, one long and external, the other short, meeting the first at a slight angle and connected to the ventricular wall at the base of the conus of origin of the pulmonary artery, which is long. The left auriculo-ventricular or mitral valve consists of two membranous flaps with chordae tendineae and musculi papillares as in a Mammal. A transverse section across the ventricles shows a very thick walled left ventricle with a thinner walled right ventricle, the cavity of which is crescentic, embracing the left ventricle. Ornitho- rhynchus alone among Mammals shows a similar section. Three semilunar valves guard the base of the aorta and pulmonary artery. The common carotids run up the demi-canal on the ventral aspect of the cervical vertebrae, hidden by the muscles. The jugulars are connected by an anastomosis across the base of the skull, and there is a remarkable subcutaneous venous plexus in the neck. The connections of the veins of the kidney and of the hind limb are characteristic. The femoral vein coming from the hind limb passes through the kidney between its anterior and middle lobes. Close to the internal border of the kidney it receives on its anterior border the efferent vein of the anterior lobe : on its posterior of the middle and posterior lobes. It then passes on as the common iliac vein, and, joining its fellow, becomes vena cava inferior. Towards the outer border of the kidney the femoral vein gives off an afferent renal vein, which enters and ramifies in the anterior lobe of the kidney ; and a second — the renal portal, or hypogastric vein — which traverses the middle and posterior lobes of the kidney, gives off branches in its course, receives the sciatic vein, and then issues from the posterior lobe of the kidney and unites with its fellow. Into the point of union falls the caudal vein : from it issues a large coccygeo-mesenteric vein, which receives veins from the cloaca and large intestine, runs along the mesentery of the large intestine, and joins the portal system. Each lobe of the kidney receives a small artery. As pointed out by Jourdain, the calibre of the veins is out of all proportion to that of the arteries. . He states that each renal lobule contains a central vein, or rootlet of the efferent vein, and is surrounded by a number of venules derived from the afferent vein. A capillary network connects the two systems of vessels. As there are no valves in the renal portal veins, with rare exceptions, such as certain Ratitae, Bustard, Swan, the blood from the viscera and hind limbs can pass freely either through the iliac veins and thence to the vena cava, or through the coccygeo-mesenteric vein to the hepatic portal system. It can hardly be doubted from these facts that the kidney, as in Reptilia, receives venous blood. The thyroid is a paired gland, and lies close to the origin of the common carotid arteries. The thymus is also paired, and may be found in young birds as a long gland, one on each side of the neck. The spleen lies on the right side of the proventriculus. The supra-renal capsules are two yellow bodies closely connected to the iliac veins at the anterior end of the kidneys. A "fcfis If T Of THE Tf tt TJNIVEE. COMMON PIGEON. 57 The generative system is the same as in all birds. A rudiment of the right ovary is rarely detectable ; but there is frequently a rudiment of the oviduct of that side. There is no intromittent organ. Birds, W. K. Parker and A. Newton, Encyclopaedia Britannica (9th ed.)} iii. Aves, Selenka and Gadow, Bronn's Klas. und Ordn. des Thierreichs, vi. Abth. iv. (in progress). Points in Anatomy of Columbae, Garrod, P. Z. S. 1874. Columba livia. Yarrell, History of British Birds, 4th ed. by Newton and Saunders, iii. p. 13; or Seebohm, British Birds and their Eggs, ii. p. 405. Pigeon. T. J. Parker, Zootomy, London, 1884, p. 209. Integument. General account, Jeffries, Proc. Nat. Soc. Boston, xxii. 1882-83. Skin of foot. Hanau, Inaugural Dissertation, Bonn, 1881. Epitrichium and beak. Gardiner, A. M. A. xxiv. 1884. Feathers. Studer, Inaugural Dissertation, Berne, 1873. Id. Z. W. Z. xxx. 1878. Waldeyer, Festgabe zu Jacob Henle, Bonn, 1882. Pterylography. Nitsche, Ray Society, 1867 (transl. by Sclater). Uropygial gland. Kossmann, Z. W. Z. xxi. 1870-71. Cutaneous sense organs. Merkel, Endigungen der sensibeln Nerven, Rostock, 1880; cf. Carriere, A. M. A. xxi. 1882. Central nervous system, Stieda, Z. W. Z. xix. 1867. Special sense organs. Eye. Leuckart, Grafe and Samisch, Handbuch der Augenheilkunde, Leipzig, 1876, ii. p. 144. Pecten. Denissenko, A. M. A. xix. 1 88 1. Cramptorfs muscle. Exner, SB. Akad. Wien. Ixxxviii. Abth. 3, 1882. Harderian gland. MacLeod, Archives de Biol. i. 1880. Ear. Hasse, Z. W. Z. xvii. 1867. Retzius, Gehororgan der Wirbelthiere, ii. Stockholm, 1884. Alimentary canal. Gadow, J. Z. xiii. 1879 (Columbae, p. 148). Salivary glands. Reichel, M. J. viii. 1882. Glands of Proventriculus. Nussbaum, A. M. A. xiii. 1877 (p. 744). Gizzard. Wiedersheim, A. M. A. viii. 1872. Bursa Fabricii. Forbes, P. Z. S. 1877. Retterer, Journal de PAnatomie (Robin), xxi. 1885. Cf. Martin Saint- Ange, £tude de 1'appareil reproducteur, Paris, 1854 (Memoires pre*- sentds par divers savants, &c. a 1' Academic des Sciences, Paris, xiv. 1856). Histology of oesophagus and proventriculus. Klein, Strieker's Histology (Sydenham Soc.), i. p. 535> P- 554- Syrinx. Variations of vocal organs in Passeres, J. Muller, transl. by Bell, Oxford, 1878. Of Ratitae. Forbes, P. Z. S. 1881. Trachea, Convolutions of: Forbes, P. Z. S. 1882. Lungs. Aeby, Bronchialbaum der Saugethiere, &c., Leipzig, 1880, p. 93. Minute Anatomy, Schultze, Strieker's Histology (Sydenham Soc.), ii. p. 68. In Apteryx, Huxley, P. Z. S. 1882. Air-sacs. Strasser, M. J. iii. 1877. Campana, Recherches d' Anatomic, &c. Anatomic de 1'appareil pulmonaire, &c., chez le poulet, Paris, 1875. Milne Edwards, Lemons sur la Physiologic, &c. ii. p. 531. Heart. Sabatier, Le Cceur, Paris, 1873. Right auriculo-ventricular valve, &c., Gegenbaur, J. Z. ii. 1866, p. 375. Arterial system. Barkow, Meckel's Archiv f. Anat. u. Physiol. 1829. Carotid arteries. Garrod, P. Z. S. 1873. Venous system. Neugebaur, Nova Acta, xxi. 1845. Points in venous system. Wade, J. L. S. xii. 1876. Renal-portal system. Jourdain, A. Sc. N. (4), xii. 1859. Lymph hearts. Stannius, Archiv. f. Anat. u. Physiol. 1843 ; in the chick, Budge, same Archiv, Anat. Abth. 1882. 58 DESCRIPTIONS OF PREPARATIONS. Thymus. Afannassiew, A. M. A. xiv. 1877, p. 346. Kidney. Heidenhain, A. M. A. x. 1874, p. 20. Hufner, Zur Vergleich. Anat. u. Physiol. der Harnkanalchen, Dissertation, Leipzig, 1866. Oviduct. Blasius, Z. W. Z. xvii. 1867. Albumen glands (also of Amphibia), Loos, Z. W. Z. xxxv. 1 88 1. On the mechanism of flight. See Garrod, P. Z. S. 1875, and in Anat. of Passerine Birds, with 4 Plates, P. Z. S. 1876. Strasser, J. Z. xix. 1885. Marey, Animal Mechanism (Internat. series), xi. 1874. Pettigrew (same series), vii. 1874; and Trans. Roy. Soc. Edinburgh, xxvi. 1872. On muscles of wings. Schopss, Meckel's Archiv f. Anat. u. Physiol. 1829. Sundevall, Brit. Ass. Reports, 1855. Rolleston, Homology of Muscles of Shoulder Joint, Tr. L. S. xxxi. 1870 (cf. a paper by Fiirbringer, M. J. i. 1876). Garrod, P. Z. S. 1873 and 1874. Fiirbringer, M. J. xi. 1885. On muscles of hinder extremities. Garrod, P. Z. S. 1873 and 1874. Gadow, Zur Vergleich. Anat. der Muskulatur des Beckens, &c. der Ratiten, Jena, 1880. Value of innervation in determining homology of a muscle. Fiirbringer, J. Z. vii. 1873, p. 240. Cunningham, Journal of Anat. and Physiol. xvi. 1882. II. THE SKELETON OF THE COMMON PIGEON (Colmnba livia). A Bird's skeleton has characters both peculiar and well-marked. The bones undergo extensive anchylosis especially in the skull, pelvic region, hand and foot The bone-substance is dense, and is stated to contain a large proportion of lime phosphate. The cancellated tissue with the marrow is frequently absorbed and its place taken by air derived in the skull from the nasal passages and tympanic cavities : in the rest of the skeleton from extensions of the air-sacs connected with the lungs. All the bones in the Pigeon contain air, i. e. are pneumatic, save the caudal vertebrae, the fore-arm and hand and hind-limb. The skull presents the following general features : — It has, like the Reptilian skull, a single condyle. The cranial surface is smooth, polished, and its sutures obliterated. The orbital cavities are large and separated one from another by a thin vertical inter-orbital septum, formed chiefly by the mesethmoid. The orbit is bounded in front by the lacrymal bone and the homologue of the lateral mass of the ethmoid in man. The beak consists mainly of the praemaxillae, which are continued backwards below the orbit by a slender bony rod. The anterior part of this rod is the maxilla, the posterior which articulates with the quadrate, a jugo-quadrato- jugal bone. A moveable quadrate articulates with the skull on the one hand, and the lower jaw on the other. The two rami of the mandible are anchylosed at the symphysis. A ring of bony plates developed in the sclerotic of the eye is seen suspended in the left orbit. A similar ring is found in many Lacertilia, the Chelonia, and extinct Ichthyosauria. The vertebral column is divisible into a cervical, dorsal, so called COMMON PIGEON. 59 sacral, and a caudal region. The articular surfaces of the centra are typically procoelous and cylindroidal, i. e. concave from side to side and convex from above downwards anteriorly, curvatures which are reversed posteriorly. The cervical and dorsal vertebrae have synovial joints. A ring of fibres binds together the edges of the opposing surfaces. Between them is interposed a fibre-cartilaginous meniscus thick at its circumference, thin centrally where it is perforated for the passage of the ' suspensory ligament ' which unites the vertebral centra. The free caudal vertebrae have usually flattish centra. They articulate as in Mammalia by inter- vertebral discs (annuli fibrosi), which have centrally a ' nucleus pulposus/ the homologue of the suspensory ligament and formed as it is from notochordal cartilage. The length of the neck in every Bird is at least equal to the height from the ground at which the legs carry the body and to the distance from the root of the neck to the last caudal vertebra. The actual length varies, and depends chiefly on the number of vertebrae present, and not on the length of their centra. In Mammalia while the number of vertebrae is nearly invariable, the length of their centra is very variable. The atlas is ring-like and articulates with the occipital condyle by a deep cup completed as in Reptilia by the odontoid process of the axis. This latter vertebra^ has a neural spine, and an inferior spine, a structure present also in the two next vertebrae. The odontoid is anchylosed to its centrum. The third vertebra has its neural arch deeply emarginated before and behind, a peculiarity repeated in the seven following vertebrae. It has also, like the next nine vertebrae, cervical ribs anchylosed to the superior and inferior transverse processes, and inclosing a canal which lodges the vertebral artery and vein with the main trunk of the sympathetic. The fourth vertebra has no neural spine, and its centrum has ventrally a pair of down- growths which form a demi-canal for the protection of the common carotid arteries. These features are repeated in the succeeding vertebrae to the tenth inclusive. The eleventh and twelfth have inferior spines ; the two following — the thirteenth and fourteenth — have neural spines ; their ribs are free but not connected to the sternum, and the last pair carry recurrent or uncinate processes like the first four pairs of dorsal ribs. By some anatomists the thirteenth and fourteenth vertebrae are counted as dorsal. Five vertebrae make up the dorsal region defined by the presence of free ribs connected to the sternum. They have large neural spines and transverse processes with keel-shaped centra. The three first have their centra anchylosed, and the ligaments connecting their neural spines, inferior spines, and transverse processes ossified. The fourth dorsal is free : the fifth unites with the sacrum. The sacrum contains, as it always does in Birds, vertebrae derived from four regions. The first is a dorsal vertebra with ribs. The six vertebrae succeeding it are lumbar, of which the three 60 DESCRIPTIONS OF PREPARATIONS. last retain only the upper division of the transverse processes, which are present also in all the remaining sacral vertebrae. The ligaments uniting these processes ossify, and a flat subcutaneous area (absent in Ratite birds) is thus formed. To the lumbar succeed two sacrals, homologues of the vertebrae so named in Lizards, &c. In some specimens of the Pigeon, both of them carry a pair of stout bony rods or ribs visible only from below. These ribs are not free. They swell at their distal ends which fuse to the outer ends of the transverse processes (upper division) of their own vertebrae and coincide with the widest part of the area mentioned above, and lie therefore just internal to the acetabulum. Either the first or the second pair of these ribs may be absent. Behind the sacral vertebrae comes a variable number of caudal vertebrae, termed for distinction's sake * uro-sacrals.' The free caudal region contains seven to eight vertebrae. They have no articulating processes. The last is thin, compressed, and up-turned. It is known as ' pygostyle ' or ploughshare bone (os en soc de charrue), and represents four to six fused vertebrae. The five pairs of ribs consist of ossified vertebral and sternal sections. Each vertebral section articulates only with its own vertebra : the first four bear uncinate processes united to their posterior borders. These processes, as in Hatteria, Iguana and the Crocodile, are pre-formed in cartilage. The last sternal section unites with its predecessor, not with the sternum. The sternum covers nearly the whole abdomen and has a deep concave internal surface. There are four borders : — an anterior bearing the rostrum in its centre with a coracoid groove on either side and ending laterally in a costal process ; a right and left costal border deeply concave and bearing the ribs ; and a posterior or xiphisternal border. This border is convex, of great extent, and interrupted, as in some other Birds, on either side the median line by an outer xiphisternal notch and an inner xiphisternal fontanelle — the inner notch of the Fowl tribe. Notch and fontanelle are in the living bird closed by membrane probably substituted for original cartilage. The border presents accordingly five processes — two outer, two intermediate, and one median. The convex outer surface of the sternum carries the keel or carina, whence comes the name Carinatae, applied to the vast majority of living birds as opposed to the Ostrich and its allies, known as Ratitae from the raft-like aspect of the keel-less sternum. The shoulder girdle consists of a scapula, coracoid, and furcula 1. The scapula is sword-shaped and thin. There is no separate suprascapula. A small conical process internal to the glenoid facet represents the meso- scapula or acromion. The coracoid is firmly united by ligament to the scapula. A prominent clavicular process rises in front of its glenoid facet, and there is a thin curved subclavicular, or subscapular process ( = praecoracoid), 1 This word is often, but incorrectly, written Furculw/w. It is written Furcula in Bronn's Klass. und Ordn. des Thierreichs, vi. Abth. 4 by Selenka, without the mention of any other form. SKELETON OF THE COMMON PIGEON. 6 1 on the internal, or true anterior, border in contact with the acromion. A rough line runs downwards from it to the broad sternal or epi-coracoidal end of the bone and gives attachment to the coraco-clavicular membrane. The coracoid fits into a groove in the sternum. The furcula, a characteristic Avian bone, is formed by the fusion of the ventral ends of the two clavicles. At its upper end each clavicle expands into a disc or epicleidium, which is tied by ligament to the acromion and to the sub-scapular and clavicular processes of the coracoid. There is thus formed a foramen triosseum through which the tendon of the second pectoral muscle, or elevator of the wing, passes to its insertion on the humerus. The spot where the ventral ends of the clavicles fuse is prolonged into a point, the homologue of the large hypocleidium in the Fowl, and united as it is to the keel of the sternum by a ligament which together with the point represents a portion of the interclavicle. The fore-limb of this specimen is in the position of rest. The humerus lies parallel to the axis of the body, its true ventral surface turned outwards ; the fore-arm is flexed on the humerus and the hand is adducted. When the wing is expanded, the hand is abducted : it is incapable of flexion. As to the humerus, its glenoid head is transversely elongated : on its radial or upper margin at the proximal end is a conical process to which the first pectoral, or depressor of the wing, is attached, and dorsally to it is the facet for the insertion of the second pectoral. On the ulnar margin proximally and dorsally is a deep pit, at the bottom of which is a pneumatic foramen. The surface of articulation for the radius is longyDblique and on the ventral surface, as in Lizards. The radius is rod-like ; the ulna stout, somewhat curved, and with a short olecranon. Its outer surface is pitted by the sacs of the secondary wing feathers. There are two carpal bones in the proximal row — a scaphoid (= radiale), and a fused lunar and cuneiform (= intermedium and ulnare). The distal carpalia are fused to the heads of the metacarpalia, forming a carpo-metacarpal bone. The first metacarpal is a mere process and carries a single phalanx ; the second is stout and long and carries two phalanges ; while the third is slight, curved and fused distally to the second, and carries but one phalanx. In no Bird are there more than these three digits. The pelvis has the three bones ileum, ischium, and pubis peculiarly disposed. The first extends backwards and forwards along the whole extent of the sacrum ; the ischium lies parallel to the backward extension of the ileum ; the pubes to the ischium, and neither of the two latter have a ventral symphysis. All three unite in the acetabulum. The centre of this cavity is membranous in the living animal. Hence in a prepared skeleton it appears to be perforated. A prominent surface — the anti- trochanter — on the posterior-superior margin of the acetabulum, works against the base of the neck which carries the head of the femur. Ileum 63 DESCRIPTIONS OF PREPARATIONS. and ischium fuse distally, and thus inclose an ileo-sciatic foramen. The obturator foramen between the ischium and pubes is long and narrow, and subdivided partially by the obturator process of the ischium. The femur is remarkably short. The head is prominent, and its neck at right angles to the main axis of the bone. The condyles are large and separated by a deep patellar groove. The external one is typically subdivided, and its outer subdivision plays between the tibio-tarsus and fibula. The former of these two bones is the largest in the limb. It has proximally a cnemial crest on the anterior surface, subdivided into a pro- and ecto-cnemial process ; and distally there are two condyles formed from a cartilage in the embryo which represents the proximal tarsalia — astragalus and calcaneum. Anteriorly and above these condyles a narrow bony bar confines the extensor tendons of the toes. The fibula is slender and pointed distally. The third section of the limb is the tarso-metatarsus, a compound bone formed by the union of a bone representing the distal tarsalia to the heads of the second, third, and fourth metatarsalia, of which the third is the longest. Behind the tarsal element lies an ento-calcaneal process, the attachment of the tendo Achillis, pierced and grooved by the flexor tendons of the foot. The first metatarsal is small, incomplete proximally, and united to the second by ligament. There are four digits in all — the first, or hallux, is turned inwards and backwards and carries two phalanges ; the three remaining digits carry phalanges increasing successively in number from three to five, the usual succession in Birds. The third is the longest digit ; the fourth is so only in a few instances, e. g. Penguin, Gannet, Pelican, &c. All the bones in the embryo contain marrow. The degree in which it is replaced by air varies much. Apteryx, Penguin, small Songsters, have air only in the skull : the Hornbill in every bone of the body. A membranous tube — the siphonium of Nitsche — conveys air from the tympanic cavity to the lower jaw, as in the Crocodile. This tube in the Raven, Thrush, &c. becomes bony. The Cretaceous toothed birds had a pneumatic skeleton, as was probably the case in the Dinosaurian Reptile, Coelurus. The outlines of the cranial bones can only be seen in a young skull. The single condyle is made up, as in some Reptiles, by the basi- and ex-occipitals. The parietals are short but wide. Two membrane bones — the basi-temporals— parts of the parasphenoid, underlie the base of the skull, and the rostrum, or anterior part of the parasphenoid, similarly underlies the mesethmoidal septum. The palatine bones extend forwards to the maxilla and articulate behind with the rostrum. The pterygoids articulate in front with the palatines and the basipterygoid processes of the rostrum, while behind they diverge and articulate with the quadrate. The maxillae have short palatal plates — ' maxillo-palatine processes' — which extend inwards above the anterior end of the palatines. The squamosal and jugal are only connected by ligament, not by bone, as in Mammalia. The Pigeon and Sandgrouse have no vomer, the Fowl has a pointed vomer SKELETON OF THE COMMON PIGEON. 63 at the anterior end of trie rostrum. The internal nares open between the rostrum and the palatines, and there is no hard palate. The Pigeon, like many birds, is in a permanent state of cleft palate (schizognathism), owing to the palatal plates of the praemaxilla and maxilla not meeting in the middle line. The lower jaw of the young Fowl has five bones in each ramus, a dentary, splenial, angular and surangular with one cartilage bone, the articular. Some Birds add a coronoid, thus attaining the standard of the Lacertilian. The symphysis in the Cretaceous toothed birds was ligamentous. The hyoid is characteristic. There are three median bones, one tongue- shaped, formed by the union of the ceratohyals, followed by the basihyal and a basibranchial. The first branchial arch is well developed, and consists of an upper epi- and a lower cerato-branchial. The joints between the several parts are synovial. Archaeopteryx (Jurassic), Ichthyornis and Apatornis (Cretaceous) had amphi- caelous vertebrae like the Geckoes and Hatteria among living Lizards. The third cervical of Ichthyornis shows transitional characters to the modern Bird, and closely resembles the corresponding vertebrae in the Tern (Marsh). A transitory amphi- caelous stage exists in the chick on the seventh day. Some of the dorsals in the Penguin, Auks, Plovers have spheroidal faces and are opisthocaelous. The cervical ribs remain distinct for a long time in Ratitae. The division between cervical and dorsal vertebrae is somewhat arbitrary. The late Professor Rolleston considered the two vertebrae with ribs not touching the sternum as dorsals, because the ribs indent the lungs, and the last pair carries uncinate processes. Professor Huxley considers them as cervicals because the ribs do not touch the sternum. But there is embryological evidence in favour of the former view. It has been shown by Miss B. Lindsay that two anterior ribs in the Fowl and one in the Gannet are continuous at an early stage with the sternum, but become separated from it subsequently by atrophy. The identification of two vertebrae as ' sacral ' and as homologous with the vertebrae so named in Lizards depends on the following points : (i) their predominant size in the embryo ; (2) trie presence of free ribs ossifying by separate centres in the embryo while the preceding vertebrae are devoid of them ; (3) that these ribs expand and fuse distally, as in the Crocodile ; (4) that they are in relation with the acetabulum ; and (5) that the nerve passing out between the ribs is the last and weakest factor in the plexus ischiadicus, as in Lizards. Anchylosis of three dorsal vertebrae is characteristic of the Peristeromorphae (Pigeon group), while four are similarly anchylosed in the Alectoromorphae (Fowl group), but the first is a cervical. Archaeopteryx has but five anchylosed sacral vertebrae, and the tail contains twenty vertebrae, of which the last fifteen are devoid of transverse processes, and carry each a pair of large feathers. Hesperornis has fourteen sacral and twelve caudal vertebrae, of which the last six or seven are anchylosed by their centra only, and are in other respects free. The sternum is formed from right and left plates of cartilage, constituted by the fusion of the ventral ends of the ribs. The absence of transverse segmentation universal in Mammalia is characteristic of Birds and Reptiles. The carina, according to Gotte and Hoffmann, is formed from a single or primitively double 64 DESCRIPTIONS OF PREPARATIONS. band of tissue continuous with the clavicles. It therefore represents in part the interclavicle of a Reptile. The anterior extremity of the band forms the hypo- cleidium and the interclavicular or sterno-clavicular ligament. Their view has been recently controverted by Miss B. Lindsay, who contends that the Avian sternum consists (i) of a costal sternum derived from the ribs ; (2) of a meta- sternum, apparently a growth from (i) ; (3) of anterior lateral processes (costal processes), either outgrowths of (i) e.g. Struthio, or formed from anterior ribs, e.g. Chick; (4) of a keel, an outgrowth of (2); and (5) posterior lateral (i.e. xiphisternal) processes derived from (2) also ; together with other structures some- times present. The subject requires re-investigation. The shape, disposition, &c. of the several parts of the sternum vary much in Birds. The scapula and coracoid are fused in Ratitae and connected by ligament in Carinatae. Each ossifies from a single centre. With rare exceptions (Psophia among Carinatae, Apteryx among Ratitae), the inner or anterior border of the coracoid becomes partially ligamentous. The sub-scapular process when large, e. g. Gull, Eagle, is pierced by a foramen ; when small, this foramen lies in the ligament. A foramen similarly placed exists in the complete coracoid of Psophia and Apteryx, and in the Crocodile and most Lizards. It transmits a nerve for a muscle, the second pectoral in the Bird. These facts point to a homology, as maintained by Sabatier, of the sub-scapular process of the coracoid with the praecoracoid of the Reptile, i. e. with its proximal extremity. In the embryo, as figured by Miss Lindsay, there appears to be a large praecoracoid rudiment. The ligamentous portion of the coracoid is well characterized by its tough nature and parallel fibrillation. It is invaded more or less by ossification in Ratitae, and in an old Ostrich its anterior margin ossifies, inclosing a coracoid fontanelle. In a Carinate it forms part of the coraco-clavicular ligament. The clavicular process of the coracoid gives attachment in Carinatae to the deltoideus minor muscle and principal ligament of the shoulder-joint. In Ratitae it is represented by a mere roughness or slight tuberosity. It must be considered as a process developed for the same reason and for the same purpose as the deltoid tubercle or supinator ridge in the humerus of some Mammalia, e.g. Armadillo. The structures attached to it are of prime importance in flight. The two clavicles ossify parosteally. They may be absent, as in all Ratitae save the Emu, and in some Parrots ; or fail to meet ventrally — Emu, Toucan, some Parrots and Owls. The upper end (epicleidium), small in the Pigeon, may be large, e. g. Goose, and is stated to be in this case cartilaginous in the embryo. The hypocleidium is small in the Pigeon ; it is large and directed downwards, e.g. Fowl, or backwards, e.g. Rook. A coraco-clavicular ligament unites each clavicle to the inner border of the coracoid, and a sterno-clavicular ligament unites the hypo- cleidium to the carina. Irregular ossifications may appear in these membranes. The proportions of the segments of the fore-limb one to the other vary much. In Ratitae they show scarcely a trace of the characteristic Avian position when at rest. The Cretaceous Hesperornis has only the humerus, and the limb is either absent or reduced in the extinct Dinornithidae. In Archaeopteryx the parts of the hand are free, and each digit appears to have borne a claw. A carpo-metacarpus exists in all other birds. In the embryo fowl a mass of cartilage (=carpalia i + ii) corresponds to the two first metacarpals, and a second (=carpalia iii + iv) to the third and embryonic fourth metacarpals. Uria grylle has, according to Morse, embryonic claws to the first and third digits. SKELETON OF COMMON PIGEON. 65 The bones of the pelvis are separate inter se in Archaeopteryx. In all other birds they fuse, at least in the acetabulum. The real length of the ilium is to be measured from the outer ends of the sacral ribs to the acetabulum. The great extension of the bone along the sacrum represents breadth, and is an exaggeration of a Crocodilian and Deinosauran feature. The praeacetabular section, with few exceptions, ossifies parosteally, and it is rarely shorter than the postacetabular section, as in the Ostrich, Divers, and Ichthyornis. The length corresponds with the iliac axis of Professor Huxley. As in all Sauropsida this axis makes an acute angle forwards, not backwards as in Mammalia, with the sacral axis, i. e. a line drawn antero-posteriorly through the centra of the sacral vertebrae. The dorsal iliac area is little developed in some birds, e. g. Ostrich, Divers. The backward inclination of the pubes and ischia, and the loss of the ventral symphysis is characteristic of birds among living Vertebrata, So far as the pubes are concerned, there was no symphysis in many of the extinct Deinosauria, and the same is true in some instances of the ischia also; and this is especially the case in the two groups Stegosauria and Ornithopoda, in which the conformation of the pelvis is Avian and the ischia with the main part of the pubes (post-pubes) are inclined backwards. The Ostrich is the sole example of a bird with a pubic symphysis, and in Rhea the ischia meet not ventrally, however, but dorsally imme- diately under the backbone. The union between the distal ends of the pubis and ischium of the same side, and of the latter with the ileum, does not always occur, e. g. Cassowary, Emu, Apteryx among Ratitae • the Tinnamou among Carinatae. There is a well-developed pectineal process in front of the acetabulum in the Ostrich. An examination of a young specimen shows that both ileum and pubis enter into its formation.' A similar process is found in some Carinatae; but it appears to be formed entirely by the ileum in the Fowl. The chick, however, as proved by Miss A. Johnson, has at an early period a large forward extension of the pubis which gradually dwindles away. This process appears to be the homologue of the prae-pubis (so-called) in the Stegosauria and Ornitho- poda among Deinosauria, whilst the main portion of the bird's pubis is the homologue of the post-pubis (so-called) in the same groups. There does not appear to be much ground for supposing, (as has been done) that the prae- and post-pubis represent separate bones. They are continuous one with the other in Stegosaurus (cf. Marsh, American Journal of Science and Arts, xxi. p. 169, PI. viii.), in Camptonotus (op. cit. xviii. p. 502), Laosaurus (op. cit. xvi. p. 415, PI. x.), and in Iguanodon (Dollo, Bull. Mus. Roy. d'histoire Naturelle de Belgique, ii. 1883, PI. iii.). And the solitary instance of Allosaurus, which was supposed by Marsh to have had a separate post-pubis (American Journal cited, xvii. p. 90, PI. vii. 2), is now included by that author in a group of Carnivorous Deinosauria, the Theropoda, in which the post-pubis is absent. See Marsh on Theropoda, American Journal of Science, xxvii. 1884. Baur has recently suggested that the pectineal process (in part) of Birds and of Deinosauria, or the prae-pubis in some of the latter group, is the homologue of the Os acetabuli of Mammalia (see note, p. 107, Z. A. ix. 1886). For the Os acetabuli, see Gegenbaur, Ausschluss des Schambeins von der Pfanne, &c., M. J. ii. 1876; Krause, Centralblatt fur Me- dicin. Wissenschaften, 1876, p. 817; Leche, Bronn's Klass. und Ordnungen des Thierreichs, vi. pt. 5, Mammalia, p. 576; Id. Monthly Internat. Journal of Anat. F 66 DESCRIPTIONS OF PREPARATIONS. and Histology, i. p. 363. For a figure of Iguanodon, see Moseley, Nature, xxviii. 1883, p. 441; or Dollo, op. cit. supra, PI. v. Cf. Hulke, Journal Geol. Soc. xl. 1884, p. 53; and Huxley, Anatomy of Vertebrated Animals, 1871, pp. 223-228. The process of the ischium which divides the obturator foramen into two portions is very large in the Deinosaur Laosaurus. The upper division of the foramen transmits in birds the tendon of the obturator internus muscle, a rather curious point. Of the segments of the hind-limb, the femur is remarkably short and very broad in Hesperornis and the Divers ; while the tibio-tarsus is of great length in Ratitae and Waders. The fibula is as long as the tibia in Archaeopteryx. It is so at one time in developing birds but shortens subsequently. The tarsal element of the tibio-tarsus has been found by Baur to appear in the embryo chick as two bones, a tibiale and a fibulare, corresponding to the tibia and fibula respectively. The tibiale develops an ascending anterior spur, while the tibia broadens out so as to cover the fibulare, and the fibula itself shortens. The two tarsal bones subsequently unite. According to Morse the ascending process corresponds in the embryoes of certain birds representing various groups (Alcidae, Laridae, &c.) to a separate intermedium. If this is so, there are three proximal tarsalia in some birds, each with its own ossific centre. The ascending spur of the tibiale (? = astragalus) is present in Deinosauria. The bony arch confining the extensor tendons in the Pigeon as they pass over the tarsal region of the tibio-tarsus is ligamentous in Ratitae. This ligament is united at one end to the tibia, at the other to a bony projection apparently developed on the fibulare (=calcaneum) both in the Ratite and the young Carinate. It probably corresponds to the anterior annular ligament of the human foot. The distal tarsal elements are represented by a single cartilage which corresponds eventually to the second, third, and fourth metatarsalia. The rudimentary fifth metatarsal fuses with it (Baur). The metatarsalia in Archaeopteryx are apparently free, or but slightly anchy- losed. In Penguins the three metatarsalia (ii., iii., iv.) are short and lie parallel to one another. They are not raised from the ground in them nor in Auks and Divers. Mt. iii. is the largest as a rule, but Mt. iv. is as large in some birds and much larger in Hesperornis. The calcar of the Fowl ossifies independently, but fuses with Mt. iii. Mt. i. is only anchylosed to Mt. ii. in Phaethon. All the toes are united by a common web in the embryo — a condition which persists in the Pelican, Cormorant, Solan Goose, and Divers. They rarely all retain their primitive forward position, e. g. in the Penguin, Swift, &c. The hallux is sometimes absent. Aves, Selenka and Gadow, Bronn's Klass. und Ordn. des Thierreichs, vi. Abth. iv. Birds, W. K. Parker and A. Newton, Encyclopaedia Britannica, (ed. ix.) iii. Oiseaux fossiles de la France, &c., Milne-Edwards, 2 vols. and Atlas, Paris, 1867-71. Eappareil locomoteur des oiseaux, Alix, Paris, 1874. Palatal and other characters. Huxley, P. Z. S. 1867, 1868. Pigeon. T. J. Parker, Zootomy, London, 1884, p. 182. Fossil Birds. Archaeopteryx. Dames, Palaeont. Abhandl., Berlin, ii. part 3, 1884. Marsh, Nature xxv. 1881-82. Vogt, The Ibis, 1880. Owen, Ph. Tr. 153, 1863. Cf. Baur, Z. A. ix. 1886, for complete lit. Odontornithes (Hesperornis, .Ichthyornis). Marsh, Memoir, United States Geological Exploration, 4oth parallel, COMMON RINGED SNAKE. 67 1880. Odontopteryx. Owen, Journal Geol. Soc. xxix. 1873. Dinornis,&c. Id. Ex- tinct Birds of New Zealand, 2 vols. 1879. -Dodo, Strickland and Melville, London, 1848. Solitaire, Newton and Clark, Ph. Tr. 168, 1878. Skull. Development and Structure, W. K. Parker, 'Fowl,' Ph. Tr. 159, 1869, Cf. Id. Morphology of Skull, London, 1877, p. 219. Of other Birds and groups, Monthly Microscopical Journal, 1872, 1873; Tr. L. S. (2) i. 1879; Tr. Z. S. ix. 1877; x. 1879. Structure of Pneumatic bones. Wildermuth, J. Z. xi. 1877. Vertebral column, &c. Articulations, Jager, SB. Wien. Akad. xxxiii. 1858. Axial skeleton. Mivart, Tr. Z. S. viii. 1874; x. 1879. Pygostyle. W. Marshall, Niederland. Archiv f. Zool. i. 1871-73, p. 194. Birds vertebrae. Marsh, American Journal of Science and Arts, xvii. 1879. Processus uncinatus. Behrens, Inaugural Dissertation, Gottingen, 1880. Shoulder-girdle. Hoffman, Niederland. Archiv f. Zool. v. 1879-82. Bunge, Inaugural Dissertation, Dorpat, 1880. Gegenbaur, Untersuch. zur Vergleich. Anat. der Wirbelthiere, ii. Leipzig, 1865. Sabatier, Comparaison des ceintures, Paris, 1880. Harting, L'appareil episternale des oiseaux, Naturkundigen Ver- handlungen, Utrecht, 1864. Shoulder and Elbow joint. Fiirbringer, M. J. xi. 1885. Shoulder girdle and Sternum. Gotte, A. M. A. xiv. 1877, p. 549. W. K. Parker, Ray Society, 1868. Miss Lindsay, P. Z. S. 1885. Sacrum and Pelvis. Gegenbaur, J. Z. vi. 1871 ; cf. Huxley, P. R. S. xxviii. l879> P- 399- Pelvis of Birds and Deinosauria, Baur, M. J. x. 1885. Prae-pubis, Miss Johnson, Q. J. M. xxiii. 1883. Carpus and carpo-metacarpus. Rosenberg, Z. W. Z. xxiii. 1883 Tarsus. Morse, Memoirs, Boston Soc. Nat. Hist. 1880. Baur, M. J. viii.' 1882. Cf. Id. op. cit. x. 1885, p. 446, and Z. A. viii. 1885. Foot in Birds. Forbes, Ibis, 1882. Rudimentary Hallux. Id. P. Z. S. 1882. Teeth in living Birds. Fraisse, Verhandl. Phys. Med. Gesellschaft, Wurzburg, xv. 1 88 1, SB. p. iii. Effects of artificial selection. Darwin, Animals and Plants under Domestication, i. caps. v. vi. (ed. ii.) 1875. See also his Index. 12. COMMON RINGED SNAKE (Tropidonotus Natrix), Injected and dissected so as to show the manner in which the viscera are arranged in situ. THE following external characters, for the most part discernible in this specimen, should be noted : — the absence of limbs : the transparent cornea-like structure covering the eye formed by the union of the eyelids : no external mark indicating the position of the ear : the slight furrow between the two rami of the lower jaw permitting the free extrusion of the tongue when the mouth is closed. In these points as well as in the total loss of the shoulder-girdle Ophidians differ from Lizards, but they agree with them in having the cloacal aperture transverse and a complete invest- ment of scales. These scales, with the exception of those upon the head F 2 68 DESCRIPTIONS OF PREPARATIONS. overlap one another. They are processes of the dermis with an epidermis, the outer layers of which are thick and horny. These layers, together with the outer coat of the fused eyelids and the labial organs of sense, are moulted periodically, apparently at intervals of one month, during the active summer life of the animal. There are three types of scales : the flat plates of the head with apposed edges : the series of broad ambulatory ventral scales extending from the throat to the divided scale which pro- tects the cloaca: and the triangular scales of the body, smallest, and strongly keeled near the median dorsal line. The Harderian gland of the eye and the labial glands of the mouth have been exposed by the removal of the skin. The lobed extremity of the former appears behind the eye : it extends below that organ and opens by a single duct at the inner angle. The true lacrymal gland which lies above the eye is absent in Serpents but present in Lizards, e. g. the Blind-worm. The labial glands of the upper jaw are divisible into two kinds, readily distinguishable in the freshly killed animal. The first kind is grey in colour and forms (i) the azygos rostral gland lying upon the praemaxillae and not exposed here, and (2) a series of composite glands, each with its own duct extending back to the angle of the gape. The second kind is naturally white, but assumes a yellow colour in spirit preparations. It is large, and opens by a single duct between the maxillary teeth, and is the homologue of the poison gland in the Viper. The glands of each ramus of the lower jaw form an unbroken series and are grey in colour. The integument has been divided m the median ventral line as far as the prae-cloacal scale, and then reflected to the right and left. The dark-coloured tongue formed by the hyoglossi muscles extends backwards from the chin. On either side of it is a white rod, generally considered to be the larger cornu of the hyoid bone. The common jugular veins lie to the outer side of these rods. They rise into view at the angle of the man- dible, and the right vein is well displayed in its whole course, the left only close to the heart. The trachea is seen externally to the right jugular close to the heart. This portion of it is dilated and marked by irregular ridges : the first portion lies dorsally to the tongue and has complete rings. The oesophagus, at first dorsal to the trachea, passes to the left side of the body where it becomes visible close to the heart. It dilates gradually into the stomach. Between the two jugular veins, and close to the heart, is the thymus gland, and between the latter and the heart three vessels are visible, one in front, the left aorta, one difficult to see in the middle, the carotid, and a third passing to the right, the right aorta. The left aorta reappears at the outer side of the left jugular vein, embracing the oesophagus. This organ is twisted to show the dorsal junction of the two aortae, beneath which a black bristle has been passed. The dorsal aorta in Snakes is not closely tied to the backbone as it is in other Vertebrata. COMMON RINGED SNAKE. 69 The pericardium has been cut away. The right auricle is large and has its wall removed to show the right auriculo-ventricular aperture. The left auricle is small and is crossed by the left jugular vein. The ventricle touches the small and rudimentary left lung, and rests upon the large and long right lung which lies behind the liver. Crossing the ventral surface of the right lung is the vena cava inferior, accompanied by the pulmonary artery which lies to its outer side. The liver is long and unilobar. A furrow on its outer surface lodges the vena cava inferior. The gall-bladder is large, and lies about a half inch from the posterior end of the liver. It is bent sharply upon its duct, beneath which a piece of blue paper has been passed. This duct and the bile duct unite inter se, and with the pancreatic duct in the substance of the pancreas, a globular gland lying on the intes- tine close behind the gall-bladder. The first portion of the intestine is straight, but from the pancreas onwards it is disposed in short abrupt coils. These are supported by a mesentery, but the peritoneal coat does not follow every turn of their course as is usual, but passes from the end of one coil to the end of the next succeeding. The coils are closely united by connective tissue. The lobed fat body commences about the level of the pancreas. It is fastened out on the animal's right side ; the branches of a vessel, the remnant of the epigastric vein, may be seen here and there among its lobes. The vessel in question joins the portal vein. About four inches from the liver is the right ovary with a single row of ova, and between it and the fat body is the vascular oviduct. The left ovary and oviduct have similar relations on the left side, but are placed more posteriorly. The same asymmetry is visible in the position of the two kindeys, organs consisting of a number of leaf-like lobes placed one behind the other. Close to the cloaca, the large intestine is seen lying between the two oviducts. It has been opened and a black bristle passed through it into the cloaca. The left oviduct has a white bristle similarly inserted into it. The skin and muscles behind the cloaca have been removed to show the two sacs, homo- logues of the two eversible sacs or intromittent organs of the male. The following points of anatomy may be noted, not visible in the specimen. The subcutaneous connective tissue is very scanty in amount and absent altogether on the abdominal surface. Nervous system. The olfactory lobes of the brain are swollen terminally and are long ; the prosencephala broad ; the cerebellum somewhat tongue-shaped and projecting over the fourth ventricle. The pituitary body is broad. The parts of the brain lie nearly all in one plane. There is no spinal accessory nerve. Special sense organs. There is no tympanic cavity. Among Lizards, the Geckoes (Ascalabota) Amphisbaenae and some others resemble the Ophidia in the fusion of the eye-lids. There is thus formed a lacrymal sinus. 70 DESCRIPTIONS OF PREPARATIONS. Circulatory organs. There is a sinus venosus formed by the union of the vena cava inferior and the right jugular vein. It opens into the right auricle by an aperture guarded by two valves. The left jugular opens into the auricle separately. A single pulmonary vein opens into the left auricle. The two auricles are separated by a septum, the free edge of which is produced into a right and left auriculo-ventricular valve. The ventricle has a single cavity partially sub- divided by a muscular band or septum on its anterior wall. To the right of this septum is the cavum pulmonale from which the pulmonary artery arises. The left side of the ventricular cavity is divisible in turn into a left cavum arteriosum into which the left auricle opens and a right cavum venosum from which arise to the right the left aorta, to the left the right aorta. Hence these two vessels cross at their origins. In contraction of the heart the septum isolates the cavum pulmonale completely. The three great vessels, i. e. two aortae and pulmonary artery have, as in all Reptilia, two semilunar valves at their origin. In the arterial system the right aorta gives off, first, two coronary arteries ; secondly, an arteria cephalica, the common origin of the two carotids; thirdly, an arteria collaris, which runs beneath the back-bone and ends close to the head. The left aorta gives off no branches. The subvertebral aorta has no caeliac axis ; there are several hepatic and renal arteries, and the right and left intercostal arteries arise by a common stem. The aorta is continued into the tail as the caudal artery. The pulmonary vein lies hidden by the vena cava. In the venous system, the right inferior jugular close to the heart receives two veins, a short anterior and a long posterior azygos. The vena cava inferior is formed by the union of the two efferent renal veins ; it receives the ovarian (or testicular) veins, and the hepatic veins in its course along the liver. The portal vein rises on the dorsal wall of the cloaca and receives the veins of the intestines, stomach, spleen, pancreas, the epigastric vein and the intercostal veins. It runs on the visceral surface of the liver to which it is distributed. The reni-portal veins of the kidneys are formed by the bifurcation of the caudal vein. There are anastomoses between them, the portal, and the epigastric veins. Digestive system. The tongue is partially contained in a sheath which opens anteriorly on the floor of the mouth. Leydig describes a paired and an azygos gland in connection with this sheath. The stomach gradually contracts to the pylorus, which is well developed. The last six inches of the small intestine are nearly straight ; it opens laterally into the large intestine, and there is in some specimens a short caecum (as in the Pythons) at this spot. The large intestine is about three and a half inches long; it gradually increases in calibre to the cloaca, from which it is marked off by a constriction. Respiratory system* The posterior portion of the lung is thin and not alveo- lated, and receives blood from the hepatic arteries. In Pythons the left lung is functional, though smaller than the right,. In Hydrophis cyanocincta — a marine serpent — the lung extends to the cloaca, beneath the back-bone. Urinary system. The ureter runs centrally between the reni-portal vein on the outer and the efferent renal vein on the inner side. There is no urinary bladder. Generative organs. The testes are elongate, rounded organs with a small epididymis, and placed asymmetrically like the ovaries. The vas deferens is thrown into short coils and accompanies the ureter of its own side, and opens COMMON RINGED SNAKE. 7 L finally into the terminal dilatation of the ureter, in which spermatozoa are found at the breeding season. Each ureter opens on a dorsal papilla in the cloaca.; and a groove leads from it to the apertures of the intromittent sacs. These are eversible, and their inner surfaces are clothed with epidermic spines. They are retracted by muscles. The right is often larger than the left ovary. The oviduct has a large ostium with an entire margin as in all Vertebrata, save Mammalia; The two oviducts fuse into a short vagina, which opens dorsally into the cloaca near its outlet. The two ureters open on a dorsal papilla quite close to the same outlet. The post- cloacal sacs of the female are smooth and non-eversible. Note on the poison glands. The white gland mentioned above in the English Grass-snake becomes much enlarged in those colubriform snakes in which one or more of the posterior maxillary teeth are grooved. Such serpents were termed Opisthoglypha by Bibron and Dumeril, Ophidia suspecta by Schlegeh But a serpent with furrowed teeth may be found in the same family as a serpent with none but solid teeth, e.g. Homalocranion with grooved and Calamaria with solid teeth in the family Calamaridae. Hence the terms have been abandoned as of no classificatory value. The members of certain families, however, among the colubriform snakes are always opisthoglyph, e. g. Psammophidae^ Dipsadidae, and some of them appear to be undoubtedly poisonous. The serpents belonging to the suborders, Proteroglypha and Solenoglypha, are all venomous in the highest degree. In the former there is a large furrowed tooth at the anterior end of the maxilla, and behind it a series of small solid teeth : in the latter the maxilla is much reduced, very moveable, and provided with but a single large furrowed tooth and the germs of its successors. The Proteroglypha in- clude the Elapidae, e. g. the Cobra di Capello (Naja tripudians) and the marine Hydrophidae ; the Solenoglypha, the Viperidae, e. g. the English Viper (Pelias\ and the Crotalidae, e. g. the Rattlesnake ( Crotalus horridus). In both these sub-orders the white gland (supra) reaches its maximum of development. The ligamentum zygomaticum ( = an unossified jugal) which stretches internally to the gland between the maxilla and the quadrate, develops a silvery fascia-like pouch enclosing the gland. One of the three divisions of the temporal muscle is attached to the in- ternal aspect of this pouch : and from its posterior end the masseter takes its origin and passes down to the mandible. The gland is also enclosed in a tough fibrous investment of its own Within this investment there is in Petias berus loose con- nective tissue with large lymphatic spaces said to be absent in Naja haje (Emery). The gland tubes are united together by a fibrous coat with which the loose invest- ment is continuous. The tubes are collected into bundles, and open into a single duct. This duct opens above the base of the furrowed tooth, and a fold of the mucous membrane surrounds both the base of the tooth and the aperture of the duct. Hence the poison flows down the channel of the tooth, when the contrac- tion of the muscles attached to the outermost capsule forcibly ejects it in the act of closing the jaw. The poison is not only secreted in the gland tubes, but is likewise stored within them. The quantity ejected is large at the first bite, but be- comes less and less with successive bites. In Elaps (? all species) the poison glands are much elongated, and reach far down the body. As a rule they do not extend beyond the angle of the jaw, 73 DESCRIPTIONS OF PREPARATIONS. Reptilia, Hoffmann, Bronn's Klass. und Ordn. des Thierreichs, vi. Abth. 3. Erpttologie gdnfrale, Dume'ril and Bibron, 9 vols. Paris, 1834-54. Reptiles of British India, Giinther, Ray Soc. 1861. Thanatophidia of India , Fayrer, 1874. Natrix. Bell, British Reptiles. London, 1839, p. 47. Integument with sense organs. Ley dig, A. M. A. viii. 1872 ; ix. 1873. Todaro, Atti dell' Academia dei Lyncei (3), ii. part 2, 1878 (Math. Nat. Class). Merkel, Endigungen der sensibeln Nerven, Rostock, 1880. Cf. Knauer, Z. A. ii. 1879. Organs of circulation. Jacquart, A. Sc. N. (4) iv. 1855. Briicke, Dk. Akad. Wien, iii. 1852. Rathke, ibid. xi. part 2, 1856. Heart. Sabatier, Etudes sur le Cceur, Paris, 1873. Renal-portal System. Jourdain, A. Sc. N. (4) xii. 1859. Organs of respiration. Milne-Edwards, Lecons sur la Physiologic, &c. ii. 1857; Schulze, Strieker's Histology, (Sydenham Soc.) ii. 1872. Organs of digestion. Teeth. Tomes, Ph. Tr. 165, 1875 : Leydig, A. M. A. ix. 1873; Gervais, Journal de Zoologie, ii. 1873. Glands, Leydig, A. M. A. ix. 1873; Meyer, Monatsberichte, Akad. Berlin, 1869; Emery, A. M. A. xi. 1875; Reichel, M. J. viii. 1882 ; Digestive tract, spleen, &c., Duvernoy, A. Sc. N. 26, 1832; 30, 1833. Stomach, Edinger, A. M. A. xvii. 1880. Excretory organs and genitalia. Braun, Arb. Zool. Zoot. Inst. Wurzburg, iv. 1877-78. Supra-renals. Id. op. cit. v. 1882. Reproductive organs. Martin Saint-Ange, £tudes de 1'appareil Reproducteur des Animaux Vertebra's, Paris, 1854 (Me'm. par divers savants, Acaddmie des Sciences, xiv. 1856). 1 3. VERTEBRA OF CONSTRICTING SERPENT (Python sp. ?). THE vertebral column of Ophidians is, according to most authorities, divisible into three sections. — A cervical region containing only an atlas and axis : a second region containing vertebrae very numerous and corre- sponding to the posterior cervical, the dorsal lumbar and sacral regions in other Reptiles, and bearing freely moveable ribs, important organs of loco- motion in these animals : and a third or caudal region. The last named is characterized by the anchylosis of the ribs to the centra, and the presence of bifid descending processes, which protect the caudal artery and replace the chevron bones found in all other Reptilia. The number of caudal vertebrae varies very much. These vertebrae of a constricting Serpent are selected from the second section of the column. Each one has the following characters. The centrum is short and prismatic. Its anterior surface forms a deep cup with thin prominent edges and the vertebrae is therefore pro-coelous ; its posterior surface has a ball peculiarly prominent. Both ball and cup are placed obliquely to the axis of the centrum. The inferior surface has a low ridge terminating in a knob posteriorly and representing the inferior spine (hypapophysis) seen in the anterior trunk vertebrae of the Python itself and in all the vertebrae of the common Ringed Snake and of venomous VERTEBRA OF CONSTRICTING SERPENT. 73 serpents. The neural canal is surrounded by a neural arch which is anchy- losed to the centrum as it is in L acertilia and in Chelonia with few exceptions. The neural spine is low. At its base anteriorly, there stands above the neural canal a bony wedge, the zygosphene, with articular surfaces looking obliquely outwards and downwards. In a similar position but posterior to the spine is a deep cavity, the zygantrum, with articular surfaces looking obliquely inwards and upwards. The wedge and cavity fit the one into the other in contiguous vertebrae. Similar structures are found in the Iguana among Lizards, and the Edentata with the exception of the Sloths among Mammalia. The articulating processes are remarkably large and flat The anterior pair (prae-zygapophyses), placed externally to the zygosphene, have their surfaces disposed typically, i. e. looking obliquely inwards and upwards, relations reversed in the posterior pair (post-zygapophyses). A low ridge connects the prae- to the post-zygapophysis of the same side. The size of the articular surfaces, their disposition, the depth of the cup and of the zygantrum, and prominence of the ball and zygosphene permit great freedom of motion and at the same time prevent dislocation. The fact that the cup and zygosphene are both anterior, and the ball and zygantrum posterior, constitutes a further safeguard in the same direction. Below the prae-zygapophysis lies the diapophysis or articulating surface for the rib. Its upper portion is convex in every direction, while the lower portion is concave from above downwards but convex from before back- wards. This lower portion in some Snakes, e.g. Rattlesnake, is much pro- longed ventrally. The atlas resembles the corresponding vertebra in Lacertilta and Chelonia. It consists of three pieces : one inferior, prolonged ventrally into a spine ; and two, one on either side, forming the neural arches. There is no neural spine. The odontoid process is united by anchylosis to the centrum of the axis. It carries an inferior spine united to it, at least in young specimens, by a suture and representing perhaps a sub-vertebral wedge-bone such as exists in many Lizards between the centra of two adjoining vertebrae. There are no ribs to the atlas and axis, but in a specimen of Python (sp. ?) in the Oxford Museum cartilaginous representatives of these structures exist The anterior caudal vertebrae, to the number sometimes of ten but never more, have ribs apparently forked at their vertebral end. The ventral division of the fork is perhaps an outgrowth from the centrum ; i. e. represents the lower transverse process (= parapophysis). The lymphatic heart is lodged in the space enclosed by the fork. Vertebral column. Owen and Bell, Reptilia of the London Clay, Part iii. Palaeontographical Society, 1850; De Rochebrune, Journal de 1' Anatomic, &c. (Robin), 17, 1 88 1. Skull of Common Snake. W. K. Parker, Ph. Tr. 169, 1878. 74 DESCRIPTIONS OF PREPARATIONS. 14. COMMON FROG (Rana temporaria), Injected and dissected so as to show its nervous, circulatory, and respiratory systems, together with some of its reproductive and digestive organs. THE following external points may first be noted : the two slit-like external nostrils at the end of the snout, placed dorsally and widely separated : the large eyes partially covered by the thin moveable lower lid, the homologue of the third lid or nictitating membrane of higher Verte- brata, the upper lid being immoveable : the dark round smooth spot behind each eye where the outer skin adheres to the tympanic membrane : the pigmented skin with its innumerable minute vessels, injected in this specimen and correlated with the presence of the numerous glands charac- teristic of the soft moist transpirable skin of the Amphibia : the projections of the epidermis visible to the eye only when the specimen is held in certain positions, and best developed on the dorsum proper and the dorsal surface of the hind limbs : the cloacal aperture placed somewhat dorsally between the hind limbs : the short fore-limb : the base of the second finger (= index), dilated as it always is in the male at the breeding season : and the long hind limb with its webbed foot The brain and abdominal viscera have been exposed by the removal of the skin, muscles, cranial roof, and left fore-limb. A space separates the skin from the underlying structures. It is divided into lymph sacs, fifteen in number, by vertical septa or by the union of the skin to the parts beneath. The two cerebral hemispheres of the brain are visible anteriorly : narrow in front where they pass into the olfactory lobes which are not exposed in this preparation, and broad behind where there is a diamond- shaped interval occupied in the centre by the base of the pineal gland and to either side by the optic thalami. Next come the large oval and obliquely placed optic lobes or corpora bigemina. A narrow trans- verse band behind the optic lobes represents the cerebellum. It leaves the large fourth ventricle or sinus rhomboidalis completely exposed. The following structures in the body are visible from the ventral surface. Close to the angle of the lower jaw on the left side is an aperture, the passage to the croaking sac present only in the male. The heart lies medianly. It has been turned out of the pericardium, the thin membrane seen lying immediately below it. It consists of a conical yellowish opaque ventricle, separated by a well-marked auriculo-ventricular furrow from the thin transparent right and left auricles. The conus arteriosus lies ven- trally between the auricles and is continued on into the truncus aortae which ' is extremely short in the Anura, the division of the Amphibia to which the Frog belongs. It appears to divide into a right and left half, each of which really consists of three vessels, carotid in front, aorta in the middle, pulmonary artery behind, none of them readily visible here. But the left COMMON fROG. 75 vena cava superior may be seen between the root of the left lung and the left auricle where it dips down on its way to the sinus venosus. The liver with its right, central, and left lobes lies inferiorly to the heart. Some coils of the intestine are to be seen below the right lobe and leading downwards in the middle line from the median lobe is the epigastric vein. At the lower end of this vein the left division of the bifid allantoid bladder projects sideways : the right division lies behind the vein with a small portion of the rectum visible just behind it. Turning the preparation so as to show the left side, the deeply injected left lung is observed to occupy the space between the left lobe of the liver and the cut edges of the dorsal skin and muscles. The oval yellow testis with black pigment speckled on its surface is placed dorsally, and posteriorly to the lung. At its lower end the vesicula seminalis projects touching the pylorus. The stomach and duodenum lie between the left and median lobes of the liver, and the lung and testis. The stomach tapers to the pylorus. A remarkable hermaphrodite Nematode worm, Angiostomum nigrove- nosum (=Ascaris or Leptodera nigrovenosd], is commonly found in the lungs. Its young pass through the intestine and become sexually mature in earth saturated with decaying animal matter. They are minute compared to their parent and their progeny wander back into the lung. A Trematode parasite, Polystotnum integer rimum, with six posterior suckers and two large hooks, frequently occurs in the allantoid bladder : and the multi-nucleate Holotrichous Infusorian, Opalina Ranarum, inhabits the rectum. The outer layers of the epidermis are shed periodically — a process which does not extend to the cornea of the eye. The skin-glands are of two kinds: (i) mucous glands with clear contents ; (2) glands with granular contents, and prob- ably poisonous like the similar glands in the Toad and Salamander. Chromato- blasts connected to nerves occur in the epidermis, but are most numerous in the outer layers of the corium. The concentration and diffusion of the black pigment within the cells depends on a reflex action of the nervous system set up through the eyes. In this manner the colour of the animal becomes adaptable more or less to its surroundings. The adult frog retains no trace of the organs of the lateral lines present in the tadpole. Besides free nerve-endings, the skin contains special ' touch spots' ' — flat cells lying near one another, each supplied by a nerve-filament. The two olfactory lobes of the brain, generally so-called, are connected an- teriorly by a transverse commissure : each lobe has a second root. They are solid, and it is probable that * small papillae, situated at their base from which the olfactory nerves spring, and which contain a process of the lateral ventricle, should properly be regarded as the olfactory lobes. These papillae arise prior to the solid anterior prolongations of 'the hemispheres ' (Balfour). The pineal gland is long and fila- mentous, pierces the cranial roof, and is attached to the skin at the level of a line drawn through the anterior angles of the two eyes. With the exception of the tips of the olfactory lobes, the several parts of the brain contain ventricles. The facial t F 76 DESCRIPTIONS OF PREPARATIONS. and auditory nerves have a common origin, and the ganglion of the former unites with the Gasserian ganglion. The ganglion of the glossopharyngeal is fused with that of the vagus, and the hypoglossus is the first spinal nerve. Of the latter there are ten in all. The anterior and posterior nerve-roots unite outside the neural canal. Round the ganglia, formed at the place of union, there lie sacs containing crystals of calcium carbonate — the so-called glands of Swammerdam. The Gasserian gan- glion is similarly surrounded. The sympathetic trunk commences at the Gasserian ganglion, is connected with the vagal ganglion, and passes out with the vagus through a foramen in the exoccipital. A ramus communicans connects it with each of the spinal nerves from the first to the sixth inclusive, and a single ganglion, corresponds to each ramus. But the posterior spinal nerves, especially the tenth, give off a variable number of rami communicantes, and the ganglia vary corre- spondingly. The first and second sympathetic ganglia send important (accelerator) twigs to the heart : the fifth sends a twig to the caeliac plexus, which is in connec- tion with the root of the mesenteric artery. The sinus venosus of the heart is divided into a large right moiety receiving the two venae cavae superiores and the vena cava inferior, and a small left moiety receiving the pulmonary veins. Two valves guard the aperture into the right auricle. A thin non-muscular septum separates the two auricles : its lower free edge is adherent to the two auriculo-ventricular valves, one anterior, the other posterior. Their free edges and under surfaces are tied by chordae tendineae to the walls of the ventricles. These walls are produced into trabeculae which have a fixed direction and therefore influence the course of the arterial and venous blood-currents respectively. A conus arteriosus leads from the ventricle : its walls are yellowish, semi-transparent, and contain striated muscular fibre. From it springs the truncus aortae which is extremely short and gives origin at once to a right and left branch; its walls are whitish, somewhat opaque, and contain only smooth muscle fibres. The conus is separated from the ventricle by three valves, and from the truncus also by three valves. Of these latter, one is a large right valve prolonged as a spiral fold down the dorsal wall of the conus ; the other two are small valves lying a little to the left, one dorsally, the other ventrally. A vertical septum, placed transversely, divides the origin of the pul- monary arteries from the origins of the aortae and carotids. It is continued into the sinus of the large valve, is fixed to its free edge and to the wall of the conus between the two small valves. The cavity of the conus is consequently divisible physiologically into a dorsal portion which leads to the pulmonary arteries and a ventral from which arise carotids arid aortae. Each half of the truncus contains three vessels, carotid anteriorly, aorta in the middle, and pulmonary artery posteriorly. A fibrous band replaces a vessel, the ductus Botalli or primitive union between the carotid and aorta of each side. At the spot where the carotid splits into the lingual artery, which corresponds as in Reptilia to the external carotid of Aves and Mammalia, and into the carotid so-called in Amphibia, or internal carotid of higher Vertebrata, there is a rete mirabile, the carotid gland, formed by the development of anastomoses between the lingual or external -carotid, and the first branchial artery of the Tadpole (the common stem of the lingual and carotid of the adult) where they are contiguous to one another. The two aortae unite into a sub-vertebral aorta under the backbone, and the left arch gives off a COMMON FROG. 77 large caeliaco-mesenteric artery before it unites with the right. The pulmonary artery divides into the pulmonary vessel proper and the cutaneous artery. It represents the fourth aortic arch of the Tadpole, the third disappearing. Each cava superior traced upwards breaks up successively into the external jugular, the vena anonyma, and the cutanea magna, which receives the subclavian vein. The cava inferior is formed by the union of the efferent renal veins and receives the veins of the genital glands, the fat bodies, and near the heart the hepatic veins. The epigastric vein divides proximally into three branches, one for each lobe of the liver, and a third which takes up the portal and splenic veins and then enters the left liver lobe. Traced downwards, this vein receives first a cardiac vein from the conus, secondly veins from the right and left abdominal parietes, finally the veins of the allantoid bladder. It then divides into a right and left branch traceable respectively to the right and left femoral veins. Each femoral vein gives off also the reni-portal vein which passes to the outer side of the kidneys. Into each reni-portal falls a dorso-lumbar vein, and in the female the oviducal veins. Anastomoses exist between the allantoid, rectal and oviducal veins. The spleen is a reddish-brown body attached to the mesentery at the level of the commencing rectum. A pair of lymphatic hearts opening into the subscapular veins lie posteriorly to the outer ends of the transverse processes of the third vertebra. A second pair lie one on either side the urostyle posteriorly, and opens into a vein which falls into the communicating vein between the femoral and ischiadic veins. A large lymphatic sac, cisterna lymphatica magna, lies at the back of the abdomen. Its ventral wall, formed by the peritoneum, is pierced by microscopic apertures or stomata. There is an upper but no lower lip. On the roof of the mouth are (i) the minute apertures of the intermaxillary glands, the homologues of the internasal glands of the Urodela, immediately behind the fold of mucous membrane which protects the praemaxiliary teeth; (2) the internal nares to the outer side of the vomers; (3) the Eustachian tubes close to the articulation of the lower jaw. The .broad flat tongue is affixed to the symphysis of the mandible : its free end is bifid. In the male an aperture on either side, close to the ramus of the lower jaw, leads to the croaking sac. The teeth are conical, and restricted to the upper jaw and vomers. They consist of enamel, dentine, and a bony base or pillar. New teeth are continually formed during life to replace those that are worn or broken away. A short oesophagus leads into a stomach which is at first dilated, then narrows to the pylorus, and lies on the left side of the body. It is surrounded to a great extent by a lymph sac. The duodenum is bent at a sharp angle with the stomach. The coils of the intestine lie on the right side, and end in a short median dilated rectum which opens into the cloaca dorsally to the aperture of the bifid allantoid bladder. The gall-bladder lies in the notch between the median and right liver lobes. The pancreas is a thin lobed gland lying between the stomach and duodenum ; its duct enters the bile duct which opens on the dorsal wall of the duodenum. The aperture of the larynx opens on the ventral wall of the oesophagus; it is protected by the two arytenoid cartilages and a ring-like cricoid and leads to the two large oval lungs. The lungs have much reticulated walls and are covered externally by peritoneum. The thymus is a small gland lying close behind the angle of each jaw. The thy- roid is paired, and each body lies in the angle between the two, i. e. the large and small posterior horns of the hyoid, and adheres to the vena jugularis of its side. 78 DESCRIPTIONS OF PREPARATIONS. The kidneys are concealed by the fat bodies and genitalia. The former are golden yellow in the fresh state, vascular, and of unknown use. The testes vary in size according to the time of the year. The vasa efferentia run towards the inner border of the kidney: the majority fall into a longitudinal vessel close to that border, while a few may end blindly. From this vessel transverse canals, dilated at their origin, pass outwards, receive tubuli uriniferi, anastomose, and fall into the ureter or Wolffian duct, which runs on the outer margin of the kidney. In the Toads the efferent vessels unite with functional Malpighian corpuscles, represented perhaps by the dilatations (supra] in the Frog. A solid remnant of Miiller's duct may be found running forwards on the outer margin of the kidney. It ends blindly. In the Toads this remnant is large, hollow, and uniting with its fellow, opens on the dorsal wall of the cloaca. Each ovary is divided by internal septa into fifteen sacs : the outer surface is lobed. The two oviducts commence by slit-like openings close to the roots of the lungs. At first narrow and straight they become convoluted and glandular, and finally, near their termination, thin-walled and dilated. Narrowing again, they open on papillae situated anteriorly to the openings of the ureters on the dorsal wall of the cloaca. The kidneys are semi- lunar glands covered by peritoneum only on their ventral faces. Whitish spots, especially numerous near the efferent veins, mark the position of the nephro-stomata or ciliated funnel-shaped apertures which open into the abdominal cavity and are connected by ciliated tubes with the tubuli uriniferi in the Tadpole (as in Urodeles throughout life); but in the adult, according to Nussbaum, with the reni-portal capillaries. The ureter (Wolffian duct) begins at the anterior end of the kidney, runs along the outer margin, and, in the male, is dilated and glandular just where it quits the kidney, forming the receptaculum seminis. Both ureters open on the dorsal wall of the cloaca separately. The supra-renal capsules are yellow when fresh and e in the median line on the ventral face of the kidneys. Amphibia, Hoffmann, Bronn's Klass. und Ordnungen des Thierreichs, vi. Abth. 2. 1873-1878; Huxley, Encyclopaedia Brit. (ed. ix.), i. Anura Batrachia d. Deut- schen Fauna, Ley dig, Bonn, 1877. Rana temporaria. Bell, British Reptiles, London, 1839, p. 84. The Frog. Ecker and Wiedersheim, Anatomic des Frosches, Brunswick, 3 parts, 1864, 1 88 1, 1882 (with references to literature), in process of translation. A. M. Marshall, Owen's College Course of Elementary Biology i. (ed. 2), 1885. Figures. Atlas of Practical Elementary Biology, Howes, 1885. Integument. Leydig, Allgemeine Bedeckungen, &c. A. M. A. xii. 1876. Chro^ matoblasts and change of colour, Lister, Ph. Tr. 1858, p. 627. Connection of chro- matoblasts with nerves, Ehrmann, SB. Akad. Wien. Ixxxiv. Abth. 3. 1882. Nerve- endings. Merkel, Endigungen der sensibeln Nerven in der Haut, Rostock, 1880. Cf. On the Epidermis of Salamander, Pfitzner, M. J. vi. 1880. The periodical moult of cuticle, Knauer, Z. A. ii. 1879. Digestive tract. Teeth. Tomes, Ph. Tr. 1875; Hertwig A. M. A. xi. (Suppl.), 1874. Intermaxillary gland, Wiedersheim, Z. W. Z. xxvii. 1876. Tongue, organs of Taste. Engelmann, Strieker's Histology (Sydenham Soc.), iii. 1873, p. 14. Cf. Z. W. Z. xviii. 1867-69. Glands of Stomach. Partsch, A. M. A. xiv. 1877. Cf. Swiecicki, Arch. f. Physiol. (Pfliiger's), xiii. 1876. Lungs. Schulze, Strieker's Histology (Sydenham Soc.), ii. 1872, p. 72. SKELETON OF COMMON FROG. 79 Heart, vessels, carotid gland. Boas, M. J. vii. 1881, p. 488; cf. especially pp. 502 and 540; M. J. viii. 1882. Sabatier, Etudes sur le Cceur, &c., Paris, 1873 ; Fritsch, Archiv. f. Anat. und Physiol. 1869. - Spleen. Phisalix, A. Z. Expt. (2) iii. 1885. Lymphatic system. Langer, SB. Akad. Wien. viii. Abth. i. 1866; iv. Abth. i. 1867. Urogenital system. Spengel, Arb. Zool. Zoot. Inst., Wurzburg, iii. 1876. Con- nections of nephrostomata. Nussbaum, Z. A. iii. 1880. Spermatogenesis. Blomfield, Q. J. M. xxi. 1881. Albumen glands of Oviduct. Loos, Z. W. Z. xxxv. 1 88 1. 15. SKELETON OF COMMON FROG (Rana temporaries), With Figures 6 and 7. THE skeleton, as is commonly the case in Amphibia, retains a large amount of cartilage, in the deeper portions of which much calcareous matter is deposited, but not in the form of bone. PrO. JSO. IM. #.%./ \ 1D \ • }D.MM*- |^XC°- Mck. Sf.h. Eo7\).c.\ Evu. FIG. 6. FIG. 7. After Professor W. Kitchen Parker. The cartilage is dotted ; the cartilage bones are shaded with oblique lines, and the membrane bones left white. The lettering is explained in the text. The skull is remarkably flat, and the cranium proper of small size, though the total breadth is great owing to the large size of the orbital fossae. Its parts may be readily identified with the help of Figs. 6 and 7. There are two condyles (O. c.) formed by the two exoccipitals (E. (9.) which are pierced each by a foramen for the vagus nerve (X). The otic capsules project laterally but contain only one ossification, the pro-otic (Pr. O.} which is pierced by the fifth nerve ( V). The cranium in front of the otic capsules is cartilaginous and the cartilage is pierced by the optic nerve (//), but most anteriorly of all it contains a bone, the girdle bone, os en ceinture, or sphen-ethmoid (Sp. Et.}. This bone has an internal vertical septum. On the upper surface of the cranium lie the fronto-parietals (P. F.) : on its 8o DESCRIPTIONS OF PREPARATIONS. under surface the large para-sphenoid (P. S.), both developed in membrane. Other membrane bones are the two nasals (Na.): the two praemaxillae (P. MX?) and the two maxillae (Mx^ which carry teeth in a simple series : and the quadrato-jugals (Qu. y.) which continue the line of the maxillae back to the distal ends of the quadrate cartilage (Qu.}. The apertures of the external nares (E. n.) lie just in front of the nasals. On the roof of the mouth, in front of the sphen-ethmoid, are the two dentigerous vomers ( F0.), and at their anterior ends are the internal nares (/. «.). The palatine bone (Pa.), an investing membrane bone, is here placed transversely — a rare position seen again in Ichthyosaurus : while the pterygoid (Pt.) lies parallel with the cranial axis. The latter is deeply forked behind : the outer process underlies the quadrate cartilage : the inner or pedicle is continuous with the quadrate and articulates with the ear-capsule. The squamosal (Sg.) is seen in the lateral view. It is the homologue probably of the squamosal + the praeopercular of bony Ganoids, two bones actually fused in Polypterus. The complex stapes (6V.) or columella auris, the homologue of the Perch's hyomandibular ; the cartilage ring or annulus tympanicus (A. T.), the homologue of the Ray's spiracular cartilage (?), which gives support to the tympanic membrane, and the hyoid apparatus are not present in this specimen (see p. 81, infra]. The lower jaw consists of the two typical rami. The bulk of each ramus consists of an unaltered Meckel's cartilage (Mck.)t to which there is added distally an ossification known as the mento- meckelian (M.Mck.) formed by the ossification of the lower labial cartilages. Meckel's cartilage is covered in part by two membrane bones, one on the outer side for a short distance of its distal part, the non-dentigerous dentary (D): the other on the inner side, the articular (At) as- it is termed by Professor W. K. Parker, but identified formerly by Professor Huxley as a representative of the angular, coronary and splenial elements of the Sauropsidan lower jaw. The vertebral column numbers nine vertebrae and a urostyle. The first vertebra is the only cervical vertebra : it is ring-like with a shallow centrum, two concave facets for the occipital condyles and two post- zygapophyses. The centra of the second to the seventh vertebra inclusive are pro-coelous. The centrum of the eighth is amphi-coelous, and of the ninth or sacral vertebrae biconvex anteriorly, the posterior surface being divided into two facets for the articulation of the urostyle. The neural arches are thin and narrow centrally, leaving in this region interspaces, one between the other. The articulating processes or zygapophyses are flat, and disposed normally. All the vertebrae except the first have lateral processes tipped with cartilage and varying in length, size, and direction. These appear to represent ribs fused to transverse processes. The urostyle is long and curved. Its posterior end lies just above the cloacal aperture. It is pierced laterally at its anterior end by the tenth pair of spinal nerves, SKELETON OF COMMON FROG. 8 1 and lodges the posterior part of the spinal cord. The neural canal is often open terminally. The shoulder-girdle is complete : and the glenoid cavity divides it into a dorsal scapular, and a ventral coraco-clavicular, moiety. The former con- sists of a broad semicartilaginous supra-scapula and an ossified scapula, the latter of an anterior clavicular bar separated by a fenestra from a posterior coracoid. Both clavicle and coracoid are Connected by cartilage medianly to one another and to the corresponding parts on the opposite side. In the middle line project (i) anteriorly a conical bone ending in a cartilage plate — the episternum or anterior part of the interclavicle : (2) posteriorly a flattish bone with a cartilage disc, the hypo-sternum. The pelvis has the V-shape characteristic of Anura or tail-less Amphibia. The long ilium trends backwards to a disc-like symphysis into which it enters, and which is partly cartilaginous and partly formed by the two ischia. The acetabulum is a deep cup. The fore-limb consists of a humerus, and of a radius and ulna fused together, a furrow indicating externally the line of union. The elements of the carpus are small and consist in the proximal row of a scaphoid ( = radiale) and a cuneiform ( = ulnare): of a centrale displaced to the radial side : and in the distal row of a trapezium ( = carpale i) carrying the thumb, represented by a single bone: a trapezoid ( = cp. 2) carrying the second digit, which in the male enlarges at the breeding season, and a single element ( = cp. 3, 4, and 5?) carrying the remaining three digits. The long hind-limb consists of a femur, and of a fused tibia and fibula, A long astra- galus ( = tibiale) and calcaneum ( = fibulare) united at each end form the proximal row of the tarsus. The distal row consists of a tarsal element which carries a single bone, the extra digit or sixth toe ; of a fused meso- and ecto-cuneiform ( = tarsalia, 2 and 3), while the ento-cuneiform (tarsale i) and the cuboid (t. 4 and 5) are represented by ligament. The five ordinary toes are present, the number usual in the hind foot of both Anura and Urodela. As in the majority of Amphibians, the skull of the Frog has no basi- or supra- occipital ; no basi- or ali-sphenoid ; and no epi- or opisth-otics. The palatine and pterygoid bones are here membrane bones, at first lying on the surface of a cartilage rod, not as in Teleostean fishes ossifications in the cartilage. The pterygoid however invades the subjacent cartilage. The hyoid apparatus (Fig. 6) consists of a median basi-hyo-branchial plate (Bh.) suspended to the skull by a slender hyoid arch (St. h.). Behind the lower end of this arch a process represents the remains of the two first branchial arches, or cerato-branchials (Br.) ; another process on the posterior edge of the median plate represents the third branchial arch, or cerato- branchial (Br1) ; while ossifications represent the fourth arch, the so-called thyro- hyal (B. hy.). The Eustachian tube (Eu.) passes between the ear-capsule and the outer process of the pterygoid. G 82 DESCRIPTIONS OF PREPARATIONS. There is a remnant of the notochord in the middle of the centrum of each vertebra. A small tubercle, very rudimentary in the Frog, projects in some Am- phibia, most of all in Urodela, from the centrum of the first vertebra, and fits into a pit in the basi-occipital condyle. It has been supposed in consequence that the first vertebra represents an axis, and that an atlas is fused with the skull, or rather has disappeared, leaving a slight trace behind. The urostyle in Bombinator igneus appears from Gotte's researches, to be derived from (i) three vertebrae, the xth, xith, xiith, and (2) a rod of cartilage which lies below the notochord. In the animal when -f- of adult size, the xth vertebra has a pair of nerve foramina behind it ; the xith has a similar pair, not found in the Frog, and the xiith has the neural canal opening behind its arch. The notochord behind the vertebrae atrophies. The part of the shoulder-girdle termed clavicle by Gotte, consists of a cartilage bar with a membrane bone overlying it. Bar and membrane bone constitute the praecoracoid of W. K. Parker. The bone is the clavicle, the cartilage the prae- coracoid of Gegenbaur. The bar is at first, as in the Chelonia, according to Gotte, a process of the scapula which grows ventrally, fuses medianly with the coracoid, and gives rise to a little mass of tissue which fuses with its fellow and forms the episternum as well as the connecting cartilage which unites the two halves of the shoulder-girdle ventrally. This connecting cartilage represents the posterior pro- longation of the Lacertilian interclavicle. The hyposternum, according to Gotte, is not formed, like the sternum of higher Vertebrata, from the ventral ends of ribs, but by a chondrification of the membrane uniting the epicoracoids (median cartilage borders of the coracoids). In Bombinator and Urodela other structures are added, viz. in the former a pair, in the latter sometimes more, e. g. in Menopoma three pairs, of cartilage bands lying in the linea alba and intersections of the recti abdo- minis muscles. Gotte compares them to the false ribs of the Crocodile and Hatteria, but Ruge (M. J. vi. 1880, p. 369) suggests that they are rudiments of the ventral ends of true ribs. The ileum, according to Hoffmann, is an ileo-pubis. In a young Dactylethra capensis he found the symphysial portion to contain a pubic ossification fused in the adult to the ileum. In the middle line there was a projecting rod-like epipubis, a structure generally present in Urodela. In the Frog a pubic ossification appears to be absent, but Hoffmann mentions in Rana and Bufo ' a flat, thin, fairly strong tendon/ with the same attachment as the epipubis. The obturator nerve which perforates the cartilage in Urodela and marks off the pubis, passes over, i. e. outside the pubic region in the Anura. It is not certain whether or no the cuneiform in the carpus represents the ulnare and intermedium as it does in some Urodela. The carpalia 3, 4, 5, do not always fuse in Anura. The astragalus and calcaneum are in some genera separate. The sixth toe is very commonly present. In a young R. temporaria it consists of a tarsale, and of a metatarsal with two phalanges which ultimately fuse, but remain separate in R. esculenta. The sixth and first toe bear nails in Rhinophrynus dorsalis, as they do with the addition of the second and third toes in Xenopus laevis. Skeleton in general, see Hoffmann, Huxley, Ecker and Wiedersheim, p. 78, ante. Skull of Anura. W. K. Parker, Ph. Tr. 166, 1876; 172, 1881. COMMON PERCH. 83 Skull of Frog. Id. Ph. Tr. 161, 1871 ; cf. Parker and Bettany, The Morpho- logy of the Skull, London, 1877. Basiocdpital. Albrecht, Bull. Mus. Hist. Nat. Belg. ii. 1882-83. Vertebral column. Gegenbaur, Untersuchungen zur Vergleich. Anat. derWirbel- saule bei Amphibien und Reptilien, Leipzig, 1862. Urostyle. Gotte, Entwickelungs- geschichte der Unke (Bombinator igneus\ Leipzig, 1875, w^h Atlas, p. 391. Ribs. Gotte, op. cit. p. 381. Sternum and shoulder-girdle. Gotte, A. M. A. xiv. 1877. Pelvis. Hoffmann, Niederland. Archiv. fur Zool. iii. 1876-77. Carpus and Tarsus. Gegenbaur, Untersuchungen zur Vergleich. Anat. der Wirbelthiere i., Leipzig, 1864. Sixth Toe. Born, M. J. i. 1876; vi. 1880, p. 49. Structure of toes, &c. Leydig, M. J. ii. 1876. 1 6. THE COMMON PERCH (Perca fluviatilis), Dissected so as to show its nervous, respiratory, circulatory, digestive, and reproductive systems in situ. THE following external characters are to be noted : — the laterally flattened body and pointed head : the general investment of cycloid scales : the conformation of the mouth : the large eye devoid of eyelids ; and the two dorsally placed apertures, anterior and posterior, of the nose, neither of them communicating with the mouth : the lateral line of sensory organs extending down each side of the body to the tail, and the four depressions on the under surface of each lower jaw indicating the position of the sense organs contained within one of the so-called mucous canals of the head : the opercular apparatus, composed of the operculum and the branchiostegal membrane with its supporting rays which cover the branchial arches laterally and meet on the ventral surface under the chin : and the two sets of fins, azygos and paired. The former consist of the two dorsal, the caudal and the anal fins, imperfectly seen in this specimen, because the caudal region has been removed and the body cut through about the middle of the second dorsal and the anal fins. The paired fins are to be seen on the left side — the pectoral above, i.e. dorsally, and the ventral below. A line drawn through the attachment of the pectoral fin, and at right angles to the long axis of the body, passes just in front of the ventral fin ; the latter is therefore said to be thoracic in position. When such a line passes behind the ventral fin, the latter is said to be jugular in position. In a certain group of Teleostei which preserves the duct to the air-bladder and is hence termed Physostomi, the ventral fins are placed somewhat in front of the anus, but remote from the pectoral, and are then said to be abdominal, e.g. Pike, Salmon. The brain and viscera have been exposed in situ by the removal of the G 2 84 DESCRIPTIONS OF PREPARATIONS. roof of the cranium, the right opercular apparatus and right side of the body walls. There are four divisions of the brain visible. They do not overlap one the other, nor do they fill the cranial cavity as in the young frog. The first division forms the olfactory lobes or rhinencephala from which the olfactory nerves may be seen passing forward : the second division the cerebral hemispheres or prosencephala : the third the optic lobes, corpora bigemina or mesencephalon, the largest of all the divisions in Teleostei'. and the fourth the cerebellum which is subglobular and, unlike the preceding parts of the brain, unpaired. The removal of the operculum on the right side displays the four gill- arches with their double series of gill-filaments, arranged like the teeth of a comb. Each arch is hence said to be bi-pectinate. Internally to the last or fourth gill-arch, and anteriorly to the liver from which in the natural state it is separated by a fibrous septum, lie the heart and ventral aorta. The aorta has a distinct bulb or swelling where it springs from the ventricle : this, the most muscular part of the heart, is in contact with the body walls ventrally while the thin auricle is placed dorsally to it. The liver is large and imperfectly divided into three lobes of which the left is not exposed to view. The single ovary lies posteriorly to the liver and being ripe occupies the greater portion of the abdominal cavity, having displaced the remaining viscera. The anterior three-quarters of its right side have been removed, but owing to the extreme state of distension of the organ, it is not possible to make out the transverse ovigerous lamellae which cross its interior. The commencement of the duodenum with one— the shortest of the three pyloric appendages, may be seen between the liver and ovary. The two other appendages, the stomach, ' and the loop of intestine containing the spleen in its concavity, are all alike hidden on the left side of the body. The gall-bladder has been displaced upwards and to the right. It is lying on the under surface of the liver in a depression, the homologue of the fossa cystis felleae of man, with which it does not in this fish usually come into relation save when the ovary is in a state of turgescence. The terminal portion of the intestine and the rectum pass with -a straight course down the middle line of the body to the anus. This aper- ture, into which a black bristle is inserted, is superficial in Cyclostomi, in Teleostei and Ganoidei, and it is placed in front of the genito-urinary depression, clearly visible here behind it. The air-bladder lies dorsally to the ovary, and between its upper end and the liver posteriorly, and the fourth gill arch anteriorly is a gland (really paired) the homologue of the thymus. Bands of yellow-coloured fat in a state of atrophy correlated with the hypertrophy of the ovary, are to be seen contained in the peritoneal lamellae which unite the intestine to the ventral surface of the ovary as well as along the outer or attached edge of the air-bladder. ; *; .'. COMMON PERCH. 8^ An Acanthocephalous parasite, Echinorhynchus Proteus, is not uncom- mon in the intestines. E. angustatus is also found in the same place as well as the Nematode Cuctdlanus elegans and the remarkable Cestode Triaenophorus nodulosus. A variety of other parasitic worms may be met with, a list of which is given in Zschokke's work cited below (p. 90) or in von Linstow, Compendium der Helminthologie, Hannover, 1878, p. 206, an invaluable work for all students in this difficult branch of Zoology. The epidermis consists of several layers of cells, which are united by proto- plasmic processes. The external cells generally bear a striated cuticle. Glandular cells are present and open on the surface. There are also ' Retort-cells,' or * Kolben- zellen,' with clear contents, which rise to the surface and burst. The scales are dermal ossifications, and doubly refractile. They are covered by the epidermis, and contained within pouches of the dermis. When the free border is evenly, or nearly evenly round as in the Perch, they are said to be cycloid ; when it is pro- duced into more or less prominent teeth, they are said to be ctenoid, as in many other Acanthopteri. The scales of the lateral line are modified. They are per- forated for the passage of a nerve, and channelled on the outer surface near the free edge. The channel towards the base of the scale is converted into a canal for the protection of the sense-organs of the lateral line. These organs are composed of two set% of elements : (i) short pyriform sense- cells terminating -in a sensory hair at their outer free extremity, and a nerve-fibril at their inner; (2) supporting cells which are long, and reach the cutis, and secrete a limitans externa on their outer surface, which is pierced by the sensory hairs. These elements are grouped into eminences or ridges, which are connected in the lateral line in many instances by either modified epidermis, or by non-medullated nerve fibres. The eminences may occur on every scale of the body, e. g. in the Grey Mullet ; or they may be grouped along the lateral line, while isolated eminences occur here and there on the body, e. g. in the Pike, where they are found in numbers towards the tail. In the Perch they are found only in the lateral line of the body, and in its continuation, the mucous canal system of the head. A single canal runs along the post-temporal scale. It divides, and one branch traverses the upper fork of the scale, and crosses the parietal to the opposite side of the body. A second branch runs along the edge of the pterotic bone, and divides into a supra- orbital branch which pierces the frontal bone, and runs above the nostril along the nasal bone, and an infra-orbital branch which runs along the chain of sub- orbital bones, and forward through the homologue of the lacrymal (?), and as a rule unites with the supra-orbital branch in front of the anterior nostril. The third main branch, the infra-maxillary ', runs down the praeoperculum externally, crosses the articular, and perforates the dentary bone. The position of the sense bodies on these branches is not very evident in the Perch. There appears to be none on the connecting branch, two only on the supra-orbital, one just behind the posterior narial opening, and a second between the two apertures ; four may be discovered near the free edge of the lacrymal (?), three or four on the praeoperculum, and four very evident in the dentary region. There are as a rule in the Perch no perforations corresponding to these bodies. They exist however in the Pike, and 86 DESCRIPTIONS OF PREPARATIONS. the course of the canals is therefore more easy to trace. There is much variety in fish in the arrangement, &c., of these bodies. The canals are filled with a soft mucus, secreted by goblet-cells in the lining epithelium. Other sense-organs — terminal or end-buds — occur on the general surface of the body, on the fins, barbules when present, lips, as well as in the mouth, and on the branchial arches. They generally project above the surface ; the sensory and supporting cells are of the same length, and both alike are terminated at their free extremities by short fine points The brain fills the brain-case only in the young fish. The space developed in the adult between the brain and the cranium is filled by a fatty arachnoid tissue. There is some doubt how far the olfactory lobes do really correspond to the structures called by the same name in higher animals. The apparently homo- logous parts in Lepidosteus are parts of the cerebral hemispheres, and the true olfactory lobes are very small. In some Teleostei, e. g. the Tench, the lobes are as in many Elasmobranchii connected by a long peduncle to the brain. The cerebral hemispheres are solid. They touch the optic lobes behind, and the thalami optici are hidden from view. The pineal gland varies much in form, &c., in fish. In the Perch its basal part is conical ; its middle region filamentous ; its terminal part enlarged and fixed to the skull in the frontal region. The optic lobes contain large ventricles. A peculiar fold — the fornix of Gottsche — projects from the region corresponding to the valve of Vieussens in Mammalia, and divides the ventricles almost completely from the iter a 3tio ad ivtum ventriculum. The cere- bellum varies much in size and shape in Teleostei, bu it generally leaves the fourth ventricle more or less uncovered. The sides of this ventricle are often enlarged at the roots of the fifth nerve forming trigeminal lobes, = lobi posteriores, e. g. in the JLoach, Herring, and to a certain extent the Perch. Similar enlargements often occur in relation with the roots of the vagus, forming vagal lobes, e. g. in Cyprinoids. The pituitary body is composed of two parts, an anterior downward prolongation of the infundibulum, and an appended saccus vasculosus. Close to this structure, on either side, are the lobi inferiores or hypoaria, which are remarkably developed in Tekostei, and contain ventricles communicating with the infundibulum. The optic nerves, as in all Teleostei, cross or perforate one the other. The fifth and seventh nerves are closely united at their roots. The glossopharyngeal quits the skull by a special foramen as in Elasmobranchs, Ganoids, Dipnoi, and perenni- branchiate Amphibia. The sense-organs of the mucous canals of the head are supplied by the fifth nerve ; those of the lateral line by a branch of the vagus. The anterior narial aperture is very prominent in the Perch ; it is in some fish prolonged into a tube. The posterior aperture rarely perforates the lip in Teleostei, e. g. in some Muraenoids. The folds of the mucous membrane in the Perch are, as is commonly the case, arranged in a rosette : in some instances they are parallel to one another. In the eye, the cornea is flat, the lens spherical, the retina non-vascular. There is a pecten or falciform process, highly vascular, and pigmented, which projects into the vitreous humour, and terminates in a Campanula Halleri, connected to the equator of the lens. The ear has the three typical semi- circular canals. A sacculus and recessus cochleae are differentiated, and each possess a crista acustica. The aquaeductus vestibuli is closed terminally, not open as in Elasmobranchs on the surface of the head. There are generally two large otoliths, a sagitta in the sacculus, an asteriscus in the recessus cochleae. COMMON PERCH. 87 The teeth of the Perch are fine, long, close-set, and they are anchylosed to the bone that supports them. There is much variety in the dentinal struc- tures of the Teleostei. An external cap or tip of enamel is commonly present, but the main body of the tooth is made up of dentine in one of its three chief modifications, tubular, vaso- or osteo-dentine. Anchylosis takes place by bone developed specially in connection with each tooth. The oesophagus passes with- out clear distinction into the stomach, which in the Perch is of the type known as caecal^ i. e. prolonged backwards beyond the pylorus. It is siphonal, i. e. bent on itself, in the Pike. There is a well-marked pylorus, and the first (duodenal) por- tion of the intestine is dilated. It carries three appendices pyloricae, structures which may be absent, e. g. Pike, or present in large numbers, e. g. Salmon, where they have a linear arrangement. They are rarely united in Teleostei into a mass by connective tissue, e. g. Tunny. In the Perch, according to Krukenberg, they secrete merely mucus ; in some other fish they have a pancreatic function. The loop of intestine containing the spleen is short, and there is no external mark of separation between the intestine and rectum. There is however an internal valve, a remnant perhaps of a spiral valve, which is stated to exist only in Chirocentrus (Clupeidae) among Teleostei. The folds of the mucous membrane vary much in character, and villi are rare. Peptic glands are well marked in the stomach of the Perch ; they are sometimes absent. Ciliated epithelium occurs in the pyloric appendages. The epithelium of the intestine is columnar, and contains goblet-cells ; and it throws out pseudopodial processes. The gall-bladder is never absent; its duct in the Perch opens near the base of the pyloric appendage, which crosses the oesophagus. The true pancreas is present in the Perch as a diffuse gland. The lobules are chiefly found along the veins of the pyloric appendages, and the left branch of the portal vein. The air-bladder is simple in shape, occupies the whole of the dorsal portion of the abdominal cavity, is firmly fixed laterally to the body-walls, and its ventral surface is covered by peritoneum. Its walls are thin, and there are many vaso-ganglia or retia mirabilia developed on its ventral surface anteriorly and internally. There is no trace of the air-duct which connects the bladder with the digestive tract in the Teleostei Physostomi. The inner surface of the operculum, or more strictly speaking, the posterior edge of the hyomandibular bone, carries a filamentous pseudo-branchia or opercular gill. In the adult Pike and many other Teleostei this structure is hidden under the mucous membrane. It is a functional hyoidean gill in the young, but in the adult it receives arterial blood from the hyoidean artery, and transmits it to the choroid gland of the eye. Each of the first four branchial arches carries gill-filaments arranged in two rows, i. e. they are biserial. These filaments are separated to their base as in all Teleostei. The artery and vein run on the convex side of the arch, the vein at a deeper level than the artery. In the heart there is a sinus venosus constituted by the fusion of the right and left ductus Cuvieri. It receives the hepatic veins, and its aperture into the auricle has two thin valve-like folds. The walls of the auricle are thin, and muscles arranged more or less in a network. An anterior and posterior valve guard its entrance into the ventricle. This structure has a thick wall, which in most Teleostei and Ganodei is divisible into two muscular layers, an outer and an inner, separated by a space. This space is lymphatic, and its surfaces are covered by an endothelium. Two valves, a right and a left, guard the passage from the ventricle to the bulbus 88 DESCRIPTIONS OF PREPARATIONS. aortae. The conus arteriosus of the Elasmobranch and Ganoid has been absorbed into the ventricle of the Teleostean. A slight trace of it is found in some Teleostei, but in Bntirinus (Ctupeidae) alone are there two sets of valves. The persistent valves appear to be those of the distal row of the Ganoid and Elasmobranch. The ventral aorta gives off (i) the fourth and third branchial arteries, which have a common origin, and (2) the second and first arteries separately. The veins unite dorsally on each side into an epibranchial artery, and in the same manner as the arteries rise from the ventral aorta. Each epibranchial gives off the common carotid anteriorly, and the two carotids unite by a cross vessel which passes above the parasphenoid. Posteriorly, the two epibranchials fuse to form the subvertebral aorta. By these unions a circulus cephalicus is formed from which the external and internal carotids' arise anteriorly, and into which the hyoidean artery falls dorsally. This artery rises from the ventral end of the first branchial vein, and in its course it sup- plies the pseudobranchia. In the Perch a caeliaco-mesenteric artery, which supplies most of the abdominal viscera, springs from the right epibranchial before it fuses with its fellow. The blood from the abdominal viscera (stomach, pyloric append- ages, spleen, intestine, air-bladder in part) flows into a hepatic portal system. The genital (spermatic or ovarian) vein enters the left ductus Cuvieri. The caudal vein when it enters the abdomen divides into a right and left branch. The former anastomoses with the right cardinal vein, and the latter, which is small, is also con- nected to the right cardinal. There is a renal-portal circulation. The left cardinal vein rises from the anterior part of the left kidney and is not directly connected to a branch of the caudal vein. Both cardinal veins unite anteriorly, each with the corresponding jugular vein, forming the right and left superior cavae or ducts of Cuvier. A vein — the inferior jugular — lies dorsally to the heart. It is best developed on the left side, collects blood chiefly from the inferior part of the head, and falls into the left duct of Cuvier. The thyroid gland is broken up into lobules, and lies ventrally to the ventral aorta, but some of its lobules are found distributed for a short distance along the roots of the branchial veins. They are red in colour. The thymus is paired, and each part lies behind and dorsally to the branchial arches, internally to the supra-clavicle so-called. The kidneys are placed on either side of the vertebral centra. They enlarge anteriorly and coalesce, and are perforated by the cardinal veins. This region of the kidneys in many Teleostei was found by Professor Balfour to consist of a vascular lymphatic tissue, and it does not, as is generally stated, appear to be a persistent head-kidney or pronephros. In the Perch the kidneys thin out posteriorly ; in the Pike they thicken and extend beyond the abdominal cavity into the caudal region. There are two ureters which rise from the ventral surface of the glands. They are fine tubes which pass ventrally downward behind the posterior end of the air-bladder, unite, and at the place of union develope a large simple urinary bladder. This structure, the form of which varies much in fish, does not correspond with the urinary bladder of higher animals, which is derived in development from the anterior (ventral) wall of the intestine, and either represents (Amphibia) or is a remnant (Sauropsida, Mammalia) of the allantois. The testes are paired ; they are rarely single in Teleostei. The inactive organ is semi-transparent ; the active organ varies much in size, shape, and lobulation, COMMON PERCH. according to its state. The vas deferens, as in all Acanthopteri, lies on the inner (median) side of the gland in a well-marked hilus. It becomes free posteriorly, and fuses with its fellow. In the genus Blennius the two remain unfused. The mesorchium in the Perch is single at its origin, a rare abnormality. The tubuli semi- niferi are long, simple, and radially arranged at the hilus. In the Pike they form numerous anastomoses inter se, as in Cyprinoidei, Clupeidae, &c. The ovary, as in some other Teleostei, is single. The ova is produced by lamellae arranged trans- versely. The oviducal canal is central in the Perch, &c., and lined by a flat or columnar epithelium. It is lateral in the Pike, Cyprinoidei, &c., and lined by a ciliated epithelium. When there are two oviducts they unite posteriorly. The genitalia are covered externally by a layer of flat epithelium which, according to Brock, is peritoneal. The genital and urinary ducts open into a urogenital sinus placed behind the anus, the former in front of the latter. In some instances the oviducal aperture is situate on an enlongate papilla, e. g. in Rhodeus amarus, which lays its eggs in the mantle cavity of the Anodon. In this instance the urinary bladder opens also on the papilla. NOTE. — On the ovary of Teleostei. In Salmonidae and the Eels the ova are shed into the abdominal cavity, and pass out thence through pores, usually but perhaps not properly homologised with abdominal pores. In the Smelt (one of the Salmonidae] there are two oviducts widely open at their inner and anterior end, just as in the majority of Ganoidei. Each ovary has developed from its ventral edge a lamella, which curves over and partially protects the outer side of the ovary. In all other Teleostei, as in the Perch, the ovaries and oviducts are continuous, and this is true also of the Ganoid Lepidosteus. In the case of this fish, however, Professor Balfour and Mr. W. N. Parker found that a lamella similar to that of Osmerus was formed from the ventral edge of the ovary, and met another lamella growing from the dorsal abdominal wall. These two lamellae by their union clearly bring about the disposition of parts seen in the adult. Observations on the mode of formation of the posterior part of the oviduct were not made for want of mate- rial. The same want prevented MacLeod carrying his observations on Teleostei to a perfect conclusion, but he has found with reference to the ovary a state of things similar to what obtains in Lepidosteus, It remains to be ascertained whether the ducts, generally called ducts of Miiller in Ganoids, are really so or not ; whether the ducts of Teleostei are ducts of Miiller continuous with the ovarian capsule or not; and whether in both groups alike the oviducts are not simply peritoneal folds and nothing more. Study of Fishes, Glinther, Edinburgh, 1880, or Id. Ichthyology, Encyclopaedia Britannica (ed. ix). xii. Pisces, Von Siebold and Stannius, Handbuch der Zootomie, i. Berlin, 1854. Pisces, Hubrecht und Sagemehl, Bronn's Klass. und Ordn. des Thierreichs, vi. i. (in progress). Perca fiuviatilis. Fishes of Great Britain and Ireland, F. Day, i. p. 2, London, 1880. A History of British Fishes, W. Yarrell, i. p. i, London, 1841 (ed. 2). Perch. Cuvier et Valenciennes, Histoire des Poissons, ii. p. 28, Plates i-viii. Cod. T. J. Parker, Zootomy, London, 1884. Integument. Leydig, Festschrift zur Feier des loo-jahrigen Bestehens der Naturf. Gessellsch. in Halle, 1879. Organs of Lateral Line. Merkel, Endigungen 9o DESCRIPTIONS OF PREPARATIONS. der sensibeln Nerven, Rostock, 1880; Bodenstein, Z. W. Z. xxxviii. 1882; Solger, Z. A. v. 1882; Leydig, op. cit. supra. Scales. Baudelot, A. Z. Expt. ii. 1873; Carlet, A. Sc. N. (6) viii. 1879 ; O. Hertwig, M. J. ii. 1876; v. 1879; vii. 1881. Central nervous system. Mayser, Z. W. Z. xxxvi. 1882; Ussow, Archives de Biologic, iii. 1882 ; Rabl-Ruckard, Arch, fur Anat. und Physiol., Anat. Abth. 1882, 1883 ; Id. Biol. Centralblatt, iv. 1884-85. Pineal gland. Cattie, Archives de Biologic, iii. 1882 ; Id. Z. W. Z. xxxix. 1884. Organs of special sense. Nasal. Blaue, Z. A. v. 1882. Eye. Berger, M. J. viii. 1882 ; Hoffmann, A. M. A. xxiii. 1884. Ear. Retzius, Gehororgan der Wir- belthiere, i., Stockholm, 1881 ; for abstracts, see Biol. Centralbl. ii. 1882-83. Teeth. Tomes, Ph. Tr. 166, 1876 ; Id. Dental Anatomy (ed. 2), London, 1882; Sternfeld, A. M. A. xx. 1882 ; in Scaroids, Boas, Z. W. Z. xxxii. 1879 ; cf. Owen, Odontography, Text and Atlas, London, 1840-45. Digestive tract: minute anatomy. Edinger, A. M. A. xiii. 1877 ; physiology, cf. Krukenberg, Vergleich. Physiol. Vortrage, ii. Heidelberg, 1882; in Cobitis fos- silis, Lorent, A. M. A. xv. 1878. Pancreas, Legouis, A. Sc. N. (5) xvii, and xviii. 1873. Pharyngeal pouches, rumination, of Scaridae. Sagemehl, M. J. x. 1885. Air-bladder : minute anatomy. Schulze, Strieker's Histology (Sydenham Soc.), ii. 1872, p. 78. Function. Moreau, A. Sc. N. (6), iv. 1876. Branchiae: minute anatomy. Riess, A. N. 47, 1881. Pseudo-branchia. Maurer, M. J. ix. 1883. Heart. Kasem-Beck und Dogiel, Z. W. Z. xxxvii. 1872. Conus. Boas, M. J. vi. 1880. Spleen (in Ichthyopsida). Phisalix, A. Z. Expt. (2) iii. 1885. Lymphatics. Sappey, Systeme lymphatique des Poissons, Paris, 1880. Thyroid and Thymus (with development). Maurer, M. J. xi. 1885. Kidneys. Hyrtl, Dk. Wien. Akad. i. 1850; ii. 1851. Balfour, Q. J. M. xxii. 1882 ; Emery, Archives Italiennes Biol. ii. 1882, cf. Fiirbringer, M. J. iv. 1878. Head-kidney. Grosglik, Z. A. viii. 1885; ix. 1886; Emery, Ibid. Renal-portal circulation. Jourdain, A. Sc. N. (4) xii. 1859. Sex organs. Brock, M. J. iv. 1878. Oviducts^ of Osmerus. Huxley, P. Z. S. 1883. Development of Duct. MacLeod, Archives de Biologic, ii. 1881 ; in Lepi- dosteus, Balfour, and W. N. Parker, Ph. Tr. 173, 1882. Fishes ova. Mclntosh, Nature, xxxi. 1884-85. Relations of Yolk to Gastrula in Teleostei. J. T. Cunningham, Q. J. M. xxvi. 1885-86. Parasites of fresh-water Fish. Zschokke, Archives de Biologic, v. 1884. 1 7. SKELETON OF COMMON PERCH (Perca fluviatilis), -With Figures 8 and 9. THE Perch belongs to the order Acanthopteri among Teleostei. It may be taken as an illustration of a highly specialised as opposed to a generalised type of organisation : combining as it does so many of the peculiarities of its own class, rather than showing affinities, as e.g. do the Dipnoi, to other and higher forms of life. Its skeleton is typically Teleostean. SKELETON OF COMMON PERCH. This skeleton is well ossified. The parts of the skull may be most readily identified with the help of Figs. 8 and 9. The occipital region (Fig. 8) is composed of the four typical bones, the supra-occipital (S.O.), which, as usual, is here of very large size; the two ex-occipitals (&O.\ which surround the foramen magnum, each pierced by the glossopharyngeal and vagus nerves (X.} and offering an oblique facet (*) for articulation with the first vertebra ; and a basi-occipital (B.O.\ which has a cup-shaped posterior surface and articulates with the same vertebra. The basi-sphenoid (B.S.) is small and has the Y-shape typical of Teleostei: the alisphenoid (A.S.) is placed in front and above it. Between the occipital and basi-sphenoidal re- gions intervenes the auditory capsule, which in 7*eleostei\s very large and ossi- fied from five centres. The bones corresponding to these centres are the epi- otic (Ep. O.), the opisth-otic (Op. O.~), to which are attached the upper and lower extremities of a forked bone, the post-temporal scale (not shown in the EJZt STa. S.O. PMx MX. \~^ Pr.O. FIG. 8. — From a specimen (natural size). figure) by which the fore-limb is connected to the skull : the pro-otic (Pr. (9.), a large bone pierced by the fifth and seventh (VII) nerves, and two bones which attain their relative maximum development in the Teleostei, the pter-otic (Pt. 0.), and the sphen-otic (Sp. #.),.( = post- frontal of Huxley). Between these two bones above, and the pro- and opisth-otic below, are the articular cavities (t) for the two heads of the hyomandibular (H.M. in Fig. 9). As is often the case, the parietal (P.) is a very small bone, and the frontal (F.) a very large one. The nasal (Na.) is slender. In front of the ali- and basi-sphenoid a cartilage plate, the interorbital septum (/.#•), divides the two orbits. It represents the prae- and orbito-sphenoid regions. Anteriorly to the septum is a large bone, the ecto-ethmoid (E. Et.) ( = the pre-frontal of Huxley), the homologue of the lateral masses of the ethmoid in a mammal, and pierced by the single olfactory foramen (/.). The two ecto-ethmoids are separated by a small median mesethmoid. The praemaxilla (P. MX.) 92 DESCRIPTIONS OF PREPARATIONS. is very large and dentigerous, and of a characteristic shape. The maxilla (Mx), which is edentulous, lies behind and parallel to it, and does not form the margin of the gape of the mouth. This disposition of the two bones is nearly universal among Teleostei. The base of the cranium is underlaid by two bones — a single vomer ( Vo), dentigerous in front, and a parasphenoid (P. S.) which extends back to the basi-occipital. These two bones, as well- as the praemaxilla, maxilla, nasal, frontal and parietal are preformed in membrane, the remainder belong to the cartilaginous cranium, which per- sists as cartilage beneath the parietal and frontal regions. The jaw apparatus (Fig. 9) is very remarkable. It is connected to the cranium in front by the palatine (Pa) and behind by the hyomandibular FIG. 9. — From a specimen (natural size). (Hm.\ The palatine (Pa.) is dentigerous. There is a slender pterygoid (Pt.) ( = ectopterygoid of Huxley), which bears a few teeth anteriorly and de- scends along the anterior edge of the quadrate, while a thin plate-like meso- pterygoid (M. Pt.) (= entopterygoid of Huxley) extends horizontally inwards. These bones are formed in a bar of cartilage lying in front of the mouth in the embryo. Another bar of cartilage — Meckel's arch — which lies behind the mouth segments transversely. The upper, t>r proximal portion, forms the metapterygoid (Mt. Pt.) and the quadrate (Qu). The latter has a rounded articular head for the articular element of the lower jaw, and both bones are firmly united to the bones in front and behind them. The lower or distal segment constitutes the lower jaw in part. Its upper end SKELETON OF COMMON PERCH. 93 ossifies as the articular (Ar.\ from which an angular (An.} is cut off, while the median portion remains cartilaginous and rod-like. It is surrounded by the dentary (D.\ a bone formed in membrane, and carrying teeth. The hyoid cartilage of the embryo gives origin to a pharyngo-hyal element which ossifies as the hyomandibular (ffm.) and symplectic (Sy.) bones. They are connected with the bones of the jaw as well as with some others to be described presently. The hyomandibular has two condyles for articulation with the otic region (f in Fig. 8). It is a large bone and represents, accord- ing to Kitchen Parker's most recent researches, the columella auris or stapes of higher forms. The quadrate represents the element of the same name in Amphibians, Reptiles, and Birds, and the incus of Mammals, while the malleus of the last-named group corresponds to the articular of the fish, of Amphibia and Sauropsida. A series of membrane bones well developed in Teleostei and Ganoidei are attached to the posterior edge of the hyoman- dibular and quadrate bones. These are the prae-operculum (P. Op.\ the operculum (Op.}, the sub-operculum (S. Op.\ and the inter-operculum (/. Op.}. They close in laterally the branchial cavity. The remaining portion of the embryonic hyoidan cartilage gives origin to the interhyal or stylo-hyal (S. h} which articulates between the hyomandibular and symplectic, the epihyal (Ep. k}, the cerato-hyal (C. h.}, both large flat bones, and two nodular hypo-hyals (H. h.}. The arches of the right and left side are united by a median basi-hyals (B. k.} or entoglossal. A series of membrane bones, the branchio-stegal rays (Bs. r.}, are attached to the epi-cerato-hyal. They are formed in the membranous flap, which is a continuation of the opercular apparatus to the hypo-hyal region. A thin median bone, the basi-bran- chiostegal ( = urohyal of Huxley), not shown in the figure, projects back- wards from the basi-hyal region towards the ventral ends of the clavicles, so-called, with which it is connected by ligament. It underlies the ventral aorta. The branchial arches are five in number on each side. In the median ventral line, and immediately following the basi-hyal, are three basi-branchial bones. The first arch consists, proceeding from its dorsal to its ventral end, of a small pharyngo-branchial, or superior pharyngeal bone, an epi-branchial, to which is articulated, at a sharp angle, a cerato-branchial followed by a hypo-branchial. The three next arches have a similar composition. The pharyngo-branchials are, however, both large and dentigerous. The hypo- branchial element in the third arch is applied laterally to the last basi-bran- chial, and is wanting altogether in the fourth arch. The fifth arch consists of a single bone, probably homologous with a cerato-branchial. Osseous plates bearing fine teeth are implanted on the anterior and posterior aspects of the four first arches. These plates are carried by cartilaginous rods in the case of the first cerato-branchial. The rods, known as gill-rakers, are present in some fish on the following cerato-branchials. The interlocking of one 94 DESCRIPTIONS OF PREPARATIONS. series with another, closes the gill-clefts and prevents the passage of foreign bodies of any size. The vertebral column consists of forty-two vertebrae, of which twenty- two are dorsal and twenty caudal. There are no other regions differentiated in the backbone of a fish. Successive vertebrae articulate one with another by the edges of the centra, which are united by ligaments. In the majority of vertebrae a process springs from the base of the neural arch on each side and articulates with the arch of the vertebra preceding. The neural arches are continuous with the centra, and are prolonged dorsally into neural spines. Lateral projections, or inferior, i.e. haemal arches, as it appears from development, project from the centrum of the eighth and following vertebrae. They increase in length and are turned more and more ventrally in the posterior dorsal vertebrae. The processes of the two last dorsal ver- tebrae have the ribs united to them, as is the case in the caudal series, where the ventral ends of the ribs unite and inclose the caudal canal, which lodges the caudal artery and vein. This is the case as far as the last or terminal caudal vertebra, which bears solid inferior arches, to either side of which pass the two vessels into which the caudal artery and vein divide. The two last caudal vertebrae are modified. The last but one has a short expanded neural, and a long inferior arch. The last has the centrum pro- longed into the urostyle — the ossified and undivided sheath of the notochord —which is bent upwards dorsally at a sharp angle and incloses the termina- tion of the notochordal cartilage. It has a long, low neural arch and six inferior or haemal arches, expanded and flattened, divided by an interval into two groups of three arches apiece, one anterior, the other posterior. To the neural arches of these two terminal vertebrae are attached three long and somewhat broad ' false ' spines, as they are called. These, with the haemal arches, carry the rays of the caudal fin. Simple curved and free ribs are carried on the lateral processes of the dorsal vertebrae except the two last. These ribs bend ventrally but do not meet, and in the Perch, as in all other fish, there is no sternum. Certain of the anterior ribs have epipleural bones attached to them at some little distance from their vertebral ends. These bones radiate outwards into the myocommata or connective tissue septa uniting successive muscular segments or myomeres. The Perch has four fins belonging to the median or azygos system. These fins, as in all fish, are supported by fin-rays, ossified in the Teleostei. The four fins are the first and second dorsal, the caudal, and an anal, which is in position ventral. The dorsal and anal fins are supported by a series of bones, ' fin-bearers ' or ' interspinal ' bones, which in the case of the dorsal fins alternate with the neural arches ; in the case of the anal fin with the inferior arches. These bones lie between the muscle-masses of the right and left halves of the body respectively. The folds of integument which constitute the fins proper, are supported by bony 'fin-rays.' There are SKELETON OF COMMON PERCH. 95 thirteen (or sometimes fifteen) of these in the first dorsal, and as in all Acanthopteri with two dorsal fins, they are stout stiff spines, hence called entire fin-rays. The first fin-ray of the second dorsal, and the two first of the anal fin are likewise entire, while the remaining rays as well as all the rays of the caudal fin are ' soft ' and ' jointed/ The soft rays break up into filaments at their free termination, and these filaments as well as the stem to a certain extent, are jointed. The caudal fin in the Perch appears externally to be equally developed in its dorsal and ventral half. It may be termed ' homocercal.' But, anatomically speaking, the eight or nine long fin-rays which make up the bulk of the fin are all articulated to inferior arches, and the upward-bent termination of the notochord has therefore in reality the chief part of the fin on its ventral edge. The dozen or so small ' accessory ' rays which belong to the dorsal aspect are inconsiderable in size ; and, balanced by similar rays on the ventral aspect, complete the wedge-shaped outline of the fin. The fin is, therefore, from an anatomical point of view, as ' heterocercal' as is the caudal fin of an Elasmobranch in outward appearance as well as anatomically. The paired fins are the pectoral and pelvic or ventral, i.e. the fore and hind-limbs respectively. A forked bone, the supra-temporal scale, connects the fore-limb to the skull. It is followed by two thin, plate-like bones, developed in membrane and properly belonging to the skin, the so-called supra-clavicle and clavicle. These two bones lie immediately below the mucous membrane of the hind wall of the branchial cavity. They are probably not represented in Vertebrata above the class Pisces. The two 'clavicles' meet in the median line where they are united by ligament. They bear on their inner and posterior face the true^shoulder-girdle. This consists of two flat bones, one, the scapula, more dorsal, small, and perforated by, a foramen, the other, the coracoid, large and extending nearly to the median line. There is a thin bone, the 'post-clavicle,' which is attached proximally to the clavicle, and hides the scapula when viewed from within. It is superficial and underlies the surface of the depression internal to the pectoral limb. It has been removed on the right side. To the edge of both scapula and coracoid articulates the fore-limb. The small basal cartilages which immediately articulate with the shoulder-girdle cannot be made out in this specimen. The fin contains about ten soft rays. The hind-limb lies immediately behind the median union of the pec- toral arch. The basal part of the limb consists of a long triangular bone produced by the contact of a right and left element, which appear to be homologous with the two metapterygia (right and left), i. e. the posterior basalia of the hind-limb in an Elasmobranch. The true pelvis is therefore absent. Teleostei and Ganoidei, with the exception of Polypterus, which has small pelvic cartilages, agree in this respect, and differ from Elasmobranchii, 96 DESCRIPTIONS OF PREPARATIONS. ' Holocephali, and Dipnoi, which possess a pelvis. The fin-rays are articulated to the posterior and external faces of this double bone. The first ray in each fin is entire, the remaining five are soft. The following additional points may be noted relative to the skull. A series of suborbital bones lies beneath the eye of which the most anterior is exceedingly large. These bones are removed in this specimen. In some Teleostei, e.g. the Trout, there are similar bones above the eye. The optic nerve passes out of the cranium between the arms of the basisphenoid. The bone termed alisphenoid is continued forwards for a short distance by membrane which is pierced by the olfactory nerve. The two nerves separated by the interorbital septum traverse the back of the orbit and pierce the ectoethmoids on the way to the olfactory mucous membrane. An ' ocular' canal which lodges the recti muscles of the eyes passes backwards into the forepart of the basi-occipital, beneath the pro-one and above the parasphenoid. The inner wall of the ear-capsule is replaced by thin membrane which is easily injured in dissection, and the vertical semi-circular canals then appear to lie in the cranial cavity. The skulls of Teleostei in general agree closely with that of the Perch. The size of the cartilaginous cranium varies, as may be seen on comparison of a Salmon with the Perch. The amount of persistent cartilage as compared with bone also varies. Anchylosis of the bones may take place. In Cyprinoid and Siluroid fish there is no inter-orbital septal plate, but there are orbito- and pre-sphenoid ossifica- tions in this region. In the Pike the mesethmoidal cartilage is partially ossified by two pairs of bones instead of a median one, and in the Salmon this region is not ossified but is covered by a supra-ethmoid bony plate. The Pike has a small supra-orbital bone. Praemaxillae are absent in Muraenoids. In some Teleostei, e.g. Pike, a small bone placed distally on the posterior margin of the maxilla appears to represent the jugal. The metapterygoid is absent in the Siluroid Clarias capensis. And in Siluroids generally the sub-opercular is absent. In Cyprinoids the fifth branchial arch is strongly bowed, and carries prominent teeth which work against the horny basi-occipital tooth. The group of Pharyngognathi is so named from the fact that the fifth pair of arches is fused into a single dentigerous plate. When the lower jaw is connected to the cranium solely by a hyomandibular element derived from the hyoid arch as it is in Teleostei, in Ganoidei except Lepi- dosteus, in the majority of Elasmobranchii, it is said to be hyostylic. When it is connected not only by a hyoidean element but also by a quadrate, as in Lepidosteus, or by a palatoquadrate, as in Cestradon among Elasmobranchii and probably in Holocephdli, it is said to be amphistylic ; and when it is connected by a quadrate element alone as in Amphibia and Sauropsida, it is termed autostylic. In Mam- malia the dentary element articulates with the squamosal, the articular portion of the lower jaw and the quadrate having been converted into ear-bones, i. e. malleus and incus respectively. The azygos system of fins appears in Teleostei as in other fish as a continuous fold of skin supported by embryonic fin-rays which are afterwards replaced by per- manent fin-rays. Such a continuous fold extending from the back of the head round the tip of the tail to the anus persists in Blennies, Eels, Congers, Soles, Ophididae, &c., among Teleostei, and in Dipnoi. But this condition in the SKELETON OF COMMON PERCH. 97 opinion of the late Professor Balfour has been secondarily acquired. The caudal fin is first differentiated from the continuous fold : then the posterior dorsal when present, unless the anterior dorsal is of a peculiar type, as in Lophius, when it appears before the posterior. The anal fin appears before the pelvic fins, unless the latter are of a peculiar type and adapted to special uses, as in the young of some Gadoidei. The young Elasmobranch, Lepidosteus, and Teleostean, have at first a long pointed tail, to the tip of which the notochord extends and the lateral line as well whenever this primitive condition is retained. A caudal fin is next developed on the ventral aspect and at some little distance from the extremity of this pointed prolongation as an enlargement of the continuous fold. The prolongation bends upwards towards the dorsal aspect at the same time. The growing caudal fin, as remarked by Professor A. Agassiz, has much the appearance of a second or pos- terior anal. It is supported solely by enlarged haemal arches, beyond which appear at a later period the fin-rays. Such a condition persists and forms the heterocercal caudal fin of all Elasmobranchii, living chondrostean Ganoids, and many extinct Ganoids. But in existing bony Ganoids and the Teleostei the pointed prolongation atrophies until the caudal fin becomes terminal. The upward dorsal inclination is preserved in the urostyle. The part of the caudal fin formed on the neural side of the urostyle is always inconsiderable. This outwardly symmetrical caudal fin, really asymmetrical and, anatomically speaking, heterocercal, is termed homocercal. But in a few fish, such as the Dipnoi and the Teleostei above- mentioned, the backbone retains its straight course and divides the caudal fin into two equal portions, dorsal and ventral. Such fins are known as diphycercal. In the Eel and some other eel-like Teleosteans, rudimentary haemal arches exist and point to the existence at some distant period of a caudal lobe now aborted : and in them the diphycercal tail is secondarily acquired. In Holocephali the long whip-like tail has the groove of the lateral line continued to its apex, and a small ventral lobe represents the large caudal lobe of Elasmobranchii. The late Pro- fessor Balfour believed that he had found traces of caudal haemal arches in Cera- todus, which would indicate a lost caudal lobe in that Dipnoan. It is perhaps doubtful whether a primary diphycercal tail exists among living Pisces. The number of haemal arches present in the caudal lobe varies much, e. g. the Cod and Stickleback have only two. The ' accessory ' rays are in some instances of large size, e.g. in the Cod, and the caudal fin has then a peculiar rounded appearance. The series of bones known as supra-clavicle, clavicle, inter-clavicle, and post- clavicle in Teleostei and Ganoidei appear orue and all to be derived from the integu- ment and mucous membrane of the branchial cavity. In the Teleostei they are dermal in position; in Adpenser on the contrary their outer surface resembles in appearance the bony plates of the integument, and the bone has also the same structure. Re- cent researches appear to have established the fact that the clavicle of higher forms is a process of the cartilaginous shoulder-girdle (see Gotte and Hoffmann's papers referred to under Pigeon, p. 67). Such a process exists in the Sturgeon, and is named prae-coracoid in Professor W. K. Parker's Monograph on the shoulder- girdle. 'Swirski's researches on the shoulder-girdle in the Pike appear to establish the fact that the true coracoid aborts almost completely in that Teleostean, and H 98 DESCRIPTIONS OF PREPARATIONS. therefore probably in others as well, while the clavicular process ( = praecoracoid process of his paper) increases in size, and forms what is generally termed coracoid. The dorsal angle of the scapula of the Pike bends inwards and downwards in development. In Cyprinoids, in the Salmon, &c., a curved bar of bone stretches from the dorsal angle of the scapula to a spot ventral to the scapular foramen. This bar in the Cyprinoid is derived, according to 'Swirski, from two processes of the scapula, a dorsal and a ventral, which meet and fuse. They appear in a late stage of development. In the bony Ganoidei the proximal end of the metapterygium of the ventral fin bears a cartilaginous process. Davidoff regards this as pelvic, but Wiedersheim considers it to be part of the metapterygium. The two metapterygia are always in contact at their proximal extremities, but often widely separate distally, e. g. bony Ganoidei, the Pike, &c. ; or a bar of bone may develop at the distal end of each metapterygium, and meet its fellow in the middle line, while a space separates the middle portion of the two bones, e. g. Salmon ; or the two may be in contact through- out the whole extent of their median edges, e. g. in the Perch ; or finally, complete fusion may take place. The basal cartilages or bones which articulate with the coraco-scapular are five in number. The first is included in the first fin-ray, and represents a propterygium; the fifth is long and is a metapterygium. In some Teleostei the number is reduced, or the bones may be entirely lost. A row of small radial cartilages is to be found in some instances inclosed in the bases of the fin-rays. They do not seem to be present in the Perch, at least not in the adult. The fin-rays of the pelvic fins articulate directly with the metapterygium. A row of excessively small bony or cartilaginous particles representing radials is sometimes inclosed in their bases. They do not exist however in the full-grown Perch. For the caudal canal and interspinous bones, see next Preparation, pp. 100, 101. Skull of Salmon. W. K. Parker, Ph. Tr. 163, 1873; cf. Id. and Bettany, Morpho- logy of the Skull, London, 1877. Cod. T. J. Parker, Zootomy, London, 1884. Pike (and other fish). Huxley, Lectures on Elements of Comparative Anatomy, 1864. Investing bones of Pike's head. Walther, J. Z. xvi. 1882. Loricaria, &c. Goldi, ibid, xvii. 1883. Char acinidae (and auditory bones]. Sagemehl, M. J. x. 1885. Azygos fins. Kolliker, Uber das Ende der Wirbelsaule der Ganoiden, &c., Leipzig, 1860. Caudal. Lotz, Z. W. Z. xiv. 1864 (Perch, p. 89); Development of; Agassiz, Proceedings Amer. Acad. (new series) v. 1878; vi. 1879: ix. 1882; cf. Balfour and W. N. Parker on Lepidosteus, Ph. Tr. 173, 1882, and Grassi, M. J. viii. 1882. Fatty fin of Salmonidae. De la Valette St. George, A. M. A. xvii. 1880. Ichthyopterygium and Cheiropterygium. Baur. Z. A. viii. 1885; cf. Balfour in Comparative Embryology, ii. pp. 500—511 with references. Shoulder-girdle. 'Swirski, Schultergiirtel des Hechtes (Inaug. diss.), Dorpat, 1880; W. K. Parker, Ray Society, 1868; Gegenbaur, Untersuchungen zur Ver- gleich. Anat. der Wirbelthiere, ii., Leipzig, 1865. Pelvis. Wiedersheim, M. J. vii, 1881 ; von Davidoff, M. J. v. 1879 ; vi. 1880. VERTEBRAE OF COMMON COD. 99 1 8. VERTEBRAE OF COMMON COD (Gadus morrkua). THE anterior and posterior surfaces of the centra of these vertebrae, as in all fishes with distinct vertebrae, except Lepidosteus, are concave, the anterior surface being less so than the posterior. They are therefore amphicoelous. The two concavities communicate by a fine central canal. Their surfaces are marked by two sets of lines, radial and concentric, in- dicating the directions in which ossification has taken place. More or less wedge-shaped cavities are observable around the periphery of the centrum. There are two of especial depth on the ventral surface immediately internal to the bases of the haemal arches in the more anterior vertebrae of the series, while two shallower cavities occupy the median line. In the posterior vertebrae the lateral cavities become more and more shallow, and finally disappear, while the median cavities coalesce and deepen. These cavities in the living animal are filled by connective or fatty tissue. The neural arches are continuous with the centra and coalesce to form a median spine. At the base of each arch there are two processes on either side, an anterior broad process, with a slight groove to its outer side, and a posterior process, pointed and adapted to fit more or less closely into the groove just mentioned. The anterior process becomes more and more pronounced in the anterior dorsal vertebrae, and in the first vertebra is of huge size and articulates with the skull. The vertebrae carry well-developed inferior or haemal arches projecting outwards and downwards, and continuous with the centrum. In the median vertebrae they are broad and flat, and serve to protect the air- bladder, which is largely developed in the Gadidae, and devoid of an air- duct. In these same vertebrae a smaller process is developed behind, but parallel with the haemal arches. It is probably of no significance beyond assisting in the protection of the air-bladder. The haemal arches fail to be developed only in the four first vertebrae. They carry the ribs. In the region of the tail they are united with the ribs and form a caudal canal, as in the Perch. The bones of fish are the poorest in inorganic constituents of all the five classes of Yertebrata, and the bone-cells, the characteristic microscopic structures in true bone, may in the fully formed bone be absent altogether or very rare, e. g. in the Perch or Pike. According to Kolliker, they are to be found in all Ganoidei and in the majority of Teleostei which possess an air-duct to the air-bladder. The formation of the vertebral centra in Teleostei appears to take place as follows : — The embryonic chorda dorsalis, when fully formed, is composed of central vesicular cells with a thin superficial epitheliomorphic layer produced by the peripheral aggregation of the protoplasm of certain of the cells, each apparent cell containing a nucleus. A cuticula comes next : and most externally H 2, 100 DESCRIPTIONS OF PREPARATIONS. a layer which, according to Grassi, is an elastica externa, but, according to Gotte, is composed of flat nucleated cells in a single layer, with an abundant intercellular substance. In this substance Gotte states that the first ossifications forming the primary centrum start. The chorda grows intervertebrally, and shrinks to a fibrous cord in the middle of the vertebra. Hence the amphicoelous centra. The cuticula thickens and becomes fibrous intervertebrally, as does the external layer, forming the ligaments which unite the centra peripherally. The neural and haemal arches rest with their bases on the chordal sheaths, but do not grow round the chorda as they do in the Elasmobranch. The main, i. e. secondary osseous substance, is periosteal, formed from connective tissue, which developes between the bases of the arches and imbeds them. There are four principal types of structure in the vertebrae of adult Teleos- teans. (i) Concentric laminae are laid down round the primary centrum. The bases of the arches ossify and fuse indistinguishable with the laminae ; but isolated masses of cartilage, remnants of them, may persist here and there temporarily, e. g. the Cyprinoids. (2) Delicate bony radial laminae with intervening clear connective tissue, appear first of all ; finally the bony matter becomes spongy, and the connective tissue is converted into a fatty medulla. The bases of the arches persist as cartilage, and form the well-known cross, e.g. Pike. The majority of Tekostei belong to this type. (3) The bony matter forms a delicate spongy mass ; the intervening substance consists of small round cells with a clear, firm matrix ; the bases of the arches may retain isolated masses of cartilage ( Cyclopterus lumpus} or may ossify completely ( Chironectes sp. ?). (4) There are delicate radial united by a few concentric, bony laminae. The intervening substance is hyaline cartilage : Monacanthus penicilligerus, Diodon. The differences depend upon (i) the persistence of cartilage in the bases of the arches or the total conversion of the latter into bone, and (2) either the total or partial conversion of the intervening connective tissue into bone; and in the latter case the persistence of the remnant as connective tissue or its conversion into either medulla or cartilage. The neural or superior appear before the haemal or inferior arches, and both structures appear earlier in the anterior than in the posterior region of the column. In Cyprinoid fishes certain of the arches may ossify without previous chon- drification. In the anterior region of the column the haemal arches may become entirely imbedded within the centrum, e. g. in the Perch ; and in the posterior region they become not only more and more ventral in position, but they bend more and more towards the ventral median line. Finally they appear to unite, inclosing the caudal canal. The most posterior, however, attached to the terminal vertebrae fuse into solid knobs. It may be observed that in the dorsal region ribs are articulated to their ends. These ribs are developed continuously with the arches in Elasmobranchii^ Ganodei so far as is known, and in some Teleostei, whereas in others they are developed independently, but yet in close apposition to their ends. In Amia and Lepidosteus the caudal canal is formed by the haemal arches plus ribs which are cut off from them, but which articulate with them, as in the dorsal region. And it is these ribs that unite distally, completing the canal. The presumption is, that in other cases the canal is formed by haemal arches with ribs remaining continuous with them. This view was strongly supported VERTEBRAE OF COMMON COD. TOi by the late Professor Balfour in his paper on Lepidosteus, published in connection with W. N. Parker (infra). In some Elasmobranchii^ e.g. Scy Ilium and some Teleostei, the anterior portion of the caudal canal is apparently inclosed by haemal arches, to which ribs are attached laterally. The explanation of this arrangement is as follows (Balfour) : — The canal in this region is secondarily acquired : it is a prolongation forward of the posterior part which is normally developed, and it is formed by the growth of two processes — one from each haemal arch near its base, towards one another which meet and fuse, the true haemal arch being that portion of cartilage or bone which bears the rib. In some instances, e.g. Pike, the right and left neural arches are not united by bone in the median line. It should be noted also that in the region of the trunk at least, the ribs have free ventral ends. There is no sternum. Other structures to be noted are : (i) the ligamentum vertebrale superius which runs above the spinal cord and connects the successive vertebrae. (2) A pair of cartilages which appear in development between the ligament and the spinal cord : they project posteriorly, ossify or become surrounded by bone and become continuous ultimately with the arch. They are probably homologues of the intercalaria neuralia of the Elasmobranch which lie between the neural arches. They are present also in Ganoids. (3) The interspirial bones which support the azygos fins seen in the preceding preparation (Prep. 17). These, according to Gotte, may be regarded as dissociated parts of vertebral spines, but this view is extremely doubtful. Dohrn has recently suggested, and Mayer has strongly supported his view, that the azygos fins are derived from paired rudiments ; the anal and ventral element of the caudal fins, by the coalescence behind the primitive cloaca of the two ventro-lateral ridges from which the pectoral and pelvic fins originate : the dorsal fins and the dorsal element of the caudal fin from two similarly coalesced dorso-lateral ridges. If this is so, then the interspinal bones are really equivalents of the basal cartilages in the paired fins. They are not connected with the vertebral column in most Elasmobranchii. In Dipnoi they remain articulated to the neural spines in the dorsal region, and to the haemal in the ventral region, where in other types the anal fin is formed. They appear in the young Teleostean just above the summits of the arches, but when the latter lengthen, come to lie between them. They are developed also in regions where the azygos fins are deficient; but it must be borne in mind that this system of fins is primitively represented by a continuous fold. (4) The set of bones known as epi-neurals, epi-centrals, epi- pleurals (Fleischgraten). They are formed in the intermuscular septa (myocommata), and never pass through a cartilage stage. Gotte has suggested that the series which lies on the ends of the inferior arches in the region of the trunk are ribs homologous with the ribs of Elasmobranchs owing to their position. But this view appears to have little to support it, and the mode of formation of the bones in question is against it. It may be noted that the ribs of all fish except Elasmobranchii lie at the base of the myocommata, immediately below the peritoneum. In Elasmobranchii they extend outwards in the fibrous septum which separates the dorso-lateral and ventro-lateral muscle-masses, and therefore not below the peritoneum. Hence these ribs have been regarded as not homologous with the ribs of other fish. But in Lepidosteus the free ends of the ribs, the bodies of which are normally 102 DESCRIPTIONS OF PREPARATIONS. placed, bend outwards into this same septum. It is possible, therefore, that the ribs of Elasmobranchii have undergone a change of position. Schmid-Monnard has recently investigated the origin of bony tissue in Teleostei. He finds as follows, (i) The first bone is always formed outside the cartilage. (2) All skeletogenous tissues may take part subsequently in the formation of bone, and whether cartilage or connective tissue they ossify either directly or indirectly, i.e. by means of osteoblasts. In most parts of the skeleton bone is derived exclusively from perichondrium. (3) The first:formed bone is invariably homogeneous, containing neither bone-cells nor connective fibres. Bone-corpuscles are to be found here and there in those fish in which, according to Kolliker, they are absent, and the bone resembles dentine, e. g. Pike, Perch, Lota vulgaris, Gadus aeglefinus, &c. But true dentine, i. e. tubular dentine, appears never to be formed. The great development and regular arrangement of Sharpey's fibres probably led to the supposition, coupled with the then imperfect methods of research at command. Formation of vertebrae, arches and ribs in Teleostei, Gotte, A. M. A. xvii. 1879 ; Grassi, M. J. viii. 1882 ; in Ganoidei and Elasmobranchii, Gotte, A. M. A. xv. 1878; in Elasmobranchii, Balfour, Elasmobranch Fishes, London, 1878; in Lepidosteus, Id., and W. N. Parker, Ph. Tr. 173, 1882. For discussion as to homologies of ribs and formation of caudal canal, see work last quoted. Azygosfins. Mayer, Mitth. Zool. Stat. Naples, vi. 1885. Formation of bone. Schmid-Monnard, Z. W. Z. xxxix., 1883, and Kolliker, P. R. S. ix. 1859. Cf. Kostler, Z. W. Z. xxxvii. 1882. 19. ASCIDIAN (Ascidia affinis), Dissected to show the chief features in the anatomy of Urochorda. THE animal is mounted with the base of attachment or posterior end downwards, and the oral or inhalent aperture or anterior end upwards. The ventral edge is on the left hand, the dorsal on the right, and the animal's left side is therefore turned to the observer. On the dorsal edge, and some- what anteriorly, is situated the atrial or exhalent aperture, through which a white bristle has been passed. There is an external transparent test, This test is secreted by the ectoderm cells, and is remarkable among animal tissues for containing cellulose or a substance very closely akin to it. It is prolonged inwards for a short distance at both oral and atrial apertures. The body walls, often termed mantle, have shrunk away from the test under the action of the alcohol in which the animal is preserved. This contraction does not take place in all species of the genus, and in this instance the body walls remain firmly united to the test in the region of the two apertures. They may be seen to be fibrillated, and consist of connective tissue with bundles of muscular fibres. In the genus Ascidia these bundles are more or less irregularly arranged, and interlace more or less with one another, a fact to be made out easily in the part of the body walls left near the atrial aperture. ASCIDIAN. 203 But in the genus Cynthia they are disposed in two layers, an outer longitu- dinal and an inner circular layer. The cavities of the pharynx and digestive tract have been displayed by the removal of their own and the body walls on the left side. Close to the margin of the oral aperture may be noticed, first, the circle of tentacles which guard the entrance, secondly, behind this circle two delicate bands, the peripharyngeal bands which run circularly round the pharynx, and thirdly, the cavity of the pharynx itself. This cavity is marked by trans- verse and longitudinal lines at right angles to one another, inclosing innumerable square meshes. The transverse lines are the transverse vessels which run from the ventral to the dorsal edge of the pharynx, connecting the two main longitudinal bloodvessels which correspond to these two edges respectively. The longitudinal lines are the longitudinal bloodvessels which run on the inner surface of the pharyngeal cavity. They extend from the anterior to the posterior end of the pharynx, and are connected by short vessels to the transverse vessels at the spot where the two sets Cross one another. The longitudinal vessels which pass from one transverse vessel to another, and lie in the same plane as they do, are very fine and inclose the stigmata or apertures which lead from the cavity of the pharynx to the peribranchial or atrial cavity. They are not visible without the use of the microscope. The cilia which beset the stigmata cause a current of water to flow from the pharyngeal into the atrial cavity. The pharynx has thus a respiratory function and is hence often termed branchial sac. The pharynx extends on the right side of the animal to its posterior extremity, and the digestive portion of the alimentary canal is placed to its left side, an arrangement which does not occur in all Urochorda, e.g. Cla- velinidae. Stretching along the dorsal edge of the pharyngeal cavity is a delicate longitudinal band, the dorsal lamina. Anteriorly it may be seen to come into contact with the peripharyngeal bands. At this point there is an opaque white spot, the nerve-ganglion which is underlaid by the neural gland, and in front of the opaque spot may be noted a crescentic curved line with .the concavity turned forwards. This line represents the dorsal tubercle, so-called, or the aperture of the neural gland. When the specimen is held in a certain position as regards the light, a yellow opaque line may be seen running along the ventral edge of the pharynx internally. This line is the endostyle, which extends from the peripharyngeal bands anteriorly to the posterior end of the dorsal lamina, close to which is the aperture from the pharynx into the stomach. A black bristle has been passed down the pharynx and through this aperture. The pharynx is attached to the body walls along thejine of the endo- style, but a cavity, the atrial or peribranchial cavity, which opens externally by the atrial aperture, extends round its dorsal edge throughout nearly its whole extent to the right and left sides, separating them from the body T04 DESCRIPTIONS OF PREPARATIONS. walls. This atrial cavity is produced by two invaginations from the exterior in the larva, the openings of which unite, forming the single atrial aperture. The atrial must not be confounded with the body cavity which, if present at all, is probably represented by blood spaces in the body walls, &c. The stomach is posterior and ventral : the intestine rises from its anterior ventral end, runs forwards for a short space, and then turns back upon the stomach dorsally and posteriorly. It passes to the left side of the aperture of the pharynx into the stomach, and then turns forwards and dorsally to end in the anus, which may be seen just above the dorsal edge of the pharynx internally to the atrial aperture. The cavity of the intestine is partially divided by a ridge or typhlosole which projects from its right wall, and commencing at the pyloric end of the stomach, extends in this Ascidian to within a short distance of the anus. In the substance of the body walls, which are left along the ventral edge of the intestine close to its origin, and in the concavity of its last curve may be noticed some opaque white cords. These are probably portions of the testis. The substance of the test is sometimes fibrillated : it contains cellular elements which wander into it from the ectoderm. These cells increase the quantity of matrix. Some of them may become pigmented and others frequently undergo extreme vacuolation, becoming little more than bladders, with a thin wall of protoplasm which contains the nucleus. Chemically, the matrix appears to be identical with the cellulose of plants. The terms ' mantle ' and ' branchial sac,' often used in speaking of the body walls and pharynx respectively, are better discarded, as it is a well-established fact that there is ho relationship between the Urochorda and Mollusca. The muscle fibres are unstriped and have the form of fusiform or filiform fibres. There is a well-developed sphincter both at the oral and atrial apertures, and the test at these spots is generally lobed — eight lobes round the oral, six round the atrial, aperture of the Ascidiidae. The part of the test between the thickened lobes is very thin and bendable. The mode in which the muscle fibres are disposed in the body walls is characteristic in the sub-groups of Urochorda. The connective tissue corpuscles of the body walls are often pigmented. The tentacles spring from a ridge which coincides with the line in the oral aperture where the test ends. They are simple and filiform. Posteriorly to the circle of tentacles is a plane area — the praebranchial zone — limited behind by two ciliated ridges bounding a ciliated groove. The ridges are the two peripharyngeal bands. The posterior is continuous ventrally with the ridges bordering the endostyle, and the groove of the latter is continuous with the peripharyngeal groove. The posterior is also continuous dorsally with the dorsal lamina. The anterior band forms a complete circle. The cavity of the pharynx is lined with endoderm. Its outer or atrial surface is covered by the invaginated ectoderm. The stigmata lying between the transverse vessels and the fine longitudinal vessels are usually straight, but sometimes curved. Under some conditions it appears that the direction of the ciliary current may be reversed, and water be expelled at the oral instead of the ASCIDIAN. 105 atrial aperture. The system of internal longitudinal vessels is very well developed in Asciditdae, and it is characteristic of the family to have a prominent papilla developed from the free surface of these vessels at spots opposite to the short vessels which connect them to the transverse system. The endostyle is a groove, at the bottom of which are cells bearing extremely long cilia. The sides of the groove are ridged, and the ridges are caused by the large size of certain cells which appear to secrete mucus. The mucus col- lects into small lumps or balls which are carried forwards to the peripharyngeal groove. They traverse this groove and are then conveyed by the ciliated dorsal lamina backwards to the stomachal opening. The dorsal lamina has anteriorly a median epipharyngeal groove, the extent of which varies in different Ascidians. The free edge of the lamina is curved, usually to the right. It ends posteriorly in a low ridge continuous with the endostyle. The ganglion is single, generally swollen at either end where it gives off an anterior and posterior set of nerves. The central part is fibrous, and the ganglion cells are placed peripherally. The ganglion lies in the body walls, and always between the oral and the atrial apertures, on the opposite side of the body to the endostyle. As to organs of special sense, the tentacles must be regarded as tactile ; and in many Ascidians there are orange-coloured visual (?) spots between the lobes of the oral aperture. The sub-neural gland consists of caecal ramified tubes. They open into a duct which lies between the gland and the ganglion. This duct runs forwards and opens into a ciliated depression of the prae-branchial zone. The margins of the depression are prominent, and the aperture has usually a crescentic shape, the concavity of the crescent being turned forwards. The margins may be much modified in shape. They constitute what is often spoken of as the ciliated sac or dorsal tubercle. The gland has been homologised by Julin with the pituitary body of the brain in higher Vertebrata, and it has been suggested that it has a renal function. In some cases it has secondary ducts opening into the atrial cavity. The digestive portion of the alimentary canal is disposed variously in different groups of Urochorda. The liver (?) may be represented by glandular tissue coating the stomach, and sometimes attaining considerable size, or by a system of clear tubes ramifying over the stomach and part of the intestine and opening into the pyloric portion of the former. A renal organ is probably represented by clear walled vesicles containing concretions in which uric acid has been found, and situated round the intestine and in the body walls. These vesicles have no ducts. The heart is more or less fusiform, and is inclosed in a delicate pericardium. It lies in Ascidia along the posterior ventral edge of the stomach on the left side. It gives off (i) a ventral vessel which sends a branch through the mantle to the test and then runs forwards beneath the endostyle, and is connected to the transverse system of vessels in the walls of the pharynx; (2) a dorsal vessel from its opposite end which sends a branch through the mantle to the test parallel to the one above mentioned. These two branches divide in the test, end in ampullae, and intercommunicate. The remaining branches of the dorsal vessel go to the body walls; and to the stomach intestines and reproductive organs from which the blood is collected into a large vessel running along the dorsal edge of the pharynx and connected to its transverse vessels. There are 106 DESCRIPTIONS OF PREPARATIONS. branches also (= connectives of Hancock) from the pharynx to the body walls, thence to the test and back again. When the heart pulsates so as to drive the blood from the ventral to the dorsal surface, it draws arterialised blood from the pharynx and venous only from the test ; when in the opposite direction the blood it draws has previously passed through the viscera, body walls, and test. The blood itself is a clear plasma containing rounded nucleated corpuscles, many of which are pigmented, generally' yellow, red, or brown, but white and blue are sometimes found. All Urochorda are hermaphrodite, but in most instances the male and female organs are not ripe at the same time. These organs in Ascidia are racemose glands placed on the left side of the body between the intestine and stomach. The testis is composed of delicate white tubules, ramifying dichotomously, and spread over the ovary, stomach, and intestine. Both oviduct and vas deferens run along the dorsal edge of the intestine and open near the anus. The Ascidian is an example of an animal which has lost, more or less com- pletely, the structure typical of the phylum, and even, strictly speaking, of the class to which it belongs, at the same time acquiring marked peculiarities of its own. It is an instance of what has been termed by Professor Ray Lankester Degeneration^ or 'a gradual change of structure in which the organism becomes adapted to less varied and less complex conditions of life.' Elaboration is the converse : but ' elaboration of some one organ may be a necessary accompaniment of Degeneration in all the others : in fact this is very generally the case.' It is when the total result of both processes combined leaves the organism ' in a lower condition, that is, fitted to less complex action and reaction in regard to its surround- ings than was the ancestral form with which we are comparing it (either actually or in imagination), that we speak of that animal as an instance of Degeneration.' The causes of degenerative evolution are, according to the same authority, four in number : (i) parasitism ; (2) fixity or immobility; (3) vegetative nutrition; (4) excessive reduction of size. Instances are, of the first, various Copepoda, e. g. Lernaea ; of the second, a barnacle (Lepas) ; of (i) and (2) combined, the parasitic Cirripedia or Rhizocephala ; of the third, the Turbellarian worm Convoluta • and of the fourth male Rotifers, etc. There can be no reasonable doubt that an Ascidian is one of the Chordata. The structure and development of the larva suffice to establish this point fully. The cTSiHral nervous system originates as a neural groove the side of which closes over to form a neural tube; there is a cerebral or myelonic eye; the caudal notochord is derived from the archenteron ; the pharynx becomes a respiratory organ pierced by slits, with walls richly vascular. These structures reach a certain degree of perfection. But with the fixation of the larva a series of regressive changes sets in. The notochord disappears with the swimming tail. The nervous system is reduced to a fraction of what it was in the larva. The eye is lost. The pharynx however becomes much more complicated and enlarged relatively to the remaining organs. The larval characteristics are however more or less retained and specialised in the free swimming order Larvacea. Tunicata. Herdman, Challenger Reports, vi. 1882 ; and Bronn, Klass. und Ordn. des Thierreichs, Malacozoa, iii. i. 1862. Notes on British Tunicata. Herdman, J. L. S. xv. 1881. Ascidies Simples SHELL OF EDIBLE SNAIL. IO; des cotes de France, de Lacaze Duthiers, A. Z. Expt. iii. 1874; vi. 1877. Heller, Untersuchungen, &c., Dk. Wien. Akad. xxxiv. 1875; xxxvii. 1877; and SB! Wien. Akad. Ixxvii. Abth i. 1878. Degeneration. Ray Lankester, Nature Series, 1880. Various points. R. Hertwig, J. Z. vii. 1872, Test. O. Hertwig, J. Z. vii. 1872. Semper, Verhandl. Phys. Med. Gesellsch. zu Wurzburg, vii. 1875. Tunicin = Cellulose. Cf. Watt's Dictionary of Chemistry, v. p. 918, and Suppl. 2, 1875, p. 271. Nervous System. E. van Beneden et Julin, Archives de Biol. v. 1884. Neural gland. Julin, Archives de Biol. ii. 1881. Herdman, Nature, xxviii. 1883. Endostyle. Fol, M. J. i. 1876. Tubular gland opening into stomach. Chandelon, Bull. Acad. Royale Belgique (2) xxxix. 1875. Ulianin, on Doliolum, Fauna and Flora des Golfes von Neapel, x. 1884. 20. SHELL OF EDIBLE SNAIL (Helix pomatia). WHEN the shell has its apex directed upwards, and its aperture down- wards and towards the reader, its spire, as is the case with the great majority of Gasteropod spiral shells, ascends obliquely towards the right. It is in consequence termed ' dextral.' In the living animal inhabiting such a shell, the heart is on the left, the generative, respiratory and anal aper- tures on the right of the body. The aperture of the shell in Helix as in all vegetable-feeding Gastropoda is entire, i. e. forms an unbroken curve : in carnivorous Gastropoda on the contrary it is notched or produced into a canal which lodges a process of the mantle or siphon : e. g. in the Whelk. These two varieties of aperture are known respectively as ' holostomatous ' and ' siphonostomatous.' The columella or pillar formed by the contact of successive whorls of the shell lies on the left side of the aperture in the angle between the first whorl and the peristome or free margin of the aperture. It is hollow and the external opening or umbilicus is partially hidden by the peristome, not wholly as it is in the Garden Snail, Helix aspersa. The smooth rounded off edge of the peristome shows the animal to have been adult. The shell is coarsely striated in a direction parallel to the margin of the aperture, i. e. corresponding with the lines of growth, and more delicately in a spiral direction, parallel to the five coloured bands with which it is marked. The ' apex ' or ' nucleus,' the tip of the shell or the part first formed, is smooth and semi-porcellanous in appearance, thus differing, as is so often the case, from the rest of the shell. The spiral line marking the point of contact between one whorl and its successor is known as the ' suture.' The shell increases in thickness towards the peristome, and it is probable that but little addition to its substance takes place except in 108 DESCRIPTIONS OF PREPARATIONS. that region. In this particular, as well as in being univalve and possessing a much smaller amount of organic matter, the shell of a Gastropod contrasts with that of a Lamellibranch. There is no operculum in the Pulmonata, the order of Gastropoda to which the Snails and Slugs belong. When the snail hibernates it closes the aperture of its shell by a whitish disc, the hibernaculum or epiphragma — a structure containing much organic matter and calcareous granules of variable shape irregularly scattered. It appears to be secreted by the margin of the mantle or collar (see next preparation) rather than by the foot, as has been supposed. There are sometimes two such epiphragmata, one within the other. The moveable clausilium of the pulmonate Clausilia, which covers the aperture of the shell when the animal is retracted, appears to be a structure homologous with the epiphragma. The majority of Gasteropod spiral shells are dextral. When the spire of a shell turns in a direction opposite to what is normal, it is said to be 'reversed.' In some species this is a common occurrence, e. g. in the Common Whelk ; in others it is so usual that the type of shell commonly met with is sinistral, e. g. Clausilia. The lower whorls of the shell may, in a few instances, become loose and straggling, e.g. Vermetus, Siliquaria. The reflection of the peristome over the umbilicus in Helix is due to the presence of a ' columellar lobule ' or projection of the collar. When the shell is injured at the peristome before the animal is adult, both the outer non-calcareous layer ' periostracum ' or ' epidermis ' and the calcareous layers internal to it are repaired : when at any other spot, or after the animal is- adult, only the latter, and in this case the shell is thickened internally. The colouring matter is secreted by unicellular glands of the collar. The spiral shell, as in all Gastropoda with few ex- ceptions, is a secondary shell, i. e. formed on the outer surface of the mantle, and not from the shell-gland of the embryo. A primary shell, i.e. one formed by this gland, rarely persists. It is found, however, as the ' nucleus ' of the shell in the pulmonate Clausilia, and as the internal shells of the Slugs Arion and Limax. The shell of the Helices is stated by Longe and Mer to consist of an external organic cuticle or periostracum ; of an outer calcareous layer containing colouring matter, and composed of (i) a thin external division with confused striation, (2) a thicker internal division which consists of vertical prisms ; and of an inner nacreous layer including several laminae of prisms arranged horizontally, the axes of the prisms in the different laminae being at right angles to one another. The cuticle is formed by a cutogenic apparatus situated just behind the collar. It is composed of a pallial groove into which glandular caeca open, and an epithelial organ con- sisting of long bottle-shaped cells which secrete granules. The epithelial organ disappears in the adult, and the glands of the pallial groove gradually undergo atrophy. The rest of the shell appears to be formed by the edge of the mantle ; the nacre from its surface in general. In Zonites algirus, according to Nalepa, the cells of the pallial groove develope in spring— the period at which growth of the shell takes place — into long unicellular glands, and the cells of a ridge (? = epithelial organ of Longe and Mer) just behind the groove enlarge into flask-shaped gland EDIBLE SNAIL. 109 cells. The cuticula is secreted by these two sets of gland cells. It is possible that the glands may be developed in Helix, as in Zonites, in the spring of the year, and that the atrophy of the organ in the adult (supra] merely marks a period of rest. (See Nalepa, SB. Akad. Wien. Ixxxvii. Abth. i. 1883.) A layer of nacre or mother of pearl is not generally found in the calcareous shells of Gastropoda. The calcareous matter is usually arranged in three layers composed of parallel lamellae, those of the outer and inner layer usually disposed parallel to the suture, of the middle layer at right angles to it. The direction of the layers is sometimes reversed. The lamellae are vertical to the surface of the shell, and each lamella is composed of calcareous prisms, all parallel to one another, but forming an oblique angle with the surface of the shell. The prisms of adjoining lamellae are disposed in contrary directions to one another. The calcareous matter is in the form of Arragonite (Ca CO3). Splinters of the shell are hard enough to scratch Calc-spar. The Gastropod shell is composed chemically of about 1-5 per cent, of an organic substance (Conchiolin) ; of about 95-98 per cent, of Lime carbonate ; with small quantities of Magnesium carbonate ; phosphates ; silica and alumina. The inner whorls of the shell are in some Gastropoda partially absorbed. In some cases the animal withdraws from the older region of the shell. The part forsaken is closed off by a calcareous lamina secreted by the surface of the mantle. The process of withdrawal and closing off may be repeated several times. Technical terms, structure, &c. Woodward, Manual of the Mollusca (ed. 3.), 1875, p. 28, p. 204. Keferstein, Mollusca, Bronn's Klass. und Ordn. des Thier- reichs, iii. 2. p. 899, p. 1181. Zittel, Handbuch der Palaeontologie, Abth. i, ii. p. 153. Ley dig, Haut-decke und Schale der Gastropoden, A. N. 42. 1876. Structure and formation of Shell in Helices. Longe and Mer, C. R. xc. 1880, cf. A. N. H. (5) v. 1880., Epiphragm. Keferstein, op. cit. supra, p. 1186. Binney, Terrestrial Mollusca of United States, ii. 1851. Analysis and source of lime. Barfurth, A. M. A. xxii. 1883, p. 509. 21. EDIBLE SNAIL (Helix pomatid), Dissected so as to show the position of the heart and the respiratory cavity. THE shell, with the exception of a part of the columella or spire left in situ, has been removed. The foot is completely expanded and the left upper tentacle is partially protruded. A depression on the animal's right side anteriorly above the foot, marks the position of the generative aperture. The thickened collar is seen crossing the animal's dorsum. In it, on the right, is the pulmonary aperture, and below the aperture, again to the right, the columellar lobule. The pulmonary chamber has been opened by a trans- verse incision, and its cavity exposed by turning forward the anterior flap. Its roof is for the most part thin, and is formed by the branchial fold of the mantle, the anterior edge of which has fused, as in most pulmonate Molluscs, HO DESCRIPTIONS OF PREPARATIONS. with the skin of the dorsum leaving only the small pulmonary aperture. Its floor is thickened and muscular, and represents that portion of the general integument which is roofed over by the fold of the mantle. The heart, composed of a thin-walled auricle and thick-walled ventricle, is seen as in dextral Molluscs, on the left side, and below it lies in this preparation the triangular pale-coloured kidney. The vessels have been injected with a red-coloured fluid from the auricle. The right half of the roof of the pulmonary chamber is covered by vascular ramifications : the anterior part of the left side is smooth ; its posterior portion corresponds to the pericar- dium and kidney. The pulmonary vein passes along the edge of the cut on the right side towards the auricle. It is joined, but this fact cannot be seen here, before it enters the auricle by the efferent kidney veins. The right border of the pulmonary chamber which is applied to the line of the suture in the shell is relatively thick. It is muscular, and muscular fibres pass off from it into the roof of the chamber. One of the vessels bringing venous blood to the chamber runs along this border. When the animal is retracted into its shell by the action of the colu- mella muscles, seen in the next • preparation, air is forced out of the pulmonary chamber ; conversely, when the animal expands again by the contraction of the muscular integument including the floor of the pulmonary chamber, air is drawn in But, except during the acts of expansion and retraction, interchange of air in the pulmonary chamber with the air without must take place only by diffusion. The foot is relatively large in most Pulmonata, and simple in contour. It is divided by a transverse furrow into a fore- and hind-part only in Pedipes and Auri- cula brunnea. The sole of the foot is ciliated, and in Arion its lateral walls as well. A supra-pedal gland appears to be present in all Pulmonata, as well as in some other Gastropoda (Azygobranchia, Opisthobranchid). It opens above the anterior margin of the foot beneath the head. It consists of a ciliated duct extending back- wards in the substance of the foot near its coelomic surface. Into this duct open unicellular glands. Sense cells have been stated to be present in the epithelium of the duct ; but it is very doubtful if the gland has the olfactory function which has been ascribed to it. A caudal aggregation of mucous glands, situated posteriorly on the dorsal aspect of the foot and opening into a depression of the integument, is found in Arion ater and some other Slugs. But no Pulmonate possesses a pedal gland opening near the centre of the creeping sole of the foot, such as is found in many Azygobranchiate Gastropoda. A supra-pedal and a pedal gland coexist in the Azygobranchiate Cyclostoma elegans, which leads a terrestrial life. The aperture of the pedal gland has often been taken for an aquiferous pore. Many Mollusca possess the power of suspending themselves by mucous threads, and not only those of terrestrial habit but aquatic as well, e.g. Limneidae, and the marine Litiopa and Rissoa parva. In the Limaddae the thread appears to be derived from the mucus coating the surface of the body. EDIBLE SNAIL. in The position of the anal, renal, and generative apertures in front and on the right side of the body is due to the twist of the dorsal aspect of the body or visceral dome, which is characteristic of Gastropoda Anisopleura. The presence of but a single kidney and a single generative duct is nearly equally characteristic. The pulmonary chamber must be regarded as formed by the fold of the mantle which in branchiate Gastropoda constitutes the roof of the branchial cavity lodging the gills or ctenidia. The latter are aborted, and the mantle-fold itself has become vascular and respiratory. The fore-edge of the mantle-fold in the aquatic Pulmonate Limnaeus is free, and has not undergone fusion or ' concrescence ' with the dorsum, as it has in Helix and its allies. When Limnaeus inhabits the 4eep waters of lakes it is said to admit water to the pulmonary chamber, instead of coming to the surface at intervals for a fresh supply of air, as it does when it inhabits shallow streams and ponds. An adaptation of the branchial fold of the mantle to aerial respiration occurs also in certain streptoneurous Gastropoda Aniso- pleura : namely, in the Pneumonochlamyda and in the genus Ampullaria, among Azygobranchia. In the last named, which is amphibious, and is found in tropical America, Africa, and the East Indies, the left side of the branchial cavity contains the ctenidium, and is separated by a fold from the right side, the walls of which are vascularised. The Pneumonochlamyda^ to which our English Cyclostoma elegans belongs, have lost the ctenidium, and respiration is carried on solely by the walls of the branchial cavity, as in the Pulmonata, from which order the Azygobranchia differ in such essential features as the twisted character of the visceral nerve-loop, and the separateness of the sexes. The view, which is advocated by von Ihering, that the pulmonary chamber in Pulmonata is derived from the renal organ, does not appear to be tenable. The size and extent of the pulmonary .chamber vary greatly in the Pulmonata. The pericardial cavity is a portion of the coelome closed off completely from the remainder, but communicating, as in all Mollusca where it is present, with the exterior through the nephridium, with which it is connected by a ciliated nephridial tube. The aorta divides into an intestinal and a cephalic branch, the latter passing through the infra-oesophageal collar. The system of capillaries appears to be very complete in Helix pomatia, judging from the result of injections. The vascular system of Zonites algirus has been accurately investigated by Nalepa. The arteries and capillaries have proper walls, and are lined by an endothelium. The capillaries communicate by short branches with a narrow meshwork of wide blood-spaces — the ' transition ' vessels, which open into the venous spaces or the coelome by infundibular orifices, seen also in Arion ater (=rufus) by Jourdain. The venous system is represented in part by the coelome, in part by vessel-like spaces/ the walls of which are formed of homogeneous connective tissue with scattered nuclei, but which are not lined by an endothelium. The pulmonary vein pulsates rhythmically in Zonites. There is in the same Pulmonate a nervous network in the walls of the auricle, of the ventricle, and aorta, in the last two instances derived from the genital nerve. Nerves are found also in the walls of the larger vessels. Ganglion cells connected with these nervous networks appear to be rare. They are found in the walls of the heart of certain marine Rhipidoglossa (Fissurella, Haliotis, Trochus), according to Haller. See Haller, M. J. ix. 1883, p. 61 ; cf. Dogiel, A. M. A. xiv. 1877. DESCRIPTIONS OF PREPARATIONS. The blood si Helix contains haemocyanin, a respiratory substance in which copper is present in combination with a proteid. It is colourless when deoxidised, bright blue when oxidised. Hence the coelomic fluid of a Helix exposed to air- assumes a violet tinge. The blood-plasma of the pulmonate Planorbis contains haemoglobin. Haematin is found in the liver-secretion of Helix pomatia, H. aspersa, Arion ater, and Limax. Amoeboid corpuscles occur, but rather sparingly, in the blood. The shape of the nephridium varies in Pulmonata. It has a long tubular duct in Helix, &c., opening near the anus ; whilst in other Pulmonata, such as Arion, it has a simple round opening. Its cavity is large, and its walls are lamellate. It really consists of a urinary chamber into which open acini of very large calibre. These acini are bound together externally by connective tissue. The renal cells are ciliated, and urates of Ammonium and Calcium are found in them and in the cavity of the sac. Free uric acid and Guanin occur in Zonites as well as Ammonium urate. The nephridium of Helix pomatia is supplied with arterial blood from the pulmonary chamber and by the renal arteries ; by the latter alone in Zonites. Gastropoda. Ray Lankester, ' Mollusca,' Encyclopaedia Britannica (ed. ix.), xvi. 1883. Keferstein, Bronn's Klass. und Ordnungen des Thierreichs, iii. 2. 1862-66. (Pulmonata, p. 1160.) Pulmonata. Semper, Z. W. Z. viii. 1856-57. Leidy, in Binney's Terrestrial Air-breathing Mollusca of the United States, i. 1851, p. 198. For the Slugs, see literature to Plate v (post). Helix pomatia. Cuvier, Memoires pour servir a 1'histoire, &c. des Mollusques, 1871. (Annales du Muse'um, vii. 1806.) H. aspersa. Howes, Atlas of practical Elementary Biology, 1885, Pis. xiii. xiv. Taylor, Journal of Conchology, iv. 1883, pp. 102-105. Zonites algirus. Nalepa,- SB. Akad. Wien. Ixxxvii. Abth. i. 1883. Ancylus. Sharp, Proc. Acad. Nat. Sc. Philadelphia, 1883. Onchidium (Marine Slug). Bergh, Challenger Reports, x. 1884, p. 126, and M. J. x. 1885. Muscles and locomotion of snails. Simroth, Z. W. Z. xxx. Suppl. 1878; xxxii. 1879 > xxxvi. 1882. Glands of foot. Houssay, A. Z. Expt. (2) ii. 1884. Carriere, A. M. A. xxi. 1882. Supra-pedal gland. Sochaczewer, Z. W. Z. xxxv. 1881. Id. and Simroth, ibid, xxxvi. 1882. Sarasin, Arb. Zool. Zoot. Inst. Wurzburg, vi. 1883. Mucous threads spun by Mollusca. Martens, Z. A. i. 1878. Eimer, ibid. Tye, Quarterly Journal of Conchology, 1878. Pulmonary chamber = a uropneustic apparatus. Von Ihering, Z. W. Z. xli. 1884 ; criticisms of view, see Semper, Arb. Zool. Zoot. Inst. Wurzburg, iii. 1876-77; cf. Simroth, Z. W. Z. xxvi. 1876, p. 337 et seq. On Ampullaria. Jourdain, C. R. 88, 1879. Sabatier, ibid, and A. N. H. (5), iv. 1879. Vascular system and pulmonary vessels. Nalepa, op. cit. supra. Apertures of capillaries in Arion ater (= rufus]. Jourdain, C. R. 88, 1879. Haemocyanin, properties and distribution. Halliburton, Blood of Decapoda (Crustacea], Journal of Physiology, vi. 1885. MacMunn, Q. J. M. xxv. 1885. Haemoglobin in Mollusca. Ray Lankester, P. R. S. xxi. 1872; cf. Sorby, Q. J. M. xvi. 1876. For colouring matters in general, see Krukenberg, Vergleich. Physiol. Vortrage, iii. 1 884. MacMunn, Proc. Birmingham Philosoph. Soc. iii. 1881-83. Cf. Moseley, Q. J. M. xvii. 1877. EDIBLE SNAIL. Connection between nephridium and pericardium in Helix. O. Niisslin, Beitrage zur Anat. und Physiol. der Pulmonaten, Tubingen, 1879. Cf. Haller, Marine Rhipidoglossa, M. J. xi. 1886. 22. EDIBLE SNAIL (Helix pomatia), Dissected so as to show its digestive and reproductive, together with portions of its circulatory and respiratory organs. THE body has been detached from the shell, and the greater part of the mantle and skin has been removed. The coils of the right lobe of the liver (hepato-pancreas), which occupy the upper whorls of the shell and lodge the hermaphrodite gland, are arranged with the rest of the reproduc- tive organs on the right hand : the left lobe of the liver, intestine, heart and respiratory chamber lie to the left : the nerve-collar, buccal mass and stomach occupy the centre. Between the latter and the parts to the left lies at a lower level the sole-shaped ' foot ' upon which the animal creeps. A black bristle has been passed between the cerebral or supra-oesophageal ganglia and the buccal mass which contains the crescent-shaped chitinoid jaw and the lingual ribbon or radula, structures invisible here. The buccal mass is somewhat retracted, as it is when the snail * draws in his head.' On the left side of the buccal mass may be seen the sheath which contains the connectives from the cerebral to the infra-oesophageal gangli£, a small portion of which with numerous nerves proceeding from it is also visible. From the base of the buccal mass projects the small conical ' sac' of the radula below, while the oesophagus and ducts of the salivary glands enter it above, — the oesophagus between the two salivary ducts. The oesophagus expands into the stomach which is embraced by the salivary glands. The right and left bile ducts, from the right and left lobes of the liver respectively, enter one on either side the pylorus, and at this spot there is a small projecting median caecum. Just below the heart, which has been exposed in the pericardium, and above the aorta cephalica, which has been cut transversely, the intestine passes on to the convex surface of the left lobe of the liver. It reappears on the left side of this lobe in the preparation, on the right side in the natural position of the parts in the living animal, and passes straight to the anus. A white bristle has been placed in the anus through the pulmonary aperture. To the right side of the terminal segment of the intestine is seen the wall of the pulmonary chamber with its network <• of vessels. The hermaphrodite gland or ovo-testis is lodged in the concavity of the penultimate and ante-penultimate coils of the right lobe of the liver, which occupy the lowest portion of this preparation on the right hand. From the gland a convoluted hermaphrodite duct passes up to a spleen- I 114 DESCRIPTIONS OF PREPARATIONS. shaped body, the albuminiparous gland, and from this spot the duct passes upwards again as a much thicker tube consisting of two parts — one, ovi- ducal or uterine, much plicated ; the other granulated, the vas deferens. These two parts are not separated completely from one another internally, but at the level of a cylindrical muscular organ, the dart-sac, they become independent. The vas deferens passes to the left, turns round the right upper tentacle, and then passes back again to the base of the penis at the point where the flagellum or organ for secreting the spermatophore also joins it The retractor muscle of the penis is attached to it near this same point The flagellum is of great length, and to its right lies a duct of similar, perhaps correlated length, terminated by a bulb. This is the receptaculum seminis or spermatheca, an appendage to the female organs. Lying below the penis and dart-sac are the two bundles of multifid vesicles, which enter the female duct close to the dart-sac. The generative aperture is on the right side of the head. The integument varies in thickness in different parts of the body. In the Pulmonata with external shells it is exceedingly thin and delicate where it lines the coils of the shell. It is thicker where it forms the roof of the pulmonary chamber, and thickest of all on the exposed parts of the body. The epidermis consists of a single layer of cells. These cells are columnar and provided with a very delicate cuticula, thickened in certain spots on the tentacles and oral lobes (infra, p. 121-2). The terrestrial differ from the aquatic Pulmonata, and from aquatic Gastropoda in general, in the restricted extent to which cilia are present. In aquatic forms they are as a rule found over the whole of the exposed surface of the body; in the terrestrial Pulmonata, on the contrary, they occur on the sole of the foot but sometimes not over its whole surface ; occasionally along its lateral margins (Ariori) ; round the aperture of the supra-pedal gland and that of the pulmonary chamber in some instances (e. g. Helix nemoralis, Limax margin- atus)1. The cells occasionally contain pigment. Among them are scattered on the exposed surface of the body, sensory cells (infra, p. 121), small goblet cells, and the apertures of mucous, pigment, and calcareous, glands. The mucous glands are unicellular, but project inwards into the cutis. The mucus varies in character, but not infrequently contains whetstone-shaped bodies. The glands are especially large on the collar. Pigment-producing glands are either richly pigmented epi- dermic cells, e. g. on the collar where they secrete the colouring matter of the shell, or long irregular bodies composed of more than one cell, extending out- wards between the cells of the epidermis, inwards into the cutis, where they are said by Leydig to become continuous (?) with the pigmented networks there pre- sent. The glands which produce calcareous matter resemble the irregularly-shaped pigment-producing glands, and, like them, open externally and are continuous (?) 1 Inter-cellular passages are said to exist between the epidermis cells, and are supposed to be the passages by which water enters the blood. They have been described and figured in Helix by Nalepa, SB. Akad. Wien. Ixxxviii. Abth. I. 1883. For absorption of water in Mollusca, see Schiemenz, Mitth. Zool. Stat. Naples, v. 1884. Cf. Sarasin, C. F. and P. B., « Directe Communica- tion des Blutes,' &c., Arb. Zool. Zoot. Inst. Wurzburg, viii. pt. i, 1886. EDIBLE SNAIL. 115 internally with the cells of the cutis which contain calcareous matter. The pig- ment and calcareous glands do not extend over the surface of the foot. The latter are especially numerous on the collar, and round the margins of the foot in Helix. The body-wall consists of connective tissue and muscle-fibres. Its surface in the exposed parts of the body is raised into ridges and tubercles. The muscle- fibres are non-striated long cells. They occasionally appear striated, but the cause of the striation is not known. The connective tissue consists of plasma-cells, a matrix, and fibrils. The plasma-cells are richly developed in all Mollusca, but especially in Pulmonata where three kinds are distinguishable. These are: (i) oval or round in shape with a transparent protoplasm and a round nucleus ; (2) irregular cells containing refractile granules which are not fatty in nature ; and (3) cells with granules of lime carbonate, which are immeasurably fine in the interstitial tissue of the upper coils of the visceral dome, coarser elsewhere. These cells are imbedded singly or in masses in a matrix which contains stellate cells, and connec- tive tissue fibrils, the latter much more scanty than in other Mollusca. The matrix is sometimes much reduced, and the first kind of plasma-cells then appears to form sheaths round various organs, e. g. nerves, but the stellate cells of the matrix may still be detected among them. The connective tissue membranes are generally pierced by apertures, many of which are bounded by refractile rings, the product of several cells which encircle the aperture. Glycogen has been detected in the plasma-cells of the first kind both in Anodon and Helix (Blundstone, P. R. S. xxxviii. 1884-85). Branched pigment-cells are found in the cutis and sometimes extend inwards to its deeper layers. Changeable chromato blasts have been observed by Ley dig in Limax variegatus and L. (j=-Amalid) carinatus (A. M. A. xii. 1876, p. 541). The buccal mass consists chiefly of an odontophore, using that term in a wide sense to include the muscular and cartilaginous apparatus in connection with a chitinoid (?) radula or lingual membrane bearing transverse rows of teeth. The radula is developed within a radular sac and is perpetually growing throughout life. The sac is essentially a ventral diverticulum of the buccal cavity, and its lining cells are continuous with the oral epithelium. It is crescentic in transverse section. At the blind end of the sac the lining cells are differentiated into odontoblast cells, which form a transverse ridge broken up in correspondence with the number of teeth present. This ridge contains in vertical section four to five large cells in Pulmonata and Opisthobranchia \ a number of elongated cells in other Glossophora. The singly refractile core of each tooth is secreted by these cells, the doubly refractile enamel-like outer layer by the cells lying immediately dorsal to the odontoblast cells, whilst the membrane or matrix uniting the bases of the teeth is formed by a single cell in Pulmonata and Opisthobranchia, by several in other Glossophora, placed in each case just ventral to the odontoblast cells. These matrix cells split up at their free ends into fibres. The radula is carried by a subradular membrane, developed by the cells of the ventral wall of the sac. The membrane is borne upon cartilaginous pads, and the whole is worked by a system of muscles, protractor and retractor. Other muscles flatten the radula or convert it into a groove. And in some cases the subradular membrane with the radula slides backwards and forwards to a limited extent over its cartilaginous supports. I a Il6 DESCRIPTIONS OF PREPARATIONS. The chromogen, called myohaematin by MacMunn, is found in the buccal muscles (and in the heart) of Helix, Arion^ Limax and of other Pulmonata. According to Ray Lankester, haemoglobin occurs in the buccal mass of some Mollusca. The form and number of the teeth in a transverse row are very variable in the Glossophora. In Pulmonata a median tooth may be distinguished from an indefinite number of admedian teeth. Such a dentition may be formulated thus — oo . i . oo , and is termed myrioglossate. Various technical terms have been applied to the variations in number and arrangement of the teeth already alluded to. A median tooth may be present or absent : so too admedian teeth : and in the arrangement known as rhipidoglossate lateral may be distinguished from admedian and median teeth. The size and shape of the teeth themselves are also extremely variable. In the Pulmonata each tooth is provided with a recurved hook, usually simple, but sometimes denticulated \ The more or less chitinoid jaw lies on the dorsal aspect of the oral cavity. In most Helices it is a crescentic plate ; its free surface provided with antero-pos- terior ridges. The form of the jaw varies in different Pulmonata : in aquatic Pulmonata it may consist of more than one piece, and in a few terrestrial forms it is absent, e. g. Testacella. The salivary glands vary in form and size in different Pulmonata. They are compound glands, but the ultimate acini are composed of a number of uni- cellular glands. They receive bloodvessels and nerves, the latter from the buccal ganglia. An additional salivary gland lies imbedded in the buccal mass in Helix pomatia round the entrance of the main salivary ducts. It consists of unicellular glands, partly opening into the salivary ducts, partly uniting and opening by ducts of their own into the oral cavity. The salivary extract converts starch into sugar. The muscular coats of the digestive tract are an external circular, and an internal longitudinal coat. The longitudinal ridges of the internal surface are formed by the longitudinal muscles. The epithelial cells possess a cuticle, and cilia are present in places ; in young Helices over the whole surface of the stomach. Goblet cells are present in Zonites and apparently in Helix-, and in the former Nalepa observed small cells at the base of the other cells, which they are apparently destined to replace. The liver or hepato-pancreas is a compound acinous gland. Its acini are held together by connective tissue, ramifying bloodvessels and nerves. The epithelium forms a single layer. In the ducts are found columnar cells, sometimes ciliated, but not in H. pomatia (Barfurth), as well as mucous-cells : in the acini three kinds of cells — liver-cells (Barfurth) or granular-cells (Frenzel); ferment-cells (Barfurth), club-cells, or club-shaped ferment-cells (Frenzel); and calcareous-cells. The granular cells contain each a vesicle which incloses a number of more or less coloured highly refractile granules, fat globules variable in size, 1 The character of the dentition is especially useful for classificatory purposes in the Azygo- branchiate Gastropoda. The names in use are chiefly the following. The formulae indicate the presence or absence of median, admedian, and in Rhipidoglossa of lateral teeth. Rachiglossa, o. i. o, e.g. Volutidae: Hamiglossa, T. i. i, e.g. Murex, Buccinum: Toxoglossa, i. o. i, e.g. Conidae: Taenioglossa, 3. i. 3, e. g. Cerithium, Natica, Littorina, Paludina: Cteno- or Pteno-glossa, oo. o. oo , e. g. lanthina : Rhipidoglossa, oo. 4-7. i. 4-7. oo, e. g. Turbo, Nerita (also Haliotis, Fissurella, among Zygobranchiata). There are forms of dentition which do not fall under any of the above-written terms, e.g. Patella Vulgata, the Limpet, with 3. i. 4. i. 3. EDIBLE SNAIL. 117 and albuminous bodies. In Helix the granules are clear yellow in colour ; sometimes, as in Arion, they are the only coloured constituent of the cells. These granular cells are absent in Cephalopoda. The ferment-cells vary in appearance in different Mollusca. They are absent in Chiton, Patella, Fissurella, and Pteropoda ; doubt- fully present in some azygobranchiate Gastropoda and some Lamellibranchiata. In shape they are like a club or pear, with the larger end turned towards the cavity of the acinus. They contain each a vesicle inclosing more or less coloured fluid, viscous or semi-solid bodies, fat and albumen globules. The refractile index of the contents of the vesicle is low in Helix and Cephalopoda, and their colour is similar to that of the granule-cells. It is generally more or less different in the Opisthobranchia and Azygobranchia. It is similar in intensity in Pulmonata, more intense in other groups (Frenzel). The calcareous cells are wanting in Lamellibran- chiata, Pteropoda, and in Natantia (= Heteropodd) among Gastropoda, and other isolated instances in this Class. Their size is great, and they are more or less triangular with the base external, and the apex not reaching the cavity of the acini. They contain the so-called calcareous globules, which vary in chemical characters. Lime is always an ingredient, and, according to Barfurth, in com- bination with phosphoric acid, a view opposed by Frenzel. Barfurth has shown that the percentage of inorganic ash obtainable from the liver of Helix pomatia varies at different seasons. In May the amount averaged 20.24 per cent; in September 25.72 per cent. ; after the breaking and repair of the shell it fell at once to 16.99 per cent. ; and after the formation of the epiphragma to 10.26 per cent., the normal figure during the winter rest of the animal. It may be added that the quantity of phosphates in the epiphragma is 5.52 per cent, as opposed to 0.85 per cent, in the shell. The mucus of Helix contains lime, and to a very large amount in Arion. The secretion of the liver is acid, and has beeri found to have a diastatic and a peptic action in H. pomatia. It contains enterochlorophyl in Helix pomatia, in many other Gastropoda, some Lamellibranchiata, and haematin in various Pulmonata, e. g. Helix pomatia, H. aspersa, and Arion ater, &c. Gly- cogen is found in the liver ; in Helix, in the plasma-cells of the connective tissue which are abundant; in Limax, in the liver-cells, the connective tissue being scanty. For the distribution and mode of occurrence of glycogen in the other tissues of Gastropoda, as well as in the tissues of other animals, the very valuable paper of Barfurth's, A. M. A. xxv. 1885, should be consulted. The hermaphrodite gland or ovo-testis consists of a number of cylindrical follicles opening into a common duct. The ducts from the various groups of follicles unite in their turn with the chief duct of the gland, which is lined by a non-ciliated squamous or columnar epithelium. The ova and spermatozoa are formed from the lining epithelium of the follicles, but at a different time. The albumen gland is a racemose gland. It opens by a common duct, which with its branches is lined by a ciliated epithelium. The ovi-spermi-duct is lined by a ciliated epithelium, and both portions are beset with glands. The free portion of the oviduct is also ciliated, but not the vas deferens, as is the case also with the penis and flagellum. The latter organ, with the hinder part of the penis, secretes the ' capreolus,' or spermatophore, a structure formed of hardened mucus. It is more or less elongated and its edges folded so as to form a groove in which a mass of spermatozoa are lodged. In copulation it is transferred to the duct of the recepta- I i 8 DESCRIPTIONS OF PREPARA TIONS. culum seminis, and in this species it is said to undergo resolution in ten days after transference. The dart-sac, which has extremely muscular walls, contains a pointed cuticular style, with much calcareous matter in its composition. Its use appears to be a preliminary excitant to copulation, after which act it has been found within the body. A new dart is speedily formed after the loss of the old one. The multifid vesicles secrete a highly refractile fluid, which is poured out during co- pulation, and is formed by the epithelium. The duct of the receptaculum seminis has occasionally a tubular diverticulum appended to it. The latter is constantly present in many species of the genus Helix, e. g. H. aspersa, and may possess a terminal bulb like the receptaculum itself. The genital organs are richly supplied with nerves and bloodvessels. In aquatic Pulmonata the male and female apertures open separately from one another externally. Of the accessory organs present in Helix pomatia, the flagellum is absent in all the American species of the genus, and both dart-sac and multifid vesicles, common in the European species, are rare in them. A dart-sac is found in the American slug Tebennophorus ; and glands resembling the multifid vesicles are appended to the male organs in Veronicella. With these two exceptions the organs in question are confined to the genus Helix. The eggs have, like those of other Helices, of Arion and some other terrestrial Pulmonata, a hard calcareous shell inclosing a quantity of albumen and a small ovum. Those of some terrestrial (Limax), and of all aquatic Pulmonata, are devoid of a calcareous shell, and the surface of the albumen is simply hard- ened. The eggs of H. pomatia are large, nearly a quarter of an inch in size, and are laid in earth as are those of other terrestrial Pulmonata, in masses, but not united, as are those of many species of Limax and of all aquatic Pulmonata. A few terrestrial Pulmonata are viviparous (certain species of Clausilia and Pupa, Helix rupestris, Achatinella, a Vitrind). The larva has but slight traces of the typical Molluscan velum ; in that point differing from its aquatic congeners, where it is well developed, and Arion and Limax, where it is wanting. There is no operculum. A remarkable contractile pedal sinus is present. A paired provisional tubular renal organ appears to be found in the larva of all Pulmonata. Integument. Leydig, A. N. 42, 1876. Epithelium. Flemming, A. M. A. vi. 1870. Connective tissue (interstitial), Brock, Z. W. Z. xxxix. 1883. Odontophore. Mechanism. Geddes, Tr. Z. S. x. Radula, Rossler, Z. W. Z. xli. 1885. Description of various forms of dentition and nomenclature, Gray. P. Z. S. 1853 ; A. N. H. (2), xii. 1853 ; Troschel, Das Gebiss der Schnecken, Berlin, i. 1856-63; ii. parts i-vi. 1866-79. Myohaematin in odontophore muscles, MacMunn, Proc. Physiol. Soc. in Journal of Physiology, v. 1884 ; P. R. S. xxxix. Novr. 1885. Salivary glands, &c., see Nalepa (op. cit. ante, p. 112). Sugar formation. Bonardi, Boll. Sc. Pavia, Anno 5, 1883. Liver Barfurth, A.M. A. xxii. 1883 ; Id. Biol. Centralbl. iii. 1883-84; Frenzel, Biol. Centralbl. torn, cit.; Id. A. M. A. xxv. 1885 ; its physiological action, Krukenberg, Untersuchungen Physiol. Inst. Heidelberg, ii. 1882, pp. 4 and 13 ; Fredericq, Bull, de 1'Acad. Royale de Belgique (2), 46, 1878. Pigments. MacMunn, P. R. S. xxxv. 1883, and xxxviii. 1884-85 ; Krukenberg, Ver- gleich. Physiol. Studien, ii. (2) p. 63, 1882. Glycogen, Barfurth, A. M. A. xxv. 1885. Genital organs, Baudelot, A. Sc. N. (4) xix. 1863. Histology of accessory organs, Batelli, Atti Soc. Toscana Sc. Nat. (Mem.), iv. 1880 ; cf. Journal Royal Micr. Soc. EDIBLE SNAIL. (2) i. 1881, p. 435- Development, Rouzaud. C. R. 96, 1883; Jourdain, Revue Sc. Nat. Paris and Montpellier, viii. Physiology, Dubrueil, Revue des Sci. Nat., Paris and Montpellier, i. 1873; ii. 1874; cf. Jourdain in A. N. H. (4) viii. 1871. Dart of British Heliddae, Ashford, Journal of Conchology, iv. 1883. Sperm. Leydig, Untersuchungen zur Anat. und Histol. der Thiere, Bonn, x. 1883, p. 118; development of, Blomfield, Q. J. M. xxi. 1881 ; Nussbaum, A. M. A. xxiii. 1884, p. 206; cf. Von Brunn, ibid. p. 459; Platner, A. M. A. xxv. 1885. Ova and Development. Fol, A. Z. Expt. viii. 1880. 23. EDIBLE SNAIL (Helix pomatia), Dissected so as to show its nervous system. THE collar has been divided to the right of the pulmonary aperture, to the left of its columellar lobule, which is left in situ on the right side. The attachment of the mantle below the collar to the integument of the body has been divided from right to left, and again along the right margin of the pulmonary chamber. The mantle-fold or roof of the pulmonary chamber thus freed from its connections has been turned over to the left and displays the terminal portion of the intestine at its edge, the pulmonary vessels, the triangular kidney, and the pericardium, the latter opened to show the heart. The integument covering the head and neck has been divided in the middle line, and the floor of the pulmonary chamber continuous with it removed. The whole of the viscera have also been removed, leaving only the buccal mass, the nerve-collar and the columellar muscles cut short. The buccal mass is much retracted, as it is when the animal's head is drawn in. At its base are seen, the origins of the oesophagus and salivary ducts above, and the sac of the radula below. A stout black bristle has been passed under the right cerebral ganglion. From the anterior edge of this ganglion a nerve passes to the right upper tentacle which bears the eye : a similar nerve, not visible here, passes to the lower tentacle. From its posterior edge a nerve passes along the buccal mass to a small stellate ganglion, the buccal ganglion, which lies below the salivary duct and innervates the buccal mass ; a fine black bristle has been placed beneath it. There are two such ganglia, right and left, connected by a commissure below the oeso- phagus. The right cerebral is united to the left by a commissure passing across the buccal mass : it is united to the infra-oesophageal ganglion below the buccal mass by a broad band. This band contains two nerve- connectives (not visible here), one to the anterior or pedal portion, the other to the posterior or visceral (parieto-splanchnic) portion of the infra-oeso- phageal ganglion. From the posterior margin of this ganglion three nerves pass backwards to the body walls ; a fourth, which accompanies the aorta cephalica, has been removed together with that vessel. Bundles of nerves may be seen passing down from the anterior or pedal portion of. the 120 DESCRIPTIONS OF PREPARATIONS. ganglion to the foot. The cut ends of the columellar muscles are visible in a bundle below the buccal mass. These muscles are attached on the one hand to the columella of the shell, on the other to the foot, to the buccal mass, tentacles, and nerve-collar on either side. They therefore serve as retractors of these organs. There are two pairs of cephalic tentacles in nearly all terrestrial Pulmonata^ the superior pair bearing the eyes at their apex : hence the name Stylommatophora applied to this section of the order. They are hollow, and in- and e-vaginable. The exceptions are the Australian genera Janella and Aneitea, in which the eye- bearing tentacles alone are present, and the slugs Onchidium and Vaginulus, in which the tentacles are solid but extremely contractile. The aquatic Pulmonata, on the other hand, have but one pair of tentacles, solid and contractile, with the eyes placed at the inner side of their bases : hence this section of the order is termed Basommatophora. The bands of nerve-fibres uniting the various ganglia are termed ' commis- sures ' when they unite the ganglia of the same pair, e.g. the cerebral ; ' connectives ' when they unite ganglia of different pairs, e.g. cerebro-pedal connective between the cerebral and pedal ganglia of the same side. The nervous system of H. pomatia has been worked out in detail by Bohmig. Each cerebral ganglion is divisible into three regions : from one there arise the commissure to the other cerebral ganglion the various connectives, and six nerves, one to the upper tentacle, the mouth, lip, pharynx, lower tentacle, and on the right side to the penis : from the second arises a nerve to the upper tentacle and eye : from the third nothing. The two pedal ganglia are each divisible into two regions, from which 8-9 nerves pass to the foot, and from one of the ganglia branches proceed to the connective tissue surrounding the uterus. The parieto-splanchnic region consists of a right and left pleural (=com- missural) ganglion which are in union with the connectives to the cerebral and pedal ganglia, and with the remaining ganglia of the visceral system. These are a right and left visceral ganglion and a median abdominal. The right visceral ganglion gives origin to two pallial nerves inclosed in a common sheath ; the left to a single pallial nerve. The abdominal ganglion gives origin to three nerves ; one distributed to the heart, nephridium, liver and (?) sexual organs ; another to the neighbourhood of the anus, and the third to the integument. The buccal ganglia supply the pharynx, intestine and salivary glands with four nerves. Although the divisions between the pedal and the various ganglia of the visceral system are not visible externally in Helix and other terrestrial Pulmonata as they are in the aquatic Pulmonate Limnaeus, several facts may be noted which clearly show without further analysis the compound nature of the infra-oesophageal ganglion. These are (i) the distribution of the nerves given off by it ; (2) the passage of the aorta cephalica through its centre ; (3) the relation of the otocysts to the region termed pedal ; (4) the presence of double connectives on each side between it and the cerebral ganglia, which can be clearly discerned through the common connective tissue sheath. In Zonites algirus two pedal nerves run backwards in the foot nearly parallel one with another. They are connected by transverse commissures from place to place, and posteriorly they break up into a number of branches which anastomose with one another. Minute ganglia occur at the nodal points of this network, but are EDIBLE SNAIL. 1 21 not common in the course of the two main nerves. The nerves to the margins of the foot originating from the pedal nerves form a similar network with nodal ganglia. In Vaginulus, according to Semper, the pedal nerves run parallel to one another, and are provided at stated intervals with ganglia from which spring transverse commissures. The similar transverse commissures oiLimax appear from Simroth's description to be rather irregular, and they give origin to a network of fibres with nodal ganglia which unite them together. The nerves passing across the foot in Helix anastomose, but they are very irregular in arrangement ; and in Arion they break up into a fine network, and there is nothing to compare with the commissures of Limax either in the direction or in the size of the nerves ; but the networks of nerve-branches are furnished as usual with nodal ganglia both in Helix and Arion. The regular arrangements of transverse commissures in Zonites and Vaginulus recall the ladder-like structure of the pedal nervous system in Chiton, Haliotis and Fissurella among marine Gastropoda. It is not certain, however, whether or no they are strictly comparable with one another. The ganglion cells of Helix vary from 0*4 mm. in the visceral ganglia to o'i6 mm. or 0-007 mm. in the cerebral. Some of the larger cells have a con- nective tissue capsule. The majority are unipolar, but bi- and multi-polar cells are also found. The central region of the ganglia is occupied by Leydig's ' Punkt- substanz,' which is composed really of a network of very fine fibres derived from the ganglion cells. The nerves are composed of an outer sheath of vesicular connective tissue, and an inner membranous sheath continuous with internal septa dividing the nerve-fibrillae into bundles. The fibrillae and fibrillar bundles originate from the ' Punkt-substanz.' The organs of special sense are sensory epidermic cells, the tentacular ganglia, the ganglia of the oral lobes, the eyes, and the otocysts. The sensory cells are most numerous on the tentacles, oral lobes, and on the sides of the foot, where they have the form of cylinders with a fine point composed of delicate hairs, which do not project above the level of the cuticle of the ordinary epidermic cells. At the actual edge and on the sole of the foot, parts covered by mucus, the hairs are distinctly separate, and project beyond the cuticle of the surrounding cells. Hence there is a closer resemblance with the ' Pinselzellen ' of aquatic Pulmonata and other Mollusca in which the cell is terminated by a distinct head and a bundle of long projecting hairs. The nerve which enters each tentacle ends at its summit in a knob composed of nerve-fibrils and ganglion-cells, the whole surrounded by a muscular sheath derived from the retractor muscle of the tentacle. There are masses of small ganglion-cells underlying the epidermis, which is modified in the region of this terminal ganglion. The ordinary epidermic cells acquire an exceedingly well- developed cuticle vertically striated. A few goblet cells occur among them, and numerous sense-cells connected each with a fibril from a ganglion-cell. The sense- cells are. flask-shaped, and end (?) either in a single point or several hairs. They do not reach as far as the cuticle of the epidermic cells. These tentacular organs appear to be olfactory in function. As long as one or both pairs are present the snail recoils from strong-smelling liquids, e.g. turpentine. If both pairs are removed it creeps into them. The ganglionic structures in the large oral lobes or lips, and the smaller series of oral lobes are similar to the tentacular ganglia. The DESCRIPTIONS OF PREPARATIONS. overlying epidermic cells have also a very thick cuticula. A tentacular ganglion is found on the outer side near the bases of the two tentacles of Basommatophora, and an oral ganglion in the oral lobes, e. g. of Limnaeus. The eyes are situated close to the apex of the upper tentacles, a little to their inner side. The optic nerve is a branch of the tentacular nerve in the Slylom- matophora, an independent nerve in the Basommatophora. The retina is composed of a single layer of cells derived by invagination from the epidermis. The cells are of two kinds, pigmented and non-pigmented, both continuous basally with the nerve-fibrils derived from a peripheral optic ganglion. The pigmented cells are widest at their inner, the non-pigmented at their outer ends. The inner ends of the non-pigmented cells end in a flask-shaped (Stylommatophora] or a fine (Basom- matophora) visual rod. The inner ends of the pigmented cells form transparent processes. Each non-pigmented cell is surrounded by a zone of pigmented, and the visual rods are encased in the transparent processes of the latter. The retina is continuous at its anterior margin with a layer of transparent cells which lines the cornea (=pellucida). The optic cavity is filled by a structureless substance, the vitreous body (Carriere), commonly called the lens. By Hilger a vitreous body is distinguished from a lens in all the Gastropoda he examined except the Stylom- matophora, which, according to him, possess a lens but no vitreous body. It is probable that the lens, so called, is a differentiated portion of the vitreous body, and not an independently formed structure. The cornea consists of an outer layer of transparent cells with underlying connective tissue. The otocyst is small in size, and is to be found on the pedal ganglion close to the spot where the anterior bundle of pedal nerves quits it. To the naked eye it appears like a minute white dot. It consists of a connective tissue capsule nearly, but not quite, spherical, and an epithelial lining with a central cavity. The latter contains a fluid and a number of otoliths, as is usual in Gastropoda. The otoliths are more or less oval, and consist of a small quantity of an organic substance and a large quantity of calcium carbonate. The cells of the epithelium are not clearly separated from each other in the fresh state, according to Leydig. At the spot opposite to the auditory nerve (or the auditory canal) they are of larger size. All the cells are ciliated, and the cilia are short. It is doubtful whether or not long auditory or sensory hairs are present such as are found in Natantia (—Heteropoda) and in Cydas (a Lamellibranch). Leydig believes them to be present. The otoliths are in constant motion in the living animal. The otocyst is connected to the cerebral ganglion by a delicate filament. This, according to Leydig, as in some other Gastropoda, is a narrow canal lined by cells, and continuous with the cavity of the otocyst. Nerve-fibrillae are therefore not present. But if this is the case, it is difficult to see in what way auditory impulses can affect the nervous system. And, judging from the analogy of other Mollusca, auditory nerve-fibrils are probably present. The connection of the otocysts with the pedal ganglia is only accidental. Both sets of structures are enveloped by a common investment of loose connective tissue. An osphradium or olfactory organ in connection with the organ of respiration has not been detected in Helix pomatia and many of its allies. In Helix per sonata Sarasin found a nerve arising from the region of the right visceral ganglion which passes beneath the epithelium of the pulmonary chamber, the cells being more , EDIBLE SNAIL. 133 columnar where the nerve is in contact with them than elsewhere. It bends round the anterior edge of the pulmonary aperture, and at the spot where it bends swells into a small ganglion with large ganglion-cells. A nerve extends from the ganglion to the glandular cells of the epidermis (?) of the collar. The homologous nerve, but devoid of a ganglion, is present in Sucdnea amphibia, Bulimus detritus, B. decollatus and Limax dnereoniger. It is distributed to the glandular epithelium as in H.personata. The corresponding nerve is apparently present in H.pomatid1. The osphradial apparatus is well developed in the aquatic Pulmonates, Limnaeus, Planorbis, and Physa. In the first of the three, which has a dextral shell, the nerve is derived from the right visceral ganglion ; in the other two, which have sinistral shells, from the corresponding left ganglion. The nerve ends in a ganglion with large ganglion cells, in connection with a ciliated canal or depression in the pulmonary chamber above and behind its orifice. In branchiate Gastropoda the nerve is also derived from the visceral ganglion, and ends in a similar ganglion lying immediately beneath the epidermis close to the ctenidium. The epidermic cells above this ganglion are large and columnar. The apparatus is paired when the ctenidia are paired. See general account of the Class. Nervous system : of Helix pomatia and Limnaeus, Bohmig, Inaugural Disserta- tion, Leipzig, 1883 ; of terrestrial Pulmonata, Leydig, A. M. A. i. 1865 ; of aquatic Pulmonata, De Lacaze Duthiers, A. Z. Expt. i. 1872; of Mollusca, von Ihering, Vergleich. Anat. des Nervensystems, &c. der Mollusken, Leipzig, 1877. Homology of the ganglia. Spengel, Z. W. Z. xxxv. 1881. Cf. Ray Lankester, 1 Mollusca,' Encyclopaedia Brit. (ed. ix.) xvi. p. 636, and Fig. i D. Pedal nerves : of Zonites, Nalepa, SB. Akad. Wien. Ixxxvii. Abth. i. 1883, p. 282; of Limax, Helix, Arion, Simroth, Z. W. Z. xxxii. 1879, pp. 304-318; of Vaginulus, Semper, A. M. A. xiv. 1877, p. 123. Histology. See Bohmig, op. cit. supra; Vignal, A. Z. Expt. (2) ii. 1883. Cf. Haller, Marine Rhipidoglossa, ii. M. J. xi. 1886. Organs of spedal sense. Sensory epithelium. Flemming, A. M. A. v. 1869. Tentacular ganglion, &c., Id. A. M. A. vi. 1870 ; Sarasin, Arb. ZooL Zoot. Inst. Wurzburg, vi. 1883. Oral ganglion (=Semper's organ), Sarasin, op. cit. Eye, Hilger, M. J. x. 1885. Carriere, Sehorgane der Thiere, Leipzig, 1885 ; Z. A. ix. 1886. Otocyst. Leydig, A. M. A. vii. 1871. Cf. Simroth, Z. W. Z. xxvi. 1876, pp. 278-281 ; De Lacaze Duthiers, A. Z. Expt. i. 1872. Osphradium (= olfactory organ) : of Mollusca, Spengel, Z. W. Z. xxxv. 1881 ; of aquatic Pulmonata, Simroth, Z. W. Z. xxvi. 1876, p. 308; De Lacaze Duthiers, A. Z. Expt. i. 1872 (=organe nouveau d'innervation) ; of terrestrial Pulmonata, Sarasin, op. cit. supra. Regeneration of eye, &c. in Pulmonata. Carriere, Studien iiber die Regenerations- erscheinungen bei den Wirbellosen, i. Wurzburg, 1880. 1 Simroth has described (Z. A. v. 1882, and with figures in J. B. Mai. Gesellsch. x. 1883), in the slug Parmacella Olivieri, a groove with projecting edges extending from the pulmonary aperture forwards to the left, and lying in the furrow between the edge of the mantle-fold and body. Ganglion cells underlie the groove and its edges, which are supplied by the right pallial nerve, as well as by a branch from the left pallial nerve. The position of the organ external to the pulmonary chamber appears to render an homology with the osphradium impossible. Zonites has a sac opening close to the pulmonary aperture and lying in the roof of the pulmonary chamber. It is beset with gland- cells and supplied by the ' olfactory ' nerve, which is, however, devoid of a ganglion (see Nalepa, op. cit. ante, p. 239). 134 DESCRIPTIONS OF PREPARATIONS. 24. SHELL OF FRESH-WATER MUSSEL (Anodonta cygned). THE shell of Anodonta and of all Lamellibranchiata is bivalve. The two valves correspond to the two sides, right and left, of the body, and they resemble one the other. The shell is therefore equivalve. Each valve presents a short straight margin, the hinge-line, along which it is united to its fellow, and which coincides very nearly with the whole of the dorsal aspect of the animal, whilst the long curved margin coincides with its ventral aspect. The external surface is marked by concentric lines parallel with the margin and usually considered to mark distinct periods of growth — a point which cannot be regarded as certain. The areae inclosed by the lines diminish progressively in size. The smallest area corresponds to the original shell. This region of the shell is sometimes remarkably prominent and is known as the beak or umbo. If a line be drawn to the ventral margin from the centre of the umbonal region and perpendicularly to the hinge-line, it divides each valve into a smaller anterior and a larger posterior portion. Hence each valve is inequilateral. Those Lamelli- branchiata which move from place to place by means of their distensible foot, move invariably with the anterior part of the shell-valves foremost, and never in the reverse direction. The shell of the Brachiopoda is also bivalve : but it contrasts with that of Lamellibranchiata in several points : the valves are dissimilar, i. e. the shell is inequivalve : they are divisible into symmetrical right and left halves, i. e. are equilateral : and finally one valve is dorsal, the other ventral according to the usual view. At any rate they are not right and left as in Anodonta. The larger or ventral valve is perforated by a peduncle of attachment, and is uppermost in the natural position of the Brachiopod. The shell of Anodonta surpasses in size the shells of all other European fluviatile bivalves. It also shows with remarkable distinctness the three layers of which the Lamellibranch shell is usually composed — the epicu- ticula or periostracum, the prismatic layer and the layer of nacre or mother-of-pearl. The epicuticula is purely organic and is composed of conchiolin like the organic substratum of the rest of the shell. It is lamin- ated, and the ridges on the outer surface of the shell are formed by it alone. It constitutes the free border of the shell which is flexible in the natural condition. The prismatic layer is calcified and is visible as a dark border on the inner surface of the shell close to the margin. With a lens it has a shagreen-like appearance due to its division into minute polygonal spaces. The calcareous matter is in the form of more or less regular prismatic needles. The nacre covers the whole inner surface of the shell within the dark border mentioned above. It is iridescent in appearance owing to the diffraction of light by the irregular free edges of the many delicate calca- SHELL OF FRESH-WATER MUSSEL. 125 reous lamellae which enter into its composition. The calcareous lamellae alternate with organic layers. An elastic ligament unites the two valves along the hinge-line. It serves to open the shell and is antagonised by the two adductor muscles. See next preparation. The inner surface of each valve is marked by three principal muscular impressions, as they are termed, two anterior and one posterior, placed somewhat dorsally. Of the two anterior impressions, the larger is due to the attachment of the anterior adductor muscle, as well as of the anterior retractor of the foot ; while the smaller, placed nearer the free margin of the shell and more dorsally, corresponds to the protractor of the foot. The larger portion of the posterior impression gives attachment to the posterior adductor muscle, but its irregular process, extending dorsally towards the hinge-line, denotes the point of attachment of the posterior and larger retractor of the foot. A continuous line extends between the impressions of the two adductors parallel to the margin of the shell and at some little distance from it. This is the pallial line. It gives attachment to a series of muscular filaments which extend outwards into the free edge of the mantle, and are attached to the spot where the epicuticula commences. The epicuticula being flexible, contraction of these muscles brings the edges of the epicuticular layers of both valves into firm contact, and at the same time retracts somewhat the free edges of the mantle lobes. As the pallial line describes an even curve throughout its whole extent, Anodonta is said to be integripalliate. Muscles pass up from the foot and are attached to the ridge which borders the ligament internally, and in shells with hinge-teeth (infra] to the teeth, which are merely developments of the ridge present in A nodonta. A muscle also passes across from the ridge of one to the ridge of the other valve. And small muscular bundles are attached also to the inner part of the ligament. The valves of the shell are inequivalve in the Ostreidae^ one valve being smaller than the other. Each valve is nearly equilateral in some of the Pectens, e.g. Pectunculus. It is a rare thing for the anterior portion of the valve to be larger than the posterior. Anodonta^ like some of its immediate congeners and some of the oldest Lamellibranchiata, geologically speaking, is devoid of ' hinge-teeth ' on the inner aspect of the valves. The hinge-teeth consist of ' cardinal-teeth ' placed below the umbones : of ' anterior lateral teeth ' in front of the ' cardinal ' teeth and 'posterior lateral' behind them, and below the ligament. Unio, which belongs to the same family as Anodonta, and, like it, inhabits the freshwaters, possesses anterior and posterior lateral but no cardinal-teeth. The presence of these inter- locking teeth gives a dissimilar appearance to the two valves when viewed from within. The variations in the character and arrangement of the hinge-teeth have furnished recently a basis for the classification of Lamellibranchiata (Neumayr). 126 DESCRIPTIONS OF PREPARATIONS. In those Lamellibranchiata which possess siphons, the pallial line is incurved beneath the impression of the posterior adductor muscles, forming a bay or sinus. Such shells are said to be sinupalliate. The epicuticula varies in thickness in different Lamellibranchiata, and it is sometimes complicated in structure as in Mytilus edulis. It is rather thin in Ano- donta. The prismatic layer consists of more than one layer of calcareous prisms, which vary in size and external shape. The different layers of prisms are held together by remains of the organic substance (conchiolin) in which the calcareous prisms are deposited. At their first appearance these bodies are, in Anodonta at least, more or less rounded and irregular in shape, and at some distance apart. They increase in size, and at the same time new prisms appear between those first formed. In some instances, e. g. Cyprina islandica, the prismatic layer consists chiefly of a mass of granular calcareous matter : in other instances it is very distinctly lamellate, like the nacre, e. g. Astarte borealis. It may closely resemble the Gastropod shell, and consist of vertical laminae, e. g. Cardium. The nacre consists of alternating lamellae of conchiolin, and of conchiolin containing calcareous deposits, which appear exactly in the same manner as the prisms of the prismatic layer. In the region of the various muscular impressions it is ' transparent,' and here the muscular fibres are directly continuous with the shell-substance, and not separated from it by an epidermis. According to F. Mttller, the inner surface of the nacre is covered by a soft layer. The Cyclas cornea of the freshwaters differs from other Lamellibranchiata in the structure of its shell. There is no prismatic layer, and the organic portion of the nacreous layer is reticulate. Short wide unbranched canals extend into the sub- stance of the shell and are lined apparently by extensions of the mantle. In this point they differ from the branching canals so often found in the substance of the shell in other Lamellibranchiata, which, according to Kolliker, are due to algal parasites. Ehrenbaum, however, who mentions them in various genera, says nothing as to their contents. The structure of the calcareous parts of the Lamellibranch shell does not appear to be by any means so uniform as is often supposed. The ligament in Anodonta consists of an outer and inner part. The outer is laminated and passes gradually into the epicuticula. The inner is striated radially, and consists of radial fibre composed of two different alternating sub- stances refracting light differently. Hence in cross sections this inner part appears to be laminated like the outer : but the mode in which it splits proves its real radial structure. It must probably be regarded as continuous with the nacreous layer. Where it borders the nacreous layer internally there is a distinct nacreous ridge, the homologue in Anodonta of the region which is produced into teeth in Unto, &c., and, like the teeth, giving attachment to muscles ascending from the foot (F. Miiller). The border of the ligament itself is connected with numerous isolated bundles of muscles ; and just as the ' transparent ' portion of the nacre in the muscular impressions is continuous with muscle fibres, so here the fibres of the inner part of the ligament are continuous with muscle fibres. The relations of the ligament to the shell-valves show that, strictly speaking, the valves ought to be regarded as parts of a continuous structure. The dorsal region of this structure does not undergo calcification, or only to a very slight extent, inas- SHELL OF FRESH-WATER MUSSEL. 127 much as the economy of the animal requires that it should remain flexible. It is an adaptation of an originally univalve shell. The whole shell is generally regarded as a cuticular secretion, the cells near the edge of the mantle forming the epicuticula ; those of the part a little remote, the prismatic layer ; and those of the general surface the nacre. Tullberg appears to take the view that the organic part of the shell is produced by a fibrillation of the cells. F. Miiller believes that it increases by intus-susception, and states that in Anodonta the surface of the mantle is separated from the shell over a large space extending from the pallial line and adductor muscles as far as the attachment of the muscles to the ridge bordering the ligament (supra). Judging from analogy it is more reasonable to class the shell as a cuticular formation. This conclusion is borne out by Osborn's experiments on the Oyster. He studied the formation of shell by snipping away portions of the already formed shell, and placing on the ex- posed surface of the mantle a disc of thin glass such as is used in microscopic work. He found that a gummy film was formed on the surface of the glass by the columnar cells of the mantle, which hardened in forty-eight hours into a tough leathery membrane. Crystals of lime, for the most part scaly, appeared in this membrane, which became stony in six days. It is probable that the gummy films formed by the mantle-cells are charged with lime carbonate, which eventually crystallises, its crystalline form being modified by the organic matter of the films, as is the case with crystalline products formed in the presence of colloid matter. The calcareous substance of the shell is chiefly composed of Calcium car- bonate. Traces of Calcium phosphate, silica, alumina are sometimes found. The carbonate of lime is sometimes in the form of Calc-spar, e. g. Ostrea, Pecten, and such shells retain their integrity when fossilised. It is sometimes in the form of Arragonite, in the nacreous layer only, e.g. Pinna, or in the prismatic also, as in a very large number (Sorby). The Arragonite usually dissolves away when the shell is fossilised, and then either the inner layer only or both layers are lost as the case may be, leaving a stony nucleus or cast. The thickness of the shell does not depend upon the amount of lime in the waters in which the animal dwells, but rather on the workings of its tissues, modified by surrounding influences, whether chemical or non-chemical. This may be readily seen by a comparison of the dense shell of a Pearl Mussel ( Unio mar- garitifer), from the mountain-streams of Westmoreland, with the thin shell of Anodonta from Oxford waters so much richer in lime. Structure of the shell. Ehrenbaum, Z. W. Z. xli. 1884; F. Muller, Z. A. viii. 1885 ; in Anodonta, F. Muller in Schneider's Zoologische Beitrage, Breslau, i. pt. 3, 1885 ; cf. Bronn, Klass. und Ordnungen des Thierreichs, iii. i, p. 330 : and Sorby, Presidential Address, Geological Society's Journal, xxxv. 1879 ; of Cydas cornea, Leydig, Archiv f. Anat. und Physiol. 1855, and F. Muller, op. cit. supra. Hinge teeth. Neumayr, SB. Akad. Wien. Ixxxviii. Abth. i. 1883. Formation of shell in Oyster. Osborn, Biological Studies from the Laboratory of Johns Hopkins University, ii. 1882 ; cf. on Embryo Oyster, Ryder in Bull. U. S. Fish Commission, ii. 1883. p. 383. On l Molecular Coalescence' and on the effect of Colloids upon the form of Inorganic Matter, Ord, Q. J. M. xii. 1872, and St. Thomas's Hospital Reports (N. S.), ii. 1871. 128 DESCRIPTIONS OF PREP A RA TIONS, Formation of Pearls. Pagenstecher, Z. W. Z. ix. 1858, von Hessling, Die Perlmuscheln und ihre Perlen, Leipzig 1859, and in Z. W. Z. ix. 1858, p. 453. Composition of she!! in relation to lime in water, &c. Voit, Z. W. Z. x. 1859-60. Vegetable parasites in shell, &c. Kolliker, Z. W. Z. x. 1859-60. 25. FRESH-WATER MUSSEL (Anodonta cygnea), Removed from its shell and suspended so as to show the general external features of the animal. THE animal is suspended by the anterior adductor muscle, and the two folds (right and left) of the mantle have been turned back and fastened over the dorsal area. The mantle-folds are free throughout their whole extent, but they are united indirectly at the posterior end behind the foot by the attachment of the branchiae. The mantle cavity is in this way divided into two chambers, an inferior large infra-branchial chamber, and a superior small supra-branchial chamber. In the living animal the ventral edges of the mantle-folds are applied more or less closely to one another, even when the foot is protruded, and water finds its entrance to the infra-branchial chamber through the inferior siphonal notch. The edges of this notch are covered with tentacles, and hence the portion of the mantle in question is readily recognised. It is posterior, and lies immediately below the attachment of the branchiae. The supra-branchial chamber opens above the attachment of the branchiae by the superior siphonal notch, which has smooth non-tentaculate edges. A white bristle is passed into it in this specimen. The two notches in many Lamelli- branchiata are prolonged, the inferior into the inhalent, the superior into the exhalent, siphon. The two chambers, however, in the Anodon com- municate one with the other, not only indirectly through the cavities of the branchiae, but directly also along the posterior part of the base of the visceral mass, where, as may be seen in this specimen, the innermost branchial lamella has a free edge. In some Lamellibranchiata with an aborted foot this gap does not exist, and consequently the two chambers do not communicate directly. The free edges of the mantle folds are thickened, and correspond to the collar in the Snail. Two main lips, best developed posteriorly, run along the free edge, and they inclose two somewhat smaller ridges not always discernible. The outermost ridge is prolonged round the superior siphonal notch, and unites at some distance from it on the dorsal surface with its fellow to form the dorsal ridge or raphe. The foot projects in the middle line. It is continuous with the visceral mass, which contains coils of the digestive canal, liver (=hepato-pancreas), and generative organs, and is much dilated. The foot proper is purely muscular, and may be distinguished by its yellow tint and comparative FRESH-WATER MUSSEL. 129 thinness. At its anterior edge a black bristle has been passed into the mouth. A ridge or lip above the mouth, and another below it, are prolonged respectively into the right and left pairs of labial tentacles, which project like wings in this specimen. Behind these tentacles, at the sides of the visceral mass, and between it and the mantle-folds, are the branchiae or gills. There are two gills or ctenidia on each side. Each gill is composed of two lamellae, an outer and an inner. The inner lamella of the inner gill is attached most anteriorly to the side of the visceral mass above the outer tentacle. It is then free for a part of its course, but posteriorly unites with its fellow, as may be seen in this specimen. The junction of these two inner lamellae inter se separates the infra- from the supra-branchial chamber, and is the cause in part of the posterior indirect union of the mantle-folds (see also Preparation 26). The outer lamella of the outer gill is fused in its whole extent to the mantle, and it assists where the ctenidial axis becomes free posteriorly in dividing the supra- and infra-branchial chambers, and in causing the indirect union of the mantle-folds. The internal surface of the mantle is covered with ciliated epithelium, and the plasma-cells (p. 115) of its connective tissue contain glycogen. In many forms its ventral edges coalesce or ' concresce ' leaving an aperture for the foot ; or the process may be carried even further when that organ is aborted, e. g. in Asper- gillum, and then the only entrance to the mantle cavity is through the inhalent siphon. When an Anodonta is removed from its shell, there is seen (best in a fresh specimen) a reddish streak running from near the anterior to near the pos- terior adductor. This streak is the red-brown organ of Keber, the pericardial gland of Grobben. According to the latter, it consists of a series of caeca com- municating with the pericardial cavity and lined by a continuation of its epithe- lium. The homologous structure in Cephalopoda forms an appendix to the branchial heart, is similarly formed, and a portion of its epithelium, that in the peripheral caeca, is excretory in structure. The foot is in some Lamellibranchiata aborted, e. g. Oysters. In others it is very small, and its shape varies much within the limits of the class. The ventral edge has, according to Griesbach, in the Anodonta three pores of fair size, one placed anteriorly, two somewhat posteriorly. He states that water finds its way through them to the vascular lacunae, a fact disputed by other authorities. Similar pores exist in other Lamellibranchiata and Mollusca (?). Water has been sup- posed to find its way into the blood-system in one of three ways — through the nephridium into the pericardium, through special pori aquiferi, and through inter- cellular passages between the ectoderm cells or epidermis. But the great dis- tension of the foot in Anodonta^ when protruded from the shell for the purposes of locomotion, is due apparently to the action of a circular muscle surrounding the vein which conveys the blood from the foot on each side of the body to the median infra-cardiac sinus. By this means the return of the blood is prevented whilst the heart at the same time continues to drive blood into the foot. Certain K 130 DESCRIPTIONS OF PREPARATIONS. regions of the mantle appear to act as blood-reservoirs from which a store of blood may be drawn under these circumstances (Fleischmann). It may be added that in Solen ( Ceratisoleri) legumen, where me of the blood-corpuscles are tinged with haemoglobin, these corpuscles do not escape from the blood even when the animal is greatly irritated and consequently strongly contracted. Any direct pas- sage of water from without into the blood, or escape of blood, must consequently be regarded as an extremely doubtful occurrence. In some specimens of An- odonta there exists a small pit on the posterior margin of the foot. This, accord- ing to Carriere, is a remnant of the byssus gland, which is so well developed in Dreissena among fresh-water Bivalves. The young Anodonta (= Glochidium) has a byssus filament, but the gland which secretes it disappears, and the true byssus gland appears later. The filament referred to is adhesive, and clings to anything which it touches. The labial tentacles are vascular, richly supplied with nerves, and ciliated. Their opposed surfaces are covered with fine parallel ridges. The upper tentacles appear to rise in part from the mantle, the lower from the sides of the visceral mass, but the two tentacles of the same side undergo partial concrescence of their surfaces, and the furrow between them does not extend back between the mantle and visceral mass. In some instances they are of very large size. Embryology does not favour an homology with gills. Professor Ray Lankester however has suggested that they with the gills are homologues of the prae- and post-oral ciliated bands of the Echinoid and Ophiurid larva, Pluteus, or the Tornaria larva of Balanoglossus. The gills require careful examination. In Anodonta they appear to be pro- foundly modified from their original structure. Each gill consists of two lamellae, an outer and an inner. The inner lamella of the outer gill and the outer lamella of the inner gill arise close together and along the line between them run the afferent and efferent blood-vessels. This line represents the original (ctenidial) axis of the gill. Posteriorly the axis is free for a short distance, but anteriorly it is part of the under side of the organ of Bojanus and the side of the body. The space between the two lamellae, i. e. outer and inner, of each gill, is the ' interlamellar ' space, and examination shows that it is crossed by numerous ' interlamellar ' junctions. If the surfaces of the lamellae be regarded attentively, they are seen to be perforated by many series of apertures. Hence the lamellae resemble a fenestrated membrane, but may be regarded as composed of a number of parallel vertical filaments united from place to place by ' interfilamentar ' junctions. The latter is the view justified both by a comparison with certain other forms as well as by embryology. Nucula and Yoldia (Arcacea) have each ctenidial axis bearing two series, an outer and inner, of gill-filaments which are lamellate in shape. In Mytilus and other bivalves they are filamentous, and the filaments of the outer series are folded on themselves, the folded part being external, while the inner filaments are similarly folded, but the folded part is internal. The filaments are also united laterally in Mytilus, Area, &c., by peculiar long cilia into a lamella, the component filaments of which are easily separated. Solid interlamellar junctions are sparingly developed. Further steps in complication are, the development of tubular interfilamentar junctions, e. g. in Dreissena, the union of the reflected portions of the filaments to neighbouring parts and inter se, and the great development of interlamellar junctions. The FRESH-WATER MUSSEL. 131 individual filaments, originally tubular, become nearly solid at the same time by the development of rods of a condensed gelatinous tissue in their interior. The vascular channels then run chiefly in the junctions. These changes are carried to a great extent in Anodonta. The surfaces of the filaments are clothed with cilia. The ciliated cells are distinguishable into three sets : frontal, with medium-sized cilia ; lateral-frontal, forming a single row of large cells with very long cilia ; and lateral with much the shortest cilia. The vascular channels in the gills have been stated by Kollmann to posses an endothelium. This is certainly the case in many forms, e. g. Area ; but Bonnet has failed to prove the fact in Anodonta. The vascular channels have a very complex arrangement in Anodonta, and have been fully described with figures by the last named author. The tissue of the branchiae in the Unionacea (Unio, Anodonta, &c.), contains a very large amount of lime carbonate. Hence the brittleness of these structures. From the account given above it is clear that the two gills so-called of each side in the Lamellibranch are really parts of a single gill, a highly complex and modified ctenidium. The cilia covering the surfaces of the gills cause the currents of water to flow from the outer free surfaces into the interlamellar spaces and thence outwards through the superior siphonal notch. The cilia at the margins of the lamellae are said however to cause currents towards the mouth and thus subserve ali- mentation. Lamellibranchiata, Bronn, Klass. und Ordn. des Thierreichs, iii. i. 1862 ; Do. with Anodonta as a type, Ray Lankester, ' Mollusca,' Encyclopaedia Britannica (ed. ix), xvi. 1883 ; cf. Haren Noman, Niederland. Archiv fur Zool. i. Suppl. 1881-82. Anodonta, Howes, Atlas of Practical Elementary Anatomy, London, 1885. Keber's organ. Grobben, Arb. Zool. Inst. Wien. v. 1883, p. 40 ; cf. Griesbach, A. N. 43, 1877. Absorption of water, &c. Schiemenz, Mitth. Zool. Stat, Naples, v. 1884 (with lit.) ; cf. Schiller, A. M. A. xxv. 1885. Movements of foot in Lamellibranchiata, Fleischmann, Z. W. Z. xlii. 1885 ; for valve of vein from foot, see p. 419, and Fig. 5, p. 420. Byssus gland. Carriere, Arb. Zool. Zoot. Inst., Wurzburg. v. 1882. Mode of attachment. Cattie, Tijdschr. Nederl. Dierk. Vereen, vi. 1882-85. Gills. Posner, A. M. A. xi. 1875 ; Peck, Q. J.M. xvii. 1877 ; Mitsukuri, Q. J. M. xxi. 1 88 1 ; circulation and endothelium of, Kollmann, Z. W. Z. xxvi. 1876; A. M. A. xiii. 1877, and Festschrift zur Feier des 3oo-jahrigen Bestehens der Universitat zu Wurzburg, 1882, p. 42 (with lit.) ; Bonnet, M. J. iii. 1877. Rods of gills. Bonnet, op. cit. p. 321 ; cf. Kollmann, SB. Bayer. Akad. 1876, p. 163. Development of gills, De Lacaze-Duthiers, A. Sc. N. (4) v. 1856. Ciliary-currents, Sharpey, Encyclopaedia of Anat. and Physiol. i. 1835-36, p. 621. K 132 DESCRIPTIONS OF PREPARATIONS. 26. FRESH-WATER MUSSEL (Anodonta cygnea), Dissected so as to show the viscera in situ. THE animal is suspended by the apex of its foot and fastened on its left side. The mouth is superior, the anus inferior ; the heart to the left hand, and the foot to the right. The right mantle lobe, gills, and body wall have been removed to show the stomach, coils of intestine, liver, and reproductive glands in situ. The mouth lies between the anterior adductor muscle and the base of the foot. It is fringed by the upper and lower lips, which are prolonged into the corresponding labial tentacles, not visible here. The oesophagus is wide, and leads into the stomach, which has a small cavity on its dorsal wall, and in front of the pylorus a depression, the entrance to a small diverticulum, or caecum, generally, but incorrectly, said to lodge the 'crystalline style/ This organ lies in the stomach, and reaches its full development at the approach of autumn. The liver, or hepato-pancreas, surrounds this portion of the alimentary canal. It looks in this preparation white, owing to the removal of its natural dark colouring matter by the action of the alcohol. Its ducts open into the stomach. A black bristle is passed along the first or descending segment of the intestine which passes straight towards the foot. The first coil curves round towards the dorsal or haemal aspect. The second coil reverses this direction, and curves concentrically and ventrally by the side of the first and beyond it anteriorly. The third coil passes over the first segment of the intestine on its right side, and ascends towards the dorsal aspect between the first segment and the first 'coil. It then runs along the dorsal surface, perforating the ventricle of the heart, and finally opens by an anus behind the posterior adductor muscle into the supra- branchial chamber. A ridge or 'typhlosole' runs along the posterior wall of the first or descending segment of the intestine. Another com- mences by a swollen or club-shaped end on the ventral or neural wall of the ascending segment, at the spot where it bends abruptly towards the heart, and is continued as far as the anus, as may be seen here. The ingestion of food depends solely on the currents of water set up by the ciliated epithelium lining the alimentary canal from mouth to anus. The generative gland, in this preparation yellow, is mixed up with the liver, surrounds the coils of the intestine with the exception of a small portion of the second coil, and extends between the ascending and descending segments, and even behind the latter. It must be borne in mind that both liver and genital gland are paired organs. The sexes are separate in Anodonta^ and, like other Lamellibranchiata, it has no accessory organs of generation such as occur in the Snail. The cut edges of the ventricle of the heart, which surround the FRESH-WATER MUSSEL. 133 intestine, may be seen just behind the spot where the ascending segment of the intestine bends backwards. The lower wall of the ventricle is particularly evident, and there is a well-marked space (ventricular cavity) between it and the intestine. A portion of the pericardial space is visible as a triangular cavity ventrally to,. i.e. to the right in this preparation of, the ventricle. Still more to the right is the organ of Bojanus or nephridium. The two parts of its cavity may be seen in section. Next to the pericardium is the non-glandular thin-walled duct, the walls of which are more closely apposed than in nature, leaving a mere chink. The glandular portion of the organ with its lamellae extends far forwards ; it surrounds the tendon of the posterior retractor pedis muscle, which is seen just in front of the posterior adductor muscle, and reaches to the anterior edge of this muscle and even a little to its ventral surface. Behind the posterior adductor the attachment of the left gill to the mantle is displayed. It corresponds to the junction of the tentaculate and non-tentaculate portions of the mantle which constitute the inferior or inhalent, and the superior or exhalent, siphonal notches respectively. The muscular portion of the foot is well seen in partial section. In locomotion it swells up and protrudes from the shell into the soft mud of the streams and ponds inhabited by these creatures. The rate of motion is slow, and the animals, as they move along, leave a deep furrow behind them. The two adductor muscles, as seen in this preparation, are equal in size; hence the name Isomya applied to the great group among Lamellibranchiata to which Anodonta belongs. In others, e.g. the freshwater Drcissena, the anterior ad- ductor is small compared to the posterior, hence Heteromya ; and in a third order, the Monomya, to which the Oyster and Scallop belong, the anterior adductor is absent altogether, though it is present in the Oyster when the shell and muscles first develope. The anterior adductor is formed in the mantle region above, and in front of the mouth. The posterior adductor may perhaps correspond to the columellar muscles of the Snail. The crystalline style occurs among Monomya only in Anemia, but is found in most other Lamellibranchiata, either in a special caecum or in the intestinal tract, as in the Unionacea. It is transparent in most instances. Hazay has recently made some researches on its formation and function in the Unionacea. He finds that from spring to autumn the stomach is full of a gelatinous mass in which the crystalline style is slowly differentiated, a process complete by October. The remaining jelly, apparently superfluous food-material, passes into the first section of the intestine, and by November the stomach is empty of it. In the intestine the jelly becomes a thick, compact hyaline body, which is gradually con- sumed during the winter months. The style appears to act as a stopper, closing the entrance of the pylorus, but it is itself gradually consumed ; any remains serve 134 DESCRIPTIONS OF PREPARATIONS. as the nucleus of a new style. The chemical reactions of both bodies prove their albuminoid nature. The liver in Lamellibranchiata secretes a diastatic as well as a peptic or tryptic ferment ; of the two latter, sometimes apparently only the former, as e. g. in the Oyster and Edible Mussel, or both, as in the Scallop (Pecten Jacobaeus) in which the extract is active in alkaline, neutral and acid solutions alike. The liver contains no calcareous cells, only granular and ferment cells (see pp. 116-17). The heart consists of a median, thick-walled ventricle, and a thin-walled auricle on either side. The auriculo-ventricular apertures are valved. An aorta arises from both ends of the ventricle; the anterior passes above, the posterior below the intestine. The blood spaces are for the most part lacunar, but vessels are found in the walls of the intestine, labial tentacles, and gills (?). There is a median venous sinus lying between the two nephridia. From it blood passes to the nephridia, thence to the gills, and so to the auricles. The blood is colourless, and has colourless corpuscles. The pericardium surrounds the heart. It com- municates with the glandular portion of the nephridium, two apertures at the anterior end leading one into each gland. It is thus directly connected with the exterior. The renal organs or nephridia (= organs of Bojanus) are paired. They lie ventrally to the pericardium, and are divisible into a duct or non-glandular, and a secreting or glandular portion. The former opens by a pore at the side of the body, and is covered by the inner lamella of the inner gill, where it is attached to the side of the visceral mass. It lies under the pericardium, and opens posteriorly into the glandular portion which underlies it. There is a communica- tion anteriorly between the ducts of opposite sides. Other communications have been stated to exist between the ducts and the organ of Keber. The glandular portion contains numerous lamellae, and is greenish in life. It communicates anteriorly with the pericardium (supra). The outer surface of the ducts has a cylindrical epithelium ; the inner has an epithelium, of several layers, the outer- most cells rounded and ciliated. A similar ciliated epithelium exists in the glandular portion of the organ. The superficial cells of the latter contain yellow- brown urinary concretions. Guanin not uric acid is stated to be found in them. The relations of the organs are illustrated in PI. vii. fig. 4. Kollmann has recently described ciliated funnels (nephro-stomata ?) on the lamellae of the glandular portion of the organ. Their number appears to be great, as many as 200 in an Anodonta specially examined from this point of view. The funnels, however, appear to be blind, and not to lead into any system of canals or into the blood- lacunae of the lamellae into which their blind ends project. The generative organ of each side opens by a pore in front of the open- ing of the nephridium. The organs are simple racemose glands, and alike in both sexes. They can only be distinguished by the characters of their products. The testis, however, is whitish in colour ; the ovary reddish. In the few herma- phrodite forms the gland may be divided into a distinct male and female part, e.g. Cydas\ or the two elements, male and female, may lie side by side in the same caeca as in Ostrea edulis. In the case of the Oyster the two generative products ripen at different times, and hence self-impregnation does not occur, a rule which obtains in most hermaphrodite animals. Anodonta appears to be occasionally hermaphrodite. FRESH-WATER MUSSEL. 135 In the Unionacea an external difference of shape between the male and female individual is noticeable, and this difference may be considerable. Structure and strength of adductor muscles. Coutance, De I'e'nergie et de la structure musculaire chez les Mollusques Acephales, Paris, 1878. Von Ihering, Z. W. Z. xxx. Suppl. 1878. Plateau, A. Z. Expt. (2) ii. 1884. On attachment of muscles in Anodonta. F. Muller, Schneider's Zoologische Beitrage, i. pt. 3, 1885. Action of nerves on adductors. Pawlow, Pluger's Archiv, xxvii. 1885. Crystalline style. Hazay, Malacozool. Blatter, iii. 1881, p. 196. Krukenberg, Vergleich. Physiol. Vortrage, ii. p. 63, 1882, Heidelberg. Digestive tract. Langer, Dk. Akad. Wien. viii. pt. 2, 1854. Von Hessling, Die Perlmuscheln, &c., Leipzig, 1859, p. 266. Liver, Frenzel. A. M. A. xxv. 1885 ; physiology of liver, Krukenberg, Untersuch. aus dem Physiol. Inst. Heidel- berg, ii. 1882, p. 402. Circulatory system. Langer, op. cit. supra. Sabatier, A. Sc. N. (6) v. 1877, P- 35- Nephridium. Griesbach, A. N. 43, 1877. De Lacaze-Duthier, A. Sc. N. (4) iv. 1855. Ciliated funnels or nephrostomata. Kollman, Festschrift zur Feier des 3oo-jahrigen Bestehens der Universitat zu Wurzburg, 1882, Basle, p. 36. Urinary products. Krukenberg, Vergleich. Physiol. Studien, i. 2, p. 14, et seq. Differences between male and female shells. Bronn, Klass. &c., iii. i. p. 406. Hazay, Malacozool. Blatter, iii. 1881, p, 170. 27. FRESH-WATER MUSSEL (Anodonta cygnea), Dissected to show the nervous system and the route along which the ova pass from the genera- tive gland into the interlamellar cavity of the external gills, where, as in the pouch of a marsupial mammal, they are lodged, and go through certain stages of their development. PART of the foot has been removed on the animal's left side to show the pedal ganglia in situ ; the union of the inner lamellae of the inner gills behind the foot has been divided, and the glandular portion of the nephridium opened to show the nerve cord uniting the left cerebro-pleural to the left visceral ganglion in its entire length. The left cerebro-pleural ganglion is seen lying upon the tendon of the retractor pedis anterior muscle, and just anteriorly to that of the protractor pedis muscle. It is connected to the ganglion of the same name on the right side by a commissure passing above the mouth, under which a slip of blue paper has been placed. It is united by a connective which passes obliquely to the left hand with the pedal ganglion of its own side. The right and left pedal ganglia are united closely inter se. They lie within the periphery of the ' visceral mass/ a division of the body which in these Molluscs is less sharply marked off from the ' foot ' proper than in many other members of the phylum. Four or five delicate nerves may be seen passing off from the ganglion into the muscular portion of the foot. The auditory vesicle may be found appended close to the line limiting the 136 DESCRIPTIONS OF PREPARATIONS. viscera from their muscular envelope. Its nerve in Anodonta is derived from the cerebro-pedal connective (infra}. It is not seen here. A second connective passes backwards (downwards here) to the visceral ganglion, which lies upon the inferior surface of the posterior adductor muscle. It passes, first, between the fibres of the anterior retractor and of the protractor pedis muscles ; and secondly, after skirting the inner edge of the orifice of the reproductive gland, through the glandular portion of the nephridium externally to the common tendon of the posterior retractor pedis muscles. The corresponding connective of the other ganglion, i. e. that of the right side, comes into view in front of as well as behind this last mentioned tendon. A slip of blue paper has been passed under both cords just before they enter their respective halves of the ganglion. The visceral ganglion is really paired, but its two parts are closely united in the median line. It gives off posteriorly to the right and left a stout pallial nerve which skirts the mantle. Similar nerves are given off by the cerebro-pleural ganglia, but are not seen in this preparation. Both sets of nerves unite in a circumpallial plexus with ganglia here and there. Other nerves (not visible) pass to the anus which lies in the median line behind the posterior adductor muscle, and to this muscle itself. Anteriorly the ganglion gives off two stout nerves right and left to the gills. These nerves are beset in reality with ganglion cells, and are in relation with a modified epithelium. The whole represents the osphradium or olfactory apparatus of Spengel. There are no ganglia in a Lamelli- branch corresponding to the buccal ganglia of the Snail ; but ganglia may be developed along the course of the pallial nerves, and on the siphonal nerves in those genera where siphons are well developed. The lamellae of the glandular portion of the nephridium, the inter- lamellar space of the outer left gill, its interlamellar and interfilamentar junctions, and a large bloodvessel running between the outer and inner gills at their base, are all well seen (see ante> pp. 130-31). The ova must be extruded from the generative organ in part by the contraction of the foot compressing the visceral mass. They escape from the generative orifice where the inner lamella of the inner gill is attached anteriorly to the visceral mass, and thence pass on probably as follows. The free portion of the inner gill lamella is converted into a canal by the apposition of the visceral mass to its edge. Behind the foot the ova pass between the united inner lamellae of the inner gills of opposite sides below and the organ of Bojanus above, into the cloaca. This space is small in the Unionacea relatively to their ovaries : it must fill rapidly with ova under pressure, and the shell being closed, there is no other path for them to take but the one which leads into the interlamellar space of the outer gill. This space is open to the cloaca behind the limits marked by the osphradia. Spermatozoa are sometimes found free in FRESH-WATER MUSSEL. 137 the interlamellar spaces of the gills, and as the animals are usually dioecious, it is probable that they are drawn in by the currents of water inhaled by the female, and that the ova are impregnated after their ex- trusion. The ova may be found in great abundance during the autumn and winter months. They are nourished by a substance formed by the epithelium of the spongy interlamellar junctions, and develope into the Glochidium. This Glochidium is eventually set free from the parent. It possesses a shell triangular in outline with an incurved tooth in the centre of the free edge of the valve. The margin of each lobe of the embryonal mantle has four sense organs, and a long embryonal byssus filament protrudes from between the valves, by means of which the young animal attaches itself when it quits the egg-membrane. It eventually fixes itself to the fins, tail, &c., of fish, e. g. Stickleback, by means of its valve-teeth. The epidermic cells of its host grow round and enclose it, and it then undergoes a metamorphosis. The permanent mantle is formed, and the rudimentary byssus gland, homologous with the byssus gland, e. g. of Pinna, appears. The embryonal byssus gland of the Glochidium is not homologous with the permanent gland as usually supposed. The nerve ganglia and the otocysts are derived from the ectoderm. The single adductor of the Glochidium disappears ; and the two adductors of the adult are new formations, as is also the shell of the adult. According to Spengel, the supra-oesophageal ganglia of the Lamellibranch represent the cerebral plus the pleural ganglia, i. e. portions of the parieto-splanchnic ganglia so-called in the Snail, and independent ganglia in Limnaeus and many other Gastropoda. It is a noteworthy point that in Lamellibranchiata there is no connective between the pedal ganglia and the ganglia here called visceral. Supposing that the latter ganglia were the homologues of the pleural and visceral ganglia of e. g. Limnaeus, as ordinarily maintained, the absence of such a connective would be an abnormality. It would not be, on the supposition that the pleural ganglia are fused to the cerebral. The homology of the visceral ganglia with the ganglia of the same name in Gastropoda, is also probable from their connection with an osphradial apparatus. As to organs of special sense. ' Certain of the epidermic cells are modified into tactile cells, continuous basally with nerve filaments and furnished at their outer free ends with a bundle of fine projecting tactile hairs or setae. These tactile cells are most plentiful on the papillae of the mantle edge, especially in its siphonal region. They are more sparingly present round the edge of the cloaca, on the edge of the fore part of the mantle, on the labial tentacles, inner surface of mantle, and on the foot. Eyes are not present in Anodonta at any period of its existence, whether larval or adult. Many larval Lamellibranchiata possess larval eyes at the base of the velum close to the oesophagus. For the eyes of adult Lamellibranchiata, see general account of the Class. An otocyst or auditory organ lies near to each pedal ganglion. It is sur- 138 DESCRIPTIONS OF PREPARATIONS. rounded by plasma-cells (p. 115), and consists of a fibrillated connective tissue coat, within which is a nervous layer (?) formed by the auditory nerve, and most internally a layer of ciliated epithelium. Whether special auditory hairs are present, as in Cyclas and some Gastropoda, is uncertain. The cavity of the vesicle contains a fluid in which floats a spherical calcareous otolith, single as in all Lamellibranchiata. The nerve is derived from the cerebro-pedal connective, according to Simroth, not from the pedal ganglion, as ordinarily stated. See Z. W. Z. xxvi. p. 270, PL xvii. fig. 56; and for otocyst, PL xviii. figs. 62 and 68. It is not certain if the auditory nerve is similarly derived in other Bivalves. It is in other Mollusca. The only specimens of Anodonta at my command were not sufficiently fresh to enable me to decide whether or no the nerve-supply is invariably derived as Simroth describes. It is possible that in some cases the nerve passes through the pedal ganglion, but without being derived from it. Nervous system. Duvernoy, Memoires de PAcade'mie des Sciences de 1'Institut, xxiv. 1854, p. 87 (with plates referred to). Histology. Vignal, A. Z. Expt. (2) i. 1883. Homologies of, and osphradium. Spengel, Z. W. Z. xxxv. 1881, P- 373- Tactile cells. Flemming, A. M. A. v. 1869; vi. 1870, p. 453; cf. ibid, xxiii. 1883. Otolithic vesicle. Ley dig, Lehrbuch der Histologie, 1857, p. 278, with references, p. 283. Simroth, Z. W. Z. xxvi. 1876. Generative organs in Lamellibranchiata. De Lacaze-Duthiers, A. Sc. N. (4) ii. 1854. Of Mytilus. M'Intosh, A. N. H. (5) xv. 1885. Occasional herma- phroditism in Anodonta. De Lacaze-Duthiers, A. Sc. N (4) iv. 1855. Passage of ova to gills. Von Baer, Meckel's Archiv. (Archiv. f. Anat. und Physiol.), 1830. Possible passage of ova from one Mussel to another. Von Hessling, Z. W. Z. x. 1859-60, p. 358. The different parts that act as Marsupia. Bronn, Klass. &c. iii. i, p. 442. Development with literature. Balfour, Comparative Embryology, i. p. 220. Schierholz, Zur Entwickelungsgeschichte der Teich- und Fluss-muschel, Berlin, 1878; cf. Id. Z. W. Z. xxx. 1878. Post-embryonal development of Najaden (Anodonta]. Schmidt, A. N. 51. Cyclas cornea. Ziegler, Z. W. Z. xli. 1885. 28. COMMON COCKROACH (Periplaneta orientalis), FEMALE, Dissected to show its digestive, renal, nervous, and reproductive systems, AMONG external features characteristic of the class Insecta, the head with the antennae, the three pairs of jointed thoracic limbs, and the seg- mented abdomen may be noted. The short tegmen or wing-cover of the female of this species is visible above the second limb on the right side, and posteriorly, on the same side also, at the extremity of the abdomen, one of the two short jointed cerci anales found in many Orthoptera, &c. The dorsal body walls or terga, and the fat body which abounds be- tween the viscera, especially in the abdomen of these insects, even in their COMMON COCKROACH. 139 adult state, have been removed, and the digestive tract fastened out upon the left side of the body. The digestive tract is divisible into three regions, which correspond to the stomodaeum, archenteron, and proctodaeum of the embryo. The first of these includes oesophagus, crop and gizzard, the second, the chylific stomach and caeca, and the third, the intestine and rectum. The narrow oesophagus expands directly into the crop, which occupies about three- fourths of the entire length of the body, and is distended with food. The digestive tract as a whole, however, is little more than twice as long as the body, a comparative shortness compensated partly by the character of the food, and partly by the large quantities devoured. A muscular subconical gizzard follows the crop. This organ is not developed in the larvae of In- secta with a perfect metamorphosis, e. g. Colooptero, in those species which possess it when adult, but is developed in larval Orthoptera as well as Odonata. The posterior end of the gizzard is elongated and projects into the chylific stomach. Eight ' pyloric ' caeca arranged in a whorl mark the commencement of this region, and a very much larger number of long and slender Malpighian or renal tubes its termination. Pyloric caeca are found in most Orthoptera, and in the Plecoptera. Their walls are glandular, and the size of the caeca varies with their state of distension. The intestine consists of a short, narrow ' ileum,' and a long, somewhat dilated colon. The ileum is not clearly visible in this preparation. The colon is ridged and beaded owing to the contraction of its muscular coats. It ends in a rectum, which shows six longitudinal ridges alternating with furrows. The lobed labial salivary glands are to be seen on either side of the anterior end of the crop : and on the right side in this preparation the right salivary receptacle, a .pellucid bladder reaching a little further back than the gland. An azygos nervus recurrens, derived from the 'ganglion impar' or ' frontale ' of the stomatogastric system, lies on the dorsal wall of the crop, and ends in a triangular ganglion placed a little way in front of the middle point of its length. From this ganglion a nerve may be traced passing down the sides of the crop to the gizzard. The paired ganglia of the system are not to be seen. The six terminal ganglia of the ventral chain are visible in the abdomen. The two first are more closely apposed to each other than are any of the succeeding four. The last ganglion is more or less cordiform, larger than those which precede it, and gives off nerves to the lower por- tion of the digestive and generative tubes. Each ovary consists of eight ovarian tubes or ovarioles inserted in pairs, on the inner edge of the oviduct, one set of tubes along its ventral, the other on its dorsal margin. The tubes are beaded, owing to the swellings caused by the ova. These ova increase in size the nearer they are to the oviduct. The tips of the ovarioles are in the natural state united j 40 DESCRIPTIONS OF PREPARATIONS. by short filaments to a common ligament. This ligament is probably attached, as in other Insecta, near the heart. The two oviducts open beneath the last ganglion of the ventral chain into a short vagina. The spermatheca consists of a short peduncle terminated by two slightly curled vesicles. It opens into the vagina behind the last nerve ganglion. And opening in turn behind it are the right and left colleterial glands, often, but wrongly, termed sebaceous. They secrete the material which forms the cocoon. The Cockroach, according to Cornelius, moults seven times before it becomes adult. The first moult occurs immediately after hatching ; the second a month later ; and the remaining moults at intervals of a year. The adult stage is reached in the fifth year. The young animal differs from the adult principally by inferiority of size, by the smaller number of facets in the cornea of the eye, by the absence of wings, imperfection of the genitalia, and in this family by lightness of colour, a feature, however, in which great differences exist between adult individuals of this species. The absence of a quiescent stage and of a period of abstention from food, such as exist in Insecta with a perfect metamorphosis, is probably the reason why the fat body persists, instead of being utilised as a storehouse of force during the internal changes undergone by the organism. A Gregarine Clepsidrina blattarum is often found in the body cavity of the Cockroach, and some remarkable Flagellate Protozoa in its intestine. The body is divisible, as in all adult Insecta, into a head, thorax, and ab- domen. The head is broad transversely, and compressed antero-posteriorly. It is carried vertically, not horizontally as in many forms. Its dorsal surface or Epi- cranium is convex, and is marked by a Y-shaped epicranial suture, as in the Earwig. This suture is in some specimens indistinct. The branches of the Y end in a translucent spot of unknown function placed superiorly to the inner side of the articulation of the antennae. The front of the head (= clypeus) is flat and broad, and a labrum is moveably articulated to it, closing in the mouth anteriorly. The antennae are long, filiform and many jointed. The joints are beset with hair (i.e. are setose), and the basal joint is attached to a soft membrane, which closes the socket. For the minute anatomy of antennae, see p. 145, infra. Behind the antennae are the reniform compound eyes. There are no ocelli. The mouth is constructed for biting, and consists of three pairs of jaws : the mandibles, the maxillae, and the labium. Each mandible is of one piece, tri- angular, attached by two condyles to the head. Its inner edge has at the base a grinding surface or mola, and in front of the mola and at the tip strong teeth. Neither teeth nor mola are fashioned alike in the two mandibles. They appear to interlock more or less perfectly. Each maxilla is composed of (i) a basal part, the cardo, placed horizontally and articulating with the head ; (2) a stipes, which is placed vertically, and bears (3) a five-jointed palpus on its outer edge articulated to a basal piece representing the palpiger of some Insecta; (4) a hood-shaped galea in COMMON COCKROACH. 141 front, which is articulated by a distinct joint to the stipes ; and (5) a lacinia. This lacinia is attached to the stipes by an imperfect joint; it ends with two stout conical teeth, at the base of which there rises on the internal edge a finger-shaped process, terminated by three or four recurved blunt teeth. Its inner edge is beset with stout hairs. The labium closes in the mouth posteriorly. It consists (i) of a large basal sub-mentum^ to the fore-edge of which is moveably articulated (2) a mentum-, (3) of two three-jointed/^/ attached each to a basal prominence, repre- senting a palpiger borne on the external angle of the mentum ; and (4) a ligula. The ligula is divided almost completely by a median cleft, at the base of which is a small triangular piece. Each half of the ligula bears two processes articulated to it : an outer, the paraglossa or lamina externa of Gerstacker ( = galea ?), and an inner, the smaller of the two, the lamina interna of the same author (= lacinia?). A labium of this character is to be regarded as primitive. It is found in many Orthoptera, in Termes, Perla, Aeschna, and the incomplete stages of Ephemeridae, and shows clearly the origin of the labium from a pair of limbs fused medianly. The antennae are processes of the pro-cephalic lobes : while the mandibles, maxillae, and labium belong to three fused segments distinct in the embryo. The head of the insect is consequently often regarded as formed of four segments, one prae-oral and three post-oral The labrum (supra] is formed in some Insecta by the fusion of two processes. They are, however, hardly to be regarded as limb-rudiments. The antennae are not modified limbs like the remaining appendages of the head, and are compared by Balfour with the paired processes of the prostomium in the Chaetopod Polychaeta. The embryonic rudiments of the labium are extremely large, and in some Insecta are turned backwards parallel to the thoracic limbs. The epipharynx, which lies on the internal surface of the labrum, is not trace- able in Periplaneta as a distinct process. The hypopharynx (= lingua), on the contrary, is very large. It lies on the inner or oral surface of the labium, and the salivary duct opens towards the base of its posterior surface. The thorax is composed of three limb-bearing segments — pro-, meso-, and meta-thorax. A distinct neck1 intervenes between the thorax and the head. The pro-thorax has a large tergal (dorsal) plate, the pro-notum, which overlaps the head in front and the meso-thorax behind. Its sternal element is small, as are the two lateral elements, epi-merum, and episternum, which lie in front of the articulation of the limb. These parts are larger in the two segments behind. The tergal elements (meso- and meta-notum) of the meso- and meta-thorax are sub-equal in the Cockroach. In most Orthoptera, some Neuroptera, the meta- is larger than the meso-thorax, a condition the reverse of what obtains in most Insecta. The limbs increase in length progressively from before backwards. Each limb is made up of ten joints — a large coxa articulating with the thoracic ring, a small trochanter followed by a femur, a tibia, and a tarsus of six joints. The last tarsal joint is minute, and furnished with two claws, as is usual in Insecta that possess claws. The meso- and meta-notum carry wings. Those of the first pair in the male are 1 In the neck there are certain chitinous pieces, or cervical selerites, one dorsal and median, with a longitudinal depression, two ventral and transverse, and two lateral. The first and second elements are inconspicuous, the third large. They are placed obliquely. Their significance is doubt- ful. See Huxley, Anatomy of Invertebrated Animals, p. 403. 143 DESCRIPTIONS OF PREPARATIONS. stiff, semi-opaque, and coriaceous in texture. They act as wing-covers or tegmina, and extend so far as to cover the fifth abdominal somite. Those of the second pair are membranous and large : each wing has an anterior triangular area of stiff texture and a thin posterior area. In repose, the base of the posterior portion closes like a fan, and then the remainder of the wing is folded once lengthwise, the edge of the fold being internal, the anterior firmer area of the wing lying uppermost and protecting the thinner posterior area. In some Cockroaches the tip of the wing is folded transversely. The base of attachment of both pairs of wings is broad. In the female there is a pair of short tegmina, while the hind pair of wings is re- presented only by two small triangular areae of the meta-notum marked by a few ridges. Certain Cockroaches, e. g. Polyzosteria, are wingless in both sexes. The abdomen is flattened dorso-ventrally. Its outline is somewhat different in the two sexes. It is made up of a number of distinct segments or somites, com- posed each of a dorsal tergum and a ventral sternum connected at their margins by a softfleura/ membrane, hidden by the projecting free edges of tergum and sternum alike. The portions of the terga and sterna exposed to light and air are hard and dark, but the membranes which connect successive terga and sterna are colourless and pliable. The first seven terga in both sexes are well developed, the 8th and 9th very narrow, and generally hidden by the 7th ; the 10th is triangular and pro- jecting. The ist sternum is rudimentary, and represented by a narrow band at the base of the 2nd sternum. This and the succeeding sterna to the 7th inclusive are well developed in both sexes. In the male the 8th and 9th sterna are external, and the last named has articulated to its free margin a pair of unjointed setose styles, while the ioth sternum is internal, and has developed in connection with it variously shaped copulatory processes, which surround the aperture of the vas deferens. In the female the 7th sternum is very large, and its posterior extremity is cleft medianly in the adult (imago). The two halves are boat-shaped, and are connected by a distensible soft skin. They retain the egg-capsule, which the female carries about for a long time. The S^ and 9tl1 sterna are internal, and bent at an angle inter se. The 8th has the vaginal aperture. There are three pairs of processes in connection with these sterna, homologous with the parts of the oripositor in e.g. a Cricket, or of the sting of a Bee. The first pair, according to Huxley, are developed from the 8th, the second and third from the 9^ sternum. The 10^ sternum in the female appears to be obsolete. The anus opens terminally in both sexes, and lies between two triangular podical plates, probably representing an i itjl somite, as in the Dragon- fly. At the outer angles of these plates arise two many-jointed setose appendages, the cerci anales, or cercopoda, supposed by Packard to represent a pair of rudi- mentary abdominal legs. The nervous system consists of a supra-oesophageal ganglion supplying the antennae, translucent white spots, eyes and labrum, and connected by very short commissures to an infra-oesophageal ganglion supplying the mandibles, maxillae, and labium. The ventral chain consists of three thoracic and six abdominal ganglia, united by double commissures. The oesophageal commissures have under the neurilemma a continuous layer of ganglion cells, and from them spring the two roots, one on each side, of the ganglion frontale. The ventral commissures also contain ganglion cells1. The stomatogastric system consists (i) of an azygos ganglion 1 According to Nussbaum a ridge of cells is developed from the hypoblast in the embryo Peri- COMMON COCKROACH. 143 frontale, triangular in shape, lying in front of the supra-oesophageal ganglion, and giving off posteriorly a nervus recurrens, which courses along the dorsal wall of the oesophagus and crop ; (2) of a couple of paired ganglia, lying on either side of the nervus recurrens anteriorly, and connected to the under side of the supra- oesophageal ganglion and to the nervus recurrens; (3) of a triangular ganglion terminating the nervus recurrens, from which two nerves, one on either side, run obliquely down the walls of the crop, and end in the muscular walls of the gizzard. Each nerve, near its termination, has a small ganglionic enlargement. There is also a sympathetic system in connection with the ventral chain. A fine nerve springs from either the right or left commissure, connecting successive ventral ganglia, and about midway between the ganglia. It runs dorsally between the commissures, and just above the ganglion behind its place of origin it divides. Each branch swells into a. long spindle-shaped ganglion, and then joins the lateral nerve of the same side coming from the ventral ganglion. With the exception of the chylific stomach, the digestive tract is lined by a chitinous cuticula, which is for the most part beset with setae. The chitinous coat is continued into the salivary ducts and receptacles. In the ducts it is striated, as in the tracheae (infra). The finest branches of the ducts in the acini of the gland are smooth-walled, and their terminal dilatations are lined by a coat which is not chitinous. The walls of the digestive tract consist of an external membrane under- lain by an outer layer of circular, and an inner layer of longitudinal1 striated muscle-fibres. Then follows a layer of connective tissue cells, and a layer of columnar cells, which secrete the internal cuticula where it is present. The chylific stomach is lined by columnar cells, with a striated border and rounded glandular cells lying in depressions. The gizzard has on its internal surface six longitudinal projections or teeth, and behind each tooth two cushions, the first with an uneven, the second with an even surface. In the interval between two teeth are three parallel folds, and between each of these and the teeth a smaller fold. The posterior part, which projects into the chylific stomach, contains six principal folds in a line with the cushions, and between each of them a small accessory fold. The circular muscle layer is strongly developed, and according to Wilde, longitudinal fibres run from the teeth to the cushions, and on the outer surface of the posterior part, whilst radial fibres are found only in the anterior and posterior parts. The same authority states that in some Orthoptera the internal cuticula is cast off at each moult. According to Krukenberg (Untersuchungen Physiol. Inst. Heidelberg, ii. 1882, p. 26), the salivary glands secrete a purely diastatic ferment, the chylific stomach both diastatic, peptic and tryptic ferments. The ridges in the rectum consist of elongated hypodermis cells, underlain by a mass of connective tissue cells, richly supplied with tracheae. They are structures highly developed in the larvae cf Dragon-flies. These animals take in and expel water from the rectum, which thus becomes an important respiratory organ. The Malpighian vessels consist of an outer homogeneous layer, a single layer planeta, immediately beneath the ventral nerve-chain. These cells eventually ensheath the nervous structures. Nussbaum suggests that the ridge represents the chorda supraspinalis in Lepidoptera ; cf. infra, p. 160. 1 There is pparently a contradiction in Basch's paper (cited below) as to the position, &c. of these 1 yers. Cp. pp. 241, 249, 252 of his paper. 144 DESCRIPTIONS OF PREPARATIONS. of gland cells with large ramified nuclei, and internally a porous chitinoid (?) membrane. The secretion contains (i) yellow-brown globules, which if numerous make the vessels opaque; (2) clear white globules; (3) uric acid crystals. The tubes open in Periplaneta into the lower portion of the chylific stomach, but in most Insecta into the intestine. Rathke, however, states that in Blatta Germanica they arise as outgrowths from the intestine. The corpus adiposum, or fat body, is a whitish glistening tissue, arranged more or less in masses, and consists of cells containing oily drops, albuminous bodies, and in some cases uric acid. This tissue originates by proliferation from a layer of cells within the hypodermis, if the observations of Wielowicjski on Corethra are to be trusted. . Sub-hypodermic cells have been described also by Viallanes in the larval Eristalis and Musca. If the abdominal terga are removed as a connected piece, the heart and surrounding tissues may be found on its inner surface. The heart consists of a tube, divisible into an anterior aortic membranous portion, which runs forward into the thorax, and a posterior muscular portion ending blindly behind, and divided into a series of chambers by lateral apertures. Its muscles are arranged circularly or spirally, and at the apertures in a figure of oo . The apertures are valved. The chambers contract successively from behind forward, and according to Cornelius, there were eighty such contractions in a minute in a Blatta which had just undergone ecdysis. The heart is suspended to the back of the abdomen by muscular fibres. Its walls are connected to a network of elastic fibres and interposed pericardial cells, which resemble the cells of the fat body, and the whole is limited on the surface turned to the viscera by a tissue partly fibrous and partly composed of the paired alary muscles. These muscles are striated, and arranged segmentally in fan- shaped bundles. The handle of each fan is attached laterally to one of the terga ; its expanded portion is spread out below the heart, and the muscle-fibres end in tendons reticularly arranged. The whole structure, as pointed out by Graber, forms a pericardial sinus, which expands and contracts rhythmically like the heart. Numerous tracheae ramify in it and upon the heart. A very similar structure, with alate muscles, covers over the ventral nerve cord, and forms a pulsating sinus, but the contractions run from the anterior to the posterior extremity. There are nine stigmata, or respiratory apertures, two thoracic, and seven abdominal. According to Bela Deszo, the stigmata and apertures into the heart correspond numerically in Insecta^ Myriapoda, and Arachnida. This statement can, however, only be true as far as concerns the abdominal stigmata. The two thoracic stigmata are situated, one in the meso-, the other in the meta-thorax, in front of the articulation of the coxa. The abdominal stigmata are placed immedi- ately under the lateral expansions of the terga, upon conical papillae. With the exception of the first, these papillae are concealed by lateral prolongations of the sterna. The last of the series is the largest. The entrance into the trachea, which rises from each stigma, is protected (i) by hairs which cross the aperture, (2) by a special apparatus for closing the tube, which in the Orthoptera is continuous with the lips of the stigmatic aperture. These lips are prolonged inwards as two valves. A process arises on the outer, i.e. visceral surface of each of these valves. A muscle passes from one to the other process round one of the margins, and when it contracts, squeezes the valves together, and thus narrows the aperture. The details COMMON COCKROACH. 145 of the structure appear to differ somewhat in the thoracic and abdominal stigmata. In most Insecta the apparatus for closing the trachea is separate from the stigmatic lips. It is apparently absent in the Rhynchota and abdominal stigmata vi Diptera. The tracheal wall consists (i) of a layer of polygonal cells continuous with the hypodermis of the body walls ; (2) of an external supporting membrane ; and (3) of an internal chitinous coat or intima secreted by the cells. This chitinous coat, except at the ultimate terminations of the tracheae, and in certain tracheal dilata- tions, is marked by fine transverse lines, which, as usually explained, are due to a spiral thickening of the coat with intervening thin membranous portions. But Macloskie has adduced reasons for believing that the spiral thickenings are really spiral crenulations, i. e. tubular or channel-like folds open to the trachea by a slit or fissure. He points out that such an explanation is in harmony with the structure of the pseudo-tracheae in the proboscis QiMusca^ and would also account for the lengthening and shortening of the tracheae themselves during the respiratory expansions and contractions of the abdomen. The testes undergo atrophy in the adult male. In the wingless, i. e. immature, male they are to be found as numerous pyriform vesicles placed dorsally, as is usually the case with the genitalia of Insecta. They are attached by short pedicles to a common duct. The ductus ejaculatorius opens on the iotl1 sternum. It is dilated anteriorly, and to the dilated portion two glands are attached, — one, a mushroom-shaped gland composed of short caeca with viscid granular white con- tents, the other composed of dichotomous moniliform tubes, united by a common investment into a long mass overlying the last ganglion. The spermatozoa have straight rod-like heads and long flagella or tails. For the female generative organs, see description of Plate viii. and literature cited there. NOTE ON THE STRUCTURE OF ANTENNAE. Hauser has recently investigated the function and structure of antennae. He concludes upon experimental grounds (effect of strong-smelling substances; the power of finding odorous food or in certain instances the female ; according as the antennae are present or removed) that antennae possess an olfactory function. The sensory apparatus consists (i) of a conducting antennary nerve arising from the supra-oesophageal ganglia in connection with certain lobes (Bellonci, Flogel) \ (2) of a hypodermis-cell, in union basally with a nerve-filament, and terminated by a freely projecting rod, but varying in other respects in different groups ; (3) of a supporting and protective apparatus in the shape of either a groove or a chitinous cone, open round the base of the projecting sensory rod to the cavity of the antennae, and hence filled with blood plasma. The cone is generally open at its tip. The groove is in some instances (Orthoptera, Apis) closed by a superficial delicate membrane but when it is freely open, escape of liquid is impossible, by reason of the minute size of the aperture. In the Diptera a number of sensory cells are contained within one common depression. In some Hymenoptera, grooves and cones occur side by side. The organs are generally restricted to certain joints of the antennae, e.g third joint in Diptera Brachycera, terminal joints in Lepidoptera Rhopalocera Hauser points out that the number and perfection of the organs are in correlation with the habits of the insects examined: The Syrphidae (Diptera Brachycera) L - j46 DESCRIPTIONS OF PREPARATIONS. .with larvae feeding on vegetable food, &c., have but few (one to four) ; those with larvae feeding on dung have many; the Honey-bee has 14,000 to 15,000 grooves and 300 cones on one antenna, Ichneumon 3000 grooves : but in the phytophagous subsection of the Hymenoptera grooves are absent, and cones are present in com- paratively small numbers. The Lamellicorn beetles have vast numbers on the foliate expansions of the joints, e.g. male Cockchafer about 39,000, female 35,000. Hauser was unable to find any organs in the antennae viHemiptera and Neuroptera ( Chrysopa), nor in the Carabidae, but in the latter and in other Coleoptera he found identical structures on the maxillary and labial palpi. They may, however, in these situations have a gustatory rather than an olfactory function. Cones especially appear to be present in larvae, not only on the antennae, but also on the palpi. In many instances, however, the organs appear to be absent. Cone-like structures occur on the antennae of Lithobius forficatus and Julus terrestris among Myriopoda ; see also Bourne (G. C.) on Sphaerotherium, J. L. S. xix. Dec. 1885. Insecta, MacLachlan, Encyclopaedia Britannica (ed. ix.) xiii. Anatomia degli Insetti, Camerano, Turin, 1882. Die Insekten, Graber, ' Naturkrafte ' series, xx., Munich, 1877. Elementary Text-book of Entomology, W. F. Kirby, 1885. Intro- duction to Entomology, Kirby and Spence, 4 vols. (vols. iii. and iv. 1826, which con- tain orismology and anatomy, have never been reprinted). Guide to the study of In- sects, Packard (ed. 3), Salem, 1872. Modern Classification of Insecta, Westwood, 2 vols., London, 1839. Entomologist's Text-book, Idem, 1838. Orthoptera. Orth. Europaea, Fischer, Leipzig, 1853 ; see for Family Blattinae, pp. 84-88. Recherches anatomiques, etc., sur les Orthopteres, Le'on Dufour, Me'moires, etc., des savants Strangers, Acad. Roy. de France, 1841. Classification of Orthop- tera and Neuroptera, and Genealogy of Insects, Packard, American Naturalist, xvii. 1883. The Cockroach. Beitrdge zur ndheren Kentniss von Periplaneta orientalis, Cor- nelius, Elberfeld, 1853 ; cf. Hagen, Stettin. Entomol. Zeitung, xv. 1854, p. 378 ; and Huxley, Anatomy of Invertebrated Animals, 1877, p. 398. Structure, movement, and function of Insectan limbs. Dahl, A. N. 50, 1884. Adhesive organs. Simmermacher, Z. W. Z. xl. 1884; cf. Dewitz, Z. A. vii. 1884; viii. 1885 ; and Reviews by Emery and Graber in Biol. Centralbl. iv. 1884-85. Glands of feet. Dahl, A. M. A. xxv. 1885. Folding of wings. Scudder, American Naturalist, x. 1876 ; De Saussure, A. Sc. N. (5) x. 1868. Circulation in wings of Blatta, Moseley, Q. J. M. xi. 1871. Thoracic muscles. Luks, J. Z. xvi. 1883 ; Structure of muscle, von Limbeck, SB. Akad. Wien, xci. Abth. 3, 1885. Somites of head. Packard, American Naturalist, xvii. 1883. Structure of antennae, Hauser, Z. W. Z. xxxiv. 1880. Mouth-parts of Insecta. Brulle, A. Sc. N. (3), ii. 1844 ; Chatin, A. N. H. (5), xiv. 1884. Brain of Cockroach. Newton, Q. J. M. xix. 1879. Of Locust. Packard, American Naturalist, xv. 1881. Stomatogastric and sympathetic systems. Kostler, Z. W. Z. xxxix. 1883. Nervous system of Insects in general. Leydig, Vom Bau des Thierischen Korpers, Tubingen, i. 1864 and Atlas. Lobi olfactorii, Flogel, Z. A. vi. 1883, p. 539; Id. Z. W. Z. xxx. (Suppl.) 1878, described under Blatta, p. 566. Digestive tract, &c. Basch, SB. Akad.Wien, xxxiii. 1858. Salivary glands. Kupfer, LARVA OF PRIVET HAWK MOTH. 147 Festgabe zur Karl Ludwig, 1874. Gizzard of Orthoptera. Wilde, A. N. 43, 1877. Appendices pyloricae. Graber, SB. Akad. Wien, lix. Abth. i. 1869. Rectum. Chun, Abhandl. Senck. Gesellsch. x. 1875. Malpighian tubes. Schindler, Z. W. Z. xxx. 1878. Corpus adiposum. Targione Tozzetti, Bolletino della Soc. Entomol. Ital. iii. 1871; iv. 1872. Its formation in Corethra. Wielowicjski, Z. A. vi. 1883. Heart. Graber, A. M. A. ix. 1873. Ventral pulsatile sinus. Id. op. cit. xii. 1876. Blood corpuscles. Id. SB. Akad. Wien, Ixiv. Abth. i. 1871. Cf. on heart. Poletajewa, Z. A. ix. 1886. Stigmata. Krancher, Z. W. Z. xxxv. 1880; Landois and Thelen, Ibid. xvii. 1877 (for P. orientalis, see Taf. xii. fig. 12). Primary number in Insecta. Packard, American Naturalist, viii. 1874. Tracheal System of Insecta. Palme'n, Morphologic des Tracheensystems, Leipzig, 1877. Minute Anatomy of Tracheae. Macloskie, American Naturalist, xviii. 1884. Testes of Periplaneta. Rajewski in Hoffman and Schwalbe, Jahresbericht fur Anat. und Phys. 1875, p. 425. Male genital armature of same. Huxley, Anatomy of Invertebrated Animals, 1877, p. 406. Female organs, &c. see end of description of PI. viii. Histology ', Histolysis, and Histogenesis of Insecta, Viallanes, A. Sc. N. (6), xiv. 1882. Cf. Ludwig, Untersuchungen zur Anat, &c. der Thiere, Bonn, 1883, passim. Development of Blatta germanica. Rathke, Meckel's Archiv (Arch, fur Anat. und Physiol.), 1832. Cf. A Note on same. Patten, Q. J. M. xxiv. 1884. 29. LARVA OF PRIVET HAWK MOTH (Sphinx Ligustri\ THIS and the two following preparations are intended to illustrate the various points of external anatomy in which the larva, pupa, and imago of a Lepidopteron, an insect with perfect metamorphosis, differ from one another. The larva (or caterpillar) belongs to the eruciform type of Packard. It possesses a distinct head, but it has a somewhat vermiform appearance owing to the great homonomy or similarity of the remaining somites, and the fact that the segmentation of the antennae and feet, and the form of the mouth-parts are not obvious to 'the unassisted eye. The greater part of the covering of the head is made up by the two large ' parietal scales ' of Lyonet, corresponding to the epicranium of the imago. A triangular plate, the ' frontal scale ' of the same author, the representative of the clypeus, lies anteriorly between them. The parietal scales are each marked laterally by a dark stripe, at the inferior end of which may be found the six ocelli usually present in a caterpillar, as six whitish spots. The antennae and mouth-parts may be seen with the help of a lens. The antennae are three-jointed, and are situated below and internally to the ocelli. The joints can be retracted one within the other. L Z 148 DESCRIPTIONS OF PREPARATIONS. There is a pair of large toothed mandibles, but the structures corresponding to the maxillae and labium of the Cockroach are here fused into a plate which closes the mouth behind. The plate consists of a median and two lateral lobes. Each lateral lobe represents a maxilla and carries a jointed appendage, probably the maxillary palp. The median lobe, representing the labium, carries a pair of jointed appendages, the homologues apparently of labial palpi, and a central tubular projection, the spinneret, upon which opens the common duct of the two silk glands. The three first somites carry each a pair of five-jointed limbs. They represent the thoracic region of the imago, while the remaining ten somites represent its abdomen. Of these ten, the two first somites have no appendages but like the first body-somite (pro-thorax), and unlike the second and third (meso- and meta-thorax), are pierced on each side by a respiratory foramen, the spiracle or stigma. The succeeding four somites also possess spiracles and carry unjointed, sucker-shaped limbs, armed ter- minally and on the inner side with spines, and known as ' pedes spurii,' or 1 prolegs.' The two following somites have spiracles but no prolegs, and the last of the two (the eighth abdominal) carries a dorsal horn characteristic of the family Sphingidae, with the exception of a few species. The body is terminated by two somites, of which the first, representing the large ninth abdominal somite of the imago, is a narrow ring difficult to make out; while the last, the thirteenth larval somite reckoning from the head, is perforated by the anus and carries a pair of anal prolegs. The anus itself is covered by a triangular anal valve. The action of the spirit has destroyed the natural colour of the cater- pillar. In life it is of a bright green colour, with yellow spiracles, the dorsal horn black above and at the tip, yellow below, and the abdominal region ornamented by seven oblique stripes, lilac above and white below. It feeds upon the privet, ash, &c, and exists in the larval condition from the end of July to the middle or end of September. During this period, it changes its skin six times, increasing much in size after each moult, notably the last. When full grown it has a brief period of quiescence, and then becomes very restless. Its colour, especially on the dorsal surface, tends to a brownish pink, and the pulsations of the heart become very visible and rapid. At last it buries itself in the earth, penetrating to a depth of six to eight inches. Here it hollows out a smooth-walled and waterproof chamber by moistening the soil with the secretion of its silk-glands, and pressing it with its body. It lines the chamber with silk and in about three days' time throws off the caterpillar skin and appears as a pupa or chrysalis. The labrum is represented by two tubercles connected by a soft skin lying in iront of the mandibles. In Pieris it is a corneous plate hinged upon the frontal LARVA OF PRIVET HAWK MOTH. 149 scale. The mandibles of Sphinx are only slightly toothed. In the embryo the maxillae and labial limbs are separate as in other Insecta. A few Lepidopterous larvae are apodal, e. g. Micropteryx among Tineinae. The coxa and trochanter of the thoracic limbs are but slightly indicated. The tarsus is neither jointed nor clawed. In these points the limb contrasts strongly with the limb of an insect with an active pupa stage, e.g. Cockroach, which closely resembles that of the imago. In a few instances, e. g. in Stauropus Fagi, the second and third pairs of limbs are large with conspicuous joints. The prolegs vary much in number and character. In the embryo Sphinx Populi, as figured by Kowalewsky (Mem. de 1'Acad. Imp. St. Petersburg (7), xvi. 1871, PI. xii. fig. 10), there is a pair to each of the ten abdominal somites. In the Caterpillar they may be reduced to the last and the anal pairs ( Geometridae\ or to the anal pair alone (some Tineinae}. They are as a rule armed with hooks, alternately long and short, arranged either in a complete circle round the terminal disc or only on its inner side. The anal pair is produced in the Puss Moth (Cerura vinula) into a pair of long whip-like processes, each con- taining a protrusible filament, used to drive away Ichneumons. Abdominal limbs are also found in the Thysanura, in the larvae of Panorpatae, of Tenthredinidae among Hymenoptera, and of some Coleoptera (?). Many larval Tenthredinidae closely resemble caterpillars. They differ from them in the following points : the ocellus is single ; the antennae are 3-jointed and conical, but 7-jointed in the genus Lyda ; the maxillae are membranous, bilobed and furnished with 3-5 -jointed palpi ; the labium is small, fleshy, and provided with 3-jointed palpi and a spinneret. The majority have prolegs which vary from 6-8 pairs in number. These legs have no hooks, and there is a pair on the second ab- dominal somite, which never bears one in the Lepidoptera. The larvae of the autumnal brood rest in the cocoon without pupating through the winter. Such a phenomenon is rare among Lepidoptera. The duration of a caterpillar's life varies. It may extend only to a fort- night, or to three years in Cossus. The number of moults is generally three or four. In this process the old cuticle as a rule splits on the back of the thoracic somites, and the split extends to a variable degree both forwards and backwards. The three scales of the head may separate also. The cuticle of the oesophagus and in- testine appears to be cast at least in the final moult. Previous to pupation the caterpillar may suspend itself either by the anal prolegs, e.g. Vanessa, or secure itself by the anal prolegs and a band of silk round the thorax, e. g. Pieris ; or may fashion a silken cocoon with or without an admixture of foreign bodies ; or else it buries itself as do the majority of Sphingidae. A few members of this family, however, e.g. Chaerocampa Nerii, form a cocoon on or near the surface of the ground. Note the following points of internal anatomy. The nervous system comprises a supra- and an infra-oesophageal ganglion and a series of ten ventral ganglia, of which three are thoracic and seven abdominal. The tenth is often double, e. g. in Acherontia. The commissures between the three thoracic ganglia are generally divaricated. The respiratory sympathetic system consists of a nerve running on the dorsal aspect of the nerve-chain from one ganglion to the ganglion next following. In the abdominal region this nerve branches to right and left just in front of the ganglion," with which two short filaments connect the branches at their origin. The branches are the nervi transversi or respiratorii, and supply the tracheae and 1 50 DESCRIPTIONS OF PREPARATIONS. stigmata. In the thoracic region the nerve breaks up into three primary branches. The median divides right and left as in the abdominal region, and supplies the lon- gitudinal trunk which unites the prothoracic to the first abdominal stigma, as well as the tracheae it gives off. Each side branch runs backwards and unites with the ganglion, first of all giving off a lateral branch, the nervus lateralis transversus of Cattie, which joins the first nerve given off by the ganglion. The stomatogastric system consists, according to Newport, of a ganglion frontale and nervus recurrens : and of two ganglia in connection (like the paired system of the Cockroach) with the posterior aspect of the supra-oesophageal ganglion. The latter supply nerves to the tracheae of the head. The ocelli, as usual, have each a single lens. Beneath the lens lies a number of black pigment cells imbedding a crystalline body composed of several parts (Carriere). The stigmata possess a complete apparatus by which they may be closed. The tracheal stems arising from them are all connected on each side by a longitudinal trunk which in the meso- and meta-thoracic somites gives off an external branch furnished with the usual closing apparatus of a stigma. No external indications of stigmata, however, are visible in these somites in Sphinx, but Prof. Westwood possesses a dried specimen of Cossus in which they are clearly visible. W. Miiller- Blumenau has found an aquatic Lepidopterous larva (Cataclysta pyropalis) living in Brazil, in which all the stigmata are closed and respiration js carried on by filamentous branchiae, but there are two closed thoracic stigmata, which he places between the pro- and meso-thorax and between the meso- and meta-thorax. He states that he has also found them in many terrestrial larvae. The wings arise in close connection with the meso- and meta-thoracic branches. There are three pairs of cephalic stigmata in the embryo (Hatschek). The digestive tract consists of an oesophagus, and an intestine and rectum lined by a cuticle, and a chylific stomach or mesenteron. There are six Malpighian tubes with a beaded exterior which open as usual into the beginning of the intestine. Two short salivary glands open into the mouth, and a pair of serictaria or silk glands by a common duct on the spinneret *. The rudimentary genital organs lie one on either side the heart under the fifth abdominal tergum. A delicate filament may be traced from each organ round the intestine. Herold figures these filaments as extending in Pieris Brassicae, in the male to the anterior border of the ninth somite ; in the female to the anterior border of the eighth somite, and into con- nection with two oval bodies at its posterior border. An abundant fat body fills the body-cavity, or coelome. The blood is acid as in all caterpillars with one ex- ception hitherto examined, and contains amoeboid blood-corpuscles. It is green, and the colour is due to metachlorophyl (Poulton). There are two types of larval (or young) Insecta : one known as Campodei- form, from a more or less close resemblance to the genus Campodea among Thy- sanura ; the other as cruciform, of which a caterpillar may be taken as a good example. The Campodeiform larva has the typical regions of the body clearly de- fined, mouth parts adapted for biting, well developed ambulatory thoracic limbs, and frequently terminal abdominal jointed appendages. The outlines of the body 1 Poletajew (Z. A. viii. 1885) states that the silken thread is single in the Tenthredinidae, but double and twisted in Lepidoptera. Hence he disputes the statement that the ducts of the serictaria have a common outlet. LARVA OF PRIVET HAWK MOTH. 15! are even, and the series of somites follow one another without any abrupt change of shape. Such a larva is seen, more or less adapted of course, in most Ametabola and Hemimetabola. The condition of the adult insect is acquired in a graduated series of moults, whilst organs such as wings make their appearance ; and others, e.g. the genitalia, are brought to maturity. It is also seen in some Metabola, as in certain families of Coleoptera, e.g. Cicindilidae, Dytiscidae, &c. ; in some Neuroptera, e.g. Sialidae ; some Megaloptera. In these instances it is occasionally highly specialised, e.g. in the Ant-lion. It appears in others as a first larval form, subsequently modi- fied in accordance with a change of habit in the direction of the cruciform type, constituting what is known as Hypermetamorphosis. Such a change occurs in Mantispa among Panorpatae, in the Strepsiptera, and in various Meloidae among Coleoptera, e.g. Meloe, Sitaris, Hornia, Epicauta, Cantharis. In Mantispa the second larval form is only sub-eruciform : in the Meloidae there is a number of forms one after the other more and more degraded. (See Packard, American Naturalist, vii. 1883, pp. 938-944; Riley, op. id. xii. 1878, pp. 213, 282.) The eruciform type of larva is very generally found among Metabola. It may succeed, as just mentioned, a Campodeiform larva, and in its simplest shape, e. g. in Trichoptera, which may be termed sub-eruciform, it is little more than a persistent embryonic form such as is observable in the earliest stages of most Insecta. The caterpillar Qi\h.zLepidoptera, of \hePanorpatae, of Tenthredinidae amongHymenoptera, and the grubs of some Coleoptera are typical examples of the type. The head is well-defined, but the somites of the body are simple and cylindrical (homono- mous), and the animal has a vermiform aspect. Nevertheless a thoracic region with articulated limbs is distinguishable, and what is more the abdomen (except in Coleopterous forms) possesses functional abdominal limbs. Such limbs, but not functional, exist in Thysanura, and as rudiments in the embryoes of most Insecta, e.g. in Hydrophilus (Dytiscidae], which has a modified Campodeiform larva. The limbless grub of Aculeate Hymenoptera, and still more the maggot of Diptera, must be regarded as degenerate examples of this type : and where limbs are present in a Dipterous larva they are probably secondary and special developments. The larvae of all existing Insecta, and even the primitive Ametabolous order Thysanura have been modified to a greater or less extent by Natural Selection. The embryo however has a type of structure which is readily modified in the direction of one of the two larval types ; and it is not surprising to find larval forms, such as those of Trichoptera, which may be regarded as actually transitional between the two. Larvae of British Lepidoptera and their Food plants, Wilson, London, 1880. Figures in Horsefield and Moore, Catalogue of Lepidopterous Insects in East India House Museum, 2 vols., 1858-59, and in Dewitz, Jugendstadien exot. Lepi- doptera, Nova Acta, 44, 1883. Aquatic Lepidopterous larvae. Miiller-Blumenau, A. N. 50, 1884; Maurice, Bull. Scientifiques du Departement du Nord, iv. Anatomy. Cossus, Lyonet, Traite* anatomique de la Chenille qui ronge le bois du Saule, Hague, 1762. Sphinx, Newport (and also as to pupa and imago) * Insecta'; Encyclopaedia of Anat. and Phys. ii., London, 1836-39. Skin glands. Klemensiewics, Verhandl. K. K., zool. bot. Gesellschaft, Wien, xxxii. 1883. Glandular hairs. Dimmock, Psyche, iv. 1885. Muscles. Lubbock, on Pygaera bucephala. Tr. L. S. xxii. 1859, Stigmata. Krancher, Z. W. Z. xxxv. 1881. 152 DESCRIPTIONS OF PREPARATIONS. Serictaria. Helm, Z. W. Z. xxvi. 1876; Lidth van Jeude, Z. A. i. 1878; Joseph, Z. A. iii. 1880. Nervous system. Newport, Ph. Tr. cxxii. 1832; cxxiv. 1834 (includes Vanessa Urticae) ; Cattie, Z. W. Z. xxxv. 1880. Blood. E. B. Poulton, P. R. S. xxxviii. 1885. Protective coloration and markings. Poulton, op. cit. ; ibid. xl. 1886, and Trans. Entomol. Soc. 1884 and 1885 ; Weismann, Studies in Theory of Descent (translated by Meldola), London, 1880-82 ; Cameron, on Smerinthus, Trans. Ento- mol. Soc., 1880. On colour^ see also Hagen, Proc. American Acad. (2) ix. 1882 ; F. Miiller, Kosmos, xii. ; Wallace, Tropical Nature, London, 1878, pp. 158, 249. 30. PUPA OF PRIVET HAWK MOTH (Sphinx Ligustri\ THERE are two forms of quiescent pupae among Insects : one in which the antennae, mouth-parts, limbs and wings are free, the other in which they are coherent to one another and to the body. Of the first kind of pupa two varieties are distinguishable. In one the larval skin is simply thrown off. It is known as incomplete y exarate or liber a, and occurs in the Neuroptera, Coleoptera, Hymenoptera and some Diptera. In the other variety, known as coarctate, and occurring only in Strepsiptera and Diptera^ the larval skin is retained, and it either preserves the form of the larva, e. g. Strepsiptera^ Stratomyidae, or contracts into a barrel-shaped structure, e. g. Muscidae. The second form of quiescent pupa, known as obtected, larvate, or signate, is characteristic of Lepidoptera. The cohesion between the limbs, &c., as seen here, is due to the hardening of a sticky fluid which covers the surface at the moment when the caterpillar skin is thrown off. The obtected pupa is either angular, as in the majority of Lepidoptera with club-shaped antennae ( — Rkopalocera), and then often brightly tinted ; or it is conical, as in Sphinx and other Lepidoptera with the antennae fashioned after various types (—Heterocera). It is then, with rare excep- tions, dark-brown in colour. The pupa is plainly divisible into three well-marked and dissimilar regions, head, thorax, and abdomen. The head is globular and deflexed. The long antennae take origin from it laterally, and are bent backwards ventrally and towards the middle line. A lunate convexity below and in front of the base of the antennae marks the eye. A small median square piece in front is the labrum, and the angle projecting forwards from the convex surface that bears the eye, i.e. the gena, is formed under the larval mandible. It touches the outer inferior angle of the labrum. The pupal maxillae are large in size, and take origin below the labrum. Each has at its cephalic extremity a horn-like projection, the two projections uniting in the middle line. These projections usually but wrongly identified with the whole tongue, antliae, or maxillae of the imago, are absent in some Sphingidae, e. g. Acherontia, Macroglossa, and of great size in others, e.g. Sphinx Convolvuli. UNIVERSITY PUPA OF PRIVET HAWK MOTH. The maxillae expand into an angle below the eyes, and are then prolonged as narrow bands lying side by side in the median ventral line, as far as the tips of the wings. The labium is hidden by the maxillae. Between the maxillae and antennae may be seen the tibial and tarsal portions of the pro- and meso-thoracic limbs, whilst the metathoracic are completely hidden by the wings. The mesothoracic, or first pair of wings, are large, and hide the metathoracic pair completely, save at their dorsal origin. Both wings and maxillae extend to the middle of the sternal region of the fourth abdominal somite. The mesothoracic tergum, viewed dorsally, is of great size, whilst the prothoracic,- and especially the metathoracic, are much reduced. A depres- sion behind the outer angle of the prothoracic tergum leads to the spiracle. The abdomen consists of ten somites. The first and the ninth in par- ticular are small. The first spiracle is completely, the second partially hidden by the wings. The third to the seventh spiracle inclusive are large, but the eighth is denoted only by a well-marked scar. The tenth, somite is prominent. It bears a dorsal spinous projection, the cremaster^ which is differentiated within the anal valve of the caterpillar, and is covered with spines which vary much in different specimens. It is used by the pupa as a prop when it works its way up from its underground chamber previous to the emergence of the imago. A longitudinal depression, ventral to the cremaster, marks the anus ; and the ridges on either side of it (=sustentors of Riley) correspond to the anal prolegs of the caterpillar. Irregular scars, more or less evident on the ventral aspect of the third to the sixth somites, denote the position of the other prolegs of the caterpillar. In a male pupa the ring of the ninth somite is interrupted ventrally, and a depression with a more or less prominent tubercle on either side, marks the future aperture of the vas deferens. In a female, the eighth somite (as well as the ninth) is interrupted ventrally and bears a depression, the future aperture of the bursa copulatrix. These depressions appear to be constant in all Lepidoptera. When June approaches the pupa becomes restless and writhes in its chamber. It works its way up to the surface of the ground by means of the abdomen, the only part of the body which possesses the power of motion. The last abdominal somites of the moth become free first of all from the pupa-skin : this skin then becomes brittle and is fissured longi- tudinally in the dorsal region of the thorax. A split runs ventrally along the fore-edges of the wings. The moth emerges early in the day and suspends itself vertically while its wings expand and dry. In two to three hours it becomes capable of flight. The newly-formed pupa is soft : in colour creamy-white, and all its appendages contain large cavities filled by a blood-plasma which is extremely milky owing to 154 DESCRIPTIONS OF PREPARATIONS. the resolution of the fat body. A pupa preserved in alcohol at this stage retains its light colour. One in the possession of Mr. Poulton, has clearly defined though feebly coloured streaks, corresponding to the coloured streaks of the caterpillar. Under natural conditions the pupa hardens and becomes dark-brown. As this change of colour takes place underground, it cannot be due to the action of light. Swammerdam was the first to point out that the appendages are readily separable in a newly-formed pupa, or may be dissected out from under the cater- pillar skin when it is ready to be moulted. The operation is easy if the pupa or caterpillar are preserved in alcohol. In the caterpillar it will then be found that the angle of the gena already mentioned in the pupa corresponds to the mandible : that the pupal antennae and maxillae are folded upon themselves : that the wings are mere tubercles. All these parts expand and assume their proper position as the caterpillar skin is being cast off. It will also be found that the horn-like projection of the pupa consists of a right and left division, one belonging to each maxilla : that its labium consists of two back- wardly turned lobes united basally, and of great size in Pieris : that the coxa of the limbs are united to the thorax, the trochanter inconspicuous : that the femur and tibia are bent at an angle on one another, the former concealed by the latter, and that the joints of the tarsus are not differentiated. Though the first abdominal spiracle is hidden by the wings, it retains the character of an open functional spiracle. If the horn-like projection of Sphinx Ligustri is opened when the moth is nearly ready to emerge, the bases of the antlia of the imago will be found to fill it imperfectly. Each base forms a thick band or ribbon attached anteriorly to the head, lying under the outer surface of the projection, inside which it is folded back once upon itself. It then runs on into the straight median portion of the pupal maxilla. It appears to me likely that it is differentiated from a part, and not the whole of the outer wall of the pupal maxilla, but the histological details of the pro- cess are still wanting. Owing to the fact that the antennae, antliae, &c., of the imago are formed within the corresponding organs of the pupa, and are withdrawn from them leaving them empty when the imago emerges, the pupal organs have been spoken of as ' cases ' or ' thecae,' e. g. Ceratotheca, Glossotheca, &c. But it must be borne in mind that just as the change from caterpillar to pupa takes place by a moult, and the pupal organs are formed within the corresponding organs of the caterpillar, from which they differ essentially in shape and size, so it is with the change from pupa to imago. Indeed there is reason to believe that more than one moult takes place during the pupal stage. In Sphinx Ligustri and in some others (?) a thin pellicle may be raised from the inner surface of the last pupal -skin ; and Pro- fessor Westwood has drawn my attention to a passage in Curtis (British Ento- mology, Description of Plate 147), where that author records the fact that an imago of Acherontia Atropos cast off a complete and thin pellicle after emergence from the pupa-skin. The pellicle in question appears to be homologous with the thin skin cast by the sub-imago of Ephemeridae after it has taken flight from the water, having already just emerged from another skin. The pupal state of Sphinx Ligustri lasts for forty-two to forty-three weeks. During this period changes take place affecting all the internal organs. Changes in the nervous system continue for the first four weeks, but are then suspended until March. They have been worked out in this moth and in Vanessa Urticae by New- PUPA OF PRIVET HAWK MOTH. 155 port ; whilst Herold has worked them out in Pieris Brassicae, as well also as the changes undergone by the digestive system and the evolution of the male and female organs. In some pupae, e.g. of Cossus, the edges of the abdominal somites are fringed with short spines or adminicula to aid the movements of the animal. The apex of the cremaster varies much in character. The duration of the pupal state differs much. In small species it lasts only a few days. In Lepidoptera with two broods in a year, e. g. Papilio Machaon, the pupal state of the first brood lasts thirteen days, of the second from September to June. In the broods of various Pieridae, &c., the same ' differences may be observed. The pupal state of Sphinx Ligustri is occasionally ex- tended for a year beyond the normal : and this is often the case in some other genera. It is a very common thing for a larval Tenthredinidan (Hymenopterd) to delay assuming the pupal state within its cocoon, but it is an extremely rare occur- rence in Lepidoptera. It is said, however, that if the Cossus larva makes its cocoon in autumn, the caterpillar does not become a pupa till after winter has passed : if it makes it in June, it becomes a pupa at once and emerges as an imago in three to four weeks. Note. The wings make their first (i. e. outward) appearance in the pupa stage of insects with a perfect metamorphosis. In the newly-formed Lepidopterous pupa they are hollow sacs with a cavity continuous with the coelome. These sacs swell out as the larval skin is being cast. But wings appear from the first as external processes, gradually increasing in size with successive moults in insects such as the Earwig, Ephemeron, Cockroach, i.e. in Insecta Ametabola, and Hemimetabola. Dewitz (Berlin. Entomol. Zeitung, xxv. 1881) has found in very early stages of larval Trichoptera and Lepidoptera a pair of small meso- and meta-thoracic involu- tions of the hypodermis cells containing an internal chitinous lamella continuous with the cuticle. These involutions increase in size at every successive moult, and acquire a more or less perfect investment of mesodermic cells derived from the sheaths of either the tracheae or the nerves. They are evaginated previously to the last moult by the withdrawal of the internal chitinous lamella, and when the last larval skin is stripped off they appear as external sacs (supra]. The homology and first origin of wings are points of great obscurity. There are no traces of them in the most primitive Insectan order known, the Thysanura, a survival of forms existing apparently before the acquisition of wings. Wings are to be considered as secondary or acquired structures. It is tempting to regard them as modified tracheal gills, which they much resemble in structure, and which are also organs secondarily acquired. But it is impossible to suppose that all Insecta with a very limited number of exceptions are descended from ancestors which took to aquatic habits; l took? because the first Tracheata were without doubt terrestrial forms. And the only supposition that appears feasible is that respiratory structures similar to tracheal gills were of use to terrestrial Insecta living under conditions long passed away. The larval Calotermes rugosus (F. Miiller, J. Z. ix. 1875), one of the Termitidae, animals of subterranean habit, develops peculiar dorsal appendages devoid of tracheae on the pro- and meso-thorax. The pair on the prothorax dis- appears : that on the mesothorax acquires tracheae, and grows into the mesothoracic pair of wings. The metathoracic pair of wings develops in a similar manner but at a later period. It must be remembered that the oldest fossil Insecta known, even T56 DESCRIPTIONS OF PREPARATIONS. the Silurian Blatta, possess wings, so far as can be judged, of the ordinary structure. The first origin of these organs dates back therefore to a period and to conditions of which we have no record. Pupa and Imago of Cossus. Lyonet, Recherches sur Fanatomie et les nreta- morphoses de diffe'rentes especes d'Insectes. Ouvrage posthume, Paris, 1832. Changes in nervous and digestive systems^ and Evolution of reproductive organs. Herold, Entwickelungsgeschichte der Schmetterlinge, Cassel, 1815. Changes in nervous system. Newport, Ph. Tr. 1832 and 1834. In reproductive. Bessels, Z. W. Z. xvii. 1867. Formation of antennae of Imago. Dewitz, Biol. Centralbl. iii. 1882-3. Of wing, Id. Berliner. Entomol. Zeitung, xxv. 1881. Pancritius, Z. A. vii. 1884. Philosophy of Pupation. Riley, American Entomologist, iii. 1880. 31. IMAGO, MALE AND FEMALE, OF PRIVET HAWK MOTH (Sphinx Ligustri). IN the imago or Moth, Sphinx Ligustri reaches the last stage of its life- history, the sexually mature insect. The dissimilarity between the head, thorax and abdomen, which first appears in the pupa, is now far greater, but the outlines of these three heteronomous regions are much softened and obscured by the thick coat of hair and scales that clothes the body. The small head distinguished by its light colour carries in this stage large sensory organs, — convex, black, pigmented eyes and antennae. As the food consists of the liquid nectar of flowers the organs of the mouth are reduced and modified, the only conspicuous parts being the spiritrompe or antliae> the homologues of the galeae in the biting mouth, which are extended in one of the specimens, and the hairy labial palpi which are, as in all Lepidoptera, turned forwards beneath the head. The thorax is broad and clothed dorsally with black-brown hairs, with a streak of white hairs over the roots of the wings. These organs are composed of a thin membrane supported by thick nervures, the whole surface being covered with thickly set and variously coloured scales, arranged in distinctive patterns and characteristic of Lepidoptera. The fore- and hind-wings on each side are connected by a hook and bristle. The bristle springs from the fore-margin of the hind-wing near its root and ends in a tuft of stiff black hairs. The hook springs from the under surface near the margin and at the root of the fore-wing. The three pairs of thoracic-limbs are of but secondary im- portance as organs of locomotion and are used chiefly for support. The tarsi end with strong claws. The abdomen is large and pointed posteriorly. Its first somite is clothed dorsally with black hairs, a few white being inter- mingled. The remaining somites bear a longitudinal median dorsal grey- brown band with a central dark line ; and each somite except the last has to either side of the median band a transverse bar of pink hairs, and a fringe IMAGO OF PRIVET HAWK MOTH. 157 of black hairs to its posterior margin. This black fringe encroaches more and more on the surface of the somites at the expense of the pink band the further back the somite is in the series. The female differs from the male as follows. The antennae are more slender, and want the setae-like hairs on the brown surfaces of the joints : the thorax is rounder in front and the fore-wings have generally a more curved anterior border and less acute apex : there is no hook developed on the fore-wings for the bristle. There are five pink bands on the abdomen instead of six : the last somite is broad basally, conical and as long at least as the two preceding somites, whereas in the male the abdomen tapers gradually to a point and is terminated by two valves with a vertical slit between them. The Lepidoptera are often divided into two chief sub-groups — the Rhopa- locera with the antennae ending in a club, and the Heterocera to which Sphinx belongs. The Heterocera have various types of antennal structure. They frequently possess the retinacular apparatus binding the fore- and hind-wings together, and their posterior tibiae have four instead of two spines. The Sphingidae, the family of which Sphinx is the type, are characterised by the prismatic shape of the antennae and the long bristle-like character of their terminal joint ; by a three-jointed very hairy labial palp with a minute terminal and two broad compressed basal joints ; by a one-jointed maxillary palp ; a robust body and relatively small wings. Their mode of flight is peculiar and sustained. Hence the popular name of Hawk-moths given to these insects. Ocelli are generally stated to be absent, but Cattie affirms their existence in Acherontia Atropos. To see the form and composition of the regions of the body it is necessary to divest it of hairs and scales by careful brushing. The head has no sutures. The prothorax is ring-like and is hence often termed ' collar.' Its tergum carries at each outer angle a vesicular dilatation clothed with long hairs, the patagium of Kirby and Spence. The prothorax is united to the mesothorax by membrane, but the latter and the metathorax are firmly connected. The mesothorax is very large. Its tergum is broken up into a large scutum and a lozenge-shaped scutellum behind. The fore-wings are attached to it, but their roots are covered by concavo-convex shields, the tegulae or wing-covers1. The metathorax is small. Its scutum is narrowed medianly where the scutellum projects forward. It bears the hind wings. The nervures of these organs become plain when the scales are brushed off. They contain extensions of the tracheae and blood-channels. For their arrangement, as well as the composition of the lateral walls of the thorax, the student must consult the larger works on Entomology. The limbs consist of the same parts as in the Cockroach. The anterior coxae are free; the median and posterior are closely attached to the thorax. The trochanter is small ; the femur short. The anterior tibiae have at their proximal end a peculiar enlarged moveable spine. The median 1 The terms * patagium ' and ' tegula ' are often misapplied. They are defined as in the text above by Kirby and Spence in the Orismology, Vol. iii. of the Introduction to Entomology. Patagia are structures peculiar to Lepidoptera, whereas tegulae are found in various other orders. I58 DESCRIPTIONS OF PREPARATIONS. tibiae have a pair, the posterior two pairs, of distal posterior spurs. The tarsi are six-jointed, and not clothed with hairs like the rest of the limb. The abdomen consists of ten somites. Each consists of a strongly chitinised tergum and sternum united laterally by a soft pleural membrane. The first of the series in both sexes, the ninth and tenth in the male, the seventh to the tenth in the female, require more notice. The first is firmly united to the meta- thorax. It is constricted, and its tergum is divided into a median and two lateral pieces, a division brought about apparently by the attachment of muscles. A slight groove and a difference in the chitin mark its separation from the second tergum. Its sternum is incurved and continuous with the second sternum. The ninth somite in the male is inclosed by the eighth. It consists of two narrow lateral chitinous bands which meet with expanded ends dorsally and ven- trally. Each band consists of a median, a dorsal and a ventral piece, the whole forming an S-shaped figure. Strong muscles are attached to these pieces. The valves which inclose the male genital organ and the anal papilla are attached to the posterior edges of the ventral pieces. Each valve is trigonal, concave internally, its margins fringed with long hairs. Near the ventral edge of its inner surface is a curved chitinous lamella, feebly toothed, — the harpe of Gosse (?=harpagon of White). To the posterior edges of both dorsal pieces of this ninth somite is articulated a stout decurved pointed process terminating in two hooks, the uncus of Gosse or tegumen of White. It immediately overhangs the slender anal papilla. A band of chitin connected with the base of the uncus, and continuous from side to side, curves under the same papilla, and from its mid ventral-point project two slender rods which appear to correspond to the scaphium of Gosse. The penis projects from the cavity below, i. e. in front of these rods, and above, i. e. behind the ventral union of the pieces of the ninth somite. The uncus appears to corre- spond to the cremaster of the pupa, the anal valve of the caterpillar. The bar curving below the anus may be either a chitinisation in the tenth or anal somite, or a dissociation from the ninth. In the female the seventh somite is much elongated. Its sternum is small and triangular, its pleural membranes large and meeting posteriorly and ventrally. Somites eight, nine, and ten are inclosed by it. The eighth is short dorsally, long ventrally. It has a small pleural membrane. Its sternum is strongly chitinised, grooved ventrally, and the groove narrows anteriorly, serving as a guide to the large orifice of the bursa copulatrix. The ninth somite is soft in texture. A very narrow band represents its dorsal and lateral regions ; its ventral region is thick- ened with a linear ventral groove. The tenth somite is represented by a large papilla slit vertically. Its sides are thickened, rough and pilose, and in the slit the anus opens above and the vagina below \ 1 De Lacaze Duthiers, in his series of classical papers on the genital armature of female Insecta, places the aperture of the bursa copulatrix of the Lepidoptera behind the sixth somite, instead of behind the seventh in the sternal region of the eighth, its ordinary position in Insecta (see the table on p. 230, A. Sc. N. (3) xv. 1853% except in Ephemeridae, where it lies in the seventh intersegmental membrane. De Lacaze Duthiers does not recognise the altered first somite nor the exact position of the orifice of the bursa. Newport recognises the first somite, but has not described the differences between the male and female. His figures (Figs. 391, 392, pp. 922, 923, Article Insecta, cited below) are from a male. The correctness of the view taken above may be gathered partly from the account given of the pupa, partly from the relations of the spiracles (infra]. IMAGO OF PRIVET HAWK MOTH. 159 The structure of the body as above described is probably typical of the Lepidoptera, but the subject is one that calls for investigation. The region of the first abdominal somites seems the most variable. The spiracles are vertical slits as in the previous stages. The last abdominal spiracle, the eighth, is aborted. A slight scar indicating its position may some- times be found in the male. The remaining spiracles are the prothoracic and seven abdominal. The former is situated in the soft skin between the pro- and meso-thorax, nearer to the prothorax. The first abdominal is the largest of the whole series. It lies in the pleural membrane under the edge of the lateral piece of the first tergum, i. e. on the abdominal side of the thoraco-abdominal constric- tion1. The other spiracles are all situated in the pleural membranes of their respective somites. If the interior of the pupa skin is examined, the cuticle shed from the first portions of the tracheae may be found attached to all the pupal spiracles, with the exception of the eighth abdominal, thereby shown to be closed. As to the mouth parts there is a labrum forming a narrow band with a median process and a lobe covered with hairs at each outer angle 2. Mandibles are absent as articulated pieces, a characteristic feature, according to Walter, of all the Lepidoptera save Micro-lepidoptera (Tineinae^ &c.). But from investigations made upon caterpillars about to assume the pupal condition, it seems to me that two stout pointed projections of the genae, lying to the outer side of the lateral lobes of the labrum, are non-articulated representatives of these appendages. At any rate, they are formed at the base of the caterpillar's mandible. The maxillae consist of a cardo and stipes, imperfectly separated and immoveably united to the head. The palps are one- jointed and bear a tuft of hairs. But the bulk of the maxillae, the antliae, consists of the greatly-developed galeae. Each galea is prolonged into a long band spirally coiled when at rest ; convex on its outer, concave on its inner, face, thus forming a channel. The faces are strengthened by independent systems of chitinous spots 01 rings. The inferior edge of each channel bears a series of processes by which one galea is strongly tied to the other; while its superior edge carries one or more series of flat processes which overlap the corresponding processes of the other side. The outer surface carries hairs and ' borers ' (ppotrypes\ the latter restricted to the tip. The borers are simple in Sphinx. When well developed they appear to act by piercing the nectaries of flowers or even pulpy fruits (e. g. as by Ophideres\ and it is possible that they may also be to a certain extent organs of touch. The labium is reduced to a membrane between the bases of the maxillae. Its palps are three-jointed. The terminal joint, though small, is moveable. The second is deep, compressed and grooved where it fits round the maxilla of its side, and the first joint narrows to its base which is attached to a small cone 3. 1 Newport (article Insecta) calls the first of these two spiracles, mesothoracic (p. 923), the second, metathoracic (p. 924) ; and Millie r-Blumenau uses the same expressions, connecting the spiracles of the imago with the two closed thoracic spiracles he discovered in an aquatic cater- pillar. See ante, p. 150, and p. 199 of paper quoted. In all the specimens I have examined they are situated as above described. 2 The prominence figured by Newport (article quoted) Fig. 377, as labrum, is a process of the clypeus from which a muscle takes origin. The parts designated mandibles by him (as they are also by Savigny) are really the lateral lobes of the labrum according to Walter. The only pupa of S. Ligustri at my command in a fit state to decide the question appears to corroborate Walter's view. 3 Walter has shown that in certain species of Micropteryx (Tineinae) there is a hinged and toothed mandible ; a maxilla with moveable cardo and stipes, six-jointed palp, a thin lacinia and 160 DESCRIPTIONS OF PREPARATIONS. Changes have taken place in the internal organs. The supra- and sub-oeso- phageal ganglia have coalesced round the oesophagus. The mesothoracic ganglion appears to abort \ the prothoracic persists and is approximated to the metathoracic which has coalesced with the first abdominal. The second and third abdominal ganglia abort, but the nerves which they give off persist. The abdominal ganglia of the imago correspond to the four last ganglia of the caterpillar. A remarkable structure, the chorda supraspinalis, overlies the abdominal ganglia, extending from the spot where the nerves arise from the sixth, an aborted ganglion, to the last ganglion. It is well-developed in the imago, and has been found by Burger in a large number of Lepidoptera. It is triangular in section and is intimately united by one angle to the neurilemma of the nerve cord whilst lateral muscles are attached to its two other angles extending thence to the abdominal walls. It is composed either of reticular cell-tissue with intervening jelly, or of vesicular cells. The chorda forms a strong attachment for the muscles, and the latter roof in the ventral blood-sinus (see under Cockroach, p. 144). There are vesicles developed on the tracheal branches in the abdomen. In the digestive tract there is a pharyngeal sac which can be dilated and contracted by systems of muscles. The oesophagus is long, and has at its stomachal end a non-pedunculated sucking stomach. The chylific stomach of the caterpillar is much reduced. The Malpighian vessels have lost their bead-like caeca. The intestine is long and the colon, as in many Lepidoptera, has a large dorsal caecum. The testis is single externally, the two embryonic testes having become en- veloped in a common sheath, as is often the case in Lepidoptera 2. But there are two vasa deferentia : they are long much contorted tubes, and each duct receives an accessory tubular gland. Each ovary consists of four long convoluted ovarian tubules united terminally by a ligament, both inter se and to the dorsal wall of the abdomen. The bursa copulatrix is large and pyriform, and a slender canal starts from its neck on the ventral anterior surface, curves round the neck and enters the side of the vagina. At its entrance a slender caecum is attached to the vagina which probably represents a spermatheca. There are two accessory tubular glands which dilate each into a pyriform enlargement, then fuse, and their common duct opens dorsally into the vagina near its exit. This exit is immediately below the anus 3. enlarged galea ; a labium with mentum, three-jointed palps, paraglossae and ligula. In other species of the same genus the mandible is a simple lobe, the laciniae of the two maxillae are lost, and the galeae are transformed into more or less typical antliae capable of being coiled up and closely united, and there are no paraglossae. Mandibles as more or less simple lobes articulated to the head appear to exist in most Micro-lepidoptera. 1 So thinks (Newport Phil. Trans. 1834, P- 394), but with reference to Vanessa Urticae he says (p. 416) that the prothoracic and mesothoracic, his second and third, ganglia fuse. Herold states the same fact with reference to Pieris Brassicae. 2 Cholodkowsky has shown that when the external capsule is removed, the apparently single testis consists of eight testicular follicles, four to each vas deferens. The follicles are arranged in various ways within the sheath. Hepialus Humuli retains the testis in the primitive condition, i. e. four free fol- licles on each side. In Pygaera anachoreta, &c., each set of follicles is contained within a separate capsule : but in most Lepidoptera the follicles of both sides are contained in a common capsule. 3 There can be little doubt that the canal connecting the bursa copulatrix to the vagina, itself represents the primitive vagina. The aperture of the bursa is in the typical position of the vaginal aperture. The canal in question has to pass round the neck of the bursa to gain its destination. The bursa in all other insects is a dorsally-placed appendage to the vagina. These IMAGO OF PRIVET HAWK MOTH. 161 The origin of a perfect metamorphosis, such as that of Sphinx, with three well- marked stages, larva, pupa, and imago, is probably due to the operation of more than one cause. The Cockroach is an example of an insect which always lives from birth to the adult state in the same manner and under the same conditions ; and this is the case with all Ametabolous Insecta. At birth such insects differ from the adult only in a few points, and the differences are gradually abolished. On the other hand, the differences between a caterpillar and a butterfly are very great, and the change from one state to the other is effected in an abrupt manner. It has been pointed out by Balfour (Comp. Embryology, i. p. 353) that a pupa-stage might easily arise (i) from a change at first small, then greater, in the character of the mouth-parts which would necessitate a more or less prolonged period of quiescence, the more prolonged the greater the change ; (2) from the operation of climate and other natural causes on such a period of quiescence. The question however is a very difficult and complex one, impossible to treat in a short compass, and the student must refer to the list of works given below relating to the Metamorphosis and Genealogy of Insecta. But it must be carefully borne in mind that, among the Micro-Lepidoptera, certain minute moths are known with mouth-parts con- formed to the type of biting mouth-parts (note, p. 159); that it is possible that the period of pupation is accompanied by more than one ecdysis, and is therefore essentially a period of abbreviated development ; that in some Hymenop- tera Aculeata there is, as it were, a preparatory stage previous to the true pupa- stage (Packard, Guide to Study of Insects, 1872, p. 66) ; that the changes from the larval mouth-parts to those of the adult are not so very great in all Metabola, e. g. in the Tenthredinidae • and that grades may be traced among Metabola, e. g. the Trichoptera and some Neuroptera afford examples of a comparatively simple perfect metamorphosis, the Lepidoptera a more complex example ; whilst in some Hymenoptera, and especially in the Diptera as a class, the metamorphic changes are of a very profound character. Nor must it be forgotten that our knowledge of many details of anatomy, embryology and of life-histories is still very imperfect. British Butterflies and their Transformations, Humphreys and Westwood, London, 1841. British Moths, lid., 2 vols., 1843-45. Tineina (Insecta Britan- nica\ Stainton, London, 1842. Natural History of Tineina, Id., 13 vols., London, 1855-73. Butterflies, &c., Scudder, New York, 1881. Anatomy of Danais Archippus. Burgess, Anniv. Memoirs, Boston Soc. Nat. Hist. 1880. Cossus. Lyonet, Ouvrage posthume (see p. 156, ante), Paris, 1832. Sphingidae. Butler, Tr. Z. S. ix. 1877. Anatomy of, &c., Baltzer, A. N. 30, 1864. Mouth-parts. Kirbach, A. N. 50, 1884. Walter, Beitrage, J. Z. xviii. 1884. Antlia. Breitenbach, J. Z. xv. 1882. In Ophideres fullonica. Darwin, Q. J. M. xv. 1875. Mode of Action of mouth-parts. Burgess, Amer. Naturalist, xiv, 1880. Palpus maxillorum. Walter, J. Z. xviii. 1884. Wing and Scales. Semper, Z. W. Z. viii. 1857; Deschamps, A. Sc. N. (2) iii. 1835; cf. Dimmock, Psyche, iv. 1883. On Wings in Insecta. Adolph, Nova three facts point strongly to the conclusion stated. The vaginal aperture below the anus is therefore secondary, but the mode in which it arises has not been worked out. It should be noted that in the caterpillar the somites in this posterior region are much abbreviated. M 1 62 DESCRIPTIONS OF PREPARATIONS. Acta, 41, 1879; 46, 1884; Amans, Comparaisons des organes-du vol dans la se'rie animate, A. Sc. N. (6) xix. 1885. Stridulation of Acker ontia. Moseley, Nature, vi. 1872. Nervous system. Chorda supraspinalis. Burger, Arch. Zool. Niederland. iii. 1876-77. Cattle, Z. W. Z. xxxv. 1881. Nussbaum, Z. A. viii. 1884, pp. 17, 48. See works cited ante, pp. 152, 156. Eye. Carriere, Sehorgane der Thiere, pp. 152, 186, Miinchen und Leipzig, 1885. Grenacher, Sehorgan der Arthropoden, Gottingen, 1879, p. 103. Carriere, :Q. J. M. xxiv. 1884 ; Id. Z. A. ix. 1886, and Hickson, Q. J. M. xxv. 1885. Malpighian vessels. Cholodkowsky, C. R. xcviii., xcix. 1884. Stigmata. Krancher, Z. W. Z. xxxv. 1880. Sex apparatus in Nematois metallicus. Cholodkowsky, Z." W. Z. xlii. 1885. Testis. Cholodkowsky, Z. A. iii. 1880; vii. 1884. Clasping organs of Rho- palocera. Gosse, Tr. L. Sc. (2) ii. 1883; White, Ibid. i. 1879. Connection of genitalia to heart. Landois, Z. W. Z. xiii. 1863. Ovary. Brandt, Das Ei, Leipzig, 1878. Female apertures. De Lacaze- Duthiers, A. Sc. N. (3) xix. 1853. Metamorphosis. Martin Duncan, Transformation of Insects (ed. iii.), Cassel and Co., London (n. d.). Sir John Lubbock, Metamorphosis of Insects (Nature Series), 1874. Balfour, Comp. Embryology, i. p. 348 et seqq. Keferstein, Betrachtungen iiber die Entwickelungsgeschichte der Schmetterlinge u. deren Variation, Erfurt, 1880 (not seen). Genealogy of Insecta. Packard, American Naturalist, xvii. 1883. Brauer, Verhandl. K. K. zool. bot. Gesellschaft, Wien, xix. 1869; xxviii. 1878. Id. Sys- tematisch. Zoologisch. Studien, SB. Akad. Wien, xci. Abth. i, 1885 ; see Emery, Biol. Centralblatt, v. 1884-85. Mayer, J. Z. x. 1876. Cf. Wood Mason, Trans. Ent. Soc. 1879. 32. COMMON CRAYFISH (Astacus fluviatilis), FEMALE. THE body of this Crustacean, like that of all Podophthalmata, consists of two great divisions, an anterior, the cephalothorax, covered dorsally and at the sides by a large continuous shield, the carapace, and a posterior, the abdomen, consisting of six separate metameres or somites, and of an azygos terminal flap, the telson, the last somite of the body, upon which the anus is situated ventrally. The cephalothoracic carapace is divisible into two regions by a well-marked curved line, with its concavity looking forwards, which is known as the cervical groove. The part anterior to this line corresponds to the head, the part posterior to it to the thorax, and they are known respectively as cephalo- and omo-stegite. The omo- stegite is marked dorsally by two longitudinal and short branchio-cardiac grooves connected anteriorly by a curved transverse groove. Within the area inclosed by these three grooves lies the heart. The lateral areae of the omostegite, known as the branchiostegites, roof in the branchial chamber. COMMON CRA YFISH. 1 63 They are formed by two flaps, right and left, which are homologous with the pleura of the abdominal somites, and like those pleura have an outer and inner lamella and a free ventral edge. The cephalostegite bears a median anterior projection, the rostrum, and to either side of the base of this rostrum an eye is visible, pedunculate as in all Podophthalmata. The two first antennae (antennules), each with an outer and inner division, the exopodite and endopodite, project forwards in front of the rostrum, and have to either side the long annulated second antenna (antenna) with a pointed scale or squame, the exopodite, at its base. The appendages in relation with the mouth, the two mandibles, the two pairs of maxillae, and three anterior pairs of thoracic limbs or maxillipeds, can only be clearly identified by dissection. The third pair of maxillipeds, however, is conspicuous and lies between the first and largest pair of ambulatory thoracic limbs. Of these ambulatory thoracic limbs there are five pairs. Unlike the three pairs of maxillipeds they consist of a single stem, the endopodite, the exopodite being lost in all Decapoda. Each limb of the first pair is very large and is often spoken of as c Chela/ but is perhaps better termed, with Professor Huxley, ' Forceps.' It is chiefly used in prehension not in crawling, the function of the remaining limbs. These limbs are slender and consist each of seven joints, the typical number in the higher Crustacea, and known, counting from base to tip, as coxopodite, basipodite or basis, ischiopodite, meropodite, carpopodite, propodite, and dactylopodite. One of the coxopodites bears the genital aperture in both sexes in all Macrura, the group to which Astacus and the Lobster belong, in the Hermit Crabs and the Arthrostraca or Sessile-eyed Crustacea. The oviducal aperture in the Macrura is on the third coxopodite, and in this specimen a black bristle has been inserted into it. In the male, the aper- ture of the vas deferens is on the fifth coxopodite. The membrane connecting the coxopodite and basipodite is the spot at which the limb separates when the animal throws it off in consequence of either fright or injury. The basi- and ischio-podite are united in the forceps, so that the joints of this pair of limbs are reduced to six. The three first pairs of ambulatory limbs are chelate. The chela or claw is produced by the enlargement of the outer angle of the propodite into a process equal in length to the dactylopodite. Foreign objects are seized between this pro- cess and the moveable terminal joint. The chela of the Scorpion is fashioned on the same plan, but the produced angle of the propodite is in this animal the internal angle. The claw in the Squillidae is of a different type. The two terminal joints of the limb are elongate, the propodite is more or less grooved, and the dactylopodite bends backwards and fits into the groove. Note that the space between the bases of the thoracic limbs becomes wider and wider posteriorly, and that the sternum of the last pair M 2 164 DESCRIPTIONS OF PREPARATIONS. is separate and moveable ; whereas the anterior sterna are calcified as a continuous whole. The six abdominal somites are all free and connected one to another by soft intersegmental membranes. Each somite bears attached a single pair of appendages, swimmerets or pleopoda. The somite forms an un- broken ring. Its ventral region between the attachments of the limbs is the sternum, while the opposite dorsal area is the tergum. The flap pro- jecting ventrally and laterally is the pleuron, and a small space between the socket for the limbs and the base of the pleuron is known as epimeron. A typical swimmeret, e. g. that of the fourth somite, consists of a basal protopodite bearing two processes, an inner, the endopodite, and an outer, the exopodite, both fringed with setae. There are two joints in the protopodite, a small basal coxopodite and a larger basipodite. The endo- podite consists of a simple unjointed basal and a terminal jointed or annulated portion. The exopodite is similar but its parts are smaller. The last pair, often termed par excellence the swimmerets, are somewhat modified. The protopodite consists of a single joint The endopodite and exopodite are expanded into broad thin plates, and the latter is divided by a transverse joint. This pair of limbs together with the telson make up the caudal fin, by means of which the animal deals a powerful stroke upon the water, and darts backwards whenever the tail is suddenly flexed. The first pair of abdominal limbs is either wanting altogether in the female, or one limb is present without the other. It is rare for both to be present. When present they are reduced to slender filaments with a minute basal joint or protopodite, and a jointed terminal portion which perhaps represents an endopodite. The two first pairs of these appendages are modified in the male for sexual purposes, and are to be seen in the two following pre- parations. The English and Irish Crayfish is widely spread over the Continent. It constitutes the variety of Astacus flnviatilis known as A. torrentium, the £crevisse a pieds blancs. There is a Crayfish which is found also widely spread over the Continent and which resembles A. torrentium very closely. This constitutes the variety of A. fluviatilis known as A. nobilis> the Iicrevisse a pieds rouges. The body walls or integument consist of (i)^uticular structures for the most part calcined ; of (2) a single layer of ectoderm or hypoderm cells (=chitinogenous cells); and of (3) connective tissue imbedding pigment cells, bloodvessels and nerves. The cuticular structures differ somewhat in the hard calcified and the soft intersegmental regions. The former is composed of four distinct strata : (i) a cuticle, (2) a pigmented calcified layer, (3) a non-pigmented thick layer containing a very large proportion of calcareous matter, and (4) a non-calcified softish and COMMON CRAYFISH. 165 thin layer. The organic substratum is chitinoid. The salts are Calcium carbonate and phosphate, chiefly the former. (1) The cuticle is structureless, resistent, and of a yellowish tint. It shows areae corresponding to the outlines of the ectoderm cells, and inside these areae slight ridges. The latter are supposed by Braun to be due to the coalescence with the young cuticle during its formation of processes similar to those of the intestinal cuticle. Both areae and ridges are more distinct in the newly formed cuticle. (2) The pigmented layer consists of a system of fine lamellae, parallel to the surface, and alternately more and less refractile, the latter being mere lines. The lamellae are perforated by minute close-set vertical pores. The pigment lies between the more refractile lamellae in the form of bluish-black-granules, sometimes furnished with processes. (3) The third layer makes up the chief substance of the hard parts. Its refractile lamellae are thick, well calcified, and perforated by pores continuous with those of the pigmented layer. The lamellae become thinner towards the inner surface, where they merge insensibly into the fourth layer. (4) This layer is very distinct in the Lobster. It is not recognised by Braun, but according to Vitzou it is composed of delicate non-calcified lamellae, and the innermost show impressions or outlines corresponding to the ectoderm cells. It is said to possess but few vertical pores. There are certain variations in different regions from this typical structure. The knobs on the opposing surfaces of the chelae, certain spines, e. g. those on the posterior dorsal edge of the basal joint of the exopodite in the last pair of swimmerets, are colourless. The pigmented layer is absent, and the cuticle is greatly thickened in these places. In the articular and intersegmental membranes the cuticle is present, but the remaining layers are represented by non-calcified lamellae not distinguishable into systems. The inner wall of the branchiostegite is extremely thin, and is composed of a membrane resembling the cuticle. Areae are absent here as they are on the eye-stalk, and in the oesophagus and stomach. They are very well marked in the intestine, and each area supports 3-6 pointed ridges. The cuticular structures of the alimentary canal are composed of a cuticle and underlying lamellae, which are thickened and calcified only in the gastric ossicles. Vertical pores are found also only in the same structures. There are integumental setae or hairs and glands. The setae are either hollow or solid. The hooked setae on the inner wall of the branchiostegite are solid, and certain solid setae in the oesophagus and stomach are probably simply processes of cuticle. Hollow setae are either closed or open at the base. The former are most numerous, and their stem is beset with solid barbs, i. e. they are feathered. The latter are chiefly confined to the antennae. They are short, thick-walled, and their stems are destitute of barbs. A pore-canal pierces all the layers of the integument except the cuticle, and leads to the base of each seta. The cuticle is continuous with the base of the hair, which is generally lodged in a slight depression. The pore-canal contains a process of the ectoderm, in which nuclei are present. It is prolonged internally below the integument by the hair-tube. This structure has nucleated walls, and expands at its inner extremity into a bulb in which cell-outlines are distinguishable. It is about half the length of a hair, and contains a hair papilla which fills its cavity. A formation of new hairs precedes the formation of a DESCRIPTIONS OF PREPARATIONS. new cuticle, &c. The barbs, when present, are first formed, then the walls of the stem, both alike from the papilla and the walls of the tube. The tip of the hair is slightly hooked, and fits into the pore-canal of its predecessor. When the old integument is cast off the new hairs are therefore evaginated mechanically. The soft substance contained in the stem atrophies and a new hair-papilla is formed, as is believed by Braun from the walls of the tube. The glands of the integument are (i) the glands of the roof of the branchial cavity, and (2) the cement glands found only in the female. The former were dis- covered by Leydig. They lie in the substance of the branchiostegite and open singly on the roof of the branchial cavity. They are tubular in structure and are but slightly lobed. "The gland cells are columnar and pointed. The cement glands were discovered and investigated by Braun, and have been found by him in other Decapoda. In the Crayfish they extend over the anterior two-thirds of the ventral surfaces of the abdominal pleura, and thence they spread along the edges of the sterna almost continuously. They also cover about a third of the base of the exo- and endo-podite of the last pair of swimmerets. These cement glands undergo a periodical development, 5-8 weeks before the eggs are laid in November or December, and then they give the parts a whitish appearance. The glands' them- selves are tubular with rounded or polyhedric cells supported by a basement membrane : their ducts rarely open singly, but as a rule in groups. Lereboullet, who first observed the white appearance of the abdomen, but who failed to recognise the glands, states that their secretion coagulates on exposure to water. Shortly before oviposition the female flexes the abdomen ; the cavity thus formed becomes filled with a transparent viscid fluid which glues the edges of the opposing somites together. The ova pass into the chamber, are impregnated and suspended to the setae fringing the abdominal limbs, to the abdominal sterna and intersegmental membranes by a hardened layer of the cementing fluid. It is possible that the fluid may set the spermatozoa free from the coat that binds them together. The ectoderm is composed of a single layer of cells columnar or cubical in certain places. In the inner lamella of the branchiostegite, in the abdominal pleura, and the exo- and endo-podite of the last pair of swimmerets, places where the cuticular structures of the two surfaces are near together, the ectoderm cells are enlarged at intervals, and their bases are connected from one to the other surface by bundles of nucleated fibres which appear to be, but are probably not, continuous with the ectoderm cells themselves. A layer of ectoderm cells also intervenes between the attachment of the muscle fibres to their chitinoid, so-called 'tendons,' which are processes in reality of the cuticular structures, and are moulted with them, as has been proved in the case of the tendons of the adductor mandibulae and of the muscle which adducts the dactylopodite of the forceps. Beneath the ectoderm is a layer of fibrillated connective tissue, processes of which extend inwards, accompanied by large-celled connective tissue. The pigment cells of this layer are stellate, and contain some a yellow, others a red, pigment, together also with groups of quadrate or oblong crystals of a deep blue colour. Vessels are present, and probably nerves, as irritation of the newly formed integument causes movements on the part of the animal. The moults occur in the warm part of the year (May-September). According to Chantran, they take place (in A- nobilis ?) as follows. The young animal is hatched in May-July. It moults once COMMON CRAYFISH. I(57 in 10 days' time; then four times at 20-25 days' interval up to September. Three more moults in May-July following complete the first year of life. There are five in the second year, two in the third, and the animal is now becoming adult. The male is sexually mature after the ^th-iyth moult, and henceforward under- goes ecdysis twice yearly. The female is sexually mature in the fourth year and only moults once a year. The process is very probably in abeyance in old Crayfish, but it is not certain* at what age they cease to grow. According to Carbonnier there are 2-3 moults in the first year, and then the process becomes annual. When the time approaches for a moult, the old integument becomes softer and thinner, and the ectoderm cells lengthen. The gastroliths and hairs are previously formed. The first step, according to Braun, is the formation of minute processes, 2-5 to each cell, which afterwards form (?) the ridges of the areae (see supra]. Vitzou did not observe this formation in any Decapod, but states that the inner surface of the cast-off integument is covered by a soft transparent substance. A section taken through the old and new integuments shows the presence of a new and very delicate cuticle, an outer system of continuous lamellae, and an inner system traversed by vertical lines, corresponding to the contours of the subjacent ectoderm cells. Vitzou concludes from this fact, and from the subsequent shortening of the ectoderm cells, that the lamellae are developed by a specialisation, chemical and physical, of the outer ends of these cells. And his opinion is borne out by the fact that in sections taken parallel to the surface from the carapace of Crabs polygonal areae corresponding to the ectoderm cells may always be seen. The outlines are brought more distinctly into view by treatment with silver nitrate. But in most instances, e. g. Lobster, these areae are obliterated by early fusion of the cell-walls. Tullberg appears also to be of the opinion that a conversion of the outer ends of the cells takes place during the formation of the new carapace (cf. Zoolog. Jahresbericht, 1882, Arthropoda, p. 12). Vitzou has proved the presence of glycogen in the con- nective tissues preparatory to and during the moult. Certain tendons, the cornea of the eye and the lining membrane with the hairs of the auditory sac, are regenerated at the same time as the rest of the integument. The endophragmal skeleton is broken up previously to being cast off (Mocquard, C. R. xcvi.). The new cuticle of the alimentary canal is formed at a comparatively late period. The pores in the lamellae are due to the development of hair-like processes from the cell-surfaces. They are very close-set, and subsequently disappear leaving the pores (Braun). The old integument splits in z&Macrura across the back, between the cephalo- thorax and the first abdominal somite. A split is said to occur also along the limbs. The animal lies upon its side and withdraws first the cephalothorax and then the abdomen with their several appendages. In the Crabs a split occurs along a circular suture between the tergal and pleural regions of the cephalothorax. The animal retains the horizontal position, and the abdomen is freed before the cephalo- thorax (Vitzou). The appendages in front of the mouth, and those which are modified into mouth-parts must be disarticulated for the purpose of examination. The eye-stalk has a short basal and a long terminal joint. It was formerly sup- posed that it represents a limb, and therefore a somite. It is however derived DESCRIPTIONS OF PREPARATIONS. from a process of the procephalic lobes, and the eye is not stalked at its first appearance in those forms of Crustacea in which it is stalked in the adult when they pass through a complete series of developmental forms commencing with a Nauplius or Metanauplius. The series of true appendages appears to commence with the first antennae. The eye itself surmounts the terminal joint. It has a convex, soft and trans- parent cornea marked out by faint lines into square facets. The visual structures are arranged in two layers. To each corneal facet corresponds an eye-element, i. e. a crystalline cone and a retinula. The former is derived from cells of the ectoderm of the procephalic lobes, the latter from the supra-oesophageal ganglion (?). The crystalline cone is formed from four crystalline cells. The nuclei of the cells and a small quantity of protoplasm lie immediately under the corneal facet. The outer part of the crystalline cone is less refractile than the inner part, which is long and has the shape of a four-sided pyramid terminated by a long pointed piece which has four projections fitting into the rhabdome of the retinula. The retinula is composed, as in Palaemon, of retinal cells grouped round a rhabdome formed of four square chitinoid rods. Each rod swells out posteriorly, is red in colour, and is marked by alternate light and dark striae which do not correspond in position in adjoining rods. The retinula cells surround the rhabdome and contain black pigment granules. A basement membrane pierced by the nerve fibres, one to each retinula, separates the retinulae from the optic ganglion. Nucleated yellow brown pigment cells intervene between the retinulae, and two (or more) black pigment cells between the crystal- line cones. The ectoderm cells at the margin of the eye are elongated and filled with black pigment \ The protopodite of the first antenna is three-jointed. The basal joint is tri-hedral and lodges the auditory sac. The exo- and endo-podite are both annulated. The number of joints in the exopodite vary, but, distal to the first eight, they bear on their ventral surfaces the olfactory setae so-called. The auditory sac is curved, and possesses delicate chitinous walls. Its aperture, which is permanently open, but is protected by numerous setae springing from its outer margin, lies in the dorsal surface of the joint. The auditory setae are arranged in two rows which meet at the closed end of the sac. The largest setae are one-fiftieth of an inch long. They are hollow and are moveably attached to the wall of the sac. This attachment is by a membrane delicate on one side, stout on the other forming the ' tooth ' of Hensen. A narrow plate, the ' ligula ' of Hensen, is developed in the shaft of the setae, on the side opposite the tooth. The shaft with the exception of the ligula is beset with five solid barbs. The nerve fibre is furnished with a ganglion cell close to its ter- mination, which is fine and delicate, and according to Hensen attached to the ligula. The otoliths are numerous and irregular in shape. According to Hensen's observations on Palaemon antennarius, these otoliths are foreign particles collected by the forceps and scattered over the base of the first antenna, whence some find their way into the sac. A specimen which moulted and was confined in a basin of 1 For the terms descriptive of the structures found in the eyes of Arthropoda see the general description of that phylum. The Crayfish's eye would be described, using the terms there given, as polymeniscous, diplostichous, retinulate, furnished with vitrellae, and perhaps also as exo- chromic. But the origin of the pigment cells has not been exactly ascertained. See Lankester and Bourne, Q. J. M. xxiii. 1883, on Eyes of Scorpio and Limulus. COMMON CRAYFISH. 169 filtered sea-water was supplied with crystals of uric acid. Whilst the cast-off auditory sac contained the usual otoliths, the new sac contained a large proportion of uric acid crystals. The foreign bodies thus obtained are kept in situ apparently by a gelatinous substance. When the auditory sac is a closed one, there are either no otoliths (Brachyura), or e. g. in My sis, there is one otolith in the shape of a rounded laminate body, apparently a secretion. The olfactory hairs occur 2-3 on the first two joints that possess them, the sub- sequent joints having an anterior and posterior row with 7-8 hairs in a row. They are about -5-^5- of an inch long, * shaped like a spatula with a rounded handle and somewhat flattened blade' (Huxley). They are two-jointed, and contain a soft granular tissue. Jourdain, who has recently investigated these structures in various Crustaceans, states that a small hyaline body projects from the free extremity and that a nerve fibre is traceable to the base of each hair, and may sometimes be seen to have a swelling (? ganglion cell). He terms the hairs ' poils a batonnet,' and divides them into ' poils & batonnet cylindriques et a batonnet stipite's.' The former are long and cylindrical and usually many-jointed, the latter are usually three-jointed and somewhat fusiform. According to him the Crayfish possesses the cylindrical variety, but the hairs are short. The protopodite of the second antenna is two-jointed. The basal joint bears a ventral tubercle, to the inner side of which is the aperture of the green gland. The second joint is divisible into two parts more or less moveable, and bears an exopo- dite in the shape of a scale or ' squame.' The endopodite is long and many- jointed. The mandible or first appendage of the mouth consists of four joints. The basal (= coxopodite) is long and forms a three-sided pyramid. The base of the pyramid projects inwards over the sides of the mouth. The oral side of the base bears two stout obscurely separate teeth ; the posterior side and outer angle form a sharp ridge with several teeth, but the anterior side i& hollowed out, the hollow in- vading the centre of the base. The other three joints are small, the terminal dilated and fringed with setae. They are articulated to the anterior side of the pyramid not far from the base. They constitute the ' palp,' which is not an exopo- dite, a structure rarely present in the mandible of the adult, as e.g. in some Cope- poda. In the larval form known as Zoaea, the first Zoaea-st&ge has no palp to the mandible. It sprouts out in later stages. In the Nauplius the mandible has the typical biramose character, but when Penaeus, the only Decapod with a Nauplius stage changes to the Zoaea, the mandible is reduced to its basal portion, and the palp is evolved at a later period. The Phyllopod mandible is similarly reduced, and never gains a palp. But the temporary suppression of a limb or part of a limb is by no means an uncommon phenomenon in the higher Crustacea (cf. the account in Balfour's Comparative Embryology, i., of the evolution of Sergestes, p. 398, Phyllosoma, p. 396, and Squilla, pp. 402-3). The other appendages of the mouth are best taken in reverse order. The most perfect is the third maxilliped. The protopodite is divisible into a coxopodite which bears a podobranchia and coxopoditic setae (cf. p. 182) and a basipodite. This joint, as in the forceps, is continuous with the basal joint of the endopodite which is divisible into an ischiopodite, meropodite, carpopodite, propodite and terminal dactylopodite. The exopodite or palp is short and articulates with the I70 DESCRIPTIONS OF PREPARATIONS. basipodite. It is slender, and has a long basal joint and a many-jointed filament. The second maxilliped has a smaller and softer endopodite, and a larger exopodite. There is a podobranch but no coxopoditic setae. The basipodite is distinct, and the meropodite very long. The first maxilliped has the coxopodite and basipodite imperfectly separate and expanded into thin setose plates. The endopodite is short and two-jointed ; the exopodite large, with a much elongated basal joint. The podobranch is reduced to the stem and lamina, and is known as the flagellum or epipodite of Milne-Edwards. The three pairs of maxillipeds belong to the thorax. The second maxilla has a very thin lamellate coxopodite and basipodite, each partially subdivided by a fissure. The endopodite is small and simple. The exopo- dite forms a large plate, the scaphognathite, which is kept in perpetual motion and bales the water out of the branchial chamber, into which it runs posteriorly and inferiorly. For its homology see Glaus, Untersuchungen zur Erforschung der genealogischen Grundlage des Crustaceen-sy stems, Wien, 1876, p. 42. The first maxilla has two thin simple and foliaceous expansions. The first represents the coxopodite; the second is usually termed basipodite, but the homology is doubtful. The endopodite is extremely reduced, and the exopodite absent. It is present in Euphausia and in the Zoaea of Penaeus up to a certain stage, and then atrophies. The sides of the mouth are formed by a soft lip. In front it is overhung by a leaf-like projection, the labrum, attached basally to the epistoma or broad trian- gular calcified area in front of the mouth and extending forwards to the base of the antennae. The sides of the mouth are overhung by the bases of the mandibles, and behind the latter are two small soft lobes united by the posterior margin of the mouth. These lobes are the paragnatha, metastoma, or lower lip. Neither labrum nor metastoma represent appendages in development, but it is possible that the latter represents a dissociated portion of the first maxilla (Glaus, p. 15, Neue Beitrage, &c., infra). The ventral region between the bases of the mouth parts, and again between the bases of the first and second antennae becomes calcified, forming a series of sterna. The antennary sternum constitutes the epistoma. The rostrum in Nebalia is a moveable process. In Squilla it is jointed to the fore-edge of the carapace and moves with the somite of the first antenna. In Nebalia it appears to belong rather to the region of the first than to that of the second antenna, to which Professor Huxley allocates it. The ventral and basal lateral parts of the sternal wall of the mouth-parts and thorax give origin to a number of internal cuticular folds or apodemata. These apodemata constitute the endophragmal skeleton and give attachment to muscles as well as protection to the thoracic portion of the nerve-chain and the sternal blood- sinus. Their arrangement is too complicated to be explained without the aid of figures, and the student may consult Professor Huxley on the Crayfish, p. 157, where will be found both description and figures. There remain for examination the two first pairs Of abdominal appendages in the male and the telson. The first appendage is unjointed and cylindrical. Its apex forms a plate slightly bifid. The sides of the plate are rolled upon themselves, the anterior half surround- ing the posterior, giving rise to a canal open at each end. It is not certain what parts of the typical limb are represented in this appendage. COMMON CRA YFISH. 171 The second appendage has a protopodite divisible into a coxopodite and long basipodite. The endopodite consists of a large basal piece and terminal jointed filament. The apex of the basal piece is prolonged upwards as a plate to the inner side of the filament, and the inner edge of the plate is rolled upon itself. The exopodite is present and has the usual structure. Both pairs of appendages are used for the transmission of sperm : see p. 186. Both the male and female alike possess no appendages to the first abdominal segment in the Parastaddae — the Crayfishes of the S. Hemisphere. The telson is a plate moveably articulated to the last somite of the abdomen. It is divided in nearly all the Potamobiidae or Crayfishes of the N. Hemisphere, but not in the Parastaddae^ by a transverse suture, so that the posterior half is moveable upon the anterior half. The anus is situated on the ventral surface of the basal portion. There is some question as to what the telson represents. Hartog (British Assoc. Reports, 1882, p. 575) believes that it represents the last somite of the Nauplius — a post-anal plate united to the furcae anales, the latter representing paired terminal outgrowths elsewhere developed into limbs by the formation of joints. He points out that in the Copepoda — in his view the primitive Crustacean group — the anus is a terminal dorsal slit; that the tergum of the last somite forms a supra-anal plate, whilst the furcae project one on either side of the anus. Supposing the supra-anal plate to become adnate to the furcae, the anus becomes first terminal and ventral, and finally by growth of the plate ventral. He homolo- gises two setose knobs projecting at the sides of the telson in Astacus with the furcae. It has been pointed out by Claus (Untersuchungen, &c. supra, p. 12) that in the JProtozvaea-stage of Penaeus the anus is terminal between two furcal processes. In its youngest Zoaea a short transverse bridge (= supra-anal plate of CopepodaT) connects the furcae dorsally above the anus. During subsequent growth this bridge enlarges, the anus becomes more and more ventral, and the furcae are lost, be- coming the two posterior setigerous processes of a broad bilobed terminal plate. In a young Phyllosoma (cf. Claus, op. cit. p. 51) the same facts may be observed, but the processes become obsolete, the setae alone persisting at the outer angles. There is much variety in the shape of the last somite of the abdomen in Crustacean larval forms. It is very frequently a broad bilobed plate, more rarely, as in the young Astacus itself, a simple plate. In both cases the anus is ventral. Claus concludes that the telson represents the terminal furcal somite of the abdo- men in Phyllopoda. It is possible that it may represent a region rather than a somite. For in Nebalia (Claus, Z. W. Z. xxii. 1872, p. 329) there are two somites behind the sixth abdominal somite, the last bearing a pair of furcae and the anus, and in most Phyllopoda the abdomen contains a large number of somites. When Astacus quits the egg it only differs from the adult in certain points summarised here from Professor Huxley's account. The cephalothorax is relatively large and convex in shape ; the short rostrum is bent down between the eyes ; the thoracic sterna are relatively wide ; the chelae of the forceps are slender, and the tips of all the chelae are strongly incurved, the young Crayfish attaching itself by those of the forceps to the empty egg-case : the dactylopodites of the two last pairs of thoracic limbs are hooklike : the first pair of abdominal appendages is un- developed : the sixth is included within the telson, which is a simple broad oval plate usually notched in the middle of its hinder margin. Setae are few in number and are mostly uncalcified prolongations of the cuticle not sunk in pits, and devoid DESCRIPTIONS OF PREPARATIONS. of barbs. It is noteworthy that at the stage of development in which the first and second antennae and mandibles are present as rudiments, and which therefore corresponds to the Nauplius of other Crustaceans, a cuticle is formed and then moulted. A similar phenomenon occurs in other cases, e. g. in Nebalia, in My sis within the incubatory pouch of the mother, and has been noted in the Isopoda, e. g. Asellus. It is perhaps a general phenomenon in Crustacea with a shortened de- velopmental history. The majority of Decapoda differ from Astacus in having a well-marked metamorphosis. The prawn Penaeus is hatched as a Nauplius, most other Decapoda as a Z^aea. The Lobster (Homarus], placed by M. Milne Edwards but not by Faxon among the Astacidae, starts in the Mysis stage, i. e. with the thoracic feet biramose and natatory. One or two fresh water Decapoda have a shortened metamorphosis : one or two Land Crabs none at all. Much has been written as to the ancestral character of the Nauplius and Zoaea. The subject will be found discussed in Balfour's Comparative Embryology, i. p. 41 7, to which must be added some recent remarks of Claus in his Beitrage, &c., Arb. Zool. Inst. Wien. vi. 1885, p. 91. It is at present impossible to say how the Nauplius has been derived, and what are its affinities. It is apparently not a simple organism, for in the Copepoda Natantia its body is divided into three somites, but the segmentation disappears before hatching. Of its three pairs of appendages the first is uniramose and for the most part sensory in function, the second and third are biramose and natatory. The three pairs correspond to the first and second antennae and mandibles of the adult. It is generally agreed that the second antennae are post-oral appendages. They are innervated from a post-oral ganglion in the Nauplius, and in the adult Apus, Limnetis, Branchipus among Phyllopoda. Daphnia much resembles Branchipus. It may be added that in the Nauplius they lie at the sides of the mouth, and as a rule develope a masticatory hook. The first antennae on the contrary are generally held to be prae-oral appendages. Claus in his Beitrage (supra] points out (i) that they retain their uniramose cha- racter in Entomostraca, whilst in Malacostraca they become in most instances secondarily biramose ; (2) that they are sensory in function ; (3) that they arise from a region of the head morphologically unlike (as in Chaetopoda) the somites of the body. He appears to regard them (i) as limbs, and (2) at the same time to compare them with the prae-oral tentacles of Chaetopoda and the antennae of Myria- poda and Insecta. As to this second point however it may be remarked that they do not originate from the procephalic lobes as do the antennae of Myriapoda and Insecta. Ray Lankester first drew attention to the innervation of the first and second antennae in Apus (Q. J. M. xxi. 1881), and Pelseneer (Q. J. M. xxv. 1885) has in- vestigated the nervous structures microscopically. He finds that the ganglia of the first antennae are contained in the supra-oesophageal ganglion, but are separate from the mass of ganglion cells supplying the eyes. They are connected by a transverse commissure. The nerve to the first antenna runs backwards accompany- ing the oesophageal commissure for some distance, as it does in Limnetis. He says that in Branchipus and Daphnia the corresponding nerve arises from a group of cells distinct from the rest of the supra-oesophageal ganglion. It may be observed that Rathke states that in the embryo Crayfish first and second antennae alike are supplied from a ganglionic rudiment distinct from an anterior rudiment apparently- COMMON CRAYFISH. 173 supplying the eyes. The natural conclusion is that the supra-oesophageal ganglion consists of two distinct parts in Apus, &c., i.e. of a true prae-oral ganglion supply- ing the eyes, and a second pair of ganglia shifted forwards supplying the first antennae. The shift forwards explains the backward course of the nerves. Whilst in ApuSj &c., the second antennae have a post-orally placed ganglion, in higher Crustacea this ganglion also has shifted forwards ; but there is evidence to show (embryo Astacus) that there is a distinct ganglion, supplying in this instance both first and second antennae, which fuses with an anterior rudiment to form the supra- oesophageal ganglion of the adult. It seems probable therefore that the first antennae, like the second, are in reality primitively post-oral appendages, or at any rate are homologous with the limbs borne by the post-oral somites of the body. The Arachnida afford us an instance of a group of Arthropoda in which all the appendages are embryonically post-oral. The first pair of appendages however is invariably shifted in front of the mouth during growth. In Scorpio and Limulus at least their nerves are said not to come from the supra-oesophageal mass but from the commissures. The ganglia however have not been investigated microscopically. It is possible that these animals retain an archi-cerebrum, i.e. a supra-oesophageal ganglion not fused with other ganglia placed posteriorly to it. The Crustacea, on the contrary, evidently possess a syncerebrum, i. e. a supra-oesophageal ganglion fused with one or two posteriorly placed ganglia. The following table shows in parallel columns the post-oral somites and their appendages in the Crustacea, Arachnida, Myriapoda, and Insecta. The antennae of the two latter classes are not included as being apparently processes of the pro- cephalic lobes. The post-oral appendages of Crustacea and Arachnida which become prae-oral are printed in a different type, and the first antennae of the former Class are marked with a note of interrogation as being somewhat doubtfully homo- logous with the succeeding appendages. See, for remarks on the descent of Arthropoda, Balfour, Comparative Embry- ology, i. p. 451 ; for a discussion on the relations of the Arachnida and Crustacea, Kingsley, Notes on the Embryology of Limulus^ Q. J. M. xxv. 1885, p. 556. TABLE OF POST-ORAL SOMITES AND THEIR APPENDAGES IN ARTHROPODA. Number of Segment. Crustacea. Arachnida. Myriapoda. Insecta. I. II. First Antenna (?) uniramose in Nau- plius and adult En- tomoftraca, second- arily biramose in Decapoda, &c. Chelicerae. Mandible : with a small palp in Chi- lopoda : non-pal- pate in Diplopoda. Mandible : non-pal- pate. Second Antenna : biramose in Nau- plius as a rule, but uniramose in many adult forms. Chelae in Scorpio and Thelyphonus. Pedipalpi in Spi- ders. Maxilla : palpate in Chilopoda : 4-lobed plate in some adult Diplopoda, but simple appendage in embryo. Maxilla: palpate. 174 DESCRIPTIOiVS OF PREPARATIONS. Number of Segment. Crustacea. Arachnida. Myriapoda. Insecta. III. Mandible : bira- mose in Nauplius : not palpate in Zoaea,nor inPhyl- lopoda, terrestrial Isopoda and Am- phipoda, nor Cu- macea, &c. A3-articulate'palp' in adult Decapoda. ist pair of limbs bearing outer ele- ment of so-called 'labium' in Scor- pion. Limb - like with basal parts in con- tact, Chilopoda : short, 4 - jointed with claw in lulus (Diplopoda), in - |i nth do.: anal so- mite in many Or- thoptera, &c. ? Cerci anales - limbs. DESCRIPTIONS OF PREPARATIONS. Number of Segment. Crustacea. Arachnida. Myriapoda. Insecta. XVIII. 5th abdominal so- mite with pleo- pod in a Malacos- tracan. 5th caudal somite in Scorpion. Anus behind it. XIX. 6th do. Telson in Scorpion. A jointed filament in Thelyphonus. XX. Telson. 7th abdominal so- mite, Nebalia. XXI. 8th abdominal somite in Nebalia, but there are only 6 abdominal ganglia in this animal. In Apus numidicus (Phyllopod} the total number of somites is said to reach 46. The following notes of the differences observable between the Lobster (Ho- marus) and the Crayfish, with which it is sometimes grouped in the family Astacidae, may be useful to the student. The squame of the second antenna is relatively small ; the fifth thoracic sternum is fixed ; the first abdominal appendage in the male is two-jointed, and the terminal joint is lamellate but only curved to form a gutter, not rolled up as in Astacus ; the second abdominal appendage in the same sex has a small plate to represent the rolled up lamina of Astacus : in the female the first abdominal somite bears appen- dages well developed ; the telson has no transverse suture. The number of podo- branchiae is as in Astacus, but the stem is completely split into a plume and lamina, and the branchial filaments are stiff and close set ; the single arthrobranch of the second maxilliped in Astacus is absent, and the total number of these organs is hence reduced to ten, but there are four pleurobranchiae. The infra-oesophageal ganglion is small relatively to the thoracic ganglia, and more distinctly constricted at the sides than in Astacus. The caecum of the mesenteron is small and bilobed : the caeca of the liver are short and the anterior portion of the gland large. The first part of the intestine is smooth, the terminal portion plicated, and at the junction of the two parts there is a dorsal caecum as in Amphipoda and most Decapoda. The testicular lobes are long and only joined by a commissure, and the vasa deferentia are shortened. The largest Gregarine known, Porospora gigantea, inhabits the alimentary canal : see Schneider, A. Z. Expt iv. 1875 ; E. van Beneden Q. J. M. xii. 1872. The Lobster quits the egg in the J/jw-stage, cf. Rathke, A. N. H., 1841 ; Sars, Vidensk. Selsk Forh. Christiana, 1874; Smith, Trans. Connecticut Acad. of Arts and Sciences, 1873. The anatomy of a Crab (Callinectes) and its developmental forms or Zoaea is given in W. K. Brooks's Handbook of Invertebrate Zoology, Boston, 1882, p. 1 68 et seqq. For the Zoaea of the Brachyura, see also Balfour, Comp. Em- bryology, i. p. 398. Crustacea, Woodward, Encyclopaedia Britannica (ed. ix.) vi; Gerstaecker, COMMON CRA YFISH. 177 Bronn's Klass. und Ordn. des Thierreichs, v. i, pt. 2 in progress. Histoire Naturelle des Crustace's, M. Milne-Edwards, 3vols., Paris, 1834-40. British Stalk-eyed Crustacea, Bell, London, 1856. British Sessile-eyed Crustacea, Spence Bate and Westwood, 2 vols., London, 1863-68. The Crayfish. Huxley, International series, xxviii. 1880. Classification and Distribution of Crayfish. Id. P. Z. S. 1878. Revision of Astacidae. Faxon, Mem. Harvard Mus. x. No. 4, 1885. Figures, Atlas of Practical Elementary Biology, Howes, 1885. Limbs, &c., and Phylogeny. Glaus, Arb. Zool. Inst. Wien. vi. 1885; Boas, M. J. viii. 1883. Homologies of Limbs. Ray Lankester on Apus, Q. J. M. xxi. 1 88 1 • Packard, Amer. Naturalist, xvi. 1882. Tegumentary Skeleton. Milne-Edwards, A. Sc. N. (3), xvi. 1851. Integument. Vitzou, A. Z. Expt. x. 1882 ; Braun, Arb. Zool. Zoot. Inst. Wurzburg, ii. 1875. Moult of Astacus. Chantran, C. R. Ixxi. 1870; Vitzou, op. cit. supra. Change of Tendon of Adductor mandibulae. Baur, Miiller's Archiv (Archiv f. Anat. und Phys.), 1860. Self- Amputation of limbs. Dewitz, Biol. Centralbl. iv. 1884-5; *n Crab, Fredericq, Arch, de Biologic, iii. 1882 ; Id. A. Z. Expt. (2) i. 1883. Regeneration of limbs in Astacus. Chantran, C. R. Ixxiii. 1871. Glands of branchial cavity and cement glands. Braun, op. cit. supra. Cf. Nebeski on Amphipoda, Arb. Zool. Inst. Wien, iii. 1881. Organs of Special Sense. Eye. Grenadier, Sehorgane der Arthropoden, Got- tingen, 1879 ; Carriere, Sehorgane der Thiere, Miinchen und Leipzig, 1885. Cope- podan (azygos) eye. Hartog, A. N. H. (5), x. 1882. Functions of facetted eye. Exner, Biol. Centralbl. i. 1881-2; Notthaft, Senckenberg, Abhandl. xii. 1881. Regeneration of Eyes in Astacus. Chantran, C. R. Ixxvi. 1873. Ear. Hensen, Z. W. Z. xiii. 1863. Olfactory hairs. Jourdain, Journal de 1'Anatomie, xvii. 1881 ; Leydig, Miiller's Archiv (Archiv f. Anat. u. Phys.), 1860. Development of Astacus. Rathke, Ueber die Bildung, &c., des Flusskrebses, Leipzig, 1829; Lereboullet, Mem. de PInstitut (Savans Strangers), xvii. 1862; Bobretzky, cf. Hofmann und Schwalbe's Jahresbericht, ii. 1873; Reichenbach, Z. W. Z. xxix. 1877, and Q. J. M. xviii. 1878 ; Schimkewitsch, Z. A. viii. 1885. On various larval forms, see Balfour, Comp. Embryology, i. 1880, p. 380, &c., and lit. cited. For Penaeus, add Brooks, A. N. H. (5), xi. 1883, and Faxon, Amer. Naturalist, xvii. 1883; see also Conn, on Significance of larval skin in Decapoda, Studies from Biological Laboratory, Johns Hopkins University, iii. 1884. 33. COMMON CRAYFISH (Astaciis fluviatilis), MALE, Dissected so as to show the nervous, circulatory and digestive systems in situ, and in the relations they hold to each other and to the external body walls. THE animal has been bisected longitudinally and the left half of the body removed together with its appendages, their muscles, the left lobes of the liver, and green gland. The series of nerve ganglia and connecting N 178 DESCRIPTIONS OF PREPARATIONS. commissures lies above the sternal elements of the various somites, the heart beneath the central area of the omostegite, whilst the reproductive anjd digestive organs occupy a position midway. The appendages of the right side are left in situ. The modified first and second pairs of abdominal limbs show that the specimen belongs to the male sex. The supra-oesophageal and twelve post-oral ganglia, six thoracic and six abdominal, are seen in profile. The great length of the commissures between the first-named ganglion and the first of the post-oral series or infra-oesophageal ganglion, whence the six pairs of appendages connected with the mouth are innervated, may be noted as corresponding with the elongated antennary sternum. In the abdominal series a slip of blue paper has been placed beneath the commissures uniting the fifth to the sixth ganglion. A white bristle has been introduced through the mouth into the oesophagus and stomach. This latter organ is large and extends forwards behind the eyes, but its anterior wall has been displaced a little backwards. It is divisible into a larger cardiac and a smaller pyloric portion, the pyloric situate just in front of a piece of blue paper placed under the hepatic artery. The lower edges of the liver lobes of the right side may be seen below the posterior cardiac region and extending backwards below the commence- ment of the intestine. The aperture of the common duct of the left liver lobes lies at the apparent termination of the hepatic artery, a projection in front of it marking the so-called pylorus. The part of the digestive tract into which the liver ducts open constitutes the mesenteron, and represents the archenteron of the embryo. The oesophagus and stomach are differen- tiations of the stomodaeum, whilst the intestine, which runs straight to the anus on the ventral surface of the telson, and is of great relative length, is formed from the proctodaeum of the embryo. The mesenteron and its appended glands are lined by endoderm, the oesophagus, stomach and intestine by ectoderm which secretes a chitinous cuticle. This cuticle is cast by the animal whenever it moults. The heart is a rounded body lying below the cardiac area of the omostegite. Of the arteries, to which it gives origin, the left hepatic passes obliquely downwards in front to gain the liver lobes, the sternal artery obliquely downwards behind to pass between the commissures uniting the third and fourth thoracic ganglia. Both these arteries have pieces of blue paper placed beneath them. Another piece of blue paper has been placed under the superior abdominal artery in the region of the fourth ab- dominal tergum. The left anterior lobe of the testis lies below and in front of the heart : the posterior azygos lobe beneath and behind it, whilst numerous coils of the right vas deferens come into view between the testis and intestine. The powerful and complex flexor muscle of the abdomen may be seen below the intestine and above the nerve chain extending into the thorax to be attached to the endophragmal skeleton (see COMMON CRA YFISH. 1 79 ante, p. 170) as far forwards as the oesophagus. It is by means of this muscle, by which the abdomen is suddenly flexed, that the animal executes its rapid backward darts through the water. The ambulatory thoracic limbs are employed for the slower movements of crawling ; and the muscles that move these limbs on the right side may be seen passing through the intervals of the endophragmal skeleton from their origins upon the epimera displayed in Preparation 36. 34. COMMON CRAYFISH (Astacusfluviatilis\ Dissected to show the heart and the origins of the chief vessels in situ. THE cardiac region of the omostegite and the adjoining region of the cephalostegite have been removed, as well as the tergal regions of the two first abdominal somites. The walls of an arterial pericardial sinus which surrounds the heart have also been removed, and that organ with the origins of the principal vessels has been consequently exposed. The heart is hexagonal in outline and compressed. Five arteries spring from its an- terior border. One in the middle line supplies the eyes and first antennae. Two others, one on each side this median artery, pass obliquely forward to the second antennae. And finally, two hepatic arteries, not seen here, spring one from each outer and inferior angle and pass downwards between the anterior testicular lobes and the intestine to the liver. One of them is shown in Prep. 33. A single dilated trunk takes origin from the posterior border of the heart. It divides at once into two branches. One of these, the superior abdominal artery, lies immediately above the intestine in the median dorsal line. A slip of blue paper has been placed under it. The other, the sternal artery, seen in Prep. 33, passes between the commissures uniting the third and fourth thoracic ganglia. It then divides, and one division passes forwards, the other backwards beneath the nerve-cord as the inferior abdominal artery. The muscular walls of the heart are pierced by six main valved inlets, which permit blood to enter but not to pass back. Two of these inlets are on the dorsal surface and are visible in this preparation about the middle region. Two other inlets are ventral, and the remaining two are placed, one on the right, the other on the left side. In addition to these main inlets, Dezso has described others of small size, four pairs on the dorsal and two pairs on the ventral surface. Six elastic alae cordis connect the heart to the non-muscular pericardium. Their main function is probably to antagonise the contractions of the heart itself. They may also serve to suspend it in the sinus, but the arteries are probably, in this as in other animals, the chief means by which it is kept in position. The polygonal aspect of the Decapod heart and the presence of N 2, T8o DESCRIPTIONS OF PREPARATIONS. arteries serve to distinguish it from the similarly unilocular and non- vasiform heart of some Entomostraca, while both these forms are in turn distinguished by their compressed shape from the vasiform structure found in Squilla and most Arthrostraca as well as in other classes of Arthropoda. The tissue of the heart consists, according to Dezso, of muscle cells. The striated substance is present only on one side of the cells as in Nematoda, and the cells are so grouped that their striated sides form the axis of a cord of cells. The nuclei are numerous and nucleolated. This histological condition must be regarded as a persistent state of what is an embryological phase in the development of muscular tissue in Arthropoda. The cords formed by the cells cross in all directions, and the heart-walls are spongy in texture. The inlets ' represent blood-spaces in the walls leading into a ventricular cavity.' The pericardium is non-muscular. It lies upon the heart, and consists of elastic connective tissue with a few scattered nuclei, and an outer layer of ordinary connective tissue. Bipolar ganglion cells, each in its own capsule, are to be found in the posterior half of the dorsal surface of the heart, frequently in groups of three or more. The muscular tissue of the heart contains myohaematin, as does that of the Lobster and the Crab, according to MacMunn. A medium-sized artery possesses three coats — a structureless intima, a middle coat composed of circular fibres, probably of connective tissue as they are not striated, and a homogeneous adventitia with numerous nuclei, regularly arranged. The intima disappears (?) in the smaller vessels. The middle coat is best marked in the large and medium-sized vessels, while the adventitia increases in thickness, and becomes both lamellate and fibrillate in the medium-sized vessels. Many of the arteries possess a sheath of cellular connective tissue, which is best seen in the superior abdominal artery. The capillary system is well developed, and forms networks : one of the easiest to demonstrate is the one on the surface of the supra-oesophageal ganglion. The capillary has a structureless wall with an oval nucleus here and there. Its cavity is very narrow hardly admitting a blood-corpuscle. The venous channels, according to Haeckel, have distinct walls composed of a thin plate of homogeneous nucleated connective tissue intimately united with the connective tissue coats of the various organs. They are always, according to him, well-defined channels, not irregular spaces. However this may be, they must, strictly speaking, be considered as constituting the peri-visceral cavity, rather than a system of vessels independent of that cavity. The venous space in the sternal canal is connected to the spaces which lead to the branchiae. The efferent branchial canals are distinct vessels, six in number, which ascend the walls of the thorax and open with widened apertures into the pericardial sinus. The blood-spaces of the branchiae are inter-cellular spaces, see p. 183. It is possible that, as in Phyllosoma, &c., the blood circulating in the branchiostegites may return to the heart without passing through the branchiae. The blood-corpuscles are colourless and amoeboid. The plasma contains haemocyanin (p. 112), and a red lutein, or lipochrome known as tetronerythrin, both of which are found also in the blood of other Crustacea, and the former in that COMMON CRAYFISH. iSi of many other animals as well. Tetronerythin is a pigment commonly distributed in the animal kingdom and is found in various tissues, e.g. in the integument and muscles. It has been supposed by Merejkowski to have a respiratory function, but this is doubtful. The yellowish-red granules, seen sometimes in the blood-corpuscles of Decapod Crustacea , are perhaps formed of it. Heart. Bela Dezso, Z. A. i. 1878. Circulatory System. Astacus, Krohn, Isis, 1834. Stomatopoda, Schizopoda and Decapoda. Claus, Arb. Zool. Inst. Wien, v. 1884. Structure of 'vessels ; &c. Haeckel, Mtiller's Archiv (Archiv f. Anat. und Phys.), 1857- Blood of Decapod Crustacea. Halliburton, Journal of Physiology, vi. 1885. Haemocyanin and Tetronerythrin, ibid. On the latter see also Merejkowski, Bull. Soc. Zool. France, viii. Haemoglobin in Crustacea, see lists in Halliburton, op. cit. ; also van Beneden on Lernanthropus, Clavella, and Congericola. Bull. Acad. Roy. Sc. Belgique (2), 49, 1880; Id. Z. A. iii. 1880. Myohaematin. MacMunn, P. R. S, xxxix. 1885. 35. COMMON CRAYFISH (Astacusfliwiatilis), Dissected so as to show its digestive, reproductive, and respiratory systems in situ. THE greater part of the tergal region of all the segments of the body has been removed, together with the heart and its vessels and in the abdomen the thin stratum of extensor muscles. The stomach occupies a central position anteriorly, and is clearly divisible into a wider cardiac portion in front and a narrower pyloric portion behind. An arcuate plate, the 'cardiac ossicle,' crosses the cardiac portion at the point of greatest width, and receives the insertion of the major part of the anterior gastric muscles which spring from the base of the rostrum, A pyloric ossicle crosses the pyloric portion of the stomach in a similar manner, and gives attachment to the posterior gastric muscles which take origin posteriorly from the carapace.. The other stomachal ossicles can only be studied when the stomach is properly opened. To the right side of the pyloric portion of the stomach is seen the end of the adductor mandibulae muscle separated from its attachment to the carapace ; and behind it, as well as to either side, are the two lobes of the liver. On the left side, in front of the liver, is to be seen a small portion of the sac of the green gland. The paired anterior lobes of the testis lie in the middle line between the liver lobes. The azygos posterior lobe overlies the intestine. At the spot where these three lobes unite the right and left vas deferens take origin as slender tubes, the calibre of which rapidly widens. They are disposed in many convolutions which intrude some way into the abdominal cavity before they turn downwards, to open on the coxopodites of the last pair of thoracic limbs. The intestine takes a jS2 DESCRIPTIONS OF PREPARATIONS. straight course, as in all Crustacea, to the anus. The branchial cavity is just exposed on the left side, but largely on the right by the removal of a great part of the branchiostegite and of the united thoracic epimera which separate the viscera from the branchial cavity on the inner side. The branchiae are seen lying in the cavity thus exposed. They may be distinguished as podo-, arthro-, and pleuro-branchiae. Podobranchiae are attached one to each coxopodite from the second maxilliped to the fourth thoracic limb inclusive. The arthro-branchiae are divisible into an anterior and posterior, or external and internal set, and they are attached to the membranes uniting the coxopoditeS to the body. The second maxilliped bears an anterior arthrobranch only, the third maxilliped and the thoracic limbs to the fourth inclusive possess both sets. The fifth thoracic limb bears no branchiae, but a pleuro-branch is attached to the epimeron of its somite. There are also two rudimentary pleuro-branchs, one on the third, another on the fourth, epimeral regions. The number and arrange- ment of the branchiae varies much among the Decapoda. A podobranchia consists of a broad basal portion convex posteriorly and in- feriorly, beset with setae and articulated to a coxopodite. The stem of the branchia bends at right angles to this base and divides into an apical plume and a lamina. The free extremity of the plume is simple and filiform. At its base it gives origin to a number of cylindrical branchial filaments. Similar filaments spring also from the outer and anterior surfaces of the stem itself. The lamina originates about the middle of the stem. It is folded upon itself. The edge of the fold looks forwards, and the leaves of the fold are one external, the other internal, the latter extending downwards .towards the base of the stem to a distance greater than the former. The folded edge of one podobranchia fits into the space between the leaves of the foregoing podobranchia. The surface of each leaf is plaited longitudinally ten or. twelve times. The edges and surfaces of the leaves, especially of the plaits, are beset with small elevations each bearing a single minute hooked seta. The epipodite (so-called) of the first maxilliped represents the base, stem and lamina of a podobranch. It is slightly folded, but the edge of the fold is posterior, whilst the internal edge of the lamina and its posterior surface bear hooked setae. The structures known as coxopoditic setae are long and slender filaments arising close to the bases of the podobranchiae, with acute apices and their terminal portions beset with foliaceous scales. They ascend vertically, lying among the branchiae, and it is suggested by Professor Huxley that they exclude parasites. The Crayfishes (Paras- tacidae) of the S. Hemisphere differ from those (Potamobndae) of the N. Hemisphere in having (i) the laminae of the podobranchs rudimentary; (2) some at least of the branchial filaments, the setae of the stem, and the coxopoditic setae terminally hooked; and (3) a few branchial filaments upon the epipodite of the first maxilliped. An arthro- and pleuro-branch have a structure similar to the base, stem and apical plume of a podobranch. The anterior of the two rudimentary pleurobranchiae is often a mere papilla, but the posterior resembles in structure a branchial filament. Of the two varieties (? species) of A. fluviatilis, the A. nobilis differs from A. torren- COMMON CRAYFISH. 183 Hum in possessing three instead of two rudimentary pleurobranchiae. The Crayfishes of the S. Hemisphere, with the exception of Astacoides from Madagascar, have four functional pleurobranchiae. It is remarkable that these organs are entirely absent in Cambarus (the only other genus besides Astacus of Potamobi'idae\ which is dis- tributed East of the Rocky Mountains from the Great Lakes to Guatemala, and is found also in Cuba. Inasmuch as the branchial filaments are all cylindrical in the Crayfishes, the branchial plumes are tricho-branchiae. The Crayfishes in this respect agree with all Decapoda Macrura except the genera Gebia and Callianissa, the Prawns, Shrimps, and Mysidae. The branchiae of the last-named are either absent or rudimentary, In the other Macrurans mentioned, as in the Hermit Crabs and Brachyura, the filaments are replaced by lamellae, and the branchiae wet phyllo-branchiae. Claus has pointed out (Arb. Zool. Inst. Wien, vi. 1885, pp. 39-47) (i) that the relation of the coxopodite to the body-walls is by no means a constant one in Crustacea, and the arthrodial -membrane has limits often ill-defined; (2) that in the branchiferous Schizopoda (Euphausia, &c.) the branchiae, which are all trifid, form a row of podobranchiae ; (3) that in the larva of Penaeus there are three rows of branchial rudiments, which he terms distal, middle and proximal. The last is a double row, but one of the elements appears at a later period than the other. The distal rudiment on each limb represents the podobranch + epipodite(= lamina) : the middle and the first proximal, the anterior and posterior arthrobranchs, and the second proximal the pleurobranch, in Astacus. It may be noted that the primitive distal rudiment from the second maxilliped to the third thoracic foot inclusive pushes out a basal bud. The bud becomes a branchia, afterwards lost on all the feet save the second maxilliped, while the primitive rudiment forms an epipodite. The three sets of rudiments which first appear belong distinctly to the original basal joint of the limb, but the arthrodial membrane develops in such a manner that they come to lie subsequently on the coxopodite, the arthrodial membrane, and the epimera. The cavity of the branchial stem is divided by a septum into an outer and inner channel communicating at the apex. The latter is continuous with one of the six blood passages or ' branchial veins ' which open into the pericardial sinus with widened mouths. The former is continuous with bloodvessels coming from the sternal sinus lodged in the sternal canal. Each branchial filament is similarly divided by a septum incomplete at the apex. According to Haeckel the blood- spaces are intercellular spaces, or lacunae, and not true vessels. The tissue is spongy. The cells are pyriform, one end pointed and united to the cuticula, the other swollen and nucleated, and attached to other cells (cf. Haeckel, Arch. Anat. u. Phys. 1857, p. 554). The oesophagus and stomach (=stomodaeum), and the intestine (=procto- daeum) are lined by a chitinous coat. This coat consists of a superficial delicate cuticle similar to that of the carapace, and a deeper lamellate layer, sometimes pene- trated by pores, especially where it is much thickened. There are numerous setae in the stomach, principally in its pyloric portion, and in the intestine. They are of two kinds, hollow hairs, similar to those of the carapace, very plentiful in the pyloric portion, and solid continuous processes. Minute ridges secreted by the chitino- genous cells, and corresponding three or more to a single cell, occur in the intestine jg4 DESCRIPTIONS OF PREPARATIONS. where the chitinoid coat is divided into areae corresponding, as on the carapace, to the individual cells. The cuticular coat of the stomach ends at the pylorus with five projecting processes ; that of the intestine commences with six elevations pro- longed into ridges which traverse the tube in a spiral fashion. Beneath the chitinoid coat is the single layer of chitinogenous, or ectoderm cells, large in size, and then a fibrous membrane followed by a layer of cellular connective tissue which incloses muscle fibres, both longitudinal and circular, in the stomach and intestine. The chitinoid layer of the stomach is thickened in the dorsal and lateral walls to form certain ossicles which, according to Vitzou, have the same structure as Jthe carapace. Some of these ossicles have simply a supporting function, others constitute the 'gastric mill.' The latter are named and arranged as follows. There is a cardiac ossicle crossing the cardiac region transversely, articulating laterally with a ptero-cardiac piece, and extending forwards into a softer disc upon which the anterior gastric muscles are principally inserted, and backwards into a narrow uro- cardiac piece which is produced inferiorly into two accessory or cardiac teeth, rudi- mentary in the Lobster. A pyloric ossicle crosses the pyloric region transversely and dorsally, and gives insertion to the posterior gastric muscles. It articulates in front with a prepyloric ossicle which is bent downwards so as to form with it an acute angle backwards way, and articulates in turn with the urocardiac piece (supra). Close to this articulation the prepyloric ossicle is produced into a bifid median tooth, single in the Lobster. The pyloric ossicle articulates laterally on each side with a zygocardiac ossicle lying in the walls of the cardiac region. This ossicle articulates at its outer extremity with the corresponding extremity of the ptero-cardiac ossicle (supra] of its own side. Its inner extremity bears the great serrated lateral tooth. Just below the anterior end of this tooth projects an infero-lateral tooth borne by the 'lateral cardiac piece' of Milne-Edwards, one of the supporting bars of the stomachal walls. The stomach possesses extrinsic and intrinsic muscles. The former set includes the gastric muscles above mentioned, as well as the anterior lateral, the pos- terior, superior and inferior dilators. The latter set includes various muscles. One system of fibres unites the ptero- and zygo-cardiac ossicles. The remainder, according to Mr. T. J. Parker, act as constrictors, especially a layer which embraces the pyloric region. This region has its cavity, more particularly in the posterior part, narrowed by the bulging inward of its side-walls and the development of a median ventral ridge. The surfaces of these parts are beset with setae and form a most efficient ' filter.' There is on either side of the stomach, at the entrance of the oesophagus, a round white spot caused by the presence of a flattish papilla having the same structure as the rest of the wall of the stomach. Forty days before a moult in the adult, or for a shorter period in the young according to age, the chitinogenous cells of this papilla develope a number of minute knobbed processes which raise the overlying chitinous cuticle and eventually break up into corpuscle-like bodies. Beneath the cuticle, and between the ends of the chitinogenous cells and their pro- cesses, lamellae of calcified organic matter are laid down forming the gastrolith. The lamellae are pierced by pores. Their substance consists of Calcium carbonate, with a small admixture of phosphate, and of organic substances partly soluble in water, partly insoluble, and perhaps chitinoid in nature. During the development of the gastrolith the papilla becomes more prominent and changes its shape. The fully COMMON CRAYFISH. 185 formed stone is slightly concave and smooth on its stomachal or inner face, convex and marked by ridges on its external face. The lamellae of this face are the last formed and the hardest. The stone is eventually cast off into the stomach, previous to the moult, and ground down. It is to be regarded as a cuticular structure forming a storehouse of calcareous matter preliminary to the moult, which is not effected healthily unless the gastroliths are previously and well developed. If they fail to be developed properly, the animal, so it is said, usually dies. The cuticular linings of the stomach and intestine are thrown off and regenerated at each moult. The parts of the gastric mill are said to be first of all broken up in all Decapoda. The remaining parts are moulted entire. Tubular glands occur in Astacus, as in all Decapoda^ in the walls of the oesophagus and the terminal dilated portion of the intestine. Glands have also been found by Braun on certain of the mouth-appendages, e. g. in Astacus, to the number of twenty on the ' median flattened ' joint of the first maxilla, ' opening on its outer surface,' and on the 'lingula' (=metastoma). They resemble the glands of the branchial cavity (see ante, p. 166) rather than those of the digestive tract. The mesenteron (=archenteron), or median portion of the digestive canal, is extremely short as in all higher Crustacea. It has a dorsal caecum, and receives on each side the common liver duct. Its cells form no cuticle. The liver, or hepato- pancreas, derived as an outgrowth from the mesenteron, consists of a right and left gland, each consisting of an anterior, a median, and a posterior lobe. The secretory portion consists of innumerable tubular caeca lined by cells. These cells in all Crustacea possess a fringe of fine hairs affixed to a membrane, which is probably porous. The protoplasm is distinctly striated, especially in Isopoda. In Decapoda there are two kinds of cells, one that secretes coloured drops of fat and contains masses of small globules; the other, fine and coloured granules (= ferment-cells). In the Isopoda there is one kind of cell only forming both fat and fine granules, whilst the globules are absent (Frenzel). There are cells in reserve destined to replace those which are destroyed. The secretion is acid in reaction and con- tains, in many instances, cholesterin, in the Crayfish haematin, as well as a diastatic, peptic, tryptic, and possibly a fat-destroying ferment. Glycogen has been found in the gland (cf. Vitzou, A. Z. Expt. x. p. 554). The ' green gland,' or renal organ, opens on the inner side of a ventrally-placed papilla upon the basal joint of the second antenna (p. 169). The duct, lined at its commencement by chitinous cuticle, widens out into a thin walled sac which, to- gether with the gland itself, lies in the thorax at the base of the antenna, the sac dorsally to the gland. The latter forms a disc-like body composed of a tube coiled upon itself and divisible into three sections : (i) a long whitish tube which opens into the sac ; (2) a green-coloured tube opening into (3) a triangular yellow- brown lobe. The coils are so disposed that the third section of the tube lies centrally and dorsally ; the green section forms the outer circumference, and the white section lies between the two others. The bloodvessels are derived from the antennary and sternal arteries, and are especially numerous on the terminal lobe. Nerves derived from the supra- and the infra-oesophageal ganglia are distributed to the excretory sac. This sac and the tube are lined throughout by a single layer of epithelium supported by a fine structureless tunica propria. In the green section the cells have a striated cuticle, and the protoplasm is striated as in the tubuli DESCRIPTIONS OF PREPARATIONS. contort! of the Mammalian kidney. Guanin and uric acid have been stated to occur in the gland. Grobben compares the terminal lobe with its rich vascular supply to the Malpighian capsule ; the remaining sections to the tubuli uriniferi of the Vertebrate kidney. The shell gland of the Phyllopoda and Copepoda consists of a similar terminal lobe and tube, but it opens on or close to the second maxilla. The ovary is a trilobed gland like the testis, but the fissures between the lobes are not so deep. The oviducts originate as do the vasa deferentia, but they are wide, short, and straight. The ovary is lined by a delicate cuticle. The ovum is developed by the growth of a single cell out of a small mass of cells. It has a vitelline membrane, and when ripe, is set free into the cavity of the ovary. It is fecundated externally to the body, and is suspended to the feet, sterna, &c., of the abdomen during the development of the embryo. For the mode of suspension and the cement glands, see ante, p. 166. The testis is a tubular gland. The tubes branch, and the ultimate branches end in a number of short stalked vesicles varying in size according to the state of their contents. The lining cells multiply and are differentiated into spermatozoa. These bodies are disc-like structures with a number of slender curved rays attached to the circumference of the disc. The structure of the disc is still somewhat obscure. The spermatic fluid is milky, and contains a viscid substance which agglutinates the spermatozoa into thread-like spermatophores. The sperm is shed through the channels of the first pair of abdominal limbs. The second pair are worked to and fro in these channels as if to keep them clear. The male throws the female on her back and deposits the sperm on the ventral surface of the last pair of swimmerets and on the thoracic sterna round the oviducal apertures, parts which are approximated during oviposition. Branchiae. Huxley, P. Z. S. 1878; Glaus, Arb. Zool. Inst. Wien, vi. 1885. Digestive tract-, structure of walls, glands, and epithelium. Vitzou, A. Z. Expt. x. 1882; Braun, Arb. Zool. Zoot. Inst. Wurzburg, ii. 1875, iii. 1876-7; Frenzel, A. M. A. xxv. 1885. Ossicles of stomach and muscles. Mocquard, A. Sc. N. (6) xvi. 1883 ; T. J. Parker, Journal Anat. Phys. 1877 ; Albert, Z. W. Z. xxxix. 1883. Working model of gastric mill. Roth, Nature, xxi. 1879-80. Pyloric filter. Huxley, The Crayfish, p. 58. Gastrolith. Braun, Arb. Zool. Zoot. Inst. Wurzburg, ii. 1875 ; Chantran, C. R. Ixxviii. and Ixxix. 1874; its chemical composition, Dulk, Miiller's Archiv (Arch. f. Anat. und Phys.), 1835. Structure of liver. Frenzel, Mitth. Zool. Stat. Naples, v. 1884. Its Chemical action. • Krukenberg, Untersuch. Phys. Inst. Heidelberg, ii. 1882. Its Colouring matter. MacMunn, P. R. S. xxxv. 1883. Green gland. Wassiliew, Z. A. i. 1878; Grobben, Arb. Zool. Inst. Wien, iii. 1 88 1. For guanin and uric acid in it, cf. Griffiths, P. R. S. xxxviii. 1885. Testis. Lemoine, A. Sc. N. (5) x. 1868; Rougemont, Organes ge'nitaux, &c., Astacus, Bull. Soc. Sc. Nat. Neuchatel, ii. 1880 (not seen). Spermatozoa. Grobben, Arb. Zool. Inst. Wien, i. 1878; cf. note in Huxley, The Crayfish, p. 354. Spermato- genesis. Sabatier, C. R. c. 1885 ; Hermann, C. R. xcvii. 1883. Fecundation. Chantran, C. R. Ixxi. 1870; Ixxiv. 1872. Structure of ovum. Waldeyer, Eierstock und Ei, Leipzig, 1870, p. 85; cf. general account \>y Ludwig, Arb. Zool. Zoot. Inst. Wurzburg, i. 1874. Cement glands. Braun, Arb. Zool. Zoot. Inst. Wurzburg, ii. 1875 ; iii. 1876-7. Their secretion and fixation of ova. Lereboullet, A. Sc. N. (4), xiv. 1860. COMMON CRA YFISH. 1 87 36. COMMON CRAYFISH (Astacus fluviatilis), MALE, Dissected so as to show its nervous system. THE supra-oesophageal ganglion and the twelve post-oral ganglia of the adult Crayfish, of which six belong to the thorax and six to the abdomen, have been displayed by the removal of the whole tergal region of the body, of the viscera of organic life, and the endophragmal skeleton in the thorax. The oesophagus, through which a black bristle has been passed, and a small terminal portion of the intestine remain in situ. The supra-oesophageal ganglion was seen by Rathke to be made up of two rudiments in the embryo ; of which the posterior, or the one placed nearest to the mouth, was the larger and supplied the first and second antennae. The ganglion itself in the adult gives off nerves to the eye, the eye-muscles, to the integument of the head, and the first and second antennae, besides furnishing two azygos nerves, one anterior, the other posterior, to the stomato-gastric nerve. The superior and inferior roots of this nerve (infra}, together with the two commissures to the sub- oesophageal or first post-oral ganglion, are seen passing over a piece of blue paper placed under them in front of the oesophagus. The left end of this blue paper rests on the secreting portion of the left antennary or green gland. The infra-oesophageal ganglion is the largest of the post-oral series and innervates no less than six pairs of appendages, viz. the mandibles, the two pairs of maxillae, and the three pairs of maxillipeds or foot-jaws. In the developing Crayfish, as shown by Rathke, this mass is represented by six pairs of white specks. It is followed by five thoracic ganglia, which remain distinct and correspond in the adult as well as in the embryo of Macrurous Decapods to the five pairs of thoracic feet. Each ganglion is connected to its successor by two longitudinal commissures, showing the primitive bilateral composition of the chain. The commissures between the third and fourth ganglia are widely separate for the passage of the sternal artery seen in Preparation 33. The fourth and fifth ganglia are approximated. The first abdominal ganglion is some distance behind the last thoracic, and all the six abdominal ganglia are equi- distant one from the other. A slip of blue paper has been placed under the third and fourth, and another under the commissure to the last of the series. The commissural cords are clearly double. The third ganglion is seen to give off a pair of nerves on each side, while another pair springs from the commissures immediately behind the ganglion. The anterior nerve on each side goes to the swimmerets ; the posterior and the com- missural pair to the muscles of the same, i. e. the third somite. All the abdominal ganglia resemble the third pair in these points with the DESCRIPTIONS OF PREPARATIONS. exception of the terminal ganglion, which may be seen to give off a large number of nerves. Accurate investigations have shown that of these there are five pairs and one posterior, median and azygos nerve. This azygos nerve supplies the termination of the intestine ; and the nerves to either side of it, i. e. the fifth or innermost pair, are destined for the telson. The two outermost pairs of nerves, i. e. the first and second, go to the exopodite, and the third pair to the endopodite of the last enlarged pair of swimmerets. The fourth pair, according to Krieger, supplies muscles in the same manner as do the commissural pairs of nerves corresponding to the five foregoing ganglia. The infra-oesophageal and the thoracic ganglia lie in the sternal canal formed by processes of the apodemata. The roof of this canal has been cut away to expose the nerve chain, but parts of the apodemata may be seen on either side of it in the shape of vertical tubular processes. The two eyes with their stalks, the bases of the first and second antennae, are shown by the removal of the overhanging rostrum. The surface of the basal joint of each first antenna thus exposed is the one that contains the aperture into the auditory sac. The aperture itself is concealed by setae. The summits of the branchial plumes are well seen in this specimen between the branchiostegite and the epimera of the thoracic somites. From the internal aspect of the epimera, the muscles which move the limbs upon the thorax are seen trending downwards and bifurcating as they pass between the sections of the endophragmal skeleton to their insertions. The sub-oesophageal ganglion gives off six inferior and four superior or dorsal nerves. The six inferior are destined for the mouth-parts. The mandibular nerve accompanies the commissures round the oesophagus for a certain distance. The last nerve, which goes to the third maxilliped, arises at some distance behind that for the second maxilliped, and the part of the ganglion from which it springs has a certain amount of distinctness or individuality. Of the four superior nerves the first is of considerable size and innervates the scapho-gnathite. The three remain- ing nerves are fine, and their destination unknown. Each of the five thoracic ganglia gives off two pairs of nerves : an anterior large pair destined for the limb and the gills belonging to the somite, and a posterior fine pair destined for the corresponding thoracic muscles. The median azygos nerve given off by the last abdominal ganglion divides, ac- cording to Lemoine, into two branches, a posterior anal branch and an anterior intestinal branch. The latter subdivides into (i) a branch to the anal end of the intestine ; (2) a branch which courses along the ventral surface of the intestine, to which it gives twigs from spot to spot ; and (3) a branch which turns round the in- testine and runs upon its dorsal aspect. Lemoine traced these two last-mentioned branches as far forwards as the genitalia. COMMON CRA YFISH. 1 89 The nerve-factors which make up the stomatogastric system are derived from two sources, from the supra-oesophageal ganglion itself, and from the commissures connecting it to the infra-oesophageal ganglion at the spot where these commissures come into contact with the walls of the oesophagus. The nerves derived from the first-named source are two, a superior azygos nerve and an inferior azygos nerve. The former has a small ganglion close to its origin ; it is short and runs upwards, i.e. dorsally. The latter is long and runs backwards and downwards. Two small ganglia, the mandibular or oesophageal ganglia, recently investigated by Krieger, give origin to the second set of factors named above. They are semi-oval and lie on the ventral side of the commissures, and from each of them spring at least three nerves. One is external and bends down, branching on the oesophagus ; the two others, the superior and inferior roots, are internal and pass forwards between the commissures. The inferior roots unite together to form a single trunk, to which the superior roots then unite. The single nerve thus formed is joined by the inferior azygos nerve from the supra-oesophageal ganglion, and constitutes a median nerve which runs upwards in front of the stomach, giving off one after another three branches to that organ (Lemoine). It then unites with the superior azygos nerve l from the supra- oesophageal ganglion. The single trunk formed by this union bends round the stomach on to its dorsal aspect. Close to the point of union it gives a nerve to the stomach and twigs to the anterior gastric muscles through which it passes. It then forms the stomatogastric ganglion, from which spring two nerves, an upper, the car- diac nerve of Lemoine, lodged in the integument and going to the heart, and an in- ferior or gastro-hepatic which lies on the dorsal wall of the stomach. The latter passes backwards and divides posteriorly into the terminal branches. Between these and the stomatogastric ganglion is a slight swelling from which rise the two lateral branches. Both lateral and terminal branches pass downwards. The latter supply the posterior gastric muscles. They eventually distribute themselves to the liver so-called and various stomachal muscles, and anastomose both with the lateral branches and faepostero-lateralrvetve which arises from each oesophageal commissure dorsally to the oesophageal ganglion, and passes upwards on the oesophagus. These various nerves give off numerous branches which have recently been investigated in detail by Mocquard in various Decapoda. The ganglia consist of central masses of Leydig's ' punkt-substanz ' formed by dense networks of fine nerve-fibrils, and external masses of ganglion cells, varying in size. The ganglion cells themselves differ in the same respect : the smallest possess but little protoplasm. Each cell is contained in a connective tissue capsule. Their processes, though numerous, originate in most instances from one surface or pole. The nerves are tubular, and, according to Krieger, consist of an external sheath and homogeneous fluid contents, but Freud states that there are delicate fibrillae im- bedded in this homogeneous substance ; and he traces the same distinction, viz. a homogeneous matrix and imbedded fibrillar network, in the bodies of the ganglion cells. The nerves branch repeatedly. There is a tough elastic perineurium or common investment, composed of decussating fibres and covered within and without by cellular connective tissue. Histology of nerve-cord, nerves, &c. In Astacus, Krieger, Z. W. Z. xxxiii. 1880 ; 1 This root is not mentioned by Mocquard, but is figured by both Krieger and Lemoine, and I have found it myself more than once. 19o DESCRIPTIONS OF PREPARATIONS. Freud, SB. Akad. Wien, Ixxxv. Abth. 3, 1882. In Decapoda, Yung, A. Z. Expt. vii. 1878. Stomatogastric system with figures, and posterior intestinal nerves. Lemoine, A. Sc. N. (5), ix. 1868; Mocquard, A. Sc. N. (6), xvi. 1883. 37. COMMON STARFISH (Asterias or Asteracanthion rubens), Dried. THE animal consists of a central disc, which is prolonged into five lobes, the so-called arms, rays, or radii. The interval or part between each radius is known as an interradius. Two surfaces may be distinguished : — one flat or somewhat concave, the ventral, oral, actinal, or ambulacral surface ; the other convex, and termed the dorsal, aboral, abactinal, or anti-ambulacral surface. In the centre of the ventral aspect of the disc is the membranous peristome with the mouth. Five sets of spines, the mouth-papillae, project over this area interradially, giving it a pentagonal appearance : and there radiate from it the five avenues or ambulacral grooves, one to each ray, which lodge the locomotor feet and hence give this aspect of the animal the name of ambulacral. The feet in question have been removed from two of the grooves for a short space, but are left in situ and in a dried condition elsewhere. Examining the exposed part of each groove attentively, it is seen to be formed by two series of narrow parallel ossicles — the so-called ambulacral or vertebral ossicles, the long axes of which are at right angles to the axis of the ray. The ossicles of one side of the groove are inclined at an obtuse angle, open ventrally, to the corresponding ossicles of the other side ; and their dorsal ends are articulated moveably together. The summit of the angle is median. The groove lodges the radial water- vascular vessel, the inferior transverse vertebral muscles, the radial perihaemal spaces and bloodvessels, and the radial nerve-cord, with the feet. Between the ossicles are a series of pores, one pore between each pair, formed by the apposition of two grooves in adjoining ossicles. The ampullae of the feet which lie on the dorsal side of the ossicles communicate through these pores with the ventrally placed feet. The two first pores lie in the same straight line, while the succeeding, to very near the tips of the arms, are arranged in a zig-zag fashion, being alternately near to, and remote from, the axis of the ray. Hence there appear to be four rows of pores and four rows of feet to correspond. In the majority of Asteroidea, however, the pores retain a straight linear arrangement for the whole extent of the grooves. The edges of a groove are bordered immediately by a series of fine moveable spines, borne by the adambulacral ossicles (infra). In this COMMON STARFISH. 191 specimen there are three rows of such spines, but in many instances there are only two. At the oral end of each groove they form the mouth-papillae above-mentioned, the spines of one side in one groove meeting the spines of the adjoining side of the contiguous groove interradially. Externally to this series of spines comes another series in triple row of stout spines, fixed like all the other spines of the body. They are borne by the median set of interambulacral ossicles (infra}. This series of spines, and the series of moveable spines, both extend to the tip of the ray, ceasing at a spot where a circlet of spines denotes the position of the eye-speck and terminal feeler or tentacle in the living animal. A third series of stout spines in a single row, borne by the inferior marginal ossicles (infra), borders the ventral aspect of the ray on either side. Turning to the dorsal or aboral surface the perisoma or integument with its network of calcareous ossicles and membranous soft intervals may be first noticed. The spine-bearing ossicles form more or less regular lines parallel to the axes of the rays. One line in the middle must be noted particularly. Attached to the soft intervals and at the bases or tips of the spines may be seen scattered pedicellariae. Other pedicellariae are grouped round and on the series of moveable spines of the ventral surface. These two sets of pedicellariae differ remarkably from one another. Both however are to be regarded as modified spines, not zooids, like the polymorphic aviculariae and vibracula of Polyzoa. (See Flustra, Preparation 48, post.) In one of the interradii is a circular calcareous plate, the madreporic tubercle or madreporite, the surface of which is marked by grooves radiating from the centre. By the removal of the perisoma from the disc, it may be seen that a canal with calcareous walls — the stone-canal, — curved like the letter S, leads from this plate to the ventral surface, where it opens into the circum-oral water vascular vessel. The two rays, one on either side the madreporic plate, constitute the bivium as ordinarily defined ; the three remaining rays the trivium. The central one of the three lies opposite the madreporic interradius and is often spoken of as the anterior ray. The perisoma has been removed from the dorsal surface of three rays, and the ambulacral ossicles can be seen from their dorsal aspect. Their median ends form a prominent vertebral ridge with median furrow. Each first ambulacral ossicle is large, broad, and pointed medianly, and projects over the peristome radially. It appears to be formed by the fusion of two ossicles. The pore is large. At the outer end of this ossicle and to the outer side of the pore is an enlarged first adambulacral ossicle, which is, however, smaller than the ambulacral. This adambulacral with its fellow adjoining the same interradius carries the oral papillae or spines which project over the peristome interradially. In the majority of Starfish the first adambulacral itself projects interradially over the peristome beyond 1 92 DESCRIPTIONS OF PREPARATIONS. the corresponding ambulacral ossicle and with its fellow in the adjoining ray forms two teeth. The ambulacral is in this case a support for the teeth. The first type of mouth is termed ambulacral, the second adambulacral. The former appears to be characteristic of those Starfish in which the ambu- lacral pores are arranged in zig-zag-, the latter of those in which they maintain a linear arrangement. A small plate, somewhat indistinct in this specimen, overlies interradially each set of first adambulacrals. It is known as the odontophore, and appears to be homologous with the oral plate of many other Echinoderms. See the general account of the Phylum. A single row of minute adambulacral ossicles articulating immediately with the outer ends of adjoining ambulacrals can be made out in nearly the whole length of the arm ; as well as the transverse rows of five inter- mediate ossicles, which unite them with a longitudinal row of inferior marginal ossicles bordering the ventral aspect of each arm. The median intermediate ossicle of the five is enlarged and spine-bearing, and is con- nected to the intermediate ossicle in front and behind. In some Starfish, e. g. Astrogonium, there is a series of well-developed dorsal superior marginal ossicles. The perisoma or integument consists of two layers, an outer and an inner, between which exists a system of irregular channels which are ultimately continuous with the system of perihaemal spaces. The calcareous ossicles, spines and pedi- cellariae, belong to the outer layer with the single exception of the ambulacral ossicles which belong to the inner one. The outer layer passes across the ambu- lacral groove from side to side, inclosing a space between itself and the ossicles. In this space are lodged (proceeding from the dorsal to the ventral wall) the water- vascular radial canal, the transverse ambulacral muscles, the right and left- perihaemal space with the bloodvessels, and the nerve-forming ectoderm. The outer surface of the perisoma is covered by a ciliated epithelium (ecto- derm) which possesses a distinct cuticle pierced by pores for the cilia. The consti- tuent cells are (i) supporting cells, (2) gland cells, (3) sense-cells. At the base of the epithelium is a network of nerve-fibrils with ganglion cells. In the ambulacral nerve and circum-oral nerve-ring the supporting and sense-cells are of great length, and the nerve-fibre layer much thicker, and the fibres parallel to one another. These structures are continued on to the terminal feeler or tentacle, the first formed tube- foot of the arm, as well as on to the paired feet; but in the latter the nerve-fibres are arranged in longitudinal bundles, and gland-cells occur plentifully on the terminal sucking disc. The eye-speck is placed on the ventral surface of the feeler and is com- posed of several eyes, the number increasing with age. Each eye consists of a conical depression in the epithelium over which the cuticle passes uninterruptedly. The walls of the depression are formed of sense-cells, some of which contain pigment, and its central cavity is filled with a clear liquid. If the radial nerves are divided close to the ring, the animal loses the power of coordinating the movements of the several arms : and if the eye-specks are removed, it ceases to react to light. The pedicellariae in the Asteroidea possess only two blades with the exception COMMON STARFISH. 193 of Luidia, where they possess three as in Echinoidea. But the stalk of the Asteroid, unlike that of the Echinoid pedicellaria, is formed entirely of soft structures. There are in the Asteroidea, which possess four rows of ambulacral feet, two kinds of pedi- cellariae. Both agree in having three calcareous pieces, a basilar piece bearing two blades opened and shut by muscles. But in one kind which occurs chiefly isolated the blades are articulated opposite to one another, in the other the blades cross one another at a certain point like the blades of a pair of scissors. The latter are termed crossed pedicellariae. They occur in rings or semi-rings upon the spines. In other Asteroidea there is a form of pedicellaria termed ' valvulate,' in which the blades are broader than they are long. The pedicellariae are used to take hold of objects such as algae until the feet can be applied. The apical system of plates is not traceable in this nor in the majority of adult Starfish. It is disguised by the formation of new plates and ossicles during growth. See the general account of the Echinodermata and Asteroidea. The ossicles are developed from a calcareous network as is usual in Echinodermata, and the calcareous matter is chiefly Calcium carbonate. The series of ambulacral ossicles end at the tip of each arm with a terminal plate which supports the tentacle on its ventral surface, and is formed in the young Starfish at an early period. All new ossicles and plates are added between it and already existing ossicles and plates. A series of plates on the dorsal surface con- nects it to a radial plate of the apical system in the young Starfish and some adults. And when the radial becomes indistinguishable, this median row of plates may remain conspicuous. In Asterias it is often well marked, but the ossicles are similar to the other ossicles of the general perisoma which are arranged in linear series. According to Viguier, ten muscles correspond to each pair of ambulacral ossicles : four vertical muscles (two on each side) uniting the ambulacral and adanv bulacral ossicles : four longitudinal (two on each side, one superior and one inferior), and two transverse muscles. One of the transverse muscles is ventral and deepens the groove when it contracts : the other dorsal and antagonistic to the ventral. The dermal branchiae of the dorsal surface are delicate contractile tubular processes of the perisoma. The madreporite contains tubular ciliated canals radiating conformably with its superficial furrows. These canals send up minute vertical non-ciliated tubes which open in the furrows by ciliated funnels. Of these openings A. rubens has about two hundred. The radiating canals join the stone-canal. This canal has at its dorsal end an ampulla, which is seven-lobed. Its cavity is partially divided by a septum, but in many Asteroidea it is completely broken up into many tubes. To the circumoral water-vascular ring are attached in pairs nine racemose vesicles or Tiedemann's bodies. A tenth is missing, and the stone canal opens where it should be. The bodies consist of branched canals lined by cubical cells. The radial vessels spring from the circumoral. They give off laterally, corresponding to the interval between two ambulacral ossicles, the branches for the feet. These open into the bases of the feet in such a manner that water cannot regurgitate. A branch from the base of every foot passes through a corresponding ambulacral pore and swells into an ampulla on the dorsal aspect of the series of ossicles. Longitudinal muscles exist in the walls of the ampullae and feet, circular in other parts of the water-vascular system. O 194 DESCRIPTIONS OF PREPARATIONS. The so-called heart or plexiform organ (gland) is composed of a plexus of vessels lying to the inner, i. e. adcentral, side of the stone canal inclosed in a peri- haemal space. At its ventral end it is connected to a circumoral plexus which lies in an incomplete septum dividing the circumoral perihaemal space into an outer and an inner space. The radial vessels are similarly lodged between a right and left space. At its dorsal end the plexiform organ is connected with a peri-anal ring which gives oif the ten vessels to'the genitalia and two vessels, one on either. side of its dorsal end, to the intestine. These vessels are also lodged in spaces. The vascular system contains brown cells, which also occur, but more sparingly, in the water- vascular system. The perihaemal spaces are probably a part of the coelome. They are lined by an epithelium which in the radial spaces forms an interrupted ventral band of columnar cells which were supposed by Lange to be nervous. They are said to communicate both with the coelome and the channels in the perisoma. The water-vascular oral ring lies to the dorsal and outer side of the two peri- haemal spaces, and below or ventral to them is the thickened nerve-ring just as in the arms. The coelome is lined by ciliated epithelium. Echinodermata, Encyclopaedia Britannica (ed. ix.), vii. System der Asteriden, Mtiller und Troschel, Brunswick, 1842. Stelle'rides du Muste, Perrier, A. Z. Expt. iv. 1875; v. 1876. Species of genus Asterias, Bell, P. Z. S. 1881. Skeleton. Gaudry, A. Sc. N. (3), xvi. 1851 (for figure of Asterias rubens, PI. 18, Fig. i); Agassiz, Memoirs Harvard Museum, v. 1877; Viguier, A. Z. Expt. vii. 1878. Names of ossicles belonging to ambulacral series. J. Miiller, Abhandl. Akad. Berlin (Classis Physica), 1853-54, p. 162 ; p. 210. Oral and apical system of Asteroids. Sladen, Q. J. M. xxiv. 1884 (contains general references). Oral ossicles. Viguier, A. Z. Expt. vii. 1878; Ludwig, Z. W. Z. xxxii. 1879; cf. remarks in Carpenter on ' oral and apical system of Echinoderms,' Q. J. M. xxii. 1882. Pedicellariae. Perrier, A. Sc. N. (5) xii. 1869. In Echinoids, Id. A. Sc. N. (5), xiii. 1870; Geddes and Beddard, Trans. Roy. Soc. Edinburgh, xxx. ; Foettinger, Archives de Biologic, ii. 1881 ; Sladen, A. N. H. (5) vi. 1880. In Euryalidae, Lud- wig, Z. W. Z. xxxi. 1878. General Minute Anatomy. Beitrage zur Histologie der Echinodermen, Ha- mann, pt. ii. Die Asteriden, Jena, 1885. Nervous system, eye, perisoma, feet, &c. Hamann, Z. W. Z. xxxix. 1883, p. 170. Mode of locomotion. Romanes and Ewart, Ph. Tr. 172, 1881, p. 836. Function of eye and pedicellariae. lid. J. L. S. xvii. 1884, p. 131 ; cf. Romanes, Jelly-fish, Star- fish, and Sea-urchins, Internat. Series, 1. 1885. Fission in Asteroidea and Ophiuroidea. Kowalewsky, Z. W. Z. xxii. 1872 ; Simroth, Ibid, xxviii. 1877; Haeckel, Ibid. xxx. (suppl.) 1878; E. von Martens, A. N. 32, 1866, p. 68. COMMON STARFISH. 195 38. COMMON STARFISH (Asterias, or Asteracanthion rubens), Dissected so as to show its digestive and motor systems. ONE of the rays, the central ray of the trivium, has been cut short, and more or less of the dorsal or anti-ambulacral integument removed from each of the other four, and from the central disc. In the interradial space which is opposite to the ray cut short, is seen the madreporite ; and a little to the left of a line drawn along the axis of the central ray to the madreporic tubercle, and near the centre of the disc is seen the small piece of dorsal integument in which the anus opens. It lies on this aspect in the interradius, to the left of the madreporite (see note, Plate xi.). From the intestine, and close to the anus, arise two diverticula. They bear several irregular caecal ampullae and reach a short way into two interradii, the interradius between the central and left ray of the trivium and the interradius between the left rays of the bivium and trivium respectively. The internal surface of these diverticula is longitudinally plicated and they are probably highly extensile. They are generally considered to be the homologues of the respiratory trees of the Holothuroidea. The intestine itself cannot be seen. It is short and arises from the pyloric division of the stomach. This pyloric division is pentagonal in outline, and a single trunk may be seen to arise above each angle of the pentagon. Each trunk enters the corresponding ray, and divides into two branches which, with their foliaceous glandular ampullae, fill up the greater part of the cavity of the ray. The saccular dilatations of the anterior or cardiac division of the stomach are to be seen lying below the trunks of origin of the complex caeca, and bulging for a short distance into the cavities of the rays. These sacculi can be evaginated so as to enclose the animals on which the Starfish feeds, e.g. young oysters, cockles, &c., too large to be drawn into the disc. Each sacculus is retracted after protrusion by a pair of muscles attached to the sides of the vertebral ridge of the ambulacral groove. The Asteroidea are the only group of Echino- dermata which possess a radial development of caeca to the digestive tract as seen here. The two divisions of the arborescent caeca have been separated in two of the rays to show the ampullae of the feet. These are arranged in two symmetrical rows on either side of the vertebral ridge of the ambulacral ossicles, thus corresponding to the arrangement of the pores seen in the preceding preparation. On the ventral surface the mouth is to be seen placed centrally. The ambulacral feet are variously contracted, some more, some less. They have sucker-like ends, which are not supported by calcareous plates as they are in Echinoidea and most Holothuroidea. In the left ray of the trivium the feet are completely retracted, and in the left ray of the bivium the series of o % I96 DESCRIPTIONS OF PREPARATIONS. adambulacral moveable spines has closed completely over the groove, thus protecting the soft parts which it lodges. The pedicellariae may be seen strewn among the spines of the perisoma, and upon the moveable adambulacral spines. It is possible that some of the minute elevations among the spines are the incompletely retracted tubular respiratory processes of the integument. The oesophagus is short and longitudinally plicated : the cardiac and pyloric divisions of the stomach are partially separated by a circular fold, and the intestine has a narrow plicated aperture into the stomach. The epithelium of the stomach is stated by Hoffmann to be ciliated. The arborescent caeca in the arms with their ducts are suspended to the dorsal perisoma by a couple of mesenteric bands. The cells in the caeca form entero- chlorophyll, and tryptic, peptic, and diastatic ferments. They pour their secretion into the stomach. The development of the genital glands is periodical : when sexually mature they reach far down into the arms. They are branched glands alike in both sexes, and are surrounded by a blood sinus. Each gland has a single duct in A. rubens, but this duct opens by a sieve plate, i.e. a plate pierced by many pores. There are ten plates corresponding to the number of the glands. They are placed inter- radially and dorsally close to the bases of the arms. In most instances the genital plate has only one aperture. These calcareous genital plates have been supposed to be the homologues of the genital, i.e. basal plates of the apical system in Echinoidea. It is certain however that this is not the case : and in Starfish in which the apical system is retained in its typical form, no relation is observable be- tween the genital apertures and the basals. Moreover in certain Starfish there is a more or less numerous series of genital glands, each with its own aperture, extending up the sides of the arms to a greater or less degree. In Asterina gibbosa the genital apertures are ventral. The duct in Asterina pentagona has unicellular glands by which the coat of the ovum is formed. These are probably present in other Star- fish as well. Impregnation is external. General Anatomy of soft parts. Ludwig, Z. W. Z. xxx. 1878; cf. Carpenter, Q. J. M. xxi. 1 88 1. Add on generative organs. Ludwig, on Asterina gibbosa, Z. W. Z. xxxi. 1878. General minute Anatomy. Beitrage zur Histologie der Echinodermen, Hamann, pt. ii. Die Asteriden, Jena, 1885. Digestive ferments. Krukenberg, Vergleich. Physiol. Vortrage, ii. 1882, and lit. cited, p. 78. On colouring matters. Id. op. cit. iii. 1884, and lit. cited, p. 179. Enterochlorophyll. MacMunn, P. R. S. xxxv. 1883; Id. P. R. S. xxxviii. 1885. 39- COMMON EARTHWORM (Lumbricus terrestris, s. Agricola), \ Suspended by the anterior extremity to show its external characters. THE anterior region of the body tapers to a conical point: the posterior is flattened dorso-ventrally and tapers abruptly. The first somite so- COMMON EARTHWORM. 197 called or prostomium (infra) lies entirely in front of the mouth and corre- sponds to the praeoral somite in o'ther Chaetopoda. The mouth is ventrally placed in the second segment and is therefore subterminal. The anus on the other hand is posterior and terminal. The whole body is distinctly divided into a series of somites or segments, separated by well-marked intersegmental furrows. Vertical transverse septa or dissepiments which divide the body into compartments, correspond internally to these furrows with the excep- tion of the first five or six. The compartments communicate with one another round the supra-nervian blood-vessel. The chitinoid cuticle has been loosened by maceration in the spirit and may be seen lying in folds in the posterior region of the body. In life it is iridescent owing to the presence of two sets of fine parallel superficial lines crossing each other at angles of 75° to 80°. On the ventral surface of the fifteenth somite count- ing the praeoral as the first, are two white tumid elevations. These cor- respond to the apertures of the vasa deferentia. If the worm be held so that the light falls upon it obliquely, two somewhat raised lines may be seen running down the body on each side parallel to one another. The more ventral of these lines corresponds very nearly in position with the apertures of the vasa deferentia. Both lines indicate the position of the setae which constitute the locomotor apparatus of the Chaetopoda. There are in each somite of the common Earthworm two setae in each line, implanted a small distance apart, but they are often lost by accidents. The outer row is wanting in some instances as far back as the clitellum (infra). The inner row usually commences on the fourth or fifth somite. Both rows may be absent in the posterior somites. The spot corresponding to the inner or ventral row on the twenty-sixth somite of this specimen is somewhat swollen. The setae are here, as in the region of the clitellum, peculiarly long and delicate in a sexually mature worm, and are generally retracted. They are supposed to act as accessory copulatory organs. The dorsal and lateral parts of somites thirty-one to thirty-eight, and especially of somites thirty-three to thirty-six, are white and swollen, and the swollen region is bordered by a prominent well-marked edge on either side of the median ventral line. These swollen somites constitute the clitellum, an organ especially characteristic by its great development of the terrestrial Oligochaeta. It is glandular and secretes a plentiful mucus from which the cocoon is formed. The prominent ventral edge acts as a copulatory organ. The development of the clitellum depends on the age and the state of sexual activity of the individual. Vejdovsky believes, from his observations on the growth of the prostomium or praeoral somite in the individual produced by fission of Aeolosoma tenebrarum as well as in embryoes of Rynchelmis, that it is really an outgrowth of the buccal somite and not a separate somite. In Aeolosoma the buccal somite developes the I98 DESCRIPTIONS OF PREPARATIONS. supra-oesophageal ganglion, the two cephalic provisional nephridia, and the pharynx, the prostomium growing forwards by degrees. * It is not separated by a furrow from the buccal somite as it is in all other Chaetopoda except Chaetogastridae (see Vej- dovsky, op. cit. infra, p. 162). In Typhaeus from India it is apparently absent, and the mouth therefore terminal. In Microchaeta Rappifat somites are secondarily annulated, as in the Leeches. The number of annuli in the first and in each somite from the ninth onwards is three : but in the second to the seventh somite inclusive it is either six or seven. The cuticle is thin, transparent, and variable in thickness in different regions of the body. It is said to consist of two layers, an outer longitudinal and a thicker inner circular fibrillar layer. It is pierced by minute pores, the orifices of unicellular hypodermic glands. The hypodermis or epidermis consists of a single layer of cells. In the inter- segmental furrows they are shorter than elsewhere, and are non-glandular : whereas in other regions glandular and non-glandular cells occur intermixed. The non- glandular cells are either cylindrical with several basal processes, or more or less globular with a slender external process and one or more basal processes. The latter kind of cell lies at a deeper level than the former. The glandular cells vary some- what in shape and in the nature of their contents, probably in accordance with their state of activity. According to Professor Ray Lankester, processes of pigment cells belonging to the subjacent connective tissue pass up, especially in L. olidus, between the hypodermis cells, as in the Leech. In the clitellum non-glandular cells are not to be found in the common Earthworm and some of its allies ; in others they are pre- sent in reduced numbers. Its glandular cells are divisible into a more superficial layer with coarsely granular contents which stain readily with carmine, and a deeper layer with finely granular contents which do not stain with the same reagent. In Allolobophora (Dendrobaena) rubida, Vejdowsky traced nerve-filaments into appa- rent continuity with the glands. Capillaries penetrate between the clitellar glands, and are very numerous in the common Earthworm: in some species they are few in number. In Perionyx and Megascolex the general hypodermis is vascularised as in the Leech. An elastic basement membrane separates the hypodermis from the mus- cular layers of the body-wall. It is highly developed in Perionyx and Perichaeta. The muscles are disposed in an outer circular and an inner longitudinal layer. Each muscle-fibre corresponds to a single cell with a well-developed cortical fibrillar substance and a central medulla or protoplasm which is scanty, and in which. a nucleus has been found in L. olidus and Phreoryctes, two worms in which the fibrillar cortex is sometimes deficient at one spot, so that the muscle-cell becomes coelomyarian. In Z. terrestris and some other Lumbrici and a few other Earth- worms, as well as in Serpula and Protula among Polychaeta, the simple layer of lon- gitudinal cells found in the lower Oligochaeta is disposed in parallel folds, the bi- pinnate bundles of Claparede, which are held together by intervening connective tissue. The muscle-cells of the longitudinal coat in other Chaetopoda are arranged in more or less regular groups or bundles. The longitudinal coat is divided in Oligochaeta into distinct tracts by the projection internally of the sacs of the setae. The bundles of protrusor or parieto-vaginal muscles attached to the bases of the sacs in question are derived however from the circular coat. Bundles of muscular or inter-follicular fibres pass from the bases of the sacs in the dorsal row to the bases COMMON EARTHWORM. 199 of the sacs in the ventral row. They appear to be special formations. Branched pigment cells are found most plentifully in the connective tissue of the circular layer of muscles. They are more numerous on the dorsal and lateral aspects of the body than on the ventral ; more numerous in its anterior than in its posterior half. The body-cavity is lined by a peritoneal epithelium which varies in character in different regions and where it coats different organs. The septa dividing the body into compartments are composed of connective tissue, of radial muscular fibres continuous with the fibres of the circular coat, and of circular fibres surrounding the digestive tract, the supra-neural blood vessel and nerve cord. Aeolosoma has only one septum separating the head from the rest of the body. The most anterior septa in the common Earthworm are replaced by the muscular bundles which pass to the pharynx from the body-wall. The perforations in the septa, by which one division of the body-cavity communicates with another, are in some instances near the body-wall. The divisions of the body-cavity communicate with the exterior in most Earth- worms by dorsal pores, situated quite close to the anterior intersegmental furrow of each somite. In the genus Lumbricus these pores commence in the sixth, seventh, or eighth somite. They become occluded in the clitellum by the pressure of the gland cells. They are simple perforations of the body-wall. Their aperture is sur- rounded by a circular sphincter muscle, and is opened by an anterior and posterior band of longitudinal fibres. The peritoneal cells at the margin covering the longi- tudinal divaricators are massed in a small heap. The function of the pores appears to be that of expelling coelomic fluid, or lymph. A cephalic pore is found in the lower Oligochaeta except Aeolosoma^ either on the ventral or dorsal aspects of the prostomium, or else anteriorly and terminally. Dorsal pores are found in Enchytraeus^ JVais, &c. They are absent in Polychaeta. The setae are S-shaped, with the outer end more pointed than the inner. They are implanted in sacs or trichophores, which are simple invaginations of the hypodermis, composed of three (?) cells, one basal and two lateral. New setae are produced in the same sac as the old, and each seta is the product of a single cell. The genital setae of the tenth to the fifteenth somites, of the twenty-sixth somite and of the clitellum, are produced in the sacs of the ordinary setae present before the worm becomes sexually mature. The sacs enlarge as soon as the ordinary setae drop out, and from the enlarged cells are produced the long delicate genital setae. The corresponding setae in the aquatic Oligochaeta are produced in new sacs, the setae and sacs previously present undergoing atrophy (Vejdovsky). The setae of Urochaeta are bifid at their apex : the genital setae of Acanthfodrilus (? all species) and of Urochaeta, and the setae in general of Rhinodrilus, are variously ornamented. Consequently the setae of terrestrial Oligochaeta are not invariably simple as Clapa- rede supposed. The form of the setae in aquatic Oligochaeta is generally either simple and hair-like, or else slightly curved with a bifid apex. In the genus Anachaeta the trichophores are present but form no setae. The same thing occurs in Urochaeta with certain setae. In Branchiobdella (if it is an Oligochaete) all traces of the sacs are lost. Genital setae are always developed on the clitellum, and their form is variable, but different as a rule to those of other parts of the body. The arrangement of the setae in two rows, an -outer and inner or dorsal and ventral, is the common one in Oligochaeta^ but it is sometimes departed from. The 200 DESCRIPTIONS OF PREPARATIONS. number of setae in each somite is also not constant. Among Earthworms Lum- bricus has a pair in each of the four longitudinal rows (i. e. eight) in each somite. In some species of Acanthodrilus, in Titanus, the eight setae are sometimes separated by equal intervals, and there are consequently eight rows, i. e. four on each side. The eight setae when thus separated may not remain in line from somite to somite, but alternate in position, as in the posterior somites of Urochaeta. Megascolex has very numerous setae, interrupted only in the median dorsal and ventral lines, whereas in Perichaeta and Perionyx single setae are implanted at intervals round the circumfer- ence of each somite. It is difficult to say whether the concentrated or scattered mode of distribution is the more primitive (see under nephridia, p. 205). The genus Acan- thodrilus affords an instance of variations in this particular in different species. The clitellum of Earthworms appears to fulfil two functions. It is a copulatory organ, and in L.foetidus Perrier observed the formation of a resistent membrane from its surface during sexual congress. It also secretes the cocoon in which the ova are contained together with spermatozoa and albumen. The cocoon is stripped off forwards by contractions of the body : and is charged with its contents in tran- sit (?). The two ends of the cocoon close of their own accord as soon as it is freed from the body. In some Earthworms the clitellum surrounds the body, and there is no ventral furrow, as occasionally happens in L, terrestris. Its position and extent are variable not only in different groups of terri-colous Oligochaeta, but within cer- tain limits in the same genus and even the same species. In the aquatic Oligochaeta it includes only the somite in which the vas deferens opens, and it does not secrete during congress. It is apparently absent in Moniligaster among terrestrial Oligochaeta. Phosphorescence has been noticed in various Lumbrid, and appears to be due to a fluid excreted from the hypodermis (see Vejdovsky, p. 67). System und Morphologic der Oligochaeten, Vejdovsky, Prague, 1844. Id., Monographic der Enchytraeiden, Prague, 1879. Lombridens Terrestres, Perrier, Nouvelles Archives du Museum d'Hist. Nat. de Paris, viii. 1872. Studies on Earth- worms (first of a series), Benham, Q. J. M. xxvi. 1886. Lumbricus terrestris. Claparede, Z. W. Z. xix. 1879; Horst, Tijdskr. Neder- land. Dierk. Vereen, iii. 1876 ; cf. A. N. 43, (ii), 1877, p. 481 ; Brooks, Handbook of Invertebrate Zoology, Boston, 1882, p. 140; Howes, Atlas of Practical Elementary Biology, London, 1885. Plutellus, Perrier, A. Z. Expt. ii. 1873. Urochaeta, Id. op. cit. iii. 1874. Pontodrilus, Id. op. cit. ix. 1881. Megascolex (= Pleurochaetd] Moseleyi, Bed- dard, Trans. Roy. Soc. Edinburgh, xxx. pt. 2. Megascolex, Perichaeta, Id. A.N.H. (5), xiii. 1884. Do. with Perionyx and Typhaeus, 'Earthworms from India,' Id. op. cit. xii. 1883. Perichaeta, Moniligaster, ' Earthworms from Ceylon,' Id. op. cit. xvii. 1886. Acanthodrilus capensis, Id. Proc. Roy. Soc. Edinburgh, 1885. Micro- chaeta Rappi, Id. Tr. Z. S. (to appear) ; and Benham, op. cit. supra. Hypodermis. Von Mojsisovics, SB. Akad. Wien, Ixxvi. Abth. i. 1877 ; Ray Lankester, Q. J. M. xx. 1880, p. 303. Body wall. Beddard, Proc. Roy. Soc. Edinburgh, 1884. Muscles. Rohde, Schneider's Zool. Beitrage, Breslau, i. 1885. Dorsal pores and musculature. Ude, Z. W. Z. xliii. pt. i. 1885. Peritoneal epithelium. D'Arcy Power, Q. J. M. xviii. 1878. Cocoon and congress. Perrier, A. Z. Expt. iv. 1875, notes, p. xiii. ; cf. Ratzel, and Warschawsky, Z W. Z. xviii. 1867-69, p. 547. COMMON EARTHWORM. 2OI Phosphorescence in Earthworms. Vejdovsky, op. cit. supra, p. 67 ; Cohn, Z. W. Z. xxiii. 1873. In Polychaeta. Panceri, Atti Acad. fis. mat. Naples, vii. 1878 (cf. Journal de Zool. v. 1876, p. 94); Robin, Bull. Soc. Philomath. (7), vii. In Polynoe. Jourdan, Z. A. viii. 1885. General account. Mclntosh, Nature, xxxii. 1885, p. 476 ; cf. Panceri, A. Sc. N. (5), xvi. 1872, and Secchi, on Spectrum, ibid. In decaying organisms. Pfliiger, in his Archiv f. Physiol. xi. 1875. Vegetable mould and Earthworms. Darwin, London, 1881 ; Hensen, Z. W. Z. xxviii. 1877. Turriform heaps. Trouessart, C. R. 95, 1882. Absence from N. W. Prairies of N. America. Nature, xxix. 1883-84. Habits. Ibid. xxx. 1884. 40. COMMON EARTHWORM (Lumbricus terrestris, s. Agricola), The first forty-one somites of the body including the clitellum, dissected to display so far as possible the reproductive system as well as the portions of the digestive, and circulatory systems con- tained in this part of the body. THE integument has been divided down the middle dorsal line and fastened out on either side. The digestive tract occupies the centre of the preparation. It consists of a buccal cavity, not seen here ; of a pharynx, oesophagus, crop, gizzard, and intestine or stomach, and a short rectum. The pharynx occupies the first five somites of the body. It has a rough exterior owing to the number of muscular bundles which pass between it and the body walls and have been severed in dissection. The oesophagus is narrow and extends through ten somites. About a quarter of an inch from its posterior end there is very visible on the right side, one of the three calcigerous or oesophageal glands, — the glands of Morren. The crop occupies a large space in the sixteenth and seventeenth somites. It has thin walls and dark-coloured contents. The gizzard comes next and lies in the seventeenth and eighteenth somites. It is smaller than the crop and has light-coloured muscular walls ; and is followed by the intestine or stomach. The first portion of this tube is distinctly sacculated laterally, a feature which becomes less and less marked posteriorly. Its walls have a darkish appearance due to the modified and pigmented ' chloragogen ' cells of the peritoneum which coat it and form the tissue, often miscalled hepatic. The dorsal blood-vessel is clearly seen on the dorsal surface of the crop and intestine in the middle line. It has a distinctly moniliform appear- ance which is more marked in some other terrestrial Oligochaeta. On either side of the anterior part of the oesophagus may be seen four of the five or six pairs of ' hearts ' which connect the dorsal vessel with the supra-neural vessel, in the somites behind the pharynx. " The large pendulous vesiculae seminales lie anteriorly to the crop and to the outer side of the oesophagus. The two pairs of spermathecae may be seen, especially on the left side, in the line of the outer row of setae. Remnants of the septa may also be dis- cerned on the internal surface of the body- walls, and in most of the posterior 202 DESCRIPTIONS OF PREPARATIONS. somites the nephridia or excretory organs are left in situ on either side of the intestine. The stomodaeum in Aeolosoma is restricted to the first somite of the body ; in other Oligochaeta it extends backwards through a variable number of somites, and is differentiated into a buccal cavity and a musculo-glandular pharynx. The muscular tissue is most developed on the dorsal aspect of the pharynx, and is irregular in arrangement. The pharyngeal glands appear to correspond to the septal glands of the Enchytraeidae : they are large in Megascolex Moseleyi and Pontodrilus. The oesophagus and stomach-intestine jointly represent the archenteron. The oesophagus of Perichaeta has three sets of glands attached to it. The gizzard is absent in Pontodrilus and all aquatic Oligochaeta except some Naidomorpha. It is in some instances placed more anteriorly in the body, e. g. in Urochaeta, Antaeus, &c., than it is in Lumbricus. Digaster possesses two, Moniligaster five, gizzards. The second gizzard in the former is separated by a whole somite from the first : the last four gizzards of Moniligaster follow one another closely, but are six somites behind the first. In these cases a longer or shorter tube intervenes between the gizzards, as is the case also between the gizzard and the sacculated region of the intestine. It is difficult to say whether the last-named tubular region should be considered as intestine or oesophagus. Glands are attached to it, e. g. three pairs in Urochaeta, one pair in Microchaeta, which appear to bear a lime-secreting function. They may be regarded as either homologues or analogues of the glands of Morren, according to the view which is taken as to the oesophageal or intestinal nature of the tube from which they originate. Calcigerous glands are entirely absent in some instances, e. g. in Megascolex, Typhaeus, Pontodrilus, &c. A pair of ventrally placed caeca open into the commencement of the sacculated intestine of Perichaeta, whilst Typhaeus and Megascolex possess a series of dorsal paired glands, five in the former, fifteen to sixteen in the latter, opening into it more posteriorly. The walls of the digestive tract consist of the following layers from without inwards : (i) an epithelial peritoneum ; (2) a longitudinal, (3) a circular, layer of muscle-cells ; (4) a layer of capillary bloodvessels, and (5) the epithelium. The peritoneal cells in the region of the stomach and the stomachal vessels are pyriform and branched at their attached ends. They are coloured in all instances — green, black, or in the Earthworm brownish-yellow : hence chloragogen cells. They have usually a well-defined membrane, and contain concretions. They appear to be set free into the coelome, and then degenerate ; and their products, soluble and in- soluble, are probably excreted through the nephridia. The longitudinal layer of muscle-cells is thin, but is better developed in the gizzard where the circular layer attains an extraordinary development. The capillaries are well developed everywhere, but more especially in the pharynx and oesophagus. The lining epi- thelium is columnar. In the Earthworm it secretes a cuticle in the pharynx, oesophagus, crop and gizzard, but in the stomach is ciliated. In many of the lower Oligochaeta it is ciliated throughout the whole tract. The cells in the oesophagus from the n^to the 13^ somite are glandular. To this region belong the calci- gerous glands. The first pair of these glands in the Earthworm is a pair of hollow diverticula, whilst the second and third pair consist solely of enlarged gland-cells (Claparede). They secrete a milky fluid which owes its milkiness to the presence COMMON EARTHWORM. 203 of Calcium carbonate in the form of rhombohedra (especially in the first pair), or of minute spherules. The Calcium carbonate has a chemical action, not a mechanical, as supposed by Claparede, and partly neutralises the acids of the humus swallowed by the worm, partly acts on the humus itself. The dorsal wall of the stomach-intestine is invaginated in nearly all terrestrial Oligochaeta to form a ridge or typhlosole. This structure is absent in Megascolex, and in the aquatic Criodrilus and semi-marine Pontodrilus, which in many respects resemble the terrestrial forms. The ridge is complicated anteriorly, simple pos- teriorly, and consists of all the layers entering into the wall of the intestine. In some cases (Urochaeta^ Allolobophora cyanea, &c.) a longitudinal supra-intestinal vessel runs along the cavity of the typhlosole, but in Lumbricus its walls are supplied by vessels derived from the dorsal longitudinal vessel. It is kept invaginated, according to Claparede, by muscles passing across from one to the other side in each somite. The epithelium covering it internally is often found enlarged with smaller cells between the bases of the large cells. The large cells are probably thrown off into the cavity of the tract and there resolved. The same changes have been noted in the intestinal epithelium of Rhynchelmis. The digestive tract ends with a short thin-walled rectum, which is lined by cuticle and extends only through a single somite. Thie anus is dorsal in Criodrilus, and there are in this worm at least seven somites with an embryonal nerve-cord traceable behind it. A longitudinal dorsal and ventral mesentery are present in Criodrilus. The dorsal is aborted in other instances, but the ventral usually persists. The contents of the oesophagus are acid ; of the stomach-intestine alkaline. There is a diastatic and tryptic ferment, and a peptic as well, but the latter must be inactive in the living animal owing to the alkalinity of the stomach (Krukenberg). The vascular system of the Earthworm consists of a median dorsal vessel, of a supra-neural (= ventral) vessel, and three neural vessels, two lateral and one sub- neural1. The dorsal, supra-neural, and sub-neural vessels branch anteriorly and anastomose on the pharynx. In the region of the oesophagus six pairs of dilated and pulsatile vessels or 'hearts' pass from the dorsal to the supra-neural vessel. A small seventh pair gives origin to a lateral oesophageal vessel on each side which runs forward to the pharynx. The supra-neural vessel is connected from place to place with the capillary network in the walls of the intestine, and by this means indirectly to the dorsal vessel. A vascular loop unites the dorsal to the sub-neural vessel in each somite of at least the intestinal region. The integument and nephridia are provided with a rich supply of capillaries. The blood supply of the latter is said to be connected with the supra- and lateral, neural vessels ; of the former, with the dorsal and sub-neural vessel. But there is some uncertainty about 1 Howes figures in his Atlas of Practical Elementary Biology (PL xii. fig. 2) a sub- or infra- intestinal vessel applied to the inferior aspect of the intestine. Such a vessel exists in some Poly- chaeta and Oligochaeta, but not in the common Earthworm, so far as I can find from actual observation, by dissection of fresh specimens and of mounted sections. Ben ham also appears to have failed to find it (Q. J. M. xxvi. p 253). It seems to me that the dark streak on the inferior aspect of the intestine which looks like a vessel is due to the attachment of the ventral mesentery in which the supra-neural or ventral vessel is suspended, and the consequent absence of chloragogen cells. Dark lines due to the same cause mark the attachments of the septa to the intestine. If the chloragogen cells are removed by a scratch, the scratched spot has a similar dark appearance. 204 DESCRIPTIONS OF PREPARATIONS. these points. In Urochaeta and Criodrilus the dorsal and supra-neural vessels give origin to the integumentary capillaries. Dilatations occur in the nephridial capillaries, especially in the middle and posterior regions of the body. Similar dila- tations occur elsewhere. The most remarkable variations in the circulatory system of Oligochaeta are the following. The dorsal vessel, in two sp. of Acanthodrilus from New Zealand is double from the pharynx backwards, an embryonic feature seen also in some Polychaeta : in Megascolex and Microchaeta the sections of the dorsal vessel in the anterior somites are double between the septa, single where they perforate them (Beddard, Proc. Roy. Phys. Soc. Edinburgh, 1885). In some Earthworms, e. g. certain sp. of Acanthodrilus, Urochaeta, the posterior hearts connect a supra-intestinal vessel to the supra-neural or ventral vessel ; and in some instances, e. g. Urochaeta, Enchytraeidae, &c., the dorsal vessel has moniliform pulsatile dilatations, and in the Lumbriculidae blind lateral and sometimes branched processes. Integumental capillaries are wanting in aquatic Oligochaeta, except at the posterior extremity of the body in some Naidomorpha, the Tubificidae, and Criodrilus. The blood-vessels of Lumbricus are lined by an endothelium, the cells of which vary in character in different regions, suggesting a distinction into arteries and veins. The contained liquid is coloured with haemoglobin, as in most Oligochaeta, except Aeolosoma, Chaetogastridae, and most Enchytraeidae. It contains floating corpuscles, flattened, fusiform, sometimes nearly circular, which, according to Ray Lankester, are ' the nuclei of the endothelial cells set free from the walls of the vessels.' Vejdovsky, however, points out that the Lumbricidae, like the Leeches, have minute valve-like or irregular masses of cells connected to the walls of the vessels by fine processes. They are also present but confined to the dorsal vessel in aquatic Oligochaeta. Some of these cells are set free, according to him, and form the corpuscles which are absent only in Enchytraeidae and Naidomorpha. The coelomic fluid contains amoeboid cells, ' large colourless vacuolated cor- puscles, with a ragged outline often produced into filaments and provided with a large nucleus.' The excretory nephridia, or segmental organs as they used to be called, are found one pair to each somite throughout the whole extent of the body from the fourth somite onwards. Each organ consists of a ciliated funnel or nephrostome, a convoluted tube, and a terminal muscular duct. The funnel is bilobed, but one of its lips is very small : its free margin is formed by a single row of ciliated cells. It opens into the cavity of the somite anterior to that in which the duct opens externally \ an arrangement common to all Oligochaeta, but by no means so in Polychaeta. The tube, therefore, to which it is connected passes backwards through a septum. The convoluted tube forms typically three loops. The first portion of it is clear-walled and ciliated; the second rather wider, its cells granular, and their cilia peculiarly long ; the third part is wider still, its cells coarsely granular and not ciliated. The tube appears to perforate the cells, i. e. is intra-cellular as in the corresponding region of the Leech's nephridium. The duct is long, with muscular walls, and is dilated. It corresponds to the more or less pyriform vesicle into which the glandular part of the nephridium opens in some 1 The Oligochaete Plutellus is an exception to this rule. Its nephridia (and oviducts) open both externally and internally in the same somite, as in many Polychaeta. COMMON EARTHWORM. 305 terrestrial and most of the aquatic Oligochaeta and the Leeches, and is probably therefore derived from an invagination of the hypodermis, as is the vesicle of the aquatic forms, the rest of the nephridium being mesoblastic. The pore by which the duct opens is generally placed slightly in front of the setae in the ventral row, but it may be placed in a similar position relative to the dorsal row. There seems to be much variability in the position of the aperture in the same individual, and the apertures of the right and left organs in the same somite may not correspond (cf. Claparede, Z. W. Z. xix. 1869, p. 620 ; Hering, Z. W. Z. viii. 1856-57, note, p. 401). The variation is perhaps due to the loss of nephridia originally present in each somite, and opening in different positions relatively to the setae. Perrier observed in Plutellus an alternation in the position of the nephridial apertures between the upper and lower rows of setae. Beddard has noted a similar but irregular alternation in a species of Acanthodrilus in which the dorsally placed nephridia have the duct represented by a large thin-walled muscular sac, with a small diverticulum ; whereas those ventrally placed open laterally into a long muscular sac, much as do the nephridia of aquatic Oligochaeta. There is there- fore an anatomical difference between the two sets of nephridia opening dorsally and ventrally. The last-named anatomist has also described the nephridia of a New Zealand Acanthodrilus (A. multiflorus), in which there are four nephridia on each side in each somite of the body, corresponding to the setae. The two ventral nephridia on each side are intertwined and anastomose (?). In the pos- terior region of the body the organs open externally, each by a single aperture placed dorsally to each of the eight separate setae ; in the anterior region, however, there are innumerable minute orifices in a single circular row, and in the same line with the setae, due to the branching of the ducts which appear to form a circular canal round the somites in this region. It is quite possible therefore that in other Earthworms certain of the nephridia in each somite are aborted, but that it is sometimes one, sometimes another pair which is missing. Benham states (Q. J. M. xxvi. p. 256) that numerous small nephridia occur in each of the somites of a small Perichaeta from the Philippine Islands. Details are wanting at present. The aquatic Oligochaeta have a single pair of nephridia in each somite opening near the ventral set of setae, and composed of the same parts as in Lumbricus. The duct, which is sometimes glandular, is usually vesicular, and the gland-tube opens into it laterally. The funnel is absent in Chaetogastridae. Vejdovsky finds that both funnel and gland-tube are produced by the growth of peritoneal cells covering the septa : the duct by an invagination of the hypodermis. Nephridia of the ordinary shape appear to be absent in Megascolex and Typhaeus, and their place is taken -by tufted bunches of tubes, the structure of which has not been investigated. But they appear to be closely similar to the first. pair of nephridia of Urochaeta, and the nephridia in general of Microchaeta. Each organ in the last named consists of a muscular vesicle opening externally to which are attached (i)a series of U-shaped loops, each loop containing intra-cellular tubes, and (2) a single simple tube which perforates a septum in the usual manner and opens into the coelome by a small contracted funnel. The mode of con- nection of the tubes to the vesicle has not been ascertained as yet. The Earthworm is hermaphrodite like most of its congeners. The organs are figured in PI. xii. The male apparatus consists of two pairs of testes, three 2o6 DESCRIPTIONS OF PREPARATIONS. pairs of vesiculae seminales, and a paired vas deferens. The testes are situated near the nerve cord on the posterior faces of the anterior septa dividing the tenth and eleventh somites. Each testis is described by Blomfield as ' a white trans- lucent body of irregularly quadrangular form, rarely more than one-tenth of an inch in diameter, much flattened, and attached by one side to the coelomic epithelium of which it appears to be a local modification.' In the Chaetogastridae the spermato- zoa begin to develope in the coelome and complete their growth in the sperma- thecae, but in most Oligochaeta, as in the Earthworm, they develope in vesiculae seminales which are caecal outgrowths of the septa. This is their probable origin in' Lumbricus. The immature vesiculae in this worm are, according to Blomfield, 'six small light-coloured vascular growths on the three septa 9-10, 10-11, 11-12, arranged in three pairs.' The first and second pairs are anterior outgrowths of the septa to which they belong : the third is a posterior outgrowth of the septum 11-12 (Benham, note, p. 259, Q. J. M. xxvi. 1886). The cavities of the vesiculae are traversed by connective tissue trabeculae and capillary bloodvessels. The sperm- polyplasts and fully formed spermatozoa are lodged in the interstices of the trabeculae, in which may also be found the Gregarine Monocystis Lumbrid in all stages of its life-history. When the Earthworm is sexually mature the first four vesiculae 'form a central body covering in the rosettes and testes of the tenth segment.' A similar coalescence occurs in the eleventh somite between the two last vesiculae. Each vas deferens consists of two ciliated funnels or rosettes; one situated in the tenth, the other in the eleventh somite, which join a common duct. The latter opens near the ventral set of setae in the fifteenth somite. It therefore traverses three somites. The two ovaries occupy a similar position to the testes but are lodged in the thirteenth somite. Each is invested by a layer of flat peritoneal cells. When mature it has the form of a pear with a long stalk — the latter being formed of a single string of ripe ova. It consists, when immature, like a testis, of a finely gran- ular protoplasm with scattered nuclei, in other words, of a syncytium. The ova of the aquatic Oligochaeta, e. g. Tubificidae, break away from the ovary and ripen in the coelome or in an ' egg-sac ' similar to the vesiculae seminales. Each oviduct commences with a wide ciliated aperture vis a vis to the ovary. The duct per- forates the septum between somites thirteen and fourteen, and opens in the fourteenth somite near the ventral row of setae. A receptaculum ovorum, or 'egg-sac,' is attached to the oviducal funnel and receives the ripe ova. It contains 1-5 ova, and like a vesicula seminalis, it is a diverticulum of a septum, — that which separates somite thirteen from somite fourteen. It projects into the latter (Bergh). The female apparatus is completed by two pairs of vesicular sperma- thecae which open in the inters egmental furrows between somites nine and ten, and ten and eleven, and are derived as in aquatic Oligochaeta from invaginations of the hypodermis (Bergh). They receive the spermatozoa in congress and secrete a clear fluid, which, according to Vejdovsky (op. cit. pp. 154-5), forms the spermato- phores. The clitellar glands are usually supposed to form these structures. The spermatophores are 1^—2 mm. long and 0-5-0-7 mm. wide, somewhat spirally coiled, and with an open hollow at one end in which the spermatozoa lie parallel to one another. In some worms they are closed. They are usually attached on or about the clitellum. COMMON EARTHWORM. 307 The testes of Microchaeta are four in number as in Lumbricus, and are con- tained in the vesiculae. The organs in Earthworms usually described as testes appear to be the vesiculae. Of these there are generally two pairs ; in certain forms, however, only a single pair, which extends through three somites in Urochaeta and Typhaeus, through twelve to fifteen in Titanus, and thirty or more in Urobenus (Benham). There is always a single pair of ovaries, which in Plutdlus are situated in front of the testes. The ovaries may be lobed, e. g. in some species of Peri- chaeta. There are four separate vasa deferentia in Moniligaster and Acanthodrilus, each with its own aperture : and they each appear to possess, as does the vas deferens of the aquatic Oligochaeta, a terminal vesicle homologous with a nephridial vesicle. The ciliated funnels are in front of the vesiculae in Pontodrilus Marionis, while Anteus is stated to possess no sperm-ducts at all. The genital apertures are sometimes approximated ventrally and the oviducal aperture is single, and median in Perichaeta, Perionyx, and sometimes in Megascolex. The position of the male apertures with reference to the clitellum varies, and has been used by Perrier as a means of classifying Earthworms. The apertures in question are situated (i) in front of the clitellum, — Ante-clitelliani, e.g. Lumbricus \ (2) within it, — Intra-clitelliani, e. g. Anteus, Titanus, Urochaeta : (3) behind it, — Post- clitelliani, e. g. Perichaeta, Megascolex, Plutellus, Pontodrilus, Acanthodrilus, Peri- onyx : (4) or there is no clitellum even in sexually mature specimens, — Aclitelliani, e.g. Moniligaster. Megascolex has been termed by Beddard Infra-clitellian, for though the vasa deferentia open within the clitellum, yet they occupy a non- glandular ventral area. As to the other accessory organs of generation, the setae of the male somites are sometimes much enlarged, especially in the genera Acanthodrilus and Eudrilus; copulatory papillae are present in some species of Perichaeta; and prostatic glands opening into the sperm-ducts are found in the worms just named. The spermathecae are always in front of the vesiculae ex- cept in Microchaeta and Eudrilus. They vary in number and size even in the same genus, and are sometimes complicated as in Perichaeta Houlleti by the addition of glands. There may be more than two spermathecae in each somite : in this case they are arranged in transverse rows. See Beddard, A. N. H. (5), xvii. 1886, pp. 91-92 and 93-94: Benham, Q. J. M. xxvi. pp. 263, 280. In the aquatic Oligochaeta the nephridia atrophy away in the somites in which the sexual glands and ducts are undergoing evolution. In Lumbricus and many other Earthworms the two sets of structures co-exist. Vejdovsky is inclined to suppose that the larger size of the coelome is the cause of this persistence. The evolution of the genital ducts in the Earthworms remains to be worked out \ In many aquatic Oligochaeta it is accurately known. The vas deferens has in them a funnel, and a ciliated tube derived from peritoneal epithelium, and an atrium which is invaginated from the hypodermis like the nephridial vesicle, but is not represented in most Earthworms. The spermathecae are invaginations of the same kind. The oviducts are either simple slits (Aeolosoma, Chaetogastridae, &c.) or a pair of funnels (Lumbriculidae), and are obviously degenerate. Vejdovsky is inclined to conclude (op. cit. p. 161) that the sexual ducts either correspond to all the parts of an ordinary nephridium (vasa deferentia of aquatic Oligochaeta, of Acanthodrilus and Moniligaster) or to some of them (spermatheca to the nephridial 1 Bergh states that he intends to publish on the subject. 20$ DESCRIPTIONS OF PREPARATIONS. vesicle : the oviduct of all, and vasa deferentia of Lumbricus, &c. to the nephridial tube) : and that they are excretory organs developed only at the time of sexual maturity. In the lower Oligochaeta the sexual apparatus atrophies after the sexual products are discharged, but the worms appear to live on and may propagate themselves asexually. It is probable that the development of the sexual apparatus of the Earthworms will be found to be similar to that of the aquatic Oligochaeta. The only thing that must be regarded as certain at present is the fact that the sexual ducts of Oligochaeta are not, as in many Polychaeta, ordinary nephridia, which take on a sexual function at the reproductive season. It is possible that the resemblances traceable between them and the nephridia are merely homo- plastic1 and not homogeneous : i. e. the structures may not be identical. And the fact that the vasa deferentia of the Earthworms extend through several somites before they open externally may be held to support this view. It is true that the nephridia of the aquatic Phreatothrix extend through several somites. They form, however, U-shaped tubes and their external and internal apertures are always in two con- tiguous somites as in other Oligochaeta (Vejdovsky, op. cit. p. 124). The terrestrial Oligochaeta do not multiply by fission and gemmation as do many of the aquatic forms. They appear, however, to have considerable powers of regeneration : and can form new anterior and posterior somites, supra- and sub- oesophageal ganglia, and pharynx in the place of parts removed by excision. They have been occasionally found with a bifid posterior extremity. Digestive tract. Claparede, op. cit. ante, p. 600 ; in general, Vejdovsky, op. cit. pp. 100-112. Chloragogen cells, Vejdovsky, op. cit. pp. 110—112. Secretion of calcigerous glands, Robinet, C. R. 97, 1883. Digestive ferments. Krukenberg, Unter- such. Physiol. Inst, Heidelberg, ii. 1882, p. 37 ; Fredericque, Bull. Acad. Belg., (2) 47, p. 217. Colouring matter from intestine. MacMunn, Proc. Philos. Soc., Birmingham, iii. 1881-3, P- 3^9- Vascular system. Jaquet, Mitth. Zool. Stat. Naples, vi. 1885 ; Vejdovsky, op. cit. pp. 1 1 2-1 20. Double dorsal vessel. Beddard, Proc. Roy. Phys. Soc., Edin- burgh, 1885. Blood and blood corpuscles. Vejdovsky, op. cit. p. 118; Ray Lankester, Q. J. M. xviii. 1878, p. 72 ; Blomfield and Bourne, Q. J. M. xxi. 1 88 1. Endothelium of vessels. D'Arcy Power, Q. J. M. xviii. 1878. Coelomic, i. e. amoeboid cells. Kiikenthal, J. Z. xviii. 1885. Nephridia. Claparede, op. cit. ante, p. 615; Gegenbaur, Z. W. Z. iv. 1853. In general and development. Vejdovsky, op. cit. pp. 120-129. Of Acanthodrilus multiporus. Beddard, A. Sc. N. (6) xix. 1885. Of a N. Z. Acanthodrilus. Id. Z. A. viii. 1885. Sexual organs. Hering, Z. W. Z. viii. 1856-57 ; Bergh, Z. A. ix. 1886. In general and development. Vejdovsky, op. cit. pp. 125-151. Development of sperma- tozoa, vesiculae seminales. Blomfield, Q. J. M. xx. 1880; cf. Jensen, Archives de BioUv. 1883. Spermatophores. Fraisse, Arb. Zool. Zoot. Inst, Wurzburg, v. 1882; Vejdovsky, op. cit. p. 154. Accessory organs. Vejdovsky, op. cit. p. 135. Homo- logics. Id., op. cit. pp. 157-161. Regeneration of excised parts. Miss Fielde, Proc. Acad. Nat. Sc. Philadelphia, pt. i, 1885; cf. on Lumbriculus, Billow, Z. W. Z. xxxix. 1883. Bifid Earthworms; 1 For 'homoplasy,' see Ray Lankester, 'Use of the Term Homology,' &c., A. N. H. (4) vi.' 1870. COMMON EARTHWORM. 209 Bell, A. N. H. (5) xvi. 1885. Regeneration in marine Chaetopoda. De Quatrefages, A. Sc. N. (3) ii. 1844, p. 100; Claparede, Ann^lides du Golfe de Naples, 1868, p. 30. For lit. of subject in general, see Milne Edwards, Lesons sur la Physiologic et 1' Anatomic compare'e, Paris, viii. 1863, p. 301 et seqq. ; or Fraisse, Die Re- generation von Geweben und Organen bei der Wirbelthieren, Cassel und Berlin, 1885 ; cf. Horst, Z. A. ix. 1886. 41. COMMON EARTHWORM (Lumbriciis terrestris, s. Agricola), Dissected so as to show its nervous system. THE integument has been divided down the middle dorsal line and fastened out on either side. The entire digestive tract with the exception of the buccal cavity, most of the nephridia or excretory organs, and the septa dividing the body into compartments, have been removed. Of the reproductive organs only the spermathecae or receptacula seminis, two globular white sacs, have been left in situ on the right side. They open re- spectively between the ninth to tenth and tenth to eleventh somites, on a level with the dorsal row of setae. The two lobes, making up the supra- oesophageal or cerebral ganglia, are pyriform, and have their broader ends apposed to each other in the middle line. A thick nerve passes off from each of their outer or narrower ends. It bifurcates, and ends in a plexus in the prostomium, on which are situate numerous sense-bodies. A right and left oesophageal commissure surround the passage from the buccal cavity to the pharynx and connect the supra-oesophageal to the first ganglion of the ventral nerve-cord. This cord extends to the posterior extremity of the body. It takes the shape of a thick band in which ganglionic enlarge- ments are recognizable with difficulty for a space corresponding with that , occupied by the pharynx, oesophagus and reproductive organs. Posteriorly to the fifteenth somite it becomes more slender, and the ganglia more distinct. Finally, for a length nearly equal to the posterior half of the animal, it becomes thicker and moniliform, the ganglia being plainly marked but closely apposed. The terminal ganglion is, contrary to what is seen in some Vermes and many Arthropoda, smaller than those which precede it. The two rows of paired setae are well seen on each side in most of the somites. The enlarged inner copulatory setae of the two somites, fifth and sixth in order anteriorly to the clitellum, as well as of the clitellum itself, can be readily distinguished. In the interval between each inner row of setae and the nerve-cord in the fifteen anterior somites, a longitudinal mus- cular fascicle is seen passing forwards. It is inserted on the outer ends of the supra-oesophageal ganglia and on the commissures, and acts as a re- tractor muscle to these parts. Between the two rows of setae of the eighth P 2io DESCRIPTIONS OF PREPARATIONS. to the thirteenth somite inclusive, may be seen, on the right side, the cap- sulogenous glands of D'Udekem. These structures appear to be merely enlarged setiparous sacs, and not glands. They have been supposed to secrete the albumen surrounding the ova in the cocoon. The nephridia are left in situ on either side of the nerve-cord in some of the posterior somites. The supra-oesophageal or cerebral ganglia belong developmentally to the pro- stomial region of the first somite of the body, but generally shift in the adult backwards even as far in some instances as the 3rd or 4th somite. They retain a position in Microchaeta Rappi, &c., in the first (buccal) somite. The oesophageal commissures are composed of one or two fibrous cords. The ventral nerve-cord lies internally to the longitudinal muscular coat, except in the Lumbriculidae, where it lies next to the circular coat. The cerebral ganglia of Aeolosoma are, however, in continuity with the hypodermis, but a ventral cord is absent in this genus. In a few Chaetopoda, in Protodrilus, Polygordius, and Histriodrilus (•= Histriobdella\ the ventral cord as well as the cerebral ganglia are similarly continuous, and the three genera in question have been made into a separate group of Archi-annelidae (Hatschek). Saccodrrus, in which the nervous system is also hypodermic in posi- tion, is in other respects not so archaic a type, and has been classed by Foettinger apart from other Chaetopoda as Archi-chaetopoda. The nerve-cord is composed of (i) an external layer of polygonal epithelial cells belonging to the peritoneum; (2) a coat of longitudinal muscle fibres, which does not extend up the commissures to the supra-oesophageal ganglia; (3) of a neurilemma, formed by the cord itself; and (4) the nervous matter proper, with a neuroglia or supporting connective tissue. The ganglion cells are found on the anterior surface of the supra-oesophageal ganglia, and as a layer on the ventral surface of the ventral cord, but not in the oesophageal commissures. They have no sheaths. In the first part of the cord they form a perfectly continuous layer (a fact denied by Vignal), with a right and left and two median aggregations, while in the posterior region these aggregations are separated from one another. This continuous arrangement is found only in Lumbriddae and Lumbriculidae among Oligochaeta. The number of cells does not appear to be much increased in the ganglionic enlargements, which are due chiefly to the greater amount of fibrous matter present at the origin of the chief nerves. According to Claparede the cord is divided into a right and left half by a median connective tissue septum. The nerve-cord has in Lumbriddae three special blood-vessels running longi- tudinally within the muscular sheath, one subneural and two lateral, one on either side. The three are connected by ventral loops just behind each ganglionic enlargement. The lateral vessels give off a branch which accompanies the paired nerves (infra), the median a branch which accompanies each septal nerve (infra), and all three are connected to a capillary plexus ramifying in the substance of the cord and round the ganglion cells. The presence of these vessels was supposed to be a distinctive feature of the Oligochaeta terricola, as opposed to Oligochaeta limi- cola, but the subneural vessel is absent in the terrestrial Megascolex, Perichaeta Houlleti, and Microchaeta Rappi, in the semi-marine Pontodrilus, but present in the aquatic Criodrilus. COMMON EARTHWORM. 311 The nerves originating from the cerebral ganglia usually break up into a plexus, in which ganglion cells are interpolated. The network is especially well-developed and visible in Tubifex, and in Limnodrilus and Anachaeta its ganglion cells are aggregated into special ganglia. The first ganglion of the ventral chain generally gives off many nerves ; the following ganglia a pair of nerves on either side. A single septal nerve arises also on each side, between successive ganglionic enlargements, and is distributed to a septum. A pharyngeal plexus or ' Vagus ' system is derived in most, probably in all, Oligo- chaeta from the oesophageal commissures. In some of the lower Oligochaeta there are distinct pharyngeal ganglia. In the Earthworm the system consists usually of a more or less elongated mass lying on either side of the pharynx, continuous with a rich plexus, the fibres of which are ultimately lost among the muscular structures. Ley dig and Vignal state that ganglion cells are found in all parts of this plexus ; Claparede, on the contrary, that none occur in the plexus so far as it is visible to the naked eye, but that they are present on the finer branches among the muscular bundles. A cord of ganglion cells in continuity anteriorly with the cerebral ganglia runs down each side of the body in most Oligochaeta. It is especially easy to see in JVatSj and is contained in the hypodermis. «It is said to supply the muscles of the head, sacs of the setae, and the nephridial apertures ; and is, perhaps, in connection with the ganglion cells, which have been found in the walls of the digestive tract in some instances. A zone of ganglion cells encircles each somite in the Naid Slavina appendiculata, is connected to the lateral cords, and supplies the tactile eminences (infra]. In Lumbricidae the lateral cords are to be detected clearly in the young posterior somites, but they are resolved in the older anterior somites into scattered cells. As to organs of special sense. Oligochaeta never possess otocysts. Eyes are found only in some Naidomorpha \ supposed gustatory organs in the pharynx of Enchytraeidae and Limnodrilus ; and olfactory (?) organs as a couple of ciliated pits on the head in Aeolosoma, Ctenodrilus, Parthenope. The last-named structures occur also in some Polychaeta. Tactile organs, however, are commonly distributed, and in various forms; as (i) a hypodermis cell furnished with an external tactile seta, and continued basally into a nerve fibril, in its turn often connected to a ganglion cell, — found in numbers on the prostomium of Aeolosoma, Chaetogastridae, and Naidomorpha ; (2) tactile papillae, in which the hypodermis cell is protrusible, and furnished with short setae ( Chaetogastridae] ; (3) tactile eminences, apparently composed of aggregations of tactile hypodermis cells, arranged fifteen to twenty in number in a zone on each somite of Slavina appendiculata ; and (4) goblet bodies, or aggregations of very delicate hypodermis cells provided with sense-hairs, found most plentifully on the prostomium and buccal somite of Lumbricidae, more sparingly on the anterior somites of the body, but especially round the setae. A single goblet body is found also on each side of the somites of the Lumbriculidae, seated on the lateral cord of ganglion cells. They become enlarged on the clitellum of the Lumbriculid Rhynchelmis, where gland cells occur among the sense cells. See on the subject Vejdovsky, op. cit. pp. 96-100. On the dorsal aspect of the nerve-cord there are to be found three remarkable P 3 2 i 2 DESCRIPTIONS OF PREPARA TIONS. giant tubular fibres. They are found in all Oligochaeta except in Aeolosoma, Phrae- oryctes, and Branchiobdella, and the number given is usually that found in the Earth- worm. They occur also in many Polychaeta, where their number varies from one as in Eunice, to six as in Glycera. These fibres do not extend into the oesophageal commissures, and they taper anteriorly and posteriorly. Each fibre is composed of a doubly contoured sheath with clear contents. No connection between the giant fibres and nerve fibres has ever been demonstrated. On the contrary the giant fibres are separated from the nerve-cord by the inner neurilemma, and they are im- bedded in a connective tissue sheath containing reticulate cells. They appear to have a purely supporting function, and the apparatus is hence termed ' Neurochord ' by Vejdovsky (op. cit. pp. 86-87), who compares it physiologically with the noto- chord of Chordata. Nervous system. Claparede, op. cit. ante, p. 585. Ley dig, Vom Bau des Thierrischen Korpers, i. Tubingen, 1864, pp. 139, 168. Id. Tafeln zur Vergleich. Anatomic, Tubingen, 1864, Taf. i. fig. 5; Taf. iii. fig. 8; Taf. iv. figs. 7 and 8; Taf. v. figs, i and 2. Vignal, A. Z. Expt. (2), i. 1883, p. 373. Of Oligochaeta in general. Vejdovsky, op. cit. pp. 79-96 ; and on Neurochord, p. 87. Organs of special sense in Oligochaeta. Vejdovsky, op. cit. pp. 96-100. 42. MEDICINAL LEECH (Hirudo medicinalis), Suspended to show the external form of the body and the coloured bands which differentiate the variety H. medicinalis from the variety //. officinalis. THE animal is suspended by the anterior extremity which is formed by the funnel-shaped buccal cavity or anterior sucker which leads to the mouth, and is not separated from the rest of the body by a constriction in Leeches with jaws (Gnathobdellidae) as it is in Leeches with a protrusible proboscis (Rhynchobdellidae). The body itself has a flat or slightly concave ventral surface and a convex dorsal surface : and it is terminated by a disc- like solid posterior sucker which is formed by the fusion of posterior em- bryonic somites, according to Leuckart seven in number. The body is annulated, and, according to Whitman, H. medicinalis has in all one hundred and two rings, representing twenty-six somites. Conse- quently the annuli do not represent somites : they are, on the contrary, due to a secondary and imperfect division of them. It has been pointed out by Whitman that certain of the annuli or rings bear what he terms seg- mental papillae. These organs resemble in histological structure the eyes, but reduced in size and deprived of pigment, and they may be regarded as the metameric or serial homologues of those organs. In Hirudo and some other Leeches there are normally fourteen segmental papillae, eight on the dorsal and six on the ventral aspects of the annuli upon which they occur. The eight dorsal papillae are arranged as a median pair, with three organs to either side of it, an inner, outer, and marginal organ. The MEDICINAL LEECH. 213 first pair of eyes replaces the median pair of papillae, the remaining four pairs a papilla of the inner series. The first and second annuli of the body carry a pair of eyes ; the third annulus does the same, but it is fol- lowed by an annulus with neither eyes nor segmental papillae : and the two therefore go together. The fifth annulus bears eyes ; the eighth and eleventh only segmental papillae, and these three annuli (fifth, eighth, and eleventh) are each followed by two annuli which have neither eyes nor papillae. Whitman therefore concludes that the first and second somites are represented by a single annulus, the third by two annuli, the fourth to the sixth inclusive by three apiece. The seventh and succeeding somites, up to the twenty-second inclusive, are each composed of five annuli, the first of them bearing the segmental papillae. The twenty-third somite has three, the twenty-fourth to the twenty-sixth two annuli apiece. The buccal annuli are the fifth and sixth, the post-buccals the seventh and eighth, and these two pairs of annuli are fused ventrally. A pair of nephridial pores opens on the ventral surface of the last annulus of the sixth to the twenty-second somite inclusive. This annulus can be readily recog- nised in H. medicinalis by the fact that it is the one that carries a large black spot in the middle of the three light lines which traverse the body lengthwise on each side of its dorsal aspect. The male orifice lies between the second and third rings of the tenth somite, i. e. the thirtieth and thirty- first annuli of the body. The female orifice occupies a corresponding position in the eleventh somite. The ninth, tenth, and eleventh somites constitute the clitellum or region which secretes the cocoon. The anus lies either in the last annulus or in front of it. It may be added that papillae occur also on the sucker, but they do not afford any clue to its composition. The grouping of the annuli, as above detailed, is characteristic of the genus Hirudo. Slight differences are observable in allied genera. The medicinal Leech varies much in its colouration : and no less than sixty-four varieties have been enumerated. The variety H. medicinalis has in the natural state the dorsal surface greenish grey, with three rust-red longitudinal streaks on either side. The middle one of these three streaks has a black spot more or less distinct on each annulus, and it may be readily seen that one of these spots at regular intervals is much enlarged. It marks the last annulus of a somite. There is also a small black spot in the same annulus interrupting the inner light line. The ventral surface is greenish yellow spotted with black, or else black. The action of the spirit soon de- stroys these bright colours, as it has done here. H. officinalis has a median dorsal green band bordered by a red or brown line. The lateral dorsal regions are green with black and reddish-brown spots, sometimes grouped in two longitudinal lines. The amount of black pigment is very variable, and sometimes the red prevails. The ventral aspect is green, and as a rule not spotted. There are many intermediate forms between these two 214 DESCRIPTIONS OF PREPARATIONS. varieties, but however much the colouration may change, the form of the teeth distinctive of the medicinal Leech remains constant (Leuckart). The somite is not always composed typically of five annuli in Leeches. In Branchellion it has three, in Pontobdella four. The Gnathobdellidae appear to agree with Hirudo. The ten-eyed Leeches of Japan possess six segmental papillae on the dorsal, and six on the ventral surface of the annulus that bears them. The clitellar somites form the cocoon which contains the ova, a certain number of spermatozoa with albumen, the latter absent in Piscicola. The substance of the cocoon is secreted by the clitellar glands (infra]. When it is fully formed, the animal withdraws its head, and the two ends of the cocoon close up. The openings are plugged by hardened albumen, through which the young Leeches eat their way when ready to escape. Hirudo^ like Aulostoma, lays its cocoon in damp earth. The cocoon is usually attached to some foreign object in the water. Its shape is variable among Leeches. The surface of the body is covered by a delicate cuticle perforated by pores, which are the apertures of unicellular glands. This cuticle is continually under- going regeneration, the old one being peeled off, as may be readily seen in a Hirudo kept in confinement. The epidermis or hypodermis of the medicinal Leech consists of mallet-shaped cells about j^Vff of an inch long. The heads of the mallets are placed superficially beneath the cuticle. The handles are consequently separated by spaces, into which processes of pigmented connective tissue cells with capillaries make their way. The nucleus lies in the handle of the mallet. The hypodermis cells vary much in character in different Leeches, and in Rhynchobdellidae the pigment cells and capil- laries do not always intrude between them. From the hypodermis cells are produced numerous unicellular glands. In Hirudo these glands pass into the dermis, and their ducts are consequently long. Those of the general body-surface appear to be simply mucous glands. The more deeply situated glands are (i) clitellar glands, which occur in the clitellar region in groups of four to five, the glandular part lying in the longitudinal muscle layer ; (2) prostomial glands of doubtful function, but probably found only in Gnathobdel- lidae, with ducts opening round the edges of the buccal cavity, and their contents not staining with borax carmine; (3) salivary glands, which belong to the buccal cavity itself, with ducts opening on the ridges which bear the teeth, and with contents which stain with borax carmine. The hypodermis cells are also modified to form sensory cells, with which nerves are continuous. In the head region are found the goblet-like bodies and the eyes, the visual nature of which is doubted by Carriere. Of the latter there are ten, two on the first and second annulus, and two on the first annulus of the three following somites. Each eye consists of a projecting cap of short non-pigmented hypodermis cells, covered by cuticle and containing at their outer ends small re- fractile bodies : of an external layer of pigmented cells, separated by a lamina (?) from a layer of nucleated clear cells with well-defined walls and protoplasm largely replaced by a vacuole, and lodging a corpuscle of unknown significance : and of a MEDICINAL LEECH. 215 slender core of hypodermic cells as they appear to be from a comparison with the segmental papillae and from their histo-chemical properties (Whitman). The nerve enters at the side but near the base, and at its point of entrance there appear to be ganglion cells. The goblet-bodies and segmental papillae have a bulb-like thickening of the hypodermis cells, no pigmented coat, and relatively few clear cells. The former are exceedingly numerous (about sixty in number), especially on the anterior margin of the prostomium. They are situated on branches of the same nerves as the eyes. The Leech is sensitive to light, but the prostomial region appears to be equally if not more sensitive to other impressions (touch, taste ?). The eyes of Clepsine differ from the eyes of Hirudo, and consist, according to Carriere, of a semi-globular cup of large pigmented retinal cells, con- taining a number of transparent and smaller cells. They are situated in the muscular coat. The eyes of Nephelis appear to be similar ; those of Pisticola, a Leech in which they occur on the posterior as well as the anterior sucker, appear to be more simple, judging from Leuckart's figures. The visual nature of these eyes seems certain. The sub-hypodermic tissues consist of a connective jelly-like matrix imbedding nucleated corpuscles with very fine branches, and more or less pigmented, as well as ' vaso-fibrous ' tissue: The latter consists of nucleated branched cells, containing a plentiful supply of pigment granules. Their branches sometimes become tubular, and the nuclei project into the lumen, and even drop into it. These cells occur in all parts of the body, and their processes pass out between the hypodermic cells. According to Professor Lankester, the tubular cells may become continuous with the thin-walled capillaries, in the cavities of which free nuclei may sometimes be detected. The botryoidal or so-called ' hepatic ' tissue, which surrounds, but not in immediate contact, the walls of the alimentary tract appears to be merely a modifi- cation of the same tissue. It consists of largish vessels with cellular walls. The individual cells are swollen up, nucleated, and pigmented. The vessels thus formed partly end caecally, partly form a plexus continuous with the ordinary thin-walled capillaries. The corresponding pigmented cells of Rhynchobdellidae never become tubular, and usually remain more or less rounded. In the same group, vacuolated cells and fat cells are found in the connective tissue matrix. They do not occur in Gnathobdellidae. The muscles of the body are imbedded in the connective tissue matrix, leaving a sub-hypodermic layer free. They are arranged in an outer circular and a deep longitudinal series. Between the two are diagonal fibres* In Hirudo connective tissue with longitudinal fibres intervenes between the circular and diagonal layers. A set of dorso-ventral and radial fibres takes the place of the septa of Chaetopoda. The fibres of this series near the centre of the body pass between the caeca of the first portion of the alimentary canal. The outer ends of the radial fibres are branched, and end close under the hypodermis in the sub-hypodermic connective tissue. The muscle fibres themselves possess an outer fibrillate layer and an inner granular medulla with a nucleus. Their ends are often much branched, especially on the walls of the alimentary canal. The amount of connective tissue matrix present, as compared with muscular fibres, is very variable in Hirudinea. Where there is little of it, as in Clepsine and Nephelis, the worm is perfectly firm and rigid to the touch when living. In Aulo- 2i6 DESCRIPTIONS OF PREPARATIONS. stoma and Haemopis, where it is plentiful, the animal is always limp. Hirudo occupies a middle position between these two extremes. Anatomy of the Hirudinea. A. Gibbs Bourne, Q. J. M. xxiv. 1884. Remy Saint-Loup, A. Sc. N. (6) xviii. 1884. Leuckart, Die Parasiten, i. 1863, Leipzig. Moquin-Tandon, Monographic de la famille des Hirudine'es, with Atlas (ed. 2), 1846, Paris. Leeches of Japan, Whitman, Q. J. M. xxvi. 1886. Leech, Mclntosh, Encyclopaedia Brit. (ed. ix,), xiv. Genera, &c. Diesing, SB. Akad. Wien, xxxiii. 1858; Id. Systema Helminthum, 1850, i. p. 433, and Leuckart, op. cit. Somites, Whitman, op. cit. supra ; Proc. Amer. Academy, xx. (n.s. xii.) 1885, or American Naturalist, xviii. 1884. Cuticle, hypodermis, vasifactive and botryoidal tissue. Ray Lankester, Q. J. M. xx. 1880. Dark-green pigmented network of Hirudo. Joseph, Z. A. vi. 1883. Muscle. Shore, Nature, xxvi. 1882, p. 493. Nerve endings in ditto. Hansen, Archives de Biol. ii. 1881. Eyes and Segmental papillae. Whitman, Q. J. M. xxvi. 1886. Eyes. Car- riere, Sehorgane der Thiere, Munchen und Leipzig, 1885 ; Ranke, Z. W. Z. xxv. 1875. Ditto and goblet-bodies. Ley dig, Arch. f. Anat. und Physiol. 1861 ; Id. Tafeln zur Vergleich. Anatomie, Tubingen, 1864, Taf. ii. figs. 5, 7; Taf. iii. figs, i, 2, 3. Sense-cells of hypodermis. See Gibbs Bourne, op. cit. p. 434, PI. 27, fig. 15. Pigment of skin. MacMunn, Proc. Birmingham Phil. Soc. iii. 1881-3, p. 389. 43. MEDICINAL LEECH (Hirudo medidnalis), Prepared to show its laterally sacculated stomach and the intestine. A stiffening injection of gelatine was thrown into the digestive tube, the specimen hardened in spirit and then dissected. THE integument has been divided down the middle dorsal line and reflected to either side : the portions of the vascular system interposed be- tween the digestive tract and the body-walls cut away, and the entire cavity of the 'crop '.and its diverticula exposed by the removal of their dorsal wall. Anteriorly to the crop is the pharynx with a villous exterior. This appearance is due to the presence of numerous unicellular salivary glands and to the cut ends of radial muscular fibres. The pharynx is succeeded by the crop which has thin walls. Lateral diverticula or caeca which occupy five-sixths of the entire body are appended to it on either side. These caeca are really segmental dilatations of a central tube. There are eleven of them in all. The first is small. The second appears to be double. The third has much the same appearance but not so well marked. It is due to the great development of a partial septum which may be seen in the suc- ceeding seven caeca projecting backwards from the anterior wall. In some species of Clepsine the caeca are very distinctly bifid at their outer ends. The last pair of caeca, the only pair in Aulostoma, are of very great length and bending sharply almost immediately at their commencement so as to become apposed to each other along the middle line, are prolonged back- MEDICINAL LEECH. 217 wards to a point on a level with the commencement of the rectum, and nearly as far as the end of the body. The form of the caeca, it should be noted, depends very much on their state of distension. This region of the digestive tract serves, first, as a crop or reservoir for the blood which forms the sole food of the animal, and, secondly, as a place where slow changes go on in the various constituents of that fluid. The oxy-haemoglobin is extracted by degrees from the corpuscles : it is reduced to haemoglobin and crystallised, and changes gradually take place in the corpuscles. Between the two last caeca lies the stomach or * gastro-ileal' (Gratiolet) section of the alimentary canal. A black bristle has been passed into it. It communicates by a narrow aperture with the crop, and at its commence- ment there is a pair of small caeca, one on the right, the other on the left, directed forwards. These caeca and the stomach are much larger in the Horse-leech (Aulostoma gulo). The stomach is very vascular, and it has a villous interior with a spirally arranged valve. The contents of the crop enter into it very slowly, and the red colour of the blood then changes from dark-red to green. The posterior end, colon or intestine, of this section of the alimentary canal is little vascular and contains no valve. It is followed by a short rectum of small calibre which terminates in a dorsally-placed anus, as in all Leeches except Acanthobdella where it opens in the centfe of the posterior sucker. From a developmental point of view the pharynx and rectum must be regarded as invaginations from the exterior, i. e. as stomodaeum and proctodaeum, while the rest of the canal is archenteron, i.e. lined by endoderm or hypoblast. The ventral nerve chain may be seen in part through the walls of the stomach. The spot where the walls of the caeca meet centrally (i. e. in the median tube) corresponds very nearly in most instances with a ganglion. On either side of the crop and adherent to the inner surface of the reflected body-walls may be seen remains of the botryoidal tissue. It is arranged chiefly in four bands, two dorsal, two ventral, close to the walls of the digestive tract but separated from it by a layer of vaso-fibrous tissue and capillaries. A section taken through the pharyngeal region shows, according to Gibbs Bourne, the following structures passing from within outwards: (i) Pharyngeal epithelium composed of minute cells; (2) three ridges, one dorsal, two lateral, composed of salivary ducts and radiating muscles, the branched ends of which abut upon the hypodermis ; (3) a circular layer of muscles very dense and compact ; (4) the longitudinal muscle layer of the body wall, in which occur blood sinuses, vessels, and salivary glands ; (5) the diagonal and circular muscle layers of the body ; and (6) the hypodermis and cuticle. To the three pharyngeal ridges correspond the three muscular jaws. They bear at their edges in the medicinal Leech about 80- 90 fine chitinoid teeth. These teeth contain lime carbonate both in Hirudo and DESCRIPTIONS OF PREPARATIONS. Aulostoma. Haycraft (P. R. S. xxxvi. 1883-84, p. 478) has proved that the secretion of the pharyngeal glands has the power of arresting coagulation in blood apparently by destroying the coagulation ferment. The muscle cells of the ' crop ' are for the most part disposed in a transverse direction, and their ends are branched. Its epithelium is low, columnar, and the cells, according to Gibbs Bourne, may be seen giving off clear droplets into the blood. In Aulostoma the alimentary epi- thelium is ciliated. Its food consists of worms, &c. Moquin-Tandon states that when young this Leech possesses distinct lateral caeca to the crop in addition to the posterior pair. The region of the crop is not sacculated in Nephelis, Trocheta, or Pontobdella. In the last two it is constricted at intervals, and the last-named possesses a posterior azygos caecum underlying the stomach. Trocheta is carni- vorous like Aulostoma. The body-cavity or coelome is in all adult Leeches almost obliterated by connective tissue growths. This process of obliteration of the coelome is termed by Gibbs Bourne diacoelosis. The remains of the coelome are much more conspicuous in the Rhyncho- than in the Gnathobdellidae. In the medicinal Leech its chief remains are in the form of the dorsal and ventral sinus, the latter lodging the nerve-cord. They are in direct connection only by means of the dorsal sinus of the gastro-ileal section of the alimentary canal. Other remains of the coelome are found in the network of vessels surrounding the testes, &c. The blood-vessels, which with their branches have muscular walls, are represented by a right and left longitudinal trunk which anastomose at the anterior and posterior extremities of the body. These trunks give off in each somite latero-abdominal vessels which anas- tomose ventrally, and two sets of dorsal vessels, short latero-lateral and long latero-dorsal branches. In the gastro-ileal region the latter anastomose inter se dorsally by means of their posterior branches. Anteriorly to this region they are connected only through the capillary system of thin-walled vessels. There is a superficial network of fine capillaries which penetrate the hypodermis, and more deeply pass into an intermediate layer connected with the lateral vessels and with the botryoidal tissue. The latter forms the deepest layer, connected on the one hand with the vessels given off from the lateral vessels, on the other hand with the sinuses. It tends to form a secondary coelome (= metacoelosis, Gibbs Bourne). The dorsal and ventral sinuses, according to Gibbs Bourne, communicate with (i) the cutaneous network, (2) the capillaries of the crop, and (3) of the stomach, and (4) the sinuses (moniliform hearts of Brandt) which surround the nephridial funnels and lie upon the testes. The thin-walled capillaries of the Leech possess no endothelium and no nuclei. The walls of the vessels in the botryoidal tissue are formed solely by the pigmented cells themselves. The blood-plasma in the Gnathobdellidae •, but not in Rhyncho- bdellidae, is coloured red by haemoglobin. It contains amoeboid corpuscles, and here and there Prof. Ray Lankester detected nuclei set free from the walls of the developing capillaries, as in the Earthworm. But they appear to be very scanty in number in the Leech. Bite of Leech, Carlet, C. R. 96, 1883. Lime in teeth, Schneider Zool. Beitrage, i. 1885. Alimentary canal. Gratiolet, A. Sc. N. (4), xvii. 1862, p. 182; Gibbs Bourne, op. cit. p. 492. MEDICINAL LEECH. Digestion of blood by the common Leech. Stirling, Journal Anat. and Physiol. jxvi. 1882. Vascular system and coelome. Gibbs Bourne, op. cit. p. 453 ; Jaquet, Mittheil. Zool. Stat. Naples, vi. 3, 1885. 44. MEDICINAL LEECH (Hirudo medicinalis), Dissected so as to show its nervous system. A PART of the pharynx with the jaws has been left in situ, and a black bristle also passed down it through the oesophageal or nerve-collar. The supra-oesophageal ganglion is seen above the pharynx, part of the glandular and muscular walls of which have been removed to show it in situ. It lies immediately behind the dorsally placed jaw, which is very visible in this preparation. It is connected by commissures forming an extremely narrow ring to the ventral chain of ganglia. This chain, counting the infra- oesophageal ganglion as the first of the series, numbers twenty-three ganglia in all. The infra-oesophageal ganglion and the second ganglion, which is closely apposed to it, are to be seen with difficulty here. The first ganglion, easily seen, is the third of the series. The longitudinal commis- sures between the third and fourth, the fourth and fifth ganglia increase in length, though they are shorter than those connecting the ganglia belong- ing to the middle region of the body. The sixth ganglion is concealed by the prostate gland at the base of the muscular penis. It is close to the seventh ganglion. The ganglia at the posterior extremity of the body, beginning with the nineteenth ganglion, are again closely aggregated to- gether. The last ganglion is much larger than any of the series except the first. It can be readily seen with the naked eye that nerves are given off laterally from each ganglion. They are in reality paired, but one branch is dorsal, the other ventral. There are no nerves given off between the gan- glia in the Leech, as there appear to be in the Earthworm. But in the Leech the ganglion cells are really aggregated in the ganglia, not scattered along the whole cord as in the last-named worm. In front of the supra-oesophageal ganglion lie three minute ganglia closely connected to it, one median and two lateral. The nerves given off by these ganglia supply the three jaws. The nerves originating from the supra-oesophageal ganglia supply the eyes and the goblet-shaped organs of Leydig. The infra-oesophageal ganglion, according to Leuckart, is composed of three ganglia in the embryo which fuse in the adult. As figured by Leydig, it is composed of two halves, right and left, connected by five transverse fibrous commissures. 220 DESCRIPTIONS OF PREPARATIONS. Through the four interspaces between these commissures and in front of the first one of the series small bundles of muscle fibres pass vertically. According to Ley- dig each ventral ganglion is pierced by a central hole which transmits a similar muscular bundle. The terminal ganglion of the chain is composed of seven em- bryonic ganglia. According to Vignal the distinction between these ganglia can be traced in the adult. There appears to be a larger number of fused embryonic ganglia in some other Leeches. The infra-oesophageal ganglion gives off five nerves on each side. The two nerves given off by the other ganglia of the chain on each side are dorsal and ven- tral respectively. The ventral nerve of the third ganglion in the chain and its successors has a minute ganglion at the spot where it first bifurcates. The ganglion cells in the Leeches tend to accumulate in masses on the outer surface of the ganglia. This follicular appearance is much more marked in some other instances than it is in Hirudo, e. g. in Haemopis and Nephelis. At the point of origin between the two nerves of each ganglion on either side lies a large ganglion cell, the connections of which are not known. The fibrous commissures between the ganglia of the ventral chain are three in number, two large and lateral, one small, median and dorsal. The latter is the intermediary nerve of Faivre, its dis- coverer. Each of these commissures has its own sheath, and all three have in ad- dition a common sheath, the outer surface of which is pigmented. The three com- missures, according to Vignal, fuse centrally in each ganglion. Brandt discovered in connection with the ventral surface of the crop a median nerve which bifurcates posteriorly in correspondence with the last pair of caeca. This nerve forms a rich plexus on the walls of the crop, and the filaments of the plexus are in connection with numerous ganglion cells. The exact mode of con- nection of this sympathetic system, which appears to be chiefly in relation with the muscles of the crop, to the ventral chain is not known. It takes place probably through branches of the nerves given off by the ganglia. The ventral chain of ganglia is contained within a ventral blood sinus in all Hirudinea. Nervous system. Leydig, Bau des Thierischen Korpers, Tubingen, 1864; Remy Saint-Loup, A. Sc. N. (6), xviii. 1884. For figures, see also Leydig, Tafeln zur Vergleich. Anatomic, Tubingen, 1864, Taf. i. figs. 4, 6, 7 ; Taf. ii. figs, i, 3, and 5. Cp. figures of other genera on same plates and on Taf. iii. figs. 4 and 5. Histology only. Vignal, A. Z. Expt. (2), i. 1883, p. 343. Lateral ganglion cell, Leydig, Tafeln, &c., Taf. ii. fig. 3, 1. Sympathetic nerve. Id. op. cit. Taf. i. fig. 4 ; Taf. ii. fig. 5. Hermann, Centralnervensystem von H. medicinalis, Miinchen, 1875; Hoff- mann, Untersuchungen iiber den Bau, &c. der Hirudeen, Verh. Ak. Amsterdam, 1880; Kohler, Systeme nerveux de Nephelis, Nancy, 1883, have not been acces- sible to me. MEDICINAL LEECH. 221 45. MEDICINAL LEECH (Hirudo mcdicinalis), Dissected so as to show its reproductive and segmental organs or nephridia in situ. A BLACK bristle has been passed into the pharynx through the nerve- ring, and the ventral chain of ganglia is visible throughout the greater part of its extent. In the middle line, covering the sixth ganglion of the ventral chain, is seen a globular body, projecting chiefly to the left side of the nerve- cord and connected posteriorly with a median siphon-shaped muscular tube. The globular organ has walls partly muscular, partly glandular, and is called consequently the prostatic part of the male intromittent apparatus. The glands appear to secrete the material which forms the spermatophores. The median siphon-shaped tube is the copulatory organ or penis, and its walls contain both circular and longitudinal muscle fibres. From the base of the prostatic body passes to right and left a ductus ejaculatorius. These are each connected respectively to what is easily seen with attention to be a mass of coiled tube of a yellowish colour. The coiled tubes, according to Leuckart, contain at the height of the reproductive season numberless minute globules. The masses in question may be regarded provisionally as vesiculae seminales. Each qoiled tube is continuous with a duct, the vas deferens, which passes backwards parallel to the nerve-cord. It is slightly tortuous. From its inner or median side nine branches arise, each passing to a testis. The nine pairs of testes are globular bodies lying close to the ventral nerve-cord, one behind each ganglion from the eighth to the sixteenth inclusive. They are therefore segmentally arranged. The pro- static apparatus, copulatory organ, and vesiculae lie in the tenth somite (Whitman), the male aperture being median and ventral in the second annulus of that somite. It is difficult to see unless the penis is protruded, as it sometimes is in this Leech and in the common Horse-leech, Aulostoma gulo, when the animal is killed by chloroform. The female organs lie in the eleventh somite (Whitman), therefore in the somite interposed be- tween the first pair of testes and the male intromittent apparatus. Close behind the seventh ventral ganglion may be seen two roundish bodies, the capsules which contain the true ovaries, lying one on either side of the longitudinal nervous commissures. The oviduct is continuous with or rather perforates these capsules. Its anterior part is forked, one branch of the fork passing under the nerve-commissure to the left ovarian sac. The posterior part is single and may be distinguished by its yellow colour. It enters the base of an oval sac, the vagina, which has muscular walls and a cuticular lining and opens by a median ventral aperture in the second annulus of the eleventh somite. 222 DESCRIPTIONS OF PREPARATIONS. Externally to each vas deferens and alternating in position with each testis, is a row of globular sacs only a very little less in size than the testes themselves. These are the vesicles of the segmental organs or nephridia. Each vesicle opens by a pore on the last annulus of its somite. In front of it, and in part externally to it, is a loop-shaped body. This is the chief part of the nephridium, representing the main and apical lobes of that organ. A narrow caecal process passes inwards from it and lies upon a testis. This is the testis-lobe which ends in the nephridial funnel or nephrostome. The funnel lies in a vascular sinus, the perinephrostomial sinus, or the moniliform heart of Brandt. The two pair of nephridia lying behind the last pair of testes in the twenty-first and twenty-second somites (Whitman), possess these sinuses as well as funnels, but this is not the case with the five first pair of nephridia which correspond to the somites six to ten (Whitman). There are in all seventeen pairs of these organs. The azygos and median position of the generative pores is a noteworthy feature, as is also the development of a muscular intromittent organ. It is doubtful however whether they are of any importance in determining the affinities of Hirudinea. The ductus ejaculatorius, according to Leuckart, has a stratum of circular muscle fibres in its walls. The vas deferens ' is surrounded by a space or sinus packed with cells, ' which possess a rather degenerate appearance ' (Bourne). The sacs in which the true ovaries are lodged contain similar amoeboid corpuscles. It has been suggested by Gibbs Bourne that these spaces represent a portion of the original coelome, and the cells original blood corpuscles, and that they have been closed in by the growth of the connective tissue before haemoglobin appeared in the blood-plasma. The true ovary is a filamentous body, 10 mm. long in some instances. It is coiled within the ovarian capsule. The oviduct opens into the capsule, but it is not simply continuous with the walls of that space. On the contrary, its anterior end is disposed in coils within it. These facts support the view quoted above from Gibbs Bourne as to the nature of the capsule. The median or single portion of the oviduct is surrounded by glands, by which the albumen mixed with the ova in the cocoon is secreted in all probability. The vagina has muscular walls, is lined by cuticle, and receives the spermatophores in congress. These bodies are stated to contain not only spermatozoa, but corpuscles similar to those found in the coils of the vesiculae seminales. They are resolved in the vagina, and the spermatozoa, now free, are said to penetrate into the ovarian capsules. The ovarian capsules are sometimes of great length, as in Nephelis and Clep- sine. In these genera * egg-strings,' produced by the continuous division of a cell, lie free in the capsular cavity. The formation of the string from a ridge of the epithelium lining the capsule has been observed in Nephelis. The ova sometimes degenerate, and Schneider states that they are destroyed by amoeboid cells in the ovarian capsule. The same thing occurs with the spermatozoa. But the full MEDICINAL LEECH, 223 account of his observations on these points, and on the origin of the ova in various Leeches, has not been accessible to me. Joseph has recently discovered in Clepsine that the vasa deferentia and oviducts arise independently of the sexual glands. In Branchiobdella, (which is probably an Oligochaete), both ovaries and testes are proliferations of cells lining the coelome, and the generative products are carried away by ducts with open mouths, which are, perhaps, modified nephridia. It is quite possible that the glands and ducts have the same origin in the Hirudinea. The extension of the vas deferens through several somites, and the presence of nephridia in the same somites, creates a difficulty for this view, — the same difficulty, however, that recurs in the Earthworm. The Rhynchobdellidae possess neither the tubular intromittent organ nor the muscular vagina. The nephridial funnel of Hirudo appears to be degenerate. It is imperforate and multilobed. The lobed ciliated cells which compose it are set upon a vesi- cular dilatation containing a debris of cells. The various lobes of the nephridium are made up of nucleated cells, varying in size and character. These cells are perforated by intracellular ductules with independent walls. The ductules pass from one cell into another, and are branched, especially in the cells of the main lobe, some of the branches remaining caecal. The main duct has cellular walls, its lumen perforating the cell, as in part, at least, of the Earthworm's nephridium. The vesicle has thin walls with muscular fibres, and is contractile. It is lined by a ciliated epithelium. The cells of the gland are surrounded by a rich network of capillary vessels connected with the lateral blood-vessel, and through the testicular sinus with the main ventral blood sinus. The whole gland is invested with vaso- fibrous tissue. Nephridial funnels appear to be present in all Leeches. They vary in com- plexity. Among the Gnathobdellidae, they are perforate in Nephelis and Trochaeta, and in these genera they open into special spaces developed in the botryoidal tissue, termed by Gibbs Bourne ' metacoelorne.' They are present and usually perforate in Rhynchobdellidae. In Clepsine they open into the ventral blood sinus, and in Pontobdella into a dorso-ventral sinus. In all Leeches the dilatation following the funnel appears to be present. As to the gland, in Pontobdella, Branchellion, and Piscicola, the tubules form a network continuous on both sides of the body and across the ventral median line. The funnels and external openings, however, of Pontobdella are metamerically arranged. Branchellion and Piscicola require further examination. There is no terminal vesicle in Rhynchobdellidae. The Gnathobdel- lidae in general appear to agree more or less closely with Hirudo. Gibbs Bourne has found in the ductules of the nephridium in the medicinal Leech minute clear structureless bodies ; in the liquid of the vesicle, and sometimes in the main duct, bunches of needle-shaped crystals, soluble in nitric acid. According to Leuckart, the embryo of Hirudo has four pairs of nephridia in front of the first persistent pair of the adult, and three pairs behind the last per- sistent (or seventeenth) pair. Genitalia. Leuckart, Die Parasiten, (ed. i.) i.' p. 672. Ovary and ova of Nephelis Aulostoma, Piscicola, Pontobdella. Schneider, Das Ei und seine Be- fruchtung, Breslau, 1883. Egg-strings of Clepsine, Whitman, Q. J. M. xviii. 1878; of Nephelis, Jijima, Q. J. M. xxii. 1882. 224 DESCRIPTIONS OF PREPARATIONS. Sinus ofvas deferens and the ovarian capsule. Gibbs Bourne, op. cit. p. 473. Spermatophores. Cf. Robin, A. Sc. N. (4) xvii. 1862. Nephridia-. of Hirudo, Gibbs Bourne, Q. J. M. xx. 1880; xxii. 1882, p. 337; of other Leeches, Id. Q. J. M. xxiv. 1884, p. 478 ; Schultze, A. M. A. xxii. 1883. 46. TAPEWORM (Taenia serrate?), With the cysts of Cysticercus pisiformis, s. C. Taeniae serratae. IN the upper part of this preparation is suspended a portion of the great omentum of a Rabbit (Lepus ctmiculus). Seven pyriform sacs may be observed attached to it, a group of three on the left-hand side and a fourth at the right-hand corner being especially conspicuous. These sacs are connective tissue cysts which lodge each a single individual and rarely more of the cystic stage (Cysticercus pisiformis) of T. serrata. See next preparation. The presence of the parasite, like the presence of any other foreign body, has had an irritating effect on the tissues of the omentum, and the consequence is the formation of a protective capsule. The brain and the eye are the only two parts of the body in which this capsule is never formed. In the lower part of the preparation is suspended a Tapeworm, Tcenia serrata, which inhabits the intestines of the dog. When a dog devours a rabbit, and swallows an encapsuled Cysticercus pisiformis, the head and neck of the latter develop into the Tapeworm, as it is seen here. The head and neck are displayed in the middle line, bent forwards and downwards. With the aid of a simple lens the head may be seen to possess a slight median projection, the rostellum, at the base of which is a circlet of chitinous hooks. The possession of these hooks constitutes the difference between an * armed ' and an ' unarmed ' Tapeworm. ' Unarmed' Tapeworms are not found in the Carnivora. The head expands below the hooks, and bears four rounded pit-like suckers, one of which is turned towards the observer. It then contracts into a short unjointed neck. The neck begins to broaden out into the body of the worm, and it is at the same time divided by transverse lines into a series of joints. The first joints are almost linear: they then become broader and deeper, and finally their depth becomes greater than their breadth. Their posterior margin be- comes at the same time remarkably prominent. The two last joints are of considerable length and comparative narrowness. They are ripe and ready to be detached. If the lateral margins of the joints are attentively examined, one or other is seen to present near its centre a prominent papilla. This papilla is the projecting edge of the porus genitalis. It alternates generally, but not invariably, in succeeding joints from the right to the left side. TAPEWORM. 225 The head and neck are often termed ' scolex/ the joints, * proglottides/ and the whole Tapeworm, ' strobila.' There are five Tapeworms ordinarily found in the intestines of the dog. Of these, Taenia Echinococcus, derived from the cystic form Echinococcus veterinorum, which occurs in a variety of Mammals, consists as a rule only of a head and neck with three or, at the utmost, four joints. It is not much longer than an intestinal villus. Taenia elliptica, derived from the Cysticercus T. ellipticae, which inhabits the dog-louse (Trichodectes Cants), is recognisable at once by the elliptic shape of its ripe joints and the two pori genitales, one right and the other left. It also possesses a peculiar round rostellum beset with four irregular rows of sixty small hooks. The other three Tapeworms are not so easy to distinguish. They are T. serrata, derived from the C. pisiformis of the Rabbit ; T. marginata, from the C. tenuicollis of the Sheep, &c. ; and T. coenurus, from Coenurus cerebralis of the Sheep, and perhaps of the Rabbit. T. serrata has the largest head (1.3 mm.), and a circle of 38-48 hooks, alternately large and small, the large hooks being 25 mm. long, and the length of the anterior branch of their forked roots very great ; the proglottis, when ripe, is 8 mm. long and 3 mm. broad ; the uterus, when full of ova, has 8-10 main lateral branches beset with numerous short ampullae. There are about 325 joints between the neck and the first ripe joint. T. marginata has scarcely any constriction behind the head : the number of hooks is 22-42, the hooks being smaller than those of T. serrata: the proglottis is larger and longer; the uterus has about eight side branches which are much branched laterally. There are about 510 joints in front of the first ripe joint. T. coenurus has a pyriform head ; 24-32 hooks, which are small ; the proglottis is smaller than that of T. serrata ; the uterus has 20-25 simple branches. There are about 200 joints before the first ripe joint. The egg-shell of all three is about -027 mm. in diameter, but in T. serrata it rarely bears processes as it does in T. marginata. The rostellum of Taeniae often attains a greater size than it does in T. serrata. It has a special system of muscles. The same is true of the suckers which are composed of radial and equatorial fibres almost exclusively. The former deepen the cup, the latter contract its margin. The surface of a Tapeworm is bounded by a vertically striated cuticula which, according to Griesbach, is a product of the gelatinous connective tissue of the body. The striae are due to pores *. Beneath the cuticle is a system of transverse or circular fibres which Leuckart considers to be muscular ; Griesbach as elastic. The substance of the body in Solenophorus consists of a gelatinous matrix which forms a system of trabeculae with lacunar spaces "representing the coelome. The matrix contains elastic fibres and nuclei, small rounded, as well as stellate cells, and therefore closely resembles the connective tissue of Mollusca (Griesbach). A fine granular protoplasm with nuclei (? cellular : see infra under T. lineata) covers the trabecula immediately below the cuticula, but does not appear to extend 1 It has been said that protoplasmic processes extend into these pores from the granular protQ- plasm covering the subjacent connective tissue trabeculae. Q 226 DESCRIPTIONS OF PREPARATIONS. continuously into deeper strata. There is a thin superficial layer of longitudinal muscles and two deeper layers, an outer of longitudinal and an inner of circular fibres, as well as a system of dorso-ventral fibres. The deep longitudinal and circular layers surround a nucleus or core of connective tissue which lodges the generative organs, the longitudinal excretory canals, and longitudinal nerves. In Taenia lineata the cuticula has the same structure as in Solenophoms. There is a similar matrix which is finely granular. It contains large granular, oval, round or amoeboid cells, small fusiform or stellate cells, and scattered nuclei with or without traces of surrounding protoplasm. Tubular or vasiform spaces filled by a granular material lie immediately beneath the cuticula : they sometimes occur empty. There is also a layer of vertically spindle-shaped sub-cuticular cells which give origin to (?) the cuticula. They lose their individuality if the specimen is preserved only in alcohol. The deep longitudinal muscle-fibres are grouped in bundles and retain no trace of their formative cells, as do the dorso-ventral and circular muscle-fibres (Hamann). T. serrata has not been investigated by modern methods with reference to these points. The term ' parenchyma ' is often used in speaking of the connective tissue substance of Cestoda. It is better discarded as the tissue in question is not cellular in structure. The nervous system of T. serrata has been carefully investigated by Niemiec. It consists, as in some other Taeniae, of the following parts. A nerve-ring lies a little below the base of the rostellum. It gives off nerves to the muscles of the hooks, and contains eight slight ganglionic enlargements from each of which originates a stoutish nerve passing backwards. Two pairs of these nerves (A, A : A, A\ situated at opposite extremities of the same diameter of the ring, unite each with one of the two lateral principal ganglia. The other two pairs (B, B : B, B) unite with the secondary transverse commissure and the polygonal commissures (infra]. The two lateral principal ganglia are connected by a primary transverse commissure in the middle of which is a voluminous central ganglion. This ganglion gives origin to a slender secondary transverse commissure which lies at right angles to the primary transverse commissure and forks at either end. Each branch of the two "forks unites with one of the nerves B, B, &c. supra, and with the superior polygonal commissure. The lateral principal ganglia and the nerves J?, B, &c., are united by two ring-like polygonal commissures, one superior, the other inferior. The points of union of the nerves B, B, &c., with the secondary transverse and the polygonal commissures are swollen and form secondary ganglia, from which, as well as from the lateral principal ganglia, nerves are given off to the suckers The nerves B, B, &c., are continued backwards through all the joints as slender filaments, two on each surface. Three longitudinal lateral nerves, a median stout nerve with a slender nerve on either side, extend backwards in a similar manner, and lie to the outer side of the longitudinal excretory trunks. They are said to degenerate in the ripe joints by Hamann. It is possible that branches may originate in some instances from the lateral nerves. Riehm states that they possess a swelling close to the posterior margin of each joint in Dipylidium pectinatum (=T. pectinata in part) of the Rabbit from which nerves pass both inwards and outwards. Similar swellings exist in Ligula and Schistocephalus (Kiess- ling). Their ganglionic nature is by no means certain. TAPEWORM. 337 The nervous system of Bothriocephalus and Ligu/a is said to be simpler than that of the Taeniae. The excretory system of T. serrata appears to correspond in its main features with that of most Taeniae, a general description of which is given in the account of the Class Cestoda. Ciliated funnels have been detected by Fraipont in this Tapeworm and in its Cysticercus (Archives de Biol. i. 1880, p. 439). Leuckart figures the anterior anastomosis between the longitudinal vessels as consisting of a ring-like vessel with branches in connection with it (Parasiten (ed. 2), i. p. 379, fig- J53)- Two longitudinal vessels are certainly present, perhaps four. P. J. Van Beneden does not figure a cross anastomosis at the posterior margin of each joint, nor does he mention the presence of valves ; points which lack of material has prevented me from determining. He mentions, however, that treatment of the scolex with acetic acid causes an evolution of Carbon dioxide in the excretory canals which escapes by the foramen caudale or aperture of the pulsatile vesicle. Peculiar rounded or elliptical bodies of a bright refractile appearance are found in the head and neck, and in the joints, especially the young joints, of all Cestoda. These bodies are very numerous in T. serrata. They are found principally in the superficial part of the connective tissue, but may occur also in the more central part where they are absorbed on the evolution of the sexual organs. They often show concentric lines like those of starch granules, and under the action of an acid they give off a gas, Carbon dioxide, which exists in com- bination with lime. Hence the name Calcareous bodies given to them. They contain a small amount of organic matrix, and are believed to be either calcified cells or portions of calcified cells. They lie, according to Griesbach, in the lacunae of the connective tissue, and he appears to think that they may enter the excretory system through direct communications between its cross anastomoses and the coelomic lacunae. It is certain that the excretory canals contain calcareous particles, and in certain Trematoda their branches have appended ampullae, in which lie calcareous bodies similar to the calcareous bodies of the coelomic lacunae in the Cestoda. The function of these structures is unknown ; it may be partly excretory, partly skeletal. For the generative organs of a Taenia, see PI. xiv. (post\ figs. 2 and 3. The structure and development of the ovum in T. serrata have been carefully studied by E. Van Beneden. It consists of a delicate shell containing a germ or ovum-cell together with a quantity of a hyaline, homogeneous and colourless albumen or deutoplasm (= secondary yolk). The germ segments into two cells, one transparent, the 'embryogenic globe,' the other a 'granular cell,' which segments no further. The former of the two divides, and the result of its division is a number of cells of which (i) three are larger and constitute the 'albumino- genous layer ; ' (2) the remainder are smaller and constitute the ' embryonic mass.' The three cells (i supra) enlarge and surround together with the ' granular cell,' the 'embryonic mass,' and secrete a delicate superficial cuticle, the cell-limits becoming indistinct. In the ' embryonic mass ' there are three, four, or some- times five flattened cells placed laterally and containing, unlike the remaining cells, nucleolated nuclei. These cells constitute a c chitinogenous layer.' They give origin to (i) a superficial homogeneous coat; (2) a coat or shell of radially placed juxtapposed chitinoid cylinders which increase in length at the expense of 228 DESCRIPTIONS OF PREPARATIONS. (3) an internal faintly striated coat in which lie the degenerating nuclei of the chitinogenous cells. The remaining cells of the ' embryonic mass ' become the hexacanth embryo or proscolex, as it is called, and are arranged in an incom- plete superficial set of more granular cells, and a contained set of clearer cells which protrude at one pole, but are probably grown over subsequently. The three pairs of hooks belong to the superficial set of cells. When the proscolex is mature the original egg-shell and the albuminogenous layer of cells disappear, and it remains invested solely by its chitinoid coat. The ova of some other Taeniae, e. g. T. mediocanellata s. saginata, appear to have a similar development. The ovum sometimes undergoes regular and equal segmentation, e. g. in T. bacil- laris, and then the albuminogenous layer is formed by a layer of numerous cells raised from the surface of the embryonic mass. For figure of proscolex, see PI. xiv. fig. 4. Van Beneden regards the albuminogenous layer of cells as the homologue of the ciliated coat of cells or ' embryophore ' of some species of Bothriocephalus, of Schistocephalus, and Ligula. Moniez, on the other hand, considers the chitino- genous set of cells as the homologue of the same structure ; but he missed the albuminogenous coat in T. serrata, &c., though he appears to have detected it in other Tapeworms. Leuckart's account agrees essentially with Van Beneden's. The latter thinks the successive coats of cells, formed as above, are to be con- sidered as layers of ectoderm cells thrown off one after the other. The last joints of a Taenia contain the uterus alone of all the genitalia, laden with the proscolices contained within their chitinoid coats. They are detached either singly or in small numbers. For their subsequent fate, see next Preparation. Some Cestoda possess a uterine aperture, and the ova are conse- quently discharged at an earlier stage, e. g. Bothriocephalus. Parasites of Man and diseases resulting from them, Leuckart, transl. by W. E. Hoyle, Edinburgh, i. 1886; the German original * Parasiten des Menschen,' i. (ed. 2), 1 88 1 ; ii. 1876. Parasites, Cobbold, London, 1879. Vers Intestinaux, P. J. Van Beneden, Paris, 1858 (or Supplement aux Comptes Rendus de 1'Acad. des Sci., ii. 1861); Vers Cestoides, Id. Mdm. de FAcad. Roy. Belg. xxv. 1850. Les Cysticerques, Moniez, Travaux Inst. Zool. Lille, iii. i, 1880; Les Cestodes, Id. pt. i, ibid. iii. 2, 1881. Lists of Parasites and Hosts, von Linstow, Compendium der Helminthologie, Hannover, 1878. Cestodes of Hares and Rabbits, Riehm, Inaug. diss. Halle, 1881 (Zeitschr. f. d. ges. Naturw. Giebel, 54, 1881). Taenia lineata, Hamann, Z. W. Z. xlii. 1885. T.perfoliata, Kahane, Z. W. Z. xxxiv. 1880. Solenophorus, von Roboz (Beitrage, &c.), Z. W. Z. xxxvii. 1882 ; Griesbach (Beitrage), A. M. A. xxii. 1883. Triaeno- phorus, Megnin, Journal de 1'Anat. et Physiol., 1881. Ligula and Schistocephalus, Kiessling, A. N. 48, 1882. Suckers, Niemiec, Recueil Zool. Suisse, ii. 1885. Connective tissue, &c., Griesbach, on Solenophorus (supra). Ditto, and muscular tissue, Hamann on T. lineata (supra}. Nervous system, Niemiec, Recueil Zool. Suisse, ii. 1885; cf. Lang, Mitth. Zool. Stat. Naples, ii. 1881. Excretory system, Fraipont, Archives de Biol. i. 1880; ii. 1881. Pintner (Untersuchungen, &c,), Arb. Zool. Inst. Wien, iii. 1881. TAPEWORM IN CYSTIC STAGE. 229 Coelome, Fraipont, op. cit. ; cf. Van Beneden and Ray Lankester, Z. A. iv. 1 88 1 ; v. 1882. Griesbach on Solenophorus (supra]. Sexual organs of T. mediocanellata (=sagmata] and T. solium, Sommer, Z.W. Z. xxiv. 1874; of Bothriocephalus, Id. and Landois, Z. W. Z. xxii. 1872. Development of proscolex from ovum in T. serrata, E. Van Beneden, Archives de Biol. ii. 1881. For genera! account^ see Moniez, Travaux Zool. Inst. Lille, iii. 2, and Leuckart (supra]. 47. TAPEWORM IN CYSTIC STAGE (Cysticercus pisiformis, s. C. Taeniae serratae), Mounted as a preparation for the microscope. THE worm has been removed from its connective tissue sac or capsule, and the head with a small portion of the neck has been evaginated by gentle pressure from the vesicle, or proscolex, within which it is generally retracted. The vesicle has delicate walls, and in life is distended by a liquid which contains only a trace of albumen and is little more than a solution of salts, chiefly of sodium. Under the microscope, using a J-inch objective, the principal features of the head may be readily made out: the slightly projecting rostellum : the circlet of hooks and the suckers. The hooks are arranged in alternation, one large with one small. The free part of a large hook is longer and somewhat straight as compared with the corresponding part of a small hook. The imbedded portion or root is in both cases forked, but the anterior (or rostellar) branch of the fork is of remarkable length in the large hooks. In the neck, especially in the part not completely evaginated, may be noticed numerous clear rounded bodies. These are the calcareous bodies, or concretions : see ante, preceding preparation, p. 227. They do not exist in the walls of the vesicle. The life-history of Taenia serrata is briefly as follows. The ripe proglottides are scattered among the grass by the way-side, &c., and are swallowed by a Rabbit, the soft tissues being digested in the stomach, and the chitinoid shells containing the hexacanth embryoes or proscolices set free (cf. p. 228) : or the proglottides decay naturally and set free the contents of their uteri, which are then eaten with her- bage. The chitinoid shell of the proscolex, under the combined influence of warmth and the gastric juice, becomes brittle, and either break up spontaneously or is broken by the movements of the embryonic hooks. The embryo, now free appears to bore its way into the tissues : and it has been found in the portal blood by Leuckart. The embryoes of T. marginata have been similarly found by Leis- ering in the portal capillaries of the Lamb. The parasite sojourns for about four 230 DESCRIPTIONS OF PREPARATIONS. weeks in the liver, in which it creeps about. It then escapes into the abdominal cavity, and becomes encapsuled either on the omentum or mesentery. The Cysticercus of some Tapeworms, e. g. of T. solium, is generally found in the connective tissue of muscles, &c. In these instances the hexacanth embryo probably migrates through the tissues, not through the blood-current. The changes undergone by the embryo are as follows. It grows in size, and in the case of C. pisiformis becomes elongate, and the hooks are lost. The central cells enlarge and become clear. The subcuticular muscles are differentiated, and at a later period a system of more deeply placed muscles. Between the two layers of muscles intervenes a layer of cells. It is from a meniscus-like thickening of this cell-layer, which is developed at the anterior end, the end where the hooks remain attached in C. Arionis, that the head and neck of the future Tapeworm are developed. The head may be formed however, as in Archigetes Sieboldi, at the op- posite end, the one at which Moniez asserts that it is always formed *. At the same time a network of excretory vessels appears. It opens to the exterior by a pos- teriorly placed pulsatile vesicle. The finer vessels end in ciliated funnels. When the organism has attained a length of 4 mm. the central clear cells break down and give rise to a central accumulation of liquid. In some instances this change takes place at a much earlier period : in others, e. g. in Piestocystis or in the Cysti- cercus Taeniae ellipticae^ or of Tetrarhynchus, in a word, in the majority of Cestoda the central cells do not thus liquefy 2. When C. pisiformis is about 2 mm. long, the meniscus of cells above-mentioned begins to develope. It grows inwards, pushing before it the deep layer of muscles which form a receptaculum capitis. When it has reached a certain size, a de- pression appears externally, and thenceforth the meniscus grows inwards as a hollow cone. The cavity of this cone widens at its inner or deep end, and here the ros- tellum, hooks, and suckers are developed ; but it is only when the cone is wholly or partly evaginated that they appear in their ordinary shape. In the position in which they are developed they are, as it were, inverted. When the head and neck are evaginated the Cysticercus appears to be divisible into three parts, (i) the head and neck proper, or scolex ; (2) the basal part of the invagination or hollow cone, and (3) the proscolex or vesicle. When such an encapsuled Cysticercus is trans- ferred to the stomach of a dog, the inclosing cyst is digested together with the parts, (2) and (3) supra : and only the head and neck pass on into the intestine. Here the head attaches itself by its hooks and suckers, and in forty-eight hours growth has proceeded so far that there are well-marked indications of 12-18 joints. 1 Leuckart states that the excretory system of Archigetes opens externally near the posterior end of the scolex or sexual worm, i. e. the end to which the pfoscolex or vesicle is attached. The excretory system of the scolex and proscolex, e. g. of a Cysticercus, opens externally at the end of the proscolex opposite to that at which the. scolex is attached. If Leuckart's statement is correct, it shows that the scolex of Archigetes develops from the proscolex in an unusual position and tends, therefore, to invalidate the assertion of Moniez referred to in the text. 2 Moniez states that the proscolex when i cm. long and less than i mm. broad divides trans- versely, the two parts being connected by a slender pedicle which ruptures. The anterior half develops the head : a rudiment of a head (?) was observed only once in the posterior half. The point of attachment of the pedicle to the anterior half is marked by a persistent depression, the foramen caudale. The fate of the posterior half is not known for certain. Non-division of a Cysti- cercus is, according to Moniez, the reason why the proscolex sometimes attains such large dimensions. TAPEWORM IN CYSTIC STAGE. 231 The proscolex in T. serrata gives origin to a single scolex, and the resulting organism is therefore termed Cysticercus. When it produces a number of scolices, the resulting organism is a Coenurus, e. g. C. cerebralis of the Sheep, the cause of the disease known as ' sturdy,' ' gid,' or ' staggers ; ' and when scolices are produced not directly by the proscolex, but indirectly from ' brood-capsules,' which originate from the proscolex in the first instance and remain attached to it, the organism is an Echinococcus. For Figure of Coenurus, see PI. xiv. fig. 5. The cyst in which the Cysticercus lies is formed by the irritated tissues of its host. Its inner surface is covered by a layer of epithelioid cells : its walls are com- posed of connective tissue and contain blood-vessels. The whole structure is produced, apparently, by the metamorphosis of lymph-cells. The life-history of T. serrata and of other Cestoda is generally supposed to in- clude three successive generations : two asexual, the proscolex and the scolex ; one sexual with numerous individuals, the proglottides. The last-named are supposed to be produced, one after another, by posterior gemmation of the scolex, from which they are detached in many instances either singly or in groups. Many interesting features of resemblance between a fully-formed proglottis and a Trematode have been pointed out by P. J. Van Beneden (cf. Vers Intestinaux, p. 251, and PI. xxvii). But the facts do not appear to necessitate the view that the proglottis is an individual : and Riehm especially has drawn attention to certain particulars. The setting free of a proglottis may be paralleled with the setting free of the hecto- cotylised or sexual arm of many male Cephalopoda : the formation of new pro- glottides with the re-development of this arm, or to the building up of a complete worm from two or three somites, as in the Oligochaete Lumbriculus variegatus. Con- sidered as an organism, a proglottis is unable to maintain its own existence : it has no organs of adhesion, and as a rule it is placed by its detachment under destructive influences. If however it remains under favouring circumstances, i.e. within the intestine, it may increase in size (Vers Intestinaux, p. 249 ; Vers Cestoides, pp. 123, 143), just as the fragments of a Nemertean may continue to live and mature sexual products. However, both Leuckart and Van Beneden are inclined to regard this fact as decisive for the zooid-nature of a proglottis. But it is not clear from Van Beneden's account to what the increase is due ; nor perhaps is it absolutely certain that dissection of the host may not cause separation of the joints from the strobila. Such separation often occurs with the slightest disturbance (cf. Vers Ces- toides, p. 139). Turning to anatomy ; — though the deep longitudinal muscles do not extend across the interval between successive joints, and are interrupted even in Ligula, yet there are longitudinal muscles which do so (Riehm); the nervous system is continuous throughout the worm : so too the excretory system, but the transverse anastomoses of the Taeniae do not exist in other forms ; the ovaries of Ligula appear to be continuous (Moniez), and discharge of the proscolices from the uterus may take place long before the joints are detached, e.g. in Bothrio- cephalidae and Ligulidae. Moreover the primitive terminal joints may remain barren or develope sexual products at a relatively late period. The degree to which seg- mentation is marked externally is variable : and in Triaenophorus and Ligula is scarcely discernible, much less so in the latter than in the former. The formation of joints is usually held to depend on the evolution of the sexual organs. It is 232 DESCRIPTIONS OF PREPARATIONS. however well marked in the young Ligula, much less so in the sexual worm : so too Triaenophorus. It exists in the barren posterior region above-mentioned, but this may be considered as arrested in growth. The setting free of the joints may well be an adaptation, and is possibly due to the completed development of the embryoes and consequent regressive metamorphosis of the genitalia. Other regressive changes appear to occur at the same period. Hamann states that the nerves degenerate in the ripe joints of T. lineata : and Megnin contends that the scolex may lose its hooks and its suckers, and may even atrophy away com- pletely. See Journal de 1'Anat et Physiol. xvii. 1881. Such changes may occur, according to him, in T. serrata and T. solium. Donnadieu found that Ligula sometimes undergoes digestion when its ripe ova are discharged. Facts such as these show that no absolute conclusion can be based on the growth of joints after their detachment in the tapeworms of some fish (supra]. As to the two supposed asexual generations, the proscolex and the scolex, the question appears to turn on the following points : the asexual reproduction of the proscolex as proscolex ; the mode in which the scolex takes origin from the pro- scolex : the character of the connection between proscolex and scolex : and the apparent necessity for two hosts. The proscolex does not generally multiply itself asexually, but gemmation takes place in the Staphylocystis of Villot, in Echinococcus (?), and in an Echino- coccoid form discovered by Metschnikoff (cf. Leuckart, ' Parasiten,' i. p. 464) ; and it is possible that an imperfect fission may sometimes occur, e. g. in Coenurus. Such increase may be compared with the fission of a Trematode Sporocyst, e.g. of Fasciola fopatica, an undoubted representative of a generation, or with its gemma- tion as in Leucochloridium : with the division of the embryo Lumbricus trapezoides, an immature individual : with the formation of germs by the tail, i.e. an organ, of the CVrazT-wz-larva, Bucephalus polymorphic, or of Cercaria cristata, according to the observations of Ercolani : or with the separation of parts of the hydrocaulus or stem which develope into individuals in a Campanularian, the Schizocladium of Allmann. In other words, the occurrence of asexual reproduction does not neces- sarily prove that the proscolex is an individual. The scolex is derived from a local thickening of a layer of cells, part of the body-wall of the proscolex : but its muscular layers and its excretory system become completely continuous with the corresponding structures of the proscolex. This mode of origin is utterly unlike the way in which germs originate from the walls of a Trematode Sporocyst or Redia, viz. by the growth and division of a cell from the layer lining the coelome, or from the mass filling it at first (cf. Thomas, Life History of the Liver Fluke, Q. J. M. xxiii. 1883, p. 125, and figs, referred to) : nor does it resemble the mode in which buds are formed in the Metazoa with the participation of all the germinal layers. With reference to this last point, however, it must not be forgotten that the germinal layers are at no period distinct in the Cestoda. It may be added that in a form discovered by Gruber (Z. A. i. 1878) in Cyclops serrulatus, there is no apparent proscolex at all, and the scolex developes without imagination. The proscolex and scolex may remain permanently attached to one another as in Archigetes Siebvldi. The connection may only be severed when the first joints are detacfied, as in many Phyllobothrians and Phyllacanthians which infest fish. TAPEWORM IN CYSTIC STAGE. 233 Or the scolex may separate from the proscolex, and either enter an intermediate host, e. g. some species of Tetrarhynchus, or even in rare instances remain free (cf. Moniez, Travaux, &c. iii. i, p. 142). Finally, the proscolex may be digested in the stomach of the second host, but so also are the barren joints formed by the scolex of Cysticercus fastiolaris (= Taenia crassicollis) while still encapsuled in the Mouse. There is evidently much variety observable with reference to the character of the connection between the two structures. It is a general fact that two hosts are necessary for the evolution of a Cestode. Archigetes however is an example of the sufficiency of a single host. Ligula attains immature sexuality in a fish, its first host, so that a few days' sojourn in its second host, a water-bird, brings about sexual maturity. Taenia solium may exist in the flesh of man in the Cysticercus-stage, in the alimentary canal as a Tapeworm ; and it has been shown by Riehm and Leuckart that the scolex of C. pisiformis may de- velope into a jointed and sexless worm in the intestines of the Rabbit, but whether it would ever attain sexual maturity under these conditions is doubtful. That the proscolex may develope in an alimentary canal is proved by P. J. Van Beneden's discovery of proscolices with scolices in all stages of growth in the intestine of the Lump-fish (Cyclopterus). The same thing possibly occurs also with other Tape- worms inhabiting fish. Why there should be two hosts is a most obscure question. Leuckart appears to think that there is a physiological connection between the two, inasmuch as one is the prey of the other. Moniez has broached the idea that a change from one to another host is, in these animals, which are so completely parasitic and therefore dependent, necessary to maintain their vigour. The facts adduced certainly weaken the generally adopted view as to the existence of an Alternation of Generations in the Cestoda. There can be little doubt that the strobila is to be regarded as a single organism, and the same state- ment is probably true of the scolex and proscolex1. It is evident that the Cestoda are profoundly modified by parasitism, not only structurally but developmentally. The study of lower forms in the class may throw more light on the many obscure points connected with its evolution, but unfortunately there are great intrinsic diffi- culties in the way of such an investigation. Development of scolex and life-histories. Leuckart, op. cit., Moniez, Les Cysticerques, and P. J. van Beneden, Vers Intestinaux and Vers Cestoides; see p. 228, ante. Archigetes Sieboldi, Leuckart, Z. W. Z. xxx. (Suppl.), 1878 ; Gruber, Z. A. iv. 1881. A Gemmating proscolex^ Staphylocystis, Villot, Tenias des Musaraignes, A. Sc. N. (6), viii. 1879; Urocystis, Id. A. N. H. (5), vii. 1881. On the question of Alternation of Generations in Cestoda ', see Riehm, Zeitschr. f. d. ges. Naturw. (Giebel), 54, 1881, p. 590; Moniez, Les Cysticerques, op. cit. pp. 135-154 ; Leuckart, Die Parasiten (ed. 2), pp. 488-490 ; P. J. van Beneden, Vers Intestinaux, p. 242 et seqq. Comparison with Trematoda. Van Beneden, ibid, p. 251 et seqq. Parasitism. P. J. van Beneden, Animal Parasites and Messmates, Internat. Series, xix. 1876. 1 Moniez regards the scolex simply as an organ of adhesion. To this view Niemiec opposes the character and degree of development attained by the central nervous system which is always lodged in it. DESCRIPTIONS OF PREPARATIONS. .48. BROAD-LEAFED HORNWRACK (Flustra foliacea), With Figure 10. A seaweed-like Polyzoon, universal in European seas and widely spread over the world. It is colonial, like the vast majority of Polyzoa, and the colony or zoarium forms erect fronds which spring from a small basal portion, spread over some foreign object in the fashion of a Membrani- pora. The fronds vary in outline in different specimens : they are flexible owing to the small amount of calcareous matter deposited in the chitinous walls of the zooecium, i.e. the resistent part of the ectocyst of the individuals making up the colony. They are composed of two lamellae, in which the zooecia are placed back to back. Some species of the genus, however, have the zooecia disposed in only one lamella. The zooecia themselves are arranged in parallel longitudinal rows, and the zooecia in one row alternate in position with the zooecia of the adjoining rows to either side. An individual zooecium is elongate in shape, with one, the distal, end curved or rounded, and the other, the proximal, end somewhat contracted. The margin of the rounded distal end bears two spines to either side the middle line, and in some cases a fifth single spine placed centrally. A whitish spot near this end, readily visible under a lens, marks the position of the retracted fore- part of the individual or tentacle-sheath which bears the circlet of tentacles. New zooecia are added to the colony at the curved edges of the fronds. Flustra foliacea generally grows upon stones or shells at moderate depths, but may occur in deeper water (sixty-two to seventy fathoms). It possesses a strong and peculiar smell when first taken from the water, and its fronds form a favourite resting place for other Polyzoa and various Hydroids. The anatomy of Membranipora (Flustra) membranacea has been well studied by Nitsche, and is illustrated in Fig. 10 A. It is abundant and universal on our coasts. The colony is composed of a single lamella in which the zooecia are disposed in parallel series, alternating as in Flustra foliacea. The zoarium is attached to the fronds of Laminaria and Fucus, and may attain the length of several feet. The zooecium or cell is in outline a parallelepiped. One surface, the lower, is attached ; the other, the upper, is free. It has two elongated sides and two short ends. Of the two ends, one, that to the left in the figure, is proximal, i. e. near to the original parent zooid ; the other distal, i. e. near to the growing free edge of the colony. A spine (Sp.) projects from the free surface at each of the two proximal angles. The mouth so-called, or aperture made by the retraction of the flexible fore-part or tentacle-sheath of the zooecium, is placed distally on the free surface. It is crescentic in outline, and its proximal edge or lip (op.) is thickened, forming the operculum, a structure from which the suborder Cheilostomata takes its name. BROA D-LEA VED HORN WRA CK. 235 The wall of the zooecium is composed of a cuticle or ectocyst and an endo- cyst. The former (e.) is chitinoid, and the chitin of the base of the two spines, of the two ends with their angles, are strengthened by calcareous matter, and a cal- careous plate also occupies the greater portion of each side. Each end is per- forated in two places, each side in four places, by a set of pores. These pores are grouped within circular areae with raised margins, and are known as ' rosette ' or ' communication ' plates (r.). The endocyst (/.) is thin and delicate. Over the rosette plates it consists of columnar cells. Elsewhere it is membranous, containing scattered nuclei surrounded by masses of protoplasm. Two cords of fusiform cells, applied to the endocyst of the lower surface, connect the rosette plates at one end with the corresponding plates at the other end. These cords anastomose inter se. op.m. -pv.l! Fig. 10. A. Longitudinal section, magnified, of Membranipora (Flustra) membranacea, after Nitsche. B. Avicularium (typical), after Hincks. C. Vibraculum of Scrupocellaria scruposa, after Hincks. and with the lateral rosette plates. Other cords, the funiculi laterales (/), com- posed of superficial fusiform cells surrounding a granular axis, run from one to the adjoining rosette plate, and are connected also to the funiculi of the stomach (/".). The fore-part of the body or tentacle-sheath1 when retracted is contained within the basal part. In Fig. 10, A. it is represented as fully expanded. Its 1 The term ' polypide ' is often used to denote ' the zooid, consisting of alimentary canal, with tentacles, nervous ganglion, &c., which is developed within the zooecium ; ' and then ' zooecium ' ( = cell, auct; cystid, Nitsche; Brutkapsel, Reichert) denotes 'the chamber in which the polypide is lodged' (Hincks). The term zooecium is used here and in the general account of Polyzoa to denote the thickened part of the cuticle which persists after the death and decay of the rest of the organism. It is simply a skeletal structure. The term polypide is discarded altogether. For the view that a Polyzoon consists of two distinct animals, i.e. zooecium, and tentacle-sheath + digestive tract, is now known to be erroneous, and the use of the word only tends to maintain the error. DESCRIPTIONS OF PREPARATIONS. proximal portion is retained in a folded condition by certain ligaments and muscles. The summit of the fold forms a circular 'diaphragm' (d.\ which contains a muscular sphincter, and closes over the tentacle-sheath (/.) when it is invaginated. The walls of the tentacle-sheath are homogeneous and contain nuclei, and are probably covered, as in other Polyzoa, by a delicate cuticle. It is crowned by a circlet of hollow, ciliated tentacles (T.\ usually eighteen in number. The mouth (m.) lies in the centre of the circlet, and the cilia on the oral surfaces of the tentacles which work towards the mouth are longer than those on their aboral surfaces. The digestive tract consists of a more or less cup-shaped pharynx or oesophagus ('•)• Their tips are either blunt or tapering. Elongation takes place by an onward flow of ectosarc followed by the extrusion into the ectosarc of a current of endosarc. The flow is, as a rule, not even but more or less sudden, as though there were a surface-resistance to be over- come. The posterior part of the body, where the protoplasm is receding, has often a lobed or mulberry-like appearance, as shown in Fig. 13, C. Similar lobes may be observed on a pseudopodium during its retraction. Any single lobe in the posterior mass might commence to elongate, and grow into a pseudopodium : and the direction of the animal's movement, indicated in C. by the arrows, might be thus reversed. In the extended condition this species may measure the ^ of an inch. An Amoeba has been observed dividing into two. The protoplasmic bridge or filament connecting the two halves, gradually becomes more and more delicate and lengthened, and then finally snaps. It is uncertain whether or no Amoeba Proteus ever forms spores in the shape of minute Amoeba or Amoebidae, such as have been observed in Pelomyxa palustris. \ Protozoa, Ray Lankester, Encyclopaedia Brit. (ed. ix.) xix. 1885. Infusoria ( = Ciliata\ Saville Kent, Manual of the Infusoria, 2 vols., and Plates, London, 1880-82. For Paramecium, see.vol. ii. p. 483 ; PL xxvi. Figs. 28-30. Amoeba. Leidy, Fresh-water Rhizopoda of North America, United States Geological Survey, xii. 1879. For A. Proteus, see p. 31 ; Pis. i; ii; iv, Fig. 25 ; vii, Figs. 13-19; viii, Figs. 17-30. Of. Gruber, Z.W.Z. xlL 1885. s a PLATE I. COMMON RAT, Mus decinnanns. DESCRIPTION OF THE PLATES1. PLATE I, COMMON RAT (Mus decumanus), a I Dissected so as to show, superiorly, the cerebrospinal nervous system -lodged in the cranio- spinal cavity, and, inferiorly, portions of most of the organs of Vegetative life. CHARACTERS distinctive of Mammalia shown in this figure are the following : the epidermic exo-skeleton in the form of hairs ; the suspension of the lungs freely in closed ' pleural ' sacs ; the perfect diaphragm (c] sepa- rating the cavities of the thorax and abdomen ; the smooth external surface of the kidney ; the single aorta crossing the left bronchus ; and the presence of an omentum or epiploon (w). The scalpriform incisors characteristic of the order Rodentia are con- cealed in this profile view by the lips, but the figure shows well the great size of the masseter muscle which is crossed by^the duct of the parotid gland (/) and by the facial nerve. The great size of the organs of special sense relatively to the entire bulk of the animal, and of the hind limbs relatively to the fore limbs, are characteristic, though not universally nor exclusively, of Rodents. Points of less classificatory importance are furnished to us by the presence of a vena cava descendens on the left side ; of smooth cerebral hemispheres (Jt) ; of a uterus all but completely bifid (y and j/); of a Har- derian gland (/) ; of a hibernating gland (k] ; and of a double lacrymal gland (^-and/). The left halves of the parietes of the craniospinal, thoracic, abdominal, and pelvic cavities have been removed to expdse the parts shown in this figure. The integument has been removed from the greater part of the facial region, but a narrow strip has been left connecting the concha of the 1 Plates I, VI, VIII, and X, are taken from the specimens described in the first part of this work. Plates II, III, XI, XII, and XIII, are from specimens of the same animals as described in the first part, but prepared differently, and therefore often displaying new points. In describing these Plates repetition is avoided as much as possible. Plates IV, V, VII, IX, and XIV, relate to animals or groups not described before. The descriptions of these Plates are therefore supplementary to the descriptions of specimens in the first part. 262, DESCRIPTION OF THE PLATES. ear with the upper eyelid. A similar strip has been drawn as left in situ overlying the costal attachment of the diaphragm. a. Left eye. b. Left ear. c. Diaphragm forming a contractile dome-shaped floor between the abdo- minal cavity below and the thoracic above. d. Eleventh dorsal or anticlinal vertebra. e. Spinal cord. The part where it widens into the medulla oblongata is concealed by the large external ear. f. Part of Harderian gland, which discharges its secretion by a duct open- ing under the rudimentary third eyelid or nictitating membrane. This gland is found in most mammals, with the exception of Chirop- tera and Simiadae. g. Intra-orbital portion of lacrymal gland. g'. Extra-orbital portion of lacrymal gland, lying upon the masseter muscle. It sends a duct with some glandular tissue inlaid in its walls to enter the orbit at its posterior angle, and receive the duct of the intra-orbital portion (g\ : h. Cerebral hemisphere of right side. i. Vagina. j. Parotid gland. Its ducts are seen to converge from its constituent lobules, which are loosely aggregated from the neighbourhood of the ear to that of the acromion, and to cross, when united, the ramifica- tions into which the motor nerve of the facial muscles is seen to break up. The buccal pouch is wanting in the true Mures. Some lymphatic glands have been removed from the space between the masseter muscle and the parotid gland. k. Portion of ' hibernating gland ; ' a gland found in many Rodentia, Chiroptera, and Insectivora, and spreading in them into the axillary, the nuchal, the thoracic, and even occasionally into the abdominal regions. /. Submaxillary gland and duct. m. Heart ; the line ends upon the left ventricle. The apex of the heart is not turned so much to the left as in man and in some of the lower mammals, e.g. the moie. The fold immediately below the point . where the line abuts upon the ventricle is formed by the cut edge of the pericardium. n. Left auricle. o. Phrenic nerve. /. Aorta. A bristle has been passed between it and the left azygos vein, and abuts on the diaphragm where the left phrenic nerve enters it. Behind this bristle are seen, passing from the aorta to the sternum, first, the third lobe of the right lung ; secondly, the oesophagus ; \ v COMMON RAT. 263 thirdly, the fourth lobe of the right lung within its own pleural cavity, in relation with which is the phrenic nerve ; and, lastly, the lobules of fatty tissue, already spoken of, in apposition with the fourth and fifth of the six sternal bones. q. Left azygos vein joining the vena cava superior of the same side, and receiving some veins from the masses of fat just mentioned in connec- tion with the pericardium. r. Root of left lung : the lung of this side has been removed ; it consisted of a single lobe, as is often, though not always, the case in Rodentia, Marsupialia, and Insectivora, though very rarely in Carnivora and 'Quadrumana ; see Cuvier, Le9ons d'Anatomie Comparee, torn. vii. ed. 2nie, 1840, pp. 156-163. s. Kidney. /. Spleen. tt. Stomach. v. Liver ; the line abutting upon its left lobe. w. Omentum or epiploon. x. Coecum. The entrance of the small intestine into the coecum is not seen, but we observe that the coecum becomes smaller in calibre where it is bent on itself superiorly. x ' . Convolutions of intestines. y. Upper end of left cornu of pregnant uterus, passing into the Fallopian tube, which together with the ovary fills up the space between this convolution of the uterus and the kidney. yf. Lower portion of same uterine cornu distended with foetuses. %. Bladder contracted into a conical shape and receiving the ureter at its base on the left side. - z . Outlet of urinary organs through a perforated clitoris distinct from the vagina. TT. Rectum. A. Flexor muscles of the tail, which arise from the internal surface of the pelvic bones. b. Anterior portion of ilium, the posterior part of which has been removed, together with the pubis and ischium. From its internal surface the caudal flexors are seen to take origin, and in front of them and in a line with the point on which the letter b is placed, the cut end of one of the great veins returning blood from the hind limb is seen. PLATE II. PIGEON, Columba livia. COMMON PIGEON. 365 PLATE II. COMMON PIGEON (Columba livia), Dissected so as to show the main points characteristic of Aves, and the arrangement of the principal muscles of flight. THE characters distinctive of Birds shown in this figure are the follow- ing : feathers ; epidermic scales to the feet ; musculature of the wing ; characters of the brain ; oesophageal crop (absent in some Birds) ; large size of duodenal loop ; pancreas and the number of ducts to this gland ; the two coeca ; extremely short large intestine as compared with the length of the small intestine ; heart resting in a deep notch of the liver ; single aorta crossing the right bronchus ; mode of division of the innominate artery ; deep indentation of the lungs by the ribs ; tubular structure of the lungs ; and trilobed kidney adapted to the pelvic fossae. There is no diaphragm as in all lower Vertebrata : the cloaca, as in Crocodilia, Chelonia and Amphibia, receives separately the rectum, urinary and generative ducts : the testes are permanently retained within the abdo- men as in a few Mammals and all lower Vertebrata : the bladder is absent as in all Lacertilia and Ophidia. In some Mammals a portion of the stomach is purely receptive, but no Mammal developes an oesophageal crop ; and it is as rare for a Mammal to possess two coeca as it is for a Bird to possess one. The only Mammals — the Prototheria — which have a cloaca, have also a sinus urogenitalis. a. Right cerebral hemisphere. Its surface is smooth, contrasting with that of the transversely laminated cerebellum seen behind in the median line. b. The crop, which is bilocular in the Columbidae. A window has been made in its right wall to show its division into two compartments. c. Right lobe of liver. d. Heart. The ventricular portion is more acutely conical in most Birds than in Mammals, and the auricles are smaller in relation to the ventricles. e. Loop of duodenum in which are contained the longitudinally arranged lobes of the pancreas. Into this loop of intestine three ducts open from the pancreas and two from the liver, which has no gall-bladder in this species. Two of the pancreatic ducts open near the middle of the distal segment of the duodenum close to each other and to one of the gall-ducts ; the third pancreatic duct opens near the distal end of the loop, and the second gall-duct near its proximal end. e\. Terminal segment of small intestine ending in the large intestine at 266 DESCRIPTION OF PLATES. f. Two long coils of small intestine have been removed between this terminal segment and the distal end of the duodenum. f. Large intestine, two coeca marking its commencement. In the small size of the coeca the Columbidae contrast with the majority of Gallinaceae. g. Terminal dilatation of the large intestine receiving the vas deferens and ureter dorsally on each side. In this cloacal arrangement Birds resemble Reptiles and Amphibia ; in all Mammals there is a sinus urogenitalis developed, into which these ducts open. In the absence of a urinary bladder Birds resemble Snakes and many Lizards. h. Testis. i. Kidney divided into three lobes, which are conformed to the sinuosities of the pelvic bones. j. Vas deferens, dilating before its termination in the cloaca. k. Ureter. /. Teres major muscle, the subscapularis and great part of the scapula having been removed. m. Right jugular vein receiving the veins from the oesophagus, and by virtue of these vessels, as also of a branch of anastomosis with the left jugular, attaining, as is usual in Birds, a larger size than that vessel. n. Right jugular vein in thorax. o. Vena cava inferior, entering the auricle to the right of and posteriorly to the entrance of the vena cava superior of the right side. /. Lung, showing on its exterior surface indentations corresponding with the ribs. q. Right bronchus entering the lung. Between the bronchus and the vena cava inferior we see a portion of the glandular proventriculus, and immediately above the bronchus and below the arch of the aorta, which has been displaced a little upwards, the junction with the jugular of the fragment of vein left to represent the subclavian trunk. r> Right innominate artery, which is seen to break up into three main divisions, the- common carotid, the axillary and the pectoral arteries. y. Portion of inner tuberosity of humerus which overhangs the pneumatic foramen of the bone. z. Gizzard. s. Great pectoral muscle, the main depressor of the humerus and wing. Its main tendon is seen turned back at x ; two other tendons which it gives, one to the long extensor, the other to the short extensor of the alar membrane, are not shown in this figure. t. Second pectoral, the main elevator of the humerus. u. Coracobrachialis inferior, a muscle arising from the inferior and outer three- fifths of the distal part of the coracoid, and inserted into the internal and COMMON PIGEON. 367 proximal lip of the cup-shaped pneumatic cavity of the humerus. The opposite lip of this cavity receives the tendon of the teres major /; and from the triangular space between the muscular bellies of these two muscles, the subscapularis muscle, together with the upper portion of the scapula, and a small muscle, the serratus anticus, which passes between the fibres of the subscapularis to be inserted into the inferior edge of the scapula, have been removed. v. Coracobrachialis superior, a bicipital muscle with a very extensive origin; arising, superiorly, from the inner surface of the vertebral end of the clavicle ; inferiorly, from a facet on the lateral aspect of the upper surface of the sternal rostrum ; and between these two points of origin from the upper and inner surface of the ligament connecting the coracoid, clavicles, and sternal rostrum. Its tendon, which is joined by that of the subscapularis, is inserted proximally and anteriorly to that of the preceding muscle u. w. One head of the extensor plicae alaris anterioris longus, arising from' the upper end of the clavicle in continuity externally with a head of the extensor brevis. These muscular bellies appear to be divarications of the deltoid. wi. Muscle in connection with the long alar extensor tendons. Its fibres have in the natural condition of the parts much the same direction as those of the muscle w and of the deltoid ; but its origin is mainly from the fascia which covers the biceps in front, and being interposed between that muscle and the tendon of the great pectoral, it is continued up into the tendinous expansion by which the posterior layer of the tendon of the great pectoral connects itself more or less intimately with the coracoid head of the biceps and obtains an insertion into that bone. The muscle wi is inserted mainly into the inner of the two tendons at its distal extremity. This tendon is prolonged down to be inserted into the radial process of the carpo-metacarpal bone which carries the pollex. It is more or less intimately connected with the two other long extensor tendons from the muscle w and from the great pectoral, which are here drawn as one ; as also with the extensor brevis which is not shown in this figure. x. Tendon of great pectoral muscle turned back. The posterior portion of this tendon receives at its lower edge the tendon of a cutaneous muscle which is figured as attached to its outer angle, and higher up it receives the main tendon of origin of muscle wi, and is ultimately prolonged either separately or in connection with the tendon of the biceps up to the coracoid. xi. Biceps. Its tendon is seen running upwards to be inserted into the internal anterior process of the upper end of the coracoid ; it has a small insertion into the humerus also, which is not shown here. PLATE III. COMMON FROG, Rana temporaria. COMMON FROG. 369 PLATE III. COMMON FROG (Rana temporaries), Injected and dissected to show the chief features of the circulatory organs, and especially the connected systems of the renal-portal and epigastric veins, together with certain portions of the muscular system, the renal and reproductive organs. THE integument has been turned back on the right side, together with the musculo-cutaneous vein, the superficial branches of which extend from the knee to the shoulder ; part of the muscular wall of the body has been removed on that side, but part has been left in situ ; and the main trunk of the musculo-cutaneous vein is seen crossing a slip which the obliqims externus muscle receives from the scapula ; on the left side the muscular and cutaneous elements of the wall have been turned back whilst remaining in their natural connection with each other and with the epigastric vein ; the shoulder girdle has been cut through the middle line, and fastened out on either side so as to expose the lungs, heart, and great vessels ; the liver has been removed with the exception of a small part of its substance, as have also the stomach and intestines down to the lower end of the rectum. a. Intermandibular space. The skin is left in situ anteriorly in the sym- physial angle ; immediately posteriorly to its cut edge is seen part of the mylohyoid or submaxillaris muscle ; and posteriorly again, and at a deeper level, the converging hyoglossi in the middle line, and on either side of them the geniohyoids. b. Tetradactyle hand. The first finger, i. e. second digit, or * thumb ' so- called, has its basal joint more or less tumid in this, a male specimen. c. Muscles of thigh. The line points to the sartorius, which is bordered externally by the vastus internus, and internally by the adductores and recti interni. See Ecker, Die Anatomic des Frosches, p. 1.15. d. Point where the musculo-cutaneous veins, constituted by factors from the regions of the head and face, but mainly from those of the back and flanks, turn inwards to pass over a slip going from the scapula to the external oblique muscle and join the subclavian vein. As the skin is moist and glandular this vein has probably to a certain extent a respiratory function. See Ecker, /. c., p. 88. e. Vein, called ' epigastric ' by Rathke, * umbilical ' by Bojanus and Jour- dain, ' vena portae accessorial and 'vena abdominalis inferior s. an- terior I by other authors. This vein is mainly constituted by the convergence of the two transverse branches from the femoral veins seen at /in the figure, but it receives twigs also from the abdominal parietes, and a factor of especial significance in the shape of the vesico-hemorrhoidal vein from the allantoid bladder and rectum. DESCRIPTION OF PLATES. The occasional pathological distension in liver diseases of the veins of the anterior abdominal parietes in the human subject shows that an arrangement may exist in a rudimentary condition in the higher Vertebrata similar to that shown here to exist functionally between the epigastric and the parietal veins ; and the con- nection with a vesico-hemorrhoidal vein, whilst it may be held to foreshadow the arrangement of the umbilical vein in the foetus of Mammals, puts prominently forward the fact that anastomoses exist between the portal and systemic veins. For the 'renal portal' of the Frog, see Jourdain, A. Sc. N. (4), xii., 1859, p. 180. /. Point where the transverse branch of the femoral vein of either side fuses with its fellow to form the trunk of the epigastric. g. ' Renal portal/ or renal inferent vein of the right side, being the other branch of the bifurcating femoral vein, which is thus seen to be con- tinuous with the portal systems of both liver and kidney. Conse- quently there are two channels open for the return of the blood to the heart. h. Bifid allantoid bladder distended, with ramifications of the vesico- hemorrhoidal veins. i. The rectum, which is cut short. j. A vesicular, and in this species glandular, dilatation developed upon the Wolffian duct, by which both testicular and renal products pass down to the cloaca. A vein passes directly into the kidney from it. k. Vena cava inferior, constituted mainl/ by the efferent kidney veins, but receiving also those of the testes and fat bodies. /. Testis of left side. It has, together with its fellow and with the kidneys, been displaced a little to the right side. m. Fat bodies. n. Spleen. To the left and a little above the spleen are seen the cut ends of two vessels, one of which receives a factor from that organ, coming itself from the intestine, and the other of which takes origin from the stomach. Both veins join a branch of the epigastric, and are distri- buted to the liver, a small portion of which is seen left immediately above them. o. Gall bladder left attached to the epigastric vein by a vein which passes from it to that vessel. p. Lung of left side. The cavity seen on the outer side of either lung has its outer wall constituted by the internal abdominal muscle (homo- logous with the internal oblique and transversalis), which arches in- wards in a dome shape, and is connected with the oesophagus and pericardium, the coracoid and hyposternum. In the natural condition of the parts these cavities are however mainly occupied by the lobes of the liver, which nearly entirely cover the lungs in an anterior view. q. Heart The conus arteriosus takes origin from the base of the ventricle, a constriction known as the,/ -return Halleri marking the line of sepa- COMMON FROG. 271 ration of the two organs. The conus bifurcates into the two great divisions of the truncus aortae, which are each subdivided by internal partitions into three canals. These canals become three tubes, the carotico-lingual, the aortic, and the pulmonary trunks, of which the first is most internal and anterior, and the last the most external and posterior. q'. Conus arteriosus, with the auricles one on each side. It inclines to the left, and is attached on that side to the ventricle by the frenulum bulbi of Briicke. Denkschrift Akad. Wienj Bd. iii. 1853, p. 355. r. Lingual branch of the first of three trunks arising just internally to a rete mirabile known as the ' carotid gland,' from the outer side of which the carotid artery, called sometimes the ' ascending pharyngeal,' passes to the back of the oesophagus in close apposition with the second main trunk or aorta. s. Convergence of hyoglossi muscles. /. Geniohyoid muscle. ii. Left external jugular vein passing down to unite with a vein formed by the union of the cutaneous vein with the vena anonyma, and thereby constitute the vena cava superior of the same side. PLATE IV. SKATE, Raja Batis. SKATE. 373 PLATE IV. SKATE (Raja Batis). THIS plate, illustrating the anatomy of the Skate, is introduced for the purpose of supplementing the description of the Perch, a member of the order Teleostei, p. 83. The Skate belongs to the more generalised order Elasmobranchii. In addition to points characteristic of this order detailed below, note in the examination of a specimen the following : — the exo- skeletal spines on the dorsal surface resembling teeth, not only in form but in structure, and attached to a basal plate of bone : the great extent of naked skin : the spiracle or visceral cleft between Meckel's arch (mandi- bular arch) and the Hyoidean arch— the partial homologue of the Eusta- chian tube — which opens from the mouth behind the eye : the minute apertures of the aquaeductus vestibuli, or pedicles of invagination of the inner ear, placed at the posterior and dorsal aspect of the cranium, one on either side : the skeleton, cartilaginous with the exception of the bodies of the vertebrae : the communication between the pericardial and abdominal cavities in the shape of a bifurcated canal : the presence of a posterior division of the kidney or metanaphros, with a certain number of ureters passing off from the inner side of the organ to open into the dilatation, at the posterior end of the Wolffian or mesonephric duct : the union of a certain number of these ureters into a single duct, especially in the male : in the female the open conjoined mouths, situated ventrally at the root of the liver, of the two oviducts, M tiller's ducts or Fallopian tubes ; the ' nida- mental ' gland situated on each oviduct ; the thicker posterior uterine portions of the ducts : the two oviducal openings, one on each side, into the cloaca : in the male the connection of each testis to the anterior part of the corresponding Wolffian body or mesonephros, thus forming an epididymis ; the convoluted anterior portions of the Wolffian or mesonephric ducts form- ing vasa deferentia, and the claspers placed to the inner side of the two lobes of the ventral fins. The Rays are peculiar in the slight development of the azygos system of fins, which is restricted to small lobes on the dorsal side of the extremity of the tail : in the enormous expansion forwards, outwards, and backwards, of the pectoral fins which gives the body its great width ; and in the bilobed form of the ventral fins. The skin of one Ray at least possesses the minute close-set denticles which constitute shagreen. The degree of development and the arrangement of the large cutaneous spines varies much in different species. The praenasal cartilage is always large, often extremely prolonged, and forms the pointed anterior extremity of the body. T 274 DESCRIPTION OF PLATES. FIG. i. Skate, Raja Batis, female, ventral view from a specimen, dissected so as to show the heart, gills, and digestive tract in situ. a. The line points to the spot where the conus arteriosus springs from the ventricle. This structure lies in front, with the auricle and ductus Cuvieri behind (see Fig. 2, infra}. b. The line points to the base of the ventral aorta at the spot where it springs from the conus, and gives off the two posterior innominate arteries. Each of these vessels divides into three branches — the three posterior branchial arteries which run on the outer side of the three posterior (II-IV) branchial arches and supply with venous blood the two gill-laminae, anterior and posterior, borne by each arch. These laminae, together with the fibrous septum which supports them and is continued to the skin separating the so-called gill-pouches inter se, are seen on the right side of the diagram. c. The line points to the anterior termination of the ventral aorta where it gives off the right and left anterior innominate arteries. Each of these arteries divides into the two anterior branchial arches which sup- ply— the anterior, the single gill-lamina (=opercular gill of Garioidei, pseudobranch of Teleostei} borne upon the posterior aspect of the hyoidean arch ; the posterior, the two gill-laminae borne by the first branchial arch (I). The fifth branchial arch in the Rays and most Sharks, as in Teleostei and Ganoidei, bears no gill-laminae. The mode in which the branchial arteries arise from the aorta is characteristic of Rays. d. The first of the five external gill-slits. The remaining four are seen arranged in a curved line behind. Gill-slits uncovered by an oper- cular fold are characteristic of all Sharks and Rays : their completely ventral position, of the latter only. e. The aperture into the olfactory pit. This aperture is placed ventrally in nearly all Elasmobranchii. A groove leads from the pit to the corresponding angle of the mouth. Such a groove exists in the em- bryoes of all higher Vertebrata. The outer edge of the groove re- presents the fronto-nasal, the inner edge the praemaxillary, processes seen in the embryoes of Vertebrata which possess praemaxillary and maxillary bones. If the roof of the olfactory pouch is examined in a specimen it will be seen to possess two series of transverse folds. /. The line rests upon the upper jaw, which is cartilaginous, and repre- sents, as in all Elasmobranchii, a palato-ptery go-quad rate cartilage. The transverse slit of the mouth and the under jaw (= distal end of Meckel's cartilage) are seen with the rows of diamond-shaped teeth set edge to edge. The retention of a ventrally placed mouth is characteristic of the Elasmobranchii. g. Jelly tubes or sensory ampullae. Only a certain number of these SKATE. 275 structures which are peculiar to Elasmobranchii have been figured. They run towards the head, where their inner ends are situated in close contact. Their length varies, the most anterior being short, the posterior long. Their apertures are easily found. The tubes are dilated at their inner extremities and possess in this species numerous lateral saccules. The nerves pass along the septa between these sac- cules, radiating from the centre. They form a plexus .in the walls of the saccules, the ultimate fibrils of which are probably continuous with hair cells in the lining epithelium. The jelly filling these tubes is firm, and is secreted by goblet cells scattered on the walls. The lateral line is not indicated in this figure. See Merkel, Endigun- gen der Sensibeln Nerven, &c.; Rostock, 1880, p. 33, PI. II. Fig. 10 ; and Leydig, Beitrage zur Mikr. Anat. &c., der Rochen und Haien, Leipzig, 1852, p. 37. The ampullae are also described by the same authors, see Merkel, p. 43 ; Leydig, p. 41. h. Middle lobe of the liver, which is trilobed in this species, cut short. The right and left lobes are seen on either side of the aperture made in the abdominal walls. i. Gall-bladder, in the fissure between the right and central lobes. The letters h and i are placed over the coracoidal bar. k. Stomach, which has the form of a siphon. Blood-vessels are seen arising from the concave edge of the organ and uniting to form a vessel which passes towards the liver and is one of the factors of the portal system. /. The spleen. m. The intestine with its anterior wall removed to show the spiral valve formed by its mucous membrane. Such a valve is found in all Elas- mobranchii, in Ganoidei (rudimentary in Lepidosteus) and Dipnoi. The inclination, &c., of the folds varies much in the different genera. n. Rectum. The line points to the spot where the rectal gland, found in all Elasmobranchii, opens into it on its dorsal aspect. o. Dilated duodenum or Bursa Entiana. It receives the bile and pancreatic ducts. The pyloric aperture is placed on a nipple-like projection. /. The line points to the fold which lies dorsal to the anal aperture into the cloaca, and separates it from a recess into which open the two oviducts laterally and the urethral canal medianly. In the male there is a urogenital papilla on the dorsal wall of the cloaca. q. Right porus abdominalis or external aperture of one of the two canals by which the abdominal cavity communicates with the exterior in Elasmobranchii, Dipnoi, some Ganoidei, and a few Teleostei. r. The two-lobed ventral fins. In a male the claspers would be situated between the inner lobes of these fins and the root of the tail. s. Tail cut short. T 3 276 DESCRIPTION OF THE PLATES. FIG. 2. Heart, with ductus Cuvieri and base of the ventral aorta of Raja Batis. The walls of the several sections of the heart have been removed in part to show points of internal structure. Ventral view. From a specimen. a. Ventricle, single as in all Pisces, Amphibia and Reptilia, with the exception of the Crocodile. The walls are thick, and the inner surface shows numerous bands of muscular tissue (columnae carneae). The cavity is curved and the entrance into it from the auricle is guarded by two large membranous valves, seen in the right upper corner of the diagram, — the left corner in the natural position. b. Conus or bulbus arteriosus. The aperture of the ventricle leading into it lies on the opposite side to the auricular aperture. Its walls are thick, composed of striated muscular tissue, and it is rhythmically contractile. Its cavity contains three longitudinal rows of pocket- shaped membranous valves, four in each row. The distal valve in each row is the largest. The number of valves in each row appear to vary in the different species of Rays. The conus must be regarded as a part of the ventricle. It is present in Ganoidei and Dipnoi among Pisces, and in Amphibia. The number of rows of valves and of valves in each row that it contains is very variable. The distal valves persist and form the valves that guard the entrance to the aorta in those Vertebrata in which the conus is not present as a separate division of the heart. c. The line points to the left posterior innominate artery. It and its homologue on the other side of the body spring from the base of the median aorta. The aorta is composed chiefly of membranous tissue, and the muscular tissue present in its walls is un'striped. (Cf. Fig. i, b and c). d. d. The right and left pouches of the large thin-walled auricle. This structure is single as in all Pisces except Dipnoi. Its walls are thin and its muscles form a network of trabeculae. e. e. The right and left ductus Cuvieri. These two vessels bring back to the heart the venous blood of the whole body. They fuse in the middle line posteriorly to the auricle forming the sinus venosus. The aperture of this sinus into the auricle is guarded by two membranous valves. The cavity of the whole heart makes an S -shaped curve, much more distinct in some Sharks than in the Ray. This curve is an embryonic feature in other Vertebrata. FIG. 3. Brain, with the roots of the chief nerves of Raja Batis. Dorsal view. From a specimen. a. The Lobi olfactorii connected by long olfactory tracts to the cerebral lobes from which they are outgrowths. Each lobe is lengthened out SKA TE. 277 laterally, corresponding with the elongated nasal pit. In some Rays they are secondarily lobed. The olfactory tracts vary much in length in the Elasmobranchii. They are said to be wanting in Raja miraletus. b. The cerebral lobes or main part of the fore-brain. They are much compressed dorso-ventrally, triangular in shape, with the outer angles swollen where the olfactory tracts take origin. They are slightly notched in front and grooved ventrally, indicating their bilaterally symmetrical structure. Lateral ventricles are wanting, but the degree to which they are obliterated among Elasmobranchii is very variable. A triangular spot, shaded in the diagram, behind the cerebral lobes marks the position of the third ventricle. The filamentous pineal gland (Epiphysis cerebri) has been removed. c. The mid-brain, corpora bigemina, or optic lobes. These bodies are large and hollow, and lie above the aquaeductus Sylvii or passage between the third and fourth ventricles, with which their cavities communicate. d. The cerebellum, or roof of the anterior portion of the fourth ventricle, the ventricle of the hind-brain. It is large in size in all Elasmobranch fishes, and often complexly convoluted. It consists here of two lobes in front and a long triangular lobe behind. It contains a large cavity, freely open to the fourth ventricle, which is just visible behind it, being for the most part covered by the posterior triangular lobe. e. The convoluted corpora restiformia with which are connected the roots of the fifth, facial, and auditory nerves. f. The commencement of the spinal cord. The medulla oblongata or the sides and floor of the hind-brain is remarkably short in the Rays. It is usually elongated in Elasmobranchii. g. Optic nerves. There is a superficial chiasma to these nerves in Elas- mobranchii, Ganoidei^ and Dipnoi. h. Third nerve or oculomotor. It springs from the base of the mid-brain, but its superficial origin is hidden by the saccus vasculosus. i. Fourth nerve or trochlearis. It arises, as in all Vertebrata, from the roof of the aquaeductus Sylvii behind the mid-brain, and in front of the cerebellum. j. Roots of the fifth (trigeminus), seventh (facial or ' portio dura '), and eighth (auditory or 'portio mollis') nerves. The facial rises dorsally, the auditory posteriorly and ventrally, while the remaining roots belong to the trigeminus. This close connection of the nerves at their roots is generally found in Pisces. The trigeminus and facial are also closely connected in the Gasserian ganglion in Anura (Am- phibia). The facial and auditory nerves also are often closely con- nected in Amphibia and Reptilia at their roots. 7g DESCRIPTION OF THE PLATES. k. The roots of the vagus nerve or pneumogastric. The glossopharyngeal nerve forms the most anterior of these strands which rise close together from the medulla oblongata. FIG. 4. Longitudinal section through an Elasmobranch embryo at a time when the neural canal, notochord, and alimentary canal are established. From Balfour, Development of Elasmo- branch Fishes. London, 1878, Fig. i c, p. 58. ep. Epiblast or superficial layer of cells formed from the blastoderm, and extending, with the mesoblast and hypoblast, round the ovum, to form the yolk sac. This growth or extension is greatest at the anterior end of the embryo. It is slight at the posterior end, where was situated the aperture of invagination, blastopore, or anus of Rusconi, beneath the embryonic rim. nc. Neural canal which is formed first as a groove in the blastoderm (medullary groove), and is then converted into a canal by the closure of the sides of this groove. The medullary groove extends to the posterior end of the blastoderm and the dorsal edge of the blastopore. On its closure to form the neural canal a pore persists at this spot ; and when the dorsal edge of this pore, continuous with the edge of the blastopore, grows down over the yolk, a communication is left between the neural and alimentary canals. x. Communication between the two canals named, known as the neuren- teric canal. It is a common feature in vertebrate development, and lies, as a rule, behind the spot where the anus is formed. Hence there is a post-anal extension, not only of the neural, but also of the alimentary canal. See Balfour, Comp. Embryology, vol. ii. p. 367, 268 ; p. 634-636. Cf. Spencer, Q. J. M. Suppl. 1885, p. 136-7. ch. Notochord. A rod of cells differentiated at an early period from the layer of cells below the medullary groove, and therefore probably to be considered as of hypoblastic origin. m. Mesoblast. al. Alimentary canal, roofed in by cells which are derived together with the mesoblast from the original mass of cells underlying the epiblast. The mesoblast cells are wanting immediately below the medullary groove, but form two plates to either side of it. Both mesoblast and hypoblast extend with the epiblast round the non-segmented part of the ovum or yolk to form the yolk sac. n. Nuclei in the yolk. These nuclei appear spontaneously (?) in the pro- toplasm of the yolk, i. e. in the network between the secondary yolk or deutoplasm. A portion of protoplasm separates round each nucleus. ' This process begins at an early stage, and the cells thus formed are added at first to the lower layer of blastoderm cells, and subsequently SKATE. 279 to the hypoblast. They thus contribute to its growth, and especially to the formation of the floor of the alimentary canal. The chief works on the Anatomy of the Skate and Elasmobranchii are the following : — Anatomy of Skate, T. J. Parker, Zootomy, London, 1884. Zoological account of Rays. Cf. Day, Fishes of Great Britain and Ireland, ii. 1884, p. 336. Couch, Fishes of British Islands (coloured figures), i. 1862, p. 78. Bell, British Fishes, ii. 1841, p. 550. Giinther, British Museum Catalogue of Fishes, viii. 1870, p. 434. Skull. W. K. Parker, Tr. Z. S. x. 1879. Visceral arches. Dohrn, Mitth. Zool. Stat. Naples, vi. 1885. Paired fins. Balfour, P. Z. S. 1881. Paired and azygosfins. Mivart, Tr. Z. S. x. 1879 (see general remarks). Azygos fins. Mayer, Mitth. Zool. Stat. Naples, vi. 1885. Exoskeleton. Hertwig, J. Z. viii. 1874. Benda, A. M. A. xx. 1882. Teeth. Tomes, Ph. Tr. 1876. Id. Dental Anatomy, ed. 2, 1882, p. 215. Central nervous system. Rob on, Dk. Akad. Wien, xxxviii. 1878. Viault, A. Z. Expt. v. 1876. Systeme ganglionaire. Vignal, A. Z. Expt. (2), i. 1883, p. xvii. Spinal cord, Stieda, Z. W. Z. xxiii. 1873 ; Sanders, P. R. S. xl. 1886. Cranial nerves. Marshall, Q. J. M. xxi. 1881. Cf. Beard, Branchial sense- organs, Q. J. M. xxv. 1885. Vagus nerve. Rohon, Arb. Zool. Inst. Wien, i. 1878. Gustatory organs. Todaro, A. Z. Expt. ii. 1873. Organs of lateral line. Solger, A. M. A. xvii. 1880. Do. and ampullae. Merkel, Endigungen der sensibeln Nerven in der Haut, Rostock, 1880. Intestinal spiral valve. T. J. Parker, Tr. Z. S. xi. 1880. Rectal gland. Blanchard, Journal de 1'Anat. et Physiol. xiv. 1878. ^Function of do. Id. Bull. Soc. Zool. de France, vii. Rudimentary gill slits. Van Bemmelen, Mitth. Zool. Stat. Naples, vi. Histo- logy of gills. Droscher, A. N. 48, 1882. Rudiment of air bladder. Miklucho- Maclay, J. Z. iii. 1867. Valves in conus arteriosus of Elasmobranchii and Ganoidei. Stohr, M. J. ii. 1876 ; for Raja Batis, see p. 219. Venous system of Skate. T. J. Parker, Trans. New Zealand Institute, xiii. 1880. Urinary apparatus and genital ducts. Balfour, Development of Elasmobranch Fishes, 1878, pp. 249-286. Ovary. Balfour, Q. J. M. xviii. 1878. Cf. J. Muller, Berlin, Abhandl. 1840, for forms of ovum, and Giinther, Study of Fishes, 1880. Copulatory organs. Petri, Z. W. Z. xxx. 1878 ; see Bolau, Z. W. Z. xxxv. 1881; and Schneider, Zool. Beitrage, i. i. 1883. Pori abdominales. Bridge, Journal of Anat. and Physiol. 1879. Cf. Ayers and Gegenbaur, M. J. x. 1884. Chemistry of horny fibres in Mustelus and egg-case of Scyllium. Krukenberg, Mitth. Zool. Stat. Naples, vi. 1885. PLATE V. CELLAR SLUG, Limax flaws* CELLAR SLUG. ' 281 PLATE V. > CELLAR SLUG (Limax flavus, s. variegatus], Dissected so as to show its digestive, circulatory, respiratory, nervous, and reproductive systems. THE muscular envelope has been separated from the foot along the left side, and turned over to the right, together with the shield- shaped mantle and the organs it overlies. The buccal mass and nerve collar, together with the salivary glands, have been displaced a little to the left, on which side of the animal's body the stomach and bilobed liver have been fastened out. Some of the nerves, muscles, and arteries have been cut away. The oesophagus and buccal mass have been pulled a little forward through the nerve-collar, and occupy much the same posi- tion relative to it that they do when in life the buccal mass and head are thrust forward. The two first convolutions of the intestine have been un- coiled, and it has thus been drawn as taking a much less sinuous course than it does in nature from its commencement at the pylorus 4x> the point where it comes into relation with the dorsal integument and shield, and hooks round the muscle which retracts the buccal mass and tentacles. The generative organs have been detached from their normal corfnections, and are arranged on the right side ot the animal's head. Their volume, as drawn here, is small in comparison with that which they attain in the breed- ing season. The upper tentacles, the nerves which supply and the muscles which retract them, have been cut through, and turned forward so as to lie between the generative apparatus on the right hand and one of the salivary glands on the left. The right lower tentacle is seen between the right upper tentacle and the vestibulum of the reproductive system. a. Locomotive disk or ' foot ' passing upwards at the sides into the general muscular envelope of the various organs of the animal's body, from which it is limited off by a furrow. Its internal circular coat is raised into two corrugated ridges along the greater part of the middle line of the body by the underlying supra-pedal gland. This gland is found in many Gastropoda^ and is of very large size in this Slug. Its aperture lies above the foot and below the head, its duct is long, and lined by ciliated epithelium of two kinds, of which one is perhaps sensory (? olfactory). The gland cells are aggregated on either side of and below the duct. The supra-pedal gland must be carefully distinguished from the pedal which secretes the mucous thread by which certain Prosobranchiata suspend themselves to the surface of the water, and which opens on the sole of the foot an- teriorly. See p. no, and lit. p. 112. DESCRIPTION OF THE PLATES. b. Shield and organs in connection with it. c. Stomach and bilobed liver, arranged upon the animal's left. d. Generative apparatus arranged upon the animal's right. e. Nerve-collar, consisting of two cerebral ganglia placed above, or rather at the sides of, the oesophagus, and two pairs of ganglia placed below it, and united with the upper pair by connectives. The two cerebral ganglia are connected by a flat commissure, and with the infra-oesophageal ganglia by a double connective, the posterior cord of which joins the posterior part of the mass or visceral ganglia, whilst the anterior cord joins the anterior or pedal ganglia from which nerves pass off to the foot. f. Stomatogastric ganglion of right side placed below the oesophagus, where it enters the buccal mass together with the duct of the salivary gland. The ganglion is connected by a long and delicate com- missural cord with the supra-oesophageal ganglion of its own side, and it gives off nerves to the buccal mass, to the oesophagus, and to the duct of the salivary gland. g. Salivary gland. h. Buccal mass containing the ' tongue ' or ' odontophore/ i. Semper's organ. It is of very large size in this Slug, small in Helix, Arion, Lymnaeus, and has only been detected in Pulmonata. It consists of four to five lobes composed of cells, resembling those in the salivary or supra-pedal glands, held together by a network of connective tissue and a membrane. According to Sochaczewer (Z. W. Z. xxxv. 1 88 1, p. 35), it receives two nerve twigs from the labial nerves, and is not, as Semper supposed, richly supplied with nerves. A ganglion has been detected by Sarasin (Arb. zool. zoot. Inst. Wurzburg, vi. 1883, p. 95) lying in the oral lobes and sending processes into this gland. It occurs also in Pulmonata Basommato- phora, in which Semper's organ is absent. j. Coecal projection at pyloric end of stomach. k. Liver consisting of two lobes opening each by a single duct close to the pylorus. /. Intestine passing from the pylorus to end a little in front of and above the respiratory inlet. Its two first convolutions have been separated from the liver and reproductive apparatus. As it approaches the dorsal integument and shield it describes a curve like an Italic 5. In the first loop of this 5 is seen the origin of the retractor muscles of the buccal mass and labial tentacles ; at its opposite extremity arises a straight coecum,^. m. Respiratory orifice, with the ' rectum ' curving round to open a little above and anteriorly to it. To the right of the rectum is seen the duct of the renal organ. CELLAR SLUG. 383 n. Portion of dorsal integument, an incision immediately to the right of which would disclose the shell. Internally to it is the respiratory sac, with the ramifications of the pulmonary veins. o. Renal organ, placed to the right of the heart in the natural position , of the parts, and giving off a duct which passes backwards, to run in company with the rectum and open near the anus. See enlarged figure by Professor Leidy in Binney's Terrestrial Molluscs of the United States, vol. i. PI. I. Fig. iv. p. Ventricle of bilocular heart. q. Hermaphrodite gland. r. Hermaphrodite duct. s. Albuminiparous gland. t. Vas deferens becoming distinct from oviduct v sooner than in Helix or Arion, and richly beset with prostatic glandules. «. Penis, with part of its retractor muscle left attached to it ; the muscle originates at a spot on the under surface of the muscular envelope of the viscera, close to the arterial outlet of the heart. v. Oviduct, like the vas deferens, glandular above, and membranous below ; and opening ^into a dilated vagina. w. Receptaculum seminis, opening in this species, though not in the closely allied Limax cinereus into the vagina. x. Pedal portion of the suboesophageal nerve mass, enclosing, together with the visceral, an orifice through which the anterior aorta passes. The line is drawn to a spot where in Helicidae the otic vesicle is readily found, but where in Limax it is not easy to convince oneself that it exists, even as a rudimentary organ, without the use of reagents, such as the oxalic acid recommended by de Lacaze Duthiers. y. Coecum passing off from intestine just before it comes into relation with the pulmonary cavity, and extending back nearly to the termination of the body cavity. z. Retractor muscle of the buccal mass and tentacles. Its fascicles pass with the oesophagus through the nerve-collar. They have been cut away in this Preparation. Anatomical and general account of Limacidae, Simroth, Nacktschnecken, &c. Z. W. Z. xlii. Figures of the anatomy of Limax, Leidy, Binney's Terrestrial Molluscs of the United States, i. PI. I. Reproductive system, Baudelot, A. Sc. N. (4), torn, xix., 1863, PI. III. Fig. 17. I I FRESH-WATER MUSSEL. 285 PLATE VI. FRESH-WATER MUSSEL (Anodonta cygnea), Dissected so as to show its muscular and nervous systems, as well as certain other organs in relation with them. THE animal has been taken out of the shell ; the gills and the mantle have been removed on the left side, together with the labial tentacles and parts of the pericardium, as well as of the organ of Bojanus or nephridium of the same side. a. Right mantle lobe, free along its ventral edge. af. Fimbriated portion of mantle corresponding to the inferior siphonal notch by which water is drawn into the branchial cavity. a'. Dorsal raphe along which the two halves of the mantle meet. b. Foot. The muscular portion is strongly contracted. c. Gills of right side. c Union of external gill to the mantle between the inferior and superior siphonal notches. d. Anterior adductor. e. Posterior adductor. f. Posterior retractor of the foot, inserted into either valve, anteriorly and superiorly to the posterior adductor, the scar or muscular impression of the two being more or less .confluent. Its muscular expansion in the foot is especially well developed along the free or ventral edge of the foot, and it inter-digitates very freely with the protractor pedis, though it lies for the most part at a lower level than that muscle. g. Protractor of the foot. This fan-shaped muscle spreads over the external surface of the foot, from an insertion into the shell, a little superiorly to the point where the pallial line joins the impression of the anterior adductor. It acts, consequently, as an antagonist to the preceding and succeeding muscles. Its impression is distinct in this animal from that of the adductor. h. Anterior retractor of the foot. The fibres of this muscle take origin from a point in the shell, towards the dorsal aspect of the anterior adductor, though some way from its dorsal border. They spread thence into the foot especially along its anterior edge, and down as far as its anterior angle, for the most part at a deeper level than the preceding muscle. Some of the fibres, however, spread over the visceral mass dorsally. h! The line points to the position of the smaller retractor muscles with 286 DESCRIPTION OF THE PLATES. insertions just anteriorly to the umbones, whence they radiate over the regions of the stomach, and towards the pericardium. i. Left cerebro-pleural ganglion lying in the angle between the anterior retractor and adductor, and the protractor pedis, above the entrance to the mouth. j. Connective passing from the left cerebro-pleural ganglion to the left pedal. The pedal ganglion of each side gives off twelve nerves, six from its neural, and six, more slender, from its lateral surface. They are not figured in this plate. f. Auditory vesicle appended to pedal ganglion. This vesicle is ordi- narily said to be appended to a branch given off from the most backwardly placed but one of the posterior pedal nerves. But Simroth states that its nerve is derived from the cerebro-pedal connective and so figures it. See Z. W. Z. xxvi. 1876, PI. XVI. Fig. 56, and p. 138, ante. It is not always symmetrically developed on both sides, and, when present on one side only, it has been found to contain two otoliths. It is situated in a part of the foot narrow from side to side, at the junction of its anterior two-thirds to its posterior third, and near to its purely muscular portion into which the viscera do not enter. k. Commissure between the left cerebro-pleural and visceral ganglia. It passes between the fibres of the retractor pedis and the protractor through the upper part of the foot, internally to the generative orifice, / ; then through the glandular portion of the nephridium, s ; and across the tendon of the retractor pedis posterior where it bifurcates for insertion into either valve of the shell. /. Left visceral ganglion. Two nerves are figured" in connection with it, one, a parietal nerve, going to the mantle, the other, a branchial nerve, going to the gill. m. Rectum ending in the cloaca. A delicate nerve is figured by Duvernoy, op. cit. p. 138, ante, as passing to it from the visceral ganglion. n. Heart ; the letter pointing to the slit left by removal of the left auricle. o. Pericardial space into which opens the glandular portion of the nephri- dium. /. External opening of the nephridial duct. q. Opening by which the glandular portion of the nephridium communi- cates with the duct r. Wide opening by which the ducts of the two nephridia communicate. This opening does not exist in Unio margaritifer. s. Secretory or glandular portion of the nephridium, reaching from the level of the anterior end of the pericardial space to the under surface of the posterior adductor. It opens into the pericardium by a canal FRESH-WATER MUSSEL. 287 along which a bristle has been drawn as passing. The glandular por- tions of the two nephridia communicate freely with each other, as do also the excretory sacs in this species. /. Orifice of the duct of the generative gland. This orifice is concealed in Anodonta, though not in Unio, by the attachment of the inner gill- lamina to the visceral mass. See V. Baer, Meckel's Archiv. 1830, P. PLATE VII. LAMELLIBRANCHIATA. LA MELLIBRA NCHIA TA. 389 PLATE VII. LAMELLIBRANCHIATA. FIG. i. The common Oyster (Ostrea edulis), dissected so as to show the principal features of its anatomy. THE animal has been removed from its shell and dissected on its left side, the one that corresponds to the flat or free valve of the shell. It may therefore be compared without difficulty with the figure of Anodonta^ Plate vi., which has been dissected in the same way. f a. a. a. Right lobe of the mantle which has been left entire, the left lobe being cut away save at its oral end. a '. Anterior dorsal angle. a '. Anterior ventral angle, which is produced a little beyond the dorsal angle. The part included between these two angles is fitted into a deep recess in the right valve, and the ligament of the shell, which is internal, corresponds to the straight edge uniting them. b. The two oral tentacles of the left side. The tentacles are not quite symmetrical in this animal. Note the deep bay which lies between them and the cut edge of the left mantle which is remarkably thickened in this region. c. The gills or branchiae. There are four of these as in Anodonta. They are not symmetrical inter se, and are fluted, J. e. the lamellae are not flat but undulated. c. The spot where the attachment of the gills to the mantle lobes ends. This attachment, divides the mantle cavity into an inhalent, oral or infra-branchial chamber, of great length but shallow in the Oyster, and an exhalent, aboral or supra-branchial chamber, which is deep and of less extent than the oral chamber. Note the four lines of apertures into the interlamellar spaces of the gills. d. The single adductor muscle, the sole adductor present in the families Ostraeidae and Aviculidae. The Lamellibranchiata have been divided into the Monomyaria with one, and Dimyaria with two adductor muscles, as in Anodonta, but the division is not a good one. This single adductor corresponds to the posterior adductor &t Anodonta. It is distinctly divisible into two parts, an opaque portion close to the peri- cardium, and a translucid portion behind. This division is observable in the adductors of many Lamellibranchiata. both anterior and pos- terior. Coutance has stated that the muscular fibres of the translucid portion in Pecten are striated, of the opaque, smooth ; that the former contract rapidly, the latter slowly; that the opaque portion is more U DESCRIPTION OF THE PLATES. like a ligament in function. But the histological difference between the parts does not extend to other Lamellibranchiata. e. The left auricle, and /. The single ventricle of the heart. The auricles are fused together in the middle, while their two extremities are free, i. e. they receive blood from the gills by two channels and communicate with the ventricle by two short vessels, the ventricular orifices of which are guarded by two valves apiece. The ventricle is not perforated by the intestine, a point in which the Oyster resembles Anomia and Teredo. The heart lies in a pericardium situated anteriorly to the adductor and com- municating with the exterior through the nephridium. f. The body of the animal which contains the digestive tract, the liver or hepato-pancreas, the organs of generation, and a part of the renal organ. A short ' oral process ' of Hoek, which contains a loop of intestine (Fig. 2, c.}, projects from it just in front of the visceral gan- glion,^. g. The left visceral ganglion, of very large size, and when seen from the surface instead of sideways, bilaterally symmetrical. It gives off posteriorly and laterally nerves which branch repeatedly, possess a muscular sheath (Hoek), and are connected at the margin with the pallial nerve. This nerve is complete, i. e. extends along the whole edge of the mantle. Anteriorly the ganglion gives off in addition to the cerebro-pleural commissure two nerves, one of which passes between the adductor and the pericardium to the mantle. The other is the branchial nerve, and is seen passing along the commissure of the two left gills. Between this nerve and the cerebro-pleural commissure, and behind the spot where they cross one another, is a depression, the common vestibule of the renal and generative ducts. h. Cerebro-pleural ganglion of the left side sending forwards branches to the pallial nerve. There is no foot and in consequence no pair of pedal ganglia. These ganglia are either fused with the cerebro-pleural or else aborted ; but there is a nerve-cord passing beneath the mouth from one to the other cerebro-pleural ganglion which represents the cord connecting the two pedal ganglia of Anomia. In this animal they are approximated to the cerebro-pleural ganglia and are of large size. i. Commissure between the cerebro-pleural and visceral ganglia. Two nerves going to the body originate from it. j. Large funnel-shaped anus. The organs of generation as described by Hoek extend over the surface of the body and the anterior surface of the pericardium. There are two glands, a right and left, but they are connected peripherally. They form a system of anastomosing and interlacing channels beneath the integument from which caeca descend vertically LAMELLIBRANCHIA TA. 291 into the body. In these caeca the generative products both male and female