£1 ELEMENTS OF COMPARATIVE ANATOMY. ELEMENTS or COMPARATIVE ANATOMY. BY CAEL GEGENBAUE, Professor of Anatomy and Director of the Anatomical Institute at Heidelberg. TRANSLATED BY F. JEFFREY BELL, B.A., Magdalen College, Oxford. THE TEANSLATION EEYISED AND A PBEFACE WRITTEN BY E. RAY LANKESTER, M.A., F.R.S., Fellow of Exeter College, Oxford, and Professor of Zoology and Comparative Anatomy in University College, London. LONDON : MACMILLAN AND CO. 1878. CHARLES DICKENS AND EVAN'S, CRYSTAL PALACE PRESS. 3 2 L t PREFACE TO THE SECOND EDITION. Ouk knowledge has been so much increased in extent and exactness in almost every department of Comparative Anatomy since the time when I converted my " Grundziige " into the first edition of this smaller manual — the " Grundriss " — that the publication of a second edition hardly seemed an easy task. Nevertheless, I gladly undertook it, for I had observed so much new evidence of the importance of the doctrine of development in anatomical enquiry. The road along which science may travel forward success- fully seems indeed to be growing easier, yet the distance which we have made is but short in comparison with that which lies in front of us, and far beyond our view. Every question solved leads again to fresh problems, and renders unstable even what seemed to have taken a definite form. There are, therefore, great difficulties in giving such a comprehensive presentation of the subject as a text-book ought to supply. I have tried as much as possible to evade these difficulties where I have been unable to overcome them. Much remains unaltered, because recent in- vestigations appear to demand fundamental changes, the concrete expression of which cannot be immediately taken in hand. I have somewhat modified the arrangement of the matter. I can hardly be blamed for separating the Brachiopoda from the Mollusca, and treating them as forming an independent phylum. Nor indeed is the change a real one, for even in my " Grundziige '* I drew especial attention to the great difference that obtained between them and the " other Mollusca." The Tunicata have vi PREFACE. been treated in the same way, but this does not require any apology at the present date. By treating the subject more concisely I have been able to increase the real matter to a certain extent, without enlarging the size of the book. I have, of course, only dealt with what has seemed to me to be of capital importance ; many, and even important, details have been omitted, owing to the limits imposed by the aim of the book. I have endeavoured to correct some previous mistakes and to supply omissions. If any such have been retained, or have newly crept in, I shall be fairly judged, I know, by anatomists, who will remember the vast extent of our science and the object of this work. I hope that I have satisfied them, and if I have my toil is well repaid. Heidelberg, November, 1877. C. Gegenbaur. PREFACE TO THE ENGLISH TRANSLATION. It is a great pleasure to me to be able to place in tbe bands of my pupils in Oxford and London an Englisb translation of Professor Gegenbaur's " Grundriss der Vergleicbenden Anatomie." I have to tbank tbe energy and industry of Mr. Jeffrey Bell, of Magdalen College, Oxford (now one of the staff of tbe British Museum), for the translation which he undertook and carried through at my request, when I found tbat my time was too fully occu- pied with other work to allow of my completing it myself within a suffi- ciently short period from the date of publication of the German work. My share of the present work has therefore consisted in a careful revision of the MS. and proof-sheets, which has been by no means a mere formality, but enables me to give the assurance that the original work is faithfully rendered in the translation. The chapter on the Tunicata I took occasion to translate myself. That Professor Gegenbaur's work will be of great service to those English students who do not already read German cannot be doubted. A\ro have some excellent treatises in the English language on animal morpho- logy, notably the Manuals of the Anatomy of Vertebrate and Invertebrate Animals, by Professor Huxley. But we do not possess any modern work on Comparative Anatomy, properly so-called ; that is to say, a work in which the comparative method is put prominently forward as the guiding principle in the treatment of the results of anatomical investigation. The present work therefore appears to me to form a most important supplement to our existing treatises on the structure and classification of animals. It has, over and above this, a distinctive and weighty recommendation in that throughout and without reserve the Doctrine of Evolution appears as the living, moving investment of the dry bones of anatomical fact. ISTot only is the student thus taught to retain and accumulate his facts in relation to definite problems which are actually exercising the ingenuity of investigators, viii PREFACE. but he is encouraged, and to a certain extent trained, in the healthy use of his speculative faculties ; in fact the one great method by which new know- ledge is attained, whether of little things or of big things — the method of observation (or experiment), directed by speculation — becomes the con- scious and distinctive characteristic of his mental activity. Thus Ave may claim for the study of Comparative Anatomy, as set forth in the present work, the power of developing what is called " common sense " into the more precisely fixed " scientific habit " of mind. I have made no notes nor additions of any kind to the original text, with the exception of a few references to English works likely to be useful to the English student. These additions are indicated by brackets. "Whilst the work is thus presented to the reader precisely as its author designed that it should be, there can be no objection to the introduction in this place of a few remarks suggested by the fact that this English translation is intended for the use of English students, and that it is therefore desirable, in order to prevent confusion and perplexity, to point out certain statements of fact, or of interpretation of fact, in which Professor Gegenbaub differs widely from authorities usually followed in this country. I shall, moreover, refer to some recent additions to knowledge published since this work left Professor Gegenbaur's hands. It will be understood that the following paragraphs are intended as a supplement necessitated by the special objects of this translation, and are by no means to be regarded as conceived in the spirit of criticism or discussion, which would assuredly ill befit a writer who is making known to a new audience the teachings of a master to whom he is deeply indebted. Nuclei of Cells. — In the first place, it seems necessary to notice that, whilst the last German edition of this work was in the press, very important additions to our knowledge of the nucleus of organic cells or plastids were being made. Though these investigations are not yet complete they tend to modify what is said concerning the nucleus on pages 15 and 16. The student is referred to an article by Mr. Priestley in the Quart. Journ. Microsc. Science, vol. xvi. (1876), for an account of the observations of Auerbach, Strasburger, Hertwig, and Van Benedex, and to part iii. of the same Journal, vol xviii. (1878), for original observations on the same subject by Dr. Klein. Reproduction of Infusoria. — A most important modification in the current views as to the reproduction of the Infusoria has resulted from the same line of study as that just mentioned, when carried into the domain of unicellular organisms. 0. Butschli and Engelmann have shown that we are not at present in a position to assert that the process of con- jugation in the Infusoria is followed by a production of spores (see § 70). It results from their investigations that conjugation in the Infusoria is attended by a deiinitc breaking-up of the nucleus and so-called nucleolus (paranucleus) of the conjugating individuals ; but that the conjugating PREFACE. ix individuals separate, and after expelling portions of the broken-up nuclear structures (probably as effete products), proceed to re-form the nucleus, or nucleus and nucleolus characteristic of the species. The so-called Acineti- form embryos appear to be parasites, the rod-like bodies occasionally observed in the nucleus are also parasites, whilst the striated structure and spindle- shape exhibited by the nucleolus or paranucleus in such forms as Para- moecium and Stylonichia at the period of conjugation, are simply due to changes in this body which are exactly paralleled in the nuclei of egg-cells and other tissue-elements of multicellular organisms, when those cells are about to divide by transverse fission. The process of conjugation in the Infusoria may be, and probably is, attended by an exchange of nuclear material between the conjugating individuals, and is so far comparable to sexual congress, but it results in a simple " rejuvenescence " of the conjugating individuals and not in a production of spores. Reproduction by fission and by the modification of fission, known as gemmation, has been accurately observed in Infusoria, but of the formation of " spores " in this group Ave are at present ignorant, in spite of all that has been written on the subject. Origin of Male and Female Reproductive Elements from different Germ-layers. — In § 95 Professor Gegenbaur has described the observations of Ed. Van Bexeden on the development of the sexual products in Hydractinia, and has adopted his generalisation, so far at least as it applies to the Hydromedusa\ From more recent observa- tions (Ciamiciax, Zeitschr. firr wiss. Zoologie, vol. xxx. p. 501, 1878) it appears that in other genera of hydroid polyps the same arrangement does not obtain. In Eudendrium ramosum the ova appear to develop from the ectoderm, and the sperm from the endoderm ; in Tubularia mesem- bryanthemum both ova and sperm are ectodermal in origin according to Ciamician ; Van Beneden found the ova to be endodermal and the sperm ectodermal in Hydractinia, whilst Kleinenberg ascribes both to the ecto- derm in Hydra. Nervous System and Sensory Organs of Medusa?. — During the past year a considerable addition has been made to knowledge on these points, by the researches of the two Hertwigs (" Das Xerven system und die Sinnesorgane der Medusen." Leipzig, 1877). It is no longer possible to deny the existence of differentiated nervous tissue in the Medusa1 — the central organ having the form of a ring situated along the line of insertion of the velum in the Craspedota, and of a series of isolated ganglia, usually eight in number, placed on the edge of the disc in the Acraspeda. (See for an abstract of recent researches on this subject, Quart. Journal of Microsc. Science, vol. xviii. p. 310.) Cirri and Elytra of Aphroditacerc. — The statement in § 105, that the elytra of the chsetopodous Worms, allied to Aphrodite, are formed by the metamorphosis of the dorsal cirri of the parapodia, appears to be contradicted x PREFACE. by tlie fact, that in Sigalion the elytra and dorsal cirri exist side by side on the same segment. Homologies of the Kami of the Appendages in Astacus. — The view taken by Professor Gegenbaur, as to the homologies of the parts of the appendages immediately following the month in Astacus, differs somewhat from that which is current in this country. In Fig. 122, p. 239, the mandible, two maxilla?, and three maxillipedcs of the right side of Astacus lluviatilis are figured. This woodcut was kindly re-drawn for the English edition by the author, at my request, and gives a more complete outline of the parts in question, than does the older cut of the German edition. Throughout the series of appendages, three divisions are distinguished by the letters a, c, d. Taking the lowest figure first (the third maxillipede) we find the endopodite marked a, the exopodite marked c, and the letter d placed with the single epipodite (podobranchia, Huxley) to its inner side, whilst the double arthrobranchia (Huxley) not forming part of the appendage proper, but a distinct respiratory development, is seen on its outer side. In the next figure (the second maxillipede), a indicates endopodite, c exopodite, and d is placed close to the double arthrobranchia on its outer side, whilst the modified epipodite is seen to the inner side again, of this. In the figure of the first maxillipede, a is placed near the foliaceous endopodite, which has a detached outstanding segment, c near the filamentous exopodite, and d near the broad epipodite. The same explanation of the lettering holds good for the next appendage, the second maxilla. In the next appendage — the first maxilla — the absence of the letters e and '/, indicates that the author regards the whole appendage as reduced to the representative of the foliaceous endopodite a of the two inferior appendages — a view with which few will disagree. In the case of the mandible, however, Professor Gegenbaur marks the " palp " with the letter c — considering, therefore, the basal piece of this appendage to represent the endopodite, and the palp to represent the exopodite. The more usual opinion on this matter is that the mandible, together with its palp, corresponds to the simple foliaceous first maxilla. The jointed palp, mounted on its solid basal biting-piece, cor- responds to the jointed eudopodite a of the last maxillipede. The question of the presence or the absence of a representative of the exopodite in the Decapod's mandible, is a matter of considerable importance in reference to possible comparisons between the gnathites of Crustacea and Tracheata. The actual development of the parts in question from the nauplius-form of appendage, must be the idtimatc test of the homologies of their rami in the Crustacea. Blood-corpuscles of the Mollusca. — The statement on p. 375, that "the form-elements of the blood are always colourless" in the Mollusca, is one which I may be allowed to correct, since I have published an account of the blood-corpuscles of Solen legumen (Proc. Poyal Society, No. 140, 1873), which, besides colourless amoeboid forms, comprise a vast number of PREFACE. xi oval ones, deeply stained by haemoglobin. The number of these corpuscles is so considerable as to give the blood of Solen legumen a bright blood- red colour. I may add here that I have observed similar though larger corpuscles impregnated with haemoglobin in the blood of species of Area. Homologies of the Arms of the Cephalopoda. — The view that the sucker-bearing arms of the cuttlefish are to be regarded as appendages of the head homologous with the tentacles on the head of Gasteropods (p. 326), is one which, it will be well for the student to remember, is not that usually taught. He should make himself acquainted with the older and the neAver view, and the grounds on which they are based. Without entering into a discussion of the arguments which may be adduced in favour of this or of rival interpretations of the parts, it must suffice here briefly to mention that the arms of the Cephalopod (the development of which had been made known by Kolliker), were shown by Professor Huxley, five-and-twenty years ago, to correspond to the fore-part of the foot of the Gasteropoda, and the ganglion, from which they receive their nerve supply, was then considered as corresponding to the pedal (Morphology of the Cephalous Mollusca, Phil. Trans. 1853). This view was maintained in the earlier editions of Gegenbaur's work. It has been abandoned in the present edition, in deference to the statements of Mr. Jhering (" Vergleichende Anatomie des JSTervensystems und Phylogenie der Molluscen, Leipzig," 1877). The whole of that author's work, both statement of fact and speculative superstructure, appears to me to call for very cautious treatment, involving the rejection of some of his principal conclusions. Origin of the Limbs of Vertebrates. — Professor Gegenbaur is inclined to regard the skeleton of the limbs and limb-girdles of Vertebrata as derived from gill-arches and their branchial rays (§ 357). The student is reminded that another possible derivation of these organs is from primitively continuous lateral fins — supported by cartilaginous rays, and comparable to the primitively continuous dorsal median fin. The specialisation and con- centration of the lateral fin on each side in two regions, thoracic and pelvic, woidd be competent to give rise to the two pairs of fins, such as we find in the Elasmobranchs. Mr. Balfour (" Development of Elasmobranch Fishes," 1878) is led to adopt this view by the observation, that in the embryo dog- fish the lateral fins have precisely the same mode of origin as has the dorsal median fin, arising " as special developments of a continuous ridge on each side, precisely like the ridges of epiblast, which form the rudiments of the unpaired fins." This view of the nature of the vertebrate limbs has been independently worked out with great care from the point of view of comparative anatomy, by Mr. J. K. Thacher (Median and Paired Fins, Transactions Connecticut Academy, vol. iii. 1877). In the important memoir just cited, Mr. Thacher shows very plausibly how the Elasmobranch fin, and not only the fin, but the supporting limb-girdle also, may have xii PREFACE. been derived from the gradual shifting, atrophy, hypertrophy, and con- crescence of primitively similar cartilaginous rods, which formed a series on each side of the body, identical in character with the primitive median dorsal series. According to this view, the " archipterygium " of Professor Gegenbaur is not antecedent to, but is derived from the type of fin found in Elasmobranchs. (See also on this subject, Huxley, On Ceratodus, Proc. Zool. Soc. vol. 1876, p. 24.) Eelation of the Malleus and Incus to the Mandibular and Hyoid Arches. — Investigations directed to the development of the skull led Professor Huxley some years since to adopt the conclusion of Eeichert and of Goodsir, that the small bones of the Mammals' tympanic cavity were derived from the upper ends of the anterior visceral arches. At first it appeared probable that the malleus and incus were both derived from the upper end of the cartilaginous mandibular arch, the lower part forming Meckel's cartilage. This led to the suggestion that the malleus corresponds to the articulare of the lower jaAv of other Vertebrata, whilst the incus was considered to be the representative of the quadratum, since it articulates with the malleus just as the quadratum does with the articulare (Croonian Lecture " On the Theory of the Vertebrate Skull," Proc. Loyal Society, vol. ix. p. 398). Further investigation led Professor Huxley to a modification of his views. The embryological evidence is not quite complete, but the relations of the parts in question in the developing Frog, in certain Lizards, and in Mam- malia, have led him to the conclusion (" Manual of Vertebrate Anatomy," p. 85, 1871) that whilst the malleus is formed from the uppermost extremity of the mandibular arch, and therefore represents, not articulare, but quad- ratum, the incus is developed from the uppermost extremity of the second or hyoid arch, and corresponds to the hyomandihular of fishes. The stapes is also developed from the upper portion of the hyoid arch, just below the incus. The incus may therefore be spoken of as the supra-stapedial portion of the hyoid arch, and in certain Vertebrata it exists as a mere cartilaginous supra-stapedial rudiment. These vieAvs in their later form have not been adopted by Professor Gegexbaur. He observes (§ 402) that the homologies of the ossicula auditiis of the various classes of Vertebrata have not yet been satisfactorily determined. In § 352 he maintains the earlier determination of the homo- logy of the mammalian malleus with the articulare of other Vertebrates. Concerning the homologies of the incus and the stapes, he considers it advisable, in the present state of knowledge, to make no statement. The student is advised of these differences of interpretation of structural fact, in order that he may the more carefully make himself acquainted from original sources with the details of development, relation to nerves, and other features of the parts under discussion. Nomenclature of the Lobes of the Brain in Fishes. — In the earlier editions of the present work, Professor Gegexbaur, led by the result PREFACE. xiii of investigations carried out by his pupil Miklucho-Maclay (" Vergleich. Neurologie der Wirbelthiere," 1870), modified the current nomenclature of the lobes of the Fish's brain, so that the large bispherical part, which was usually considered as the mesencephalon in the Teleostei and Selachii, was assigned to the thalamencephalon — or second of the five cerebral segments — whilst the unpaired large projecting lobe, usually considered as the metencephalon (cerebellum, fourth segment), was identified with the mesencephalon of higher Vertebrates, and the cerebellum was considered as being represented by a small transverse plate, often overlapped by the folded mesencephalon, and usually of no larger size than the piece similarly identified in the frog. In the present edition Professor Gegenbaur has modified this system of nomen- clature, and has returned to the older and usually accepted method of naming the parts of the Fish's brain. Thus in Fig. 281, d marks the two spherical masses which were in former editions assigned to thalamencephalon, and are now, as is usual with other anatomists, designated mesencephalon, the expansion between them and g being the reduced area of the thalamen- cephalon. The letter b is now referred to as metencephalon (cerebellum) : this was previously referred to as mesencephalon ; the myelencephalon prosencephalon, and rhinencephala retain their names, which had not been affected by Maclay's system. "Whilst Professor Gegenbaur has returned to the usual system of naming these parts, he still considers that the facts on which M aclay's nomenclature was based possibly point to homologies other than those indicated by the names ; so that the Fish's cerebellum does not necessarily agree with that of higher Vertebrata. He remarks : " The mesencephalon is usually considered as being confluent with the thalamencephalon in Selachians ; and a part which really represents it, so far at least as relations of position are con- cerned, is customarily called by the name ' cerebellum.' " In translating the German terms, Vorderhirn, Zwischenhirn, Mittelhirn, Ilinterhirn, and Nachirn, I have adopted Professor Huxley's equivalents, namely Prosencephalon, Thalamencephalon, Mesencephalon, Metencephalon, and Myelencephalon. In the edition of Quaix and Sharpey's Anatomy, published in 1867, a similar but not identical series of terms was suggested. For the "primitivenHirnschlitz," — the early strongly-marked sinking in of the cerebral roof which separates the prosencephalon from the thalamencephalon —we have no special term in use ; " primitive cerebral cleft " is the transla- tion which has been adopted. It is worth while pointing out to the student, in connection with this subject, and in fact in relation to the whole of the chapter on the Vertebrata, that Professor Gegexbaur assumes some small amount of familiarity on the part of the reader with descriptive human anatomy ; reference to a manual treating of this subject, on the part of the student who has not previously mastered it, is indispensable. Nomenclature of the Parts of the Digestive Tract. — The transla- tion in the present work of the simple word "Darm," and its compounds xiv PREFACE. Vorderdarm, Mitteldarm, Hinterdarm, Kopfdarm, has caused me some perplexity. It lias been variously rendered in the translation by "gut," "enteron," "enteric tube," " alimentary canal," "digestive tract." The fact is that, whilst Ave have no definite nomenclature at present in use in English which recognises the true morphology of the canal which com- mences with the mouth and ends with the anus, the nomenclature in use in Germany is of very doubtful advantage, since it has not a sound morpho- logical basis, but is altogether superficial. " Darm," for which our readiest equivalent is "gut," is used indifferently for the whole or for any part of the physiological entity which reaches from oral to anal aperture, Uut the English word "gut " is associated rather with the hinder than with the fore- most portion of this tract. It will probably be found most convenient to speak of the physiological whole as the "alimentary canal," or "digestive tube ;" and these terms I have endeavoured consistently to make use of in this sense, though sometimes the term "enteric tube" has been similarly applied. The division of this tube or canal into pharynx, oesophagus, stomach, and intestine ; or, again, into fore-gut, mid-gut, and hind-gut (Vorderdarm, Mitteldarm, Hinterdarm, p. 48), is one based upon superficial adaptations of form, and does not admit of a comparison of the parts so designated in the various phyla of the Animal Kingdom. The pharynx and the oeso- phagus of the Vertebrata are developed from the endoderm of the embryo ; the parts which receive the same names in the Mollusca and the Arthropoda are developed from the ectoderm. The hind-gut of the Vertebrate is endo- dermal in origin, ectodermal in the Arthropod, and partly endodermal partly ectodermal in the Mollusca. In fact there is no attempt to recog- nise the facts of embryology in the terminology applied to the alimentary canal. Under these circumstances I have proposed (Quarterly Journ. Microsc. Science, April, 1876, and "Notes on Embryology and Classification," London, 1877, p. 11), to distinguish the primitive digestive space which develops from the endoderm (in fact the gastrula-stomach) as the " enteron." The anterior passage leading into this from the mouth, and formed by an ingrowth of ectoderm, I have termed the " stomodamm," and the corresponding passage leading from the anus I similarly propose to call the "proctodeum." These three primary factors of the alimentary tract are most equally developed in the Arthropoda and some Mollusca. In Vertebrata the stomodamm. is exceedingly small, if indeed its true homo- logue exists at all (excepting in the Tunicata). The proctodeum is also in them evanescent. The middle portion of the alimentary tract formed from the primitive enteron (archenteron), which does not entirely coincide with that part to which the term " Mitteldarm " is applied, does not in all the various animal phyla take up the whole of the primitive enteron. This, in fact, only occurs in some of the Coelenterata, which may therefore be said to possess in the adult condition an archenteron. In other groups the PREFACE. xv primitive enteric .sac gives off the foundations for a variety of other structures, so that what is left of it as the central element of the alimentary canal is a changed and broken-up enteron, which maybe called "metenteron" as opposed to the unchanged " archenteron." It is to these three morphological factors then, the metenteron, the stomodseum, and the proctodeum, that we are called upon to assign the various adaptational swellings, constrictions, and outgrowths of the alimen- tary tract of higher animals. These distinctions arc not recognised in Professor Gegexbaur's work. It will be sufficient here to point out that the exact limit of stomodieum and of proctodeum in any particular case, can only be ascertained by direct observation of the process of development. The metenteron is that part of the alimentary canal with which the most important digestive glands are connected, such as the liver, and from its walls they arc formed as outgrowths. The stomodeum gives rise to salivary glands, and usually to masticatory sacs (gizzards), but these latter may form also in the metenteron. The proctodeum forms the cloacal chamber, where such exists, and always receives the openings of glands (such as the Malpighian filaments of insects) which are excretory rather than accessory to digestion. These explanations will be sufficient to make clear to the reader the sense in which the words " enteron " and " enteric " have occasionally been employed in the translation. Classification. — At the present day, naturalists have learnt to recog- nise in their efforts after what was vaguely called the " natural " system of classification, an unconscious attempt to construct the pedigree of the animal world. The attempt has now become a conscious one. Necessarily classifications which aim at exhibiting the pedigree, vary from year to year with the increase in our knowledge. They also vary according to the importance attached by their authors to one or another class of facts as demonstrating blood-relationships. Probably no two zoologists of the present day would agree, within wide limits, as to the classification which comes nearest to expressing the pedigree. Accordingly it is by no means desirable that students should be taught to accept any one scheme of clas- sification as finite. They should be taught to look upon these schemes as the condensed expression of an author's views — as the epitome of his teaching, facilitating the recollection and the comparison of conflicting solutions of the vast series of unsolved problems of morphology. I propose here, for the convenience of the student, to place side by side the general outlines of the schemes of classification adopted by Professor Huxley in 18G9 (No. I.), that adopted by Professor Gegenbaur in the present volume (Xo. II.), and that which I have made use of in my lectures during the past year (Xo. III.). I have taken the older classification adopted by Professor Huxley rather than that more recently put forward by him, because it is one with which XVI PREFACE. my experience as teacher and examiner has shown me that English students are thoroughly familiarised. I. SUB-KINGDOMS. Protozoa. (Rhizopoda, Gregarinida, Radiolaria, Spongida.) Infusoria, ccelentkrata. (Hydrozoa, Aclinozoa.) Axxuloida. (Scolecida, Echinoderma.) Axnulosa. (Crustacea, Araclmida, Myriapoda, Insecta, Clue- tognatha, Annelida.) MoLIiUSCO'iDA. (Polyzoa, Brachiopoda, Tunicata.) Mollusca. (Lamellibrauchiata, Bran- chiogastropoda, Pulmo- gastropoda, Pteropoda, Cephalopoda.) Vertebrata. (Pisces, Amphibia, Rejitilia, Aves, Mammalia.) 11. III. PHYLA. PHYLA. Protozoa.* + Protozoa.* CffiLEXTERATA. ♦ PORIFERA. Vermes.! Nematophora. * * * Echixoderma. Platyfielmia.|| Gephyrea.|| BRAcxnoropA.J Echixoderma. Arthropoda. Enteropxeusta.|| Nematoidea. || MOLLUSCA. Ch.etognatiia.|| TUXICATA.§ Appendiculata.^" MOLLUSCA.** Vertebrata. VERTEBRATA.ft Seeing that one of my chief objects in superintending the translation of the treatise to which these few pages are introductory, has been to he able * The Protozoa in Nos. II. and III. inclnde the same organisms as in No. I., excepting that the Infusoria are included in that phylum in Nos. II. and III., and that the Sponges are excluded, being in No. II. placed under the Ccelenterata, and in No. III. forming the phylum Porifera under the " grade " Coelentera, as shown in the genealogical tree on the adjacent page. f The Vermes of No. II. include all the Annulo'ida of No. I. excepting the Ecliinoderma, which are raised to the rank of an independent phylum. They also include the Annelida (Cha^topoda, llirudinea, and Gcphyrea) from amongst the Annnlosa of No. I. and the Polyzoa from amongst the Mollusco'ida of the same series. J The Brachiopoda, raised to the position of a distinct phylum in No. II., are placed among the Mollusco'ida in No. I. and amongst the Mollusca in No. III. § The Tunicata, considered as an independent phylum in No. II., arc found amongst the Mollusco'ida in No. I. and form a section of the Vertebrata in No. III. The Platyhelmia, Gephyrea, Enteropneusta, Nematoidea, and Chatognatha form in No. III. a number of independent phyla. Together with the Polyzoa (included in No. III. under the Mollusca), the Rotifera, and the Cluetopoda, included under the Appcndiculata, they constitute the scries of phyla which are in No. 11. massed together as " Vermes." Tf The Appendiculata include animals with lateral locomotive appendages, and usually a segmented bod}'. The group is, excepting that it has the addition of the Rotifera, nearly co-extensive with the Annulosa of No. I. ** The phylum Mollusca in No. III. includes the Polyzoa and Brachiopoda, excluded from it in both No. I. and No. II. ft The Vertebrate phylum in No. III. includes the Tunicata, which it will be seen by reference to page 70 are already placed on the Vertebrate stem by Professor Gegenbanr. PREFACE. xvu to place the work in the hands of the students of my oavii classes, I need not apologise for adding here further details of the classification which I find it most convenient to adopt in teaching. I have arranged the chief phyla first of all in the form of a genealogical tree, and secondly in a series exhibiting their subdivisions into classes, etc. This classification is of course to a large extent only a modification and adaptation of systems already put forward by other naturalists. <3 -H <3 c3 13 e8 & o s 0 ft P. o d o -. A ft O 3 © a ft o u a c3 d o •4-) O c3 ^3 Grade II. CCELOMATA. <3 •a o P4 Grade I. CCELENTERA. Grade II. ENTEROZOA. o N o ■*-> o e3 o n o o u Grade I. PLASTIDOZOA. A xviii PREFACE. PROTOZOA. GRADE A. GYMNOMYXA. Class 1. Gymuomyxa. GRADE B. COR TWA TA . Class 1. Lipostotna (Grognrinre). 2. Suctoria. 3. Ciliata. d. Flagellata. 5. Probosciclea (Noctiluca). rORIFERA. NEMATOPTIORA. Class 1. Calcispongia?. 2. Fibrospongia\ 3. Myxospongiro. Class 1. HydromcdnsEe. 2. Scyphomcdusa?. 3. Anthozoa. 4. Ctenophora. ECHINODERMA BRANCH. AMBULACRATA. Class 1. Holothuridoa. 2. Echinoidea. 3. Asteroidca. BRANCH. TENTACULATA. Class 1. Crinoidea. 2. Blastoidca. 3. Cystidea. LATYHELMIA. BRANCH. CILIATA. Class 1. Plauarise. 2. Ncmertina. BRANCH. SUCTORIA. Class 1. Trcmatoidea. 2. Cestoidea. 3. Hirudinca. GEPHYREA. Class 1. Ecliiuridic. 2. Priapulida3. 3. Sipunculidse. 4. rhoronida?. PEEFACE. Xlx VERTEBRATA. BRANCH. UROCHORDA (TUNICATAt. Class 1. Larvalia. 2. Saccata. BRANCH. CEPHALOCHORDA. Class. Leptocardia. BRANCH. CRANIATA. GRADE A. CYCLOSTOMA. Class. Cjclostoma. GRADE B. GNATUOSTOMA. Grade a. Hcterodactyla branchiata. Class 1. Pisces. 2. Dipnoi. Grade fi, Pentadaetyla branchiata. Class. Amphibia. Grade y. Pentadaetyla Ivpobranchia. Class 1. Keptilia. j = Branch. jfonocondyla. 2. Aves. ) 3. Mammalia. = Branch. Amphieondyla. Al'PENDICULATA. BRANCH. CILETOPODA. Class 1. 01igocha;ta. Class 2. Polychasta. BRANCH. ROTIEERA. Class. (Orders only.) BRANCH. GNATHOPODA (ARTHROPODA). GRADE A. MALACOPODA. Class. Peripatidea. GRADE D. CONDYLOPODA. Class 1. Crustacea. 2. Jlexapoda. 3. Myriapoda. 4. Arachnida.* * Following Prof. Ed. Van Beneden, I include Limulus, the Euryptorina, and Trilobites ufader the Arachnida as Branchiopulmonata. xs PREFACE. MOLLUSOA. BRANCH. EUCEPHALA. GRADE A. LIPOGLOSSA. Class. Scolecoinorpha (Neomeiiia). GRADE B. ECIIINOGLOSSA. Class 1. Gastropoda.* 2. Cephalopoda.! 3. Scaphopoda. BRANCH. LIPOCEPHALA. Class 1. Tentaculibranchia (Polyzoa). 2. Spirobrauchia (Brachiopoda). 3. Lamellibranchia. The phyla Enteropneusta, Chsetognatha, and Nematoidea, containing respectively the genus Balanoglossus, the genus Sagitta, and the various families of thread-worms, do not admit of subdivision into classes or orders. * Includes tbc Chitons as a separate archaic grade " Amphomoca," tho remaining Gastropoda, all of which are asymmetrical, being placed in a higher grade as " Cochlides." t Includes Siphonopoda (Cuttles and Nautilus) and Pteropoda. E. KAY LAKKESTEE, Exeter College, Oxford. September, 1878. TABLE OF CONTENTS. Paragraph 1. 2. Introduction 3 — 10. Scope of Comparative Anatomy General Part. 11. Structure of the animal body 13 Organs and organism ......... 13 12. Differentiation ........... 1-4 13 — 15. Earliest stage of the animal organism ..... 15 The cell 15 16. Differentiation of the animal organism .... 18 17. Origin of the tissues 20 18. 19. A. Vegetative tissues 21 Epithelium 21 20 — 23. Connective substances 23 24. Form-elements of the nutrient fluid 29 25. B. Animal tissues ...... .... 30 26. Muscular tissue 31 27. Nervous tissue .......... 32 28. 29. Origin of the organs 34 30. Systems of organs 37 a) Integument . . " . . . . . . . .37 31. b) Skeleton 38 32. 33. c) Muscles 39 34. 35. d) Nervous system 40 36 — 38. e) Sensory organs 42 39. f) Respiratory organs of the integument (Dermal gills) . . 45 40. g) Excretory organs ......... 46 41. h) Alimentary Canal 47 42. Respiratory organs of the enterou ..... 49 43. 44. i) Vascular system 50 45. 46. k) Reproductive organs 52 47. Metamorphosis of the organs ....... 54 Development and degeneration ....... 54 48. Correlation of the organs 57 XX11 CONTESTS. Paragraph 49. 50. Fundamental forms of the animal body 51. 52. Metamerism of the body .... 53 — 55. Comparison of the organs ..... 56—58. Classification of the Animal Kingdom . 59. Bibliographical aids in Comparative Anatomy . Pa so 58 61 63 61 75 Special Part. First Section. Protozoa. 60. General review of the group . . ...... 75 Bibliography 77 61 — 70. Organisation of the Protozoa 77 Second Section. Ccelenterata (Zoopliyta) 71. General review of the group Bibliography . 72—78. Form of the body 79. Appendages . 80. Integument . 81. 82. Skeleton 83. Muscular system 8 k Nervous system 85. Sensory organs 86—93. Alimentary canal 91-98. Generative organs 89 90 91 101 103 105 108 108 109 111 119 Third Section. Vermes. 99. General review of the group Bibliography . 100—103. Form of the body 104. 105. Appendages . 106. External gills . 107—111. Integument 112. Skeleton . 113. 114. Muscular system 115—121. Nervous system 122. 123. Sensory organs Tactile organs 124. 125. Visual organs 126. Auditory organs 127—132. Alimentary canal 133. Enteric branchia 134. 135. Accessory organs of (he al 136. Coelom imeutary canal 125 127 128 132 135 136 142 142 145 152 152 153 156 156 163 164 165 CONTENTS. xxiii Paragraph Pa£e 137—141. Vascular system 166 142 — 145. Excretory organs 172 146—156. Gonerative organs 178 Fourth Section. Echinoderma. 157. General review of the group 192 Bibliography . . . . . • • • • • .191 158—161. Form of the body 194 162. Appendages 199 163 — 167. Integument and dermal skeleton 200 168. Muscular system • • 207 169. Nervous system 203 170. Sensory organs 210 171 — 173. Alimentary canal 211 174. Organs appended to the alimentary canal 215 175. Ccelom 216 176. Vascular system 217 Blood-vessels 217 177. 178. Water-vessels ,219 179. Excretory organs . 224 180, 181. Generative organs 22* Fifth Section. Arthropoda. 182. General review of the group 228 Bibliography ........... 232 183. Form of the body . . . 234 184. Appendages 237 185. Appendages of the Branchiata ....... 238 186. 187. Branchice 240 188—190. Appendages of the Tracheata 243 191—193. Integument 248 191. Muscular system 251 195—200. Nervous system 252 201. Sensory organs 260 Tactile organs . . . . . . . . . .260 202. 203. Auditory organs 261 204—206. Visual organs 263 207—211. Alimentary canal 267 212. Organs appended to the alimentary canal 273 1) Appendages of the fore-gut 273 213. 2) „ „ mid-gut 274 214. 3) „ „ hind-gut 276 215. Ccelom 277 216—220.' Vascular system 279 221. Excretory organs 285 222—225. Trachese 286 226—237. Generative organs .... ... 291 xxiv CONTENTS. Sixth Section. Brachiopoda. Paragraph 238. General review of the group Bibliography . 239. Form of the body . 240. Integument, shell, and arms 241. Muscular system 242. Nervous system and sensory o 213. Alimentary canal . 244. Coelom and circulatory organs 245. Excretory organs 246. Generative organs . Tape 306 307 307 308 309 309 310 310 312 314 Seventh Section. Mollusca. 247. General review of the group 315 Bibliography ........... 317 248—252. Form of the body 318 253. 254. Appendages 325 255. 256. Integument 328 257—259. Shell 329 2C0— 2G3. Gills 335 264. Internal skeleton 311 265. Muscular system 342 266—269. Nervous system 343 Central organs and nerves of the body ..... 343 - 270. Visceral nerves .^351 271. Sensory organs : 351 Tactile and olfactory organs ....... 351 272. 273. Visual organs 353 274. Auditory organs . 356 275 — 279. Alimentary canal 358 280. Organs appended to the alimentary canal ..... 363 1) Appendages of the fore-gut 363 281. 2) Appendages of the mid-gut 364 282. 3) Appendages of the hind-gut 366 283. Coelom 367 284—288. Vascular system 368 289—292. Excretory organs 375 293—298. Generative organs 380 Eighth Section. Tuilicata. 299. General review of the group ........ 388 Bibliography 389 300. 301 . Form of the body 390 302. Integument 393 303. Skeleton 394 301. Muscular system 394 305. 306. Nervous system 395 CONTENTS. xxv Paragraph Page 307. Sensory organs 397 308. Alimentary canal 398 309 — 311. Eespiratory antechamber (Branchial cavity) .... 399 312. Enteron 403 313. Vascular system 404 314. Generative organs .......... 406 Ninth Section. Vertebrata. 315. General review of the group ........ 408 Bibliography ........... 412 316. Form of the body 413 317- 318. Appendages 414 319. 320. Integument 417 321 — 323. Epidermal structures 419 324—326. Dermal skeleton 422 327. Internal skeleton 426 328—334. Vertebral column 428 335—337. Ribs 438 338. Sternum 442 339. 340. Cephalic skeleton 444 341—352. Skull 447 353—356. Branchial skeleton 468 357. Skeleton of the appendages ....... 472 358 — 360. Anterior appendages ........ 474 Shoulder-girdle ......... 474 361—365. Anterior extremity 477 366. Posterior appendages ........ 484 Pelvic-girdle 484 367—369. Posterior extremity 487 370. Muscular system 491 371. Dermal muscles .......... 492 372—377. Musculature of the skeleton 493 378. Electric organs 499 379. Nervous system .......... 501 380. 381. A. Central organs of the nervous system ..... 503 a) Brain 503 382. 383. b) Spinal chord 511 384. c) Investments of the central nervous system . . . 512 385. B. Peripheral nervous system ....... 513 386. a) Spinal nerves 514 387—392. b) Cerebral nerves 515 393. c) Visceral nerves 522 394. 395. Sensory organs 523 396. Olfactory organs . . . . . . . . . .525 397—399. Visual organs 527 400—403. Auditory organs 533 404. Alimentary canal .......... 539 405. Eespiratory antechamber (Cephalic enteron) .... 540 406—409. Branchije 541 A ** XXVI CONTENTS. Paragraph Page 410. Branchial clefts and palate of the Amniota .... 545 411. 412. Nasal cavity 547 413. Buccal cavity 548 414 — 416. Organs of the buccal cavity 549 417. Alimentary canal proper (Enteron of the trunk) . . . .555 418. Fore-gut 556 419. Mid-gut 559 420. Hind-gut 561 421. Organs appended to the mid-gut ....... 563 422. Mesentery 565 423. Pneumatic organs of the enteric tube ...... 566 424. a) Air-bladder 567 425—427. b) Lungs 569 428. Coelom 574 429. 430. Vascular system 575 431 — 436. Heart and arterial system 577 437—442. Venous system 589 443 — 446. Lymphatic system 597 447 — 449. Excretory organs 601 450 — 458. Generative organs 608 CORRIGENDA. Page 79, six lines from top, for "finer" read "firmer." „ 87, three lines from top, for " generation " read " gemmation." „ 88, two lines from top, for " spiral " read " special." „ 88, in the explanation of Fig. 30, for "broad" read "brood." Introduction. The Scope of Comparative Anatomy. § I- The department of science which has organic nature for the object of its investigations, breaks up into two great divisions, Botany and Zoology, corresponding to the two kingdoms of organised nature. The two disciplines together form the science of living nature — Biology. They are closely connected with one another, in so far as the phasnomena seen in both the animal and vegetable kingdoms rest on the same fundamental laws, and in so far as, notwithstand- ing the differences of their special arrangements, animals and plants have common beginnings, and are, in the economy of nature, closely interdependent. In both of these disciplines several kinds of investigation are possible, and from these new disciplines arise. Putting aside the realm of Botany, let us follow out that of Zoology into its further subdivisions. The investigation of the functions of the animal body, or of its parts, the reduction of these functions to elementary processes, and the explanation of them by general laws, is the busi- ness of Physiology. The investigation of the material substratum of those functions, and accordingly of the phasnomena of form of the body and its parts, as well as the explanation of the phasnomena of form by reference to function, is the business of Morphology. Physiology and Morphology have thus different duties, and their methods also are different ; but for each it is necessary, although in different ways, to keep in view the other, as well as the common aim, which is indicated in the term Biology. Morphology again is divided into Anatomy and Embryology; while the former has for the object of its investigations the adult organism, the latter has the growing organism as the object of its study. B 2 COMPARATIVE ANATOMY. Anatomy may be divided into general and special Anatomy. General Anatomy has to do with the fundamental forms of animal organisms (Promorphology), and the morphological phgenomena which arise from them. Special Anatomy takes for its object the oro-anological composition of the animal body. Histology, or microscopic Anatomy, forms one of its branches, being the study of the elementary organs of the animal body. Embryology explains, by tracing their gradual development, the complications of the external and internal organisation, and, in fact, deduces them from simpler conditions. The changes in organisation can be followed out in the embryonic life of the individual, and also in the continuous series of organisms. The discipline ordinarily known as Embryology deals with the former; and as Ontogeny (or the development of the individual) is contrasted with Phylogeny (or the development of the phylum) . As the latter includes the earlier and no longer existent conditions of animals, it also embraces Palseozoology. It is the history of the development of the series of organisms in their geological succession. § 2. Since Anatomy has for its object the composition of organisms, it may be considered as the doctrine of structure, and is divided, according to the different points of view from which structure itself may be regarded, into several divisions. When the com- position of the body itself, its forms, and the relations of the separate organs are taken as its scope, it is known as descriptive Anatomy, for it describes the objects examined, without drawing any further conclusions from them. Anatomical fact is the aim of the investiga- tion, and empiricism satisfies this aim. Owing to its relations to medicine, and so to practical requirements, the descriptive Anatomy of the human organism, so far as it is restricted to a special series of facts, has become developed into a special branch, which, as Anthropotomy, is put side by side with the similarly descriptive Zootomy. The two differ only in their subject-matter, and not in their treatment of it, for both are analytical. In proportion as either abstains from drawing conclusions from its series of facts, and giving these the value of abstractions, is it wanting in the character of a science ; for a science is constituted neither by an extensive range of observations, nor by the complication of the methods by means of which such observations are made. A critical appreciation of the scientific import of any branch of study has, therefore, little to do with the mechanical apparatus of investiga- INTRODUCTION. 3 tion, which has its value only in facilitating the discovery or demon- stration of facts. Anatomy assumes a very different character so soon as the knowledge of facts is only its means, and its aim the conclusions which can be drawn from an assemblage of such facts ; the facts of individual phenomena being regarded not by themselves exclusively, but being brought into relation with one another. This happens when what is alike in the organisation of different organisms is made the object of search, and when the facts thus acquired are compared. Anatomy thus arrives at scientific results, and shapes the results of inductive inquiry into deductive conclusions. Thus it becomes Comparative Anatomy. Its method is synthetical. The analyses of Descriptive Anatomy (Anthropotomy as well as Zootomy) provide the basis for it ; they are consequently not only not excluded from Comparative Anatomy, but are most closely embraced and logically permeated by it. The more careful the sifting of facts, the surer the basis of comparison. Empiricism is thus the first requisite, and abstraction is the second. Abstraction has no basis without pre-existing empiricism; and empiricism by itself is, from the scientific point of view, a mere stepping-stone to real knowledge. § 3. The task of Comparative Anatomy is the morphological ex- planation of the phEenoinena of form met with in the organisation of the animal body. Comparison is the method which serves for the performance of this task. It shows the way which scientific investi- gation has to go, and which it is necessary to know in order to avoid disjointed and fruitless labour. The comparative method seeks to test, in series of organisms, the morphological results of the observation of the organs of the body, places together similar characters, and separates the dissimilar from them. Thus it takes into consideration everything which can in any way be looked at as the result of anatomical observation : relation to other parts of the body, form, number, extent, structure, and texture. It thus collects series of stages for the several organs, in which the extremes may be so far different from one another as not to be recognised, but which are united to one another by numerous intermediate steps. It is clear, in the first place, from the existence of various forms of one and the same organ, that the physiological value of an organ in different stages is not by any means the same, but that an organ, as its anatomical characters are modified, may come to have very different functions. The exclusive consideration of their physio- b 2 4 COMPARATIVE ANATOMY. logical duties may thus bring organs which are morphologically connected into very different categories. Thence results the sub- ordinate importance which we must assign to the physiological duties of an organ when we are engaged in an investigation in Comparative Anatomy. Physiological value is only to be regarded at all, and then in the second place only, when we are trying to make out the relations to the entire organism of those modifica- tions which an organ may have undergone as compared with some other condition of the organ. By this examination of anatomical facts, by means of com- parison, Comparative Anatomy demonstrates the connection of entire series of organs. Within these series we find changes of the most varied range, sometimes slightly, sometimes widely extended ; modifications which affect the size, number, form, and even the texture, of the parts of an organ, and which may even lead to changes in its situation. The review of such a series teaches us then to recognise a progress presented in those several successive stages, which the modifications of one and the same organ in different animals exhibit to us. § 4. We ascribe the existence of a certain amount of similarity in the organisation of certain larger or smaller divisions of the animal kingdom to Transmission — a phsenomenon which is exhibited in the passing on of its organisation by a given organism to its posterity. The descendants repeat the organisation of the parental organisms. This is an indubitable fact. Nevertheless now and again objections are raised either to the existence of Transmission or to its signifi- cance. The similarity of the organisation of the descendants and their ancestors is then ascribed, not to Transmission, but to certain physical forces acting during embryonic life. In reply, we may ask, how does it happen that in ancestor and descendant these forces are the same — viz. all those forces of tension, of pressure, and so on, from which it is sought to deduce the building up of the embryo ? If, for example, a joint gets its ontogenetic development by the movement of the parts of the skeleton by means of muscular activity, a certain arrangement of the muscles is presupposed, and a perfectly definite structure of the muscles; and for these again, a perfectly definite number and arrangement of the morphological elements which make up a muscle. This being so we must ask, whence comes the definite arrangement of these parts ? whence arises the similarity of arrangements in the ancestors and the descendants ? We find, in fact, that we must give full recognition to the exist- INTRODUCTION. 5 ence of the transmission of properties, and recognise in it a phenomenon of general prevalence, which may indeed present modifications of, but never exceptions to, certain definite laws. We may deduce it from the conditions involved in propagation, and thus explain it to a certain degree; for it is clear that portions of an organism, if they give rise to a new organism, will carry on to it the peculiarities which the primitive organism possessed. This is clearest in the lower organisms, which are propagated by mere division. Each portion forms at once an organism like the first.- But from this there extends a continuous series of methods of pro- pagation, up to those in which generative products come into action, which are quantitatively very different, although in all cases derived from the division of the parent organism. The new organism in this case also represents in actual sub- stance the continuation of the ancestral, and will therefore possess qualities which agree with those of the latter. The amount of similarity or agreement in the organisation of animals is very various. We recognise animals which differ from one another by slight points only ; then those which are separated by considerable differences ; and again others which, in external or internal organisation, present the greatest differences. Thus agreement, as well as variation, is found in interminable gradations. We call things which are more or less like to one anothei', " related ; '; and in like manner, when organisms exhibit likeness, we use that word to denote the reciprocal connection, but in this case we give to it its full meaning of blood-relationship. We recognise similar organisms as related to one another, when we can explain the similarity of the organisation by common inheritance. But the degree of this similarity measures the degree of relationship which we can deduce from it. Relationship can be regarded as close when the differences are slight ; while when the differences are great it must be regarded as more distant. We thus substitute for the conception of the agreement, or likeness, of the organisation, that of relationship, for we regard the agreements which obtain in the organisation of a collection of organisms as inherited peculiarities. The doctrine of the Blood-relationship of Organisms orPhy- logeny is based on the law of inheritance. Comparative anatomy thus reveals the relations of affinity within the various divisions of the Animal Kingdom by pointing out what is alike and what unlike. [A full account of this most important law of inheritance and its pheenomena is to be found in Hackei/s luminous essay on the subject (Grenerelle Morphologie, vol. ii. p. 170).] 6 COMPAEATIVE ANATOMY. § 5. By means of inheritance, characters are passed on to the organism, which are afterwards matured in the course of its indi- vidual development (Ontogeny). There is no such development in the simplest forms, inasmuch as the young, which arise by division of the maternal organism, only need increase in size to make them like the maternal organism. In this case, development is the same thing as growth, which is completely coextensive with it. The farther an organism is from a primitively simple condition, or the greater the sum of characters which have been inherited from its ancestors and transmitted to their descendants, the less simple is its ontogeny ; for during it a part at least of the characters which have been inherited from its ancestors are repeated, and are presented by the developing body in several successive stages. Ontogeny thus represents, to a certain degree, palaaontological development, ab- breviated or epitomised. The stages which are passed through by higher organisms in their ontogeny, correspond to stages which are maintained in others as the definitive organisation. These embryonic stages may accordingly be explained by comparing them with the mature stages of lower organisms, since we regard them as forms inherited from ancestoi'S belonging to such lower stages. Regarded from this point of view, many of the so-called larval- stages, with their " provisional organs" — so named because they are transitory, and limited to the earlier stages of life only — are seen to be forms of great importance, and full of significance. Such organs, besides having physiological relations to the organism which possesses them, in consequence of which they are preserved as practical arrangements, and become heritable, can be recog- nised in lower grades of the existing series of animal forms, and thus reveal the phylogeny of the animal that possesses them. The "stadium larvatum" then, notwithstanding its name, often points out with great clearness the blood-relationship of an organism. At times these larval organs are not so well explained by transmission as by adaptation, and thus the estimation of their true meaning is made more difficult. The significance of these arrangements is more obvious in organisms which do not enter immediately into the " struggle for existence " in the external world, but are developed for a certain time within the coverings of the ovum, and so are less exposed to the moulding influences of the outer world. In these cases they are " provisional " arrangements, and may be with greater . certainty regarded as having been transmitted, and consequently as repetitions of lower stages. The branchial clefts which appear in INTRODUCTION. 7 the embryos of the higher Vertebrata, but by-and-by disappear, are structures of this kind. Eegarded alone they are inexplicable, for they neither lead to the formation of gills at any time, nor arc they converted (with the exception of the anterior) into definitive organs of any other kind. Comparison shows us, however, that in a large division of lower Vertebrata these branchial clefts are important organs of respiration; and as we also know Yertebrata (Amphibia), in which the clefts function only for a time as respiratory organs, and close up later on, we are able to comprehend the branchial clefts of reptiles, birds, and mammals, as arrangements obtained by transmission from lower stages, which have lost their primitive function, and remain for a short time only — during foetal life. § 6. In the sum of the characters of the organisation, which inherit- ance passes on to an organism, we find, in consequence of what has been already pointed out, a greater or smaller number of arrange- ments, which pass on into the permanent adult stage of the organism without having any recognisable function in it. Such parts are, as a rule, seeu in a more or less atrophied and rudimentary condition, which they often do not acquire until the ontogeny has run its course. In the early stages of the ontogeny they generally agree in completeness with the other arrangements which obtained in the ancestral form from which they are derived. These rudi- mentary organs commence to atrophy the earlier in proportion as they were inherited earlier, in apaleeontological sense ; and, as a rule, disappear late when their origin is a relatively late one. The fully- developed form of the rudimentary organs is consequently to be found, in the former case, in widely separated species ; in the latter, on the other hand, in species more closely allied. These organs are valuable objects, since phylogenetic relations can be very generally recognised by their aid. They show, too, how little physiological significance ought to be regarded in a morphological discussion ; for in most of them a function is not to be made out at all, or, if it can be made out, is found to be quite different to the primitive one. § 7. Comparative Anatomy forms part of Ontogeny, inasmuch as it treats of the phenomena of the organisation which appear in the course of the individual development of the animal; not only in relation to the complete stage of the organism, but in relation also to the definitive arrangements of other organisms. Comparative 8 COMPARATIVE ANATOMY. Anatomy explains the phsenomena of Ontogeny. Ontogeny by itself does not rise above the level of a descriptive discipline, and in proportion to the exactness of its investigation possesses a value as so much objective material. At the same time Ontogeny gains scientific importance by its connection with Comparative Anatomy. Its facts, which by themselves are incomprehensible, or are only teleologically explicable in a metaphysical sense, because restricted to the later events in the history of an organisation, are, by Comparative Anatomy, put in connection with the known phenomena of other organisms, and are thereby rendered explicable phylogenetically. The necessity of an exact knowledge of Compara- tive Anatomy for Ontogeny is sufficiently obvious. Just as little can the former separate itself from the latter : since from Ontogeny Comparative Anatomy gains an insight into the lower stages of organisation. To the same extent, and in the same way as On- togeny helps to form the basis of Phylogeny, does it render indispensable service to Comparative Anatomy. A " Comparative Embryology " has sometimes been put in contrast with Comparative Anatomy ; of course merely as a theo- retical division of the scope of study. A "comparative" Ontogeny of this kind must, just as much as every individual ontogeny, have regard to the organisation of the fully- developed stage ; and, in fact, without Comparative Anatomy it cannot lead to any scientific results. § 8- The relations of every organism to the outer world in which it lives, from which it obtains material, and to which again it gives it up, cause the outer world to have an influence on the organism. This influence is practically seen in the changes of the organism, which depend further on a Variability which is inherent in it. Variability comes under our observation as the capacity for adaptation, and in effect operates as a modifying and even metamor- phosing agent upon the inherited organisation. The organism is altered according to the conditions which in- fluence it. The consequent Adaptations are to be regarded as gradual, but steadily progressive, changes in the organisation, which are striven after during the individual life of the organism, pre- served by transmission in a series of generations, and further deve- loped by means of natural selection. What has been gained by the ancestor becomes the heritage of the descendant. Adaptation and Transmission are thus alternately effective, the former representing the modifying, the latter the conservative principle. The endless variation of the phenomena of organisation is, We see, consequently INTRODUCTION. 9 due to Adaptation, just as we have seen that similarity is due to Transmission. § 9. Adaptation is commenced by a change in the function of organs, so that the physiological relations of organs play the most important part in it. Since adaptation is merely the material expression of this change of function, the modification of the function as much as its expression is to be regarded as a gradual process. As a rule, therefore, Adaptation can be perceived by its results only in a long series of generations, while transmission can be recognised in every generation. Although Adaptation as a process cannot be directly observed, it is nevertheless possible to infer it with certainty by comparison. When, for example, we find a simple structure of the stomach in the Carnivorous Mammalia, and, on the other hand, a more complicated one in the Herbivora, and especially in those which take in large quantities of food, as, for example, the Rumi- nantia, we are entitled to consider the complication in the structure of these stomachs as a change caused by the food — as an Adaptation to the mode of nutrition ; and when, further, the ontogeny of the Ruminantia shows us a form of stomach which is simple in the early stages of development, and is gradually converted into the more complicated condition, ontogeny confirms the supposition we have already gained by comparison. In many . cases the influence of Adaptation on the organisation can be observed also directly ; as, for example, in many Amphibia, where the branchiae which are de- veloped during the larval stage are retained in function for an extended period, if the opportunity of escaping from the water does not arrive ; and in others, again, where the branchiae undergo atrophy, as soon as their aquatic life has been exchanged for one on land, although their nearest allies live in the water, and always retain the branchiae. In the one case we see development, and in the other atrophy, as phaenomena resulting from Adaptation. In Adaptation, the closest connection between the function and the structure of an organ is thus indicated. Physiological functions govern, in a certain sense, structure ; and so far what is morpho- logical is subordinated to what is physiological. This dependence of the form of an organ on its activity is seen in the most elementary way in the matter of size. When the function is increased, there is an increase in the size of the organ. The muscular system shows to what extent increase of activity affects size. Without exercise the muscles undergo degeneration, till they completely disappear. If they are kept in exercise, and if the demands on them are increased, they develop to a considerable size. The amount of 10 COMPARATIVE ANATOMY. development is in the closest connection with the amount of activity. But since, when a function ceases or diminishes, atrophy commences, we obtain, as a result of the process, rudimentary organs. They owe their origin to atrophy. Physiology alone, theu, can give us the explanation of the origin of these organs ; and thus again we are led to observe the great influence which it exerts on the study of Morphology. § 10. An organ can be so much changed by the gradual modification of its function that it becomes, from the physiological point of view, anew one, and is then placed in quite another physiological category of organs. This fact is of considerable importance, for it helps to explain the appearance of new organs, and obviates the difficulty raised by the doctrine of evolution — viz. that a new organ cannot at once appear with its function completely developed ; that it there- fore cannot serve the organism in its first stages whilst it is gradually appearing ; and that consequently the cause for its development can never come into operation. Every organ for which this objection has the appearance of justice can be shown to have made its first ap- pearance with a significance differing from its later function. Thus, for example, the lungs of the Vertebrata did not arise simply as a respiratory organ, but had a predecessor among fishes breathing by gills, in the swim-bladder, which at first had no relation to respiration. Even where the lungs first assume the functions of a respiratory organ (Dipnoi, many Amphibia) they are not the sole organ of the kind, but share this function with the gills. The organ is there- fore here caught, as it were, in the stage of conversion into a respiratory organ, and connects the exclusively respiratory lungs with the swim-bladder, which arose as an outgrowth of the enteric tube and was adapted to a hydrostatic function. The earlier function of an organ which by adaptation is converted to new uses is generally a lower one, and less important for the organism, in comparison with the new function which is taken on, so that the organ thus rises to a higher grade. In other cases the value of the primary function is less, because it is shared by other similar organs. It is then quantitatively lower, for a share is taken by the other similar organs in discharging the total amount of the function necessary for the full activity of the organism. The atrophy of some of the organs which are of equal value raises the value of those that remain by causing their higher development. To these facts, as to their change of functions, the difference in the classification of organs, accordingly as we make use of a physio- logical or a morphological method, is due. GENERAL PART. THE ORGANISM. 13 Structure of the Animal Body. The Organs and the Organism. § 11. In the living body we observe a number of activities of its material substratum, by which the series of phsenomena spoken of as life are conditioned. Underlying these, there are chemico-physical processes, which are accompanied by a continual degradation of the material, and consequent metastasis, or change in the arrangement of chemical elements. The body nourishes itself by replacing the material used up in metastasis by fresh matter, which is received from without; and this it assimilates, or makes like to the substances of which it is itself composed. The substances, partly taken in with the nutrient matter, partly produced by metastasis, which are of no further use in the organism, are passed out. Hence results the excretory function. If the quantity of matter assimilated is greater than that which is expelled, there is an increase in the size of the body: it grows. Thus it fulfils the first condition for the production of that material from which a new organism, like to itself, arises : and so reproduction is closely connected with nutrition. The body is, in the first place, in relation to the exterior by its surface ; this puts it in connection with the surrounding medium. Changes in the form of the surface of the body result in movements, and give rise to locomotion. The surface also is the medium by which it perceives the outer world, or has sensations. The parts of the body which preside over these processes are the instruments by which life is carried on — organs. In virtue of their existence the body is an organism, and when we also include under the term " organisms " certain bodies in which no organs can be individually separated, it is because the virtual existence of organs in them is to be assumed from the mere fact that life is carried on in them. The term ' ( organism " is therefore employed in this instance, not in an anatomical, but in a physiological sense. In the simplest condition of the organism, the vital phasnomena take place in the homogeneous substance which forms the body, and which is the seat of all these processes equally. The body 14 COMPARATIVE ANATOMY. in this case represents, potentially, a collection of organs, which only appear, in fact, when the different functions are no longer performed by every part of the body. The condition which the lower organisms permanently exhibit in regard to this matter is possessed for a short time only by the more complicated. Differentiation. § 12. The complication of the organism arises from a process of division which transfers to separate parts the physiological acti- vities of the primitively homogeneous body. What was previously accomplished by the whole body, is, subsequently to that process, carried out by particular portions. The work is then done either by a laro-e number of parts, which are distinct from but similar to one another, or the separate parts acquire dissimilar shapes, and become different from one another. In the first case the division of labour is quantitative, in the latter it is qualitative, and the separation of the different parts is in correspondence with a difference in func- tion. According to the degree in which the separation or division originally set up in the primitively indifferent body is repeated in the organs derived from it, further complications arise, which present for our observation a step-by-step progression in development. Hence there arises a difference in the value of the organs, and it becomes necessary to distinguish their higher and lower conditions. The distribution of work amongst a number of different organs leads to the perfecting of the operations of such organs. Each organ is enabled to develop in a definite direction, in correspond- ence with the particular function which is undertaken by it. The organism thus becomes more highly developed, as well as com- plicated. Division of labour leads to a perfecting of the whole organism. According as the division of labour involves only a few or many organs, a greater or less part of the organism is brought under the operation of this perfecting influence. The greater the importance, for the whole organism, of the organs affected, the more considerable is the perfecting accomplished in it through their modification. The functions which attach them- selves to definite parts of the body bring about a difference in the development of those parts proportionate to their own difference ; and thus it is that new parts and new organs arise, which are different from those already existing. The division of functions leads to the establishment of a difference, that is to a differentiation, of the parts. A part of the body which was formerly like the rest, and con- sequently not different from it — that is was indifferent — passes into the condition of being separate, becomes distinguishable, or different from the rest. And as this differentiation is connected with the division of labour, inasmuch as it is conditioned by it, it may be regarded as the product of it. Every physiological function can be again divided qualitatively into various sub-functions, by the locali- THE CELL. 15 sation of which fresh organs again arise. Thus the principle of the division of labour is the cause of very great variation in the organi- sation, and all morphological phenomena are more or less closely connected with it, and with the differentiation which is due to it. First Stage of the Animal Organism. The Cell. § 13. Living matter appears in its simplest form as an albuminous sub- stance, known as Plasma, or Protoplasm, which, by the aid of our present optical instruments, seems to be homogeneous throughout. This substance occurs in the form of small lumps, in which condition we find the simplest organisms. In those simplest forms, where the protoplasm is homogeneous, and in which only a few granules at most are present as heterogeneous elements, there is no limitation of the lump to the exterior by distinct enveloping structures ; but in organisms of little higher grade we find an envelope produced by a chemico-physical change in the most external layer of the proto- plasm. Thus the protoplasm, which is endowed with all the phamomena of life, and even of movement, is enclosed by a more or less firm envelope, which forbids alterations in form, and is the cause of a definite shape being maintained. Such structures may be combined to form complex organisms, as is the case in many of the lower plants. This kind of form-element, or morphological unit, is known as the cy tod, and is rightly distinguished from another more highly-developed form. In this higher form there arises in the protoplasm a sharply marked- off dense structure, which is called the nucleus. It is the product of the first process of differentiation of the protoplasm, which no longer alone represents the living substance. In the nucleus a small body, the nucleolus, generally appears. The nucleus, unlike the protoplasm, is not contractile, or at any rate has not a large share of contractility; but it not only takes a part in most of the vital phaeno- mena of the surrounding protoplasm, but frequently gives evidence of being their regulator. Such corpuscles of protoplasm as are pro- vided with a nucleus are called cells (cellulee). These structures, like the cytods, may form independent organisms, which are then called " unicellular/" When the cells form a complex by multi- plication, we have a multicellular organism. The smallest parts of multicellular organisms, no longer separable into constituent pieces like to one another, are cells; and are therefore the form-elements of these organisms. The same remark applies to the cytods, or the simpler condition. While these, however, are rarely present, we find cells widely distributed in the Vegetable Kingdom, and as the sole form-elements in the Animal Kingdom. 16 COMPABATIVE ANATOMY. § 14. In the indifferent condition, that is as long as changes in a definite direction do not arise, leading to the formation of definite new structures, the cells of all animal organisms appear to have essentially the same character. In them we distinguish according to what has been remarked above : first, the protoplasm, which forms the principal mass of the body of the cell ; and secondly, the cell-nucleus, surrounded by, and different from, the protoplasm, than which it is usually more dense. The share which this nucleus takes in many varied pliEenomena of the life of the cell, compels us to regard it as by no means a subordinate part of the cell-body. In addition to these parts of the cell, some persons recognise (formerly everyone did so) a membrane which differs from the protoplasm or contents of the cell, and envelopes it : from its presence arose the notion of the " vesicular form " of the cell and its name. Although it cannot be denied that in many cells there are envelopes differentiated from the protoplasm, yet this condition is never found in the earliest life of the cell, but is always the re- sult of an advanced change, and of the passage of the cell into a differentiated form. Automatic movements of the protoplasm of the cell are such common manifestations of their life, that they are always definitely apparent as a property of all cells which are not highly differentiated, that is of cells in which the protoplasm is not meta- morphosed. In free cells, and such as are not enveloped by firm membranes, this phenomenon of movement produces locomotion. Even in cells that are not free movement may be observed, consisting partly in a change in the form of the surface, partly in a change in the position of the granules in the protoplasm. That there are also properties resident in protoplasm which we may attribute to sensibility of a very low grade results from many experiments and observations. We may further observe nutrition in the cell. At times, indeed, we can see the protoplasm ingesting food; in all cases the growth of the cell is an express indication of its nutrition. This pheno- menon of growth, which may be seen in any indifferent cell, is expressed by the increase in size of the protoplasmic body, owing to the assimilation of matter from without. The growth of the cell may be quite regular, increase in size obtaining in all directions, as is the case with all cells in their earliest stages ; as long as this lasts the cell completely retains its spherical form, unless its move- ments or external influences modify it. Or growth may be unequal, and the cell become elongated by increase in size along one axis ; or, again, it may become stellate by growing along several axes. Irregular increase of this kind is ordinarily accompanied by differen- tiation of the cell-substance, and is therefore the commencement of the conversion of the cells into tissue. THE CELL. 17 §15. Another phenomenon — that of reproduction — is a result of, and is indissolubly connected with, the growth of the cell ; for multi- plication is merely the extension of growth beyond the limits of indi- vidual cohesion. There are various modes of cell-multiplication; the simplest of these is a direct result of growth. A bud is formed by the cell-body growing out on one side. This gradually increases in size,, and breaks off from the mother-cell, when it becomes a new free cell. The number of young cells which are budded off from a single cell is not always the same, also the part taken by the nucleus of the mother-cell in the process, varies. This mode of multiplication by gemmation passes imperceptibly into the more common mode of multiplication by fission. Gemmation is charac- terised by the difference in size which obtains at first between the cells that are formed and their mother-cell. If they break off at once they do not nearly equal it in size ; if they delay their separation from the mother-cell they gradually get to equal it, and then the products of division are almost or altogether equal to one another, so that there is no possibility of distinguishing mother from daughter. It is evident that in proportion to the extent to which the products of division differ from one another in size, does division become more and more like gemmation ; the whole difference therefore between fission and gemmation lies in the amount of protoplasm which is given over by the parent-cell to the one which arises from it. The difference is a quantitative one merely. Division commences by an enlargement of the nucleus ; in some cases by a formation of fresh nuclei. No form of reproduction other than multiplication by fission or by gemmation has been certainly observed in the animal-cell; a large number of the modes of cell-multiplication, which have been stated to obtain, such as the so-called endogenous cell-formation and similar processes, are merely forms of fission. As to free or spon- taneous cell-formation, so much at least is certain, that it is not as common as was once supposed. When the nucleus divides and the cell goes on growing without the protoplasm becoming mai'ked off into separate portions corre- sponding to the nuclei, the structure which is formed cannot be any longer regarded as a single cell. But it is not a compound of cells either, for this would presuppose the existence of a number of separate cells. This condition has therefore been very rightly regarded as a special one, and called a Syncytium. Structures of this kind are found in nearly all groups of animals. The same result is obtained by the Concrescence of a number of separate cells, the protoplasm of which runs together into a continuous mass, in which there are of course a number of nuclei. While the protoplasm in the above-mentioned series of pheno- mena undergoes no perceptible changes in constitution, a change c 18 COMPAEATIVE ANATOMY. in the protoplasm is essential to another kind of phenomena, to which we now pass. The substances contained in the protoplasm become separated from it, that is, are secreted from it. This process of secretion varies in character; it sometimes occurs within the protoplasmic body, and then substances differing in their chemico- physical properties from the protoplasm are formed within the cell. These substances may be distinguished by their constitution, such as fat, colouring matter, and so on; or by form, such as granules, drops, crystals. Sometimes secretion affects the surface of the protoplasm; and then the secreted substance may be fluid, in which case it will get separated and removed from the protoplasm; or it may be solid, in which case it will remain more or less intimately connected to the rest of the unaltered protoplasm. Substances which are different from the rest of the protoplasm of a cell ^arise by chemico- physical changes of the whole surface, or of a part of it. We may regard these changes in the protoplasm as differentiations, for they are differentiated from matter which was previously in an indifferent condition within the protoplasm. In this way the structure which has been already alluded to as the cell-membrane, is formed at the periphery of the cell. But this same process leads to other arrangements also, which we shall have to examine more closely hereafter. The series of vital processes exhibited by a single cell essentially agrees with those exhibited by any and every other organism, so that the cell itself is virtually an organism (Elementary organism). Differentiation of the Animal Organism. § 16. The simplest and lowest stage of the animal organism is repre- sented by the earliest stage of its development, in which it is known as the egg. Except in some exceptional cases, which only prove the rule, and which need not be mentioned here, this egg is nothing more nor less than a cell. The egg-cell does not differ from other cells in any essential points; it may increase in size, and special particles — yolk granules — may appear in its protoplasm. Although the presence of these latter alters the character of egg-cells as indifferent cells, yet it does not destroy their character as cells, which is just as little affected by this as in Fig. l. Diagram of other cells by the differentiation within their an (W aGraimlar bodieg of substances, SUch as chlorophyll protoplasm. o .Nucleus , , . . J i o m\ • (Germinal vesicle). cNu- granules, starch, pigment granules, &c. Ihis cleolus (Germinal spot), condition of the egg-cell, which on the whole DIFFERENTIATION OF THE ORGANISM. 19 is a simple one, agrees in character, although it may be for a time only, with many lower unicellular organisms (Protoplasta). The egg-cell undergoes changes, which ordinarily commence after impregnation, and which are accompanied by changes in the nucleus (the so-called germinal vesicle). In its place, and in part from the material which formed it, two new nuclei arise, and Figs. 2-5. Various stages of the so-called cleavage process (Division of the egg). the egg-cell now begins to divide. Two cells thus arise, which are either like one another, or differ from one another, in size or in constitution. In both cases something fresh has arisen from the egg-cell, and in both there is a differentiation, for two parts have arisen from it. Four, eight, sixteen cells, and so on, are formed by continued division, although of course not always quite regularly, until at last a number of cells are formed. This process of the division of the egg-cell is known as the "segmentation of the yolk," and is a constant phssnomenon, although it may present various modifications, which are always due to adaptation, and which may be so explained. This is the first course of differentiation in the organism ; in place of a single cell, a number of cells, similar to, or different from one another, arise. The functions of the organism, which were all performed previously by the egg-cell, are now performed by the separate cells. The division of the egg-cell must therefore be considered as leading to a division of its functions, although indeed this division is merely a quantitative one. The various stages of this process of division have other relations also, for they appear to agree in character with the mature stage of many lower organisms (Protista), as for example the Volvo cineae and Catallacta, in the developmental history of which there is at one time an organism composed of a number of equi-formal cells. The animal organism, therefore, even in the commencement of its ontogeny, passes through several morphological stages, which are permanent among the Protista, and the process of segmentation of the ovum may be explained as a sur- vival transmitted from early ancestors. Accordingly the teleological halo, with which it would necessarily be sur- rounded, were we limited to seeking its explanation exclusively in connection with the future organism which is to arise from this segmentation, is cleared away. The organism does not, how- ever, get a specifically animal character from this formation of 20 COMPARATIVE ANATOMY. a number of cells ; that character first makes its appearance in the course of further processes of differentiation. These processes of differentiation consist in the more or less similar morphological elements (cells) which represent the organism, acquiring, in larger or smaller groups, distinct characters : in their being differentiated, and forming the rudiments (first stages) of organs, by taking a definite order and arrangement. These organs then are made up of cells, which form their tissues. We thus arrive at the essence of the architecture of organisms ; we have tissues, which make up organs, and are themselves composed of form-elements — the cells. Origin of the Tissues. § 17. The cell, then, in those organisms which we regard as animals, constitutes the whole of the organism only for a time ; that is, so long as it is an egg-cell. By division a multitude of cells is formed out of the egg-cell, and these form the rudiments of the animal. In later stages a part only of the material formed from the ovum retains the primitive character of the cells ; the form and substance of most of the cells are altered, and therefore their physiological properties are altered ; that is, new relations are established. The new cell-complexes formed from aggregates of similarly altered cells, and their derivatives, are the tissues. The process which leads to the formation of these tissues is essentially a differentiation. This, again, affords us an example of division of labour, for each differentiated aggregate of cells has to perform a definite function for the organism, which function was not the duty of a definite set of cells when the cells were indifferent, and indeed were performed in common with all others by one cell only (the egg-cell), in the earliest condition of the individual organism. In all cases histological differentiation commences in the proto- plasm of the primitive cell ; the nucleus is less strikingly affected, but numerous changes may be seen to occur in it. When the chief part is played by a substance differentiated from the protoplasm, the nucleus becomes of but slight importance. According to the characters of their form-elemeuts the tissues are divided into several large groups: these I call Epithelial tissues, tissues of the Connective Substance, Muscular and Nervous tissue. The first two form a lower group, which, as Vegetative tissues we may distinguish from the other two, which are the Animal tissues. The difference between the two groups lies in the quality of their differentiation ; the products of the differentiation of the former having a more passive relation to the organism, while the products of the differentiation of the latter exhibit an independent activity in the carrying on of the life of the organism. The vegetative group, or tissues analogous to them, are, moreover, most widely distributed in THE TISSUES. 21 the Vegetable Kingdom ; whilst in that kingdom the animal tissues, which are the source of the arrangements characteristic of animals, are wanting. All other tissues, though often distinguished by name, are either not independent tissues at all, but only much more complicated structures formed of a variety of tissues, or are forms of tissue which should be ranged under one of the above-mentioned categories, or may be merely component parts of the tissues already named. We overlook the true conception of what a tissue is if we call structures which are made up of several tissues " compound tissues ' )> a. Vegetative Tissues. Epithelium. § 18- Cells placed side by side, and forming one or more layers which invest the surface of the body or the walls of internal spaces, are called epithelial. Epithelial tissue, then, consists simply of cells. It is distinguished from other tissues by the fact that the cells, at least so far as their arrangement is concerned, retain their primitive characters. Epithelium represents phylogenetically, and therefore, also, ontogenetically, the oldest form of tissue. The germinal layers which are the earliest organological products of the differ- entiation of the masses of cells which arise from the segmentation of the egg-cell, are layers of epithelial cells. Epithelial cells vary greatly in form, and are the starting-point of various organs. The protoplasm of epithelial cells very often loses its homogeneous character, owing to the differen- tiation of its outermost layer into a thickened membrane. In stratified epithelium this is best seen in the superficial layers, the absence of a membrane in the cells of the deeper layers being an indication of their younger con- dition. Another differentiation is the development by the superficial layer of cells on the surface exposed to the ex- terior, or lining an internal cavity of the body, of fine pro- cesses, which are capable of movement ; these processes, which vibrate during the life of the cell, are known as cilia. The hairs on these ciliated cells are sometimes in the form of a single flag ell urn, or occur in a group Fwfuate' of many as cilia. In the former case the cell runs out cells, a of into a fine process, and forms a flagellate cell; these a Hydroid are most common in the lower animals. Cilia are shown a° ■£? * ° to be differentiated products, since their movements are (collar not simply effected by the contractility which is already cell), inherent in the protoplasm. In many of the lower or- ganisms cilia are formed for a time and are again drawn in and their substance fused with the protoplasm. This shows that they 22 COMPARATIVE ANATOMY. are differentiations of the protoplasm, and that their movements are due to the same cause as the movements of the protoplasm. This indication of their identity with protoplasm cannot be seen in the more differentiated forms of cilia. Another differentiation may be seen on the outer surfaces of many epithelial cells. A membrane, instead of being formed by a change of the whole periphery of the superficial layer of protoplasm, may be formed on a definite portion of it only : in this case it is more highly developed, and may lead to a partial thickening of the outer- most layer of protoplasm. In short, a layer of varying thickness of a substance different from, but as a rule still connected with, the protoplasm, forms on the outer face of each cell. Homogeneous membranes — cuticles — are formed by the further differentiation of the substance thus secreted in a layer from the protoplasm of the cells ; that is by the part formed from each cell becoming more intimately connected with the layers formed by the cells around it than with its own cell. Where these layers are laid down irregularly and gradually undergo other changes, by means of which each fresh addition can be distinguished from the preceding one, they become laminated. The more the substance of which these cuticular struc- tures are composed differs from the protoplasm of the cells which have deposited it, the more difficult is it to make out any passage of the protoplasm into it, and the more distinctly is the formation of cuticles seen to be a process of secretion. When the cuticle is not formed regularly on the surface of the separate cells, protoplasmic processes project from the secreting cell-layer into the secreted layer, which are then traversed by corresponding canals (pore- canals) : these are ordinarily very fine. These cuticles differ greatly in con- sistency, and present every intermediate step between softness and extreme hardness. They are often converted into organs of support, when they are very firm j in which case they ordinarily consist of a substance known as "ckitin." These chitinised cuticles are very common in the Invertebrata. §19. The secreting activity of the cells of large epithelial layers may give rise to liquid, or even to gaseous bodies. The epithelia there- upon enter into new relations to the economy of the organism; they no longer produce substances destined to build up the organism, but they present an intermediate step towards that condition of epithelial structures in which parts of the epithelium enter into the formation of a tissue of definite function — glandular tissue. As there is always a direct connection between the aggregation of cells which form the secreting organs, or glands, and the epithelium, which either persists permanently, as in the majority of glands, or which is at any rate present when they are first formed, this glandular tissue is seen to be nothing more than a modifi- THE TISSUES. 23 cation of the epithelial tissue, due to its differentiation. Like it, glandular tissue always consists of cells. In the simplest stage individual cells in a layer of epithelium become secreting-cells, and function as gland-cells, by forming and secreting a substance such as is not produced by the other cells. In this way uni- cellular glands arise. They either retain their original position between the other cells of the layer or sink beneath the level of the epithelium, and open between the other cells by a fine duct formed by the membrane of the cell (Fig. 7). If the secreting surface be increased without the general epithelium taking any share in it, the sunken epithelium must increase in size, and so give rise to structures which are more or less separated off from the epithelium, such as pits, sacculi, or ceecal tubes ; and these may be again complicated by fresh growths. The tissue lying beneath the primitive epithelial layer forms en- velopes for these pits as they grow; but it con- tinues to have the same relation to them, however complicated in form the ramifications and similar proliferations of the epithelium may be, as it pre- viously had to the simple even layer of epithelium. Thus the gland in its simplest form appears as a depression of the epithelium into the sub- jacent tissue. In the more distinct forms of glands there is a further differentiation of the cells which form the gland. The constituent cells of the gland become separated into those which secrete and so represent true gland-cells, and those which connect the secreting portion of the gland with the still indifferent epithelial layer. These, in contradistinction to the secreting portion of the gland, form the epithelium of the duct. Fig. 7. Unicellular glands. Anterior salivary glands of the ant (after Stein). Connective Substances. § 20. The phaenomenon which in epithelial tissue leads to the formation of homogeneous membranes may, by being extended over the whole periphery of every cell as well as by continued repetition, become of greater importance. Even in the epithelial tissue we often meet with a fine intermediate layer, the cement-substance. As the sub- stance which is differentiated from the protoplasm of a number of cells gradually increases between the cells containing unaltered protoplasm, the cells become separated from one another, and a distinction is made between the cells which form and the inter- cellular substance which is formed. A number of very different 24 COMPAKATIYE ANATOMY. tissues present this common character in their more intimate structure. They are called connective substances, as the majority of these tissues serve to unite other tissues to organs, or systems of organs. The differences in these tissues are due partly to the character of their cells, partly to their relations with the intercellular sub- stance, and partly to the chemico-physical constitution of the intercellular substance; but all these points are not equally well marked in every part of them. Whilst this latter circumstance allows us to recognise the passage of one of these tissues into the others, the fact that such passages do periodically take place under our observation, affords a more weighty reason for uniting them than the fact that we can detect common characters in then* structure, although often hidden by various differences. The various tissues which belong to this group are : 1) cellular con- nective tissue, 2) gelatinous tissue, 3) fibrous connective tissue, 4) cartilaginous tissue, 5) osseous tissue. § 21. Connective tissue is divided into the following varieties: 1) Cellular connective tissue (vesicular connective tissue) is the simplest form. It is formed of rounded or elongated cells, which are separated by a small quantity of intercellular sub- stance only. There are often vacuolated spaces in the cells, which are filled with a fluid. The intercellular substance often has the form of cell membranes, which serve to unite the juxtaposed cells to one another, and are common to neighbouring cells. In other cases again it is more largely present, without prepon- derating in quantity over the cells. The differentiation of the protoplasm of the intercellular substance varies in degree. This tissue is most widely found in the Arthropoda and Mollusca. In the Vertebrata it forms the chorda dorsalis, or notochord. 2) Gelatinous tissue (mucous tissue) is distinguished by the soft gelatinous character of the intercellular substance • it is ordinarily hyaline, and in it Fig. 8. From the gelatinous substance of the disc of Aurelia aurita, treated ■with iodised serum (after M. Schultze). x 500. a Branched fibres in which no cells can be made out. b Cells in the homogeneous gelatinous substance : the processes are largely retracted in this specimen. THE TISSUES. 25 are placed either rounded and completely separated or filiform and branched cells, which are united to one another by their processes. Chords or tracts of cells also occur. In this way a fine network is formed, which traverses the gelatinous portion of the structure, the trabecular of which may become firmer by further differentiation, and may become broken up into fibrillar. A similar fibrillation may affect the intercellular substance, in which case fibrous bands, in which there are no cells, can be made out. This tissue is found in many of the lower animals; in the umbrella of the Medusas (Fig. 8), the integument of the Heteropoda, &c. 3) Fibrous connective tissue may be regarded as a further development of gelatinous tissue. Its morphological elements are elongated or branched cells, embedded in an intercellular substance formed of fibrous tracts and bundles. This substance is largely due to a differentiation of the walls of the cells, as is clear from its development. Development also reveals the fact that part of the protoplasm which sends off processes, is directly differentiated into fibrils and fibrous bundles ; these are therefore distinct from the earlier formed, and more or less homogeneous intercellular substance. The thickness of the fibres and the direction they take vary greatly. The fibres, which are generally curved and undu- lating, sometimes run parallel to one another, sometimes anastomose; the cells and the cell-processes are, in their earlier stages, arranged in a manner corresponding to the subsequent arrangement of the fibres. Fibrous connective tissue is distinguished as loose, or firm, according to the characters of its intercellular substance ; the firmer sort is also known as " tendinous tissue/' the fibrous bands of which are placed parallel to one another. In addition to the fibrilla?, which swell up when treated with acids and alkalies, there is another form of fibre, which is seen in the intercellular substance of fibrous connective tissue; this resists these agents more completely, and is called "elastic tissue," on account of its elasticity. It is, as may be seen from its relation to the inter- cellular substance, not an independent form of tissue, but merely a modification of connective tissue. Inasmuch as a portion of the intercellular substance arises by subsequent differentiation of the protoplasm of the original cells, as was remarked above, the morphological elements which are present in fully-developed connective tissue represent the remains only of the primitive cells. According to the quantity of pro- toplasm used, and converted into fibrous structures, and so in- corporated into the intercellular substance, the nucleus of the connective-tissue cells is surrounded by more or by less protoplasm, or the whole protoplasm disappears ; the presence of isolated nuclei in the fibrous bands of connective tissue is an indication of this. Where the protoplasm still remains around its nucleus — where, that is to say, a cell, according to the conception given above, is present, this cell may undergo fresh changes, which are of so many 26 COMPARATIVE ANATOMY. kinds that connective tissue is richer than any other in the various phenomena of differentiation. § 22. 4) Cartilaginous tissue is characterised by cells lying in a firmer intercellular substance. Its cells do not, except in a few cases, possess distinct processes, or processes which can be easily made out ; but are very nearly circular in form, or else oval or fusi- form. The amount of intercellular substance varies in amount. It is distinguished from those forms of connective tissue which are formed of simple cells placed in a homogeneous intercellular sub- stance, by its greater rigidity. Cartilaginous tissue is well adapted by the possession of this character to function as an organ of support. When the cells predominate, and there is but little intercellular substance, and when what there is is in the form of fine membranes, cartilage is seen to be directly allied to vesicular connective tissue. The protoplasm of these cells often takes on a definite arrangement, and forms bands which extend from the nucleus to the periphery, and unite together there. They are separated from one another by spaces which contain fluid (Fig. 9). In proportion as its intercellular substance is diminished, does this tissue differ more and more from ordinary cartilaginous tissue. In the protoplasm of cells of this kind, which are found forming a sort of skeleton in the Medusas, the phenomenon of streaming of the protoplasm may be seen. If the intercellular substance in- creases, it either remains homogeneous (hyaline cartilage), or it undergoes further differentiations like those of connective tissue ; but these differentiations do not much affect its relations to its cells. When the intercellular substance breaks up into fibres, we get fibrous cartilage; when elastic nets appear in it we get elastic cartilage. By gradual changes of the inter- cellular substance, as well as of the cells, cartilaginous tissue passes into fibrous connective tissue, and thus indicates its close connection with that form of tissue. The cells also become more specially modified in some cases by being elongated or producing radiating processes, which unite with those near them : as, for example, in many Selachii, or more developed still in many Cephalopoda. The intercellular substance then appears to be traversed by the processes from the cells (Fig. 10). The phenomenon, which in the cases just cited is greatly exaggerated, obtains also in ordinary hyaline cartilage, where the cells are apparently sharply marked off from one another ; for the intercellular substance may be seen to be traversed by processes, although these are, of course, extremely fine. The intercellular substance of cartilaginous tissue is always Fig. 9. Cartilage cells from the tentacle of a Medusa (Cunina). THE TISSUES. 27 distinct from the protoplasm of the cartilage cells which lie in its cavities ; but as it is differentiated from the protoplasm, it must, nevertheless, be regarded as a secreted product of the protoplasm — a layer secreted by a cell : often an intercellular substance may be seen in hyaline cartilage surrounding the cell like a capsule. This was formerly regarded as a cell- membrane belonging to the cell. As these " capsules ,; can often be shown to enclose groups of cells consisting of several gene- rations, which have resulted from the fission of a single cell, the enclosed cells were looked upon as mother and daughter cells, &c, and the phenomenon itself was regarded as a case of endogenous cell- formation. As a fact, these " capsular systems " are merely the expression of secretions, not become homogeneous, and formed by several generations of cells which arose from one another. The perfectly gradual passage of cartilaginous tissue in which such capsules may be seen, into tissues where the intercellular substances is completely homogeneous, shows that we have here to do with different stages in the differentiation of one and the same secreted substance, which has arisen in the former case by an interrupted, and in the latter case by a regular, secreting activity of the cells. In virtue of the anastomoses of the processes of cartilage-cells, cartilaginous tissue comes very close to the next form of tissue, and it is only distin- guished from it by the characters of its intercellular substance. Fig. 10. Cartilage from a Cephalopod. a Simple, b Dividing cells, c Canaliculi. d An empty cartilage capsule with its pores, e Transverse section of canaliculi (after M. Fiirbringer). § 23. 5) Osseous tissue. This, the firmest form of the connective substances, consists of an organic intercellular substance combined with lime-salts, in which there are cells with fine anastomosing processes; or it presents a ground-substance like that just men- tioned, in which, however, there are no cells, but only cell-processes. These processes traverse it as fine canaliculi. There are therefore two structural phases of osseous tissue to be distinguished. Cells enter into the composition of the one, but in the other they simply send out fine processes into the pore-canals of its solid ground-substance. The tissue containing bone-cells is the most common; it is 28 COMPAEATIVE ANATOMY. Fig. 11 Rana. bone-cells found in the skeletal organs of all classes of the Vertebrata ; whilst that form of osseous tissue with canaliculi only is found in the skeleton of many fishes, and as a general rule in the dental organs of all Vertebrata (dentine). The development of osseous tissue explains the relations of the intercellular substance to the cells. That form of it which contains cells may arise in two ways : either by the ossification of con- nective tissue, the cells in which become converted into bone-cells by the ossi- fication of the intercellular substance, which becomes impregnated with calca- reous salts, while the cells themselves become con- nected with one another by their processes, which traverse the pore-canals in the intercellular substance; or the same tissue is formed by apparently indifferent cells, which secrete a scle- rogenous substance. This substance is laid down in stratified lamellas, into which the secreting cells send fine protoplasmic processes (Fig. 11, o). The secretion of this substance is preceded by a change of part of the protoplasm of the cell. As soon as this is differentiated it does not belong any longer to the cell, and is therefore secreted from it. If some of the secreting cells (o'o//) cease to be active, while the cells near them do not cease to be so, the former gradually get to lie in a layer of intercellular substance, which finally surrounds them, and so converts them into bone-cells (o'"). The cells of the secreting layer (osteoblasts) are continuously connected by fine processes with those which are already enclosed (bone-cells). Thus each of the former is rendered capable of becoming a bone-cell. The other form of osseous tissue is developed in a perfectly analogous manner, so far as its history is accurately known, through the development of dentine. In this case also a layer of cells secretes a substance, which hardens or is sclerogenous, and into this the cells at the same time send processes, which traverse pore- canals. But the cells (odontoblasts), instead of gradually sinking into this extra-cellular substance, always remain outside it, and are connected with it by their processes only. The secreting substance is thus traversed by fine parallel canaliculi (the dentinal canals, so-called, since they were first made out in the dentine). This form of osseous tissue, notwithstanding its different appearance in later stages, Transverse section of the femur of o Osteoblast layer. o'o"Cells becoming o'" A bone-cell. %> Periosteum. m Medullary cavity. THE TISSUES. 29 is very closely allied to the former kind, for its intercellular substance also is secreted from cells — arises, that is, by the differentiation of a part of the protoplasm. The connection is still closer if we regard the earliest stage in the process. In both cases a homogeneous sub- stance is secreted, which is hardened by calcareous compounds, and into this the cells, which form it, send their processes. If this process goes on in the same way as it began, so that a complete cell never passes into the secreted layers, it leads to the formation of that form of osseous tissue which is traversed by fine pore-canaliculi only, arranged for the most part in parallel lines. If some of the secreting cells gradually pass into the secreted substance, that substance becomes an intercellular substance containing bone-cells. Morphological Elements of the Nutrient Fluid. § 24. The cells, which are suspended in the nutrient fluid of the body, and which are its forrn-elements, are closely connected in origin with the connective tissue. If it is allowable to regard this fluid as an intercellular substance, then the whole of the nutrient fluid might be compared to a tissue, which would not differ from the other tissues of the connective series in any essential point other than its fluid conditio^. Even if we admit it to possess another function in consequence of this fluid condition, yet this function must be held as falling well within the category of vege- tative functions. Apart from the importance of these considerations, the form-elements in question must be enumerated in the present place, for they take their earliest origin from the tissue which forms the walls for the vessels of the nutrient fluid. As far as its characters are known, a portion of the cells which form the mesoderm during the processes of division do not become connected with the rest, but remain isolated in the fluid which fills these canals or spaces, which fluid is known as blood. These form-elements then are the blood-cells. In the Invertebrata they appear, as a rule, in the form of completely indifferent cells, consisting of a nucleus (Fig. 12, n) and protoplasm, which latter exhibits amoeboid movements. Among the Vertebrata these morphological elements persist as lymph- cells in the Craniata, while in the blood- fluid proper there are elements which are _,. 1e> _. , , li _b i°* ilj -dIoocL corpuscles derived from these forms, but are much of°" a 'crustacean (Maja altered. These latter have lost their amce- Squinado) with protopias- boid character during differentiation, and mic processes, n Nucleus. have the form of rounded or oval discs, the nucleus of which disappears in the Mammalia, though present in the rest of the Vertebrata. 30 COMPARATIVE ANATOMY. b. The Animal Tissues. § 25. In the epithelial, as well as in connective tissues, the product of the differentiation of the protoplasm gives rise to phenomena which are limited to the sphere of vegetative operations. When a more highly contractile substance arises as a product of the division of the protoplasm, a new tissue is formed, which is known as contractile or muscular tissue. Its contractility, however, is not automatic, but dependent on stimuli, which come from the form- elements of the nervous system. The contractile form-elements of the muscular tissue differ therefore essentially from the indifferent cell formed of protoplasm, although the latter also is contractile. They presuppose the existence of another, or nervous tissue, just as it on the other hand determines the existence of the muscular tissue. These intimate re- lations explain the causal relationship which these two tissues have to one another phylogenetically. The two kinds of elements are differentiated from a single neuro-muscular cell, which is in many Ccelenterata the representative of the two tissues. (Fig. 13.) This kind of cell corresponds to an indifferent stage of the animal tissues, in which they have not yet be- come distinct tissues. The tissue which forms the starting-point of the differ- entiation is not a new structure. It is the outermost layer of the body, and consists of cells, which form an epithe- Fig. 13. Neuro-muscular cells iium> The neuro-muscular tissue of Hydra. n Processes of the • ,i p -i-™. ,. ,. P , , cells, m Contractile fibres (after 1S therefore a differentiation from the Kleineuberg). epithelial tissue, and is thus connected with a more simple condition. Cells which are hardly at all different from other epithelial cells give off a band-like process at their base, which becomes connected to a layer of longitudinal fibres underlying the epithelium. While the epithelial cells of the outer layer of the body unite, when in their indifferent condition, a low grade of sensibility with a low grade of contractility, the sensibility, when the cells become more highly specialised, remains with them, and the contractility becomes assigned to a differentiated process of the protoplasm, which now appears as a distinct appendage of the cell. Thus commences that arrangement which, in the more highly differen- tiated stages, is expressed by the connection between ganglion-cells, nerve-fibres, and muscle-fibres. By sup- posing that the fibres, which in the earlier case appear to be merely processes of the cells, retain their nuclei, and that the products of the division of the nuclei of the cell gradually become fibres, and that further the neuro-muscular cell is no longer con- THE TISSUES. 31 nected directly, but by means of a separate process with the fibre, which has at the same time itself also become independent, we can see how the more differentiated stage has been brought about. Nerves and muscles seem from this point of view to be the products of the separation of one and the same layer of tissue, which tissue we shall learn later on to know as the " ectoderm." And at the same time a physiological postulate is thus satisfied : for clearly it is impossible to imagine that nerve or muscle once came into existence with their elements totally distinct from one another, and that the connection between them, on which their functions depend, was the result of a later union. Muscular Tissue. § 26. The morphological elements of the muscular tissue are, so far as their more special characters are concerned, divisible into ' two groups. One consists of cells simple in form, the other of fibres derived from cell-aggregates, or from syncytia; the latter is indicated by the presence of numerous cell-nuclei. In either case the amount of protoplasm, which retains its indifferent character, is slight, and subordinate in importance as far as the function of the form-elements in question is concerned. Further differentiation of the contractile substance may in either case lead to the higher development of the fibre. 1) The so-called smooth muscular fibres, or contractile fibre-cells, constitute the first form. They are spindle-shaped cells, which are often greatly elongated, and then are band-like in form ; in these cells either none of the indifferent protoplasm at all persists, or what does is to be found in the long axis, or at the periphery of the cell only. In all cases such remaining protoplasm surrounds the nucleus. The contractile substance is homogeneous and limited externally by a membrane, which is often difficult to demonstrate. The reaction of these muscle-fibres to nerve stimu- lation is slow. Owing to differentiation of the contractile substance into singly and doubly refractive particles, the fibres gain the appearance of transverse striation ; such is the origin of that variety of the tissue, which is known as transversely-striated muscular fibre. There are various intermediate forms between this kind of striated tissue, which consists of fibres derived each from a simple cell, and the other more homogeneous kind of fibrous muscular tissue. 2) The elementary parts of the other form of muscular tissue are formed by cell-aggregates (syncytia). They generally arise, as it seems, from the growth of one cell, the nucleus of which multiplies, so that they may be regarded as arising from the continuous but imperfect division of one cell. Their contractile substance either has a cylindrical shape, is limited externally by a homogeneous 32 COMPARATIVE ANATOMY. membrane (the sarcolemina), and contains several nuclei, with remnants of protoplasm along its axis ; or the contractile substance forms a solid cylinder, on the surface of which, and immediately below the sarcoleniina, are the nuclei with the remains of the proto- plasm. Further, there are two varieties of this form of muscle-tissue in which the contractile substance is respectively more homogeneous or more heterogeneous. If more homogeneous the fibres resemble the so-called smooth fibrous cells, from which indeed they differ only by the fact that they do not correspond to a single cell, but to a multiple of cells, as is clear from the number of nuclei appertaining to the fibre. In the other condition, owing to the differentiation of the contractile substance, they resemble the second form of simple muscular fibres, and, like them, are transversely striped. These also correspond to a number of cells although they are derived from a single cell, and owe their elongation to its growth. The reaction to stimulus is, in transversely-striped fibres, rapid. Nervous Tissue. § 27. Nervous tissue appears (as has been already explained) at the same time as muscular tissue in the Animal Kingdom, and is distinguished by its functions, even in its lower conditions, from other tissues. It receives and passes on stimuli, converts them into sensations, and produces voluntary excitations. Two conditions are to be distinguished in the morphological characters of the elementary parts, nerve-fibres and nerve-cells. The former are mostly present in the peripheral portion of the nervous system, and are the con- ducting organs, while the latter form the central elements. 1) Nerve-fibres have not always the same relations, and their different conditions are to be regarded as stages of differentiation. a) In their simplest form they are elongated homogeneous band-like bundles composed of fibres which are so slightly separated from one another that they appear to be merely striated. For the majority of Invertebrata the relation of nerve-trunks of this kind and their branches to the histological form-elements is not thoroughly made out ; and even the question whether the numerous striations of the nerve-trunks are to be regarded as the indication of their being composed of separate fibres, is an open one. The presence of nuclei in their structures is the sole fact which points to their relation to cells. In other cases fibres united into bundles may be distinguished as individual elements of structure. The fibre consists of an apparently homogeneous substance which is limited super- ficially by a fine membrane, beneath which are the nuclei. Remains of protoplasm may be at times made out around the nucleus, which shows that the rest of the fibre is a differentiated substance. The THE TISSUES. 33 structure of these nerve-fibres is therefore histologically of a similar grade to that of muscular-fibre, and the only difference between the two is in the quality of the differentiated protoplasm, which in one case gives rise to muscle, and in the other to nerve-substance. Such fibres are to be seen in the Invertebrata as well as in Amphioxus and the Cyclostomi. In the higher Vertebrata they are present only in the sympathetic nervous system. b) Further differentiation gives rise to a second stage of the nerve-fibre. The nerve-substance, which lies beneath a more or less delicate envelope, is differentiated into a chord which traverses the axis of the fibre — the axis-cylinder — and into a fatty substance which surrounds it. The latter, known as the medullary cylinder (medullary sheath), gives a highly refractive contour to the nerve- fibre, and can be separated from the axis-cylinder only by artificial means. The homogeneous sheath which surrounds the medullary cylinder — the neurilemma — contains the nuclei which are the remains of the cells from which the fibre was formed. So far as is yet known this form obtains in the Gnathostomous Vertebrata only. 2) The other form-element of nervous tissue is represented by cells, which are called ganglion-cells, as they are principally present in the swellings (ganglia) of the nervous system. They form the central apparatus. Their substance is generally finely granular in character, with many other peculiarities which cannot be entered into more closely here. The nucleus, which as a rule is provided with distinct nucleoli, lies in the middle of the granular substance ; this latter is often limited by an external membranous and firmer layer. A very complicated structure is ascribed to these cells, and is explained by every observer in essentially different ways, so that the questions involved appear to be still far from being settled. The ganglion-cells possess processes by which they are con- nected partly to one another, and partly to nerve-fibres. They form therefore the points of origin of the nerve-fibres. It is not yet settled how ganglion-cells, which are devoid of processes, and therefore completely isolated, can be of any service. The fact is, that the belief in their existence grows less and less every day. The processes of the nerve-cells vary greatly in number, as well as in their relation to the fibres ; the only point to be noted about them is that in the differentiated fibres it is the axis-cylinder which is continued into the substance of the cell, while the medullary cylinder ceases at some distance from it, or, rather, is no longer differentiated. The relations of the axis-cylinder to the substance of the cell appear to vary greatly, and are in many points a problem still* * * Solbrig, A., Ueb. d. fein. Strucfcnr der Norvenelemente der Gasteropoden. Leipzig, 1S72. 34 COMPAEATIVE ANATOMY. Origin of the Organs. § 28. In section 13, the title of organs was given to those parts of the body which were entrusted with a definite function for the purposes of the organism, and which had a form in correspondence with this function. In this general sense every form-element is an organ, just as much as the parts, which are made up of form- elements, and have a definite function, are organs. The conception of an organ is therefore a relative one. We must accordingly separate organs into those of a lower and those of a higher order. The former are represented by the morphological units or form- elements — elementary organs — while the organs of a higher order are those which are made up of a number of elementary organs — cells, and their derivatives (tissues) — and which are set apart for a single function. There are but few of these organs of a higher order in the lowest stages of animal organisation, owing to the simplicity of the organism. But these few organs form the ground- work on which the gradual complication of the organism is raised up by continued differentiation, and in accordance with the principle of the division of labour. We may therefore call those simple organs of a higher order, from which complex organs are developed by differentiation, " primitive organs." When we examine these primitive organs more closely, we find it convenient to associate them with the earliest processes of dif- ferentiation which take place in the organism, for they can be derived from them. A collection of smaller cells arises from the division of the egg-cell, and these have not all the same position. Some occupy the inner part of the organism, and others form a layer which surrounds it, and at the same time forms the external boundary of the body (Fig. 14). If in this stage of development the taking in of food into the body com- mences, then the inner cell mass becomes con- verted into the limiting layer of the digestive cavity, and forms a primitive gut (enteron). In many observations the process of division into two layers is described as due to the Fig. 14. Separation of invagination of a one-layered vesicle. In the mass of cells formed 0^aer cases it is represented as taking place by change of the yolk „ ., , ..r. . ., , °, l into a peripheral (c) differently, so that it is impossible to make out and a central (d) por- whether there is any phenomenon common tl0n- to all cases, and, if so, how far it is common. Let us therefore turn to the results of the process, without making any generalisation about it. We now have an organism made up of two layers of cells. An outer one, or ectoderm, which forms the primitive integument, and an inner THE OEGANS. 35 one, or en do derm, which surrounds a primitive enteric cavity. The two layers pass into one another at the oral opening which leads into the cavity. The two cell-layers, which form the body of such an organism, furnish the conditions under which it is possible for it to lead an independent animal existence. The outer one is the organ of support, and may be converted into an organ of locomotion if it gives rise to cilia, and may be the seat of respiratory functions also. In so far as it per- ceives the state of the surrounding medium it is an organ of sensation too. The inner layer is nutritive in function, produces changes in the food which is taken in, and allows what can be assimilated to pass into its cells ; and these in their turn feed the outer layer of cells. What is useless is passed out again by the same opening as that by which it entered. As the functions of the two layers are dif- ferent the special characters of the morpho- logical elements which compose them are different also ; we need only call attention now to the much greater size in most cases of the cells of the endoderm, as compared with the cells of the ectoderm. This grade of organisation is to be seen in some of the lower divisions of the Animal Kingdom (Coelenterata and Vermes), where it represents a lowly stage of development. Indications of it are to be seen even in the higher divisions. This form has been called the Gastrula, on account of the dominant development of the enteron. Starting from the hypothesis that forms agreeing with a Gastrula in all essential points were the precursors of all the higher forms of animal organisation, a Gastrasa-form resembling the Gastrula has been regarded as the primitive ancestral form of all animals. This Gastrasa theory is based, first, on the existence of independent animal forms which resemble the Gastrasa; secondly, on the fact that the embryonic body which commences with a Gastrula, does not, in the lower divisions, rise very much above it, so that even apparently considerable complications of the organism can be traced back to the existence of these two layers of the body ; thirdly, the presence of these two layers of cells, forming the ectoderm and endoderm, as a general, constant, and therefore regular phasnomenon, even in the higher divisions of the Animal Kingdom, as well as their constant relation to the same functions, is a fact of the greatest importance for the hypothesis in question ; indeed the occurrence of these layers as the so-called germinal layers, which make up the embryonic body, cannot be rightly understood without a reference to a hypothetical Gastrasa-form. This hypo- thesis may therefore be regarded as justified. d 2 Fig. 15. Diagram to represent the first dif- ferentiation of the or- ganism into ectoderm and endoderm, and the formation of a digestive cavity, a Mouth, b En- teric cavity. c Endo- derm. d Ectoderm. (In transverse section.) 36 COMPARATIVE ANATOMY. We recognise then the Gastrasa as a fundamental form, and discover in the differentiation of two layers corresponding respec- tively to endoderm and ectoderm, which are present even in the highest grades of the Animal Kingdom, facts which point to such a Gastraea stage, and are due to it. But it must not be at all sup- posed that we have advanced farther than just on to the threshold of a knowledge of these relationships. The definitive explana- tion of many of the points which have considerable importance in this matter is still far distant, and but little light has yet fallen on even such apparently simple points as the origin of the Gastrula and its two layers. It is a question whether the form which pre- cedes the Gastrula is a one-layered vesicle — that is, whether the two layers of the body are due to the endoderm being formed by invagination ; or whether the endoderm is developed from a pi'imitive internal cell-layer delamination. And again, whether the two con- ditions that have been observed are independent of or derivable from one another. Further investigations will have to settle all this, and our judgment must therefore be proportionately reserved until such investigations have been made. § 29. The two layers of which the body of the lower animals is made up during their early stages, and which are, in the higher divisions, represented by the germ -layers — that is, the ectoderm and endoderm — give rise to an intermediate layer or mesoderm, in the formation of which the other two apparently take an equal share. But it is not yet definitely known what share each takes, since the earliest processes of the differentiation of the rudiments of the body still require much careful investigation, and, moreover, they do not present the same characters in all cases. These three layers appear directly after the segmentation of the ovum in the higher animal organisms, and are coincident with the first traces of histological dif- ferentiation. They represent the outline of the organism in the condi- tion of a germ, and from this the whole organism by differentiation evolves itself. These rudiments of the body exhibit great modifications in the higher divisions of the Animal Kiugdom, and the stage which is represented by the Gastrula form is more difficult to make out in proportion as the differentiations through which the organism has to pass are more considerable; but the principal features can be easily recognised as identical in all cases. The outer germinal layer (deric layer or ectoderm) forms the outer limiting layer of the body, and the inner (lower) germinal layer (enteric layer, glandular layer, or endoderm) the foundation of the gut or enteron. The middle layer (mesoderm) afterwards arises between them. As the ectoderm and endoderm are the first organs marked off in the course of development, the germinal layers are to be regarded as primitive organs, which have been transmitted from the earliest THE ORGANS. 37 stages in the differentiation of the animal organism to later, and there- fore higher stages, and which, following the law of the division of labour, give rise to series of new organs. We do not yet know enough of the details of the organological differentiation of the germinal layers to be able to give the history of every organ. However, the facts which are clearly established with regard to, at any rate, some divisions of the Animal Kingdom enable us to follow out the first steps in the process of differentiation. The organs which put the organism into relation with the outer world, such as organs of defence, of support, and of sensation, are principally derived from the ectoderm (hence called the sensory layer), also those of move- ment ; while the endoderm principally provides the organs for the preservation of the individual and of the species (nutritive layer). As the origin of the mesoderm, out of which important organs are formed, is still very obscure, the relations of these organs to one or other of the two primitive germinal layers must be left an open question. The primitive character of the organism more or less disappears as the rudiments of the body are formed out of the germinal layers, and as fresh organs which render the organism more complicated arise in it. Organs differentiated from the germ layers which act the part of primitive organs are reckoned as secondary organs. From these, tertiary organs are formed, and so on. The separate organs differentiated out of a primitive organ remain connected together, owing to the fact of these processes of separation being due to the division of a function, and of the separate functions being subordinated to the primary function, from the breaking up of which they took their origin. Combinations of organs are therefore formed, which are known as organic systems, on account of their morphological and physiological connection. This connection does not always persist in the adult condition ; and, in fact, organs primitively connected often become separated. This obtains chiefly in those organs which serve several purposes, for when the functions become independent the organs become so too. But even in these cases ontogeny indicates what was the primitive condition. Systems of Organs, a) Integument. § 30. The ectoderm, as the outermost layer of the body, forms the simplest condition of the integument of animal organisms. In the lowest organisms (Protista) there is either no integument at all, the protoplasm which forms the body being protracted into ever- changing processes (pseudopodia), or the integument is represented 38 COMPARATIVE ANATOMY. by the outermost layer of the protoplasm of a single cell, in which case we have the first example of a denser stratum of the cell becoming separable as a distinct envelope and covering for the rest of the organism. The ectoderm has the function of an organ of defence when its cells secrete a substance which invests, more or less per- fectly, the surface of the body. This substance may harden and give rise to tests or shells, or form a continuous covering for the body, like the carapace of the Arthropoda. When a mesoderm is formed, that part of it which becomes con- nected with the ectoderm takes on, in various ways, the functions of an organ of support. The calcareous deposits in the complicated integument of the Bchinodermata are examples of this. The activity of the ectoderm in producing firm organs which protect the body is seen also in the Vertebrata, where numerous and varied parts, which function as investing and protecting organs, are produced by it. b) Skeleton. §31. In proportion as the various protective organs which are formed from the ectoderm increase in size or in strength, and at the same time become connected with internal organs, they attain the function of organs of support also. Such organs we designate as the skeleton. The combination of inorganic substances (chiefly calcareous salts) with an organic base plays an important part here. The supporting function of the integument gives rise to numerous adaptations. The combination of the functions of both protection and support is clearly a lower stage as compared with the formation of internal skeletons, which indicate a higher functional differentiation, and function exclusively as organs of support. Here, too, we meet with the most various conditions. The lowest forms, the first beginning of such internal skeletal organs, are solid deposits in the tissues, the separate pieces of which have no connection with one another. The growth and union of these deposits give rise to skeletal formations, which may be also regarded as excretions. Examples of them are found even in the Ccelenterata. When a definite tissue, the properties of which specially fit it for the function of support, is brought into use, the internal skeleton assumes a higher degree of development. The differentiation of cartilage from the indifferent connective tissue is the first ex- pression of this phenomenon. As low down as the Medusas, among the Vermes and among the Mollusca, the employment of cartilaginous tissue for organs of support is commenced, and in the Vertebrata it attains to greater importance, until it is pushed aside by a second and more perfect skeletal tissue — the osseous. THE ORGANS. 39 c) Muscles. § 32. The locomotion of the body exhibits itself in its simplest phase as a change in the form of the body due to the contractility of its protoplasm. When these changes in form follow one another rapidly, and have all the same direction, the body either elongating or sending out processes which attaching themselves to some fixed point, are followed gradually by the rest of the semi-fluid body (Rhizopoda), locomotion is effected. The difference between this mode of locomotion and undefined change of form is seen to be merely one of degree. The contractility of protoplasm may pro- duce changes in position even when it is invested by a differentiated, though soft, integument. This layer of integument will follow the movements of the body it invests. In such cases, and they are very common among the Protista, special organs of locomotion cannot be said to exist, for the cilia have other functions to perform for the organism in addition to locomotive ones ; such, for example, as that of aiding in the ingestion of food. Specific organs of locomotion make their first appearance when the contractile morphological elements known as muscle-fibres are differentiated; these, in the simplest case, form a muscular layer lying beneath the ectoderm. The genesis of this earliest musculature of the body is due to a differentiation of the ectoderm (Hydroid polyps), the cells of which give off flattened processes, which form a continuous layer of contractile fibres. Each individual ectoderm-cell concerned in the formation of this layer of fibres represents accordingly a sensory apparatus, which stands in direct continuity with a contractile apparatus. The cell is indeed replaced, when the musculature is differentiated, by groups of muscles which work so as to balance one another, and completely harmonise in their action (compare Sect. 31). We cannot yet say how far this process, winch gives so deep an insight into the mode of differentiation of the tissues as well as of the organs, is repeated in the ontogeny of the higher forms of animals. In all divisions above the Coelenterata we always find the separation between ectoderm and muscle complete. It may therefore be doubted whether a process of the kind described in the Hydroid polyps always accom- panies the origin of the muscular system. But yet it is very probable that something of the kind does occur. Even though the process of differentiation in the higher organisms does not enable us to recognise these processes in their case, yet it is not to be assumed without further reason that the mode of origin of the muscular tissue was in them primitively different, for Ontogeny very seldom repeats phylogenetic processes in every detail. 40 COMPARATIVE ANATOMY. § 33. The earliest musculature of the body is closely related to the integument, from which it can with difficulty be separated. Since this is the case not in the Coelenterata alone, we have here an instance in favour of an essentially equivalent origin for this part of the musculature in all cases. Together with the integument, it forms, on the appearance of a body- cavity, a " dermo-muscular tube/5 which encloses the other organs. The arrangement of the muscular fibre seldom presents much regularity till the body becomes jointed into separate parts, placed one behind the other (metameres) ; and with the development of organs of support the muscles become differentiated into separate groups. Collections of fibres form bundles, and these again make up larger complexes, muscles. The segmentation of the muscular system corresponds therefore to the segmentation of the body, and the separate segments differ in proportion to the difference in function of the metameres. The various kinds of movement which are produced by the crossing of the fibres of the dermo-muscular layer in different strata, are, where the muscular system is more highly differentiated, effected by groups of muscles acting in opposition to one another, and completely balancing one another in their action. Locomotion by movement of the whole body is brought about by the dermo-muscular tube and the differentiations which arise from it ; the whole integument, in the first instance, takes a share in this activity. A further differentiation arises from this state of things when special appendages are formed, as limbs, on certain parts of the body. When the animal changes its place these act as the arms of a lever. They have the form of simple soft processes of the dermo-muscular tube (Ringed worms), or of jointed organs, which are supported by the integument (Arthropoda), or by means of internal skeletal structures ( Vertebrata) . The complication of the muscular system is in close connection with the development of supporting organs ; and the two form a single locomotor system in which the skeleton plays the passive part. d) Nervous System. § 34. In the lowest conditions of animal organisation the protoplasm of the cells is the seat of sensation, as well as of movement ; and this is permanently the case in the lowest organisms. As the muscular layer of the body becomes differentiated, the ectoderm becomes the principal organ of sensation. The differentiation of a nervous system is due to the further development of a portion of this layer THE OEGANS. 41 as a sensory organ, for which, reason such an organ must be at first placed superficially. This superficial position of the earliest rudi- ments of the nerve-centre has been already made out in so many forms that it may be regarded as a general phenomenon. As the sensory organ becomes differentiated from the ectoderm it sinks down into the body. The developing central organ is thus gradually covered over by other layers of the body. This arrangement, which is most peculiar, and by itself most unin- telligible, is explained if we regard it as inherited from a more primitive stage, in which the nervous system was but slightly differentiated, and was represented by the whole cell-layer of the ectoderm, or by part of it. We must consider its gradual attain- ment of an internal position to be a process due to its continued differentiation, and consequent higher potentiality ; the organ, which has become of greater value to the organism, gets hidden within the body. With regard to the structural characters of the differentiated nervous system, the central organ, which is chiefly composed of ganglion-cells, is to be distinguished in the first place from the nerves, which pass to the terminal apparatus, and consist of fibrous elements (peripheral nervous system). § 35. The earliest complications are due to the appearance of several parts (ganglia), in which are central form-elements connected with one another : the further development of these parts is very various. The ganglionic mass, which forms the central organ, is primitively dorsal, owing to the earliest separation of the central organs taking place from the dorsal ectoderm, as we have already seen. This dorsal nervous mass, which generally lies near the entrance to the alimentary canal, is differentiated into several parts, which are con- nected together by commissures; their fibres form an oesophageal ring. In the animals built on a radiate plan the number of the ganglia is increased in correspondence with the radii ; the peripheral distribution of the nerves also follows these general structural relations exactly. The nervous system in bilaterally-symmetrical animals follows the bilateral arrangement. The more primitive form is represented by a superior ganglionic mass (cerebral ganglion). Other ganglia do not seem to be formed until the metameres are formed. We then are able to distinguish dorsal and ventral ganglia ; the latter may form ganglionic masses along a continuous longitudinal trunk, or a single suboesophageal ganglion. The variations in size of these oesophageal ganglia are in the closest connection with the nerves which pass off from them. When sensory organs are developed, the ganglion which sends off their nerves becomes of considerable size, while it seems to degenerate when they grow less. The supra-oesophageal ganglia are the most important in this relation, for it is from 42 COMPAEATIVE ANATOMY. them that the nerves of the higher sensory organs arise, which in position and direction have one widely-distributed arrangement. From this form another is directly derived, determined apparently by the well-marked metamerism of the body. Whilst in unseg- mented animals possessing an oesophageal ring, the ventral parts of the body are supplied by nerves which arise from the suboesophageal ganglia, we find that the number of ventral ganglia is increased when the body is broken up into parts lying one behind the other (meta- meres). A ventrally-placed series of ganglia is formed by the development of a separate pair of ganglia for each segment ; and these, being united to each other by longitudinal commissures, form a ganglionic chain. The Ringed worms and the Arthropoda present us with this form. Further differentiation gives rise to all kinds of variations of this type. In the first place, the size of the ganglia varies with the size of those parts of the body that have to be innervated ; and in the second place, the ganglia of several segments of the ventral cord fuse into larger ganglionic masses. Even when the central nervous system is entirely dorsal, as in the Vertebrata, it undergoes differentiations of this kind. When the most anterior part of the body is developed into a head, the most anterior part of the central nervous system is developed into a special region, the brain, which is marked off from the remainder of the medullary tube, or spinal cord, which remains more equal in size throughout. As differentiation advances, variously developed regions appear in the brain. e) Sensory Organs. § 36. The sensory organs inform the organism of the condition of the outer world. Protoplasm, in its indifferent condition, charac- teristic of the lowest organisms, reacts to various stimuli from with- out, and appears to be the seat of the lowest kind of sensation. When the surface of the body is not completely marked off from the inner portion of the organism (Rhizopoda) it is used as an organ of perception, of course of the very lowest grade ; it functions therefore as a sensory organ of the lowest order. When the surface is more distinctly marked off, and a distinct outer layer of the body is established (Infusoria, Gregarinas), we get a differentiation of great importance for sensory perception. Although, indeed, particular parts of the surface in the Infusoria specially acquire the function of sensory organs, yet there is no ground for speaking of sensory " organs " in an anatomical sense, in this case, any more than there is in the still lower stages. Sensory organs only appear when a nervous system is marked off, for sen- sory organs are the end-organs of the sensitive nerves. THE ORGANS. 43 Their presence therefore presupposes that form of differentiation which we treated of above, when speaking of the nervous system. Since ontogenetic facts point to the primitive segregation of the nervous system from the ectoderm, as being most probably a funda- mental process, this same outermost layer of the body becomes also of the greatest importance in studying the origin of the sensory organs. Almost all the sensory organs are derived, directly or indirectly, from it; whence arises the permanent or temporary connection of these organs with the integument. It is very unsafe to assert what are the functions of many of the sensory organs of the lower animals. This applies to all those organs which are not comprised amongst those which fall within the domain of our own judgment, on account of our possessing them or their homologues, in which case only is it possible that the connection between their structure and specific function can be estimated. Such outstanding organs have been classed together as organs of a sixth sense. § 37. The sensory organs are divided into lower and higher. The former are commonly distributed over the integument, and are simple iu structure. Compared with the higher they represent a more indifferent condition. Modified cells of the integument, which generally belong to the epidermis, connected on the one hand with a nerve fibre, and on the other provided with a process of varying shape, which is directed towards the surface of the body, are the most common examples of the lower sort. They are regarded as the organs of general tactile perception ; but the physiological function of these organs, especially in aquatic animals, has not been definitely determined, and it is reasonable to suppose that many of them are the media of specific sensations, in which case they would resemble the higher organs of sense. The significance of these arrangements is somewhat more certain when they are connected with special organs, such as movable processes of the integument and the like ; they then appear to be tactile organs. It is still a question whether structures of this kind, especially in the lower divisions of the Animal Kingdom, are of use for perceptions other than tactile. The higher sensory organs present themselves to us as special elaborations, with one special function and capable of response only to stimuli of one special kind ; they are to be regarded as developed from the lower kind of sensory organs, and oftentimes still possess the essential structure of that lower kind. Organs of taste and smell can only be certainly distinguished in the higher divisions of the Animal Kingdom, and the function of the latter is certain only in those Vertebrata which live in the air ; in the lower divisions it is still doubtful. Even in the case of the organs of taste the greatest caution is necessary as to their real import. The value of a sensory organ to the organism determines its being protected against 44 COMPARATIVE ANATOMY. external influences. This explains the foldings-in of the portion of the integument which is about to be differentiated into a sensory organ. It is for this reason that the higher sensory organs gradually sink beneath the level of the ectoderm as they are developing, and attain a favourable position for further development. § 38. Vesicles filled with a fluid, on the walls of which a nerve ends, are regarded as auditory organs (o to cysts). In its simplest form the vesicle is directly connected to the central nervous system, or the nerve passes from it to the vesicle. These vesicles almost always contain firm concretions or crystalline structures ; and very often crystals of calcic carbonate. There are often hair-like prolongations of the end-organs in addition to them, which project into the lumen of the vesicle. This form of auditory organ, which obtains in the Invertebrata, is complicated in the Vertebrata by diverticula and outgrowths which form a labyrinth. New arrangements are produced in the form of organs for carrying and increasing the sound, which become attached to the auditory organ, although they primitively presided over other functions. Inasmuch as the labyrinth-vesicles of the Vertebrata are developed from the integument, the terminal organs of the auditory nerve which are differentiated in its walls are genetically connected with the terminal organs of the tactile nerves, which lie in the integument; they may therefore be regarded as a specific development of a lower sensory organ. The genetic relations of the simpler otocysts of most Invertebrata are as yet unknown, but all the more exact results point to the supposi- tion that they arise by a differentiation of the ectoderm. The optic organ also has a simple mode of origin. We exclude the pigment spots, which used to be often called eyes, and only recognise an eye where a nerve-ending of definite form can be detected, either under or on the surface of the body, acting as an organ for the perception of light. By the light-absorbing property of the pigment it is possible that indefinite sensations of light and shade may be produced, or other sensations altogether unlike that which we call " sight" may possibly be produced by the heat-rays alone of the light. The function of pigment in the way just noticed is doubtful, but when it surrounds a part only of a rod-like nerve-ending, and that in such a way as to leave the outermost end free, and exposed alone to the influence of light, it has, clearly enough, a definite function. Optic organs of various degrees of complexity are formed by the union of a few or of many nerve-endings; the elements which are the medium of light-perception (rods) forming a convex or concave layer. Another complication is due to the addition of organs to refract the light (lenses) ; these, too, may have all kinds of relations, but they are always, either directly or indirectly, derived from the integument. In eyes in which the surface of the THE ORGANS. 45 layer of rods is convex there are, as a rule, as many lenses as there are perceptive nerve-endings ; when the layer of rods is concave, there is one lens only. By the addition of other arrangements to the nervous apparatus of the eye, by which its functional capacity is modified or increased, this organ becomes one of the most com- plicated of the animal economy. In most of the lower divisions the optic organ, even when fully developed, still retains its primitive relation to the ectoderm. In the higher divisions it is separated from it, and gets to lie, together with its perceptive apparatus, be- neath the integument, or the perceptive apparatus is derived from the embryonic foundations of the nerve-centre. The phenomena of differentiation may be seen even in what relates to the position of the optic organ, for the parts of the body which carry the eyes, as well as the number of the eyes, varies greatly in the lower divisions of Animals. Connected with this is the occurrence of a great number of eyes on the anterior part of the body, which goes to form the head, until, finally, the number of eyes on the part in question becomes limited to two. The different position occupied by visual organs forbids us to suppose that they have had a common hereditary origin, and is in favour of these heterotopic organs having been independently differentiated from an indifferent apparatus. On the other hand, that eyes which are connected to the cerebral ganglion, or the dorsal nerve-centre, have a common genetic relation, is not to be disputed. f) Respiratory Organs of the Integument. (Dermal Branchise.) § 39. An important part is played by the integument, and therefore by the ectoderm, in the formation of respiratory organs. Before they appear the gas exchange is carried on probably by the whole surface of the body, and this mode of respiration obtains in many of the lower aquatic animals. A change of the surrounding medium is effected, partly by the movements of the body, and partly by special organs, for example cilia ; thus fresh quantities of it are con- tinually brought into contact with the respiratory surface. This is not, however, the only method of respiration in the lower animals, for the introduction of water into the interior of the body, in fact the bathing of the alimentary canal by watei', is certainly not without significance in this direction, while it is of great im- portance as being the beginning of a long series of differentiations. Certain limited portions of the surface become more developed in this direction as the function becomes localised, and, in compensation for this limitation, acquire the form of blood-carrying processes, which are called branchiae. In many cases they are differentiated 46 COMPARATIVE ANATOMY. from tlie appendages (Vermes, Crustacea). An increase of the surface, which is brought about in various ways, is the mode in which the further complication of branchiaa takes place ; it is very frequently accompanied by a reduction in the number of separate branchial organs. The importance of branchiae to the body calls into existence various kinds of supporting arrangements for these organs, which, in their lowest condition, project freely from the surface of the body. Neighbouring parts of the integument being raised up into covering lamellae, the branchiae become hidden in cavities (branchial cavities), and the same tegumentary folds give rise to afferent and efferent canals for the water, which serves for respira- tion (Mollusca, higher Crustacea). In this way the development of respiratory organs may affect other parts of the integument, the direct relation of which to respiration had been lost for a very long time. g) Excretory Organs. § 40. Just as the gaseous excretory matters are eliminated from the organism by the respiratory organs, so too are there arrangements for eliminating the solid or fluid matters which have become useless to the organism. The whole surface of the ectoderm performs this function in the lower organisms ; in the higher forms of life, on the contrary, there are special organs, dermal glands, which have this function. Of those general arrangements which function as organs of secretion we have to do here with those special ones only which eliminate the excretory matters, and which are distinguished as excretory organs from those glands which secrete matters which are of use to the organism ; these latter are either indepen- dent, or are united to definite systems of organs, of which they are, in that case, specialised parts. The excretory nature of the products of secretion of those secreting organs, which are formed by the ectoderm, is least open to doubt, for the products are removed at once from the organism by the emptying of the gland. Of the various kinds of organs which open on the surface of the body one sort attains to general importance. These are the kidney- like excreting organs, which eliminate the nitrogenous excreta from the body. These organs are distinctly derived from dermal glands, notwithstanding that in the Vermes, where they seem to have their most simple form, they penetrate deeply into the body ; nor does the fact that in many cases (Annelida, Mollusca) the organ, which in other points also is much modified, opens into the body- cavity, and so connects it with the surrounding medium, and even serves in many groups (Mollusca) as a means for introducing water, THE ORGANS. 47 affect the question of its origin. In other forms (Annulata) these organs, having a tubular form, assist in the generative functions, by serving as ducts for the generative products. The recurrence of this function for a portion of the primitive excretory apparatus (primi- tive kidney, archinephron) of the Vertebrata might be explained as an inheritance from a lower stage. How far such a view is justified is still uncertain. In any case a genetic point of contact between the primordial kidney of Vertebrata and the renal tubes of lower organisms, can only be looked for where the apparatus is, as in the Vertebrata, single on each side of the body. h) Alimentary Canal. § 41. The ingestion of nutrient matter into the body is, in some of the lowest organisms, effected by endosmotic processes, in which the surface of the body takes the principal share. In others solid nutri- ment is ingested, the soft protoplasm sending out pseudopodia, and embracing the nutrient matter which happens to come into the neighbourhood (Rhizopoda). The formation of a definite part of the surface of the body, serving for the ingestion of nutriment, is really a step towards organological differentiation (Infusoria) ; but this does not constitute an alimentary canal, which does not appear as a separate organ till the body is differentiated into cell-layers. The cell-layers, when they do appear — an inner and an outer — pass into one another at the margin of the orifice of entrance. The inner layer, or endoderm, lining a space open to the exterior, forms the wall of a digestive cavity. In the simplest form, represented by the Gastrula, the endoderm is the sole wall of the primitive enteric cavity. The formation of a mesoderm gives rise to other layers external to this one. Of these the most im- portant is a muscular layer, for by it the intestine is enabled to per- form independent movements. The opening which leads into the enteric tube serves as a mouth for the ingestion of nutrient matters, as well as for an opening for the rejection of the undigested remains of the food (Ccelenterata, many Vermes). The appearance of an anal orifice produces a further separation of functions, and converts the blindly- ending gut or enteron into a tube open at both ends, the separate portions of which take on various functions, and so undergo different adaptations. The first portion, which is connected with the mouth, forms an oesophagus, which serves for the introduction of food ; then follows the true digestive cavity, which is generally widened, or provided with cascal sacs, and is generally called the stomach, though this name is not always applied to equivalent parts. The terminal part of the whole system serves for further alteration of the food, as well as for the excretion of the remnants 48 COMPARATIVE ANATOMY. by the anus. This differentiation of the digestive tube into several unequal parts is the most important complication which it undergoes ; any further differentiations are subordinate to this. Three tracts are accordingly henceforward distinguished, as fore- gut, mid-gut, and hind-gut. In addition to the varying and numerous changes in size which the different portions of the canal undergo, other arrangements, due either to special new functions, or mere expressions of further division of labour, arise in it. Organs for seizing and comminuting the food become attached to the mouth, or mark off a portion of the oesophagus (masticatory organs). In the stomach also there are sometimes masticatory organs of this kind. When they occur at the commencement of the oesophagus, just behind the mouth, this part, which is frequently distinguished by its larger supply of muscles, is known as the pharynx. The size of the cavity of the canal is increased by dilatations, or cascal diverticula. Crops are formed in the course of the oesophagus, caecal sacs on the stomach and on the rest of the intestine, which are variously complicated in number and arrangement. When the length of the alimentary canal is greater than that of the body, it is arranged in ascending and descending loops, or in coils, and so adapted to the size of the cavity in which it is contained. Both the quantity and quality of the food ingested is of the greatest import- ance as affecting all these relations of parts ; and nowhere is the adaptation of the organ to its function — which results from the mode of life of the animal — more clearly seen than in the arrange- ments of the alimentary canal. Secretory organs are generally connected with the alimentary canal, to aid in the process of digestion ; their products dissolve, and act on the nutrient matter by chemical change. Glands of this kind are sometimes distributed over the whole canal, and some- times distinguish certain portions of it only. In their simplest form they are not differentiated from the enteric wall, and in that case are not distinctly marked-off parts. Those marked off from the wall of the enteron are separated into two chief divisions. One of them comprises the glands which open in the buccal cavity, or its neigh- bourhood, and are distinguished as salivary glands. Another group is formed in the portion which serves for digestion, and is regarded as a bile-producing organ, a liver. It is right to note that the distinguishing of these organs by names which are applied to organs of higher organisms, physiologically better understood, is merely hypothetical, for nothing is known of the physiology of most of the organs of the lower animals. This holds especially for the epithelium of the gut, which generally appears coloured, and which is often called the ' ' liver/' This organ appears under the form of an epithelium, lining a part of the digestive cavity in the Ccelen- terata, in many Vermes, and even in Insects, till at last it becomes limited to definite caecal appendages of the alimentary canal, and so attains to the lowest grade of independence. The liver presents THE ORGANS. 49 itself from this point onwards either in the form of numerous follicles, which beset a large portion of the canal, or it forms a large group of glands, which open separately or together into the alimentary canal. The differentiation of the liver leads to a gradual separation of that organ from the digestive tube, so that finally it is merely connected to the canal by its ducts (higher Mollusc a, Verte- brata). Respiratory Organs of the Enteron. § 42. The general differentiations of the primitive gut (archenteron), formed by the endoderm, which have been already mentioned, give rise, in obedience to the principle of division of labour, to organs which serve for the ingestion and digestion of nutrient matters; these do not confer any essentially new functions on the gut. But such a new function does appear when the gut acquires relations to respiration. It is not certain whether this function obtains in the primitive gut, although this is probable, for the endoderm is bathed by the surrounding medium, like the outer layer of the body, while the water which is taken in with the food may serve for respiration. The relation becomes much more definite when we note the regular streaming in of water into the rectum, which obtains in many Vermes and Mollusca. This phasnomenon is an indication of the respiratory function of the gut, but has no bearing upon the forma- tion of distinct respiratory organs, which are differentiated from the digestive tube. Such a respiratory organ is brought into existence hi the most anterior portion of the gut, by its walls being broken through by lateral pores ; by special relations of these pores to the vascular system it acquires a respiratory significance. This arrangement, which already makes its appearance in the lower divisions, occurs again in the Vertebrata. Processes, known as branchias, in which the respiratory vascular network is distributed, arise on the walls of the clefts of this cavity. A region of the primitive digestive tube is thus converted into a special portion, a branchial cavity, at the hinder end of which the tube which serves exclusively for nutrition commences. Another form of respiratory organ is developed from the wall of the gut, in the form of a diverticular outgrowth of the anterior portion of that organ. This appendage of the gut is filled with air, and in fishes has merely a hydrostatic function. As the relations of the circulation become changed it is gradually converted into a respiratory organ, and becomes the lungs; in the higher divisions of the Vertebrata new organs, namely, those for the production of a voice, are developed on the passages leading into it. 50 COMPAEATIVE ANATOMY. i) Vascular System. §43. The substances prepared by the digestive process for the nourishment of the body are, in the lowest organisms which take in solid food, merely distributed from the digestive spaces into the protoplasm of the body. When a distinct digestive tube is formed, nutriment passes through its walls straight into the parenchyma of the body, so that the mesoderm and ectoderm, with the organs differentiated from them, are nourished by the endoderm. This is characteristic of the Coelenterata and some groups of Vermes. In many others a dividing of the mesoderm occurs, which takes the form either of canalicular cavities, or of a complete splitting of the mesoderm into an outer plate attached to the ectoderm, and an inner one attached to the endoderm. Between these dermal and gastric layers of the mesoderm is the body-cavity, or perienteric- cavity (c ce lorn), in which a fluid, to be regarded as the nutrient fluid, is collected. When morphological elements are found in this fluid, they are derived from the cells of the mesoderm. This fluid is not at first exclusively nutrient ; it also subserves locomotion, by swelling out different parts of the body at the will of the animal. An important function of this kind is also played by the water, which in many cases is taken into the ccelom from the exterior. The movement of the fluid in the general cavity of the ccelom is at first effected by the movements of the body. Contractions and expansions of the body-wall cause the fluid which is shut in by the dermo -muscular tube to continually change its position ; this may be regarded as the lowest form of a circulation of the blood. In this case the passages have not special walls, nor are there any special arrangements for regulating the circulation. This simple condition persists in many divisions in which the ccelom is developed (Bryozoa); in others canalicular cavities arise, which are arranged regularly, and have the form of vessels, and may undergo further complications. Their contents form the hasrual fluid or blood (Nemertines). When in addition to these vessels a perienteric-cavity is formed, the vascular system, which is partly enclosed in it, is either completely shut off from it (many Annelida), or is placed freely in communication with it at one or more points (Mollusca, Arthropoda, Vertebrata). In the latter case the vascular cavities must have arisen as portions of the body-cavity, while in the former case the body-cavity was not formed until after the vessels. The formation of the body-cavity is therefore, in the case exemplified by the Annelids, to be regarded as a secondary process ; and the formation of a hollow space in the mesoderm has accordingly led to two different results successively; on the first occasion to the formation of blood vessels, on the second occasion to the formation of a body-cavity. THE ORGANS. 51 § 44. Certain portions of the hollow cavitary system, which forms the luenial passages, are converted into contractile vessels by the development of muscles in their walls. The earliest circulatory system arises by these producing by rhythmical action the regular in-and-out flow of the blood. But the direction of the stream of blood is not yet constant ; it can be driven first to one side and then to the other. The portions of the vascular system which are dis- tinguished by their greater contractility are sometimes extended over a large surface, and sometimes limited to shorter parts. They are the beginning of the formation of a heart. The heart is therefore an organ differentiated from the blood- vascular passages, and in its simplest form is a portion of the vessels which is able to move its contents in two directions. It is only when valves appear at the ostia of the cardiac tube, that the direction of the flow is defined ; the structure of the heart thus be- comes complicated, and is further elaborated by being divided internally into separate portions (ventricles and am-icles). Contractile organs of this kind often appear as the only differentiated parts of the blood- vascular system, formed from the cavity of the ccelorn. The blood passes directly from the heart into lacunar portions of the coeloin, between the different organs, and from thence back again to the heart (Arthropoda), or there are definite vessels going off from the heart, which sometimes traverse the body in the place of the hollow cavity, or only partially replace the lacunar passage, in that they do not on their way back to the heart reach it as vessels, but into lacunar spaces. In this case the cavity of the coeloni shows itself as a portion of the blood passage, which is only partly represented by true vessels (Mollusca). Where the vessels are completely developed and the heart differentiated, the vascular system is divided into three parts. That which leads from the heart and distributes the blood in the body is called the arterial, and its vessels arteries. The passage which takes the blood back to the centre of the circulation is formed by the veins, and the part of the passage which lies between the afferent and efferent vessels, forms a network of extremely fine canaliculi (capillaries). This intermediate portion is very frequently replaced by a lacunar system, in which case the greater number of the venous passages also have no special walls. It is very often difficult to say what should be regarded as a vessel and what as a lacuna, and the distinction often depends on very unimportant points. It is not sufficient to say that the essen- tial character of a vessel is the investment of a cavity by flattened elements derived from modified connective tissue, for these elements might just as fairly be regarded as the covering of the other organs which wall in these spaces ; it is therefore questionable to call wide internal cavities, invested by such cells, "vessels." This cannot e 2 52 COMPARATIVE ANATOMY. indeed be regarded as the sole criterion, and it should only have weight when considered together with the greater or less regularity of the lumen. But in examining this question we must bear in mind one thing ; namely, that in all these cases we have to do with spaces which are walled in by connective substances, and that vessels are differentiations of these spaces, and therefore presuppose an indifferent condition. Between the two stages, the differentiated and the undifferentiated, there are all kinds of intermediate steps. k) Reproductive Organs. §45. The phasnonienon of the multiplication of the individual is primitively closely connected with nutrition. Not only is nutrition the cause of the growth and consequent increase in size of the body, but it gives rise to a condition in which the organism converts the excess of nutrient material brought to it into the means for pro- ducing a new individual. In the lower forms, as in elementary organisms, a process beginning with gemmation, and leading on to fission, is a very common phenomenon. The manner in which multiplication is effected varies with the amount of material which is used by an organism in producing a new organism. The phenomena of multiplication by gemmation and spore-for- mation, which are so common in the lower divisions of the Inverte- brata, have some relations to sexual differentiation, which indeed does occur among the Protista. It is derived from a stage in which two similar germ-cells fuse to form a new organism (Conjuga- tion). As the two uniting cells become gradually dissimilar they become distinguished into egg- cells and sperm-cells; these are the morphological elements of the sexual reproductive matter throughout the whole Animal Kingdom, notwithstanding their numerous modifications, which are seen most markedly in the seminal cells. While the ovum retains its most essential characters, and can be recognised as such in every division, the seminal cell very early undergoes considerable changes. Like other cells it gets a flagellate process, which may be greatly developed, while the cell- body and its nucleus are so reduced that they ordinarily form a structure of no great size. In this way filamentous structures — spermatozoa — are formed from the seminal cell. Sexual Repro- duction, then, does not exhibit a real contrast to the asexual. § 46. We do not exactly know in all cases what are the relations between the place where the reproductive matters are formed and the early rudiments of the body, but from what has been observed THE OEGANS. 53 in certain Ccelenterata and Mollusca we may suppose that primi- tively the relations of the two are very different, for in these forms the ova are derived from the endoderm, and the sperm from the ectoderm. The endoderm is, therefore, the female, and the ectoderm the male germinal layer. But it is not yet known how far these relations obtain among the higher animals. As yet there are only uncertain indications, but these speak to a general agreement with the results already obtained. The parts of the body which are set apart for the formation of the sexual products gradually take on the form of glands. This is a further step in differentiation, and is connected with the local- isation of the function. In the simplest cases the two kinds of generative products are formed in special parts of the body, which function as sexual organs (reproductive glands) ; but these parts are not at first distinguished by any special characters. The organs which produce the semen are called testes, and those which produce ova, ovaries. Going a step further, we find the reproductive glands still further differen- tiated. In their simplest condition the products of these organs merely break away from the spot where they are formed, and pass into the digestive sac, or into the body-cavity, or even directly to the exterior. Gradually, however, ducts, which are often very com- plicated in character, are added on : it is probable that these ducts are not primitively connected with the germinal glands. Where these ducts can be seen to have any relations to other organs, these appear to be excretory organs (§ 40) which have entered into the service of the genital organs, and have been altered so as to correspond to this function. It becomes a great question whether the excretory ducts of the reproductive matter are not in all cases excretory organs. Receptacles which serve for the collection of the sperm are formed on the outlet-tubes (seminal ducts) of the organs which produce the sperm ; from the wall of these canals glands are differentiated, which secrete a fluid to be mixed with the sperm ; finally, there are arrangements for passing the sperm into the system of the other sex (copulatory organs). The differen- tiations of the egg-forming organ are no less varied ; the duct (oviduct) of the ovary is provided with dilatations, in which the ova get special envelopes, or are further developed. These portions of the oviduct are called the uterus. Special glands, Yolk glands, are formed from the ovary, and secrete a substance which is either taken up by the ovum or which merely forms an envelope for it. Appended organs receive the semen which is passed in copulation, and are known as receptacula seminis; lastly, other parts serve for the reception of the copulatory organ, or for the deposition or preservation of the ova. The relation of the egg-forming and sperm-forming organs to one another varies greatly, and must be considered from the stand- point of differentiation. In the lower divisions organs of both kinds are united with one another, sometimes in such a way that one 54 COMPAEATIVE ANATOMY. and the same gland produces both semen and ova (hermaphrodite gland). The ducts, also, are often more or less common to them "both. But in other forms the genital organ is divided, the products of its two parts being different ; testes and ovaries, that is, are present as separate organs, the excretory organs of which only are united more or less extensively; or each of them may have its separate orifice. All those animals which unite in themselves both kinds of reproductive organs are known as Hermaphrodites. A separation of sexes is apparently foreshadowed in various forms, by the alternating activity of the organs, at one time the egg- forming and at another time the sperm-forming organ exercising its function. The hermaphrodite stage is the lower, and the condition of dis- tinct sexes has been derived from it. This change is due to the decrease in size of one or the other organ, so that hermaphroditism is the precursor of sexual differentiation. This differentiation, by the reduction of one kind of sexual apparatus, takes place at very different stages in the development of the organism, and often when the sexual organs have attained a very high degree of differentiation. In these cases ontogeny exhibits the two kinds of organs primitively united, and so causes the individual to be hermaphrodite at a certain stage in development. The separation of the sexes affects the whole of the organism, for it produces a series of changes in each sex, which affect organs that had primitively little to do with the sexual function. Sexual differentiation is completed when the two kinds of organs are given over to different individuals. Thenceforward for reproduction, not only two different substances, semen and ova, and two different organs for producing them, are necessary, but also two individuals ; these are distinguished as male and female. Changes in the Organs. Development and Degeneration. §47. The result of the continued differentiation of a given organ is a complication by which the organ is removed proportionately further from its primitive condition. As the primitive condition is the lower differentiation, it entails a perfecting corresponding to a higher con- dition. This is clear on the principle of division of labour, which is the cause of all differentiation (cf. § 12). In obedience to this law a function can be the more perfectly carried out, the more exclu- sively the organ is related to that function. The more an organ is exercised for one function only, the more suitable are the conditions for its development in one direction, for there is no competition with DEVELOPMENT AND DEGENEEATION. 55 other directions of development. A limb which is a gill too, that is which has both locomotive and respiratory functions, is of a lower grade than an arrangement resulting from a division of the two functions, in which a part separated off from the appendage repre- sents a gill, and the rest an organ of locomotion. When the functions are united, locomotion is necessary for respiration, but when they are separated, they are independent of one another, and respiration is effected without the aid of locomotion, by the development of special organs for changing the water, these organs so far taking the place of locomotion. In both organs the independence which is necessary for their further development in one direction is gained. The organs of the body are not always developed to the same extent. One or another often remains in a lower condition, and so retains its more lowly character in an otherwise highly differentiated organism. It is not therefore wise to draw any conclusions as to the extent of the differentiation of single organs from that of the organism itself; it is better to judge organs by comparing them with equivalent organs in other organisms. The real factor in the development of an organ by differentiation must be sought for in the increased or modified function of the organ in the struggle for existence, that is in its adaptations to the ex- ternal conditions of life. It is hence that transmission acquires its importance, since it not only causes a perpetuation of inherited characters, but is enabled to effect an elevation in those characters. Degeneration or reduction is another constant phenomenon which is dependent on differentiation, inasmuch as it presupposes it. Its result is, in itself, the exact opposite to that of differentiation. For while differentiation is the cause of complications, reduction is the cause of simplifications of the organism, and is therefore the cause of organs or of organisms passing to a relatively lower stage. With regard, however, to the general organism, and its relation to other organisms, it produces the same effect as differentiation, for it leads to variety in form. Reduction, like differentiation, varies in degree; it may affect separate portions of the body, or groups of organs, or finally the whole of the body. It is different, again, according as it affects the individual, the species, or the genus. In one case it may be seen to be a definite process, in another a condition, which can only be assigned its place as one of the several stages of such a process by the aid of a comparative series of allied forms. It may affect organs in two different ways. The affected organ may be in- dependent of the general arrangements which obtain in the de- veloped organism to which it belongs, and then reduction has but a transitory or provisional signification. Reductions of this kind during the course of the development produce simplifications, but as the differentiation which is going on in other parts may be producing new and higher organs, this reduction does not hold the organism back, but is a cause rather of the progress of differentiation in another direction. The reduction of parts which belong to certain 56 COMPARATIVE ANATOMY. developmental stages in the individual are examples of this kind of reduction (Larval organs). (Cf. § 5.) The other kind of reduction affects organs which belong to the developed organism or its rudiments. It may affect the fully formed and completely functional organ as well as one just laid down, and in the primary state of differentiation. The process of reduction is seen therefore in various degrees of intensity. The process is often difficult to perceive amid the various other processes of differentiation which are affecting the rest of the organism when the organ affected is only just making its appearance : the further, however, differentiation has gone, the more striking must the process be. The reduction of an organ is necessarily connected with its function, a change in which must be regarded as the cause of the reduction. Loss of function produces retrograde changes in an organ, but of course neither process is a sudden one. Although reduction is, on the whole, the cause of the simplifica- tion of an organ, and therefore of the organism also, it is not a phenomenon which makes the organism absolutely lower in degree. Reduction may rather lead to a higher differentiation, as it does when larval organs are removed ; it may give rise to higher forms even in whole series of organisms derived from one another, by facilitating the higher development of those not affected by it. In this case again reduction precedes differentiation. This is strikingly seen in the numerical relations of parts, which become individually more perfect as they diminish in number. As reduction is a gradual process, the organs which are affected by it may be met with in various stages. These rudimentary organs are most significant indications of genetic relations, while they at the same time show us how an organ which has lost its primitive function, and which may even have no intelligible signifi- cation as regards the purposes of the organism, may persist for a very long time before it completely disappears. (Cf. supra, § 6.) Reduction may affect every organic system and be observable in every part of it. It is expressed in the form as well as in the size and number of the parts, and even in their histological characters. Its conditions are to be sought for, first of all, in the relations which alter the organism. According to the number of organs affected, reduction is more or less manifest in the organism as a whole. Inasmuch as comparison everywhere reveals to us evidence of either progressive or of retrogressive change, we may regard the organism as a thing caught in the act of moving, as arrested in the midst of a career through the most diverse ranges of form. The changes of the various organs, and the phenomena which are observed in the elementary structure in the cell, are what make up this movement. COKEELATION. 57 Correlation of Organs. § 48. The changes in the organism which are due to differentiation and reduction are the cause of a fresh series of phenomena in the factors which gave rise to them. From the conception that life is the harmonious expression of a collection of phenomena regularly conditioned, it follows that the activity of an organ cannot be regarded as really existing for itself alone. Every kind of arrangement presupposes a series of other arrange- ments ; every organ, therefore, must have intimate relations with the rest, and be more or less dependent on others. Every move- ment in a muscle presupposes the existence of a nerve ; and both of these organs presuppose the existence of a nutrient system. In this way one function has an intimate connection with other apparently dissimilar functions. This relation, which was first defi- nitely pointed out by Cuvier, and which is known as Correlation, shows us the road by which we can attain to a correct appreciation of animal organisation. By far the most important point is the conception of the organism as an individual whole, which is as much conditioned by its parts, as one part is conditioned by others. Correlation is a necessary result of this conception. Not only the general arrangements of the organisation, but also its apparently more subordinate features, exhibit intimate relations with one another, and a change which affects one system of organs, simultaneously produces modifications in some of the other organs. These are adaptations to changes, which themselves are due to adaptations. They are, however*, of a secondary character, while those which are of a primary character have their origin in the outer world. Correlation may be divided into the more and the less remote ; where less remote it is expressed in one system of organs or in other systems functionally connected with this ; when more remote, in organs which are functionally less related to it. Physio- logical principles are essential in the investigation of correlation, and it is necessary, therefore, to know what are the functions of separate organs, or at least what their value is in the animal economy, in order to be able to recognise it. So, too, it is of im- portance to know what are the habits of the animal, for the original forces, on which the various relations of the organs depend, are due to them. As the forces which cause changes in the organism either lie without the organism, or, as most of them, are to be sought for without it, they do not come within the scope of our work. Com- parative Anatomy, therefore, is limited all round by a wide but uncultivated region, in which rich harvests may be gathered for biological science, whenever its treatment is taken in hand. 58 COMPARATIVE ANATOMY. Fundamental Forms of the Animal Body. § 49. Owing to the infinite variety of the external characters of animal organisms, it is necessary to seek for fundamental forms to which this variety may be referred. We must also ascertain the con- ditions which influence and give rise to the most important modifi- cations of these forms. These results may be obtained in different ways. We will choose the shortest by commencing with the lowest stage of the animal organism. This is the stage which the Gastrula form presents to us ; by its wide distribution this form will provide us with the characters which are best adapted for our purpose. An organism at this stage is some- what spherical or oval in shape, and the mouth will be found at a point on the surface. If we imagine an axis (Fig. 16, AB) drawn straight through the digestive cavity, the pole corresponding to the opening of the mouth represents the oral, and the opposite the aboral pole. This axis (AB) is the primary axis of the body. In a body of a regular cylin- drical or spheroidal shape we can imagine as many lines as we please drawn through the body perpendicular to this axis. (Secondary axes, a b, c d.) In this instance they are all equivalent. The secondary axes are in this case indifferent to one another, and are cha- racteristic of a lower condition. The organism, either when moving freely in the water, or when fixed (by the aboral pole, of course), as it afterwards is, is differentiated by the develop- ment of a certain number of secondary axes, their development having relation to the main- tenance of the balance of the body. We here, then, have to do with a statical cause. The development of the organism along its secon- dary axes takes place through the development of external appendages, tentacles and the like, or through differentiation of the enteric cavity, or through the laying down of other organs (such as the generative glands) in the direction of those axes. In consequence all the conceivable secondary axes are no longer equivalent. Those along which organs are differentiated are distinguished from the rest. They have passed, in fact, from the previously indifferent condition to a differentiated one. The so- called radiate fundamental-form of the body which is commonly found in the Ccelenterata, arises in this way, as may be seen by Fig. 16. Diagram of the axes of the body. A B Primary axis, a b, c d Secondary axes. The lower figure is a trans- verse section of the upper one, showing its two secondary axes. FUNDAMENTAL FOE MS. 59 studying the relations of the axes to one another, as explained above (cf. Fig. 17). The importance of the mouth to the organism causes the differen- tiations which obtain around it to have a special value. These differentiations are de- veloped as tentacles of various form, and cause the parts around the mouth to be much more varied in cha- racter than those at the aboral pole. If the body grows in the direction of its primary axis, without becoming attached to the ground, the axes may acquire modified importance if locomo- tion in the direction of the animal's length be established. The pri- mary axis will remain as before, but the secondary axes will necessarily differ ac- cording" to the sig'nifi- cance of the surfaces which they connect. When one and the same surface always touches the supporting object, it becomes the ventral surface, and the oppo- site one becomes the dorsal. These two surfaces, the dorsal and the ventral, are placed under different conditions, and must therefore be differentiated, in different ways, while the two sides, or — when the body is perfectly flattened out — the two lateral edges necessarily come to differ in character from the dorsal and ventral surfaces. Such cases are instances of the development of only two inequivalent secondary axes. One connects the ventral and dorsal surfaces, and is the dorso-ventral axis (Fig. 18, a b), the other connects the sides (c d) of the body, and is the transverse axis. The surfaces which correspond to the poles of the first or dorso-ventral axis are not, while those which correspond to the potas of the transverse axis are, equivalent. A primitive condition which has disappeared in the dorso-ventral axis in consequence of the differen- Fig. 17. Kadiatc fun- damental form ; letters as in Fig. 16. The an- terior surface of the body is seen in the lower figure, and shows the appendages (tentacles) which are differentiated along the two transverse axes. Fig. 18. Diagram to show the differentiation of the secondary axes. In the upper figure a cephalic portion is indicated by the development of a pair of dorsal tentacles. The lower is a transverse sec- tion of the upper figure, and the secondary axes are consequently seen in it. 60 COMPARATIVE ANATOMY. tiation of its surfaces, is retained in the case of the transverse axis. This, the second form, which can be derived from the Gastrula, and is ordinarily known as that of bilateral symmetry, first appears in the Vermes, and prevails in all the divisions above them. When the secondary axes of the body retain their primitive indifferent character, we may imagine that there are as many similar pieces in the architectural composition of the body as there are possible secondary axes. But when the secondary axes become differentiated, the divisions of the body take on a definite numerical relation. They are known as antimeres. If two secondary axes become differentiated, and are like in character, we have four antimeres, for we can divide the body into four similar parts, along these secondary axes. But when two unlike secondary axes are differentiated the body is only made up of two antimeres ; two halves of the body, distinguished as right and left, are the parts corresponding to one another. In this way the eudipleural funda- mental form is developed. § 50. The differentiation which marks off the oral from the aboral pole, and which has been already mentioned, gives a higher significance to the former region of the body. This differentiation asserts itself in other forms, as in the primary radiate form, and in very various ways. It is not only the presence of the mouth, which favours the differentiation of organs around it, as organs for aiding in the prehension or ingestion of food, but the greater significance of the anterior end of the body in locomotion is also a cause of differen- tiation. This part takes the initiative. It has to show the way to the rest of the body, and often indeed, to lead it ; it meets with a thousand foreign objects, which it has to examine, to follow, or to avoid. It is therefore exposed to external influences other than those which act on the opposite end of the body. The dignity of the relations of this region explains how it is that the mouth is not by any means always at the anterior end of the body, and that it much more frequently is placed close to or even altogether on the ventral surface, without the anterior end of the body being less highly developed. The high specialization of the anterior region is caused principally by the development of various kinds of sensory organs, and therefore of organs which put the organism into relation with the outer world ; the region moreover often has various organs of defence connected with it, and with it is closely connected the development of the central nervous system. The whole region thus gets a higher value in comparison with the general organism, for it shelters and carries the organs which elevate and rule the latter. This anterior region of the body is therefore called the head, or chief portion. The differentiation of a head seems to depend primarily on the position of the mouth. This directs the course of movement, and to this, as a secondary cause, the anterior part of the body owes its METAMERIC SEGMENTATION". Gl various distinctions. The appearance of a head is at the same time a result which affects the whole body, for the body can now be divided into two portions at least, which differ in character. Metamerism of the Body. § 51. The planning out of the individual organism as a single struc- tural entity is only characteristic of lower conditions of development, whether permanent, as in nearly all Ccelenterata and in the lower classes of Worms, or transitory, as in the higher divisions of the Animal Kingdom. Simultaneously with the growth of the body to a considerable length, we observe the beginning of the division of the organism into ' separate segments, following one on another, noticeable externally through the occurrence of separating constric- tions, or through the regular distribution of appendicular structures or processes of the body, internally represented by the arrangement of the organs in the distinct successive compartments of the body. "We term this segmentation of the body Metamerism; the separate segments are me tarn ere s. The metamerism which thus breaks up the body is only a further example of differentiation. From the primitive homogeneous indifferent body a heterogeneous, diversified body is developed, and the separate metameres differ from one another ; not only are they something new in comparison with the earlier condition, but they are also — notwithstanding their resem- blance one to another — different from one another, owing to the position which each occupies. Metamerism is not in all cases, where it is perceptible, exhibited with equal clearness. Sometimes it is apparent in this or that organ, or system of organs, more than in another, and, again, in other organs may be altogether wanting. It is easy to recognise very various conditions of the commencement and of the incomplete carrying out of the process. Where we find it in fullest develop- ment it dominates the whole organism, and is exhibited in all organs ; so that each metamere possesses its individual system of organs, and particular systems of organs common to all metameres present a special differentiation of their structure in each metamere (ventral ganglion chain). In this manner the organism becomes built up of many component parts. And hereupon we have to take note of conditions in which independent importance is acquired by the metameres. Little by little a metamere, in virtue of the elaboration of its own set of organs, ceases to be dependent on the total organism, emancipates itself from the commonwealth, and gains the capability of leading a free existence. To this many phasnornena are traceable, which are usually called gemmation (Worms). 62 COMPARATIVE ANATOMY. § 52. An efficient cause for metamerism may be sought, as has beeu above indicated, in the phenomena of growth. We can imagine a repetition of local outgrowths, resulting in practical advantage to the organism, taking place in particular systems of organs simultaneously with the elongation of the body. In this way the external meta- merism may be brought iuto connection with the movement of the body, which was perhaps the earliest cause of this phenomenon. Many facts point to its being so. In any case there are numerous examples of the gradual elaboration of metamerism without all systems of organs being at once affected by it. Metamerism has, however, a less doubtful origin in its connection with gemmation, which is itself a kiud of growth. It seems, indeed, in many cases, as if gemmation led to metamerism, in such a way that the metameres represent buds, which remain connected with the organism, and only in some cases attain to a higher stage of individual existence. Numerous instances of incomplete metamerism prevent us, however, from attributing a general significance to this process, and it cannot in any sense be regarded as the sole cause of metamerism. Metamerism leads to perfection of the organism. By it the organism is enabled to get a larger number of organs, although, indeed, they are at first mere repetitions of one and the same arrangement. As the separate segments become more independent their action becomes more free, till at last the differentiation of a larger number of separate organs gives a larger scope for action. Differentiation, then, gains ground in every part, and alters the organs of the separate metameres in different ways, according as their functions become more various. By the development and reduction of metameric organs the metameres get to differ in value, and become differentiated themselves ; this differentiation is expressed externally by the difference in their size and form. This leads to the disappearance of the primitive similarity of the metameres. The amount, too, of their independence may be lessened, and a number of primitively separate metameres may gradually fuse into larger divisions. This gives rise to complexes of metameres, in which the fact of their being composed of separate units of the body is only suggested, and that often faintly ; sometimes a large, sometimes a small number of segments undergo concrescence. This, too, is on the whole a cause of differentiation of the organism, as the body consists in consequence of some independent and of some fused metameres. Finally, metamerism may altogether disappear, and the presence of separate organs alone indicate, and that often obscurely, the phenomenon which obtained in the primitive state. Every stage in metamerism is therefore a source of variation in the external and internal organisation of the organism. HOMOLOGIES. 6 Q Comparison of Organs. § 53. The variations in organisation among the various larger and smaller divisions of the Animal Kingdom are such as to lead us, at first sight, to perceive the points of difference rather than those of agreement. And this is more marked in proportion to the diver- gence between the particular divisions compared. It is, however, the business of Comparative Anatomy to follow out the changes in the organisation, and to discover what is " similar" in the changed and metamorphosed forms, however deeply hid it may be. An organ may be " similar " to another in one of two ways. Either in its functional relations, that is from a physiological point of view ; or in its genetic and therefore anatomical relations, that is from a morphological point of view. These two relations of an organ must be kept well apart. The change of function in one and the same organ, as well as the similarity in arrangement of organs which are morphologically very different, compels us to ascribe a subordinate value to physiological relations, when we are comparing organs. The gills of a Fish, of a Crab, and of a Cephalopocl, are organs of respiration, and have many structural points in common ; yet they are very different organs morphologically, as the relations of each of the three to the whole organism shows. By insisting on similarity of function, we bring together organs which are morpho- logically different, and so turn aside from the object of Comparative Anatomy. We distinguish, accordingly, physiological likeness, or Analogy, horn morphological likeness, or Homology, and only consider the proof of the latter as our task. The smaller the division to which the objects of comparison belong, the more obvious is the homology. Homology therefore corresponds to the hypothetical genetic relationship. In the more or the less clear homology, we have the expression of the more or less intimate degree of relationship. Blood-relationship becomes dubious exactly in proportion as the proof of homologies is uncertain. It is impossible therefore to say exactly how far homology extends throughout the Animal Kingdom. As a matter of fact, numerous investigations have discovered a larger number of homologous arrangements even in otherwise divergent groups, and have thereby extended the boundaries of homology further than was formerly thought possible. In consequence of the existence of various possible modes of morphological agreement, homology is divided into two primary groups: General and Special Homology. G4 COMPAKATIYE ANATOMY. § 54. I. General Homology is under consideration when an organ is referred to a category of organs, or when a single organ com- pared with another is taken merely as the representative of such a category. These categories always consist of several organs or parts present in the body. When we compare the body- segments of an Annelid, the vertebra, or the appendages of an animal with one another, we lay the foundations of general homology. This again consists of several subdivisions, according to the kind of category which is made use of in the comparison. 1) Homotypy has reference to organs which are fellows to one another, such as the organs of the two sides of the body ; the right kidney is homotypical with the left, and the right eye with the left eye, and so on. Whilst these examples may not show the necessity for the formation of this division, it should be noticed in addition that homotypical organs have not always the same characters. They are often so changed that their homotypy cannot be recognised, and has to be worked out. 2) Homodynamy (equivalent to the general homology of Owen, and partly also to his serial homology) subsists between parts of the body which are affected by a general morphological pheno- menon serially expressed in the organism. Homodynamy is dis- tinguished from the next subdivision by the fact that the parts in question are arranged along the long axis of the organism and define its type. The metameres therefore are homodynamous parts ; as are the segments of the Arthropoda, the primitive vertebras of the Vertebrates, etc. 3) Homonomy. This describes the relation to one another of those parts which are arranged along a transverse axis of the body, or in one segment only of its long axis. The rays of the pectoral and pelvic fins of fishes, the individual fingers and toes of the higher Vertebrata are homonomous structures. Besides these there are other subdivisions of general homology distinguishable, which are however of very subordinate importance. § 55. II. Special Homology, Homology in the restricted sense. This is the name we give to the relations which obtain between two organs which have had a common origin, and which accordingly have also a common embryonic history. As exact proofs of genetic relations are necessary for the investigation of special homologies, this mode of comparison is generally limited in the lower divisions of the Animal Kingdom to systems of organs ; it is only in the Vertebrata that it is possible to extend this method to more minute features. Thus among the Vermes or the Mollusca we can hardly indicate, with any certainty, particular parts of HOMOLOGIES. 65 the alimentary canal as homologous ; while in the Vertebrata we can confidently assert that even such unimportant structures as the casca of the intestine from the Amphibia onwards are homo- logous. The homologies of the parts of the skeleton, which are thu organs that have been investigated with the greatest exactness, are those which can be most definitely recognised. It is a large part of the main task of Comparative Anatomy to prove special homo- logies. Special homology must be again separated into subdivisions, according as the organs dealt with are essentially unchanged in their morphological characters, or are altered by the addition or removal of parts. I therefore distinguish — 1) Complete Homology, when the organ referred to is unchanged in position and connection, and is still perfect however much modified in form, size, and various other points. This kind of homology is generally found within the limits of small divisions, less often in larger ones. For example, the bones of the upper arm from the Amphibia to the Mammalia, the heart of the Amphibia and Reptilia, etc., exhibit complete homology. 2) Incomplete Homology. This consists herein, that an organ which is otherwise completely homologous with another, has other parts which are wanting in the latter added to it, or conversely, when an organ is wanting in some essential part in comparison with another organ. The heart of the Vertebrata may serve as an example. The organ is homologous throughout the division from the Cyclostomi onwards, but the homology is incomplete; for in Fishes a part, the venous sinus, which in the higher divisions is taken into the heart, and which in the Mammalia is absorbed into the right auricle, lies outside the heart. The homology between the heart of the Fish and of the Mammal is consequently incomplete owing to addition. In another case it may be incomplete owing to diminution. The reverse of the previous case may serve as an example, were it allowable to regard the heart of the fish as a reduced one. An example is presented by the pectoral fins of fishes. The skeleton of this organ in the Ganoidei or Teleostei is, owing to reduction, incompletely homologous with that of the Selachi. Parts have in this case disappeared which did primitively belong to the organ, just as in the former case parts, which although they were primitively present did not belong to the oi'gan, were added to it. Systematic Classification of the Animal Kingdom. § 56. In the general organisation of every animal we recognise a number of arrangements which it has in common with a greater or less number of other animals. These relations are partly of a more p 66 COMPARATIVE ANATOMY. general nature, affecting the position or arrangement of the most important systems of organs, and partly they affect the special development of individual organs; they extend to agreement in form, size, and number. The classifying spirit of man has formed definite conceptions of these relations of organisms to one another. All those individuals which agree in essential points he has called a species, and has united into a genus those species which resemble one another in a number of points ; these again he has united into larger divisions, families, orders, and classes. Thus arose the zoological system, which, in so far as it unites what agree, and separates what differ, has come to be the expression of our general knowledge of the Animal Kingdom. In this way the whole Animal Kingdom can be broken up into several large divisions, each of which differs from the rest by a number of special characteristics. The essential character may be recognised in all the subdivisions, and even under great individual variations. This has been called the "type." The type then means a collection of characteristics which are expressed in the organism, and which are predominant in a large division of the Animal Kingdom, and which are evident in the course of develop- ment as well as in the adult condition. Such larger divisions which differ from others in certain fundamental points of organisation are themselves called " types." Within each type we note a variation in the characters of the divisions which make it up, and this often to such an extent that what is characteristic of the type appears to be lost in some forms. In this case it is always individual development which enables us to recognise the connection of these forms with the " type." If we admit that similarity of organisation in different individuals is explicable by the fact that they have a common ancestor, and that therefore these similarities are due to affinity, we must regard less close similarities as due to a less close relationship. We therefore regard the individuals which belong to one species as more closely allied than are the representatives of different species, and within the limits of one species we shall again derive from common ancestors those individuals which are distinguished by special characters, and which we unite into a sub-species. No one has any hesitation in recognising within the limits of small groups of individuals the phenomenon of the continuation of the peculiarities of a given organism into other individuals by means of transmission ; indeed it is often possible to perceive, by direct ob- servation, that descendants are like their ancestors. By extending this conception of affinity to a wider circle, and regarding what is common in organisation as due to a common descent, and what is divergent as due to adaptations, we take our stand on the theory of Descent (cf. §§ 4 and 5). We consequently regard the large divisions known as "types," as phyla, or leading branches of the genealogical tree, and by so doing point to the cause which has determined their existence. CLASSIFICATION. 67 Within one phylum a form of animal organisation is evolved along the most varied lines, which gradually lead from the simple to the more complex, and from the lower to the higher. The categories which we distinguish as species, genera, families, orders, and classes are due to continued differentiation. These subdivisions correspond to the ramifications of the branch, and in them the divergence of character is expressed. The differences between the classes, orders, and so on, are so great that they do not seem to have any connecting links at all, but we must take into consideration the fact that in living forms we have before us only the final offshoots of developmental series of organisms, which have been ramified in very various ways, and which lived in earlier and often very far-distant periods, and which have gradually disappeared. The paheontological record proves this partly, though it may be but very slightly. In the strata of the earth remnants of forms which have disappeared, and which were the predecessors, and, in fact, the direct ancestors, of later living organisms, are preserved. Inasmuch as the living forms are but a small portion of the whole world of organisms, which has existed in the course of geological periods of development, we cannot expect that far-distant connections should be always evident, the inter- mediate steps determinable, and the genealogical connection made clear and indubitable. It is necessary to try and put the whole together out of fragments and to find lost traces of continuity. The most important part of the business of Comparative Anatomy is to find indications of genetic connection in the organisation of the Animal body. Following out this conception we have to represent to ourselves a developmental series of organisms arising in each phylum from a primitive form, which has been, during geological development, differentiated into many branches and twigs, most of which have disappeared at different periods, while some, greatly changed though they may have been, have lived on until to-day. The general character which has remained in these various stages of differentiation, and has been transmitted, with modifications, from the stem-form, is what is typical in the organisation. § 57. It is not always possible to prove, to the same extent, in all of the large divisions which are regarded as types, the common ancestry of the forms which belong to it. It is in fact very probable that several divisions have had. a polyphyletic origin, in which case the organisms which belong to them must be united together for reasons other than genealogical. Such divisions cannot be regarded as phyla. We meet with such relations in the lowest forms, in the boundary-territory between Animals and Plants. It is difficult to find a boundary line, for there are organisms which seem to belong to one as much as to the other Kingdom according to the phenomena f 2 GS COMPARATIVE ANATOMY. to which they give rise. The idea of a boundary line presupposes a rigidly-defined conception of Animal and Plant. The characteristic of the animal organism may be taken to be that differentiation affects the whole organism. This differentiation consists in its division into two layers, which have been already (§ 28) called ectoderm and endo- derm, and from which the germinal layers of the higher divisions are derived. But the exclusion of all the lower organisms, which do not undergo this division from the Animal Kingdom, would put out of our scope many phenomena which are of great importance, if we would understand animal organisation. Although it might be best to regard this world of lower and very varied organisms as a special Kingdom placed between the Animal and the Vegetable, and containing the beginnings of both, that of the Protista, yet we, as our work embraces the connections between animals and these lowest organisms, must enter into a consideration of them. We therefore unite a number of those divisions of the Protista, which arc more nearly related to animals than plants, as the Protozoa. As their genetic relations to one another are altogether unknown, the division which is formed by these organisms cannot be regarded as a " phylum." Nor is there a type common to them all. We there- fore regard them as lower organisms, which are the nearest of the Protista to Animals, and we must compare them, not with the separate divisions of the higher animal organism, but with them all together. This compels us to unite the latter into a single group, which has been called the Metazoa. The Protozoa and Metazoa are not so very sharply marked off from one another. Not a few of the Protozoa are known to be composed of a number of cells. It is the arrangement of cells in layers of definite physiological value which characterises the metazoic organism. This seems to take place very gradually, and at first the layers are incomplete. We find representatives of this in the parasitic Dicyemidas, which live in the so-called veinous appendages of the Cephalopoda, and deserve to be specially mentioned. A germinal cell gives rise to a number of cells by division, among which a single large cell becomes surrounded by a number of smaller cells, which form a continuous layer. The central cell represents the endo- derm, and is covered by the peripheral layer, which represents the ectoderm at all but one small spot. (Fig. 19.) The endo- Fig. 19. Fig. 20. Vcimi- dermal cell elongates considerably, and Gasfcrula form embryo of itg protopiasm becomes differentiated in stage of Dicyema typus A r T - . , Dieycma (after E. van various ways. It iorms the groundwork typus. Benecleu). of the elongated body, and remains covered by the ectodermal cells, which also grow, without however multiplying very much; these give off fine cilia, and form the protective and locomotor organs of the body, while THE DICYEMIDiE. 69 the endodermal cell in the axis of the body becomes an organ of nutrition, and takes on the function of repro- duction, for the germs of young forms are found in it of two different types. The organism of Dicyema is thus seen to be two-layered, and the two layers have different functions ; the inner one is morphologically least differentiated, for it consists of a single cell. It is not quite certain whether this corresponds to a primitive condition or no, for the parasitic life of the Dicyemidas may have been the cause of the reduction of a multicellular endoderm. But their development is quite easy to make out, and in it there is never more than one endodermal cell, so that the fact becomes more significant. Just as among the Protozoa the Ms multicellular forms are allied by intermediate steps to the unicellular, from which they have been developed, so among the Metazoa does Dicyema exhibit to us the commencement of the separation of the body into cell-layers, and although the arrangement is not as perfect as it is in the rest, yet it is in the same direction as that which in them arrives at full expression. Vax Beneden, Ed., Recherches but les Dicyemida?. Bull. Acad. Belg. xli., xlii., 1876. § 58. Passing" over the Dicyemida3, I recognise the following divisions of the Metazoa : 1. Ccelenterata. 2. Vermes. 3. Echinoderma. 4. Arthropoda. 5. Brachiopoda. 6. Mollusca. 7. Tunicata. 8. Vertebrata. Fig. 21. Vermiform embryo of Dicyema typus. n Nucleus of the endodermal cell (after E. van Bene- den). These divisions represent in a general way separate branches of the pedigree of animals, and each of them contains higher and lower forms in various proportion. But the degree and extent to which their organisation is developed is different in each of them. The divergence of organisation expressed in each division is indicated by their relations to one another, and it shows us how the lower forms of the higher phyla may have started from the lower phyla. These 70 COMPARATIVE ANATOMY. large divisions are therefore arranged in genealogical connection. The extent to which each division is separated from its fellows varies in each case. The relation of the various large divisions to one another is seen in the subjoined tree. Vertebrata Mollnsca Arthropoda Braehiopoda Eehinodermata Vermes Coelenterata Protozoa The more exact limitation of the separate divisions, as well as the reasons for the genealogical relations here merely indicated, will be found in the special chapters. BIBLIOGRAPHY. 71 Bibliographical Aids in Comparative Anatomy. § r>9- The principal work which is to be recommended as a scientific guide to the whole of Morphology, and especially to the questions which I have, in the foregoing paragraphs, treated very concisely, and which should be care- fully studied, is : Hackel, E., Generelle Morphologie tier Organismen. Allgemeine Grundziige tier Formenwissen- schaft, mechanisch begrtindet durch die von Ch. Darwin reformirte Descendenztheorie. 2 Bcle, Berlin, 1866. The following books also treat of Morphology philosophically: LEreKAKT, R., Die Morphologie und die Verwandtschaftsverhiiltnisse tier wirbellosen Thiere. Braunschweig, 1818. Caktjs, V., System tier thierischen Morphologie. 1853. Bronn, Morphologische Studien Uber die Gestaltungsgesctze tier Naturkorper. Leipzig und Heidelberg, 1858. a. Comprehensive Works on the whole subject of Comparative Anatomy : CrviEB, G., Leeons d'anatomie comparee recueillies ct publiees par DcmiSrii, et Dcvernot. 5 vols. Paris, 1708-1805. — Leeons, etc., recueillies et publiees par Dumeril. Scconde edition. 8 vols. Paris, 1S35-46. Meckel, J. F., System der vergleich. Anatomic C Bde. Halle, 1821-33 (incomplete— sexual organs wanting). Miene-Edwards, H., Leeons sur la physiologie ct l'anatomie comparee tie l'homme et ties animans. T. I— XII. Paris, 1857-76. Unfinished. Leydig, F., Vom Bau ties thierischen Korpers. I. Band. 1 Hiilfte. Tubingen, 1864. b. Text-books and Manuals of Comparative Anatomy : Carus, C. G., Lehrbuch tier Zootomie. Leipzig, 1818. Second edition, under the title of Lehrbuch der vergl. Zootomie. 2 Bde. Leipzig, 1834. Wagner, R., Handbueh der vergleichenden Anatomic 2 Bde. Leipzig, 1834. New edition, under the title of Lehrbuch der Zootomie. 2 Bde. Leipzig, 1843-48. (The second volume, containing the Anatomy of the Invertebrata, is by H. Frey and R. Leuckart.) V. Siebold and Stannius, Lehrbuch tier vergleichenden Anatomie. 2 Bde. Berlin, 1845. Second edition under the title of Lehrbuch der Zootomie. Band I. Heft 1-2, containing the Anatomy of Fishes and Amphibia, is all that has as yet appeared. It will be continued. Bergmann, C., and Leuckart, R., Anatomisch-physiologische Uebersicht ties Thierreiches. Stutt- gart, 1852. Schmidt, O., Handbueh der vergl. Anatomie. 7C Auflage. Jena, 1876. Owen, R., Lectures on the comparative anatomy and physiology of the Invertebrate Animals. Second edition. London, 1855.— Of the Vertebrate Animals. Pt. I. Fishes. London, 1846. Jones, Rymer, General Outline of the Organisation of the Animal Kingdom, and Manual of Comparative Anatomy. Fourth edition. London, 1871. Haeting, P., Leorboek van do Grondbeginselcn der Dierkunde in haren geheelen Omvang. Deel I — III. Tiel, 1861-74 (deals also with Comparative Anatomy). St. George Mivart, Lessons in Elementary Anatomy. London, 1873. (Introduction to human anatomy.) c Figures of the Structure of Animals : Carus, C. G., and Otto, Erlauterungstafeln zur vergleichenden Anatomie. 8 Hefte. Leipzig Wagner, R., Icones zootomies, Handatlas zur vergl. Anatomie. Leipzig, 1841. Schmidt, O., Handatlas der vergl. Anatomie. Jena, 1852. Carus, V., Icones zootomica3. Leipzig, 1857. First half (Invertebrata). Letdig, F., Tafeln zur vergl. Anatomie. Erstes Heft. Tubingen, 1864. 72 COMPARATIVE AXATOMY. d. Comparative Histology : Letdig, F., Lehrbucli tier Histologic des Mcnschen unci der Thiere. Frankfort, 1857. e. Ontogeny : Foster, M., and Balfour, T. M., The Elements of Embryology. Fart I. Macmillan and Co. London, 1874. KoLtiKER, A., Entwickelungsgeschichte des Mensclien u. der hoheren Thiere. 2« Auflage. 1 Hulfte. Leipzig, 1876. In addition to these works, numerous monographs, treatises and essays, contained in the transactions of academies, and other learned societies, or Journals of Natural History, Zoology, and Anatomy, should be consulted. [The English student should consult especially the " Journal of Anatomy " (Macmillan), dating from 1867, and the " Quarterly Journal of Microscopical Science " (Churchill), dating from 1853— E. R. L.] SPECIAL PART. PROTOZOA. First Section. Protozoa. General Review of the Group. § 60. I reckon among these certain divisions of those organisms which, owing to the simplicity of their organisation, are examples of the lowest grade* of living forms. The most essential character appears to be the absence of organs differentiated for the most im- portant functions. The imperfect limitation of this division is due to this negative character ; in it, nothing that is " typical " of all its members, either in the relation of the body to its morphological elements, or in its organisation, can be made out. The want of any histological differentiation is a reason for considering the organisms enumerated in this group, in company with others, which we are wont to regard as lowly plants, as forms of life which stand between the Animal and Vegetable Kingdoms. It is on this consideration that the plan of uniting all the lower organisms which cannot be regarded as Animals or Plants into the Kingdom of the Protista is based. If we recognise this conception, it appears unnecessary to form a division of the ". Protozoa." But the knowledge of the stages of organisation, which obtain in the Protista, are of such high value for the compre- hension of animal organisms, that to pass them over completely would not be in accordance with the aim of this book. I have, therefore, retained in this place the division of the Protozoa, and place in it a number of forms which, by the simple conditions of their organisation, and by the low grade of their differentiation, are well adapted for giving an idea of the group. First of all I exclude from it those forms which have not a nucleus, that is, which do not pass beyond the cytod stage. The development of a nucleus in the otherwise simple protoplasmic body 7G COMPAEATIVE ANATOMY. of the organism is, when compared with the cytod stage, the ex- pression of a considerable advance, and prevents us from uniting the forms that do possess it with those that do not, whatever other points of agreement there may be in their protoplasm ; even though it is quite clear that in these cytod forms (Monera) we have the beginnings of the higher stages. These beginnings vary greatly, correspond to separate divisions of more developed forms, and make it probable that the latter have had a polyphyletic origin. Of the groups of the Protista I regard the Rhizopoda, Gre- garinas, and Infusoria as Protozoa. There is no permanent limitation of the protoplasmic body in the Rhizopoda; their protoplasm sends out varying processes. The lowest division is that of the AmcebidaB, the organism of which is of the grade of a cell. The body is formed by protoplasm and a nucleus ; it is ordinarily naked, but it can temporarily surround itself with a capsule, or the capsule may be a permanent covering, open at one or two points. The organism communicates with the outer world by this opening, and by it can extend itself over its shell. If there are several nuclei present the organism forms a syncytium. The Foraminif era form the second division. It is very probable that they all have a nucleus ; and these organisms, therefore, are also similar to a cell. But the formation of a covering, provided with numerous pores, and often considerably complicated, is an indication of a tendency to higher development. The Heliozoa form a small group more nearly allied to the next division. The Radiol aria, finally, are distinguished from all o titer Rhizopoda by the possession of a "central capsule" within the body. The central capsule contains a number of nuclear structures ; these, indeed, render the Radiolaria similar to cells, but their body cannot be re- garded as a cell. Another course of differentiation seems to have affected them. Further, in some the entire capsular protoplasm contains separate cells, which are regarded by many as structures not belonging to the organism (yellow cells). The development of various kinds of supporting organs gives a peculiar character to the Radiolaria. We can make out a large number of axes in the body, by the aid of these skeletons. The second division of Protozoa is formed by the Gregarina). An outer limiting portion surrounding the nucleus, that is a body of the grade of a cell, is wanting in the earliest stages only. They pass, therefore, through the Cytod-stage. In the mature organism an envelope, differentiated from the protoplasm within, can be made ont ; and they give indications of higher differentiations of the subjacent layer of protoplasm. An anterior portion is in many separated by a constriction from the cylindrical or band-like body. The Infusoria form the third large division. I do not, as do many, include the Flagellata among them. The body, formed of protoplasm, has its external form defined by the differentiation of a cortical' layer. This cortical layer surrounds the more indifferent protoplasm, which may in many cases be soon to be rotating, and so PEOTOZOA. 77 calls to mind the streaming* of the protoplasm in certain vegetable cells. A nucleus, which may bo various in form, shows that the body of the Infusorian is equivalent to a cell. In some there are several nuclei. A higher degree of potentiality is expressed by the differentiation of various histological structures in the cortical layer. But this does not affect the character of the Infusoria as unicellular organisms, so long as they have only one nucleus, if we may suppose that the cell is no longer in its indifferent condition. In many forms there is a small nuclear structure, the nucleolus, in addition to the nucleus. The Infusoria are divided into the Suctoria (Acinita) and the Ciliata. The former have definitely arranged fine pro- cesses, capable of a small amount of movement, which serve in the ingestion of food. The Ciliata are distinguished by an invest- ment of cilia ; and subdivisions arc formed according to the way in which these cilia are distributed over the body. Bibliography. Rhizopoda : Auerbach, C., Zeitschr. f. wiss. Zool. Bd. VII.— Dimardin hi Ami. sc. I. III. IV.- Schultze, M., Ueber den Organismus der Polythalamien. Leipzig, 1854. — Carpenter, W., Researches on the Foraminifera. Phil. Tr. 1856, 1859. — The same, Introduction to the study of the Foraminifera. London, 1862. (R. S.)— Huxley, Th. H., On Thalassicolla. Ann. nat. hist. 1851. — Muller, J., Abhandl. der Berliner Acad. 1858. — HXckel, E., Die Radiolarien. Eino Monographie. Berlin, 1862.— ScHULZE,F.E.,Rhizopodenstudien. Arch. f. mikr. anat. Bd. X— XII. — Hertwig, R., Arch. f. mikr. anat. Bd. X. Suppl. — The same, Zur. Histolog. der Radiolarien. Leipzig, 1876. Gregarinse : Stein, Ueber die Natur der Gregarinen. Arch. f. Anat. u. Phyl. 1818. — Kollikeh, Beitr. z. Keimtniss niederer Thiere. Zeits. f. Zool. I. — Lieberkuhn, EVolut. des Gregarines. Acad. Roy. de Belgique. Mem. des Soc. etrangeres. T. XXVI. Ed. van Beneden, Rech. sur l'evolut. d'es Gregarines. Bull, de lAcad. royale de Belgique. 2me Ser. T. XXXI. Sur la Struct. des Greg. Ibidem. T. XXXIII. Infusoria : Ehrenberg, O. G., Die Infusionsthicre als vollkommene Organismen. Leipzig, 1838.— Dujardin, Hist. nat. des Infusoires. Paris, 1811. — Stein, Fr., Die Infusionsthierc auf ihre Entwiekelung untersucht. Leipzig, 1851. — The same, Der Organismus der Infusionsthiore. I. 11. Leipzig, 1859-66. — Claparede, E.~, et Lachmanx, Etudes sur les Infusoires et les Rhizopodes. Geneve, 1S58-61. — Engelmann, Th. W., Zur Naturfreschichtc der Infusionsthiere. Leipzig, Zeitschr. f. Zool. XI. — Morphol. Jahrb. Bd. I. — Hackel, Z. Morphol. d. Infusor. Jen. Zeit- schrift VII.— Botschli, Archiv. f. mikr. Anat. IX.— Zeitschr. f. w. Zool. XXVIII.— Hertwig, 11., Ueber Podophrya gemmipora. Morph. Jahrb. I. § 61. As the body of the lowest organisms is formed of contractile protoplasm, which changes continually in form, there is no definite boundary to the body, nor any kind of differentiated integument. We see the bodies of most of those Protista, which are not provided with an envelope, alter in contour, just like the indifferent cells of higher organisms ; processes of the protoplasm are extended from different points at different times, and the rest of the body flows after them. Thus, as the body moves, there is always a change in its surface, so that a particle of its substance, which at one moment is found in the interior, may at another moment enter into the formation of a process. The processes, Pseudopodia, have some- 78 COMPAKATIYE ANATOMY. Fig. 22. An Amoeba figured at two different moments during move- ment, n Nucleus, i Ingested food. Some vacuoles may also be noted. / /, times tliu appearance of broad lobatc prolongations (Fig. 22) separated from one another "by shallow depressions ; sometimes they flow out as delicate, sometimes ;as wedge-shaped, currents, which divide in all kinds of ways at their periphery, and so form ramified processes. These changes are always within definite limits of form in the various divisions, so that the form of the pseudopodia is the expression of the first differen- tiation of a definite morphological character of the protoplasm. The pseudopodia characterise the Rhizo- poda, the superficial protoplasm of which is able to emit them in every form of this group (Fig. 23). Neigh- bouring pseudopodia can unite with one another at any point and in various numbers (Fig. 23, x)} or may even become connected in a retiform manner. This character of the protoplasm is '•••„ V\ }.'•;.:/..' Mi' ■■■•'/,- • / n°t affected by in- ternal differentiation (skeletal organs, etc.). It is the expression of a stage in the lowest living material when it is not differentiated at its periphery. When the outer- most layer of the body is hardened, pseudopodia cannot be formed at every point. The chemico- physical changes iu the peripheral parts lead to the formation of something unlike the rest of the proto- plasm, which retains its indifferent character, and indeed still gives signs of its power of movement, but is restrained by the firmer cortical layer from any considerable exten- sion of itself. This stage is found in the Gregarime; characters which obtain in many of the Amoeba) exhibit intermediate steps towards it. A firm homogeneous membrane sometimes delicately striated, extends in these forms over the whole body, which is formed by a single cell. It passes directly into the soft protoplasm, and appears to be a cuticle differentiated from it. Like all cuticles, it I I ■.( \.i ' Fig. 23. A Foraminifer (Eotalia) with extended pseudopodia, which pass through the pores of the multiloculate shell. At x, several pseudopodia have united together. PKOTOZOA. 79 lias no contractility ; but it is extensible and elastic, and is thus able to follow the contractions and expansions of the protoplasm. In addition to this separation of the cuticular layer there is also, in the Gre- / garinas, a cortical layer separated off from the internal parts, which appears to be finer than the richly-granulated protoplasm ; this is also found in the Infusoria. Fig. 24.-. GregariuEe from the enteric canal of Opatrum sabulosum ; a is the younger stage, provided with a " pro- boscidiform" process, a An- terior portion, b Posterior portion of the body, c Nu- cleus. § 62. With the separation of the body into an external cortical layer, and an internal par- enchymatous substance, further metamor- phoses of the cortical layer are connected. Of these the first that should be mentioned are the cilia which are widely distributed in the Infusoria. They appear to be direct but actively motile prolongations of the in- tegument : if combined with a cuticle they traverse it. They either beset a limited part of the body only, as the so-called oral open- ing*, or they are extended over larger tracts, or over the whole body often, with great regularity. According to the definite distribution and arrangement of these cilia, the Infusoria have been subdivided into Holotricha, Hetero- tricha, Hypotricha, and Peritricha. It is clear that they are differ- entiations of the protoplasm, from what happens in some other groups of the Protista, where they form temporary structures only, and, like the pseudopodia, can be again withdrawn into the proto- plasm of the rest of the body. The flagella, as well as the undulating membranes, which are often formed in the region of the mouth of many Infusoria, are modifications of the cilia. The cilia appear sometimes to be modified in a peculiar way to form stiff processes, movable only at their point of attachment to the body (Stylonychia); sometimes, indeed, they are flattened and broadened. The cilia, as well as the style-shaped processes, serve as locomotor organs, and show us that locomotion is connected with the integu- ment, just as it was connected with the temporary external layer of the body, where pseudopodia were formed. Another structure observed in the cutis of many Infusoria (e.g. Paramecium), are firm rod-like bodies (Trichocysts), which, under certain influences, emit a fine stiff filament. These structures lie close to one another in the cortical layer, and in a direction per- pendicular to the long axis of the body. They call to mind the stinging cells of the Coelenterata, but they are not to be regarded as the same things, for they are not formed from cells. SO COMPARATIVE ANATOMY. & 63. In the cortical layer of the body of tlie Gregarinae, and of many Infusoria, there are indications of bands, or fibres, resembling muscles. In the Gregarinas these structures are arranged cir- cularly, or spirally, and form a layer just below the cuticle; it extends over a short part only of the " head/' which, as a rule, is separated from the body by a constriction, but it never passes into the wall of partition which separates this part from the body. Among the Infusoria these contractile bands are principally known in the larger species of some genera (Spirostomum, Stent or, Prorodon, etc.). In others they are absent. They sometimes run spirally, sometimes longitudinally. They are also present in the Vorticellinas, where they form spiral coils towards that end of the body which passes into the stalk. It is clear that these structures do not form the sole contractile system of the body, for those Infusoria which do not possess them are capable of executing powerful contractions. But that they are really contractile is shown by Spirostomum, in which the contractions of the body are not effected along its long axis but in the direction of the striated band, which describes several spiral turns. The contractile band, which runs in the interior of the stalk of the VorticellinaD, is a differ- entiation from the protoplasm of the same kind ; in Zoothammium it branches with the colony, in Carchesium it exists separately in each individual of the colony. Although these structures in many points resemble muscular fibres, and are physiologically the same, they cannot be compared with these histological elements from a morphological point of view, for neither cells nor the products of cells take auy share in forming them. They are differentiations from the protoplasm of the organism, whilst in the tissues of the Metazoa they are formed by the differentiation of a whole lot of morphological elements. The whole contractile apparatus cor- responds therefore to a muscular system in function only. The separate bands or stripes are merely analogous to muscles (myophana). § 64. The function of supporting organs of the body is performed in the Protozoa by firm structures, which either traverse the soft substance of the body as a framework, or invest it in the form of shells and tests. The latter serve as organs of defence in proportion to their size and firmness. All the structures here to be enumerated are either directly or indirectly differentiations of the protoplasm, which are formed either on the surface or in the parenchyma of the body. The more completely these secretions cover the body in the form of tests, the more do they interfere with its freedom of movement, unless there are compensating arrangements. Shells and internal supports are widely distributed in all divisions of the lower PEOTOZOA. 81 Fig. 25. Transverse section of a Foraminifer (Alveolina Quoii); the arrangement of the separate chambers in relation to one another can be seen (after W. Carpenter) . organisms ; they vary greatly in complexity, anil this complexity is sometimes in inverse proportion to that of the body. Simple shells, generally oval in form, and provided with an orifice, obtain in one division of the Amcebte (Difnugia, Arcella). The shell is sometimes soft ; but some- times it is more firm, and this firmness is increased by the addition of foreign bodies. They sometimes seem to be internal shells, owing to the extension of the protoplasm over them. The shells of the Foramini- fera are more complicated in form, owing to the for- mation of new parts around a simple rounded test, which then form separate cham- bers communicating with one another by orifices, and with the exterior by pores (Figs. 23, 25). These many-chambered shells become very firm by the addition of chalk, or, though more rarely, of silica (Polymorphiua Nonionina) ; owing to the variations in the relative position, size, and mode of connection of the chambers, these structures vie in wealth of form with the more lightly constructed internal supports of the Radiolaria. When the chambers are ranged along a straight line, rod-shaped shells, often swollen into beads, are formed ; the separate portions of which, known as " chambers/'' may be all of the same size, or may increase in size from one end to the other (Nodosaridge). A spiral arrangement of the chambers which lie in the same, or in different planes, gives rise to structures like those of the shells of the Nautilus (Fig. 23). Special modifications are due to the superposition of the spiral coils, the elongation or abbreviation of the spiral axis, and so on. The planorbis-like shells of the Milliolicla3, in which partial constrictions give the first sign of the formation of chambers, repre- sent the simplest condition of these forms. By the unequal addition of fresh chambers the spiral form is completely effaced (Acervuliuge), and can only be seen in the earliest-formed chambers. These tests are usually confounded with external shells. But this only holds for a few. The shell seems rather to be an internal support, in those cases especially where the partition-walls of the so-called chambers are frequently broken through, and the pore-canals at the same time pass through the shell to the exterior, so that the protoplasm of the pseudo- podia is able to cover the outside of the shell. When the partitions G 82 COMPARATIVE ANATOMY. are merely represented by several separate columns, or lamellas, which have wide openings between them (Fig. 25), and the cavity of the chamber itself is less than the various connections between two chambers, and where, finally, all the neighbouring chamber-spaces communicate with one another, and the whole " test " is thereby traversed by a cavitary system which communicates in all directions, then the character of an external shell is completely lost. And so, since the protoplasm is in all cases able to draw itself over the outer surface of the shell, the shells in the Foraminifera ought to be regarded as internal ones, and grouped accordingly with the skeletons of the Radiolaria. § 65. The " central capsule " must be noted as an organ of support common to all the Radiolaria, although it is not very apparent. It is a capsular-closed organ, various in form, placed in the middle of the body, and formed of a membrane which closely resembles chitin. In addition to fat-globules and structures which are re- garded as nuclei, it always contains a quantity of protoplasm, which is apparently continuous with the extra-capsular protoplasm by means of fine pore-canals. In most Radiolaria (not in Thalassicolla, Thalassolampe, and Collozoum) there is, in addition to this capsule, a skeleton, ordinarily formed of silica, which traverses the capsule when most fully de- veloped (Fig. 26). Several spicules then radiate from a common centre, and these may be connected with one another by means of a con- centric highly fenestrated framework (Fig-. 26). In some (AcanthometridEe) the organic base of the frame- work predominates, and is but slowly replaced by silica. Separate spicular bits of silica, lying freely in the protoplasm outside the central capsule, are the earliest indications of this firm skeleton in the Collidas and Polyzoa. In some they become arranged in a radiate manner,withoutbeing firmly connected together. Circular skeletons, forming an open network, arc formed by the radial spicula being connected together at equal distances by rods placed perpendicularly to them. When a very fine supporting reticulum is arranged around the radial spicula in Fig. 26. Skeleton of a Radiolarian (AcHnomma asteracanthion). Two concentrically-arranged fenestrated shells are broken through at one point, to show a third one (after E. Hiickcl). PKOTOZOA. 83 a more irregular manner, the skeleton is sponge-like. The infinite variety of form is increased by discoidal and basket-like skeletons ; or these skeletons may have a spiral arrangement. In this way a supporting organ of great complexity is formed, in which the soft parts of the body are embedded, and the separate pieces of which are developed in the protoplasm. § 66. Compared with these internal organs of support in the Rhizopoda the tests of the Infusoria form a distinct series of arrangements, for they are only secretions from the surface of the body : they resemble the tests of the Arcellas, mentioned above. The secreting matrix is in this case an anatomically distinct part of the body. But this need not be regarded as a higher stage, for it is closely allied to the lowest, i.e. to the formation of a cell-membrane. Tests are principally developed in the fixed Infusoria. They are formed by the secretion of a substance, which is primitively soft, and which gradually hardens ; this surrounds the body of the animal like a cup or an urn, except at one point, where it allows of communication with the exterior. These tests are distinguished from the merely cuticular structure, which tends to form a carapace, in that the differentiated layers attain greater firmness, and become separated from the greater part of their matrix surface. But the genesis of both structures is the same. It is identical with the formation of cysts, a process which is very common among the Infusoria, and by means of which the organism shuts itself off from the exterior for a time, so as to withstand unfavourable conditions (loss of water, and so on). The immovable stalk of Epistylis, and the external layer of the contractile stalk of the Vorticellinao and Carchesinaa must be regarded as cuticular differentiations of this kind. The tests maybe soft or hard, and membranous. Some are distinguished by the agglutination of foreign bodies — cemented grains of sand, and so on. The genera Vaginicola, Tintinnus, etc., have shells. Stentor has one in certain cases. Fenestrated shells have also been observed (Dictyocyrta). A carapace is formed from the firm hyaline cuticle in Stylonychia, Euplotes, Aspidisca, Spirochona, Coleps, etc. § 67. Organs for the prehension and alteration of food are wanting in the lowest organisms. In the Gregarinas food is taken in by endosmotic processes at the surface, and solid nutritivo matters do not reach the interior of the body. Where the body is not peripherally differentiated there is, on the other hand, a direct taking in of food, which may go on at any part of the body. This is the case in the Rhizopoda. In this case the nutritive matter is surrounded by the soft substance of the body, or it is embraced by the processes of the body — the pseudopodia. In both u 2 64 COMPARATIVE AX ATOMY. cases there is really one and the same phenomenon. Any place in the protoplasm can act as a digestive cavity by enveloping and absorbing nutritive matter, and at any neighbouring part of the surface the undigested substances can be again expelled. Even in Actinosphasrium solid food can be taken into the body ; but in it the pseudopodia are not the direct agents, for they draw the prey to the body, and cause it to pass into the yielding parenchyma of the cortical layer at some suitable point (Fig. 27) ; thence it passes to the central substance of the body. In comparison with other forms Acti- nospha3rium is characterised by taking its food directly into the more differ- entiated parts of the body, and not surrounding it by the amorphous protoplasm of the pseudopodia. In the Infusoria the arrangement is more definite. They take in food in two different ways; In the Suc- toria (Acinetina?) there is no mouth ; the radiate pseudopodia-like processes pass through the envelope of the body, and act as suckers (Fig. 30). They attach themselves by their sucker-like enlargements to the prey which has come within their reach (which consists of other Infusoria, etc.), and cause it to flow, as through a tube, into their body, where it fills the parenchyma in the form of drops. The presence of similar processes in the embryos of other Infusoria shows that this 'mode of nutrition is a very common one. A higher grade is represented in the other forms ; in the Ciliata there are not only definitely organised parts for the reception of food, but also definite parts for the ejection of what is useless. An enteric tube is, however, wanting in all of them, and these differentiations are limited to the cortical layer of the body, so that the food passes beneath it into soft paren- chyma, i.e. into the undifferentiated protoplasmic part of the body, in which there arc no passages with special walls. Temporary spaces, which act as digestive cavities for the spherical food masses, are formed in it ; these cavities are not permanent ones, as may be seen from their frequent disappearance when the protoplasm is in movement. In this point, therefore, they resemble the Rhizopoda ; for a part of the digestive apparatus, that is those parts in which the food is digested, have no organological differentiation. When the Ciliata have a mouth, it is either in the form of a simple cleft, which is in many cases only apparent when food is taken in ; or it is not directly on the surface of the bod}'', but at the bottom of a depression (vestibule), the form of which is very varied, and which at times contains the orifice of egestion also ; the sur- Fig. 27. A ctiuosphcerium. a A morsel which has been taken in as food, and just pushed into the soft cortical layer 6, by the animal, c Central parenchyma of the body. d Some balls of food in it. e Pseu- dopodia of the cortical layer. PEOTOZOA. 85 rounding region (peristoma) has often also a special form. A tubular portion, or pharynx (Fig. 28, h) often passes from the mouth into the parenchyma of the body, and from it the ingested morsel finds its way into the soft substance of the latter. The position and form of the mouth varies greatly. In many cases it can only be made out during the ingestion of food (as in Amphi- leptus, Loxophyllum), and disappears as soon as the morsel has passed into the parenchyma. There is sometimes an investment of cilia on the tubular pharynx (Paramecium aurelia and bursaria) ; or an undulating membrane (Bursaria flava) ; or a covering of rod-shaped denticles, or fine longitudinal ridges. Porodon, Chilodon, Nassula, etc., have an investment of small rods in the pharynx, arranged in eel-pot form. A regular thicken- ing of its walls has been observed in Ervilia and Liosiphon. The general presence of an anal opening is not by any means established. There is only in some few cases a permanent and dis- tinctly-marked opening; it can generally be distinguished during the expulsion of undigested food only. This anal spot is as a rule at the posterior end of the body, but is, on the whole, very changeable. It may even appear at the anterior end of the body ; thus, in Stentor it lies near the mouth, and in the Vorticellinns and Ophrydias in the vestibule. Taken on the whole it appears to consist more in the localisation of a function than in the develop- ment of an organ. The products of excretion pass through the differentiated cortical layer of the body at a certain spot, without there being any special organisation of that spot. Fig. 28. Diagram of the digestive cavity of Paramascium. a. Body- space filled with soft pro- toplasm, into which the food is taken, b Moutb. c Anns. d Contractile vesicles (after Lach- mann). § 68. In all Protozoa the outermost layer of the body has a respira- tory significance, for it is by it alone that an exchange of gases with the surrounding medium can be effected. This relation must also be borne in mind in reference to the increase of surface, which is due to the pseudopodia. The cilia of the Infusoria are of import- ance in changing the water. More definite respiratory arrangements are seen when, as in many Protozoa, water is taken into the body. Cavities, which are filled with a fluid, and which gradually contract and completely empty themselves, after having reached their maximum of distension, appear within the protoplasm ; when empty they seem to disappear. These vacuoles, like the vacuoles in the cells of certain tissues, are partly variable structures, now appearing and now disappearing, and 86 COMPARATIVE ANATOMY. partly constant. When they are constant their function is in- creased, and they often expand and contract regularly and rhyth- mically, like the cardiac systole and diastole. Contractile vesicles of this kind are often seen in the Amoebae (Difflugia and. Arcella), and are very common among the Infusoria. They are also known as vacuoles. The fluid which collects in the vesicles is drawn from the parenchyma of the body and is returned to it, or passed out to the exterior on the contraction of the vesicle. Fine communications with the exterior have been made out, so that the latter course is the probable one ; but we need on this account conclude that water does not enter by the same passage. In the Infusoria the vesicles lie in the cortical layer (Fig. 28, d d), generally just under the delicate cuticle, and at definite points. If only one vesicle is present, it lies either anteriorly or posteriorly : if two, there is one near each end of the body. Trachelius ovum is remarkable for a large number of small vesicles. No special mem- branes can be made out on the wall of the vesicle nor in the canals which pass off from it. Like the vesicle the canals can only be made out while they are filling. The vesicle and canals contract alternately. In Paramecium the canals enlarge at the commence- ment of the systole, and approach one another as the vesicle diminishes in size, so that they form a stellate figure at the moment when its systole is most complete and. the vesicle has disappeared. While the vesicle is filling the canals look like small diverticula on it, and are not again fully distended until the diastole is complete. The number of canals, which is limited in P. aurelia to eight or ten, is increased to thirty in Bursaria flava, and is much higher in Cyrtostomum leucas. In these forms the canals have a wave-like course, and. ramify at their extremities. Canalicular tracts are formed by the fusion of several spaces filled with water into longer tracts, as in Stylonychia (St. my tikis), and. they empty themselves into the contractile vesicle, by definite passages. The long canals of Spirostomum ambiguum, which also are visible for a time only, but which are longer than these, are like them, so that we can make out a continuous series from the first appearance of an apparently indifferent cavity to a definitely arranged system of tubes. Another arrangement can be put btside this formation of in- different vacuoles. When such spaces in the protoplasm increase in number they run together, and so give to the protoplasm the appearance of a network which traverses the interior of the body, which is filled with fluid (Trachelius ovum). These hollow spaces have then become perfectly different organs from the pulsating vacuoles, and they may both exist at the same time. § 69. In correspondence with their low grade of organisation the Protozoa have no sexual organs, and give but the faintest indica- PROTOZOA. 87 tions of sexual differentiation. They always propagate, therefore, by modes which are called asexual, among which the chief part is played by fission and generation. The nucleus appears to be of great importance in all their modes of multiplication. Spores have been observed to be formed within the organism in Rhizopoda. A larger or smaller part of the protoplasm of the body is used in forming them ; when a larger portion is used this mode of multiplication is allied to that mode which is so common among the Protista, in which the whole body breaks up into a number of spores, and so multiplies by division. In the Radiolaria the contents of the central capsule are active in reproduction. The nuclei in it become surrounded with protoplasm, and form flagellate swarm- spores. The mode of reproduction is most exactly known in the Gre- garinas. As a rule multiplication commences by the concrescence of two individuals ; this generally occurs very early, so that the two individuals, which form one body, the anterior end of one being attached to the posterior end of the other (Fig. 29), go on growing for some time ; or conjugation may only take place later, when the forms are mature. After this comes a condition of rest, accompanied by encystation, in which the two individuals form a rounded body, with a partition between them. Then the partition disappears, and the substance of the body, and also the nucleus, breaks up into an amorphous mass, from which numerous vesicles gradually arise. From these latter a number of germ corpuscles, called " Pseudonavicellas," on account of their shape, are formed. These gradually fill the whole cyst, and each gives rise to a single very small organism, consisting of protoplasm solely, and this, being without a nucleus, corresponds to a cytod. Each of these structures moves about in an amoeboid manner, and is gradually differentiated into a young Gregarina, after which a nucleus is differentiated in its interior, and it becomes limited externally by a cortical layer. Although conjugation has no exclusive signification in bringing about these processes, as separate Gregarinas are also able to pass through these reproductive processes in just the same way, yet it is not the less important. It points, at least in the cases where it exists, to the necessity of two individuals to bring about reproduc- tion. It is, consequently, a phasnomenon preliminary to sexual differentiation. Fig. 29. a b Two con- jngated individuals of Gregarina seenuridis. c Their nuclei. 70. Conjugation also plays a part in the reproductive proceedings of Infusoria, for it is the first step in their multiplication. In this the 88 COMPARATIVE ANATOMY." nucleus is of considerable importance; it (Fig. 30, n) is a firm structure, sometimes provided with a spiral envelope, very various in form. It lies in the cortical substance of the body, or is surrounded by a continuation of this substance, if it is more deeply sunk in the interior. It is sometimes oval or round, or it is flattened and curved (Vor- ticellinge), or it is even greatly elongated and regularly con- stricted (Spirostomum). The importance of the nucleolus, which differs from the nucleus in nothing but its smaller size, is more obscure. The act of reproduction commences as a rule with the complete or partial fusion of two individuals, which may be of the same or of dif- ferent sizes ; this fact led to the mistaking of conjugation for stages of fission or gemma- tion. This concrescence gives the stimulus to changes in the appropriate parts. The nucleus becomes divided into a certain number of parts, around which the protoplasm is disposed. In this way a number of spores are formed, each of which becomes a new individual while within the mother-cell; and then gets an investment of cilia, and escapes to the exterior. It is still a question as to the share which the nucleolus takes in this process ; and the statement that in one group of the Ciliata it has the function of a sperm-forming organ, while the nucleus has the function of an ovary, requires to be confirmed. In any case this differentiation of a male organ is not a common phenomenon, but is limited to a very narrow circle. The nucleus, therefore, and the nucleus alone, is certainly known to take an active share in repro- duction, and this share is of just the same character as that which it was seen above to have in spore formation, and as it has in gemma- tion, in which the nucleus of the bud has often been observed to arise from the previous gemmation of the nucleus of the mother-cell (Podophrya). Finally, multiplication by fission is very common, although conjugation was often confounded with this process at one time. Fig. 30. — An Acineta with part of its stalk, p Psenclopodia-like, but stiff, ten- tacles, v Vacuoles, n Nucleus, e Aciliated young form lying in the so-called broad cavity. Second Section. Ccelenterata (Zoophyta). General Review. § 71. This division is the first of the Metazoa, or organisms which are undoubtedly animals. The embryonic body separates into two cell-layers — ectoderm and endoderm ; which in many Sponges alone form the permanent body, though in many a mesoderm is developed. In the lower Acalephas the formation of a mesoderm is incomplete ; that is, the mesoderm is not an independent tissue as it is in all the higher Acalephas. The most essential character of the animals which make up this division is the arrangement of the nutritive apparatus, a cavity sunk into the parenchyma of the body, and either divided into canals or extended into wider spaces. This digestive cavity, with its appended spaces, is invested by the endoderm, and in the lower forms is the sole representative of hollow organs in the body. When several individuals are united to form a colony, the canal system, which arises from the digestive cavity, is common to all of them, and is continued into the common substance of the colony or ccenenchyma. The primary axis alone can be made out in the body, and the secondary axes are indifferent, or if present appear to be equivalent. I. Spongiae. Gastrreades.* Haliphysema, Gastrophysema. Porifera, Myxospongioa. Halisarca. * The Gastrceades represent stages -which are not permanent in the rest of the Spongiae. 90 COMPARATIVE ANATOMY. Fibrospongia). Ceraspongia3. Euspongia, Spongelia, Poterium. Ilalichondria?. Axinella, Spongilla. Corticata. Thetya. Hyalc-spongiae. Euplectella . Calcispongios. Ascon, Leucon, Svcon. II. Acalephae. 1. Hydromedusns. Hydrif ormes. Hydra ; — Cordylophora ; — Hydractiiiia ; — Coryne, Syncoryne, Eudendrium ; — Tubularia, Corymorpha ; — Campanu- laria, Sertularia, Plumularia. Medusif ormes. Sarsia, Bougainvillea, Lizzia, Oceania ; ■ — E ucope, Thaumnntias ; — Trachynema ; — iEgina, Cunina ;— Liriope, Geryonia, .aSquorea. Siphonophora. Velella, Porpita ;— Diphyes, Abyla ;— Athorybia, Agalma, Physophora, Physalia. 2. Calycozoa. Lucemaria. 3. Thecomedusas. Stephanoscyphus. 4. Medusas (Discophora). Charybdea, Pelagia, Aurelin, Ehizostoma, Cassiopeia. 5. Anthozoa. Tetractinia. Cereanthus, Cyathophyllum. Hexactinia. Antipathes, Fungia, Madrepora, Astiwa, Oculina, Caryophyllia. Octactinia (Alcyonaria). Alcyomuin, Pennatula, Yirgularia, Veretillum, Renilla, Gorgonia, Isis, Corallium, Tnbipora. C. Ctenophora. Beroe, Cydippe, Cestum, Eurhamphsea, Mnemia, Eucharis. Bibliography. Spongise : Grant, B. E., Observ. on the struct, and funct. of Sponges. Edinb. New Phil. Journal, 182C-32. — Lieberkuhn, Beitr. z. Entw. der Spongillen. Arch. f. Anat. u. Physiol. 1856. Zur Anat. d. Spongillen, ibid.— The same, Z. Anat. d. Spongien, ibid. 1857, 1859, 1863. — Schultze, M., Die Hyalonemen. Bonn, 1860. — Schmidt, O., Die Spongien des adriat. Meeres. Leipzig, 1862. Supplement, 1861. Zweites Supplement, 1867. Drittes Supplement, 1868.— Glaus, Ueber Euplectella Aspergillum. Marb. u. Leipzig, 1868. — Harting, P., Sur le genre Poterium, Natuurkund. Verhandelingen. P. II. St. 2. Utrecht, 1870. — Hackel, Die Kalkschwiimme, eine Monographie. 3 Bde. Berlin, 1872. — The same : Die Physemarien. Jen. Zeitschr. Bd. X. — Schulze, P. E., Ueber Bau u. Entw. v. Sycandra raphanes. Zeitschr. f. w. Zool. Bd. XV. — The same, Untersuch. iiber d. Bau u. d. Entw. der Spongien (Halisarca), ibid. Bd. XXVIII. Acalephse : Cavot.ini, Memorie per servire alia storia dei polipi marini. Napoli, 1755. (German by Sprengel. Niirnberg, 1813.) — Eschscholtz, System der Acalephcn. Berlin, 1829. — Lesson, Zoophytes acalephos. Paris, 1843. (Suite a Buffon).— Sars, Fauna littoralis Norvegia) I. 1816. CCELENTERATA. 91 — Feet n. Leuckaet, Boitriige zur naheren Kcnntniss wirbelloser Thicrc. Braunschweig, 1817. — Huxley, Ou the anatomy and affinities of the family of the Medusas. Phil. Tr. 18-19. — Agassiz, L., Contributions to the nat. hist, of the Acalcphse of N. Am. (Mem. of the Amor. Acad, of Arts and Sc. Cambridge, 1850).— The same, Contrib. to the nat. hist, of the United States of North America. Vol. III. IV. 1860-62. Hydromedusaj : Van Benbden, P., Mem. sur les Canipanulaires de la cute d'Ostende. (Nouv. Mem. de l'Acad. royale de Bruxelles. T. XVII.) Recherches sur l'embryogenie des Tubulaires (ibid.). — Kollikeb, Die Schwinimpolypen von Messina. Leipzig, 1853. — Leuckaet, R., Zur naheren Kcnntniss der Siphonophoren von Nizza. Arch. f. Nat. 1851. — Gegenbaur, Beitr. zur naheren Kenntniss der Siphonophoren. Zeitschr. f. wiss. Zoologie. Bd. V. — Vogt, C, Sur les. Siphonophores de la mer de Nice. Mem. de l'inst. Genevois, 1851. — Claus, Ueber Physophora lrydrostatica. Zeitschr. fur w. Zoolog. Bd. X. Neue Beobachtungen. Ibid. Bd. XIII. — Hackel, E., Zur Entwickelungsgesch. der Siphonophoren. Natuurkund. Verhandelingen. P. I. St. 0, Utrecht, 1809. — Huxley, Oceanic Hydrozoa. London, 1859. (R.S.) — Fobbes, Bd., A monograph of the British naked-eyed Medusa?.. London, 1818. (R.S.) — Hackel, Die Familie der Riisselquallen. Jenaische Zeitschrii't. Bd. I. II. (Also under the title : Beitr. zur Naturgesch. d. Hydromedusen. I. 1865). — Schxlze, F. E., Ueber den Bau unci die Entwickelung der Cordylophora lacustris. Leipzig, 1871.— The same, Ueber d. Bau v. Syncoryne, etc. Leipzig, 1873.—' Kleinenbebg, N., Hydra. Leipzig, 1S72. — Alliian, G. J., Monograph of the Gymnoblastic or tubularian Hydroids. P. I. and II. London, 1871-72. (R.S.)— The same, On the structure and develop, of Myriothela. Phil. Trans. Vol. 165. Calvcozoa : Clark, II., Prodromus of the Historv, etc., of the order Lucernavia. Journal of Bost. Soc. of Nat. Hist. 1S03. Thecomedusa?: Allmax, G. J., On the Structure and systemat. position of Stephanoscyphus mirabilis. Trans. Linn. Soc. Sec. ser. Vol. I. Zool. Discophora : Ehrexberg, Ueber Acalephen des rothen Meeres und d. Organismus der Medusen der Ostsce. Abhandl. der Berl. Acad. 1835. — Milne-Edwards, Ann. sc. nat. III. xvi. — Wagner, R., Ueber den Bau der Pelagia noctiluca und fiber die Organisation der Medusen. Leipzig, 1811. — Hackel, E., Ueber die Crambessiden. Zeitschr. f . wiss. Zoolog. Bd. XIX. — Brandt, A., Ueber Rhizostoma Cuvieri. Mein. Acad. Imp. des Sc. de St. Petersb. VII. Ser. T. XVI. — Geenacher, H., and Nolc, F. C, Beitr. z. Anatomie u. Systematik der Rhizostomeen. Abh. d. Senckenb. Gesellsch. Bd. X. Anthozoa: Ehrexberg, Die Corallenthiere des rothen Meeres. (Abh. d. Berl. Acad., 1832). — Hollard, Monographic anatomique du genre actinia. Ann. sc. nat. HI. xv. — Haijie, J., Mem. sur le genre Cereanthus. Ann. sc. nat. IV. I. — Lacaze-Dcthiers, Hist. nat. du corail. Paris, 1861. — The same, Memoires sur les Antipathaires. Ann. sc. nat. V. n. iv. Developpement des Coralliaires. Archives de Zoolog. exp. Vol. I. II. — Kollikee, Die Pennatuhden. Abh. d. Senckenb. Gesellsch. Bd. VII. — Eiben, G., Bidr. tid kannedomen om Renilla. Kongl. Svensk. Vet. Hand]. Bd. Xni.-v. Kocn, G., Anat. d. Orgelcoralle. Jena, 1871.— Mose ley, H. N., On the structure and relations of the Alcyonarian Heliopora can-ulea, etc. Phil. Transac. Vol. 166, Pt. I. Ctenophora: Will, Horn? tergestina?. Leipzig, 1811. — Milne-Edwards, Ann. sc. nat. Ser. IV. Vol. VII.— Kowalevsky, A., Entw. des Rippenquallen. Mem. de l'Acad. Imp.de St. P^teivb. VII. Ser. Tom. X. — Fol., H., Beitr. z. anatom. Entwickl. einiger Rippenquallen, Berlin, 18C9. — Eimeb, Th., Zoologische Studien auf Capri I. Ueber Beroe ovatus, 1873. Form of the Body, § 72. The forms of body of the two great divisions which make up the Coelenterata are alike in the lowest stage ; in that stage, namely, which has been already (§ 28) called the " Gastrula," on account of the development of an enteric cavity. This form represents a larval stage, in which an investment of cilia function as a locomotor organ, and which may fairly be regarded as the common funda- mental form of the two chief divisions of the Zoophyta. In this form only one axis, the primary, can be made out ; it extends from the oral to the aboral pole. The secondary axes are indif- ferent, for all the transverse axes drawn perpendicularly through the primary, at any angle whatsoever, are completely equivalent one with the other. This stage is permanent in all the Spongige ; in the Acalepha3 it passes on into a condition which is characterised by differentiation of transverse axes. Among the tSpongire the Gastrula attains, when attached by its 92 COMPARATIVE ANATOMY. ah oral pole, to a definite character of the most simple form in the Physemaria?, and in Olynthus among the Ascones. In other Calci- spongias, also, this simple form of body is retained, but more considerable changes in its internal characters obtain. The most important changes in the form of the body are due to the formation of colonies. Colonies of the most varied form (cormi) are formed by budding or by incomplete division, the separate animals (persona?) of which are connected with one another in very various ways, and. may even part]y or completely fuse with one another in different ways. When these stocks are fused they often have the appearance of single animals, and in proportion to the simplification of their external form is the complication of their internal organi- sation. The modification of the mouths of the colony affects their external form just as much as does this concrescence; the mouths may be collected into groups, or united as one, or they may completely disappear. The great variety of form in this division, which is due to the causes here only indicated, may be still further modified by numerous adaptations, due to their position in space. Nowhere in the Animal Kingdom does the form of the body appear to be so changeable as in the Spongias, so that it is impossible to separate them into large divisions, to say nothing of species. § 73. In almost all the divisions of the Acalepha3 the body developed from the Gastrasa-form is adapted to a sessile or fixed condition; the stomachal cavity which is formed when this development com- mences causes the organism to have, in all essential points, the same simple character as it has in the corresponding stage in the life- history of the Spongias. Processes, known as tentacles, are developed on the anterior region of the body-wall, which encloses the stomachal cavity ; they present to us the earliest indication of a differentia- tion of the secondary axes, and therefore establish a well-marked distinction between these forms and the Spongia?. The Hydroida or Hydroid-Polyps (Hydriformes) are the lowest of the Hydromedusa3. Tentacles are placed in many cases irregularly on the parts of the body which surround the mouth (Coryne, Syncoryne, Cordy- lophora), or their number maybe indefinite, even when the structures are limited to definite zones of the body, and encircle the mouth in the anterior region (Hydractinia, Eudendrium, Campanularia) . As the number of tentacles varies we cannot suppose that the secondary axes are definitely differentiated. It is only in a few cases that they are definitely expressed by the position of the tentacles (Stauridimn). The free portion of the body, with its tentacles, becomes more independent of the rest of the body, which forms a stalk, while the aboral pole grows out into a stalk-like part, which carries the head, and is distinguished as the " capitellnm " or "hydranth." FORM OF BODY OF CGELENTERATA. 93 Colonies (coruii) are formed from single animals by gemmation. This may either occur at any point of the surface of the body (Hydra), and end by the bud breaking off, or it may take place in the stalk- like part. The creeping cormi of the Syncorynida), Hydractinias, etc., are formed by processes of the basal part, which give off new animals, attached here and there to it. When gemmation occurs in the free part of the stalk we get free, branched colonies, which become compli- cated in very various ways (Eudendriuin, Campanularia), or become regularly branched (Sertularia, Pluuiularia) . The formation of colonies is almost always accompanied by the formation of a tubular investment, which is a secretion from the surface of the body, and which serves as a support for the whole trunk, as well as for its branches ; it is also continued, in various degrees, on to the persons of the colony. § 74, The process of gemination in the Hydroid Polypes produces, in addition to the growth of the colony by freshly-formed similar individuals (persons), structures of quite a different kind, the most differentiated forms of which are developed into Medusas. The body of these buds is of a bell-shaped or discoid form (Fig. 32, m), and by its internal organisation, as well as by the tentacles, which arise from the edge of the bell, or disc, we are able to make out secondary axes, generally two in number, which cross the primary axis at right angles to one another, and are completely equivalent one with the other. A higher grade than that of the Hydroid-Polyps is expressed in this organisation. The animals move by contractions of the bell, the edge of which is produced into a membrane, the velum, which is also contractile. These Medusa-gemmas always carry the organs of reproduction ; from their ova Hydroid Polypes again arise. (Alternation of Generation.) While gemmation of Medusas in- tended for a free life distinguishes some Hydroid-Polyps (Fig. 31, a-e, Fig. 32, a-e), in others the process stops at the formation of a Medusa- bud, the organisation of which does not quite attain that stage of develop- ment which conditions the free mode of life, and it remains therefore con- nected with the colony. Nevertheless, the normal development of the sexual organs proceeds, and in fact these rudimentary Medusas form " generative buds " (Gonophores), Fig. 31. Syncoryne with a num- ber of budding Medusas on it at different stages (a-e) of develop- ment (after Desor). Di COMPARATIVE ANATOMY, the products of which are developed in them in the same manner as are those of the free Medusae. With these are connected still simpler forms of buds, and the series ends with buds the structure of which has scarcely anything in common with a Medusa. But the series which leads to these is perfect, owing to numerous intermediate forms, so that external buds, merely containing generative products, and Medusas of a rela- tively high organisation, which only become sexually mature some time after leaving the Hydroid stock, must be regarded as the widely-separateterminalpoints of one series. This phenomenon is ex- plained by the conception of a division of labour, in which the function of feeding the stock falls to the share of the individuals which remain ses- sile, while others which are broken off take on the duty of sexual reproduction. Those buds which will become free have a higher organisation, which has been gradually de- veloped from the lower forms, and primitively resembled those that remain sessile. The separation from the stock may therefore be regarded as the primary cause of the differen- tiation of the sexual individuals in the medusoid direction, while the permanence of the sessile habit of the medusoid buds, in other cases, is accom- panied by a degeneration of their medusoid organisation. But if this organisation, as we supposed above, has been obtained by a primitive freedom of life, the medusoid buds must necessarily be regarded not as arrested in an onward development, but rather as Medusa-buds in course of degeneration. A definite conclusion on the subject is not possible, on account of the fact that the several stages of degeneration might be precisely similar to those of deve- lopment, and retrogressive metamorphoses have not been directly observed. The gemmation of generative individuals, for such must the medusiform buds and their modifications be considered to be, takes place at different points. As the formation of a stock is a secondary process, the production of budson the body of the single animalmustbe Fig. 32. A portion of a colony of a Hydroid- Polyp (Eudendrium ramosum) with bud- ding Medusa?, p p p Nutritive persons. a b c d ef Different stages in the differen- tiation of the budding Medusa;, m m Free Meduste in different positions. FOEM OF BODY OF CCELENTEEATA. 95 the primitive mode. It is found oil this region in all divisions of the Hydroid-Polyps. The stocks of the Corynidas have buds distributed over the hydranth. They are frequently placed between the tentacles. In Peunaria they are found within the circlet of tentacles ; and on the same place in the Tubularias, where they are always placed in some quantity on a common stalk, forming groups like grapes, or ears of corn. Gemmation on the body of the Hydroida is, in many cases, accompanied by a degeneration. In many Canipanulariae, Hydractinias, and others, the proliferating individual gives up its share in the duty of feeding the stock, as is expressed by the diminished size of its tentacles and stomachal cavity. The animal stock is therefore composed of nutritive and proliferating persons, the latter again bearing the buds or generative persons. This is the way in which the Dimorphism of Persons arises in these stocks, and this passes into Polymorphism, in consequence of a number of the nutritive persons undergoing still further changes (Hydractinias) . The proliferating persons of a colony present various degrees of degeneration. In the most extreme case a portion only of the individual which bore the buds remains after they are developed (e.g. in many Campanularias) . The complete degeneration of the proliferating person causes the geinmge to arise from any part of the common stock, without any relation to a Hydroid person. In the higher divisions of the Medusiformes, relations to the Hydroida are lost. Although there are many considerable complica- tions in the mode of reproduction (see below, Generative Organs), yet, so far as is yet known, there is among the Trachynemidas, yEginidas, and Geryonidaa no return to the hydroid form; and it is doubtful whether any such relation does obtain. § 75. The division of labour, which in Hydroid-Polyps is essentially limited to the nutritive and generative functions of the persons united into a colony, is extended over a larger series in the Sipho- nophora ; and we met therefore with great variety in the form of the component parts of the colony. Division of labour thus leads to Polymorphism of the persons. They all follow the medusiform type, which is developed to a greater or less extent. When it is distinctly developed, the fundamental form which shows itself in the medusa-buds of the Hydroid-Polyps predominates, so that it is clear that both groups have had a common origin. The Siphonophora, therefore, appear to be swimming Hydroid colonies, all the persons of which have passed into the medusa form ; a change which is complete in the generative persons only of the Hydroid- Polyps. The separate persons of the colony of the Siphonophora are developed on a common contractile stem, which in most forms represents the axis of the stock, and around which the persons which function as organs of the whole colony appear to be arranged. 96 COMPARATIVE ANATOMY. These are — 1) Locomotive Persons (Nectocalyces) : these conform most completely to the Medusa-type, and are united together by twos Fig. 33. Some colonics of Siphonopliora. ADiphyes campanulata. B A group of appendages of the stem of the same Diphyes. C Physophora hydrostatica. A separate nectocalyx of it. E Cluster of female generative buds of Ayalma Sarsii. D a Trunk or axis of the colony, d Air-bladder. m Nectocalyx. c Cavity in nectocalyx, invested by a contractile membrane, v Canals in the walls of this cavity, o Opening of the nectocalyx. t Bracts (in c represented by tentacles), n Stomach. i Fishing-lines, g Generative organs. (Diphyida)), or, by a larger number, to form a swimming column (Physophorida)), which occupies the end of the stem (Fig. 33, A 0, m D), and which, therefore, takes the lead when the colony is moving, and is directed anteriorly. 2) Nutritive Persons are present in the second division of FORM OF EODY OF C (ELECTEE ATA. 97 the stein, where they have the form of stomachal tubes (stomachs, sucking tubes) (B G n.) Iu some cases some of them are not fully developed, but form tubes closed at the end, which function as " tentacles." 3) Protective Persons (Hydrophyllia) : in these one can often make out the Medusa-type quite clearly, but in other cases it is much less distinct, and they then have the appearance of hyaline, lamellar pieces, overlapping the persons described under 2, 4, and 5. 4) Tentacular Persons: these form simple or elongated filaments (grappling-lines), which are arranged in tufts, are capable of great extension, and are provided with special urticating organs (Urticating batteries). The primitive Medusa form can be made out in a few only of these organs, and that faintly. 5) Generative Persons : as in the Hydroid-Polyps, these may be seen in various stages of development. Although it is in a very few cases only that they are metamorphosed into Medusas that become free (Velella — Chrysosmitra), the medusiform type is very commonly well marked among them. They are generally found in racemose bunches, just as in the Tubulariaa. The arrangement of these very variously differentiated persons of the stock of the Siphonophora differs in the different divisions, while the locomotor and the protecting persons are completely wanting in many genera. In general the arrangement or distribution of the polymorphous persons of the stock is observed to be very constant in genera and species ; gemmation from the stock takes place on one side only, the arrangement of the buds all round the stock being due to its spiral twisting. This is the cause of the aiTangement of the nectocalyces in two or more rows, as well as of the grouping of the other organs. Nutritive, generative, and tentacular individuals are generally placed together in groups, in such a way that there is one bract to a group. While in most Physophorida?, these groups are very close to one another, they are set at greater distances from one another in the DiphyidaD (Fig. 33, A B), and each group is composed of a certain number of persons which, by breaking off from the stock, may become distinct individuals (Eudoxias). The anterior end of the stem, which is distinguished by the presence of locomotor persons, becomes in many divisions par- ticularly perfect, owing to the development of an air sac. This has the functions of a hydrostatic organ, and causes the anterior end of the body to be always directed upwards while the stock is at rest (Physophorida?) . (c a.) It has an opening to the exterior, which can be closed, and by which air has been observed to escape. The greater development of these bladders, which in most Physophorida) are rather small, appears to cause a degeneration of the locomotor buds of the stock; and thus there seems to be a kind of compensat- ing arrangement, through which, however, the power of the colony to move actively is diminished. Instead of swimming, it now is driven through the water. The locomotor persons are, for example, absent in Rhizophysa, in which the air sac is increased in size. By H 98 COMPARATIVE ANATOMY. becoming greatly increased into a wide cavity, the air sac occupies tlie greatest part of the stem, and thus forms the largest portion of the colony, the separate pieces of which look like appendages placed on one side of the bladder. This character is greatly developed in the Physalidas, and is accompanied by the shortening of the stem. Another condition obtains in the Velellidas, where the air sac is placed on the end of the greatly shortened trunk, and is developed by lateral extension into a disc, the cartilaginous firm walls of which divide the internal cavity into numerous chambers by forming walls of partition. In the earliest stages of development the air sac is simple in these forms also. In Porpita, the disc remains flat and circular; in Velella it is produced into a diagonal vertical crest, into which the air spaces of the disc are not continued. The concentri- cally-arranged chamber-spaces of the air sac in Velella are connected by apertures. They open to the exterior by a number of holes placed on the surface. In Porpita, fine air passages, in the form of canals, pass off from the inferior surface of the air sac, and enter into and branch in the portion of the stem, which carries the nutritive individuals. § 76. The Thecomedusa) are polypoid Ccelenterata provided with a test, and are allied to the Hydriformes, although in organisation they resemble Medusas; they are indeed, intermediate between these two groups j for they are representatives of forms which are closely allied to the larva) of the Discophora. This larval form (Scyphostoma) seems to be more highly organised than most of the Hydroid- Polyps ; it presents, indeed, points of connection with only a few of them (Cory- morpha). It is developed, just like the Hydroid-Polyps, from a planula, which is at first free, and which after- wards becomes fixed. But the fundamental form of the body resembles not only that of many Hydroid- Polyps, but their Medusa stage also, for two equiva- lent secondary axes cross th e primary one. The organ s are arranged by fours, so that four antimeres can be distinguished in the body. Medusa? are budded off from this polyp-form, but gemmation does not, as in the Hydroid-Polyps, take place at the side, but at the end. The terminal =C Fig. 34. Young stages of Aurelia aurita. 1 Planula-form, attaching itself. 2, 3 Passage into the Polyp-form. <1 Commencement of the formation of metameres. 5 Continued forma- tion of mctarnercs (Strobila) and the differen- tiation of them (after M. Sars). FORM OF BODY OF CCELENTERATA. 99 mouth-bearing portion of Scyphostoma is gradually nipped off from the rest of the body (Fig. 34, 4). As the body grows the new por- tions, which are formed towards the aboral pole, become separated metamerically (Strobila, Fig-. 34, 5), and all are developed on a Medusa type. The polyp-body is thus divided into a number, often a large number, of Medusa?, which gradually break off (Ephyra form), and when they are free become more developed. This process, which has been observed in Cephasa, Aurelia, and Cassiopeia, does not obtain in Pelagia, the ova of which are con- verted into swimming larva?, which become young Medusa), without passing through the polyp stage. The development of Pelagia is therefore compressed into a few stages, while in the others it is extended over a large series of forms, and is a more complete repetition of the palasontological development. The polypoid must be regarded as the initial stage, and was followed by the gradual metamorphosis of the polyp into a free Medusa. On this hypothesis, the strobilation of Scyphostoma and the consequent development of a immber of Medusas, appears to be a secondary process, which could only come about gradually, and after the whole polyp-body had ceased to be converted into a Medusa. It is clear, from the growth of the polyp, while it is passing into the Strobila, that an important part must be played by the nutritive relations of the Scyphostoma stage, in giving rise to the Strobila-form, or, in other words, in producing the Medusas by gemmation ; the whole phasnomenon, therefore, appears to be causally related to the nutri- tion o'f the Scyphostoma. By the gemmation of Ephyras, i.e. of young Discophora, from the body of the Strobila, an asexual mode of multiplication is interposed in the developmental process of the Medusas ; and thus we have brought about one form of the so-called alternation of generation. By the Scyphostoma form the Medusas are closely related to the Calycozoa, which appear to be derived from them. The body, which is attached by a short stalk, is widened out like an umbrella, and agrees, as to its axes, with the Scyphostomas, and their descendants. In many points they also present relations with the Anthozoa. The Calycozoa, therefore, present us with a very important intermediate form, which has been continued on, with relatively few modifications, from the ancestral form common to several large divisions of the Acalephas. § 77. In the Anthozoa the primitive form of the body is exactly the same as that of the other Coelenterata ; and even the earliest stages of the fixed planula present no essential differences. The appear- ance of tentacles and the subsequent internal differentiation give rise to various differences, the first of which affects the fundamental number of the secondary axes of the body. In some only four tentacles appear (Tetractinia), in others six (Hexactinia), and finally, 100 COMPAEATIVE AX ATOMY. in others, eight (Octactinia). In the first two divisions this number is not permanent, but the tentacles are soon increased in number, and there is a corresponding change in the internal organisation. A larger number of transverse axes can then be made out in the organism, but their fundamental number in most cases remains the same as at first. In the Octactinia, however, the first four transverse axes persist. The body of the young animal is generally cylindrical, but this form is retained in a few divisions only (Cereanthus, Actinia). In the other forms colonies are built up, and this causes the external appearance to vary greatly. The stocks (PolypariaD) are formed either by incomplete division, or by gemmation, or by both com- bined. Longitudinal division aids in the formation of the colony to a varying extent. In many cases it is merely indicated by transverse growth, and does not lead to any division of the organism, as in many Fungiee. In others, division affects the oral surface only, and the internal parts remain continuous. When this process goes on for some time, colonies with a large number of orifices are formed, which are arranged in variously-curved rows, beset at their edges with tentacles (Moaandrina). Whilst flattened or extended racemose colonies are formed in this way, branched stocks are formed when division is combined with a considerable growth of the persons in a longitudinal direction ; and these stocks may not only vary in size but also be branched in various ways. In the same way gemma- tion may be the cause of the formation of complicated colonies. In either case there is a portion of the body (coenosarc, coenenchyma) common to all the persons, and belonging to the common stock. The basal portion of the stocks of those Octactinias, which are not fixed but set loosely in the mud or sand, are developed from this coenosarc and form a solid stalk-like portion of the stock, in which gemmation does not occur (Pennatulida?). § 78. In the Ctenophora, which is the division differing most from the other Acaleplne, the permanent form of the body is developed from a larva, which in all essential points is similar to that of the others. In the Ctenophora there are four secondary axes, perpendicular to the primary, and the most important organs are arranged conformably to these. The body therefore follows, generally, the radiate type, which is best developed in the Beroidoj. This eight- rayed form is, however, derived from a four-rayed form, each radius having been split up into two. Each of the two radii which arise from a primitive radius are equivalent to the opposite radii of the same transverse axes. The development of the body follows the poles of one of the two transverse axes. The differentiation which arises in this way is clearly seen in the Cydippida?; it is more APPENDAGES OF CCELENTEKATA. 101 distinct in the Mnemidae, owing to the presence of lobate processes directed towards the oral pole, and most distinct in Cestuin, where the form of the body has become that of a band, from its having" grown in the direction of two similar interradii. Appendages. § 79. I comprise under the head of appendages those processes of the body which are known as tentacles; they are either altogether absent or are only just indicated in the Spongias, but in the Acalepha? they are widely distributed, and largely affect the external form of these organisms, in addition to which they are of great physiological importance to its general economy. Most of them are, like the wall of the body, contractile, but there are stiff forms which are not capable of much movement (Trachynemida)) . The tentacles are the seat of a large amount of sensibility, and function as sensory organs ; in many cases they are organs of prehension ; and, finally, they serve as organs of offence by means of the urticating cells which are attached to them. The Hydroid-Polyps present the lowest condition ; in many divisions of them the tentacles are scattered over the surface of the most anterior portion (or portion lying nearest the oral pole) of the body. In many they may be seen to be arranged more regularly, and in others they form a " circlet of tentacles " (Hydractinia, Eudendrium, Campanularia). This is generally placed at some distance from the mouth, and gives a higher importance to this part, which appears to be analogous to a head; and, indeed, the tentacular portion of the body (hydranth) of the Hydroida is called a " capitellum." The development, in the Tubularia, of a second circlet of tentacles, which directly surrounds the mouth, is correlated with the higher differentiation of the whole body. The outer circlet of tentacles is moved to the edge of the hydranth, as this portion becomes flattened out into a disc. Oral and marginal tentacles can then be made out. The latter are greatly developed among the Hydromedusa3 as well as among the Medusae. The marginal tentacles, or marginal filaments, which are generally greatly elongated filamentous appendages of the edge of the bell or disc in the Hydromedusse, are always arranged in corre- spondence with the radii of the body. Where interradial tentacles are present, they generally follow the radial ones, even when there is a large number of them. Sometimes they are arranged in tufts (Lizzia), or are branched (Oladonema). In opposition to the increase in the number of tentacles, until it surpasses that of the rays, is the 102 COMPARATIVE AX ATOMY. diminution of these structures. Saphenia has only two tentacles : in some forms, only one is developed (Stenstrupia). In the Trachy- nemidae also, the tentacles are arranged radially, and many, as the ^Eo-midas, have interradial ones in addition. The attachment of the tentacles to the body is peculiar, as the tissue which supports them often sends a considerable process into the body. Reduction occurs here also. ZEginopsis has only two tentacles. In the Geryonidas a change of tentacles takes place, the young animal having filaments which are not permanent (larval tentacles) and which are different in structure from the permanent ones. The oral tentacles distributed among the Hydromedusas like- wise correspond in number to the radii of the body. They are sometimes simple, sometimes branched. They are not, however, always present, and are frequently replaced by outward growths of the edge of the mouth. They arc generally wanting in the Trachy- nemidse and ^Eginidas. Among the Siphonophora all the medusiform persons want the marginal filaments, which seem to be indicated merely as rudiments ; as, for example, in the nematophorous enlargements of the protective persons. This want of an organ important in the economy of the colony is compensated for by the "tentacles" and the "grappling- lines," which can be shown to be modifications of medusiform persons (§ 75). The marginal filaments are wanting in the divisions of the Rhizostomidas and Cyanete among the Discophora; they have four large groups of tentacles which arise from the lower surface of the umbrella, and which can be considered either as marginal filaments, or as oral tentacles. In others there are marginal filaments present, corresponding in number to the radii, and sometimes even interradial ones are present. Even in the Charabdcida?, Charabdea has four tentacles carried by the arrow-shaped processes of the bell; in Tamoya (T. quadrumana) these are represented by the same number of tufts. The filaments are more numerous in the Pelagire, whilst the Aurelias are distinguished by a very large number of fine marginal ones. Oral tentacles are developed as fine fringing processes on the edges of the arms which surround the mouth. In the Rhizo- stomidas they are distributed along the numerous grooves which carry the oral pores. Two kinds of marginal filaments may be observed in the Lucern- arias ; in one division (L. cyathif ormis) the filaments which beset the edge of the cup-shaped body are just like those which are found in the Medusa?, but they may be seen to be broken up into eight groups ; in the other (L. auricula) they form eight tufts placed on the ends of the four processes which project from the body. The tentacles of the Anthozoa are different in the large groups of that class. Eight lamellar tentacles, indented or feathered in appear- ance, surround the mouth of the Octactinia. There is a much larger number of cylindrical tentacles in the Hexactinia. They surround INTEGUMENT OF CCELENTEEATA. 103 the oral surface of the body, or are scattered over it ; tliey are sometimes produced on to lobate processes also. lu the Ctenophora, processes of no great size are also occasionally present on the edge of the mouth in some families (Calymnidas, Callianirida?) ; and there are large elevated lobate extensions of the body, which we may regard as tentacular organs, although morpho- logically they are different structures. Besides these, some genera (Oydippidss) have " grappling-lines," which resemble the marginal filaments in the Medusas, and correspond in position with the poles of an interradial transverse axis : sometimes they are provided with secondary appendages. Integument. §80. The most primitive characters of the integument of the Ccelen- terata are seen in the Spongia3, where it is conrposed of the ecto- derm, which is but slightly differentiated, and follows the various changes of form in the endoderm, which limits the nutritive cavity. The special characters consequent on this relation are referred to below (§ 87). In the Physemaria) the cells of the ectoderm form a syncytium. In the Porifera they may be sometimes seen to form a thin layer (Halisarcina, Sycon). Among the Acalephas, the ectoderm undergoes differentiation very early, so that the most external layer of cells, or epidermis, which is distributed over the whole body, represents in most cases a portion only of the primitive ectodermal layer. The investment of cilia, which in the Spongiaa is limited to the earlier stages of development, not only persists in the Acalephas during the so- called larval stages, when it has a locomotor function, but is frequently continued on into the later stages, when it is generally limited to separate parts, e.g. the tentacles. As the body increases in size, the importance of the cilia, as locomotor organs, disappears. In one class only — the Ctenophora — do they retain this function, and they are then increased in size. In the place of the general investment, as seen in the larva, structures resembling cilia are disposed in longitudinal rows, and by increasing in length and breadth become converted into movable swimming or rowing plates. The plates are attached to the body by their broader base ; it is at this point only that contractility, dependent on the voluntary influence of the animal, is manifested ; the rest and larger portion of the plate seems to be rigid. There are generally eight rows of these plates, which act as steering organs ; but in many Ctenophora there are only four rows (Cestum). The urticating capsules (nemocysts) are special differentiations of the epithelial 10 i COMPARATIVE ANATOMY. Ji elements; which are found in all the Acalephas, although they are not confined to them. They are firm capsules (Fig. 35, B), which are formed in the protoplasm of the 1 A z cell, and in which an elastic, spirally- coiled §A thread (A) is found ; this is generally emitted in the form of a stiff body, when the cap- siile is touched. These " stinging organs " are sometimes solitary, sometimes in groups, and at times they are very regularly arranged. They often become greatly complicated, as in the stinging knots of the Siphonophora, where they are often arranged in spiral bands. These ' ' stinging batteries " develop on the surface, but are often provided with a special investment formed by a fold of the integument. Although these structures are scattered over the whole surface of the body, and are not absent even from the endoderm and its products, yet many parts of the body are specially characterised by them ; above all, the tentacles, or other processes of the body. The urticating capsules vary greatly in form, as does the filament in structure ; and these differences are characteristic of the different divisions. Further, the ectoderm has a secreting activity, by which tests, which more or less invest the body, are formed. They are very common among the Hydroid-Polyps, where they are formed of a firm substance allied to chitin, and are often provided with various sculpturiugs, flutings, spines, ridges, and so on. These tubular tests are especially found among the colonial Hydroid-Polyps ; they are sometimes limited to the fixed portion of the common stock (Hydractinia), sometimes continued on to the branches of the stock (Tubularia, Eudendrium, Pennaria), and sometimes they are found even on the separate persons (Campanularia, Sertularia). This provides the soft polyp-stock with an organ of support, by which it is enabled to raise itself from the ground, as well as to attach itself. Fig. 35. Different forms of urticating capsules, a Cap- sule of Corynactis ; 1 With the filament spirally coiled ; 2 Extended. B C Capsules of Siphonophora : the fila- ments extended and partly provided with hooks. D Ur- ticating cells of Medusas ; filaments still rolled up ; in one not yet differentiated, and the nucleus of the cell still visible. SKELETON OF C Dermal canals. w Ciliated chambers. The difference between the ectoderm and endoderm is represented in the same way as in the previous figure (after E. Hiickel). ALIMENTARY CANAL OF CCELENTEKATA. 113 by dermal pores only. The primitive enteric cavity in them, as in the Leucones, where it loses its layer of flagellate cells (endoderm), also loses its nutritive functions, which are confined to the radial tubes. These latter seldom remain free, but generally unite either, in part or completely, by their walls with an important layer which surrounds the primary enteric cavity. A system of canals, which are only invested by ectoderm, is formed out of the space between the radial tubes when these only fuse in part. This form may be seen in the Sycones, among the Calcareous Sponges. Innumerable modifications, including individual varieties, are present within the range of one type-form. The primary enteric cavity is altered in character by the formation of diverticula, as well as of septa, or trabecules, and may even be completely atrophied, while the canal system arising from it is developed ; this phenomenon (lipogastria) is not uncommon in the Fibrous and Siliceous Sponges. A similar atrophy may even affect the mouth (lipostomia) without affecting the stomach; in such a case the dermal pores take on the function of ingestive canals; or numerous small spaces, as in Euplectella, arise in the place of the mouth. § 88. The form of the gastric system is greatly affected by the forma- tion of stocks, a process due partly to the concrescence of free persons, and partly to budding. The union, according to the degree of its development, may then simply cause communication between the stomachal cavities, which persist for each person (Fig. 41), or lead to a complete union of the cavities ; in which case the mouths too may undergo reduction, or become reduced to one, which likewise may disappear. A special system of cavities (inter-canal system) also arises from the formation of stocks. This system is formed from the spaces which persist between the unconnected parts of the persons, or the anastomosing branches, of the body ; this, like the system mentioned above in the Sycones, is bounded by the ectoderm only, and is thus essentially distinguished from the gastric system. It is remarkable for considerable irregularities in its arrangement, and forms also wider spaces, which deceptively resemble a stomachal cavity in that they possess a mouth. From all these arrangements a significant change of function in different parts is seen to accompany the change of form in the Spongias. The physiological activity of the digestive cavity is not only shared by the secondary canals which arise from it, but even passes away from it altogether, or is limited to separate portions of it, when consequently the stomach sinks, physiologically speaking, to a lower grade. On the other hand, an important func- tion becomes localised by this change of the primitively subordinate portions of the canal system, and even the original surface of the body of the sponges gets a higher significance, in virtue of its serving as the lining of the inter-canal system. Everything distinctly 114 « COMPARATIVE ANATOMY. shows that the organisation of the Spongias is not only very vari- able, but also that to understand it, it is absolutely necessary to distinguish clearly between the physiological and the morphological value of an organ. § 89. In its earliest characters the formation of the enteric cavity of the Acalephae agrees with that of the Spongiaa, but in the matured condition there are peculiarities in the Acalepha3, owing to the greater regularity in the arrangement of the system, which is developed out of a simple cavity. The mouth, the extent of which is often increased by the development of accessory parts in its neighbourhood, leads into the digestive cavity, and serves also as an opening for the excretion of undigested matters. The principal cavity seldom remains single, but grows out into secondary cavities, which have the character of pouches, or of canals, and which also, as a rule, correspond to physiological differences, since by them the chyme which is contained in them is distributed through the body of the person, and of the stock. These accessory spaces of the digestive cavity, included with the latter under the designation " gastrovascular system," undertake the function of a circulatory system, without being, morphologically, anything else than the dif- ferentiations of a primitive enteric cavity. The gastric system of the Acalephaa agrees therefore genetically with that of the Spongia3, but is distinguished from it by the exhibition of a higher differentia- tion. This is seen in the difference between the accessory spaces and the central primary one, which forms the stomach, to which its functions are ordinarily limited, and which are not, as in the Spongire, handed over to the secondary spaces. § 90. The simplest form of the gastrovascular system is found in the Hydroida. In Hydra it forms a sjDace traversing the long axis of the body, which commences with the mouth, in the middle of the circlet of tentacles, and is continued from the next portion, the stomach, which is capable of great extension, into the thinner portion of the body, where it is narrower. This space is also con- tinued into the tentacles. In the Hydroid-Polyps which form colonies, the canal which arises from the stomach runs through the whole stock, and makes the gastrovascular system common to all the persons. In the stocks of the Siphonophora, some persons only are set apart for the ingestion of nutriment. Each corresponds in structure to the stomachal tube of a Medusa, and forms a tube capable of great extension, which is connected at its base with the general cavitary system of the stock. In this case, then, we must sup- pose that this sort of individuals has lost all the arrangements found in the body of a Medusa, with the exception of the stomach (§ 75). The gastric system of the Medusas (Hydromedusas as well as Discophorn) presents numerous variations. It always occupies the ALIMENTARY CANAL OF CGELENTERATA. 115 Fig. 43. A Thauniantias. A From the lower surface. B Seen in section. In the middle of the body is the stomach, from which the radial canals pass to the circular canal. concave portion of the gelatinous disc, and consists of a stomach placed in the middle of this cavity, and of hollow spaces which proceed from it. The stomach either lies directly beneath this surface, or is placed on a special stalk, which arises from it, and is often of considerable size. This free projec- tion of an organ, which in other animals is hidden within the body, is explained by the differentiation of the stomach of Hydrome- duste from the most anterior portion of the body of the Hydroid-Polyps, so that it does not represent a single organ, but a complete portion of the body. The mouth is generally surrounded by tentacular organs, or pointed prolongations of the wall of the stomach ; it seldom opens into a narrow portion resem- bling an oesophagus. In most Hydromedusaa the stomach is separated from the space that lies behind it by a ring which is developed at its base ; by the contraction of this ring, the stomachal cavity can be shut off from the rest of the gastro vascular system. The stomach varies greatly in form and size. It projects far beyond the edge of the bell-shaped umbrella in the SarsiadaB. The hollow spaces which are distributed in the sub-umbrella arise from the base of the stomach, or from the space which lies behind it, and have the form of narrow canals, or of wide pouch-like diverticula. The narrow canals take a radial course (Fig. 43) to the edge of the um- brella ; they are simple or regularly branched, and they open into the circular canal, which often sends processes into the marginal ten- tacles also. On their way to the margin, the radial canals may form diverticula, which are functionally connected with the generative apparatus (cf. § 96). In theiEginidas and Discophora the gastric cavity passes directly into the radial enlargements; these latter are derived from simpler canals. Narrower canals sometimes, indeed, alternate with wider spaces. The canals are branched (Fig. 44, gv), or form, as in the Rhizostomidaa, a peripheral network. As the gelatinous substance I 2 Fig. 44. Aurelia aurita. Half of the lower surface is seen, a Marginal bodies. b Oral arms, v Gastric cavity, gv Canals of the gastrovascular system, which branch towards the edge and unite into a circular canal, ov Ovaries. 116 COMPARATIVE ANATOMY. of the umbrella is continued into the wall of the stomach in the Discophora, the stomach is not very sharply marked off from the rest of the gastric system. Its wall is always continued into arm-like appendages, which, as a rule, project into folded membranes (oral arms) ; the mouth is placed between these. Division of these arms leads to further modifications, which give rise to greatly ramified appendages. In this case numerous grooves, which gradually unite, lead to the mouth, in correspondence with the form of the arms. In the Khizostomidas the mouth remains open during an early period of development only, and afterwards becomes closed by the gradual union of the " arms," which limit it, and in which the grooves form branched canals, which open at the ends of the ramifications of the arms by numerous fine pores (polystomia). In the Lucernariae the structural conditions of the gastro- vascular system closely resemble those of the Medusas. A stomachal tube, projecting from the concave surface of the umbrella, and pro- duced into four angles, leads into a wide space, which is continued into four pouches, and may be elongated into four canals, which pass into the stalk. The four pouches correspond to widened radial canals, and are, as in the Medusas, connected with one another at the edge of the umbrella, and so form a circular canal. In others this character is modified in such a way that the stomach is continued into the body, in a tubular form ; and at its end, which projects into the stalk, gives rise to radial canals, which whilst becoming enlarged run outwards towards the margin of the disc. The gastrovascular system in the larva? of the Discophora and in Scyphostoma is very similar in character. § 91. The gastric system of the Anthozoa ex- tends by means of an oesophagus from the centre of the tentacle- bearing surface of the body into the interior, where it opens into the digestive cavity. From this part canals pass upwards along the sides of the oesopha- gus into the tentacles. Owing to the width of the canals connected with the stomach, the intermediate tissue is reduced to a mere partition (s), which extends in rays from the wall of the body to the wall of the Fig. 45. Transverse section through a portion of the stock of Alcyoninni, in which two individuals, A A, are cut through just below their junction with the ccenen- chyma, and a third, B, somewhat lower, v Wall of oesophagus, c Eadial canals (chambers of the body. cavity), s Septa, o Ova. Part of the coenenchyma traversed by canals is seen to contain calcareous bodies. ALIMENTARY CANAL OF CCELENTERATA. 117 oesophagus. The canals thus appear to be chambers (c) attached to the oesophagus, which unite into a common central space the digestive cavity, or stomach (B), and so communicate with the oesophagus. The number of these chambers is eight in the Octactiniae, and varies in the other Anthozoa, but is arranged according to the same law of numbers, as is expressed in the other characters of their organisa- tion, as for instance in the number of the tentacles. The septa of the gastrovascular system are usually continued for some distance along the wall of the digestive cavity, and terminate as elongated bands or pads. When, therefore, the stock is calcified interradial lamellae are formed, passing inwards from the wall between the gastric lamellae. In the colonial Anthozoa, the central cavity is connected in each person by means of a canal system which traverses the coenenchyma (Fig. 45), and thus every individual is directly connected with the rest. This canal system forms a network of tubes of various widths which distribute the nutritive fluid in the stock. At one point of the common trunk, in the stocks of the Octactinia3, several canals are united into a wider space, from which an orifice leads to the exterior; this, probably, serves as a means of regulating the ingress and egress of the water which flows through the gastrovascular system (Pennatula, Renilla). A similar opening has been observed in Cereanthus ; it corresponds to the pore of the Hydrae, and like it is placed at the aboral end of the body. These arrange- ments, which give to the gastric system the significance of a water vascular system, have, in many Anthozoa (Corals), the form of fine pores scattered over the surface of the body; they can only be perceived at the moment they are in function — that is, when expelling water. Similar pores are also found on the tips of the tentacles in many Actinias, etc. All these arrangements call to mind the dermal pores of the Spongiae. In the Pennatulidae and Alcyonidae (Sar- cophyton) some, and at times many, per- sons in a colony are less well-developed, and seem to have lost the function of ingesting food. It is not known whether they have any share in the taking in of water. §92. Fig. 46. The gastrovascular system of a Cydippe. A Lateral view ; the mouth turned upwards. B Seen from the oral pole. Iii the Ctenophora, the nutrient cavitary system differs in details only. A stomach, which is very wide in the Beroidae, and nar- rower in the rest, is sunk in the body along its longitudinal axis ; it passes into a space which is known as the "funnel/' by means of a narrow canal, which can be closed by muscles. Radial canals (Fig. 46) pass out from the funnel and run along the ciliated ribs or " ctenophores." The radial 118 COMPARATIVE ANATOMY. canals in the Beroi'das and CallianiridaB pass into a circular canal at the oral pole. In the latter, two canals, which run along the sides of the wall of the stomach, and which come from the funnel, also enter the circular canal. In the Cydippidas, these are very wide, and appear to form a common space around the stomach. Finally, two shorter canals, which do not pass directly from the funnel, but from the canals derived from it, run outwards and open by pores, which can be closed, at the sides of the polar areas (cf. p. 111). They are placed diagonally, and provide the gastrovascular system with a second means of communication with the surrounding water. The form of the body leads to various modifications in the arrangement of this system of canals. The various groups of canals may be branched. Thus, in the Beroidae the radial canals form lateral branched diverticula, which are present, though they are not so large, in the other forms, and are in connection with the generative apparatus. § * o In some divisions of the Acalephas there are filamentous structures, which project into the central cavity of the gastrovascular system ; they are called Gastric filaments (and, though less appropriately, mesenteric filaments) . They are found for example in the LucernaridaB and Discophora. In the latter they form tufts of filaments, placed in diverticula of this cavity, and execute vermiform movements. They have similar characters in the Lucernaridas, but are different in the Anthozoa. In the Anthozoa pad-like processes, richly provided with stinging cells, run along the free edge of each septum, turned towards the gastric cavity ; they seldom become filamentous, and are some- times limited to two of the septa (Tubipora). Nothing is known as to the function of these organs, which are differentiated very early. Although glandular organs do not seem to be differentiated in the digestive cavity of the Ccelenterata, yet there is an arrangement which should be noted here ; it may be regarded as an indication of a secreting system, perhaps analogous to the liver of higher animals. It is this, namely, that the epithelial investment of the stomach, which is present in many Ccelenterata, is distinguished by its peculiar colour. The pigmented cells are set longitudinally, and are generally placed on the projecting folds of the wall of the stomach in the Anthozoa; they are also developed in the Hydromedusaa, even in the polyp forms (e.g. Tubularia)) ; in the Siphonophora they form distinct pad-like longitudinal rows at the base of the digestive cavity of the nutrient individuals. A network of " hepatic canals," attached to the single large stomach of the Velellida), appears to be specially differentiated ; it is found on the under surface of the disc. SEXUAL ORGANS OF CCELENTEEATA. 119 Sexual Organs. § 94. Sexual differentiation is not the solo factor in reproduction among the Ccelenterata, for various forms of asexual multiplication (cf. supra, §§ 73-77) obtain among them. Sexual products have "been observed in most of them, but they are not formed in organs set .apart ; the function seems rather to be one which is being gradually localised. In the Spongias the endoderm is said to be the place where these products are formed, but in those Porifera which have a mesoderm, the differentiation appears to take place in it. The history of the ova is best known ; they arise from cells in the mesoderm, but they are perhaps endodermal cells which have passed into it. In addition to what has been directly observed in this group we must bear in mind the characters which obtain in the Hydroid-Polyps (see below). The male elements have been less widely observed. The endoderm has been said to be the place where the seminal cells are formed, but masses of sperm have been observed in the mesoderm of Halisarca, together with a sexual differentiation of the stocks. § 95. The place where the generative matters are formed — as a rule in the walls of the digestive cavity, or the spaces leading from it — is most exactly known in the Hydroida among the Acale- pha3. The material of the two kinds of generative pro- ducts is, however, provided by different layers of the body : this fact deserves to be exactly described on account of its fundamental import- ance. The first, or indifferent stage, is repi*esented by di- verticula of the wall of the body, which have the form of buds, surrounding a pro- longation of the gastric cavity, and formed by the ectoderm and endoderm. A number of the cells of the ectoderm (a) of the growing bud (Fig. 47, J. B) enlarge and become distinguished by their size from the other endodermal cells, which bound the gastric cavity (g) . These enlarged cells are pushed out towards the Fig. 47. Two female generative buds of Hy- draotinia echinata. a Ectoderm, b Endo- derm. ;/ Gastric cavity. o Ovarian germs. In A the ectoderm has begun to be pushed into the endoderm. In B the invaginated portion has been constricted off from the rest of the ectoderm (after Ed. van Beneden). 120 COMPARATIVE ANATOMY, ectoderm, and are the ova (o). They gradually form a layer of cells placed apparently between the ectoderm and endoderm, and give to the whole bud the appearance of an ovary. While these processes of differentiation are going on in the endoderm, a growth of cells from the ectoderm at the tip of the bud is extending inwards (A) ; as these cells become separated off from the ectoderm (B), they form a thin lamella, which grows around the ovarian layer, but which has no further function except in another kind of bud. In the male bud, in fact, the ectoderm has the same characters, but the endoderm does not undergo any change, and simply forms a layer of cells, investing the gastric cavity without being differen- tiated into ova. The depressed portion of the ectoderm being developed to a great size, forms by constriction a layer between the ectoderm and endoderm (Fig. 48, ABC), the cells of which give rise, later on, to the morphological elements of the sperm. In this B Fig. 48. Three male generative buds of Hydractinia echiuata. a Testes. Other letters as in Fig. 47 (after Ed. van Benedeu). way the male products of generation arise from the ectoderm, just as the female products are formed from the endoderm. The fact that even in the female buds the ectoderm is depressed, leads us to sup- pose that the buds were primitively hermaphrodite. It is not yet known how far the generative products have separate origins in the rest of the Acalephse. The possibility of cellular elements having passed from one layer to another at a very early period of develop- ment may account for the fact that the endoderm appears to be the layer in which the products of both sexes are formed. Hydra appears to form an exception, for in it the generative products are formed in external bud-like organs, which are differentiations of the ectoderm. Among the Hydromedusas we not unfrequently meet with a separation of the sexes, not only into different persons, but even into different colonies ; in the Siphonophora hermaphrodite colonies only are found as the rule, but there are exceptions to this. The generative products give rise to more or less considerable swell- SEXUAL ORGANS OF CCELENTERATA. 121 ings in tlio parts of tlie body wliore they are formed, but as they are only present when the generative matters are being formed, they may be regarded as "temporary organs." There are great peculiarities in the structural relations of the parts which enclose the generative products, but they are connected by a large number of intermediate stages. In those Hydroid colonies that give rise to free Medusas (cf. § 74), the Medusas carry the generative organs ; the Medusas form the generative animals of their proper Hydroid- Polyps, and elaborate the semen or ova in the walls of their stomach, or in the radial canals, or, lastly, sometimes in the circular canal. In some this production does not take place until a long time after they have broken away from the Hydroid colony ; in others it happens earlier. The latter bring us to those in which the reproductive matters are formed while the Medusas are still attached to the Hydroid stock. Next, then, comes that stage in which the Medusa is not only not broken off, but is not completely developed. All the organs which are of use in the full and indepen- dent mode of life — mouth, gastric cavity, tentacles, swimming-bell, &c. — appear in a state of atrophy. We have in fact medusoid buds, in which the sexual products arise. In others the medusoid form is completely lost, and quite simple structures appear on the Hydroid stock in the form of generative capsules, into which, at the most, a process of the gastrovascular system still projects. Such are the structures described above in Hydractinia. These generative buds arise, like the medusoid buds, and the Medusas themselves, sometimes on the common stock, sometimes on the body of the Polyps, and often only at certain parts of the latter, as, for example, between the outer and inner circlet of tentacles in the Tubularia. Where the proliferating Polyps are provided with a sheath, the generative buds are always enclosed by the same test as the Polyps themselves. Thus the phenomenon of the budding of Medusas can be followed back to a stage in which the bud has the appearance of a mere generative organ of the hydroid stock. The Siphonophora resemble the Hydroid-Polyps. The formation in them of sexual animals of the Medusa-type, with the simul- taneous formation of other medusiform persons, helps to explain the phasnomenon known in the Hydroid-Polyps as alternation of generation, as being a division of labour. In some of the Siphono- phora the generative animals become free Medusas, in the walls of whose stomachs the generative products are formed (Velella, Chry- somitra). Most of them have only medusiform buds, which are found in very various stages of degeneration (cf. Fig. 33, B g E). The stomach of the Medusa becomes gradually represented by the generative organs only, and the Medusa-bell degenerates into a mere covering for them. Thus they occur arranged either singly (Diphy'idas), or grouped into racemose tufts (Physophoridas), which are placed on the stem of the stock, or, it may be, on definite persons belonging to it. Ed. van Bexeden. De la distinction originelle du testicule et de l'ovaire. Bull. Acad. Bclg. 2me Ser. T. xxxvii. 5. — G. Koch, Morph. Jahrb. Bd. ii. p. 83. 122 COMPARATIVE ANATOMY, § 06. In those Medusae which have no longer any relation to Hydroids, the generative matters are formed in the wall of the gastro- vascular system, just as in the Medusas of the Hydroid-Polyps, and the Siphonophora. These matters are generally formed in the radial canals (iEquoridas), or in the pouch-like diverticula of the stomach (^Eginidas) . When the canals are narrow the genitals form freely projecting diverti- cula, which, when much de- veloped, may have the form of ruff-like folds. The radial canals form lamellar enlargements intheGeryo- nida?, when the generative matters are developed. In all forms the lower wall of the canal, or that placed away from the umbrella, forms the genital region (Fig. 85, g). The germinal matters in some cases reach the exterior through the stomach, and in other cases by a rupture of the tissue. In the Discophora the rgans have always much the same re- lations, and differ but little in position and form. They consist of four or eio-ht frills, curved in a semi- lunar form, and arranged in a rosette on the inner surface of the umbrella (cf. supra, Fig. 44, ov) ; the frills are formed by diverticula of the gastrovascular system. They are placed in depressions on the lower surface of the disc, or hang freely down from it, often in numerous folds. The Lucernarias have the generative organs in the form of eight radially-arranged longitudinal ridges, placed on that part of the body which corresponds to the sub-umbrella of the Medusas, where they form projections into the pouches of the gastrovascular system. They represent, therefore, an intermediate form between the Hydro- medusas and Discophora. Fig. 49. Diagram of a radial vertical section through a sexually mature Geryonid (Carmariua hastata) ; on the right it is taken through the whole length of a radial canal, and on the left through the lateral wings of a genital lamella, at an interradial plane, b Marginal vesicles, c Cir- cular vessel, g Generative products, h Clasp of the mantle, h Stomach. I Gelatinous umbrella. p Stomachal stalk, r Eadial canal, s I Its inner ; r s Its outer wall, u 1c Cartilaginous ring, v Ve- lum. Z Tongue-like process (after E. Hackel). generative SEXUAL ORGANS OF CCELENTEEATA. 123 § 07. The generative organs of the Anthozoa are very similar in cha- racter. They are connected with the gastric cavity, so that the gene- rative matters pass to the exterior through the gullet. The septa of the gastric spaces, or their ridges, which project into the central stomachal cavity, generally function as organs of this kind. In the Alcyonarians the sexual products are formed at the free edge of these processes, either in the stomach or at some distance from it, at the base of the gastric space; two septa are sterile. They are distinguished by the presence of the ridges mentioned above, (§ 93), which project some way forwards. The other ridges do not always carry sexual products, for in many Alcyonarians they are found on four only, or only on two of them. In the Actinias the generative products are formed within the gastric ridges. The same thing happens in the Antipatharia (Gerardia). The Madreporinse, also, may be mentioned here, for in them the generative products are developed in the ridges, which project far into the base of the gastric cavity, and they there form a special process on each of the two surfaces of the ridges (Astroides calycularis) . The sexes are usually separated, and in different persons, but there may be hermaphrodites also (Cerianthus). In the colonial forms both dioecious and monoacious conditions have been observed ; in others these characters vary very greatly (Corallium rubrum). When the persons of a colony are dimorphic, those which are the more developed are at the same time those which are functionally sexual, while the others are sterile. But in some Peimatulida3 the persons without tentacles are the only ones in which there are generative organs (Virgu- laria mirabilis). § 08. Ha In the Ctenophora the peripheral portion of the gas- tric system represents the genital region. Cascal diver- ticula are developed from the sides of the canals which run parallel with the rows of nata- tory lamellae ; in these semen or ova are formed. One side of a canal is beset with . ovarian follicles, and the other with testicular lobules. Hermaphroditism repeats itself therefore in each of the radial segments of the body. The gastric system serves to carry the generative products to the exterior. This :?^%© Fig. 50. Generative organs of Beroe rufes- cens ; showing their relation to a tract of the radial canal, a Stripes running along the canal (Muscles F) b Semen-producing side, c Ovarian side, with eggs (after Will). 124 COMPARATIVE ANATOMY. arrangement therefore is exactly the same as in some of the Anthozoa, and if we compare the body-substance between two radial canals with a septum of the Anthozoa, we find that the genital regions of both sexes are arranged in just the same way as in the hermaphrodite Anthozoa. As a rule the ova of the Ccelenterata have no special coverings, and in many Spongias and Hydroida (e.g. Hydra) they appear to change in form in consequence of amoeboid movements. The seminal elements in the Acalephas are formed by a small head with a movable appendage. Third Section. Vermes. General Review. § 99. In this division a large number of animal forms, which are more or less allied to one another, are put together ; transverse axes are differentiated, whilst the longitudinal axis of the body is elabo- rated. In consequence of this an anterior and a posterior end can be made out, in addition to a dorsal and ventral surface. They differ markedly from the Ccelenterata in having two antimeres. The body is, or is not, divided into metameres ; in the more simple forms the metameres are simple in character, in the higher divisions they undergo differentiation. It is not certain whether this division does or does not represent a single phylum. The existence of a large number of small groups, represented merely by single forms, shows a considerable amount of divergence within the division ; and this is still further exhibited by the fact that almost all the higher animal phyla can be brought into more or less close connection with forms of Vermes. I arrange the various divisions of the Vermes in the following order. They might be considerably increased by the introduction of a large number of isolated genera ; but a complete classification of such a kind is not part of our purpose here. I. Platyhelminthes. Turbellaria. Rkabdocoela. Monocelis, Vortex, Mesostomum, Prostomum. Dendrocoela. Planaria, Leptoplana. 12G COMPARATIVE ANATOMY. Trematoda. Distoma, Monostomum, Tristoma, Polystomum, Aspidogastcr, Diplo- zoon, Gyrodactylus. Cestoda.* Caryophyllajus, Ligula, Bothryoccphalus, Tamia, Tetrarhynchus. Neinertina (Rlrynclioccela). Pelagouemertcs, Ncmertes, Polia, Borlasia. II. Nematlielraiuthes. Nematodes. Rkabditis, Dorylaiinus, SU'ongylus, Ascaris. Gordiacea. Gordius, Meroiis. III. Cheetognatlii.t Sagitta. IV. Acautliocephali. E chinorhynclius . V. Bryozoa.J Pliylactolasrna. Cristatella, Alcyonella, Lophopus, Pluinatella. Gyinnolosnia. Crisia, Homera, Alcyonidium, Flusfcra, Escliara. VI. Rotatoria. Hydatina, Notommata, Brachionus, Melicerta, Floscularia. VII. Enteropneusti. Balauoglossus. VIII. Gephyrea.§ Inermes. Sipunculus, Pliascalosoma, Priapulus. Cliastiferi. Ecliiurus, Bonellia. IX. Annulata. || * The Cestocla arc derived from a form common to them and to the Trematoda. The difference in organisation is due to their different kind of parasitic habits. There are several forms of which it is doubtful whether they belong to one or the other division (Amphiptyehes). f The Cha?toguathi must not be regarded as allied to the Nemathelminthes because they are put next to them ; the same remark holds gqod for the Acanthoccphali. % Pedicellina and Loxosoma are genera allied to the Bryozoa, and they might well be united with them into one division, but they must not bo subordinated to them. § In both divisions of the Gephyrea thore is a large number of very divergent forms. || Tomopteris, Myzostoma, and Polygordius are special forms, allied to, but very divergent from, the Annulata. The last mentioned unites the characters of the Nemertina and Nematodes with those of the Annelides. VERMES. 1-27 Hirudinea.* Ha?niopis, Sanguisuga, Nephclis, Clepsine. Anuelicles. Oligochaeta. Scoleina. Lumbricus, Chajtogastcr, Nais. Haliscolecina. Polyophthalmus, Capitella. CliEetopoda. Vagantia. Siphonostoma, Arcnicola, Glycera, Nephthys, rhyllorlocc, Alciopa, Syllis, Nereis, Eunice, Amphinome, Aphrodite, Polynoe. Tubicolre. Amphitrite, Hemiella, TerebeUa, Sabella, Serpula, Branchioninia. The position of the genera Neomenia and Chsetoderina is not yet settled ; but they must not be passed over, on account of the great importance of many points which have been made out in their organisation. Although they differ in not unimportant points from one another they are closely allied, and may be reckoned with the other divisions of the Vermes. I therefore unite them into a division, which I call that of the Solenogastres. It will not be possible to form any safe opinion as to their position till they are known more exactly; and for this knowledge, especially as regards their development, we must wait. Bibliography. V. Ba.er, Beitriige zur Kenntnias der niedcren Thiere. N. A. Acad. Leop. Carol. XIII. 1826.— Dujardin, Histoire nat. des Helminthes. Paris, 1845. — Van Beneden, Mciinoire sur les vers intestinaux. Paris, 1861. — Leuckart, R., Die menschlichen Parasiten. Leipzig unci Heidel- berg. I. II. 1863-76. — Claparede, Beobachtungen fiber Anatomie unci Entvrickelungsgesehichto wirbelloser Thiere, Leipzig, 1863. ON THE SEPARATE CLASSES. Flatyhelminthes : Duges, Recherches sur reorganisation et les mceurs des Planaires. Ann. sc. nat. Ser. I. T. XV. ; also Isis, 1830.— Nordhann, A. v., Micrographische Beitriige zur Naturge- schichte der wirbellosen Thiere. Erstes Heft. Berlin, 1S32.— Quatrefages, A. de, Mclmoire sur quelques Planaires marines. Ann. sc. nat. S6r. in. T. IV.— The same, Sur la i'amille des Nchnertines. lb. T. VI.— Schmidt, O., Die rhabdocolen Strudelwiirmer. Jena, 1818.— The same, Neue Beitriige zur Naturgeschichte der Warmer. Jena, 1848.— The same, Ueber Rhab- docolen. Wiener Sitzungsbericht. Math. Naturw. Classe. Bd. IX. S. 23. — The same, Ueber Dendrocolen. Zeitschr. f. w. Zoologie. X. XL— Van Beneden, Les vers cestoides. Mt'moires de l'Academie de Bruxelles. XXV." 1850.— The same, Recherches sur la faune littorale de Belgique, Turbellaries. lb. XXII. I860.— Leuckart, R, Mesostomum Ehrenbergii. Arch, fiir Nat. 1852. S. 234.— Schultze, M., Beitrage zur Naturgeschichte der Turbellarien Greif swalde. 1851.— The same. Ueber die Microstomeen. Arch. f. Nat., 1849. S. 280.— Wagenee, G., Die Ent- wickelung der Cestoden. N. A. L. C. T. XXIV. Suppl. 1854.— The same, Beitrage am- Ent- wickelungsgeschichte der Eingeweiflewurmer. Haarlem, 1857. — Stieda, Beitr. z. Anat. v. Both- ryocephalus. Arch, f. Anat. u. Phys. 1864.— Somher and Landois, Beitrage z. Anat. d. Platt- wiirmer. Zeitschr. f. wiss. Zool. 1872.— Schneider, A., Ueber Platyhelminthen. Giessen, 1873 * I would place the genus Brauchiobdclla, which is counted as one of the Hirudinea with the Scoleina among the Annelides. There is nothing leech-like in the organisation of this worm except its sucker and jaws; and these parts are really structures which are due merely to its parasitic mode of life. Rotatoria : Ehrenberg, Die Infusionthierchen, etc.— Leydig, Zur Anat. u. Entwic chichte tier Lacinularia socialis. Zeitschr. f. w. Zool. III. p. 452.— The same. Ub US COMPAEATIVE ANATOMY. — Zbller, E., Ueber Polyst. integer, z. f. w. Zool. XXII. p. 1. XXVII. p. 233.— The same, Ueber Leucochloritl. paradox, lb. XXIV. p. 564.— Sommer, Ueber de Bau v. Taenia mcdiocanellata. Z. f. vv. z. XXIV. p. 499. — Moseley, H. N., On the Anat. and Histol. of the Land Planarians of Ceylon. Trans. Royal Society. London, 1874. — M'Intosh, C, Structure of British Nemerteans, Edinb. Roy. Soc. XXV. II. p. 305. — Hubrecht, Unters iib. Neniertinen. Niederl. Arch. f. Zool. Bd.'ll.— Graff, L., Zur Kcnntniss der Turbellarien. Z. f. wiss. Zool. Bd. XXIV. Nernathelminthes : Cloquet, Anatomie des vers intestinaux. Paris, 1824.— Eberth, Unter- suchungen fiber Nematoden. Leipzig, 1863. — Schneider, Monographic der Nematodes. Berlin, 1866. — Bastian, Monograph on the Anguillulidse. Trans. Linn. Soc. vol. XXV. p. II. 1865. — Grenacher, Zur Anat. der Gattung Gordius. Z. f. w. Z. XVIII. p. 322. — Clatjs, Ubcr Leptodera appendiculata. Marburg and Leipzig, 1869. — Butschli, Beitr. z. Kenntn. der freile- benden Nematoden. N. A. L. 0. XXXVI. — The same, Z. f. w. z. XXIV. p. 361.— The same, Abhandl. d. Senkenb. Ges. IX. Chsetognatha : Kroiin, Anatomiseh-physiol. Beobachtungen iiber die Sagitta bipunctata. Hamburg, 1844.— The same, Nachtriigliche Bemerkungeudazu. Arch, fur Naturgesch. 1853.— Wilms, Observationes de Sagitta. Diss. Berol, 1846. Bryozoa : Van Beneden, Recherches sur l'anatomie, la physiologie et I'embryogenie des Brvo- zoaires. Memoires de l'Acad. Royale de Belgique. 1845 et seq. — The same, Recherches sur les Bryozoaires fluviatiles de Belgique. lb. 1847. — The same and Dumortier, Histoire naturelle des Polypes composes d'eau douce. lb. 1850. — Allman, A Monograph of the Fresh-water Polyzoa. London, 1S56. (R.S.) — H. Nitsche, Beitriige zur Anat. und Entwickelungsgeschichte der phy- lactoliimen Siisswasserbryozoen. Arch. f. Anat. u. Phys. 1868. p. 465. — The same, Beit. z. Kenntn. d. Bryozoen. Zeitschr. f. wiss. Zool. XX. XXI. XXIV. — Vogt, C, Sur le Loxosoina. Arch, de Zool. Exp. Vol. V. fickclungsges- , Uber Bau und systemat. Stellung der Riiderthiere. lb. VI. p. 1.— Huxley, Quart. Joum. of Microsc. Sc. 1852— Cohn, F., Zeitschr. f. w. Zool. VIII. p. 431. LX. p. 284. XII. p. 197. Enteropneusti : Kowalevsky, Memoires de 1' Academic de St. Petersbourg. Ser. 7. T. X. No. 3.— Agassiz, A., Mi?m. Amer. Acad. IX. Gephyrea: Grube, Versuch einer Anatomie des Sipunculus nudus. Arch. f. A. u. Phys. 1837. p. 237.— Krohn, Ueber Thalassema. Arch. f. Anat. u. Phys. 1842. — Quatrefages, A. de, Memoire sur l'Echiure. Ann. sc. nat. 3 Ser. T. VII.— Muller, M., Observationes anatomica? de vermibus quibusdam maritimis. Berol. 1852.— Schmarda, Zur Naturgcschichte der Adria. Wien. Denkschrift math. Naturvv. CI. Bd. 3. 1852.— Lacaze-Duthiers, H., Recherches sur la Bonellia. Ann. sc. nat. 4 Ser. T. X.— Theel, H., Rech. sur les Phascolion. K. Vet. Acad. Handl. XIV. Annulata : Audouin and Milne-Edwards, Classification des Annelides et description des cedes qui habitent les cotes de la France. Ann. sc. nat. T. XXVII— XXX. 1832-33.— Milne-Edwards' Article : Annelides, in Todd's Cycloptedia, I. 1835.— Grube, De Pleione carunculata. Regiomonti, 1837.— The same, Zur Anatomie und Physiologic der Kiemenwiirrncr. Konigsberg, 1838.— The same, Die Familien der Anneliden. Arch, fiir Naturgesch. 1850.— Quatrefages, Etudes sur les types inferieures de l'embranchment des anneles. Ann. sc. nat. Ser. 3. Tomes X. XII. XIII. XIV. XVIII. 1828-52. (The results are given in the " Histoire nat. des Anneles " of the same Author.) — Leydig, Zur Anatomie von Pisicola geometrica. Zeitscher. fiir. w. Zoolog. I.— Buchholz, Beitriige zur Anat. der Gattung Enchytrseus. Konigsberger rhysikal.-ffikonomische Schriften. III. 1862.— Claparede, Recherches anatomiques sur les Annelides, etc. Geneve, 1861.— The same, Rech. anat. sur les Oligochetes. Geneve, 1862.— The same, Glanurcs zooto- miqucs parmi les Annelides. Geneve, 1864.— The same, Les Annelides Chftopodes du Golfe do Naples. Geneve et Bale, 1868. Supplement, 1870.— The same, Histol. Untersuchungen iiber d. Regenwurm. Zeitsch. f. wiss. Zool. XIX.— Perrier, E., Etudes sur l'organis. des Lombriciens terrestres. Arch, dc Zool. III.— Gheefp, A., Unters. iiber die Alciopiden. N. Acta. Ac. Leop. Carol. XXXIX. Solenogastres : Tullberg, P., Neomenia, a new genus of invertebrate animals. Bihang till K. Svenskavet. acad. Handlinger III.— Graff, L., Anat. v. Chwtoderma. nitid. Zeitsch. f. w. Zool. Bd. XXVI.— The same, Ueber Neomenia u. Chtetoderma. lb. Bd. XXVIH. Form of the Body. § 100. The radiate form of body, which obtains in most Ccolenterata,, is never developed in the Vermes. It is replaced by the eudipleural form, which is generally known as that of bilateral symmetry (cf .^ supny p. 59). It predominates in all higher divisions of the Animal Kingdom. Although in some stages, as for example in the scolex-form of many Cestoda, this differentiation of the secondary FOIiM OF THE BODY OF VERMES. 129 axes is not expressed, so that similar characters to tliose which obtain in the Coelenterata can be made out, yet I do not regard this condition as one that has been inherited by the Cestoda, for they can only be derived from forms, which like the rest of the Platyhelminthes, possessed the original eudipleural form. Their condition, which depends on the equal development of the secondary axes, is at once explained by their loss of the power of locomotion, and by the attachment of the body by a point, which corresponds to one pole of the primary axis. A head, which has a mouth placed as a rule somewhere on its ventral surface, can generally be distinguished at the oral pole. In most Platyhelminthes the mouth is some distance from the head ; in the Turbellaria, indeed, it is generally some way back on the ventral surface of the body. The aboral end of the body carries the anus ; this, when present, has ordinarily a dorsal position. In the fixed Vermes the form of the body undergoes considerable modifications. It is greatly influenced by the development of a covering, as in the Bryozoa. The aboral end of the body, by which the animal is attached, can no longer carry the anus, which is accordingly placed nearer to the anterior portion of the body, which is not enclosed by the cell. § 101. Another phenomenon which is first seen among Vermes is the segmentation of the body. Already in the Rotatoria the hinder portion of the body 1. 2. is adapted to locomotion by being broken up into a number of segments. In this we see an indication of a condition which becomes very important in the higher divi- sions. In the Cestoda it is further de- veloped. A differentiation is occasioned by the growth of the body in the direction of its primary axis. The anterior and pos- terior parts of the body no longer enclose the same organs. Thus in the Caryophyl- lafsi the hinder portion of the body alone contains the generative organs. In Ligula this hinder portion of the body is consider- ably developed by the great repetition of the generative organs. In the Tseniadae a very large series of these generative rarhynchns) ; asexu: ? i -i • ,i i • i i r (nurse). 2. The same iu the organs are developed in the hinder end of >oint_forming stage (strobila)> the body, and each corresponding area in which the last joints (pro. forms a joint, which is gradually marked glottids) are breaking off, one off on the outer surface, and has the rela- by°ne(afterP.J.vanBeneden). tions of a metamere to the other joints (Fig. 51). In this way the Tasnia-chain is formed, the last meta- meres of which (the so-called proglottids) break off at a certain Fit?. 51. Taenia (Tot- al form 130 COMPARATIVE ANATOMY. stage of development, and form more or less independent individuals. This process represents therefore a process of gemmation, the pro- duct of which is the Taenia-chain ; each separate joint is a metamere as compared with the general organism of the chain, but it is also to be regarded as a separate person, since it is capable of an indepen- dent existence, the slight duration of which is explained by its struc- ture, which is adapted to a parasitic habit. The metamerism of the body, seen in the Cestoda, may be derived from a process of gem- mation, and gemmation itself is correlated with the elongation of the body. It is an intermediate stage between the two phenomena ; and there is, therefore, no well-defined antagonism between them. Where metamerism is faintly expressed, it becomes more and more nearly a case of simple elongation. In many divisions we may find examples of this incomplete metamerism. It is indicated in various systems of organs in the Nemertina. In the Gephyrea also, metamerism is not a general phamomenon, for several systems of organs are not affected by it. On the other hand, it obtains generally in the Annulata, and gives to the organism a multifid appearance. In them it is not unfrequently associated with a process of distinct gemmation. In the embryonic body there are generally fewer metameres than in the adult. The freshly-developed segments are formed in front of the last one. The elaboration of particular metameres gives rise to a large number of modifications. Such also result when a number of metameres undergo concrescence, and the primitive arrangement is only indicated by certain systems of organs ; this gives rise to conditions which it is difficult to distinguish from those in which metamerism is just commencing. When metameres are developed the organism becomes one of a higher grade of organisation, although indeed this is not the only path towards such a stage, for we meet with differentiations of other kinds which lead to higher conditions. The more definite differentiation of the ventral surface owing to the development of a groove, as in the Solenogastres (in Chastoderma this is found in the posterior region of the body only) is an example of this ; it represents the first stage in the formation of that pedal surface of the body which is seen in the lowest Mollusca. § 102. There are various other modifications in certain smaller divisions, which are to be attributed to adaptations to changed external conditions of life; this is especially the case in the entoparasitic Platyhelminthes. The " cystic form " must be regarded as the most important of these modifications ; this, which is intercalated into the developmental history of the Cestoda, is, in its phylogenetic history, just as certainly due to the organism having entered into relations which at first were strange and abnormal to it, as is the general parasitism itself referable to habits, which were only secondarily acquired. The phylogenetic history of the cystic form FOEM OF THE BODY OF VEEMES. 131 Fig. 52. Young Taenia, with head pushed in. aHead. b Envelope, c The six embryonic hooks, remaining at one point of the envelope (after V. Siebold). Fig. 53. The same Taenia, with head protruded. Letters as in Fig. 52 (after V. Sic- bold). is based on the notion that certain abnormal external conditions of life gradually became normal, in consequence of the adaptation of the organism to them, and that it did not arise by a simple antagonism to the worm's primitive ontogenesis, which now includes this cystic form as a normal part of its cycle. What has happened is this — that the process of adaptation has seized upon and exaggerated a normal inherited phase of the worm's ontogeny, and in virtue of the continuation of conditions favourable to the appearance of this exaggerated phase, its appearance has become a normal phenomenon. The variations of the cystic form are all readily deducible from the first developmental stage of the Cestoda. The embryo is generally provided with three pairs of hooks, and a cestoid head may be observed to be differentiated within it (Fig. 52, a) ; when fully developed this is pushed out, so that the envelope, which at first was external becomes the portion of the body below the head (Fig. 53, 6). In the Cysticercus-form the embryo grows into a vesicle filled with fluid, from the walls of which the head is budded out. When the head is protracted, the vesicle forms a terminal appendage of the body (Fig. 54). When a number of buds are formed on the wall of the vesicle, in which protractile heads are dif- ferentiated, we have the Coenurus-form. When the buds break off into the interior of the vesicle, and there form new vesicles, on the walls of which the same budding process goes on, leading to the for- mation of systems of vesicles, placed one within the other, and when the youngest of these can again bud off tasnia-heads on its inner wall, we get the Echinococcus-form. Notwithstanding the various characters of their final products, these processes of gemmation may be derived from a common ground-form. They are by no means unparalleled among thePlatyhelminthes, for in not a few an asexual multiplication occurs, which is very similar to these in many points. It is very common among the Trematoda, where the embryo gives rise to an asexual stage known as the " sporocyst." The body-parenchyma of this sporocyst becomes differentiated, generally into similar structures, in which in their turn the larvas known as " Cercarias " are produced, and these are developed into the sexually mature form. The variation in the forms of the separate generations seems to be due, in a general k 2 Fig. 54. cercus Cysti- cellu- losce ; head pro- truded (nat.size). a The caudal ves- icle, filled with fluid, c The an- terior part of the body, d The head (after V. Siebold). 132 COMPAEATIVE ANATOMY. way, to the degenerations correlated with their parasitic habit, and in special points to their relations to their different hosts. Parasitism, in fact, determines the whole phasnomenon in question, which is spoken of as "alternation of generations," but by no means explained by that phrase. § 103. Gemmation is a common process among the Bryozoa also, where it leads to the formation of colonies. As in other Vermes and Ccelenterata, the buds are developed from the wall of the body. Accordingly as the bud remains at the side of, and on the same level with, its mother, or grows at one end and raises itself from the ground, flattened or upwardly-growing ramified cormi are formed. At the edge of the flattened colony the youngest buds often form the rudiments of several individuals (persons), which are by-and-by separated from one another. We observe in the case of develop- ment by gemmation as also in development from the ovum, that the anterior portion of the body, which carries the crown of tentacles, develops inside the hinder portion of the body, which forms the cell round the animal. The proposal has therefore, though without reason, been made to regard the two portions as separate " indi- viduals. " All the persons of a Bryozoan colony are not equally well developed. Iu many, the parts belonging to the cells and muscles only are developed; these give rise to the so-called Avicularia, which function as prehensile organs for the colony. The Vibra- cularia, which are long spike-shaped structures, continually moving, are parts further modified. Lastly, some persons may serve only for the reception of ova, and form the so-called marsupial cap- sules. Thus we meet here again with a polymorphism, which is due to a division of the physiological work of the colony. Appendages. § 104. The appendages have the form of actively mobile processes of the body, which may be used for the most varied purposes, accord- ing to their relation to it, and their more special line of development. As low down as the Turbellaria processes are found on the portion of the body which represents the head. In many Planarians lateral lobate processes are developed as tentacles, or feelers; in others the dorsal surface of the body is distinguished by similar processes (Thy sanozoon) . While the parasitic mode of life of the Trematoda, Cestoda, and many Nemathelminthes causes structures of this kind to disappear from them altogether, such structures are largely developed in the free-living Annulata, and prove to us the great influence of the outer APPENDAGES OF VERMES. 133 .,> •world on the organism. They are especially developed in the Chosto- poda, the cephalic region of which is provided with contractile pro- cesses, either at the sides or in the middle line (Fig. 55, t t'). These processes are simple, or are further differentiated by segmentation, or even distinguished by the possession of secondary processes. By adaptation to the most varied conditions of life they are converted into very various structures, and serve for all kinds of functions. In the tubicolous Chastopoda, where the cephalic region is that por- tion of the body which comes into closest relation with the surround- ing medium, the tentacles are converted into an important apparatus. They form tufts of contractile filaments on the cephalic lobes, where they are arranged in one or more rows (Terebella [cf . infra, Fig. 79, {], Hermella); or they may be con- verted into strong plume-like struc- tures (branchial tentacles) by the development of an internal support (cartilage), and. be beset with secon- dary branches ; these, in addition to their respira- tory function, may also aid in the movement of the whole apparatus when seizing food (Serpulaceas). In some, these bran- chial tentacles are arranged in two groups, and resemble an open fan. In Siphonos- toma they form short simple filaments, with two longer delicate feelers. In others the base of the two halves of the tuft, which are separated on the dorsal line, is drawn out into a spirally- coiled ridge, on which the separate filaments are arranged (Sabellidas). When optic organs are formed on the separate filaments of the branchial tufts, the tentacles acquire new and important relations (Branchiomma) . Some of the branchial filaments undergo other kinds of changes. In some Sabellida3 one or two of the primitively similar branchial tentacles (Protula) lose their respiratory function ; in others they are converted into club-shaped organs, one of which is largely developed, and serves as an operculum to close the tube in which the animal Fig. 55 Head of Nereis Dumerilii. aa Tentacles. t1 t" t"1 tIV tv Feelers, p Parapodia. ph Pharynx. 7)i Jaws, i (Esophagus, gl Glands (after Claparede). 134 COMPAEATIVE ANATOMY. lives. In Filigrana the stalk of the operculum retains some of its primitive characters by being feathered. But this feathered arrange- ment may be lost (Serpula), and then the operculum, during its development, passes through stages which are permanent in other forms. A calcified layer is often secreted in this apparatus, which owes its origin to adaptation ; it covers the free flattened end like a disc. In some cases the widened opercular stalk takes up the ova, and functions as a brooding pouch (Spirorbis spirillum). Thus we find one and the same organ passing through a series of the most varied relations, far removed from its original significance, and caused by certain external relations. In addition to the feelers there are special tentacles in the Chastopoda, which are shorter, but contractile (Fig. 55, a). The tentacles of the Bryozoa are structures of this kind ; they have the form of filamentous ciliated and contractile processes of a discoid or lobate extension of the oral end of the integument (lopho- phore). The discoid form of lophophore, in which the mouth is placed in the centre, is the most common. In the other case, the lophophore is drawn out into two processes, so as to have a horse- shoe shape (Fig. GO, B br). In Pedicellina and Loxosoma, the tentacles, which beset the edge of a discoid surface, which carries both mouth and anus, are simpler in character ; they are not hollow internally, like the tentacles of the other Bryozoa. § 105. Another group of appendages is represented by the locomotor processes developed in Chastopoda, which are lateral processes of the metameres of the body, the foot-stumps or parapodia (Figs. 55, 56, p). They are always arranged in pairs, of which there may be one or two on each segment. When there are two, one pair occupies the dorsal, and the other the ventral portion of the side of the body. They carry setas, and often also filamentous appendages (cirri), which vary greatly in form, and may be larger than the parapodia, or may even take the place of these appendages, when the latter are atrophied. The dorsal and ventral appendages of either side are sometimes closely approximated; there are all kinds of intermediate steps between this stage, and that in which they are completely fused (Syllidae). Such a fused appendage occupies the side of the body, and carries the secondary appendages (seta3 and cirri), which, in others, are distributed to the dorsal and ventral parapodia. The cirri appear to be atrophied in the Tubicola?, where they cannot have any physiological significance, owing to the body occupying a tube, which has sometimes the form of a shell. The parapodia are developed in very various degrees, and are com- plicated by their relation to groups of setas. A metamorphosis is effected by a widening of the ends of the separate, or of fused parapodia, or rather of their cirri, to form swimming-plates (Phyllo- docida)). The elytra are special appendages of the parapodia EXTEKKAX BEAXCHLE OF VEEMES. 135 formed by the metamorphosis of their dorsal cirri ; they are scale- like lamella^ which lie on one another on the dorsal surface, and alternate with short processes (Aphroditidae). Although the parapodia of the Annelida, which function as locomotor organs, appear to be the rudiments of the appendages, which are more per- fectly developed in the Arthropoda, they are not independent, for they have no special muscular apparatus of their own, like the appendages of the Arthropoda, and they are principally set in action by the general movement of the metameres to which they belong. External Branchiae. § 106. The appendages on the head, as well as those on the metameres, of the Chastopoda undergo various changes in adaptation to the respiratory function. Although in most Vermes this is per- formed by the general surface of the body, in the Chretopoda it is confined to definite parts, which are thereby converted into branchiae, as may be seen from their relation to the vascular system, and from other points in their structure. The cephalic tentacles are the first to enter into these respiratory relations (§ 104). In some (Pectinaria, Terebella) these structures contain a perienteric fluid, and are not distinctly branchiae. They are more definitely branchial in the Pheruseida3 (Siphonostoma). In the Sabellidas they are still further differen- tiated in the manner described above, and the separate gill-filaments are beset with secondary pinnules, by which their surface is further increased in size. Just as gills are formed by the special development of the cephalic tentacles, so, too, gills are formed from the appendages of the separate segments of the body, by the modifica- tions of the cirri which are attached to the parapodia, or by the formation of special appendages. When simplest the cirri are not altered in character, but enclose a continuation of the ccelom, so that the perienteric fluid can enter into them. The presence of cilia on the cirri is also of importance for their respiratory functions. The exchange of gases is promoted by the walls of the cirri becoming considerably thinner at certain points. As a rule it is the dorsal appendages which are developed in this way. The so-called elytra of the Aphroditidae also belong to this series of processes. They communicate freely with the ccelom. Cirri become more definitely related to the respiratory function by the continuation of the blood-vascular system into them. They then form branchiae. These either retain the condition of simple processes — sometimes they are lamellar in form — or are variously branched. In Cirratulus they are greatly elongated single filaments. The branched form includes the more delicate branchiae, which are either comb-like 136 COMPARATIVE AX ATOMY. (Eunicidse, Fig. 56, A B), or are arborescent (Fig. 82, br) (e. g. in the Amphinomeas). As a dorsal cirrus is frequently present in ad- dition to these branchia?, they appear to be independent organs; Fig. 56. Diagrams of vertical sections of Annulata, showing the appendages. A Sec tiou of Eunice. B of Myrianida. p Nemopodium. p ' Notopodium. br Branchia?. br1 Cirri. this is the more probable since they frequently are separated from the parapodia, and arise directly from the dorsal surface. They are distributed in varying number over the body. They are sometimes found on all the segments of it, but they are gene- rally less abundant at the tail (Eunice sanguinea, Amphinome). Sometimes they are limited to a number of segments and become gradually rudimentary (Arenicola, Hermella). In the tubicolous forms, the mode of life leads to the development of the anterior, and the disappearance of the posterior gills. On the three anterior segments of the Terebellida) there are branched branchial tufts (Fig. 70, br), in Pectinaria two comb-like branchia), and in Branchiosabella and Sabellides simple filiform appendages at the same point. In other divisions, also, of the Vermes the respiratory function is assigned to processes of the body. This is true of the tentacles of the Bryozoa. In the Gephyrea there are special developments as respiratory processes ; in Sternaspis the hinder end of the body carries vascular appendages. Finally, even in the Hirudinea there are lamellar extensions of the integument arranged metamerically (Branchellion). Integument. 107. The integument of the Vermes, which is separated off from the ectoderm, is closely united to the muscular system, by which it is continued into the parenchyma of the body when a coolom is wanting". This obtains in most of the Platyhchninths (Flat-worms) and Hiru- dinea. Where the coclom is present, the integument, with the muscles, forms a dermo-muscular tube, as in the Acanthocephali, Gephyrea, and most Annulata. If we separate the dermo-muscular tube into its two constituen t INTEGUMENT OE VEKMES. 137 parts, we find the muscular portion to be, as a rule, the larger ; and the layer which corresponds to the true integument to be propor- tionately feebly developed. The proper integument is formed, as a rule, of a layer of cells, the elements of which are often so slightly separated that they form a syncytium. This layer corresponds to an epidermis. In the Turbellaria it is everywhere provided with cilia. In many the cilia are placed on an apparently homogeneous layer, which resembles a cuticle. But the cilia must be regarded as processes of the cells. Even in those forms, such as the (Jestoda, in which there are no cilia in the adult, there is an investment of cilia during the embryonic stages. The embryos of the Trematoda also have it. In many Annelids there are ciliated spots at various parts of the body, or large tracts may be clothed with cilia. The part that this investment of cilia plays in locomotion is best seen in the smallest forms. In the young state it is generally the sole organ of locomotion. By the growth of processes of the body the cilia-bearing surface is increased, and so the cilia become of greater importance in locomotion. This is their character in the larvas of the Gephyrea and of most Annelids. The cilia are arranged on ridge-like processes, which surround certain tracts of the surface of the body, as lines or circlets of cilia; the arrangement of these is generally characteristic of the various divisions. One or more circlets of cilia surround the body ; and by these the larvse of the Chastopoda are divided into mesotrochal, telotrochal, and poly- trochal forms. Even if the surface of the body bears other cilia also, those of the circlets are more powerfully developed, and their action essentially aids in more rapid locomotion. Of these circlets of cilia (Fig. 57, G D v) one is more remarkable than the rest, it appears Fig. 57. Arrangement of the ciliated bands in the larva) of the Echinoderma, A B, and of the Vermes, 0 D. v Anterior ; w Posterior circlet of cilia. o Month, i Enteric canal, a Anus. in the very earliest stage, and divides the body into an anterior and a posterior portion. The former represents the upper part of the future head of the worm, while from the other portion the whole of the rest of the body is developed. The primitive circlet of cilia 138 COMPARATIVE ANATOMY. persists in one division of the Vermes., the Rotatoria. While the posterior portion is differentiated into a more or less jointed body, the anterior part, which carries long cilia on a discoid thickening, is developed into a special organ, which is characteristic of this division. This wheel-organ — so-called from the movement of its cilia — varies greatly in character. It may be either permanently simple, and retain its primitive state, or it may be broadened out into lobate processes (Tubicolaria), or form tentacular prolongations (Stephano- ceros), which frequently have a locomotor function in the larval stages only, and in the later fixed mode of life are used to bring food to the animal by means of the currents produced by the action of the cilia. In the Bryozoa, also, a circlet of cilia precedes the development of the tentacles, which are budded out internally to it. Owing to the position of the mouth, this circlet of tentacles does not resemble the more common form ; but it nevertheless has close relations with the arrangements in some divisions, e.g. the Gephyrea, the larvas of which have a circlet of cilia surrounding the oral region. Further, in Polygordius, which except in this point resembles the Nematodes, there is a circlet of cilia ; so that we recognise in this circlet of cilia an arrangement which may have been transmitted from an ancestral form common to many divisions of the Vermes. § 108. When cilia are absent the epidermic layer is covered by a cuticle, which varies greatly in character, and is a product of the secretion of the epidermic cells. This cuticle is a thin or even a soft layer in the Trematoda and Cestoda among the Pla- tyhelminthes. It has the same character in the An- nelida, but in them it may be very greatly developed (Fig. 58, c). It is also pre- sent in the Acanthocephali. When this layer is thickened pore-canals may be seen in it. In the class of the Nemat- helminthes it is very greatly developed, and is thicker than the subjacent matrix. Very often several layers, differing from one another, can bo made out in it ; they are formed of a substance which appears to be closely allied to chitin. When separate portions of the cuticular investment are very firm, a kind of dermal skeleton is formed in the Annulata, which is Fig. 58. Vertical section through the integu- ment of an Annelid (Sphferodorum). c Thick cuticular layer with wide pore-canals, m Mus- cular layer, m' Muscles for the tuft of setae, S, which occupies the ventral parapodinm, p, while the dorsal one, d, is represented by a swelling, which contains glandular tubes. INTEGUMENT OE VERMES. 139 morphologically the same as the chitinous carapace of the Arthro- poda, although it is not so hard. The dermal carapace of the Rotatoria resembles completely the chitinous skeleton of the Arthropoda. Although it may not become as strong, yet the rigidity of the most anterior segment, as well as of the succeeding ones, which are connected together by softer inter- mediate pieces, gives to it the character of a true skeleton, which serves for the origin of muscles. The cells of the Bryozoa are also cuticular structures ; they are sometimes gelatinous (Lophopus crystallinus), soft and flexible ; sometimes, owing to calcareous deposits, they are much harder. The latter kind are found in most of the Gymnok-emata. They differ from the tubes of many Rotatoria, and of tubicolous Annelids, by their close connection with the body ; whereas in Rotatoria and Annelids the tubes are formed by a secretion which is detached from the surface of the body. But the fact that in many Rotatoria the body-wall loses its connection with the hinder portion of the tube, shows that there is no well-defined boundary between these struc- tures. We see, in fact, that there are intermediate steps between typical cuticular structures and other secretions, which are ordinarily, though wrongly, grouped in contrast with them. The firm cell of the Bryozoon is not developed over the whole body. It only surrounds the hinder portion, and is continued into a more delicate chitinous layer, which invests the anterior tentacle- bearing portion, but is frequently wanting. This difference in the differentiation of the integument leads to a difference in the power of movement of these two regions of the body, and allows the anterior portion to be retracted, and to hide itself and its crown of tentacles in the hinder or cell-bearing portion. There are various differentia- tions of the cell, which, more or less, bring this relation of the parts to a state of perfection. § 109. Those special structures, the aciculi, setee, hooks, and so on, which often play an important part in the economy of the animal, must be regarded as differentiations of the integument, of the class of cuticular formations. The structures in question are most extra- ordinarily varied in character, and may be divided into two groups, according to their relations to the surface of the body. In the first group they have the character of simple elevations of the integument. A thicker cuticular layer is formed on papilliform processes, which may take the form of a wart, or when elongated, of a hair or seta. Even when it is very firm, it is still only apparently an indepen- dent structure, for it is nothing more than a modification of the cuticle, into which it passes at its base. The firm papillas and aciculi which are found in the skin of many Trematoda, and some- times variously extended over the anterior region of the body, are organs of this kind. So too are the fine and closely approximated aciculi, which cover the body of the Solenogastres as far as the i 140 COMPARATIVE ANATOMY. ventral groove ; the aciculi of the Echinorhynchi, and lastly, the hooks of the Cestoda, which are in many arranged in a crown (Figs. 59, 60), or placed in the wall of four protractile tubes (Tetrarhynchus). These be- gin as thickenings of the cuticle, but they form an in- termediate step towards the second group, for as they be- Fig. 59. Head of £ Jf come chitinised they sink Taenia coenur us i • n . • i (vesicle-form :Coduu- Fig. 60. alcdeBit. do.wn mto tlie matrix, and mis cerebralis seen ferent hooks from the still deeper. from in front). The crown of hooks of the In the Second group the four suckers and the same form. Repre- „_j.„, „„ „„•„ v i • crown of hooks in seuting stages of do- ^ 01 acipCuh n° loi}ger &T1^ the midst of them velopment (after V. Oil the Surface, but in special can be seen. Siebold). depressions, which may be very well compared with glands. The secretion is formed from one or more cells, and gets a definite form as it becomes gradually chitinised ; varying in different regions of the body. As a rule setee are first formed when metameres are. These structures vary greatly in size and form, and are very different in the various genera and species. With the exception of the Hirudinea they are found in all the Annulata. They are almost always arranged in tufts (cf. Fig. 58, s), two or four of which are connected with the parapodia of each metamere. They function partly as locomotor organs, working like oars in the swimming forms (Vagantes) ; when they are meta- morphosed into hooks they may serve as seizing or clenching organs (Tubicolae). They are best developed in the Aphroditidas, where some of the finer setas form a felted layer, which covers the back and elytra. The " rod-like bodies " in the integument of the Turbellaria are special structures, as are the similar structures in the Annelides ; in many cases they call to mind the "urticating capsules of the Acalepbte." § no. An organ, the function of which is still somewhat uncertain, belongs to the category of differentiations of the integument ; this is the so-called proboscis of the Nemertina. It forms a tube, which is enclosed, in a special sheath, placed above the enteron, and is often coiled ; this tube opens in the anterior part of the body above the mouth, whence it can be protruded by eversion. Several divisions can be made out in this tube, one of which has stylets at its base — generally a larger stylet in the middle, and at each side several smaller ones in special pouches, which are sometimes regarded as reserve stylets, and sometimes as structures of a supernumerary character. The portion of the tube behind the stylets is glandular in character, and is provided with an excretory duct, which is placed close to the stylet. A muscle which arises INTEGUMENT OF VERMES. HI from the body-wall is attached to the blind end of the tube; it is to be regarded as a retractor. In many Nemertina (Linous, Nemertes, etc.) the stylets are absent. In some the tube is small (Polia involuta), and so far resembles the structures in other Platyhelmmthes, which may perhaps be regarded as the first stage towards the highly differentiated pro- boscis of the Nemertina. Such are the stylets present at the anterior end of the body of the Cercarire, which serve for boring, and are placed either on the surface, or at the bottom of a deep folli- cular depression. Lateral stylets having the same relation to the median one as in the proboscis of the Nemertina are often observed ; so that we may conclude that primitively there was the same organisation in this respect in a large division of the Platyhelmmthes. Even in certain Nemathelminthes we find similar arrangements, so that we have here to do with a wide distribution of similar characters. In some this arrangement is only found in the young stages, and disappears in the adult organism (Trematoda) ; in others it not only persists, but is connected with important differentiations (Nemertina). § 111, The integument of the Vermes attains to a higher position through the differentiation of glands as special organs of secretion. Organs of this kind have been recognised in nearly all the divisions of the Vermes, and are very common among the Annulata. In most cases they appear to be unicellular, and sometimes lie immediately beneath the integument, and sometimes, when a distinct coelorn is wanting, in the deeper parts of the body. Among the Platyhelmmthes, unicellular dermal glands are known in the Trematoda. They are generally placed in groups in the anterior part of the body, and in the hinder part also, where they are connected with the suckers. The glands are greatly developed in the Hirudinea, and especially in the Blood-leeches, where they are scattered in the parenchyma of the body, and open on to the skin by long ducts. They appear to be developed in relation to the gene- rative function. Unicellular glands are also present in the integu- ment of the Scoleina, where they are placed between the cells of the matrix. In many cases the glands take up a deeper position, and their ducts, only, pass between the epidermic cells. In the Gephyrea tubular glands also are connected with the integument, aud tubes are also found in the Annelides (Fig. 58, d). A glandular layer is developed on one portion of the body of the Lumbricida), as a clitellus; but this organ does not appear to be so simple in structure, for the tubes are invested by a special epithelium, and are sometimes lobate in form. Glandular tubes, containing masses of rod-shaped bodies, are very common among the Chastopoda (Spio, Aricia), In the Nemertina there are also glands which secrete a viscous fluid. In many cases the secretion of the dermal glands is used to form an investment for the ova. 142 COMPARATIVE ANATOMY. Skeleton. § 112. When somewhat firmer than usual the integument in many divi- sions of the Vermes plays an important part as an organ of support ; these relations have been already referred to. Organs which possess this function, without any subsidiary relations, are more worthy of consideration. Supporting organs of this kind are seen in the cartilaginous pieces in the cephalic segment of a number of tubi- colous Annelides ; from these pieces processes pass out and ramify, as fine bands, in the feather-like plates. This is to be regarded as the formation of an internal skeleton, but it presents analogies only to other similar arrangements. This also holds for the branchial skeleton of the Enteropneusti, which is made up of a lattice-work of homogeneous rods (cuticular structures). Its arrangement and development call to mind the branchial skeleton of the lowest Vertebrata (Amphioxus), but it cannot be said to have any very close relations to it. Muscular System. § H3. The muscular system of the Vermes is connected with the integu- ment, and forms in most of them the largest part of the covering of the internal organs. In some it is only slightly developed. The general arrangement of the fibres follows one of several types, which may be thus characterised : 1) Circular, longitudinal, and radial fibres form a connected mass of muscle, in which the two former are separated into layers, and are traversed by the radial fibres. The circular fibres form an outer and an inner layer, between which the longitudinal fibres are placed. The radial fibres run from the interior of the body to the surface. At the lateral edges of the body they pass directly from the dorsal to the ventral surface. This arrangement of the muscles is found in the Platyhelminthes and Hirudinea. In addition to these muscles there are fibres which run obliquely ; but they are not present in the Nemathelminthes nor in the Turbellaria rhabdoccela. 2) The longitudinal fibrous layer is alone present. This is the case in the Nematodes, Chastognathi, and in Polygordius. The longitudinal muscles are distributed in various ways. The muscular fibres either pass directly below the epidermal layer (matrix of the cuticle) in the form of flat bauds, the broad sides of which are approximated, or they have their edges approximated, and their MUSCLES OF VERMES. 143 surfaces therefore directed inwards and outwards. In either case the muscles are grouped in particular ways. They are separated into two lateral masses by a dorsal and median line, formed by other tissues ; and these masses consist of fibres lying directly on one another (Gi-ordius, Trichocephalus). In the majority of Neinathelminthes there is a further differentiation, due to the inter- position of other organs at each side of the dermo-muscular tube. This lateral line (Fig. 61, A r) is, in very many Nematodes, Fig. 61. Transverse section of Ascaris lumbricoides, A, and of Hirudo, B c Cuticular layer, m Muscular layer, r Lateral line with the excretory organ. p p Upper and lower median line, p ' Oblique fibres, v Enteron. d Dorsal. I Lateral vascular trunk, s Vesicle of the excretory organ, n Ventral nerve-chord. enlarged into a lateral tract, which is more or less developed; it is present in the Chsetognathi also. 3) The muscular system of the body consists of a layer of external circular, and internal longitudinal fibres. Neither are separated into distinct tracts in the Gephyrea, and Acanthocephali, although in the former the separate longitudinal or transverse muscular bands are frequently placed at some distance from one another. On the other hand, the Annelides, owing to the arrange- ment of the longitudinal fibres into two dorsal and two ventral layers, have a distinct lateral field or groove ; the longitudinal layer is the thicker. A layer of transverse fibres, generally represented by distinct bundles, passes from the ventral median line to the lateral grooves. In addition to these muscles, which are present throughout the body, there are also separate muscles for special organs. We need only mention here the muscles which move the bundles of setee, and which are probably nothing more than fibres separated from the muscular mass, which extends over the whole body. The suckers found in the Trematoda, Cestoda, and Hirudinea, are special differentiations of the dermo-muscular tube, which agree with one another in all the essential points of their structure. U4 COMPARATIVE ANATOMY. § 114. The muscular system of the Bryozoa consists of an external layer of circular, and an internal layer of longitudinal fibres (Phylactolasma). The circular layer is frequently separated into distinct bands. The muscles which connect the protractile portion of the body with the cell are the best developed. When the walls of the cell are very strong the circular bands are separated (Flustra), and form bundles which pass from the side walls of the cell to its superior free edge. Some of these are inserted into the portion of the cell, which functions as an operculum. When longitudinal muscles are present, some of the muscular fibres are separated off behind the invaginated portion of the body, and pass inwards to the duplicature of the body- wall, whence most of them are produced on to the base of the tentacles. They form the retractors of the anterior part of the body (parieto-vaginal muscles). The Vermes differ considerably from one another in the structure of the form-elements of their muscular system. The muscular fibres are more or less elongated structures, which as a rule are the product of a single cell, even where they are very long, as may be inferred from the presence of a single nucleus. The lower forms of the Platyhelminthes have pale fibres often difficult to make out, which may be branched. In the higher Platyhelminthes they form tubes, the contractile substance forming a hollow cylinder, which contains indifferent protoplasm and the nucleus. The contractile portion of the fibres sometimes presents a fibrillar striation. This is seen in the Hirudinea, Acanthocephali, and Gephyrea. In the last two of these divisions the fibres of each layer form a network. Among the Nemathelminthes the simplest condition is seen in Gordius. The muscular fibres are broad thin bands, with their surfaces applied to one another. In others, special differentiations of the fibres may be seen forming rhomboidal plates, which are frequently continued into elongated fibres. The contractile substance is fibrillated and striated, and lies on the outer side of the fibres, while the portion of the fibre directed towards the coclom is formed of protoplasm, which remains indifferent, and encloses a nucleus. With this are allied the special metamorphoses of the fibres into canalicular, or flattened cylindrical forms. Each fibre has a very deep groove ; this it either retains for its whole length, or it becomes cylindrical towards its ends ; its open part being always directed towards the body-cavity. The walls consist of contractile substance, broken up iuto fibrillar. Protoplasm fills the small space of the groove, and a delicate membrane is produced from the edges into a pouch-shaped organ, which projects from each muscular fibre into the body- cavity, the greater part of which is filled up by these pouch-like appendages of the muscular fibres (Ascaris lum- bricoides, Fig. 61, A). From the pouches, oblique fibres run to the median lines ; they often have a fibrillar character, and have been KEKVOUS SYSTEM OF VEEMES. 145 regarded as nerves. In some parts they exist as distinctly muscular fibrillae. If the pouch is not developed, these fibres are attached to processes of the muscular fibres, which often become converted into tubes flattened at the sides. Both these conditions are found not only in the same genera, but even in the same individual, where they gradually pass into one another. In the last-mentioned form of the muscle-cells, a large number of fibres are generally placed closely side by side in the muscle-tube. The muscular fibres of the Chaetognathi are distinctly striated transversely ; there are traces of such striation in many other Vermes. Nervous System. § H5. The close relation between the nervous system and the general organisation is shown by the general arrangement of this system. The centres and peripheral parts are simple when the body is not divided into metameres ; while, when the body is segmented, meta- merism is exhibited with the greatest regularity in the central organs of the nervous system. In all worms the most important central organs of the nervous system are placed in the anterior part of the body, and generally near the commencement of the alimentary canal. A de- velopment of the nervous tissue from the ecto- derm has been made out in several divisions at least. The central organ above the fore- gut is the most primitive portion of the nervous system, whatever modifications it may present. When a head is separated off it lies in it, and always innervates the sensory organs that are developed in the head; it varies in the degree of its de- velopment with these organs. Nerve - trunks, radiating thence to the periphery of the body, appear in various degrees of elaboration, in proportion to the extent of the area of their distribution. Two different conditions may be developed from this arrangement. The first consists in the ventral connection of the superior central organs. This gives rise to an oesophageal nerve-ring. The second is distinguished by the development of two longitudinal trunks, which approach one another on the ventral surface, and have central elements placed in them. The primitive form of the nervous system js retained in most of the Platyhelminthes, since they possess two large ganglionic masses, connected by a transverse commissure in the anterior region of the body. These " cerebral ganglia " (Fig. 62, g), with the two L Fig. 62. Anterior portion of the body of Mesostomnm Ehrenbergii. (j Cerebral ganglia. n Lateral nerves. n ' Nerves to the ante- rior end of the body. dEnteron. o Mouth pur- rounded by a sucker (after L. Graff). 146 -COMPARATIVE ANATOMY. longitudinal nerve-trunks (n) which pass out from them, form the principal portion of the nervous system ; from it finer branches pass off to the sensory organs of the integument (it '), to the derino- muscular tube, and to the internal organs. The longitudinal trunks pass along the lateral edges of the body, and are placed closer together, or are more widely separated from one another, according to the breadth of the body. In the dendroccelous Turbellaria, as well as in many Trematoda, these lateral longitudinal trunks are only slightly developed, so that it is difficult to separate them from the other nerves, which arise from the cerebral ganglia, although they are not unfrequently distinguished from the other nerves by their larger size. The Rotatoria come nearest to the Platyhelminthes. The central organ is a ganglionic mass lying on, but never surrounding, the oesophagus. In some it is distinctly separated into two lateral halves. The peripheral nerves arise from this cerebrum ; and as they are not collected into longitudinal trunks, the simplest form, which is most like to that of the Turbellaria, obtains in this group. The nervous system of Pedicellina appears to be of this low grade, for it is placed on the stomach, and does not form an oesophageal ring. It is not quite certain whether or no the ganglionic masses, which lie on the oesophagus in Echinoderes, are separated from one another. If there is a dorsal commissure the arrangement would be similar to that in the lower Platyhelminthes. The nervous system of the Bryozoa is more highly developed ; its single central mass is a simple ganglionic swelling, lying between the mouth and anus, and sending out, in addition to large branches for the tentacles, two nerves which form a ring around the com- mencement of the oesophagus. Where the nervous system is most exactly known, as in Alcyonella, there is no doubt about the oesophageal ring. Prom the lateral part of the central nerve-mass a lobate process goes to the lophophore, and, like the rest of the oesophageal ring, gives off nerves to the tentacles. In addition to this nervous system in each individual, a colonial nervous system has been recognised in the stock, but it is not quite certain that this system does exist. § 116. The nervous system of the Nemathelminthes appears to be ar- ranged in a special manner, so far as the facts about it are agreed upon. It consists of a central organ placed on the oesophagus, and surrounding it as a ring, from which nerves radiate forwards, as well as backwards. This distribution of the nerves corresponds to the arrangement of the ganglionic cells of the oesophageal ring. The nerves which run forward from it may be separated into six fibrous bundles. Two run in the middle of the lateral tracts, and four in the direction of the secondary median lines. Ganglionic cells lie both on their origin and their course. The nerves which pass back- KEEVOUS SYSTEM! OF VERMES. 147 wards consist of a dorsal and a ventral trunk, which pass along* the corresponding median lines. In addition to these, two chords arise from the ventral portion of the oesophageal ring, and, converging posteriorly, unite to form a mass of ganglionic cells (G. cephali- cum). The median nerves run along the whole length of the body. Both send fibres into the matrix of the integument. It is clear that this arrangement is a modification, speaking generally, of the simple conditions of the nervous system of other Vermes; but it is so peculiar that any special comparison is altogether impossible. The same holds for the nervous system of the Acanthocephali. A small " ganglion " placed at the base of the sheath of the pro- boscis gives off branches an- teriorly as well as posteriorly. Its relation to the dorsal cen- tral organ of other Vermes is obscure, as it is placed between the bundles of the ventral re- tractors of the sheath of the proboscis. § H7. In the second form of the nervous system two longitu- dinal trunks are predominant ; these arise from the cerebral ganglia, and pass backwards. This arrangement is first seen in the Nemertina, and is di- rectly related to what obtains in the Turbellaria, in which there are often two greatly developed longitudinal nerves passing backwards. The size of these two peripheral longi- tudinal trunks is dependent on the length of the body. As there are ganglionic cells in them, they are not exclu- sively peripheral organs. The cerebrum, too, in the Nemer- tina is more largely developed, for several large segments can be made out in each of the two ganglia. The commis- sure between the two halves is traversed by the organ which we have already de- scribed above as the proboscis. Although in most of them the longitudinal trunks (Fig. Go, n) run exactly along the lateral edge of j. 2 Fig. 63. Head of a Nemertine (Omnia- toplea alba). g Central nervous system. n Lateral trunks, o Eye-spot, p #'#" Pro- boscis, ps Its sheath, i Enteron. e Lateral organ, d Dorsal vascular trunk. I Lateral vascular trunk (after Carm. M'Intosh). 148 COMPARATIVE ANATOMY. the body (embedded within the muscular layers), in others (CErstedia) they approach one another ventrally, and are distinguished by swell- ings at the points where nerve-branches are given off. This is an anticipation of the future development of ventral ganglia, the elements of which are already pi'esent in the longitudinal trunks. The ventral approximation of the longitudinal trunks shows us how the central system got its ventral position, which becomes further developed by the formation of ganglia. But the ventral approximation of the longitudinal trunks, by surrounding the oeso- phagus, leads to the formation of an oesophageal nerve-ring, when the ventral longitudinal trunks meet. It is an open question whether the oesophageal ring of the Bryozoa, and of the Nemathel- minthes arose in such a way as this, and we would only just remark that even if they had a similar origin it does not follow that there is any connection between them and the oesophageal nerve-ring of other worms. For in the Nemertina we see that the origin of this arrangement is due to the two peripheral longitudinal trunks, which are not present in the other case. In the Nemertina the oesophageal ring is not closed ; in the Annulata it is closed by transverse connections between the primi- tive longitudinal trunks. These have gained a central significa- tion, owing to the large number of ganglion cells in them ; the longitudinal trunks now appear to commence as commissures, which connect the primitive dorsal nerve-centre (cerebrum) above the oesophagus with the ventral one, which is formed from the longitudinal trunks. § H8. The cerebral ganglionic mass is not always present in the Gephyrea. It is well marked in Sipunculus and Sternaspis, and in the former it is divided into two parts. In Bonellia and Priapulus, however, only fibrous elements surround the oesophagus, so that in comparison with the other two a change has occurred, in which the central elements must be supposed to have become degenerated, or to have taken on a ventral position. The consequent arrangement corresponds to a great development of the commissure, which was formerly present between the two halves of the superior ganglia. Instead of the two ventral longitudinal trunks there is a single nerve-chord, in which the fusion of two trunks is only a matter of inference. This ventral chord generally lies within the coelom, but in some it is placed outside the muscular layer, just below the integument (Priapulus). As a rule, there are no collections of ganglion cells into special swellings, or expressions of metamerism ; it is only in Echiurus that they are present, and in it they are but feebly developed ; in other cases (Sipunculus, Sternaspis) there is a terminal thickening, of the chord which gives off fine filaments. The ventral chord gives off filaments on either side, which are frequently irregular in origin ; they are the peripheral nerves. NERVOUS SYSTEM OF VERMES. 149 Similar nerves are given off from the oesophageal ring to the alimentary canal. The concrescence of two separate structures in the ventral chord of the Gephyrea is the reverse of the permanent separation of the two halves of the ventral chord, which obtains in other divisions of the Annulata. It would not, however, be safe to regard these stages as lower ones, until observation shall have shown, which it has not done yet, whether or no they are preceded by an earlier stage, like that in the Gephyrea. The connection of two separate ventral chords by means of transverse commissures would be more easily explicable if the ventral chord were previously single. The nervous system of Sagitta has a peculiar character. Lateral commissures from the cerebral ganglion in the head, pass backwards and downwards to the ventral surface of the body, and pass into a large ventral ganglion, which lies just below the integument, aud gives off peripheral nerves to all sides. § 119. A higher grade of differentiation is seen in the nervous system of the Hirudinea and Annelides. The cerebral ganglia are connected by commissures with a ventral chord, and so far these groups re- semble the Gephyrea. In many Annelids the two halves of the ventral chord are homogeneous, and only indicate their metameric character by giving off nerves. In most, however, there are central- form-elements regularly distributed along it. The ventral chord then appears to be broken up into separate ganglia, which are connected with one another by longitudinal commissures. Each gan- glion, again, is broken up more or less regularly into two halves, which are connected together by transverse commissures. The two ventral chords then form a chain of ventral ganglia (Fig. 64.). In many Hirudinea the longitudinal chords of the ventral medulla are, during the early stages, sepa- rated from one another. Later on they are placed very close to one another, and almost seem to be a single chord. In this case, therefore, the separa- tion of the chords must be regarded as the more primitive condition. The longitudinal chords are still closer in the Scoleina, and in the Nereidas, Amphinomidas, and Euniceas among the Chasto- poda; but in all these cases there is not a real fusion, but only a close approximation, which appears to be still closer on account of the connective tissue investing the two nerve-chords. In the tubicolous Annelids the ganglia -bearing longitudinal trunks are separate ; in the Serpulida) especially, the lateral portions Fig. 64. Anterior portion of the ner- vous system of Capitella capi- lata. g Cerebral ganglion. o Optic nerves, c (Esopha- geal commissure. b Ventral chord, with two ganglia. n Nerves passing off from them (after Claparede). WO COMPARATIVE ANATOMY. of the ganglionic chain are widely separated from each other an- teriorly. In the Sabellidas and Hermellidre the chords are closer together, and indeed in the anterior portion of the ventral nerve- chord the transverse commissures are much shorter than in the posterior. Finally, the Terebellidas close the series, for in them the transverse commissures between the ganglia are distinct in the posterior portion only, while in the anterior portion the ganglia of either side are completely fused. With regard to the ganglia, the perfection and greater develop- ment of the cerebral ganglia in Hirudinea and Annelides must be mentioned as contrasting with the lower Vermes. The two halves are very seldom united into a single mass; in Enchytrasus, this condition appears to be due to degeneration. A breaking-up into distinct lobate segments, of which the Nemertina present a simple case, may give rise to a great diversity in form. The lobes frequently have the form of rounded projections, at times almost pedunculated. The cerebral ganglia are then complexes of smaller ganglia. In the ganglia of the ventral chord also, remarkable differen- tiations may arise, partly from an increase in size, and partly from concrescence. In the Hirudinea, the first ganglion is generally very large, and always larger than the rest; it corresponds to a large number of separate ganglia united with one another, as may be seen from the segments which compose it, as well as from the nerve- bi*anches which arise from it. A similar condition is found at the end of the ventral chord, where the larger ganglion present, and innervating the sucker, is produced by the concrescence of several primitive ganglia (seven in Clepsiue), which corresponds to as many metameres as form the sucker. This phenomenon of the approximation (by shortening of the longitudinal commissures) of separate ganglia obtains also in the Scoleina, but here the inde- pendence of the parts is often clearly recognisable, owing to the presence of separate transverse commissures. The Hermellidas among the Chretopoda are an example of this, for in them the first seven ganglia on each side are in direct contact with one another. The length of the commissures, as well as the number of ganglia, is directly connected with metamerism. In the Lumbricidaj with small rings they are very close together, so that the whole venti-al choi'd presents a compact series of .swellings and constrictions. The ganglia in Clymene and Oirratulus are still closer together. Owing to this close association of the ganglionic structures of the ventral chord, it has been supposed to be analogous to the spinal chord of the Vertebrata. And the ventral ganglionic chain has there- fore been called the " ventral medulla." Although we may allow the analogy, there is no reason at all for supposing that there is any homology. Position, development, and structure forbid such a supposition. As to structure, let it be here noted that the ganglion cells in the ventral chord are in the periphery of the ganglia, while the inner part is essentially occupied by fibrous bands. NERVOUS SYSTEM OF VERMES. 151 § 120. The cerebral ganglia give off, chiefly, the nerves of the higher senses, and are developed in different degrees, according to the perfection of these latter. The tentacular nerves, and those of the organs of sight, deserve to "be particularly noticed (Fig. 64, o). The nerves which rise from the ventral chain go off, as a rule, from the ganglionic swellings ; in many divisions, indeed, there is an apparent origin from the longitudinal commissures, but the nerve may then be always referred to the nearest anterior ganglion. This happens in the Scoleina, in the Siphonostoma3, in Aphro- dite, as well as in the Nereidas, and others. Very often the lateral branches of the ventral medulla form small ganglia, which are generally placed at the base of the parapodia, and from which finer nerve-twigs take their origin (e.g. in the Nereidse). These ganglia are often connected together by longitudinal com- missures, and thus there arises a separate part of the nervous system which is co-ordinated with the ventral nerve-chord (Pleione). The visceral nerves are differentiated in the same way. In the lower divisions of the Vermes nerves pass from the superior single ganglion to the alimentary canal. This has been observed in the Turbellaria as well as in the Trematoda. In the Annelides not only are these nerves more developed, but they become in- dependent to a certain degree, owing to the deposition of ganglia in them. We divide the apparatus, which has in this way become a special system of visceral nerves, into an anterior and a posterior portion. The former is distributed over the oral region, and is specially developed in the Chastopoda which are provided with a protractile proboscis (Phyllodoce, Glycera, etc.). The posterior portion, which is less developed, passes on to the enteric tube. In the Hirudinea there is an azygos enteric nerve ; in the Lumbri- cidas a nerve is continued from the oesophageal commissure, on either side, to the ganglia placed on the enteron ; these ganglia have been observed to vary in number. These two portions of the visceral nervous system, notwithstanding their distribution in parts which are physiologically connected, must be kept apart, for the anterior portion is distributed in parts, which are movable at will, while the latter alone correspond to a true enteric nervous system, and in view of their physiological relation can be called a sympathetic nervous system. Leydig, Ueber d. Nervensystem der Anneliden, Arch. f. Auat. Ph. 18G2. Hermann, E., Des Centralnervensystem von Hirudo raedicinalis. Munich, 1875. § 121. The nervous system in the Solenogastres differs in several points from the forms which have been already mentioned as ob- taining in the Vermes. The cerebral ganglion, which in Ohastoderma 152 COMPARATIVE ANATOMY. is composed of four lobes, gives off four nerve-trunks, which pass backwards. Two of them have a central, and two a lateral course. They unite in a ganglion near the end of the body. In Neomenia there is a considerable complication. The cerebrum gives off a commissure, which surrounds the oesophagus, and of these also a commissural chord on either side, each of which passes to a ganglion at the side of the oesophagus ; from each ganglia a lateral nerve-trunk is given off. The lateral nerves unite in a terminal ganglion (branchial ganglion). A commissure passes from each of the lateral ganglia to a ventral ganglion, which gives off a ventral nerve- trunk, which is connected with its fellow of the opposite side by a number of transverse commissures. If the lateral and inferior pair of ganglia be regarded as parts separated off from the cerebrum, this form would be seen to approximate very closely to Chastoderma, and the only difference would lie in the oesophageal commissure, in the transverse commissures of the ventral trunks, and in the exclusion of the latter from any share in the terminal ganglion. In any case we have in Neomenia a further development of the simple characters of Chsetoderma. This is not the place to indicate further points of comparison, for as yet we are only beginning to obtain any exact knowledge as to the structure of these animals. Sensory Organs. Tactile Organs. § 122. The sensory organs of the Vermes are of a high grade of differentiation. The organs of tactile sensation appear in the form of fine modifications in the structure of the integument with which the peripheral nervous system enters into connection. Of this kind are the true tactile organs, while the coarser arrangements, such as the processes of the integument, are only bearers of them. The essential part of these organs consists in the connection between the sensitive nerve-fibres and the modified cells of the integument ; these cells, as a rule, project beyond the surface of the integument, as stiff setiform processes (tactile setre, or rods). These arrange- ments are most exactly known in the Rotatoria and Annelida, but they have been recognised in other divisions. Tactile setae are widely distributed among the Turbellaria and Nemertina, where they are sometimes found over the whole body, and sometimes are richly developed on the head. They are found on the tentacles of the Bryozoa; and arc widely distributed on the cephalic segment in the Lumbricida), and in the Chastopoda. They appear in the Cha)topoda on the true tentacles and antenna?, as also VISUAL ORGANS OF VERMES. 153 on those appendages of the parapodia, which are known as cirri, as well as on the structures which are formed from modifications of these cirri (cf. § 106). The appendages just mentioned are pro- vided with a large number of end-organs of sensitive nerves, and thus become complicated tactile organs, which are of a somewhat high grade, on account of their power of movement. A special complication of the tactile rods obtains in some Hiru- dinea, where groups of these structures are embedded in the base of cup-shaped organs. There is a large number of such organs in the head, and they are scattered over the hinder rings of the bod}r. The arrangement of the sensory parts in depressions of the surface of the body justifies us in supposing that we have here to do, not with a special tactile organ, but with a sensory organ of general character. The tactile papilla) are less differentiated than the tactile rods or setas. They are developed in places where the body is covered by a stronger cuticular layer, and are conical or wart-shaped elevations of the cuticular layer, which are' traversed by a pore-canal. We find these tactile papilla) in the Nematodes, where they are grouped in a regular manner, some near the oral, and some round the genital orifice. § 123. Very little is certainly known as to their function, but organs which may be regarded as sensory are formed by parts of the body which either carry cilia, or have their epithelium distinguished by some other peculiarity; such are the cephalic pits of many Nemertina, and the similar parts in Polygordius. The clefts at the side of the head lead into a narrow ciliated canal, which is con- nected, either directly or by meaus of a fibrous chord, with the cerebral ganglion. Perhaps the apparatus presented by the pro- boscis of Balanoglossus may be regarded as an organ of this kind. It is uncertain whether these organs serve for the perception of the conditions of the surrounding medium, and possess a function analogous to that of olfactory organs. Visual Organs. § 124. The visual organs of the Vermes offer numerous examples of the gradual evolution of an organ from an indifferent condition. In many of the lower Vermes, Turbellaria, Trematoda, Nemertina, and Kotatoria we often find, at the place where other forms have dis- tinctly developed eyes, pigment -spots only, which are arranged symmetrically, and either placed directly on the brain, or close to it. Nothing is known as to the mode of termination of nerves in these organs, so that it is uncertain whether such "eye-spots" should be regarded as organs for the perception of light. 154 COMPARATIVE ANATOMY. We can bo more certain about function when tlio pigment-spot is merely a covering for the special end-organs of the sensitve nerve. These end-organs have the form of specially modified cells, which traverse the pigment either singly or in groups ; judging from what obtains in cases where the optic organs are more exactly known, we may confidently say that these structures are in direct connection with nerves. They are the so-called crystalline rods, or crys- talline cones. Eyes of this kind are common enough in the Turbellaria, among the Platyhelminthes (species of Mesostomum and Vortex) ; they are as a rule found in pairs on the upper surface of the head. Many marine Planaria have a larger number of regularly-ar- ranged, well-defined, pigment-spots on this pai-t, some of which surround a crystalline body. These eyes frequently appear in the early stages of the embryo as pigment - spots ; this is also their condition in many Trematode larvas, although, in many, distinct crystalline bodies may also be made out (Amphistoma subclavatum, Monostomum mutabile). In the entoparasitic forms of this division the visual organs disappear, while they are persistent in many of the ectoparasitic forms (Dactylogyrus). They are also persistent in Polystomum. They are absent in all stages of the Ccstoda, unless, indeed, we are to regard the red pigment-spots, which in some lie behind the suckers, as rudiments of such organs. In the Nomertina, where eye-spots are frequently present, true eyes have been observed in but few cases (Polia coronata, Nemertes antonina). Eye-spots and true eyes of simple form are found on their oesophageal ring, in the free living Nematodes (Enoplus), while they are absent in nearly all the parasitic forms ; here, therefore, degeneration of sensory organs goes hand in hand with parasitic habit. The visual organs in the Rotatoria are placed immediately on the cerebrum. There is one ciystalline rod to each of two connected pigment-spots; or there is only a single visual organ with one ' crystalline rod. Others have a pigment-spot and nothing else. The complex pair of eyes in Sagitta is distinguished by a large number of radially-arranged crystalline cones, and with these are found characters which remind us of what obtains in the Annulata. § 125. Among the Annulata the optic organs of the Hirudinea occupy the lowest position. The eyes, which are present in many, lie, as in the Platyhelminthes, on the surface of the cephalic portion of the body, and are, as in them, generally arranged symmetrically and in large numbers. In their structure they agree in so remarkable a manner with the cup-shaped structures mentioned in speaking of the tactile organs, that in them a condition appears to exist, in which a specific sensory organ is evolved from the indifferent organs of sensation, which are found in the integument. VISUAL ORGANS OF VERMES. 155 Fig. 65. Head and most anterior segments of a Myria nida. a Eyes, b Tentacles, c Unpaired cephalic tentacle, d Cirri. Among the Aimclidos we find the eyes of the Chastopoda gene- rally hidden beneath the integument, and placed on the cerebral ganglion in one or two pairs : a single eye is seldom present. Generally one pair is consider- ably developed, and the second often reduced to a pigment-spot. Where these visual organs are specially developed they stand out on the surface of the integument (Sylluke, Nereida), Fig. 65, a) ; and may attain to a highly compli- cated structure. This is the case in the Alcioptc, the pelagic mode of life in which is in con- nection with the high grade of development of this sensory organ. This influence of the mode of life is also seen in their nearest allies, the Phyllodoceida3, which live at the bottom of the sea, and have rudimentary or very simple eyes. The spherical bulb (Fig. (36) presents this, the highest degree of development in the Alciopidas only. The integument (c) covers the anterior, strongly-curved seg- ment, immediately behind which there is a spherical lens (/). The hinder segment, the innermost layer of which forms the layer of rods (b), surrounds a homogeneous vitreous body (li). A layer of pigment (p) separates the layer of rods from the parts of the retina which lie more to the exterior; outside all these is the expansion of fibres of the optic nerve (o'). While in the simpler forms of eye the ter- minal organs of the nerve lie in the integument, they are here pressed together into a concave layer. Influential in the development of this arrange- ment is the multiplication of the perceptive elements, and the formation of refracting media. Just as the eyes are completely wanting in the ma- jority of the Scoleina which live in the dark, so also these organs undergo degeneration in the Tubicola among the Chastopoda. The eyes which are present in the larvas, and even in later stages, disappear, or are represented by mere pigment-spots, when they enter upon the fixed mode of life. The development of visual organs on the branchial tufts of the head is an adaptation of another kind, which is seen in certain Fig. 66. Eye of an Alciopid (Neophauta celox) (after Greeff). i Integument, cover- ing the anterior segment of the bulb, c. I Lens, h Vitreous body, o Optic nerve, o' Expansion of the optic nerve, p Layer of pigment, b Layer of rods. 156 COMPARATIVE ANATOMY. Sabellidae (Branchiomma) ; in them the eyes are either placed in large numbers on the pinnate branches of the branchial filaments, or at their ends only. In other Annelides there is a similar change in position as compared with the primitive one. In many there are eyes at the posterior end of the body, as well as on the cephalic segmeiit ; and finally, in the genus Polyophthalmus there is a pair of eyes on each metamere, in addition to those on the head. We here find an arrangement which is not only of importance as bearing on the estimation of the metameres, but is also a proof that visual organs may be developed at points which in other forms only carry sensory organs of a lower kind. Auditory Organs. § 126. We consider as auditory organs in the Vermes organs which, as in the Ccelenterata, consist of a vesicular capsule, in which there is a firm, large concretion, or a number of smaller ones. The wall of the capsule is frequently invested with cilia, as may be seen from the trembling movements of the auditory stones (otoliths). The difficulty of making out the nerve-branches in the lower Vermes — in which, indeed, these organs are most largely distributed — has generally caused the connection of the auditory vesicles with the nervous system to be missed. These auditory vesicles are generally unpaired in the Turbellaria, in species of Monocelis, Convoluta, Proporus, Derostomum. They generally lie close to the cerebral ganglia, and are found as a rule in those genera which are devoid of eyes or eye-spots. In the Nenier- tina they have only been observed in some cases (CErstedia). In the rest of the Platyhelminthes these auditory vesicles are not, apparently, present, and they are also wanting in the Nematodes. Only in the Annelida do they appear again, where they are paired, and as a rule placed at the sides of the brain (Alciopidie Arenicola, Fabricia, Amphiglena, etc.). Alimentary Canal. § 127. The alimentary canal of the Vermes forms a tube, which is either embedded in the parenchyma of the body, or, when a ccclom is present, in it^ it has a general adaptation to the form of the body. The mouth lies, as a rule, at the anterior end of the body, and is always placed on the ventral surface. Where an anus is present, ALIMENTARY CANAL OF VERMES. 157 i a; !| ■ce* J* -&-* it is placed, as a rule, at the hinder part of the body, and is some- times ventral and sometimes dorsal. A differentiation of the enteric tube into several functionally different portions can always be made out ; accessory organs for the prehension of food are also often present at the entrance into the digestive cavity. The three portions, which are here present for the first time, are distinguished as fore-, mid-, and hind-gut; the last is absent when there is no anus. The primitive form of enteron agrees with the characters which are seen in the Gastrula- form (§ 28). It makes its appearance in all in the embryonic commencement of the organism as a cascal cavity, with but slight complica- tions, which opens on the surface at one point only; it persists iu this form in the lower Vermes. This opening serves for the inges- tion of food, and also for the ejection of its undigested remains ; it is mouth and anus at the same time. This arrangement is very common among the Platyhelminthes, being the only condition of the alimentary canal among the Trematoda, and the dominant one among the Turbellaria. In the rhabdoccelous Turbellaria the alimentary canal is distinctly marked in its anterior portion only, and has the form of a simple blind tube extending through the body. The simple mouth varies in position ; it may be in the anterior portion of the body, or towards the middle of the ventral surface, and lastly, even in the pos- terior portion; it leads into a muscular pharynx, which is seldom absent (Schizos- tomeas), and which is, in many cases, pro- tractile. This is the portion of the alimentary tract, which, under many modifications, can be most clearly traced in most divisions of the Vermes. § 128. ■rS IIP ffV- Fig. 67. Prorhynchus fluviatilis. o Mouth. oe (Esophagus, protrac- tile like a proboscis. i Enteron. gl Glands opening into the enteron. c Ciliated pits, x Spike in the organ above the oesophagus, which ends CEecally at y. ov Ovary, in which there are, in the anterior parts, ova at various stages of development. In the dendroccelous Turbellaria the gut is adapted to the broad form of the body. The mouth is (Fig. G8, o) placed ventrally, and often near the middle. The muscular pharynx (p) is often metamorphosed into a proboscidiform organ, which is capable of great enlargement in size, and is cylindrical or drawn out into lobes. It leads into an enteric cavity (v), which occupies the middle of the flat body, and which is broken up into numerous branches, which pass towards 158 COMPARATIVE ANATOMY. the edge of the body ; au elegant mesh-work may be formed by the connections of these with one another (Thysanozoon). Owing to the free communication that these branches have with the central cavity, the chyme is distributed in the body, and so the enteric canal takes on the function of a vascular system. The land Plana- rians are remarkable, inasmuch as their enteric tube consists anteriorly of a median canal, while it is divided posteriorly. Nu- merous and regular transverse processes pass off from both divisions of it. The enteric tube is branched in many of the Trematoda. The gut commences by a mouth, which is generally placed in the anterior region of the body, and, as a rule, has the surrounding parts meta- morphosed into a sucker (Fig. 69, s) ; this is followed by a muscular pharynx (b), from which the enteron proper is given off. This is, when most simple, a ceecal sac (Aspidogaster, Gasterostomum), and corresponds to a low grade of develop- ment; this is very common among the Trematoda at certain stages of their de- When more differentiated, the enteron divides into two branches, which pass backwards, and either give off greatly ramified branches into the body (Distoma hepaticum), or form simple cascal sacs (c) (Distoma flavescens, D. lanceolatum) . The two branches may unite again and form an arrangement like that which obtains in some Planarise. It is clear, from the homogeneity of its struc- ture, as well as from its contents, that even in the Trematoda this branching of the gut is merely an enlargement of the tract in the body, and not the formation of heteronomous segments. The texture of the wall is in correspondence with the low stage of this form of enteron, for only the epithelial investment is indepen- dent, being bounded, exteriorly, by the tissue of the parenchyma of the body — connective tissue. Complete degeneration of the gut is clearly due to adaptations to definite modes of life, in which the food passes through the integument by cndosmosis. This phenomenon, brought about by parasitism, Fig. 68. Digestive apparatus of Eurylepta sangui no- lent a. o Mouth. $> Pharynx. v Stomach. gv Ramifica- tions of the digestive cavity. n Nerve ganglion (brain) (after Qnatref ages) . velopment (Redia-f orin) . Fig. G9. Alimentary canal of Distoma flavescens. o Mouth surrounded by a sucker, s. s' Ventral sucker. b Muscular portion of the oeso- phagus or pharynx.- c Bifur- cated enteric tube. ALIMENTARY CANAL OF VERMES. 159 attains its "highest development in the sporocyst forms of the Trema- toda. Finally, the absence of an enteric canal is the rule among the Cestoda, where the enteron is not present for a time even. The enteron is altogether wanting in the Acanthocephali, and for the same reason — namely, parasitism. Among the Platyhelminthes there are forms which are distin- guished by the possession of an anus, and which may be contrasted with those which indicate their lower condition by not possessiug Such are the Microstomeas among the Tur- one. bellaria rhabdoccela, and the Neinertina, the enteric tube of which has pretty much the same form throughout, and which begins by an elongated ventral mouth, which lies behind the central nervous system. In Malacobdella the mouth is placed at the anterior end of the body. A muscular, but generally feebly-developed, pharynx, leads into the intestinal tube, which is provided with a large number of lateral diverticula. This fills the greater part of the body-cavity, to the walls of which it is attached by muscular fibres. The lateral diverticula of the enteric tube are sometimes regularly arranged, and are the first indications of metamerism. This is best seen in Pelagonemertes ; and so far this form calls to mind the dendroccelous Turbellaria. \ v § 129. In the Nemathelminthes all three portions of the alimentary tube are generally present. In corre- spondence with the form of the body, it forms a long- tube, which traverses the body, beginning by a mouth in the centre of the anterior end of the body, and ending by a ventrally -placed anus, which is more or less near the caudal end. The most anterior portion (oesophagus) forms a narrow canal, the walls of which pass gradually backwards into a thick-walled pharynx (Fig. 70). This is distinctly marked off from the rest, and is distinguished by a musculature, which enables it to act as a sucking- organ. The layer of chitin, which invests the tract from the mouth to this portion, not uufrequently forms ridges or tooth-like organs. The mid-gut (chyle-stomach), which succeeds the pharynx, is, as a rule, the largest part ; it has simple walls, often formed by a single layer of cells, which in some (Heterakis vesicularis, Oxyuris vermicularis) is provided in places with a muscular covering of annular fibre. A cuticular layer generally lies outside the epithelium ; an internal cuticle, which is traversed by pore-canals, appears to be present also. In many, the mid-gut forms a ca^cal diverticulum in its anterior portion. This portion Fig. 70. Alimen- tary canal of a Nematode (Dia- gram). 160 COMPAEATIVE ANATOMY. of the intestine is attached by laterally-directed fibrous chords to the body-wall, as a rule, along the lateral lines. The hind-gut, which arises from the mid-gut, is the shortest portion of the whole canal, and is distinguished from the part in front of it by its diminished breadth. In the Gordiacea the enteric canal is present in the entoparasitic larval stages only, and undergoes retrogressive metamorphosis when the sexual organs are developed. In Gordius even the mouth dis- appears. The organism, when it is free, uses up the material which it ingested by its enteron during the earlier stages, in forming genera- tive products, after it has given up its parasitic habit, and the ingestion of food. The enteric canal of the Cha> tognathi resembles in many points that of the Nemathelminthes, but the enteron is connected to the body-wall in a different way, namely along its dorsal and ventral median lines. Setiform hooks, arranged in rows at the sides of the mouth, serve as organs of prehension. § 130. Although the digestive organs of the Bryozoa are sharply marked off into the three primitive di- visions, they are exceedingly simple in character. The mouth, which is surrounded by the tentacles, or placed in the centre of the lobate process which carries them, is in one division (Phyiactolasmata) over- hung by a movable process — the epistom. Thence it passes straight back to an oesophageal portion (Fig. 71, ve), which in some is widened out or even converted into a gizzard by the development of denticular processes in one part of The second portion is separated from the fore-gut, which is invested with cilia, by a constriction (v), and forms the mid-gut. It functions as a stomach, and forms a cajcum, which generally descends some way down into the coclom. Fig. 71. Organisation of Bryozoa. .4 Paludicella Ehrenbergii. B Plu- matella fruticosa. br Tentacular branchiae. oe (Esophagus (fore-gut). v Stomach, r Hind-gut. a Anus, i Covering of the body (cell), x Posterior; x' Anterior chord, at the insertion of which into the body the generative products are developed. t Testes. o Ovary, m Retractor muscles of the anterior portion of the cell. N walls (e) ; this, after a simple course, passes to the body- wall (e1), where it finds its opening. Iu the Chastopoda simple forms of looped canals are most common, the separate canals in them forming sometimes coiled bodies, and sometimes presenting a very few coils. The funnel-like internal opening, which has been recognised in many, has in some (Alciopa) just the same relations to the septa of the coelom as it has in the Scoleina. In many of them the relation to the generative system can be similarly recognised. In addition to the more secon- dary relations which the looped canals of the Annelides have to the generative system, either at certain points only, or for a greater dis- tance, their relation to the excre- tion, as well as to the introduction or expulsion, of water must be borne in mind. That these organs have a close relation to the function of excretion is shown by the glandular investment of their walls, and the glands which open directly into them. They, in fact,, resemble the chief trunks of the excretory organs of the Trematoda. A relation between the perienteric fluid and the surrounding medium, either by the outflow of the former, or the entrance of the latter, is rendered possible by the internal opening of the looped canal. From the direction of the ciliary move- ment in the canals, and at their internal orifices, which is in nearly all cases towards the exterior, it is probable that solids also may be moved in this direction. Further investigation, however, is necessary in order to confirm this supposition. Fig. 8± brious A looped canal of Lum- uot too highly magnified. a Internal opening. bbb Clear portion of the canal, arranged in two double loops, cc Narrower portion with glandular walls, d Widened portion, which becomes narrower at d', and at d" is continued into the muscular portion, e. e' External opening. Generative Organs. § 146. We meet with a larger number of intermediate steps in the sexual differentiation of the Vermes than in that of any other division. The lowest stages are hermaphrodite, but this arrange- GENEEATIVE ORGANS OF VERMES. 170 ment is not uufrequently connected with great complications in comparison with which the arrangements seen in dioecious Vermes are very simple. The simplest state is seen in the Bryozoa, the generative pro- ducts of which are developed either on the inner face of the body- wall from simple aggregations of cells, which give rise to seminal elements or to ova ; or they arise on a chord which extends from the enteric canal to the inner wall of the body (funiculus) (Fig. 71, n). The mature generative products pass into the ccelom, and are thence passed out into the surrounding water by the orifice of communication mentioned above. The two sexes arc ordinarily united in the same individual, the male and female germinal glands being separate. In all phylactolcematous fresh-water Bryozoa, special bodies, (statoblasts) formed of an aggregation of cells, are developed in the body-wall, at the points where the ova are formed ; these break off, just like the ova, and form free-living buds. Various differentiations give rise to complicated shell-structures around them. § 147. Hermaphroditism obtains also in the Platyhelmmthes generally (Turbellaria, Trematoda, Cestoda). The two groups of sexual organs are, as a rule, united at a common orifice, being other- wise separated from one another, and embedded in the paren- chyma of the body. The secreting glands (testis and ovary) are generally small and simple in character. The excretory duct and the glandular organs connected with them, as well as the diverticula, or pouch-like appendages of the former, which act as places for the development of the fertilised ovum, or as receptacles for the semen, take by far the greatest share in the complication of the apparatus. As to the male organs, the testes vary in number, and are generally indistinctly marked-off spots for the formation of the semen, which reaches the common duct by narrow seminal ducts. A widened portion of the former has the function of a seminal vesicle, and its end is converted into a protractile organ, which serves as a penis. The ovary forms the most important part of the female organs. An organ, generally widely branched, is connected with its ducts — the yelk-gland : in the lobules of this gland cells are produced. The cells of this gland are used in building up the embryo, a number of them together with the egg-cell forming the egg. The origin of the yelk- gland is probably to be found in the division of labour of a primi- tively very large ovary, a portion only of which has continued as ovary, while the cells of the other parts have ceased to be ovarian germs, but becoming surrounded by the products of the fission of the egg-cells, are taken into the future body of the embryo. The oviducts and the ducts of the yelk-gland unite in a canal of varying length, which, according to the number of eggs to be developed, is n 2 180 COMPARATIVE ANATOMY. either extraordinarily long, or short and simple, or provided with diverticula. These cavities are known as uteri ; for in them the egg is not only enclosed in its shell, but as a rule passes through the early stages of embryonic development. A diverticulum of the female excretory duct, which has generally the form of a stalked vesicle, receives the sperm during copulation. In some cases there is a second diverticulum also ; it serves apparently for the reception of the male organ (Bursa copulatrix). The most important complications of this system are seen in the parasitic Platyhelminthes. The preservation of the species is here subject to innumerable difficulties, owing to the animal living in different hosts at different stages of development, and to the wanderings which this mode of life entails ; consequently a large number of ova have to be produced, and the certainty of fecundation insured. 148. The more special characters of this generative system exhibit extraordinary variation. The male portion has, in most Turbellaria rhabdoccela, the form of two elongated testicular tubes, from each of which a vas deferens is given off (Fig. 85, t). In the Trematoda, also, the testicles are, as a rule, but few, and rounded or lobate ; these are represented in the Tur- bellaria dendrococla, as well as in several rhabdoccela (Macrostoma), and Cestoda by a number, and often a very large number, of small follicles, scattered in the parenchyma of the body (Fig. 86, t) ; these are connected together by long efferent ducts. They may form a single row on either side (land Pla- narians). The excretory ducts either form a common vas deferens, or each passes sepa- rately to a terminal portion, which is con- tinued into the copulatory organ. The common excretory duct forms the seminal vesicle, or, as happens in a few cases, it is formed by enlargements of the separate vasa deferentia. The copulatory organ (Fig. 85, p ; Fig. 87, p1) is generally large and mus- cular, and the seminal vesicle often appears to be an appendage of it. It lies in a special cavity leading to the genital pore (penial Fig. 85. system viridis. Vasa deferentia minal vesicle, Generative of Vortex 1 1 Testes, vol vs Se- p Pro- tractile organ of copula- tion, oo Ovaries. gu Yolk-glands. rs Eecep- fcaculum seminis. v Va- gina, u Uterus (after M. Schultze). sheath in Planaria, bag of the Cestoda [Fig. 80, cZ] and Trematoda) ; glands are sometimes connected with it (Planaria). The copulatory organ is, as a rule, protractile, or can be everted, whereupon certain spines or hooks which lie on its inner surface, GENERATIVE ORGAN'S OF VERMES. 181 © c — «£. in its retracted state come to lie on its outer surface. Most Platy- helmiuthes, except the Planarias, have the penis thus armed ; it appears to be connected with a more intimate copu- lation. § 149. There are greater varia- tions in the female appa- ll a t u s . The ovaries are, as a rule, one or two elongated tubes of no great size (Fig. 85, o ; 87, ov), in which the ovarian germs are formed. Where a single oviduct is present it becomes con- nected with accessory parts as it passes to the genera- tive pore, and varies in length. Several such may unite together and form a common oviduct. In most Rhabdoccela, as in the Ces- toda (Fig. 87, od) and Tre- matoda, the duct is single, though the ovaries are double. It is shortest in the Rhabdocoela, where, as in most of the Cestoda, it has an enlarged portion, Which is clearly a receptaculum seminis. This organ appears as a unilateral diverticulum of the oviduct, and gradually becomes distinct. It is still more well-marked when it is attached to the base, or along the course of the oviduct (Fig. 85, r s), in the form of a stalked appendage. The Planarians have a double ovi- duct ; as a rule, a short portion only is common to both ducts, and functions as uterus or vagina. The oviducts are of some length in the land Planarians, the ovaries of which lie in the most anterior parts of the body. They may be provided with short lateral branches along their course, which open into lacunar spaces of the ccelom (Bipalium). This peculiar character raises the question as to whether these ciliated oviducts are parts of another system of organs ; for there is no reason for supposing that ovarian tubes have degenerated so as to form these backwardly-directed lateral branches. Their open mouths forbid us to suppose that there has been any such process. These mouths, indeed, point to an excre- tory organ having here entered into the service of the generative function. When yelk-glands are connected with the ovary they appear as two or more arborescent, ramified, or lobate organs (Fig. 73, gv), and are often widely spread out through the parenchyma of the Fig. 86. Male apparatus, with parts of the female, of Bothryocephalus latus (after Landois and Sommer). a Testicular follicles: part only are represented, ve Their excretory ducts, vd Yas deferens, c Cirrus, cl Bag of cirrus. Other letters as in Fig. 87. 182 COMPARATIVE AXATOMY. uterine organs can be made body (Fig. 87, d). Their excretory ducts come together from all sides and form with the oviduct a single common portion (d1). Special portions of the oviduct function as a Uterus, by which name parts, very different morphologically, are known. In general three different kinds of such out in connection with the oviduct. In the first the- oviduct itself is used for this purpose ; and then it is not only widened, but also greatly elongated, so that it has the form of a coiled tube, which traverses the body several times. This arrangement is found in the Trematoda, and also among the Cestoda (Trias- nophorus, Ligula, Bothryo- cephalus) (Fig. 87, u). A second form is represented by lateral diverticula, or pouch-like appendages on the course of the oviduct ; this is found in a few Khab- doccela and, in a more com- plicated form, in most Tape- Worms. In the Taeniadse a tube passes from the duct, near the opening of the yelk-gland, along the middle Pig. 87. Generative organs of Botliryo cephalus latus (after Landois and Sommer). Female portion of the system, v Vaginal canal. v' Its mouth, u Uterus (with ova), u' Its mouth. ov Ovary. od Oviduct. gl Shell- glands, d Yelk-glands (part only is shown). cl1 Duct of yelk-gland, e Vascular trunks. line of a sexually-mature proglottis, and forms, according to the quantity of ova in it, a number of arborescent branches. Finally, a third kind is formed by appendages, which are found only on the end of the oviduct, or rather in the vestibule common to the organs of both sexes, and close to the genital pore. This occurs in most Turbellaria (Fig. 85, u) ; in the Khabdoccela there are, as a rule, two such uterine pouches, which are considerably distended, and which may be branched, if they have to serve for the reception of a large number of ova. In the Dendroccela there is either only one such uterus, opening into the vestibule, which in them is greatly distended ; or it is altogether wanting, when the two oviducts take on its function (Leptoplana). The size and the number of the ova, which become mature and get their envelope at one time, is always in close connection with the condition of the organ which acts as uterus. A terminal portion of the oviduct is likewise frequently differ- entiated into a special canal, known as the " vagina ;" in some cases this is further provided with an appendage which has the function of a " bursa copulatrix. )> GENERATIVE ORGANS OF VERMES. 183 A large number of unicellular glands are attached to the point where the ducts of the yelk-gland and the oviduct unite, in Trema- toda (Distoma, Polystomum, Amphistoma) and Cestoda (Bothryo- cephalus, Tamia). This group of glands is known as the shell-gland, and its secretion serves to form the investment of the ova (Fig. 87, , DEKMAL SKELETON OF ECHIXODEKMA. 201 d d' c). The whole body is covered with cilia at a period before that when they are arranged on the ridge-like projections from the ciliated band ; bnt this general ciliation is only found during the most indifferent condition of the larva. The cilia are retained even later on, on many parts of the soft dermal layer which invests the calcareous skeleton : as, for example, in the ciliated tracts, which reach to the mouth in the Spatangidre (semitos). On other parts, such as the dermal branchias (cf. supra), ciliation appears to be correlated with the respiratory function of the integument, in which the ambulacral feet may also have a share. The extent to which the integument is calcified varies greatly. Sometimes the calcareous particles are united with one another into larger pieces, and form plates which are movably or immovably connected together : this arrangement either extends over the whole body, or is confined to definite tracts of its surface. In other cases the calcareous particles are scattered, and allow of great variations in the form of the body. In this case a large number of the characteristic features of the Echinoderm disappear from other parts of its organisation ; so that the disappearance of a calcified integu- ment is a departure from the type, and the general phenomenon of a scanty deposit of calcareous matters is not to be regarded as an early, but as a final stage in the series of forms. Calcification converts the integument into an organ of support for the body, or dermal skeleton ; in many cases this sends out pro- Jv cesses into the interior of the body. These give rise to calcified struc- tures, which form an internal skele- ton and combine with the external. The whole thickness of the peri- some is not affected by the process of calcification. A thin non-calci- fied layer of tissue is always found on the inner, as well as on the outer surface ; on the latter, however, this layer disappears at an early period from some parts, so that tlie calcified parts are exposed ; this happens, for instance, on the spine-like struc- tures, as well as on other processes of the calcareous skeleton. The lime-salts are always de- posited in a regular manner in the integumentary layer. They form delicate frameworks or retiform structures (Fig. 101), in the spaces of which soft organic substance persists. The most solid skeletal parts are thus traversed by soft structures ; when the calcareous Fig. 101. View of a calcareous net- work of a plate of the dermal skeleton of an Echinid (Cidaris). b Trabecular cut throiigh. These were directly per- pendicular to the horizontal network (somewhat highly magnified) . 202 COMPARATIVE ANATOMY. skeleton is merely represented by separate and microscopic deposits, these are generally definite in shape, and characteristic of genera and species. The calcareous skeleton of the larva forms an organ of support, which is generally made up by a framework of delicately attached, and often perforated rods. They are ordinarily found in the larval Echino'ida and Ophiurida; there are also calcareous bodies in the larvae of the Holothuroi'da. The presence of a calcareous skeleton in the larvas is clearly an instance of an arrangement which is common to the group ; but it must not be forgotten that this larval skeleton corresponds to the form of the larva, and not to that of the adult Echinoderm ; none of it passes permanently into the adult form. There is, in fact, a repeated change of the calcareous skeleton in the Holothuroi'da. § 164. As regards the special characters of the dermal skeleton, the presence of pieces, movably connected with one another, on the ambulacra! surface of the arms, is characteristic of the Aste- roi'da. Transversely - placed pairs of calcareous pieces, which gradually diminish in size, are found from the mouth as far as the tip of the arm (Fig. 100, A w) ; they form the floor of a groove — the tentacular groove. The sepa- rate pieces form a jointed series by their articular at- tachments, and the suckers pass out between the solid joints (p). These calcareous pieces are consequently known as ambulacral plates. But as special soft parts (ambu- lacral canal and nerves) are also embedded in this groove, the jointed segments do not appear to be purely dermo- skeletal parts. The ambu- lacral groove is covered by the integument, which is con- tinued laterally on to the am- bulacral plates. It consists largely of a layer of long cylindrical cells, covered by a cuticle. At the side it passes into a layer of cells, which is placed much deeper. At the lateral edges of the groove the skeleton is W\ Fig. 102. Body disc of an Ophiurid (Ophiothrix f r ag i 1 i s) ; seen from the oral surface ; the bases of the arms be- set with spicules may be seen (magnified). C Body disc. B Arms, t Calcareous plates, which coyer the canal which corresponds to the tentacular groove of the AsteroYda. 0 Genital clefts, d Masticatory plates. PEEMAL SKELETON OF ECHINODER.MA. 203 continuous with the dermal skeleton, which covers the back of the arms ; at this point there are frequently found one or more longi- tudinal rows of plates or scutes. These structures may be replaced by knobs, and are sometimes continued on to the integument of the antambulacral surface of the body ; or the integument is dis- tinguished by retiform deposits of calcareous matter and smaller tubercles, separated by the non-calcified parts of the perisome. Brisinga resembles the Astero'ida in the structure of its arms ; that is, they have an ambulacral groove. Larger flat plates, marginal plates, form the edge of the arms ; these are often distinguished by spicules and other processes. The structure of the integument of the Ophiurida resembles that of the Astero'ida. There is seldom any great development of calcareous plates on the antambulacral surface ; they are, as a rule, found near the base of the arm only in these forms. The ambulacral or ventral integument is also provided with plates around the mouth (Fig. 102). But several parts of the firm skeleton of the arms are different to those of the same parts in the Astero'ida. The pieces homologous with the ambulacral plates of the latter form a closely- set series (vertebral pieces, Fig. 100, B w), which almost com- pletely fill up the arm, and only leave a narrow canal on the dorsal, and a groove for the nerves and other organs on the ventral surface. The ccelom is therefore continued into the arms as a narrow canal only. Instead of the soft covering of the ambulacral groove, which the Astero'ida possess, there is a series of firm calcareous scutes (Fig. 100, B b) in the Ophiurida; other lateral processes of various kinds are added on to them. In the Euryalida also the leathery investment of the body covers in a skeletal structure, formed of vertebral calcareous plates attached to one another. This, as in the Ophiurida and Astero'ida, belongs to the oral surface of the body ; the plates are continued from the edges along the radii, and to the finest ramifications of them. In them, too, this skeleton forms the floor of the ambulacral groove. On the aboral surface the body-disc is enclosed in a dermis, merely impregnated with calcareous granules, which passes on to the arms, and covers them as far as the edge of the ventral groove. There is a larger number of knob-like and spicular processes in the integument, which may vary very greatly in character. A special form is very common among the Astero'ida ; namely, bundles of movable spicules attached to a common stalk (paxillas). The pedicellariae are described in § 166. § 165. This dermal skeleton is modified in the Crinoida. The dorsal integument is drawn out into a stalk, to the end of which the animal is attached. The skeleton of the stalk is formed of cal- careous plates, which lie regularly one over the other, and are con- nected with flattened basal pieces, to which other calcareous plates, 20i COMPAKATIVE ANATOMY. which form the boundaries of the body, are attached. In the young stages of the Comatulaa, a simple knob-like piece (centro-dorsal) unites the skeleton of the stalk with the body. Radial jointed pieces, which are continued into the joints of the arms, are attached to the central piece. The ambulacral groove extends along the dichotomous branches of the arms (Pentacrinus), as well as along the lateral appendages (pinnulae of Comatula) which are set alter- nately on either side of the arms. The groove becomes united with the groove of the next arm, and passes along the ventral surface of the cup-shaped body as far as the mouth. Deposits of calcareous plates are embedded in all parts of the portion of the integument which remains soft and covers over the skeleton. § 166. The differences in the dermal skeleton of the Echinoida, and the consequent changes in the form of their body, as compared with the Asteroida, are chiefly due to the calcification of the oral (ventral) perisome, that is of the portion which covers the ambulacral groove and the soft parts which lie in it, and which is permanently soft in the Asteroida. In the place of the articulated joints, there are plates which are calcified externally, and are connected with the body in various ways. In the Desmosticha the portion which corresponds to the dorsal, or aboral pole of the Starfish is a small surface, marked off by small loosely-articulated calcareous plates, placed excentrically to the anus (Fig. 103, x). This surface, which occupies the centre of the so-called apical pole of the Sea-Urchin, is surrounded by larger calcareous plates, which carry the orifices of the genital organs, the genital plates (, already described when we were treating of the Astero'i'da. The peripheral nerve-trunks pass out by openings in the five larger pieces of the calcareous ring, and extend, becoming broader as they go along the outer side of the bands of longitudinal muscles to the hinder end of the body ; near the cloaca they again diminish in breadth ; they give off fine branches along their course. Each radial nerve-trunk may be divided into two layers, which are separated from one another by a layer of connective tissue. A vessel accompanies the radial nerves ; this is separated by a wall of partition from the ambulacral vessels, which lie still more to the interior. The oral ring gives off tentacular nerves, in addition to these radial trunks. Sensory Organs. § 170. Definite portions of the integument have, in this group also, a special significance as tactile parts. The tentacles, as well as the sucker connected with the water-vascular system, may be reckoned as tactile organs, and the former become greatly developed, and so of greater importance, when the ambulacral system is reduced, as it is in the Holothuroi'da (Apoclia). Five pairs of vesicles, which lie on the roots of the radial trunks in the Synaptidaa, are said to be auditory organs, but their sensory function is as doubtful as is that of the so-called eye specks in the same genus. Visual organs are exactly known in the Asterida only ; in all other Echinoderma mere collections of pigment are regarded as eyes or " eye-spots." The eyes of the sea-stars are placed at the tip of each arm, which is ordinarily bent up, and so turned towards the light ; they occupy a pad-like elevation of the end of the ambulacral groove, the epithelial layer of which is formed of long cylindrical cells, and is very thick at this point. The rod-shaped cells contain pigment. The eyes lie on separate points of the " optic-pad." A funnel-like cavity covered by the cuticle has its walls bounded by rod-like cells, which are inclined from the periphery to the funnel ; in this way their ends form the wall of the funnel. A transparent body projects from the pigmented part of the cells into the cavity of the funnel, and so fills up the greater part of its lumen. ALIMENTAEY CANAL OF ECHINODERMA. 211 As this apparatus lies on the terminal ganglionic swelling of the radial nerves, and the cells give off fine processes to this ganglion, the two parts may be regarded as connected at this point (Astera- canthion rubens). Each eye which consists of a complex of cells is a differentiation of the epithelial layer, and resembles therefore the optic organs of other Invertebrata. Alimentary Canal. § 171. The alimentary canal, which varies greatly in character in the adult Echinoderma, has a simpler predecessor in the primitive enteron of the larval form, which is similar in all Echinoderma. Of course this does not refer to those forms in which there is no larval stage, and where the development is compressed. The first rudiment of the enteron is formed by the in-growth of the cell-layer, which invests the body of the young larva. This gives rise to a cascal tube, which is pushed down into the body, and the wall of this tube forms the endoderm, while the outer cell-layer represents the ectoderm. The organism is in fact a Gastrula. The entrance into the rudimentary enteron is regarded as the primitive mouth. A second invagination soon grows from the other side of the body towards the blind end of the enteron ; this unites with the enteron, becomes hollow, and so forms a continuous tube with the part first formed. The parts formed last are the mouth and the oesophagus, which is connected with it, and the part formed first is the mid- and the hind-gut. The orifice which becomes later the anus, and the portion of the enteron connected with it, are consequently the parts first formed. The larval intestinal canal is formed of three portions (cf. Fig. 94, A B). A wide oral opening leads into a contractile tube lying in the long axis of the body ; this is the pharynx or oesophagus. Then follows a wider part, the mid-gut or stomach, which is con- tinued into a narrow and retort-shaped tube, which is the hind-gut and leads to the anus. These three portions correspond exactly to the primitive divisions of the canal, which are distinguishable in nearly all Vermes. The mouth and anus are at first on different surfaces of the body of the larva. As the body is differentiated, especially by the development of the ciliated band, they apparently come to lie on one and the same surface, the so-called anterior side. It is, however, quite clear that the ciliated band distinctly divides two surfaces of the body; a decreased oral, and an increased anal surface turned towards the former. But before the enteron is fully developed by becoming connected with the fore-gut, a portion of it, which forms a closed vesicle, is constricted off. Two pieces are then separated off from this vesicle, r 2 212 COMPARATIVE ANATOMY. or two new vesicles are formed from the sides of the enteric caecal tube. In this way three different bodies are differentiated from the enteron. The two-paired vesicles, which lie at the sides of the enteron, represent the commencement of the ccelom; the third vesicle becomes connected with the dorsal ectoderm and opens on to it ; this is the commencement of the water- vascular system. This apparatus, like the lining cell-layer of the ccelom, takes its origin therefore from the enteron, and from that portion of it, which is without doubt its hinder part, although it appears first of all and grows inwards, from what is later on the anus. This arrangement appears to indicate that there are arrangements in the water-vascular system, as well as in the ccelom (for the two are connected together), which are phylogenetically connected with the terminal division of the enteron ; in this case this tract of the enteron is not homologous with the enteron of a Gastrula, but corresponds a priori to a hind-gut, the early development of which is clearly due to the complicated character of the organs about to be differentiated from it. These orgaus are those which are especially necessary to the organism. I consider therefore that the first formed rudiment of the enteron is not a Gastrasa-enteron, and that its orifice is not the primitive mouth, but that they are respectively the true hind-gut and anus. The median division of the enteron which is divided from the hind-gut is morphologically a part of it. The differentiation of the above- mentioned organs out of the hind-gut points to stages in which organs were connected with the hind-gut in much the same way as they are in many Gephyrea. But as yet it is impossible to prove directly that such structures have been passed on to the Echino- derma ; and it is better to regard these remarkable processes as presenting us with a problem which has still to be solved. When the body of the Echinoderm is formed in and partly from the larva, the enteron of the larva does not completely pass into it. The perisome, when formed, first grows round its middle part, aud in the sea-stars takes up this only with the hind-gut. In the Echinoida the anus also appears to be formed anew. The larval enteron is retained most completely in the mature stage of the Holothuroi'da. The fully-developed enteron is found to hang, in the mature Echinoderm, in a ccelom, which is often wide, and undergoes various changes during its differentiation, which are generally corre- lated with the characters of the perisome. As a rule the mouth always retains its position in the middle of the ventral surface of the body. § 172. The mouth in the sea-stars has a radiate form, owing to the projection into it of interradial processes ; hard papillas and spicules are formed by the perisome, and function as masticatory organs. They are specially developed in the Ophiurida, where they generally ALIMENTARY CANAL OF ECHLNODERMA. 213 form several rows one above the other (Fig-. 102, d). In these therefore the dermal skeleton forms the organs for the comminution of the food. A short, wide oesophagus follows the mouth ; this is continued into a wide mid-gut (stomach), which occupies the middle of the body. In the Ophiurida and many Asterida (Astropecten, Fig. 107. Transverse section through the arm and disc of Solaster en dec a. The radial and the interradial portions are figured on opposite sides. o Mouth. v Stomachal cavity, c Radial ceeca. g Genital gland, m Madreporic plate, s Stone- canal with its so-called heart, p Ambulacra! feet (after G. 0. Sars). Luidia) the stomach is always a blind sac, as it is also in Brisiuga. But in all Astero'i'da it is provided with diverticula or cascal saccular appendages, which are indicated in the Ophiurida by radial constrictions. The gastric casca of the Asterida extend in pairs into the arms ; they spring from the stomach, and have the form of thin-walled tubes, closely beset with lateral appendages (Figs. 107, c; 108, h), which as a rule are united by pairs into one canal before they open into the stomach. This tract represents an unpaired por- tion of the enteron belonging to each antimere (arm) of the Astero'i'da, while the cascal tubes form a paired por- tion. In Astropecten auran- tiacus these tubes arise sepa- rately from the stomach. The unpaired portion in each arm has therefore disappeared in this form, and with it the primitive condition. In most of the Asterida the short hind-gut is continued from the stomach to the anus, which is placed on the dorsal surface. The enteric tube of the Crino'i'da (Comatula) is modified; it describes a spiral coil, and its narrower short terminal portion passes into a tubular and projecting anus, which is placed interradially near the mouth. This coiled arrangement, which is apparently very anomalous Fig. 108. Asteriscus verrnculatus: opened on the dorsal surface. a Anus. i Rosette-shaped enlarged enteron (stomach). h Tubular radial appendages of the enteron. g Genital glands. 214 COMPARATIVE ANATOMY. may be also seen in young Asteroida ; in them it is only seen for a time during their development, but in the Crinoida it is continued as a permanent condition. The enteron is attached to the body-wall by radial fibres. The radial caeca of the Asteroida are attached to their body-wall by a special peritoneal fold, which extends along each caecum. § 173. In the Echino'i'da the mouth is similarly provided with Mastica- tory Organs, but they are removed from the outer surface and placed in the ccelom. They there form an apparatus, which, in the Clypea- strida, consists of five pairs of triangular calcareous pieces, but in the Echinothurida, Cidarida, and Echinida is much more complicated. Five pieces directed towards one another carry a tooth-like point, and are united with several others into a complex organ known as the " Lantern of Aristotle ; " the oesophagus traverses it. The enteric tube always describes several coils. The narrow fore-gut passes into a wider portion, which forms the longest part of the canal. It has sometimes faintly-indicated diverticula (Echinida), sometimes veritable caeca (Clypeastrida), which (as in Lagan um) project into the cavity of body which is marked off by the supporting pillars of the calcareous shell. In these forms " mesenteric fibres " extend to the body-wall for the whole length of the coiled intestine. In the Holothuroi'da the enteric tube, which is longer than the body, forms a double loop, while in the Synaptas (with the exception of the Chirodotae) it extends straight through the body-cavity, and is provided with numerous diverticula. A muscular portion of the enteron which succeeds the oesophagus is to be regarded as a special differentiation; it appeal's to function as a muscular stomach (Synaptse). This character is also seen in the Asteroida, where the oesophagus has in the same way a stronger muscular wall than the rest of the intestine. The portion of the intestine behind the muscular part in the Holothuroi'da may thus correspond to the stomach of the Asteroida. The end of the canal is widened out in the Holothuroi'da ; but this only corresponds to the hind-gut of the Asteroida, although it is called a cloaca; it has two or more arbores- cent organs opening into it. A sieve-like fenestrated lamella fastens the canal to the body- wall. This mesentery is simpler in the Synaptao which have a straight canal, while in Chirodota it is separated into three parts, in correspondence with the extent of the enteric loop ; each part is connected with an interradial portion of the body-wall. ALIMENTARY CANAL OF ECHINODERMA. 215 Appendages of the Alimentary Canal. 174. *n? . The above-mentioned radial casca of the Asteroi'da might be regarded as organs differentiated from the primitive enteron, were it not that they must be regarded differently from a phylogenetic point of view. I consider that only certain other interradial casca are to be regarded as appen- dages of this kind ; they present very various de- grees of development. In the aproctous Asterida they are absent, or are re- duced to two (Astropec- ten), while in the others they are often very greatly developed. Archaster has five such csecal sacs, divided at their ends, and in Oul- cita this division is carried still further, so that each branch forms a racemose tube. In this way these appendages acquire the form of glands, and ex- hibit relationship with a structure which is very common in the Holothu- roi'da. The structure in ques- tion is connected with the terminal portion of the alimentary canal, known as the " cloaca," and as a rule consists of two chief trunks, with short branches, which extend forwards throughout the whole length of the body-cavity (Fig. 109, ?•), and are provided with a large number of ramified caacal tubes. Although the function of these organs, which were formerly known as " lungs,"* and considered to be internal respiratory organs, is different from that of the interradial caecal tubes of the enteron of the Asterida, yet they are exactly the same morphologically, and are a further development of the simpler tubes of the Asterida. Fig. 109. Enteric canal and tree-like organs of aHolothurian. 0 Mouth, i Enteric tube. d Cloaca. a Anus. c Branched stone-canal. £> Polian vesicle, rr Tree-like organs, r' Connec- tion between them, at the opening into the cloaca. m Longitudinal muscular layer of the body. 216 COMPAEATIVE ANATOMY. The fuuction of these organs is not at all certain. The view- that they are respiratory organs is opposed to the fact that only one of them has any connection with the network of blood-vessels, while the other is merely attached to the body-wall, and projects into the body-cavity. However, the fact that water is taken up by these organs, and is again expelled, chiefly by the aid of the strong muscular wall of the hind-gut, is of importance. In some Apodia (Molpadia borealis) they are only provided here and there with branched caaca, while in others the number of cseca is increased. Thus in M. chilensis not only is one of the trees divided, but the rectum also bears a number of smaller trees. The organ is divided five times in some Lisarmatidge. They are simpler in character in Echinocucumis (E. typicus), where they form long fine tubes, provided with one short branch only. The tree-like organs of the Holothuria? are absent in the Synaptre, but there is an arrangement, which as yet is only very incompletely understood; this consists of canals, which are placed along the insertion of the mesentery and open into the ccelom by funnel-like ciliated orifices (Chirodota pellucida). Glandular organs are also present on the rectum of many Holothuria? in addition to the tree-like organs. These — the Cuvierian organs — either form unbranched cascal tubes, which are inserted singly or in thick tufts (Bohadschia, etc.), or they form racemose organs (Molpadia), or, finally, filamentous canals, beset with lobate tufts of glands, and arranged in a whorl (Pentacta and Muelleria). They secrete a substance which forms fine sticky filaments, which may serve as organs of defence. Ccelom. § 175. The development of the ccelom from a vesicular structure cut off from the earliest rudiments of the enteron (§ 171) gives to this cavity a different signification to that which it has in other divisions, where the ccelom is not formed from any part of the rudiments of the enteron. The importance of this point must not be overlooked. But it may well be supposed that the water- vascular system, which is developed in just the same way, formed an apparatus which primitively formed part of the ccelom, and was connected with the hind-gut. The two coolomatic tubes nipped off from the enteron gradually increase in size, and by becoming attached in part to the enteron, and in part to the body-wall, form the more or less spacious cavity of the ccelom. The mesenteric filaments or bands which pass from the perisome to the enteron are to be regarded as the remains of the walls of these primitive structures. As the radiate Echinoderm body becomes developed, the ccelom VASCULAE SYSTEM OF ECHINTODEEMA. 217 passes into the rays. Thus in the Asterida and in Brisinga it extends through the arms. The same thing happens in the Crinoi'da, but in them the canals are narrower. In each arm it can be divided into three parts, which are connected with special divisions of the ccelom of the disc. This latter portion is separated into several divisions by connective bands, which here and there form membranous tracts; they communicate with one another at certain points, and at other points pass into the canals. In the Echinoida and Holothuroi'da, where the organism is more concen- trated, the ccelom is more simple. In the former, however, the mesenteric filaments, and still more the calcified pillars and columns, which traverse the ccelom of the Olypeasfcridae, remind us of divisions into separate parts ; several such spaces are also marked off in the ccelom of the Holothuroi'da. In the Asterida and Echinida, as well as in the Holothuroi'da, the parietal and visceral tracts of the ccelom have been observed to be provided with cilia. The contents of the ccelom appear to be the same in character as the blood, so that in it we have to recognise a portion of the blood-cavity. In some cases communications with the exterior have been definitely observed (Crinoi'da) ; as also communications with the water- vascular system (Crinoi'da, Holothuroi'da). The first set are due to numerous canaliculi, which traverse the interradii of the perisome, and open by the so-called calycine pores. Vascular System. Blood-Vessels. § 176. The nutrient fluid in the Echinoderma is a clear, or slightly opalescent fluid, which is seldom thick or even coloured, and which is very probably mixed with the water which is taken in from the exterior. The form-elements in this fluid are simple cells. The blood-cavity is formed in the first place by a special system of canals, but also by the ccelom, which is probably connected with a third cavitary system, the system of so-called water-vessels. Owing to the uncertainty of our knowledge of this Vascular System, that is, of its mutual relations and connections, it is as yet impossible to make any generalisation which will hold for all the divisions; although indeed remarkable progress has lately been made in our knowledge of this part of the anatomy of the Echinoderma. But from the similarity of construction of these canals and spaces we may suppose that a connection between them does really exist. The close association of the haemal system and the nerve-tracts may, however, be regarded as a general arrangement. A blood- vascular trunk accompanies each radial nerve-trunk, and is continued 218 COMPARATIVE ANATOMY. into a circular canal, which surrounds the mouth. The radial vascular trunk corresponds to the ventral vessel of the Vermes, which has a similar relation to the ventral medulla. A tube which passes from the oral ring- to the stone-canal (for which see below) was formerly regarded as the heart, but this organ cannot be regarded as such. The same remark applies to the similar structure iu the Echinoi'da. We have therefore still to search for a heart as the central organ of the blood- vascular system. The enteric vessels form a second division of the blood-vascular system. In the Echinoi'da the nerves lie within the radial blood-vascular trunks ; in the Crinoi'da and Holothuroi'da they lie on their outer side, and the same is the case in the Asterida and Ophiurida. The circular vessel surrounding the mouth in the Asterida and Crinoi'da and in the Spatangidse among the Echinoi'da, where it has the form of a wide sinus, is described as having the same relations to the nerve- tract ; although in Echinus there is said to be a blood-vessel placed farther from the mouth, and above the masticatory apparatus which surrounds the oesophagus. It is probable that this separation of the blood-vessel from the nerve-ring is due to the development of the masticatory apparatus. In the Holothuroi'da the adoral blood- vascular ring is connected with the nerve-ring, but is placed inside it, and nearer the mouth. It may break up into a plexus. The aboral vascular ring found in the Asterida and Echinida does not appear to have so much morphological importance, as it is confined to a few divisions. Other vessels, which surround the generative glands, and there form wide sinus-like spaces, pass into it, in addition to the vessels from the perisome. In Comatula also a vessel, which forms a covering around the genital chord, is continued into the arms and pinnulae. In the Asteroida and Crinoi'da the vessels of the enteric canal are not independent. In Comatula they form a network, with wide meshes, in the coelom ; this is connected with the oral vascular ring*. A bundle of vessels passes from this net- work, along the axis of the cup to the centrodorsal plate, forming a special organ widened out into five chambers, the importance of which is not known. The enteric vessels in the Echinoi'da and Holothuroi'da are more independent. A dorsal and a ventral vessel can be distinguished, which have just the same characters, as the same vessels in the Vermes (cf. § 138). In Echinus the dorsal vessel is double, for in addition to the one which runs directly on the enteron there is one a little way from it, which gives off branches to the former one, as well as to the enteron. In the Spatangidas the ventral vessel has been observed to communicate with the water- vascular ring. The enteric vascular trunks of the Holothuroi'da are enlarged in the middle of their course, and the dorsal vessel passes into retia mirabilia. WATEE-VESSELS OF ECHINODEEMA. 219 Water-Vessels. § 177. Iu describing the ambulacra (§ 160), mention was made of a " water-vascular system," which took in water from the exterior, and carried it to the ambulacral organs, which it put into the condition of erection. Other organs in addition to the structures which take part in locomotion are filled by this system of canals ; and these we have already spoken of as modifications of the ambulacral feet. The probability of this system of canals being a portion of the blood- vascular system has been already pointed out. Communications have been noticed at several points; and in some cases openings into the coeloni also have been distinctly observed. It is, however, not yet certain how far these vessels have been formed from other organs. In any case we must still regard the water- vascular system as being independent, especially as its development shows that it is so, and as an important division of the system (stone-canal, etc.) arises as a structure, which is primitively quite independent of the circula- tory system. In the larvse of the Echinoderma the water-vascular system is formed by a differentiation from the earliest rudiment of the enteron ; as it gets nipped off, it forms a transparent tube, ciliated in- ternally, and connected with the integu- ment on the back of the larva, where it soon opens by a pore. When in this condition the organ has a close resem- blance to the excretory organs in the larvas of many Vermes (Sipunculidas), so that from this point of view it does not seem improbable that the water- vascular system has been differentiated from a primitive excretory apparatus. This tube, with the other rudiments of the Echinoderm (Fig. 110, A), becomes gradually surrounded by the perisouie; it then changes its form by becoming metamorphosed into a five-leaved rosette (i). The portion which still continues to open to the exterior by the dorsal pore gradually changes its position, and gets to lie on the ventral surface of the Echinoderm ; each leaf of the rosette is now developed into an elongated canal with lateral diverticula ; it is like a pinnate leaf, and forms the rudiments of the ambulacral portion of the water- vessels. In the Fig. 110. Larva of an Asterid (Bipinnaria) with budding Echinoderm. e e! d' g g' Pro- cesses of the body, b Mouth, o Anus of the larva. A Body of the embryonic Echinoderm. h Ciliated tube, i Ambulacral rosette (Rudiments of the water- vessels) (after J. Midler). 220 COMPARATIVE ANATOMY. Holothuroida tlie similarly rosette-shaped rudiment forms the oral tentacles, which have therefore an indubitable relation to the ambu- lacra! system (§ 162). The succeeding processes which are of any importance affect the central portion of the rosette, in which the canals of the five leaves have their common orifice. This is con- verted into a circular canal, which continues to form the central portion of the apparatus ; the canals in the leaves of the rosette grow out radially, and extend whilst the number of their lateral branches increases, over the ambulacra, which get larger at the same time. The adult stage may be directly derived from these arrangements, formed during the development of the Echinoderm body. A branched vascular system (Fig. Ill) has finally developed from the primitive tube, and has its ends directly connected with the suckers (p) and other such processes. The radial trunks of this system communi- cate with the circular canal (c), and this again is in connection with the surrounding medium. We have already mentioned the fact that in Spatangus the water- vas- cular ring around the mouth is connected with an enteric vessel ; and inasmuch as the contents of the two systems of canals are similar, it is very probable that they do not only communicate, but are also to be regarded as parts of one structure. The connection with the ex- terior is of a special nature, and is effected in various ways. As the Echinoderm is being differentiated in the larva, that portion of the rudimentary water-vascular system which is taken into the body of the Echinoderm remains connected with the perisome at one spot, where a porous calcareous plate — the madreporic plate (m) — is de- veloped ; this plate communicates with the lumen of that portion of the canal which is connected with it. The duct (m') leading from the madreporic plate to the circular canal, which also is a portion of the primitive water-vascular system, has ordinarily calcareous substances deposited in its walls, and is therefore called the stone -canal; its walls form a complicated cavitary system. Water passes into the stone-canal by the cribriform madreporic plate, and thence to the circular vessel. It has also connections with the ccelom. Fig-, ill of the Diagrammatic representation ■water-vascular system of a Starfish, c Circular canal, ap Polian vesicles, m Madreporic plate, m1 Stone- canal, r Radially-arranged principal trunks (Arnbulacral canals), r' Lateral branches, p Suckers, a Their ampullae (part only of the arnbulacral canals and their appendages are figured). WATER-VESSELS OF ECHLNODERMA. 221 The portion corresponding to the stone-canal is not always con- nected with the perisome. In the Holothuroi'da the connection is broken close to the dorsal pore of the larva ; the latter disappears, and the stone-canal hangs freely in the body-cavity, whence it takes up water by a very complicated and porous terminal apparatus. There are further complications of the water-vascular system, duo to the formation of contractile diverticula of the water-canals projecting iuto the body-cavity ; these must be mentioned in addi- tion to the arrangement, just sketched. These diverticula vary greatly in character; on the circular canal they form large pear- shaped vesicles (Polian vesicles) (ap) ; where the ambulacral canals pass into the sucking feet they form small ampul las (a), which always project into the body-cavity, and which may be regarded as enlargements or diverticula of the branches of the ambulacral canals. They are cavernous in structure. Both these kinds of organs serve as receptacles for the fluid passing into the canals, and owe their structure to their adaptation to the function of this vascular system ; that is to say, when the suckers are drawn hi, their ampullae are always filled, and when the suckers are pro- truded the contents of the ampullae swell them out. What the ampullae are for the separate suckers, the Polian vesicles of the circular canal are for the whole system of canals ; that is, they allow of a much more rapid action of the ambulacral structures, whether these are pushed out or drawn in, than would be possible if the quantity of fluid needed for the erection of each separate sucker must be first taken in either by the stone-canal or the madreporic plate. This activity of the ampullae of the suckers and of the Polian vesicles of the circular canal is due to the contractility of their walls, in which a muscular layer has been made out. The dis- tribution of the fluid is also regulated by muscular fibres, which here and there enclose the canals. In addition to this the ciliated epithelium which is found throughout the water-vascular system serves to distribute and continually change the water, and so without a doubt render it efficient as an organ of respiration. § 178. The arrangement already sketched in a general way applies most com- pletely to the Asteroi'da. In them the stone-canal is always inserted into a madreporic plate, which, as a rule, is placed interradially on the „.„.,„„, ,, , t ' , ■■> £ . t i t t Fig. 112. Transverse section through dorsal surface of the body. In some th£ stoue.canai 0f Astropecten cases there are several (2-5) madre- aurantiacus (after E. Teuscher). poric plates, and the number of stone- canals is then also proportionately increased; this condition, how- ever, does not remain constant in the species of the same genus. It is to be regarded as the more primitive one ; it is important 222 COMPARATIVE ANATOMY. therefore to know the earliest rudiments of this arrangement. The stone-canal always runs close to the heart-like tube. Calcareous bodies are deposited in it and form a fine network; they do not differ from those found in the perisome. They are arranged in rings ; internally to them there is a longitudinal ridge from which arise two coiled and thinner lamella3, similarly calcified. The cavities which commence at the fine pores of the madreporic plate pass between these lamelke. The ambulacral canals (Fig. 100, A a) extend along the skeleton of the arms, embedded in the ambulacral groove, and give off branches to the feet which arise between the lateral processes of the segments of the ambulacral skeleton; the ampulla of the feet pass inwards through the clefts between the calcified segments, and so come to lie within the arms (ap). At the points where the ampullre are connected with the ambulacral feet there are valves, which shut when the ampullas contract (Asteracanthion rubens) . The number of Polian vesicles varies ; they are sometimes increased in number, and form racemose tufts (Astropecten auran- tiacus), or they may be altogether wanting. In the Ophiurida the stone-canal is inserted into a plate sur- rounding the mouth ; but this plate is not formed in the same way as the madreporic plate, but so that the stone-canal takes up fluid from the body-cavity only. At the circular canal the stone-canal widens out into an ampulla, and is attached to an interradial portion. Polian vessels are not always present. The suckers have no ampulla). In the Crinoida the ambulacral water-vascular trunk runs below the radial blood-vessel, and sends branches into the tentacles of the arms, as well as of the pinnulae (Fig. 115, iv). The radial trunks meet in a circular oral canal, which sends off short canaliculi, with open mouths, into the ccelom. They take the place of the stone- canal, which is not present. As there are no ampullae or Polian vesicles either, the water- vascular system is in the Crinoi'da of a lower grade than in the other divisions. The Echinoi'da are allied to the Astero'ida. The madreporic plate always lies at the aboral pole ; it is either formed by one of the genital plates (Fig. 103, m), or by several of these, or an interradial plate is converted into the madreporic plate, or it is formed by a special plate (Clypeastridas). The stone-canal is sometimes soft (Echinus) and sometimes provided with firm walls (Cidaris). The circular canal, provided with five Polian vesicles (these are absent in the Spatangidaa), lies in the Echinida at the base of the masticatory apparatus, and gives off its ambulacral canals downwards, whence they radiate out to the ambulacra. On the inner side of the shell, and running along each of the ambulacral area3, are the branches of the ambulacral canals, which are distributed to the pores of the calcareous plates, and supply the suckers or their equivalents which arise at this point ; and give origin to transversely-placed ampullar enlargements (Fig. 116, a). In the Holothuroida, owing to the separation of the connecting piece, which later on functions as the stone-canal, from the perisome of the larva, which passes into the substance of the Echinoderm, WATEE- VESSELS OF ECHINODERMA. 223 these parts have a different relation to that which they have in the rest of the Echinoderma. The walls of the stone-canal, which hangs freely into the coelom, are more or less calcified; when they are more so they form a firm capsule. The porous parts of the canal are usually distinguished by calcification, and so repeat within the body the arrangements of a madreporic plate. The ends of each branch carry a porous piece when the stone-canal is broken up into branches ; this repetition of parts leads to the formation of race- mose structures, which are only functionally similar to a number of madreporic plates grouped around the stone-canal. The stone- canals vary in number, as well as in arrangement. Often only one is present ; in other cases, and notably in the Synaptas, there are several arranged around the circular canal. The number too of the Polian vesicles (Fig. 113, p), which are present in these forms, varies; in Holothuria and Molpadia there is one, in Synapta Beselii about fifty, and in Cladolabes about a hundred. The canals from the circular canal (0) run for- wards inside the calca- reous ring (ii), and give off branches to the oral tenta- cles (T) ; a cascal elongated tube, corresponding to the ampullas of the suckers, is connected with each of them, These tubes are of some size in the Holothurida, and lie on the outer side of the calcareous rine- : they are only feebly developed in the Synaptidse. The radial trunks going to the ambulacra are placed, in Holothuria, in the bundles of longitudinal muscles, which are thus divided into two halves. In Cucumaria they are placed on the outer side of these muscles. The branches of these canals, as in other forms, go to the feet. When the feet are atrophied the vascular branches, which go to them, are atrophied also ; but the principal trunks appear to persist even in the Apodia, for they have been observed in Synapta, although, indeed, diminished in size. Fig. 113 Longitudinal section through the anterior part of the body of Synapta digitata. R R' Calcareous ring, r Muscles passing from it to the oesophagus, o Mouth. D Enteric tube. C Circular canal, t "Canals to the tentacles T. p Polian vesicles, n Nerve-ring, n1 Eadial nerve-trunk, passing through the calcareous ring R1. m Bands of longitudinal muscles. G Ducts of the generative organs (after Baur). 224 COMPAEATIVE ANATOMY. Excretory Organs, § 179. The arrangements commonly found, among the Annulata (looped canals, or nephridia) are not found in the Echinoderma, but there are signs that these organs, or organs of the same type at any rate,, are not entirely foreign to the organisation of the Echinoderma. For in the Holothuroida two canals have been observed to run in the wall of the body; and these are beset with infundibular organs, which open into the ccelom (Chirodota pellucicla). In the Synapta) also there are organs which correspond to the internal mouths of the looped canals of Vermes, but these are not connected with canals, Finally, in the Crinoida ciliated organs have been made out in the dorsal canal of the arms, which is a continuation of the ccelom. It cannot be definitely asserted that these structures are all of the same kind, but from the characters of those first mentioned it is probable that they have relations to an excretory apparatus. We can as yet only suggest that a fundamental relation of the same kind holds for the ambulacral water-vascular system. Anyhow the arrangement of this system in the body does not justify us in regarding inquiries in this direction as barren ones, for the excretory organ of many Mollusca (Nudibranchiata) has the form of a richly-branched system of canals; and the communication of the water-vascular system with the exterior, as well as with the blood-vessels (or, what is the same thing, with the ccelom), can hardly point to its being anything else than part of an excretory apparatus. Generative Organs. ,§ 180. The asexual methods of reproduction so common among Vermes do not obtain in the Echinoderma, except in so far that the animal itself is the product of gemmation. An indication of this mode of reproduction is, however, retained in the Asteroida — in the regener- ation of lost antimeres (arms). Almost all Echinoderma — there are but few exceptions — have the sexes separate, and their organs arranged conformably with the radiate type. The male and female organs arc both very simple in character, and can only be distinguished easily when the generative products are mature, the ovaries being generally dis- tinguished by the brighter coloration of the eggs, which are yellow or red, whilst the testicular tubes are almost always white. The form-elements of the sperm are very generally fila- mentous structures provided with a small head. The generative GENERATIVE ORGANS OF ECHINODERMA. 225 system is simple in structure, the excretory ducts are not com plicated, and there is no intromittent organ, so that the surrounding water is the medium of communication in impregnation. On the whole there is a great resemblance to the corre- sponding structures found in Vermes. The lowest stages, both as to num- ber, arrangement, and as to the more special characters of the organs, arc seen in the Asteroi'da. The testes or ovaries are tubular or lobate glan- dular canals, which in some forms are arranged in two rows, and disposed conformably with the metamerism of the arms (Ophidiaster, Archaster). In others there are only two groups in each arm, which may extend along the whole cavity of the arm (this is the case even in Brisinga), or may be limited to the interradial space merely (Fig. 108, g). A comparison of these characters shows us, that there is a gradual reduction in the number of the generative glands, which corre- sponds with the gradual centralisation of the organism, which we have Fig. 114. Generative organs of an Ophiurid (Ophioclerma longicauda). The dorsal integument and the digestive organs have been removed, r Arms, g Ovarian acini. already seen to occur in the Asteroi'da. In the aproctous forms the tubes have no excretory orifices, and the genera- tive products are passed into the body- cavity. We do not yet know how they reach the exterior. In other Asteroi'da the generative glands open to the exterior by special plates, dis- tinguished by their fine pores (cribri- form plates), and placed in the dorsal interradii ; or they have a single duct with a narrow orifice (Pteraster). Each organ is surrounded by a blood sinus, which envelopes the separate lobes and lobules. The generative products pass into this sinus and are not evacuated directly. The structure and arrangement of the generative organs of the Ophiurida is the same as in the Asteroi'da. There is only one example of hermaphroditism (Ophiura squamata). The generative glands (Fig. 1 14, dach, De Apodis cancriformis anatome, lstl. — Gkube, Bemerkungcn fiber die Phyllopoden. Arch. f. Nat. 1853.— Leydig, Ueber Argulus foliaceus. Zeitsohr. f. wiss. Zool. Bd. II. Ueber Artemia salina und Branchipus stagnalis. Ibid. Bd. III.— The same, Naturgeschichte der Daphnidcn. Tiibingen, 18G0.— Daewik, A Monograph of the Subclass * Forms degenerated by a parasitic habit. f Theso too show the influence of the parasitic habit. X They have undergone modification owing to a parasitic habit. ARTHROPOD A. 233 Cirripedia. I. II., 1851,1853.— Zenker, W., Anatomisch-systemat. Studien iibei" die Krebs- thiere. Arch. f. Nat, 1851. — Van Beneden, Recherchcs sur la faune littorale de Belgique. Crustacea, Acad. Bruxelles, 1861. — Claus, Die frei lebenden Copepoden. Leipzig, 1803. — The same, Ueber den Ban und die Entw. parasitischer Crustaceen. Cassel, 1858.— The same, Beitrage zur Kenntniss der Bntomostraken. Marburg, 1800.- — The same, Ueber einige Schizopoden. Zeitschr. f. wiss. Zool. XIII. — The same, Beobacht. iib. Lernseocera, etc. Marburg u. Leipzig, 1808. — The same, Beitrage zur Kenntniss der Ostracoden. Marburg, 1808. Die Metamorphose der Squilliden. Gott. 1871. — The same, Zur Kenntn. d. Bnues u. d. Entwickl. von Brancliipus stagn. u. Apus cancriform. Gott. 1873. Zur Kenntn. d. Baues u. d. Entw. von Branchipns u. Apus. K. Gesellsch. d. Wiss. z. Gott. Bd. VIII. — The same, Z. Erforsch. d. genealog. Grundlage des Crustaccen-Systems. Wien, 1876. — Muller, Fr., Fiir Darwin. Leipzig, 180-1. Pcecilopoda : Van der Hoeven, Rech. sur l'hist. nat. et l'anatomie des Limules. Leyden, 1838. — Packard, A. S., The development of Limulus. Mem. Boston Soc. Nat. Hist. Vol. II. —Owen, R., On the Anatomy of the American King-Crab. Trans. Linn. Society. Vol. XXVIII. — Milne-Edwards, A., Rech. sur l'anat. des Limules. Ann. Sc. Nat. V. Ser. Tome XVII. Protracheata : Grube, Ueber d. Bau v. Peripatus Edwardsii. Arch. f. Anat. 1S53. — Saenger, N., Perip. cap. und Perip. Leuckarti. Verhandl. der ersten russ. Naturforscher versamml. zu Moskau, 1869 (russ.). — Moseley, H. N, On the Structure and Development of Peripatus capensis. Phil, Trans. London, 1874. P. II. Tracheata. Arachnida : Treviranus, G. R., Ueber den inneren Bau der Arachniden. Niirnberg, 1812. — Duges, Recherches sur l'ordre Acariens. Ann. sc. nat. II. i. n. 1834. — The same, Sur les Araneides. Ibid. II. vi. 1856. — Doyeee, Sur les Tardigrades. Ann. sc. nat. II. x. 1810. — Tulk (Opilioniden). Ami. nat. hist. 1843. Fror. Not. Bd. 30. — Newport, On the nervous and circulatory system in Myriapoda and macrourous Arachnida. Philos. Trans. 1813. — Quatrefages, Organisation des Pycnogonides. Ann. sc. nat. III. iv. 1815. — Van Beneden (Linguatula), Acad. Bruxelles, 1849. — Leuckart, Bau und Entwickelungsgesch. d. Pentas- tomen. Leipzig u. Heidelberg, 1800. — Ddfoce, L., Hist, anatomique et physiologique des Scorpions. Acad. d. sciences (Savants Strangers) XIV. — The same, Anat. physiol. et hist, nat. des Galeodes. Acad, des sciences (Savants Strangers) XVII. — Kittary, Anat. Unters. v. Galeodes. Bull, de la soc. imp. des Naturalistes dc Moscou, 1818; and in Froriep's zoolog. Tagesberichte Nr. 108. — Stecker, A., Anatom. u. Hist, iiber Gibocellum. Arch. f. Nat. 1876, Myriapoda : Treviranus, G. R. (Scolopendra und Julus), Vermischte Schriften. II. Bremen, 1817. — Difour, L., Recherches anatomiques sur le Lithobius fortificatus et le Scutigera lineata. Ann. sc. nat. II. 1824. — Muller, J., Zur Anat. der Scolopendra morsitans. Isis, 1829. p. 549. — Brandt, Beitrage zur Kenntniss des inneren Baues von Glomeris marginata. A. A. Ph. 1837. — Jones, R., " Myriapoda" in the Cyclopaedia of anatomy and physiology. Vol. III. 1842. — Newport, On the organs of Reproduction and the development of the Myriapoda. Phil. Trans. 1811. — The same. On the structure, relations, and development of the nervous and circulatory system in Myriapoda and macrourous Arachnida. Phil. Trans. 1843. Insecta : Reaumur, Mernoires pour servir i\ l'histoire des Insects. 1734-42. Paris. 6 vols. — ■ Swammerdam, Bibel der Natur. 1752.— Lionet, Traits anatomique de la Chenille qui rouge le bois de saule. La Haye, 1702. — Steauss-Durckheim, Considerations sur l'anatomie comparee des animaux articules, auxquelles on a joint l'anatomie descriptive du Melolontha vulgaris. 1828. — Burmeister, Handbuch der Entomologie. Bd. I. Berlin, 1833. — Newport, "Insecta" in the Cyclopaedia of anatomy and physiology. Vol. II. 1839. — Dufour, L., Recherches anatomiques et physiologiques sur les HemiptBres. Mem. Acad, des sc. (Sav. strangers) IV. 1833. — The same, Sur les Orthopteres, les Hymenopteres et les Nearopteres. Ibid. VII. 1841. — The same, Sur les Dipteres. Ibid. XI. 1851. — Pictet, Recherches pour servir a l'hist. et a l'anatomie'des Phryganides. Geneve, 1834. — Leuckart, Die Fortpflanzung u. Ent- wiek. der Pupiparen. Halle, 1858. — Brauer, Pr., Z. Anat. d. Neuropteren in Schrift. d. zool. -hot. Vereins z. Wien. — Works by Loew in various entomological journals. — Leydig's numerous researches into the more intimate structure of Insects. — Weissmait, Die Entwickelung der Dipteren. Leipzig, 1864. — Landois, L., Anat. d. Hundeflohes. N. Acta Acad. L. C. Vol. XXXIII. — The same, Anatomic der Bettwanze. Zeitschr. fiir w. Zool. Bd. XVIII. XIX. — Lowne, B. Th., Anat. and Phys. of the Blow-Fly. London, 1S70. — Muller, Fr., Z. n. Kenntniss d. Termiten. Jen. Zeitschr. Bd. X. — Kowalevsky, A., Embryol. Studien an Wiinnern und Arthropoden. M-.. j* ■jp I m \\ Fig. 131. Nervous system of Insecta. A Of Termes (after Lespes). B Of a Beetle (Dytiscus). C Of a Fly (after Blanchard). gs Supra-cesophageal ganglion (Cerebral ganglion), gi Sub-cesophageal ganglion. ' gr g2 fi Fused ganglia of the ventral chord, o Eyes. Myriapoda. It is only when the Insect passes out of its larval condition to the perfect one that changes appear. The development of some metameres, the intimate fusion of others to form larger portions of the body, the greater development of the appendages, which persist on some metameres only, with the consequent increase of the muscular supply, as well as many subordinate arrangements, must be considered as affecting the changes which take place in the nervous system. The decrease in number of the ganglia, by the shortening of the longitudinal commissures, and the consequent fusion of separate ganglia, produces a shortening of the whole ventral chord. Owing to the independent character of the head of the Insecta, in comparison with the other regions, the first ganglion of s 258 COMPARATIVE ANATOMY. the ventral chord, which, is primitively composed of three, remains embedded in the head (sub-oesophageal ganglion), and takes no share in the concrescence which affects the other ganglia ; it is only in a few cases — in Insecta degraded by parasitism — that a union of this ganglion with the rest of the ventral chord takes place. The cerebral ganglion (Fig. 131, A B G, gs) is almost always distinctly divided into two halves, each of which is again composed of several smaller masses of ganglia, which are often complicated in structure. The ganglia of the ventral chord are primitively paired, and often become closely united. On the other hand, the longitu- dinal commissures remain double, even when they are closely applied to one another. There is also a separation of the ventral chord into a superior and an inferior portion which corresponds to a physiological differentiation. The first ganglion of the ventral chord (G. infracesophageum) sands off fibres for the organs of the mouth. The three succeeding thoracic ganglia principally give off nerves to the appendages, feet, and wings ; they are consequently of some size. On the other hand, the succeeding ganglia are, as a rule, small, the last alone being an exception to this, for it is of a larger size in correspondence with its relations to the generative system. Even in the Aptera there is a fair amount of variation, for 11 ventral ganglia (Lepisma) can be made out in the Thysanura, while there are only 3-4 in the Collembola. The last portion of the ventral chord seems to form a complex of ganglia in many (Orchesella, Achorutes) . As to the Pterygota, the chief point is that of all the orders the least amount of metamorphosis is seen in the Pseudoneuroptera. In them the ventral medulla traverses the whole length of the body, and there are 5-9 abdominal ganglia in addition to the three thoracic (Fig. 131, ^1). The Orthoptera, which have 5-7 abdominal ganglia, resemble them in this. There are great variations in the Coleoptora. In some the ventral chord extends as far as the end of the abdomen, sometimes possessing 8 separate ganglia (e.g. in Cerambycidaa, Carabida), etc.); in others again the 3 ganglia of the thoracic portion are only represented by two, the second and third being fused, whilst the abdominal ganglia are also connected into one mass, which im- mediately succeeds the preceding ganglion (CurculionidaandLamel- licorniaa). In other families there are connecting links of various kinds between these conditions which represent the extremes. In the Hymenoptera we generally find the thoracic ganglia reduced to two, while the abdominal part of the ventral chord has frequently 5 or 6 separate ganglia. These are in many, however, reduced to 4 or 3, and even to one. The abdominal part of the ventral chord is, in the Hemiptera, placed in the thorax, and is represented by a ganglionic mass connected with the thoracic ganglia, which are also simple, by a commissure of varying length. The nerves for the abdomen con- sequently take a longer course, and form two longitudinal trunks NERVOUS SYSTEM OF ARTHROPODA. 259 wliicli arise from the last ganglion. A similar difference in the number of the ganglia of the ventral chord obtains in the Diptera, where the most primitive characters are seen in Pulex : 3 thoracic and 7-8 abdominal ganglia. In others there is generally a consider- able reduction by the fusion of the thoracic, or of the abdominal ganglia, or of both (Fig. 131, C). With this is connected the com- plete fusion of the ventral chord into one somewhat long knot, in the parasitic Pnpipara. We find the same characters in the Strep - siptera. As to the Lepidoptera there is more uniformity in them, a constant number of ganglia being found in the larva, while when it is metamor- phosed into the butterfly the mode of fusion appears to be essentially the same in all. § 200. The visceral nervous system of the Arthropoda shows signs of some common characters, together with great variation in particular points. Among the Crustacea nerve-filaments pass from the oesophageal commissure to the enteron, or a nerve passes to the enteric canal from the ventral chord. (In Astacus from the last ganglion also.) Even in the Arachnida nerves are given off partly from the cerebrum, and partly from the ventral ganglion to the enteron ; in the Opilionea the posterior ones are pro- vided with a large number of ganglia. In the Insecta and Myriapoda the break- ing up of the visceral nervous system into several portions has been more generally made out; we will therefore examine the arrangement of it more closely. One part forms the so-called paired system, which consists of two branches running back- wards from the cerebral ganglion to the sides of the oesophagus ; these form a simple chain of ganglia (Fig. 132, s' s") on either these ganglia varies, plexus-like connection with the unpaired system, to determine to which system they belong. The unpaired system (r r1) arises in a ganglion which lies in front of the cerebrum, and is more or less connected with it. From this ganglion a thicker nerve (r) passes backwards over the oesophagus to the stomach, and forms a plexus with the branches of the paired system ; from this plexus the neighbouring parts, especially those of the digestive system, are innervated. In many Insects this nerve (N. recurrens) forms a single ganglion (Coleoptera and Orthoptera), in others several (Lepidoptera) s 2 Fig. 132. Supra-cesophageal ganglion and visceral ner- vous system of one of the Lepidoptera (Bonibyx Mori), g s Supra-oesopha- geal ganglion (Cerebrum). a Antennary nerve, o Optic nerve, r Azygos trunk of visceral nervous system. r' Its roots arising from the supra-cesophageal ganglion, s Paired nerve with its ganglionic enlargements s's" (after Brandt). side. The number of nd it is often difficult, on account of their 260 COMPARATIVE ANATOMY. There is yet another system of nerve-branches in connection with this plexus ; it is principally applied to the large branches of the tracheae, and the muscles of the stigmata. This arrangement is brought about by a nerve-filament, which runs on the surface of the ventral chain, and is divided into two fork-like branches in front of each ganglion (Nervi transversi accessorii). The branches receive nerve-twigs from the upper chord of the ventral chain, and pass partly outwards to the branches of the trachea) and tho muscles of the stigmata, and partly backwards, whore they unite in the middle line, and at the next ganglion repeat this arrangement. Sensory Organs, Tactile Organs. § 201. The sensory organs of the Arthropoda are, for the most part, allied to those of the Vermes. All but a few indicate a connection of this kind, and these few are to be re- garded as arrangements which are developed in this division only. Tho carapace-like covering of the body of most Arthropoda requires special organs to produce the sensation of touch ; the form- elements of these are connected with ganglionic cells, and form rod-like nerve-endings. These ganglionic cells are generally struc- tures which are derived from the ecto- derm, and the whole apparatus not unfrequently retains its primitive position. These end organs, which are found in the most different parts of the body, form indifferent sensory organs, which in certain parts take on the form of tactile organs (cf. Fig. 133). Organs of this kind are mostly found on the appendages, where they present rod- shaped projecting ends. In the division of the Crustacea these tactile rods have been recog- nised in many forms, and that not only on the antennae, especially in the lower Crustacea, but even on other appendages of the body. In the Myriapoda and Insecta there are tactile rods on the antenna?, and in the latter they are also found on the tarsal joints of the feet. Fig. 133. Nerve - ending with tactile rods, from the proboscis of a Fly (Musoa). n Nerve. g Ganglionic swelling. s Tactile rods, c Fine hairs of the cuticle (after Leydig). SE1STS0EY OEGANS OF ARTHROPOD A. 261 In addition to these tactile rods special organs resembling them are found on the antennas of the Crustacea and Insecta ; they are sometimes of considerable size, and are innervated in the same way as are the tactile rods. In the Crustacea they are formed only on the inner (anterior) pair of antennas. In the Insecta they are much shorter, and are conical in form. Their position, in addition to the circumstance that they are less long than the indifferent bristles, or are placed in depressions, makes it probable that these organs have another function, and it is very easy to suppose that they arc organs of smell, or at least of a sensation very much like it. In this case the antennas, by the differentiation of special nerve-endings, have more than one function, and do not merely preside over the sense of touch. Auditory Organs. § 202. Auditory organs are not widely known in the Arthropoda, no sign of them having been seen in the Myriapoda and Arachnida ; on the other hand, in some divisions of the Crustacea and Insecta, organs may be made out, which appear to be adapted for the sensation of sound. There are two principal forms of the organ, which arc exactly correlated with the medium in which the animal lives. One form is found in the Crustacea, and consists of a saccular space, formed by an inpushing of the integument ; it sometimes remains open and is sometimes closed. These auditory vesicles lie, in most of the higher Crustacea, in the basal joint of the internal antennas. Thus in Leucifer, Sergestes, and other Malacostraca, as also in the Arthrostraca (Hyperida), a pair of these organs may be found in front of the cerebrum. As secondary structures they may also be found on other parts of the body. Thus in the Mysidas, they lie in the two inner lamellas of the fan of the tail. There are firm structures, otoliths, in the auditory vesicles, which, when the vesicles are closed (in Mysis and Hippolyta), consist of a concretion, which is held fast by fine, regularly-arranged hairs. When they are open, as they are very commonly among the Decapoda, and also in Tanais, the orifice is greatly complicated. The place of the otoliths is here taken by grains of sand brought in from the exterior ; these are regularly attached by special hairs, which arise from the wall of the auditory vesicle. They are like the other hairs of the integument, but are distinguished from them by not having their shaft directly con- nected with the floor of the vesicle, most of them standing on a fine membranous process, to which endings of nerves pass. In this they agree with the rod-like processes which carry the otoliths in the Mysidas, to which nerves likewise pass. The auditory nerve in these forms, in which the auditory vesicle is embedded in the internal 2(32 COMPARATIVE ANATOMY. antennae^ is a branch of the internal antennary nerve. Both struc- tures thus represent the end organs of nerves, which are set in vibration by the shaking- of the firm body (Otolith) which they carry, and thus produce an excitation of the nerve. The general character of this remarkable system shows us how the auditory organs arise from a differentiation of an indifferent sensory organ connected with the integument. The auditory hairs are only modifications of other " hairs ;' of the integument which contain nerve-endings, and just like those which may appear on the free parts of the body (tactile rods). The formation of the unclosed auditory vesicle, or auditory pit, represents therefore a second stage in this differentiation ; and the change into a closed vesicle is a further stage of this phenomenon. Hensen, Zeitpclir. f. wiss. Zool. XIII. § 203. The other form of auditory organ is found in the Insecta. It is principally in the Orthoptera, which are also provided with vocal organs, that an organ for receiving the waves of sound can be made out. The ordinary arrangement is a tympanic membrane, stretched on a firm chitinous ring, one surface being directed to the exterior, the other to the interior. On its inner side there is a tracheal vesicle, and on this, or between it and the " tympanum," there is a ganglionic nervous enlargement, from which specially modified nerve-endings, having the form of small club-shaped rods, arise between fine filaments. The tympanum, as well as the tracheal vesicles, serves as an organ for conducting sound. The organs of perception are represented by the nerve-endings, which are regularly arranged. In the Acridida the organ lies in the nietathorax, just above the base of the third pair of legs, and receives its nerve from the third thoracic ganglion. The Locustida and the Achetida have the organ embedded in the tibia of the two anterior legs. In the former a tympanum lies on either side of this leg", either super- ficially, or at the bottom of a cavity, which opens anteriorly by a single orifice. Two tracheal branches occupy the space between the two tympana, one of which carries the ridge-shaped nervous end- organ. This auditory ridge is, in Locusta, formed by a series of cells which grow smaller towards one end; each of these contains a ' ' rod " of proportional size. The tympanum in the Achetida lies on the outer side of the tibia of the anterior leg. Other organs, the nature of which is less definitely settled, are allied to these, as by their general structure they represent auditory oi'gans; the presence of the same pencil-shaped body in the termi- nations of nerves justifies us in at least ranking these organs with the auditory, while, further, such a relationship is implied by the ganglionic outspreading of the proper nerve along a tracheal branch. The ends of the nerves are directed towards the integument, the VISUAL ORGANS OF ARTHROPODA. 263 chitinous layer of wliicli is always provided with a number of closely- applied groups of pore- canals instead of a tympanum. Organs of this kind have now been recognised at the root of the posterior wing of the Coleoptera, as well as on the base of the halteres of the Diptera. The two forms of auditory organs in the Arthropoda are indeed very different from one another in the details of their arrangements, but there is, nevertheless, a connection, for in both cases the chitino- genous cellular layer gives rise to parts which carry the special end- organs; in the Crustacea these are connected with processes of the integument, the auditory hairs; while in the Insecta they are con- verted into the small pencils, and are consequently differentiated in another direction ; they remain within the dermal skeleton, and have no relations to the processes of it. No homology can be made out between these organs, owing to the diversity of their position, and from the fact that more complicated organs are derived from an elementary arrangement, which is distributed generally in the integument. Letdig, Arch. f. Anat. u. Phys. 1855. — Graber, V., Die tympanaleu Sinuesap- parate der Orthopteren. Denkschr. d. Wiener Acad. M. N. CI. Bel. XXXVI. Visual Organs. § 204, In the visual organs of the Arthropoda we meet with points of resemblance to certain forms of eye found in the Vermes ; to those, namely, in which a number of end-organs of the optic nerves are placed directly beneath the integument (Sagitta., Hirudinea, etc). But they have no close affinity to the more developed eyes of the Annelides, which are distinguished by the possession of a separate lens (§ 125). In Arthropoda, as in other forms, the integument is the spot at which the eye is differentiated; its mode of composition out of the elements of the integument will be understood by a reference to the sub- jacent diagram; although, of course, this does not represent the simplest condition. The articular layer of the inteo-nment forms a biconvex thickening over the eye (/) ; this forms a refractinsr but also a defen- sive organ, and functions therefore as a cornealens. The eye, which is formed from the hypodermic layer (It), lies behind this lens. Around it the hypodermic cells elongate and change their position; they become pigment cells (p). The optic cup, into which project transparent Fig. 134. Section through the simple eye of a young Dytiscus larva (after Grenacher). 264 COMPARATIVE ANATOMY. cells (•>; •J 5 a ceecal uterus which is continuous with the duct. At their hinder end they seud off a short duct (od) to the generative pore. The organs of either side are united in the same way in the testis. This is formed of a double row of glandular follicles, which unite into a coiled canal, which forms the simple excretory duct and opens at the base of the last pair of feet. The generative organs of the Decapoda resemble those of Mysis, by being similarly connected in the middle line ; they appear to be further developed by various differentiations. The female A organs are formed by two long tubes which run for- wards and backwards, and are united transversely with one another; these tubes function partly as germinal glands, but chiefly as ovi- duct and uterus. In the Crayfish the two anterior divisions have the form of short lobes, while the two hinder ones are fused into an unpaired piece. On each side a short duct passes to the genital pore, which in the Caridina has the same position as in the Schizo- poda; in the Macrura it is placed on the basal joints of the third pair of feet, but in the Brachyura on the segment of the body, which carries this pair. The Brachyura are also distinguished by a pouch-like enlargement of the oviduct (seminal pouch). In the male appai'atus the testes are formed by two much-coiled tubes, which are transversely connected with one another in front, and which, like the female organs, lie for the most part in the cephalo- thorax; in Pagurus only are they placed in the abdomen. In the latter they give off two long and closely-coiled but gradually-widen- ing ducts. Herein they agree with most of the Decapoda, but they are distinguished from them partly by the increased size of the lobes formed by the coils of the seminal canal, and partly also by the formation of the unpaired piece, which unites the glands of either side. The germinal glands are more completely united in Astacus. A vas deferens with long coils passes on each side to the outer generative orifice, which is placed, as a rule, on the basal joint of the last pair of feet ; in the Brachyura, however, it is found at the end of a penis, which is formed by a metamorphosed appen- dage. The opening of the male apparatus only is then the same Fig. 155. Homarus. deferens. Male generative organs. A Of BOfOniscus. tt Testes, vd Vas vs Seminal vesicle. o Its orifice. p Copulatory organs. 296 COMPARATIVE ANATOMY. as in the Schizopoda, while the female orifice is placed farther forwards. In the generative apparatus of the Stomapoda the testis com- mences as a fine unpaired tube in the middle line of the caudal fin ; it is continued forwards into a paired tract, from which a much- coiled vas deferens arises. Each of these passes to a penis, which springs from the coxa of the last thoracic foot. An unpaired gland, which begins in the cephalothorax, opens at the same point. The ovary passes backwards as an azygos, and afterwards as a paired, gland, as far as the cephalothorax. Each gives off an oviduct in the third thoracic segment, which opens at the base of a pouch placed in the middle line. The Decapod-type prevails in them, though modified in the female by the approximation of the orifices. In the Pcecilopoda we see a combination of the two forms repre- sented in the Crustacea. One form is followed in the median con- nection of the organs of either side, and the other in the large number of germ-sacs formed by the fine terminal branches of the network, which makes up the generative organs. The wider tracts serve as efferent passages, which are considerably widened in the female so as to collect the eggs, and are continued into the efferent duct on each side. § 229. A lower stage is seen in the female apparatus of the Protracheata. The ovary is a body divided into two halves by a septum, and sends off a paired oviduct; this passes forwards as a coiled tube, and then bends round to a widened portion, which functions as a uterus. These canals are continued backwards, and only unite to form a common short vagina near the generative pore. In the male apparatus the testes are completely separated from one another ; each of them is provided with a glandular appendage, and is continued into a long looped vas deferens. A common ductus ejaculatorius, which also opens at the hinder end of the body, is formed by the union of the two efferent ducts. § 230. The two kinds of generative glands in the Arachnida are, as a rule, unpaired; when paired they are connected transversely, and open either by one or two ducts anteriorly, and on the ventral surface. In addition to accessory glandular organs, or special enlargements of the excretory ducts serving for the storage and reception of the sperm-masses or ova, there arc external organs which carry the sexual products outwards, and which are called penes or vagina) according to the sex. The male organs repeat with slight variations the type of the female. The union of the genital glands of either side and the azygos portion of the apparatus which GENERATIVE ORGANS OE ARTHROPOD A. 297 is formed in consequence of it, calls to mind the similar relations of parts in the Branchiate, and notably in tlie Poecilopoda. The ovaries in the Scorpionea are formed of three long* tubes which bend towards, and pass into, one another at their hinder ends, while they are also connected with one another by four transverse anastomoses ; in the walls of these tubes, which often form tubular diverticula, the ova are formed. The segmented character of the organ is implied by the transverse connections, which form four wide meshes on either side, for these segments have exactly the same position as those of the abdomen. Spindle-shaped and widened oviducts are continued on from the two outer longitudinal tubes, which function as receptacula seminis for the sperm which they receive ; they open at the base of the abdomen. The testes also of the Scorpionea are a pair of winding canals, united by transverse commissures. Their double character is implied by the presence of two tubes on either side. The vas deferens of each testis opens to the exterior, after uniting with its fellow of the opposite side at just the same point as that occupied by the genera- tive orifice of the female. In addition to the vas deferens there are accessory organs on either side, which as a rule have the form of two pairs of ceecal tubes, which vary in length and function partly as glands, and partly as seminal vesicles. The separation of the germinal glands of either side is complete in the Galeodea and male Aranea. The ovaries are two tubes, on the outer surface of which the ova are developed; in the Spiders they are developed on stalked processes. In some (Segestria, Oletera) the ovaries are represented by a closed ring. A vaginal canal, which is sometimes widened out (Galeodes), is formed from the union of the two ovarian tubes, which serve to carry the ova out- wards; this canal has one or two seminal vesicles at its termination. These are found also in the Aranea, where they often open inde- pendently in front of the orifice of the vagina. The male organs in the Galeodea may be derived from those of the Scorpionea, by supposing that the transverse anastomoses between the longitudinal trunks have disappeared. Finally, in the Aranea, these longitudinal tubes are reduced to two. Bertkau, Ueber d. Generafc.-Apparatus. Araneiden. Arch. f. Nat. 1875. § 231. In the Opilionida and Acarina the circular form of germinal gland is the dominant and general arrangement ; it is derived from the transverse connection of the ovaries, which is seen in the Scorpionea. The unpaired stage of the germinal gland, which is to be regarded as the more primitive one, is implied by this arrange- ment. This circular form is most perfect in the Opilionida (Fig. 156, B 6). Just as in the Aranea and Scorpionea the ova are formed in stalked diverticula on the surface of the ring, whence they pass 1"JS COMPARATIVE ANATOMY. into the cavity of the ovarian tubes, and from it to the efferent duct, which is provided with a considerable enlargement (it) (uterus). A narrow coiled continuation of this leads to the protractile ovipositor (op). In the male there is a circular canal, a portion only of which forms the testis, instead of the ovarian ring (Fig. 156, a, t); the two ends of the testis pass into the efferent duct (vd) which completes the ring. These unite into a closely-coiled portion, from which a widened canal vesicle arises - similar to the and, like it, protractile ; the penis is attached to it. Two large tufts of acces- sory glands (gi) are con- nected with the ends of the Fig. 156 opilio. or seminal - an organ ovipositor, Phalangium Vas Generative organs A Male organs. t deferens, p Penis, to Retractors of penis. gi Appended glands (after Krohn). B Female organs, o Ovary. it Uterus, op Ovipositor. to Retractors of ovipositor. penis. The circular form of germinal gland is still retained, in its completeness, in many of the Acarina. In the female apparatus the greater part of the ring is converted into the efferent organ, owing to the limitation of the ova-producing part to a small division of it. This is most marked in Pentastomum, where the ovary is attached to a circular canal. The ovary is here differentiated from the canal. The part of the ring, which forms the efferent ducts and passes into the single portion, is often widened out into a uterus ; or the uterus is formed by the unpaired portion alone. This is the casein Pentas- tomum, the uterus of which forms a coiled canal of some length. The unpaired portion of the efferent duct is generally much shortened in the male, and the two parts of the rings connected with it are widened out into seminal vesicles. Appended glands are connected with the unpaired portion in both sexes. The great differences in the distribution of the functions of parts of this canal lead to the separa- tion of the ring into two genital tubes, the middle of the germ- producing portion of the ring becoming sterile. The two halves of the ring are then distributed to the sides, although in some cases they are still connected by a canal, or by indifferent tissue; this gives rise to organs which are only united at their orifices, or along an unpaired portion connected with them (Ixodes). The hermaphrodite generative organs of the Tardigrada are altogether unlike these arrangements. They consist of an unpaired ovary and two testes which lie beside the enteron : their efferent ducts pass into a receptaculum seminis, and open, generally provided with special glands, into the cloaca. GENERATIVE ORGANS OF AETHROPODA. ■2 'J 'J The arrangements in the Pycnogonida are just as peculiar; their generative products are formed on the wall of the coeloni, and are. passed out by special orifices (which are sometimes found on all, and sometimes on only one pair, of the feet). This character reminds us of the lower arrangements seen in the Annulata. The conversion of appendages into copulatory organs, which obtains in the Crustacea, is seen in the Aranea only from among the Arachnida; in the males of this order the palpi are organs of a complicated structure, which convey the sperm to the female generative orifice. § 232. The generative organs of the Myriapoda are, in their form and arrangements, most similar to those of the Arachnida, and, as in these forms, they sometimes open far forwards on the body, namely, on the third segment of it. The genera- tive organs of the Scolopendrida3 are placed in the hinder end of the body. In the females the generative glands are either simple externally and form an elongated tube, on the inner surface of which the ova form projections (Julidas, Scolopen- dridse, and Glomeridas), or they are double (Cras- pedosoma), in which case they are united at their anterior ends, while the oviducts open separately. In the Scolopendridas the simple ovarian tube is, as a rule, continued on by a simple oviduct ; but the double character of these organs is implied in the development of ova on both sides of the ovarian tube. The accessory organs are formed by two pairs of bodies, which sometimes open into the ovi- duct, but ordinarily directly into the genital orifice; they partly form cement glands, and partly receptacula seminis. The male organs also are often double in their efferent ducts and accessory parts only. However, mauy Glomeridas and Julidas are provided with a double testicular tube, which passes into a common vas deferens, and seems to be united into a single organ owing to the large number of its transverse connections (Fig. 157). When there is only one testicular tube, it is beset with separate follicles. The vas deferens is occasionally single (some Scolopendridas) ; but as a rule it is divided into two branches, which either open on a short papilla (Julida?, Glomeridse), or are connected together and continued into a short penis placed at the hinder end of the body (Scolopendridas). The last division of the efferent duct is provided with enlai'gements or diverticula, which serve to collect the sperm. Several pairs of glands are inserted into it just in front of its orifice. As to the general character of the generative organs, they are unmistakably approximated to the Crustacea by the possession Fig. 157. Male ge- nerative organs of Jul us. t Testicular follicles, e Efferent duct (after Steiu). 300 COMPAEATIVE ANATOMY. of separate orifices, and resemble the Araclinida in forming an annular portion. Steix, F.j De Myriapodum part, genital. Berol. 1811. § 233. Notwithstanding the s'reat variations of more subordinate cha- racters, the generative organs of the Insecta present on the whole a well-marked uniformity of structure. The organs and their accessory parts almost always lie in the abdomen, and generally open below, or in front of, the anus. The eighth abdominal segment generally seems to carry the genital orifice. In the Strepsiptera only is the female generative orifice placed some way forwards. The germinal glands are, as a rule, disposed in pairs, and retain this condition, although there are indications of the primitively single arrangement, or of a connection between the germ-glands of either side, as in the Araclinida and Myria- poda. Each germ-gland is composed of a varying number of equal parts, which are generally tubular in form, grouped into tufts, and united at an efferent duct. The ducts of the two germ- glands seldom have separate orifices. They are almost always united for a certain distance, and receive, before they unite, acces- sory organs, formed by the differentiation of a portion of their walls. In the females these organs appended to the ducts are sometimes pouch-like, or vesicular portions, which either serve for the reception of the male copulatory organ (bursa copulatrix), or as glandular organs of various kinds (cement glands), and also as a store-house for the sperm (receptaculum seminis). In the male, the paired-glands of the efferent ducts are greatly developed. In addition to them there are other parts which function as seminal vesicles (vesiculoa seminales). External organs, which are generally formed by the metamor- phosis of the terminal metameres and their appendages, are con- nected with the end of the genital duct ; in the males these form copulatory organs, and in the female vary in form (as ovipositors). § 234 In the female apparatus the complex of ovarian tubes, which is generally regarded as an " ovary," undergoes the most considerable modifications. The relations which these tubes have to the formation of ova is somewhat different to those which they have in other Arthropoda. Each separate ovarian tribe (Fig. 158) gradually widens at one end, where it is inserted into the oviduct; the opposite end is generally slender, and is often, indeed, continued into a fine filamentous process. When there are a large number of ovarian tubes, the free ends are directed towards a centre and connected together. The ova are GENERATIVE ORGANS OF ARTHROPOD A. 301 formed in these terminal filaments, the cell-masses of which represent ovarian germs; these, while undergoing continual differentiations, gradually make their way out of the ovarian tube. The ovum is a true cell at the place where it is formed, but on its way through the ovarian tube it in- creases in size, so that we find the largest po'O'g farthest from the germinal region, and nearest the oviduct, while behind them there is a continual series of smaller and younger formations up to the above-mentioned blind end of the ovarian tube. The separate eggs cause the ovarian tube to appear to be divided into segments or chambers. The gradual descent of the egg is not only correlated with its growth, but with various changes also in the substance of the yolk ; each egg is provided, especially in the last segment of the tube, with an external cuticular investment, the so-called chorion ; this is formed by the epithelial layer of the ovarian tube. As each egg passes into the so-called oviduct, a portion of the ovarian tube is degenerated, and so the egg next in front is brought close to the oviduct. The differentiation of the egg is accompanied by the growth of the thin end of the ovarian tube, which is made up for by its shortening at the other end. In many Insects a group of cells is differentiated with each egg, in addition to the epithelial layer sur- rounding it; this vitellogenous layer occupies the portion (b) of the chamber (a) behind the egg-cell (Fig. 158, B a), but is gradually used up by the latter, as it grows. An ovarian tube, or a collec- tion of such tubes, does not therefore correspond merely to a germ- producing reproductive gland, but is an organ entrusted with a much larger series of functions, and its blind end only is analogous to an ovary. The length of the ovarian tube depends on the number of eggn in it. The smallest number of chambers is found in most of the Fig. 158. A Ovarian tube of the Flea. o Ovum, fj Germinal vesicle. B Ovarian tube of a Beetle (Carabus violaceus). o Ovarian segment, divided into two por- tions, of which the ovarian cell is marked a, and the vitellogenous layer b. The ovum of the last segment has been ex- polled ; the walls of the ovarian tube are collapsed (after Lubbock). 302 COMPAEATIVE ANATOMY. Diptera, where not uufrequently there is only one (Fig. 160, o), though more commonly two or three. In many Coleoptera and Hemiptera also the number of chambers is small. The ovarian tubes are longer in most of the Hemiptera and Hymenoptera ; the largest number of chambers obtains in the Neuroptera and Orthop- tera, and lastly in the Lepidoptera, where there are four ovarian tubes made up of a large number of chambers, which look like a string of pearls. The arrangement of the ovarian tubes on the so-called oviduct also varies very greatly. They are sometimes united into tufts, sometimes broken up into groups, and sometimes arranged in rows. The so-called pseudova have been distinguished from the eggs (ova); these structures are partly characterised by the absence of the germinal spot, like the products of the female generative glands in certain generations of the Aphides and Cocciche. As the organs resemble those in which true egg-cells are formed, and as the same individual is able to produce pseudova and ova at different times, it is best not to regard the gulf between these two products of the ovary as a very wide one. These structures are links in a chain of phasnomena, which are very common among Insects ; the chain begins with the arrangement known as parthenogenesis, and extends to an apparent alternation of generations. The whole phenomenon depends on the emancipation of the ovum from the influence of the male reproductive elements. The simplest case is that in which there is no anatomical difference between the eggs, some of which are developed without previous fertilisation, while the rest require to be impregnated. The parthenogenesis of Bees, Wasps, and many other Insects is of this kind. The arrangement in which the same individual no longer produces these eggs at one and the same time is a further differentiation ; the emancipated ovarian products are then, as a rule, differently formed (pseudova). Still more peculiar is the formation of these eggs in different individuals, when whole generations can do without the influence of the semen on the reproductive elements, and at the same time fall to a lower grade of organisation (Aphides). These structures, finally, may be formed in an earlier stage in the development of the animal, and from the still indifferent germinal gland; this arrangement, just like the rest, with which it is directly allied, is derivable from sexual differen- tiation (Cecidomyia). § 235. The two, ordinarily short, oviducts seldom open separately into a depression of the integument (Ephemerida). As a rule this depression is further developed into a common efferent duct (Fig. 159, ov), the vagina; with this accessory organs, receptaculum seminis (Fig. 159, rs) and bursa copulatrix (be), are connected. The seminal receptacle is seldom absent; it is formed of a stalked and sometimes much-coiled vesicle. The receptacle is often a propor- GENERATIVE ORGAN'S OF ARTHROPOD A. 303 Fig. 159. Female generative organs of Hydrobius fuscipes. o Ovarian tubes, ov Oviduct, beset with glandular appendages, gl Tubular glands, v Vagina. be, Bursa copulatrix. rs Receptaculum seminis (after Stein). tionately wider and coiled cascal tube, which is sometimes provided with an appended gland. The bursa copulatrix is another organ, which is directly con- nected with the vagina; it is a wide cascal-sac (Fig. 159, be), which looks like a diverticu- lum of the wall of the ( \\\ ) vagina. This organ is found in some orders only, and even in them it is not generally pre- sent. The bursa copu- latrix of the Coleop- tera appears to be the most independent, and not unfrequently is of a considerable size ; in them it is generally connected to the va- gina by a canal. In the Lepidoptera also it opens into the va- gina by a narrow duct ; but it is remarkable from the fact that it has another efferent duct in addition to this one, which it sends off below the female generative pore, where it opens separately. In the Lepidoptera fertilisation is effected by means of this canal, the spermatozoa pass- ing into the receptaculum seminis from the bursa copulatrix by the above-mentioned duct, which connects it with the vagina. The open- ings of the two parts into the vagina are opposite to one another. The accessory glandular organs of the vagina either consist of a pair of simple canals which gene- rally form long loops (Fig. 160, gl) (Lepidoptera, many Diptera), or of short caecal tubes (Bugs). In others they are greatly ramified (Ichneumonidge and Tenthredinidas). The secretion of these cement- glands serve to attach the eggs when laid, and at times to unite them into masses. As a rule some portions of the integument, which have the form of valves, are connected with the female genital pore ; the markings on these valves are always exactly adapted to the male copulatory organ ; sometimes they are arranged like nippers, and consist of pi-ocesses which work laterally on one another. Fig. 160. Female generative organs of Mallophagus. o Ovarian tube, u Uterus, gl Glands (after R. Leuckart). 504 COMPARATIVE ANATOMY. § 236. The male generative organs of Insects very often repeat in their development the forms of the female organs, so that even the separate divisions of both sets of glands not unfrequently correspond. The testes, which are always paired, and seldom fused into one organ, are composed of cajcal tubes, just like the ovaries; they also vary in number and size, and are connected with one another in all kinds of ways (Figs. 161, 162, /). The testes of either side are often united in the Lepidoptcra. The Diptera and Strep- siptera, as well as many Neuroptera, have two simple, long-, and always separate testicular tubes. In many Coleoptera, also, each testis is a long, closely-coiled ca3cal tube, surrounded by a special membrane. The testes of most insects are made up of a number of tubes. Thus each testis, in most of the Hemiptera, consists Fig. 101. Testes and efferent ducts of Acheta campestris. t Testes. v Vns deferens. as sinuses, into the lamelhe of the mantle, and into the arms ; in the former they break up peri* pherally, and so come to be regularly arranged. The vascular apparatus ramifies in these spaces. The chief point as to their 312 COMPARATIVE ANATOMY. general arrangement is that the large trunks run dorsally along the euteron ; in this they may be seen to resemble the arrangements which are found in the Vermes. But the special points in this system of organs require further investigation. A saccular organ lying above the stomach is regarded as the heart ; this receives a vascular trunk which runs from in front above the oesophagus, and gives off lateral branches. The former is regarded as an afferent vessel (vein). It seems to collect the blood from lacuna?, which are placed around the enteric canal. Two lateral vessels given off from the heart are united in the Testicardines (Waldheimia) for a short distance. In the Ecardines (Lingula) they are not given off till later from a median longitudinal trunk which passes backwards on the enteron. Two arterial trunks, which have been called aortas, soon divide into two branches, one of which passes forwards and the other backwards. The anterior one represents the dorsal pallial artery, which divides into a median and a lateral branch, and supplies the mantle and the organs which lie in it. Smaller arteries are given off from the lateral branch to the lacunas of the mantle; they pass to the edge of it, and then open after having divided several times. The hinder branch of the aorta also divides into two arteries. One runs along the middle line, and forms, with its fellow of the other side, an arterial trunk which passes to the stalk. The other artery turns forwards, and again divides into two branches in the ventral lobes of the mantle where it ramifies, in just the same way as the dorsal pallial artery. On the two pairs of pallial arteries there is a pouch-like appendage, or accessory heart. The blood seems to pass from the ends of the arteries into wider lacunas which are placed in the mantle, as well as between the viscera, and around the muscles ; and these are con- nected with a complicated system of canals, which traverses the arms, and is divided into an efferent and an afferent portion. As the mantle is a secondary structure, its blood-vessels may be regarded as being so also. The pallial arteries are, therefore, of little importance, and the large trunks which accompany the enteron become of greater morphological significance. The heart appears to be a unilateral enlargement of the longitudinal trunk, as are also the accessory hearts of the pallial arteries. Excretory Organs. § 245. The excretory organs found in the Vermes, and adapted to the presence of a coelom, are also found in the Brachiopoda, where they have essentially the same characters. Like the looped canals of the Annelides, these organs have an internal and an external orifice, so that I have no hesitation in regarding these structures as homologous, EXCEETOKY OKGANS OF BKACHIOPODA. 313 evoii though their f Linction be modified. There are either one or two pairs of them. When there are four, two of them belong to tho so-called dorsal, and two to the ventral half (Rhynchonella) ; this points to tho presence of two metauieres, which have disappeared in this portion of the body. The dorsal ones are absent in Lingula and the Terebratulida. The canals, which generally open to tho exterior near the base of the arms, open into the ccolom, after having taken a bent course, by funnel-like enlargements (Fig. 166, r), which are distinguished by their radially-arranged folds. This orifice passes through the ileoparietal band, and is thus directed m il Fig. 166. Lateral view of the organisation of Waldheimia australis. D Dorsal, V Ventral surface. P Stalk. II Spirally-coiled arms, hr Branchial filaments, c An- terior wall of the perivisceral cavity, d CEsophagus. d' Mid-gut. h Liver, h ' Its openings into the mid-giit. r Internal orifice of the right oviduct (some folds only of the left oviduct can be seen), e Brachial canal, m m' m" m* Muscles to move the valves of the shell (after A. Hancock). towards the pericardial cavity. The ileoparietal band resembles therefore, in its relation to the internal orifice, a dissepiment of the Vermes (cf. supra, § 243). Although the walls of these canals appear to be glandular in character, owing to the possession of projections, villous processes, or folds, we know nothing of their function, save that they have a relation to the generative organs ; so that they appear to form an oviduct, and have indeed been hitherto regarded as being such. And as the looped-canals serve as parts of the generative apparatus in the Gephyrea and Annelides, it is not to be wondered at that they have the same relations in the Brachiopoda ; but this does not exclude the possibility of their having an excretory function also. 314 COMPARATIVE ANATOMY. Generative Organs. § 246. In some of the Brachiopoda the arrangement of the generative organs is hermaphrodite, so that the separation of the sexes seems to be an exception (Thecidium). The organs merely consist of germinal glands, in which the sperm and ova are formed. In the hermaphrodite form there are four, and in Thecidium two, glandular masses. In the Ecardines they lie in the coelom, partly surround- ing the enteron and the muscles ; in the Testicardines they form rounded masses in the cavity of the two lamella? of the mantle (con- tinuations of the ccelom) (Fig. 165, g); in either case they call to mind the character of the generative organs of the Gephyrea, and of the Annelides. In the dioecious forms they are ovaries in one, and testes in the other. It is not known what relation there is between the ovarian and seminal regions in the monoecious forms. The generative products escape into the ccelom. The excretory organs act as the efferent ducts of the generative glands, so that here too a primitively unconnected apparatus func- tions as an oviduct, or as a seminal duct, according to the sex. Seventh Section. Mollusca. General Review of the Group. § 247. The general characters of the body, and of its various systems of organs, distinctly define the phylum of the Mollusca. Owing to the absence of any distinctly-marked external metamerism, the body appears to be more compact than in the Arthropoda or Annulata among Vermes ; indications of metamerism may, however, be made out in various organs. The supracesophageal position of the central nervous system, and its connection with lower-lying ganglia, or with commissures surrounding the pharynx, when taken in con- nection with the position of the heart, which is always dorsal, are the definitely typical characters of this division ; to which we may add that in most forms shells are developed from the dorsal surface. The complete disappearance of their primitive metamerism, and the gaps that there are between the classes here united together, are completely explained by the early appearance (palasontologically speakiug) of most of the classes of the Mollusca ; while the forms still living are seen to be an exceedingly small part of the phylum, which was rich in forms, but which has been continued on in a relatively small number of divisions. As yet we know very little of the phylogeny of the Mollusca, but their metameric arrangements, as indicated by their internal organisation, point to then affinity to segmented organisms, the nearest allies of which were some of the Vermes. Although we can classify the various orders as higher and lower, yet all the systems of organs have not been developed to the same extent, so that we are able to find distinct proofs of the affinity between every single division and lowTer forms. 31 G COMPARATIVE ANATOMY. I submit the following sketch of the classification of the group, and would remark that many of the older views took note of varia- tions, which would still further separate the divisions, and especially the smaller ones, of the group. I. Placophora. . Chiton, Cryptocliiton. II. Conchifera.* Lamellibranchiata. Asiplioni'a. Ostrea, Anoinia, Pccten, Mytilus, Area, Anodonta, Unio. Siphoniata. Chama, Cardium, Cyclas, Venus, Tellina, Mactra, Sulen, rholas, Teredo. Scaphopoda.f Dentalium. Gastropoda.^ Prosobranchiata. Chiastoneura. Zeugobranchia. Fissurella, Haliotis. Anisobrancliia. Patella, Trochus, Littorina, Cyclostoma, Rissoa, Paludina, Turritclla. Orthoneura. Nerita, Janthina, Valvata, Sigaretus, Marsenia, Cypraea, Cerithium, Strombus, Pterocera, Dolium, Cassis, Tri- toniiun, Voluta, Harpa, Bueciuum, Nassa, Purpura, Murex. Heteropoda.§ Atlanta, Carinaria, Pterotracliea. Opisthobranchiata. Tectibranchiata. Bulla, Aplysia, Pleurobninclms. Nudibranchiata. Tritonia, Polycera, Aeolidia, Phyllirlioe, Doris, Phyllidia, Pleuropbylliilia. Sacoglossa. Elysia, Limapontia, Placobranclms. Pulmonata.|| * What led me most to unite all tho Mollusca, with the execptiou of the Chitouida1, into one great division, to which I have given the name Conchifera, was the considera- tion that we must recognise the great significance of the shell as affecting the whole organisation of these animals. But although the Placophora are thereby sharply marked off from the rest, I do not see that there is any sufficient reason for removing them altogether from the Molluscau phylum, for it is possible to make out in them many points in which they agree with, and are consequently allied to, the Conchifera. I regard the Placophora as the remnant of a division, the forms of which were allied to the Soleuogastres (p. 127) on the one hand, and on the other were the predecessors of the Conchifera. f The Scaphopoda form a division which is allied to the Lamellibranchiata as well as to the Gastropoda ; but they must not be regarded as a mere intermediate link. X The Zcugobranchia are, in many points, the oldest of the Gastropoda. § I regard the Heteropoda as an order which has branched off from the Proso- branchiata, and is closely allied to the Orthoneura; but which has developed special characters which are not equal in value to those of the Orthoneura. II The organisation of the two divisions of the Pulmonata does not seem to me to be so markedly divergent as to make them of equal value with the two other orders of the Gastropoda. We cannot as yet form a definite opinion as to many of the genera, e.g. Onchidium, of the Nephropneusta. mollusca. 317 Gastropoda {continued). Branchiopneusta. Lymnteus, Planorbis, Auricula. Nephropneusta. Helix, Bulimus, Clausilia, Litnax, Arion. Pteropoda.* Thecosoruata. Hyalea, Cleodora, Chreseis, Cymbulia. Gymnosomata. Clio, Pncumodermon. Cephalopoda.f Tetrabranchiata. Nautilus. Dibranchiata. Decapoda. Spirilla, Sepia, Sepiola, Loligo. Octopoda. Octopus, Tremoctnpus, Eledone, Argonauta. Bibliography. CtrriKTi, M^moires pour servir a l'liistoire et a l'anatomie des Mollnsques. Paris, 1817.— Van Beneden, Uxerciccs zootomiques. Fasc. I. II. Bruxelles, 1839. — Qooy and Gaimaed, Voyage de I' Astrolabe. Zoologie.— Delle Chiaje, Descrizione e notomia degli animali invertebrati deila Sicilia citeriore. Napoli, 18-11— 14.— Soulkyet, Voyage de laBonite. Zoolog. T. II. Paris, 1852.— Leuckart, R. , Zoolog. Untersueh. III. Giessen, 1851. — Gegenbaur, 0., U liters, lib. Pteropoden u. Heteropoden. Leipzig, 1855.— Krohn, A., Beitr. z. Entw. d. Pteropoden u. Heteropoden. Leipzig, 1860. — Lankester, Ray, Contrib. to the develop, of the Mollusca. Philosoph. Transac. 1875. — V. Jhehing, Vergleichende Anatomie des Nervensystems u. Phylogenie der Mollusken. Leipzig, 1877. Placophora : V. Middendokff, Anat. v. Cliiton. Mt5m. Acad, de St. Petersbourg. VI. vi. 1819.— Loven, S., Ofvers. K. Vet. Acad. Forhand. Stockholm, 1855. (Arch. f. Nat. 185G.) Lamellibranchiata : Poli, Testacea utriusque Sicilire eorumque liistoria et anatome. III. Tom 1791-1795. — Bojanus, Ueber die A them- urn I Kreislaufwerkzeuge der zweischaligen Muscheln. Isis, 1819, 1820, 1827.— Deshayes, Art. Uonchifera in Todd's Encyclopaedia, Vol. 1. 1836.— Garner, On the anatomy of the lamellibranchiate Uonchifera. Transact. Zoolog. Soc. London. Vol. II. 1811. — Quatrefages, Anatomie von Teredo. Ann. des sc. nat. III. xi. — Loven, S., Bidrag till kiinnedomeii om utvecklingen of Moll, acephala. Kongl. Vetensk. Acad. Handl. Stockholm, 1850.— Keber, Beitriige zur Anatomie und Physiologie der Weichthiere. 1851.— Davaine, C, Sur la generat. des Huitres. Paris, 1853. — V. Hessling, Die Perlmuscheln. Leii>zig, 1859.— Lacaze- Dutiiiers, Anatomie von Anomia. Ann. sc. nat. IV. n. — L. Vaillant, Sur la lam. de Tridac- uides. Ann. sc. nat. V. iv. — Sabatiek, A., Etudes sur la moule commune. M6m. de TAcad. des Sc. de Montpellier. 1877. Scaphopoda: Lacaze-Duthiebs, Hist. nat. organis. dovcloppement, etc. du Dentale. Paris, 1858. Gastropoda: Nordmann, Monographic des Tergipes Edwardsii. M6m.de l'Acad. Imp^riale de St. Petersbourg. IV. 1813. — Alder and Hancock, Monograph of the British Nudibranchiate Mollusca. Ray Soc. I. — VII. 1815-55. — Hancock and Embleton, On the Anatomy of Eolis. Ann. and Mag. of Nat. His. XV. 1815.— The same, On the Anatomy of Doris. Phil. Trans. 1852. Pt. II. — Hancock, Anatomy of Doridopsis. Trans. Linn. Soc. XXV. — Leydig, Ueber Paludina vivipara. Zeitschr. f. wiss. Zool. II. — Huxley, On the morphology of cephalous Mollusca. Phil. Trans. 1853.— Bergh, Bidrag til en Monograph! of Marseniaderne. Kongl. dansk. Vidensk. Selsk. Skrifter. 1853.— The same, Anatomisk Undersogelse of Fiona atlantica. Vidensk. Meddelelser for 1857. — The same, Anatomisk Bidrag til Kundskab om Aeolidieme. Danske Videnskab. Selskabs. Skrifter. 1861.— The same, Bidrag til en Monographi of Pleurophyllidieme. Naturhist. Tidsskrift. 3 Ritkke. 1 Bind. 1866.— The same, Bidrag til en Monographi of Phylli- dierne. ebend. 5 Bind. 1869.— The same, Malacolog. Untersuch. Heft. I.— X. Wiesbad. 1870-76. — Ulaparede, Anatomie und Entwickelungsgesch. der Neritina fluviatilis. Arch. f. Anat. 1857. — The same, Beitruge zur Anat. des Cyclostoma elegans. ibid. 1858.— Lacaze-Duthiers, Anatomie du Pleurohranche. Ann. nat. sc. IV. xi.— The same, Anat. et l'Embryogenie des Vermets. Ann. sc. nat. IV. xm. — Lawson, Anat. etc. of Limax maximus. Quart. Journal of Micr. Sc. 1803. — Fol, H., Sur le d(5veloppement des Hetft-opodes et des GasttSropodes pulmonics. Comptes- rendus. T. LXXXI. Nos. 11 et 13. Archives de zoologie. T. V. * In many points of their organisation the Pteropoda indicate a relationship to tho Cephalopoda, but this can only be regarded as a very distant one. f Most of the oldest fossil forms probably belonged to the Tetrabranchiata, which are represented by only one extant genus ; afc the same time these fossil forms varied greatly in character. 318 COMPARATIVE ANATOMY. Pteropoda i Esciiricht, Ueber d. Clione borcalis. Kopenhagen, 1838.— Fol, H., Sur le tle"voloppe- nient cles Pteropodes. Archives tie zoologie. T. IV. Cephalopoda : Grant, On Loligopsis. Trans. Zool. Soc. 1835. — Owew, Memoir on the Pearly Nautilus. London, 1832.— The same, Article on Cephalopoda in Todd's Cyclopaedia. I. 1836. — Vxlencibnnes, Nouvelles recherches snr le Nautile flambe\ Archives du Museum. 1841.— Peteks, Anatomie der Sepiola. Arch. f. Anat. 1842. — Kolliker, Entwickelungsgesch. der Cephalopoden. Zurich, 1814. — Van der Hoeven, BijcLragen tot de Ontleedkundige Kennis aan- gaande Nautilus pompilius. Amsterdam, 1856.— Grenacheh, Zur Entwick. d. Cephalopod. Zeitschrift f. wiss. Zoolog. Bd. XXIV. p. 419.— Fol, H., Note s. 1. tlt?veloppement des Mollusqnes pteropodes et cephalopodes. Arch, de zool. T. III.— Bobretsky, Untersuch. iiber die Entvrick- elung der Cephalopoden. Nacbr. d. k. Gesellsch. der Freunde d. Naturkenntniss etc. zu Moskau. Bd. XXIV. (Russian.) [Rat Lankestee, Development of Cephalopoda, Quart. Journ, Mic. Sci. 1875.] Form of the Body. § 248. The general form of the body of the Mollusca must be regarded as one so much altered by the relative positions of many organs, owing to the formation of shells, that it has only been possible to recognise a ground-form, which shall represent the common origin, by comparing the earlier larval stages with several mature forms. The Placophora have a worm-like larva, and a similar kind of external metamerism is indicated by the number of circlets of cilia seen in the Gynmosomatous Pteropoda. The relations thus implied are retained by the Placophora in their mature condition, at least in the dorsal portion of the body. This is separated from the ventral portion by a groove, and so defines two regions, which are found also, under the form of " mantle" and "foot," although much changed, in the Conchifera. The differentiation of a gutter-like ventral surface in the Solenogastres (cf. p. 130), as has been already explained, points to the Mollusca having genetic relations to these worms ; this supposition is confirmed by the characters of the nervous system. The Lamellibranchiata and Gastropoda, as well as the thecosomn- tous Pteropoda, develop a well-marked circlet of cilia in the region, which, later on, corresponds to the head ; this circlet is afterwards carried on a special, symmetrical, and lobed process — the Velum. The primitive significance of this circlet is clearly shown by its presence in otherwise divergent divisions, and is even still more important from the fact that we can recognise in this organ the circlet of cilia which surrounds the same part of the body in many Vermes (cf. § 107). The velum of the Mollusca may therefore be regarded as an organ inherited from a lower stage. Below the velum the rudiment of the opening into the enteric cavity is formed. As in the Placophora, the formation of a dorsal shell in the Lamellibranchiata does not prevent the enteric tube from being continued to the aboral pole of the body ; for in the Placophora the shell, as well as the mantle which carries it, is adapted to the whole body, and in the latter it is principally developed at the sides. We are able, therefore, to distinguish a primary axis, which extends from the oral to the anal pole, and which is crossed by two FORM OF BODY OF MOLLUSCA. 319 secondary and variously-differentiated axes — the dorso- ventral and the transverse. The body is therefore of the original eudipleural form, which is the dominant one in the Vermes and Arthropoda. These relations are different in the Gastropoda, where the dorsal cup-like shell gradually encloses the greater part of the body, and leaves a small portion only of the surface of the body exposed in addition to the head and foot. So that while in the previous case the shell was adapted to the body, in this case the soft parts of the body are adapted to the single shell. The body, therefore, becomes asymmetrical, and the aboral pole no longer carries the anus, which becomes lateral in position in consequence of the flexure of the enteron ; this flexure is due to the formation of the shell. All of the many variations from the symmetrical ground-form, which are seen in the Gastropod-body, may be regarded as due to this. The primitive similarity in the form of the body, due to the possession of a shell, undergoes great modifications even among the Gastropoda; the Veliger stage is not always developed, and has never yet been observed in the Cephalopoda. But even in this class the form of the body, and the disposition of its viscera, may be seen to be, in all forms, due to the possession of a shell. § 249. The velum has different functions in different divisions. In the Lamellibranchiata, where it functions for some time as a locomotor organ, but where it is never independent and soon atrophied, its function is not very great. This may, perhaps, be correlated with the rudimentary character of the future head, and this, again, with the rapid disappearance of the free mode of life in this division (Acephala). Two folds, however, which are given off laterally from the dorsal surface, become considerably developed and form a mantle; they surround the body, and excrete the shell, which corresponds A , B with the lamellas of the mantle in form and size. A space, which functions as the respiratory cavity, is developed between the edges of the mantle ; branchiae are developed from the body- wall, and project into it (Fig. 167, A br). In a few Lamelli- branchs (Asiphonia) this en- trance into the mantle- cavity is a cleft of some size, by which water passes in and out, and so carries in nutriment and removes excreta. In most, the two edges of the mantle grow together, and so shut off, more or less completely, the cavity which surrounds the gills, and cause Diagram of the relations of the foot and mantle, as seen in transverse sec- tion. A In Lamellibranchiata. B In Cepha- lophora. m Mantle, p Foot, hr Branchiae. 320 COMPARATIVE ANATOMY, the two streams of water to enter and escape from it with greater regularity. The least amount of concrescence which is observed, gives rise to an anterior larger, and a posterior smaller, orifice (Mytilidse) . The former serves as an outlet for the foot, and as the orifice of entrance for the food, while the latter, in correspondence with its position, is the orifice of exit for the f cecal matters, and of the water that has served for respi- ration. In the Chamacea) there are also two large openings behind the anterior and larger cleft, through which the foot is protruded, and which serve respectively for the entrance and exit of water ; this is an arrangement which attains a higher grade of development in a large division of the Lamellibranchiata (Siphoniata). That part of the mantle which surrounds these orifices forms an elongated tube (siphon), which undergoes other modifications in addition to its con- crescence. The respiratory tubes may sometimes be formed by Fi°\ 168. Lateral view of the mantle-cavity of a Mactra; the right mantle-lamella has been removed. br br' Branchia?. t Tentacle, ta tr Siphons. mp Posterior adductor, p Foot, c Umbo. 5)i a Anterior, separate portions of the mantle ; or there may be a respiratory tube, which is siuo-lo externally, and is only divided internally into two canals by a partition ; or the two conditions may be combined (Fio\ 1G8, tr ta) ; or, finally, two completely separate tubes may be developed : an upper one, the inner orifice of which is opposite the anus, and serves for the exit of the water, and a lower one, by which the water passes in. The investment of cilia causes the two streams to pass regularly in and out. Through these forms we arc led up to those in which the respiratory cavity is most completely closed, and the pallial tubes most developed. This is accom- panied by a diminution in the size of the cleft in the mantle through which the foot is protruded. This has become much narrower, and is placed some way from the respiratory tubes, so that the greater part of the edges of the mantle have grown together, in consequence of which the body of the animal is sac-like (Boring Mollusca). The orifice of passage for the foot is now placed at the anterior end, and tli e / P The same animal with its foot and siphons re- tracted, ins Siphonal muscle. Fig. 160. FOKM OF BODY OF MOLLUSCA. 321 two respiratory tubes in the opposite region of the body. They are continued into special divisions of the mantle-cavity, owing to the division of the latter into an upper smaller, and a lower larger cavity, which are separated by a partition. The water is brought to the lower one by the efferent tube, and passes through the gills ; streaming through the orifices in them, into the branchial plates or the intrabranchial cavity, and so into the upper division of the mantle-cavity, into which the anus also opens. The edge of the mantle is often the seat of special differentia- tions, which are generally of the form of tentacular processes, and are sometimes of a fair size. The second differentiation in the body of the Lamellibranchiata affects the ventral surface, which is differentiated even in the Placophora ; it becomes flattened, and serves as the organ by which the animal creeps along. It is formed by the development of a muscular foot, which is more or less separated from the rest of the body (Fig. 167, Ap), and which can be protruded from the cleft in the mantle, sometimes to a considerable extent. It is then hatchet or club-shaped, and functions as a locomotor organ. The two lateral surfaces of the foot are ordinarily produced into a median edge, but in some it is flat and sole-shaped, as in Chitons. Many Lamellibranchs live under conditions in which this organ is not required, and it is then atrophied, as in the fixed Oysters and Anomise, and Scallops ; in the latter, locomotion is effected by the action of the mantle and its shells. The Scaphopoda are forms allied to the Lamellibranchiata, but intermediate between them and the Gastropoda. The body, which is enclosed by a shell, is provided with a mantle-cavity, from which a trifid foot can be protruded. A part which carries the mouth is head- like in form, but is really more of a proboscis, for it does not contain the nerve-centres, and is, moreover, enclosed in the mantle-cavity. § 250. The velum is largest in the Gastropoda and the thecosomatous Pteropoda, and is absent in those only in which the earliest larval stages are not free (Land Snails). It has the form of a large, and frequently symmetrically lobate, organ (Fig. 170, A B G v), which in some is retained for a longer time, and so enables the body to continue swimming about (Macgillivraya). The development of this organ, which in its lower stages is merely represented by a circlet of cilia, appears to be correlated with the development of a shell, for when this is developed the cilia are less widely distributed. The cephalic portion of the body is alone free; and it compensates for the absence of other locomotor organs by the great development of its cilia, and of its ciliated margin. The velum increases in size, and undergoes great complications of form, in proportion to the increase in the weight of the body due to its shell. 322 COMPAKATIVE ANATOMY. The size of this velum is correlated with the differentiation of the head, from the upper surface of which it is developed ; it is in some Pteropoda only that the head, once formed, undergoes any considerable atrophy. Just as in the Lainellibranchiata, the mantle rises up in the form of a fold of the body- wall, which covers over the dorsal surface and forms the shell on its outer side. As this dorsal area of the body — which is surrounded by the mantle-fold, and the shell, which is being developed into its house — continues to bulge out, it gradually forms a blind sac, which soon contains the greater part of the viscera (visceral sac) ; in this way the viscera come under the direct protection of the shell. As development goes on, the mantle- fold becomes less intimately connected with the body, and gives rise, inferiorly, to a wider cavity, in which the growing gills are contained, aud which is homologous with the branchial cavity of the Lamellibranchiata (Fig. 167, A B). This development of a fold of the integument into the mantle, and the consequent appearance of a Fig. 170. Larvse: A Of a Gastropod; B Later stage. C Of a Pteropod (Cymbulia). v Velum, c Shell, p Foot, op Operculum, t Tentacles. subjacent space, the branchial cavity — which looks like an invagi- nation from the exterior — undergoes modifications, which are largely due to the formation of the shell. In consequence of the mantle growing unequally on either side, and not equally, as in the Lamelli- branchiata, and from the fact that it is principally developed at one point in connection with the development of the shell, the branchial cavity comes to be a single cavity, placed in the same region. This region is either beneath the hinder portion of the mantle, as in the Pteropoda (Fig. 1 70, C), or beneath the anterior portion, as in most of the Gastropoda (B). The want of symmetry, which is due to the coiling of the shell, causes the branchial cavity of most Gastropoda to lie on one side ; this is an adaptation to the larger amount of space which is afforded by the lateral portion of the shell. The production of the unilateral and asymmetrical branchial cavity from a paired and symmetrical space is proved by numerous facts ; so that we are led to think that the asymmetry of the shell is probably a secondary arrangement. A number of degenerate and more perfect arrangements have FOEM OF BODY OF MOLLUSCA. 323 their common origin in this disposition of parts. The latter are principally seen as differentiations of the edge of the mantle, and are connected with the function of the branchial cavity. Part of the edge of the mantle is produced into a groove, which serves to bring in water, and which may be converted into a tube by the folding over of its two edges. We meet with a siphon of this kind, though in all stages of gradual differentiation, in a large number of aquatic Gastropoda (Buccinum, Dolium, Harpa, Tritonium, Murex, etc.). A second siphon, formed in the same way, but smaller in size, is generally found at the opposite end of the branchial cavity ; it serves to carry the water out from it. Various other kinds of processes, or tentacular appendages, lead to fresh complications in its structure (e.g. in Strombus, Pterocera). When the shell undergoes atrophy the mantle generally does so too. This is mostly the case in the division of the Opisthobranchiata, some of which have a more or less rudimentary shell ; while in others, when adult, there is no shell at all. As all these forms had a shelled larval stage, the atrophy of the shell must have been brought about during their ontogenetic development ; and it follows that those Opisthobranchiata, which are naked in their later stages, were derived from forms that had shells. The larval shell and its accompanying mantle-fold, even though feebly developed, are therefore rudimentary organs, which prove that the naked Opisthobranchiata had the same origin as the rest of the Gastropoda. Where these rudimentary shells are retained by the adult animal they must even then be regarded as degenerate parts, and not as developing shells ; for here again a comparison with the larval forms shows that the shell had a much greater significance than have the rudimentary structures found in the adult stage of these organisms. It is of great importance also as explaining the position of the anus and of the genital orifice, which can be due to nothing but a former greater development of the shell. Within smaller divisions also we meet with series of degene- rating parts, as for example in the Heteropoda, where Atlanta has a well-developed shell and mantle, while in Oarinaria they are both rudimentary, and in Pterotrachea completely lost. A similar series is observable in the Nephropneusta. § 251. The varying extent to which the Foot is developed is of im- portance as affecting the form of the body. In the larvee of the Pteropoda and Gastropoda it has always very much the form of a short, conical, somewhat-flattened process, placed below the mouth (Fig. 170, A p). On the hinder, or dorsal surface, a shelly secretion is formed, which serves as an operculum for the orifice of the shell. Owing to its increase in size, especially in the aboral direction, the foot of' most GastrojDocla comes to have a broad lower face, which is y 2 324 COMPARATIVE ANATOMY. Fig. the 171. Diagram of the relations of mantle and foot ; vertical section. A In Lamellibranchiata. J? In Cephalo- pora. m Mantle, p Foot, br Branchiae. the reason of its being called a foot (Fig. 171, B). Sometimes, however, it is elongated, and, at others, discoid in form. In most of the Gastropoda the foot is only sharply marked off along its plantar margin. In many of the lower Prosobranchiata (Haliotis) the surface of the body above the foot is drawn out into an encircling edge (epipodium), which is distinguished from the mantle by surrounding the head. The foot of the Heteropoda is 'differentiated into a more independent organ, which springs from the ventral surface of the animal, and forms a vertical fin. The body is continued in front of, as well as behind, the foot. This arrangement is very different from the primitive one; the body has no longer a flat surface, although the end of the foot in Atlanta still carries an operculum. The structure of the muscular sole of the Gastropod foot is retained in rudiment as a sucker-like organ, which in the Pterotracheae is found in the males only. And we are reminded by this that even when fully developed the foot of the Gastropod functions as a sucker, for the animal is able to attach itself by it. The modifications undergone by the foot in the Pteropoda are still more significant. The foot, which, in the ear- liest larval stages, is formed in just the same way as in the Gastropoda, gives rise, in the Cymbulidas and Hyaleida3, to a median and two lateral pieces (cf. Fig. 170, C pp). In the Hyaleidaa the median portion is feebly developed, while the lateral lobes become large fins, which embrace the rudimentary head, just like wings ; in the Cymbulida) the median piece is also well developed. It either fuses at its base only (Cymbulia), or along its whole length (Tiedemannia), with the two lateral lobes ; in this way the large fins of these animals are produced. Fig. 172. Diagram of the rcla. tions of the mantle. A In Pteropoda. B In Cephalo- poda. p Foot. In- Branchire. t Tentacles. § 252. The greater development of the head in the Cejohalopoda is an important peculiarity as affecting the form of the body, while the mantle acquires the same relations as it has in the Thecosomatous Pteropoda, from which therefore they may be derived. The cavity, APPENDAGES OE MOLLUSCA. 325 arched over by a fold of the mantle, occupies the hinder part of the back, and so forms that region which is ordinarily known as the ventral surface. To make these relations clear we must imagine the animal placed in such a position that the aboral end points up- wards, and the head forwards and downwards (cf. Fig. 172). All the body above the head would then correspond to the dorsum of the Gastropoda. The mantle is sometimes separated from the head by a circular groove (Sepia) ; sometimes this fold of the mantle is directly continuous with the integument of the head at the sides of the neck (Octopus), so that the mantle forms a fold above the branchial cavity only. Lateral processes of this mantle function as locomotor organs (fins) ; in the Sepiada) they are generally small, and extended along its whole length; in the Loliginidse they are broader, but are limited to the aboral end of the body. The formation of the mantle-cavity and the position of the anus lead us to the conclusion that this arrangement is due to the primitive possession of a shell which covered the whole mantle ; and, indeed, the shelled Cephalopoda are by far the older forms, while the remarkable variations seen in the characters of their shells lead us to think that this structure had a very ancient origin. An organ which has the same position as the foot of the Gymnosomatous Pteropoda — the funnel — corresponds to the foot of the Gastropoda. In Nautilus it is formed of two lamellae, which arise from the ventral surface below the head, and which form a tube by being rolled over one another ; this tube projects from the mantle-cavity (Fig. 175, i). In the Dibranchiata this organ cannot be seen to be composed of two lateral parts, except in the embryo; they take their origin in the space between the mantle and the rudiments of the arms. By growing together and gradually fusing they form a tube which is similar to the one formed in Nautilus, except that it is closed. The mantle, which is also muscular, is attached to the periphery of the funnel; this effects powerful con- tractions, and so drives out the water which has entered the mantle-cavity between the funnel and the edge of the mantle ; and the animal is driven in an aboral direction by the force produced by the expulsion of this stream. The organ retains, therefore, its primitive locomotor function. Appendages. § 253. In the Mollusca the development of a cephalic region is closely connected with the differentiation of processes, which I regard as appendages, inasmuch as they are homologous with the antennas and tentacles of Arthropoda and Vermes, and when more highly differentiated are able to undertake the duties of appendages. 326 COMPARATIVE ANATOMY. These structures, which, are known as tentacles, are absent in the Placophora and in the Scaphopoda ; the processes which are found around the mouth in the latter group being structures of a special kind, and not appendages as here limited. In the Laruellibranchiata, lobate appendages (Fig. 168, t) (the so- called labial palps) are attached to the altogether rudimentary head ; they may be homologous with the more highly-developed tentacles, which distinguish the cephalic region of the Gastropoda. As in many Platyhelminthes they are, when simplest, short processes of the body, but they undergo great differentiations. In the Proso- branchiata they are generally limited to two, and are formed from the surface which is surrounded by the velum (cf. Fig. 170, B t). In many forms the eye is placed at the base of the tentacle, which may be developed into a special process. The same happens in other forms, where the optic organ is placed on an optic stalk dif- ferentiated from the tentacles, and which, when more independent, may give rise to four tentacles, as in Helix, Limax, etc. These are invaginated when they are retracted, and are so far more highly developed. Many Opisthobranchiata are distinguished by the pos- session of a pair of tentacles, which are greatly developed (Fig. 177, tt), but in addition to them there are other tentacular cephalic appendages, which characterise the various subdivisions merely, according to the way in which they are arranged, and according to the number present. They have undergone degeneration in the Thecosomatous Ptero- poda, for in these forms the tentacles are either completely absent or are rudimentary (Chreseis). The development of the parts of the foot which in them are converted into fins, does away with the necessity for the development of the cephalic tentacles, and explains why they are absent, just as, on the other hand, the distance of the fins from the head in the Gymnosomatous forms is the cause of the development of their tentacles. In these latter they have all kinds of forms, and one or more pairs of processes (Cephaloconi) are present in addition to the superior tentacles ; these lead up to the tentacles of the Cephalopoda. In Pneumodermon, indeed, two of these bodies are beset with suckers. § 254 In the Cephalopoda the large number of tentacles, which are arranged in rows on either side, and spring from lobate processes, distinguish the head of the Tetrabranchiata. In the Dibranchiata, where they form arms, there is a smaller number of them, but they are larger. The Loliginidaa, Sepiadas, and Spirulidas have ten arms. Two of them, which are longer, and in other points different from the rest, are placed outside the circle around the mouth, which is formed by the other eight ; and as they spring from pouches which are arranged in pits at the side of the head, they must be distinguished from the inner series ; so that these inner ones are always eight in number in APPENDAGES OF MOLLUSCA. 3^7 all Dibranchiata. The arms of the Octopoda, like the similar ones in the Decapoda, are connected together at their bases by a web, ex- cepting the pair which are nearest to the sides of the funnel. This connecting membrane extends farther in some Octopoda ; sometimes over a few of the arms only (four in Tremoctopus), or over them all (Histioteuthis, more completely in Cirroteuthis), and is continued right up to the tips of the arms. The suckers are special structures found on the arms of the Cephalopoda; they generally beset the oral surface of the arms in two rows (one in Eledone), and not unfrequently they are carried on stalks. Their free edge has often a cuticular thickening which has the form of a chitinous rinsr, and is sometimes toothed. Where a particular tooth is largely developed the sucker disappears, and is replaced by a hook (Onychoteuthis). In many Cephalopoda certain of these arms are peculiarly altered by function- ing as copulatory organs ; even in Nautilus the tentacles perform this func- tion. It is not always the same arm that is thus metamorphosed ; as a rule it is one of those which belong to the so-called ventral side of the animal. The mode of metamorphosis varies greatly in the different divisions ; it may merely consist in the alteration of a part of the base of the arm (Sepia), or a more or less large number of the suckers may be altered, or the tip of the affected arm may be provided with a spoon-like hollow process (Octopus, Eledone). The highest grade of this adaptive metamorphosis is seen when the arm becomes greatly increased iu size, as well as different in organisation inter- nally (Argonauta, Tremoctopus). This " hectocotylised arm " is not developed, as are the others, by a process of free gemmation, but it is formed in a vesicle, from which it is not set loose till it is mature. The greatly-coiled flagelliform end of the arm (Fig. 173, y), which is not set free till the time of copulation, has a similar covering. This appendage, with its investing membrane (x), corresponds to the modified end of the arm in Eledone and Octopus. The more highly differen- tiated copulatory arms may continue to live within the mantle- cavity of the female for some time after they are broken off; this Fig. 173. Mule of Tremoc- topus carenas. tl Superior; t~ Second pair of arms, t3 Third left arm. i4 Inferior pair of arms, h Hectocotylus. x Its terminal vesicle, y Filamen- tous appondage, set free from the terminal vesicle, i Funnel. 328 COMPARATIVE ANATOMY. is the reason why these separate arms were formerly regarded as parasitic organisms (Hectocotylus). Steexstrup, J. J., Hectocotyldarmelsen. Kongl. Danak. Vid. Selsk. Skrifter. V. E. 4Bd. Integument. § 255. The body of the Mollusca is covered by a soft dermal layer, which is, as a rule, so closely interwoven with the subjacent muscles as to form a kind of dermo-muscular tube, just as in the Vermes. The locomotor organs are formed by the great development of the musculature in certain regions of the body, and by the consequent differentiation of some parts of the dermo-muscular tube. Inmost divisions of the Mollusca there is an investment of cilia during the larval stages, which later on extends over the whole, or some parts of the body. The cilia in the velar circlet (§ 248) are those that are most markedly developed. The rest are chiefly found in the respiratory organs. In the Cephalopoda even, nearly the whole surface of the germinal disc (save the gills) is covered with cilia during development; later on a ciliated epithelium may be found on the yolk-sac also. The integument is easily distinguishable into an epidermis and cutis. In many Heteropoda (Carinaria, Pterotrachea) the latter is specially modified ; a strong transparent layer of connective tissue preventing any great amount of change in the form of the body. In the rest of the Mollusca the body is generally prevented by the shell, which is developed from the integument, from undergoing any great changes in form. The coloration of the body is due to the deposits of pigment in the integument. The most remarkable structures concerned in coloration are those which are found in many Pteropoda, and in all Cephalopoda — the " chromatophores." They are rounded cells, which arc placed at various depths in the integument; they are filled with granular pigment, and are provided with radial muscular fibres at their periphery ; when these fibres contract the cell broadens out, and the pigmented contents are thereby distributed in such a way that they become easily visible to the eye, as large, stellate, and often branched spots. Plate-like elements are found deposited in a layer, which is sometimes differentiated, and these give a silvery appearance to many parts of the body (spangled layer). The varying character of these several layers produces that play of colour which we admire in the skin of the living Cuttlefish. Other deposits are found in the integument, such as those formed of carbonate of calcium, which are common in the Gastropoda; these are either simple granules, or larger rounded concretions, or they may be rod-shaped, denticulate, or even branched ; there is often a INTEGUMENT OF MOLLUSCA. 329 large number of them, so that they may form a veritable calcareous network, as in Doris, Polycera, etc., the various species of which are distinguished by the special manner in which the various calcareous rods are grouped or arranged, as well as by the way in which they are formed. § 256. The glands are differentiations of the epidermis; they partly resemble the structures found in Vermes (unicellular glands) . When simplest, these organs are modifications of epidermal cells, which are placed between other cells, but are distinguished from them by having finely-granulated contents and a mouth (goblet-cells). They are found in the Lamellibranchiata as well as in the Gastropoda. In the Cephalopoda they are more commonly arranged in groups, and their blind ends extend below the level of the epidermis. In the Gas- tropoda— and especially in the terrestrial Pulmonata — they are found to be placed still deeper in the integument. These glands are variously modified in different parts of the body. Those found at the edge of the mantle in the shelled Gastropoda are examples; they secrete a fluid in which calcareous salts are dissolved, while others secrete colouring matters. In Aplysia the dermal glands secrete a dark-red fluid. In Murex and Purpura a layer of epithelium, which is placed between the gills and the hind-gut, and in the mantle-cavity, functions as a gland; this layer is formed by large superficially-ciliated cells. Their secretion gives rise to the substance known as Tyriau purple. Some Opisthobranchiata (^Eolidias) are characterised by the possession of urticating cells, which are placed on the ends of their dorsal papillas. The By ss us -gland of the Lamellibranchiata is a more inde- pendent glandular organ of the integument ; when it is formed the foot undergoes certain modifications ; it becomes, that is, a tongue- shaped process grooved o** its ventral surface. The groove passes to a depression at the base of the foot, at the bottom of which there is a gland which secretes the so-called " byssus." Pccten, Lima, Area, Tridacna, Malleus, Avicula, Mytilus have an organ of this kind; but it may be considered as generally present, for it is found for a time in the embryos of the Naiades and of Cyclas. Some divisions of the Gastropoda (Helicinas, Limacinao) have also a gland in their foot, which opens anteriorly and below the mouth. A large number of other kinds of glandular organs are also developed from the integument. Shells. § 257. The tegumentary investment is of special importance, for it secretes firm substances, which are laid down in layers, and which 330 COMPARATIVE ANATOMY. produce the varied forms of test and shell which characterise the Molluscan phylum. The hard structures, therefore, of this division differ essentially from those which are found in other classes of animals, by the way in which they are developed. They are products secreted from the body, and deposited on its exterior, and are of great importance as organs for the support and defence of the organisms to which they belong. They, just as much as other differentiations from the integument, imply that the outer dermal layer has a secretory activity. Although the outer layers of these structures often appear to be — as is especially the case in large shells — distinct from the organism, the shells always do form a part of it, and are at many points directly and closely connected with it; as, for example, at the insertion of the muscles into the shells. The presence of calcined spicules in the integument of the Haeophora calls to mind the relations which obtain in the Soleno- gastres (p. 139). The spicules arise in follicles, and do not reach the surface till they are larger in size, when they form slender and closely-approximated fine processes, or thicker bodies, which are distributed over the mantle. Eight large calcified plates are also arranged transversely on it, and form a series of skeletal parts, the arrangement of which implies that the body is arranged in a meta- meric fashion. As in Cryptochiton they are covered by the mantle, there is some reason for supposing that they were developed within it, and that they resemble the spicules. The plates would then be structures of the same kind, which had been greatly developed, while the spicules would be parts which had not enlarged laterally, but only vertically. This relation between the presence of a mantle and the formation of firm organs, which, when largely developed, form shells, is typical of all the other Mollusca ; and the two kinds of organs are always intimately connected. Instead, however, of the dorsal plates being developed, as in Chiton, the formation is con- tinuous, so that it gives rise to a single shell. The shell, therefore, just as much as the mantle — which we have seen to be homologous throughout the series of Mollusca — must be regarded as an organ, which is widely-distributed because inherited, great as may be the adaptive modifications which it has undergone. The multifid shell was not replaced by the undivided one by any new process, but by the development of one part, for we cannot imagine that the shell, which, as an organ investing a large part of the body, is one of so great functional importance, appeared all at once. But, if the shell was at first an inconsiderable organ, it could not have attained to that perfection of function which would have been the cause of its being transmitted as a useful arrangement. We must therefore suppose that the structure, which later on formed the shell, was primitively one of several similar organs, and that it gradually got the better of the rest. This gradual development of the shell is the only mode which is intelligible, while at the same time it connects together the multifid shell of the Placophora and the single shell of the Conch if era. THE SHELL OE MOLLUSCA. 331 Fig. 171. ropod, o Month. cavity Embryo of a Hete- trausverse section. v Velum, g Enteric p Foot, s Shell-gland (after H. Fol). § 258. The earliest rudiment of the shell appears at the aboral end of the embryo, at a spot which is distinguished by the growth of its ectoderm. A viscid substance is secreted in the gland-like invagi- nation which appears at this point (Fig. 174, s). This substance gradually fills up the invagination and reaches the surface, where it is hardened as soon as it comes in contact with the water (s '). When the invagination dis- appears its edges remain as a raised ridge, and so form the rudiment of the mantle, which is therefore very closely connected with the formation of the shell. This ar- rangement, which has been made out in the larger divisions of the Conchifera, points to the common origin of this group, while it also affords an explanation of the cause of the different ways in which their shells are formed. When this invagina- tion disappears the shell becomes external, and then the edge of the mantle either remains below it or more or less covers it. The latter arrangement shows how the external shells are connected with the internal ones, which are formed when the invagination does not disappear, but becomes still further developed in the manner already indicated. The shell is then secreted from the inner face of the walls of this organ, and it varies greatly in character in various divisions, just as do the external shells. When simplest, all the lautellas of the shell are similar in character ; in many, when in their lowest conditions, it is perforated by pore-canals. The simple condition is complicated by the appearance of layers of prisms set obliquely or perpendicularly to the lamella3. The shell increases in surface at its free edge, the deposits occurring in layers, and at the side of the mantle ; superficially they have the appearance of concentric rings. The shell becomes thicker internally by being supplied from the outer surface of the mantle. These varying modes of formation give rise to variations in the structure of the formed shell, the inner portion of which often consists of a large number of superjacent folded layers, the presence of which is the cause of the nacreous appearance of the shell (mother- of-pearl). These layers are covered by the external more compli- cated and compound layers, which are formed by the edge of the mantle. The horny covering (periostracum) of many shells is due to the same part. The shell of the Lamellibranchiata, like the mantle, is developed on either side of the body, but it is not calcified in the middle line, so that it forms two valves, which are connected with one another 332 COMPARATIVE ANATOMY. along the median line by means of the u'ncalcified portion of the shell. The "hinge" is formed at the point where the two valves pass into one another ; the non- calcified chitinons substance which connects the two shells forms the ligament. Its layers pass into those of the shells ; and the two valves are seen to be merely parts of a structure which is rudimentarily single, and later on becomes so again, and is homologous with the shells of the rest of the Mollusca. Near the ligament the valves form alternating and inter- locking processes (cardinal teeth) which serve to close the shell more perfectly. The shells of the Gastropoda are most markedly distinguished from those of the Lamellibranchiata by the absence of any uncalcified portion. They are not unfrequently internal. This internal position is generally found in those Tectibranchiata that have a rudimentary shell, and in some Pulmonata. In the latter (Helicinaa) the shell very soon becomes external, while in others (Limacina?) it remains rudimentary, and placed within the mantle. Sometimes it merely consists of a few calcareous concretions. The various stages of atrophy are also to be seen in the shells of some other divisions, as, for example, in the Heteropoda, in which the rudimentary shell of Carinaria is intermediate between the shell of Atlanta, which covers in the whole body, and that of the Ptero- trachea?, where it is altogether absent. But these latter have a tem- porary shell during their larval life, which covers in the whole body, just as it does in the Opisthobranchiata, which are also shell-less in later life. Its general presence points to its being a common heritage of all the Gastropoda, some divisions of which lose it early. The Thecosomatous Pteropoda resemble the Gastropoda in forming a shell. The animal does not always occupy the whole of the shell. Many Gastropoda withdraw themselves from the end of their shell as growth proceeds ; and the end is then shut off by a layer of shell- substance. The same thing happens in some Pteropoda (Chresei's), and indicates the commencement of that arrangement which is so much more distinctly marked in the Cephalopoda. The substance of the shell, which is a product of excretion from the mantle, varies very greatly, from the soft structures of some to the firm solid parts which form the shells of most Prosobranchiata. The soft form of shell consists of organic substance merely. Shells become firmer and horny in character when impregnated with cal- careous salts ; and when the inorganic substance forms the greater part of the shell we get strong coverings. The simple condition of the cup-shaped embryonic shells persists in some Gastropoda, and by growing regularly gets to have a more or less flattened or conical shape (e.g. Patella) ; in most, however, it becomes spiral by growing out unequally, and these spiral forms may again undergo all kinds of modifications. As the embryonic shells serve to shelter the whole body, even in those which lose them later on, we must look for the typical form in them ; from it all the other THE SHELL OF MOLLUSCA. 333 forms of shell have branched off. Derived from these we find on the one hand those which are more highly developed, and on the other those rudimentary forms which have the character of degenerate shells. § 259. The simpler shells of the Cephalopoda must also be regarded as rudimentary structures, and not as early forms ; as derived, in fact, from the more complicated and perfect forms, even if their geological succession did not indicate that the shell has undergone gradual Fig. 175. Nautilus; median section of a shell, i Funnel, t Tentacles, v Cephalic lobes. 0 Eye. b Dorsal lobes of the mantle. II Connections between the shell and the mantle, s Part of the shell still connected with the right pallial muscle, a Mantle. s Siphon, s' Siphonal canal of the shell (after Owen). reduction. Their structural characters, as well as their relations to the body, that is, to that portion of the dorsal integument which represents the "mantle," are further instances of the arrangements which we have already described. We either meet with straight (in extinct families only) or coiled shells, which are formed by the mantle, and either completely enclose the animal, or are rudimentary and contained within the mantle ; these latter have lost their signifi- cance as shells and only form internal organs of support. The well-developed shells of the Cephalopoda, as seen in the fossil Ammonites and Orthoceratites, and in the extant Nautilus, 334 COMPARATIVE ANATOMY. differ somewhat in structure from those of the Gastropoda and Pteropoda. They are divided into successive chambers, the most anterior of which is alone occupied by the animal, although the hinder ones are closely connected with it by means of a tubular prolongation (siphon), which is given off by the animal, and traverses the partitions between the chambers. The animal (cf. Fig. 175) occupies therefore the last-formed or youngest chamber only. The separate chambers correspond to an equal number of stages in its growth ; as each segment of the shell was formed the animal left the one before occupied ; and as a partition was formed a new chamber was developed. The arrangement, which was merely indicated, and rarely seen, in the Gastropoda and Pteropoda, has become typical in the Cephalopoda. The straight shells of the fossil Orthoceratites, and those of the Ammonites, which are coiled in one plane, as also the shells of the Nautilidas, are formed on this type. In these last a lobe of the mantle (Fig. 175, b) extends from the dorsal side of the animal over a portion of the shell, the greater thickness of which appears to be partly due to it. The shell of Spirula is completely covered in by the mantle, and is similar in character to that of Nautilus, except that its coils are not in contact with one another ; the shells of the fossil Belemnites are intermediate between those perfect ones which are only enclosed by the mantle and those which are placed within it. On account of this the remnants of shells, which were in all probability largely internal, are of great morphological value. In them the chambers are found in a small conical portion only — the so-called phragmocone. The separate chambers, which form the parts of the phragmocone, look like horizontal sections of a cone superimposed on one another ; they, too, are connected with one another by means of a siphon. The whole phragmocone is surrounded by thickened layers ; but these are not distributed equally over it, but form a strong solid process (rostrum) behind its apex. The broadened lamella-like portion of the thickened layers, which extends forwards over the base of the phragmocone, is known as the " pro-ostracum." The phragmocone is the homologue of the chambered shell of the other Cephalopoda, while the projecting lamella — the so-called pro-ostra- cum — is a continuation of the wall of the most anterior chamber, and the massive rostrum, which is generally the best preserved portion of the whole shell, is derived from the simple thickened layers, which are formed from that part of the mantle which is turned over the shell. The so-called " os sepias " of the Sepiaclae is a flattened shell which is completely hidden in the mantle ; its posterior tip, however, often projects, and so calls to mind the shell of the Belemnites. It consists of several layers, which are rich in organic substances, and are separated from one another by layers of calcareous deposits in such a way that it appears to be made up of superimposed lamellae. The outermost lamella, which is turned towards the dorsal surface of the animal, is pretty firm ; it passes directly into the posterior end THE SHELLS OE MOLLUSCA. 335 of the shell, aud forms the groundwork for the lamellar deposits, which often become very thick on the inner face of the shell, the carve of which is slight. These shells may be directly derived from those of the Beleinnites, especially if we take into consideration those shells which have a free projecting tip, like the shell of S. Orbig- niana. The solid tip corresponds to the rostrum of the Belemnites, while the alveolar cavities of these latter, and the pro-ostracum, which is continued on from their dorsal surface, is homologous with the rest of the shell of the Sepiada3. The partitions which form the chambers of the phragmocone in the alveoli of the Belemnites are represented by the flat or slightly concave laniellee of the shell of the Sepiada3. The layers succeed one another directly, instead of forming separate chambers. In this way the complicated shell of the Belemnites, when reduced, may be easily seen to be represented in a lower con- dition of the shells of the Sepiadre. The shell of the Loliginidas is still more reduced; it is merely formed by an elongated flexible horny-blade (calamus), which is placed in the dorsal region of the mantle. An outwardly projecting carina extends along the middle line. This rudimentary shell corresponds to the external, curved, and more highly organic portion of the shell of the Sepiadee, and is there- fore homologous with the horny-blade of a Belemnite shell. Finally, in the genus Octopus, where the mantle is not separated from the head in the region of the neck, we find a pair of thin plates, embedded in the dorsal integument ; these are the last traces of a shell formed by the mantle, and are in all respects comparable to those described as existing in the Gastropoda. Inasmuch as the shell is, even in the Cephalopoda, formed in the earliest stage by an invagination of the mantle (Sepia), the internal and external shells are closely allied ; and at the same time we may see that they are connected with the shells of other Mollusca. The shell of Argonauta is to be regarded as altogether different from all these shells, which are intelligible when closely compared; it is not secreted by the mantle, but by a pair of arms, which do not lay down lamellar deposits. In the Gastropoda we met with a special arrangement by which the so-called " operculum " was formed ; we found this on the dorsal surface of the end of the foot in many Prosobranchiata, where it served to shut in the animal, when it was retracted into its shell. The question now arises, may not this structure be also derived from one plate of the Placophora ? [E. Kay Lankester, Developt. of the Pond Snail. Quarterly Journ. Microsc. Sci. 1874.] Branchiae. § 260. The kind of respiratory organs — branchia3 — which obtains in the Mollusca is in correlation with their aquatic habitat ; these are always differentiations of the integument, and have consequently a 33G COMPARATIVE ANATOMY. primitively superficial position, although, they become covered over by folds of other regions of the integument (mantle), and so come to be placed in a special cavity — the branchial cavity. The function of respiration is part of the duty of the integu- ment, but it does not seem to be always localised in homologous regions, so that we cannot regard all the organs which appear to be gills as morphologically identical. As a rule the gills of the Mollusca are processes which are placed at the sides of the body, and when least metamorphosed arise between the foot and the mantle (cf. Fig. 167, A B br). They vary very greatly, not only as to the extent of the body which they occupy, but also in the way in which they are connected with different parts. In the Placophora they merely form a series of folds or lamella?, which encircle the body between the foot and the mantle, and appear to be formed from the epipodium (epipodial gills). In the Lamellibranchiata they form lamellar organs, which project between the mantle and the visceral sac, which ends with the foot, into the cavity enclosed on either side by the mantle (Fig. 1 76, br br') . Their free edge is directed towards the ventral surface. Almost all the Lamellibranchiata have two pairs of these gills on either side, an inner pair, which are placed mediad, and an outer pair at the sides of these. The former are often the larger. Ex- cept in Anomia, where there are a large number of other adaptive modifications, the gills are arranged symmetrically. Each gill-lamella is developed from a row of pro- cesses which bud out close to one another; in many forms these pro- cesses remain separate from one another, and form separate branchial filaments, parallel with each other (Mytilus, Avicula, Area, Pectun- culns, Pecten, Spondylus). In most, however, the gills lose this embryonic condition, owing to the concrescence of the gill-filaments. Owing to this union of the flattened filaments or lamella?, which have their surfaces directed towards one another, a gill-plate is formed; this is either effected by the mere adhesion of the filaments, or by con- crescence; in the latter case pad-like projections appear on each of the gill-filaments at regular distances from one another, and these unite together. As fine clefts are left between these junctions, by which the water passes through the gills, each plate forms a kind of lattice-work. These filaments are not simple prolongations, but loops, so that they enclose a space (intrabranchial space); when the gill- Fig. 176. Vertical section through an Anodonta. m Mantle, br Outer; br Inner gill -lamella. /Foot, v Ven- tricle, a Auricle, pp' Pericardial cavity, i Enteric canal. BRAXCHLE OF MOLLUSCA. 337 ^laments grow togetlior this space traverses the whole of the gill- plate, and communicates with the exterior by means of the clefts between the filaments. The water which enters by these clefts is collected into a canal at the point where the plate is attached ; and is carried by it to the hinder end of the body. There are chitinous rods in each of the gill-filaments, which form a special organ of support. The surface of the whole gill is covered by ciliated epithelium. Rows of large cilia extend along the ridge-like projections of the gills; between these there are finer closely-packed cilia, and the two together keep up a continual stream of water. There is a groove at the free edge of each gill-plate ; this is formed by depres- sions on each of the gill-lamellae and is invested by longer cilia ; it leads to the mouth, and so produces a current of water, which is well adapted to bring in nutriment. Great modifications of this system are brought about by the fusion of the gills of the two sides; this, which obtains posteriorly to the foot, is either effected by the direct junction of the free edges, or by the development of a special membrane, which unites the gills of either side together. This fusion is best seeu in the falci- form curved gill -plates of Anomia, where the whole branchial apparatus is separated from the greatly-reduced visceral sac, and is no longer lateral in position. [R. Holman Peck, The structure of the Lamellibranch gill. Quart. Journal Microsc. Sci. 187G.] B onnet, E., Der Bau u. die Circulationsverhiiltuisse der Acephaleukieme. Morphol. Jahrb. III. §201. The branchial apparatus of the Gastropoda, though greatly varied in details, is arranged in very much the same way as in the Lamellibranchiata ; that is to say, it is typically made up by lamella?, or by more distinctly cylindrical processes, which are arranged parallel to one another. These project from the surface of the body, and are, therefore, bathed by the surrounding medium, the water, while a current of blood passes along them internally. This simi- larity is still more marked by their position relatively to the mantle, for they have just the same relation to it as have the gills of the Lamellibranchiata. As compared with these latter they are less numerous and more confined ; their structure is comparatively much simpler. The epipodial gill is arranged circularly in the Placophora, as it is also in Patella ; but in other Patelliche (Lottia) the two pinnate branchiae are more distinctly dorsal, so that they seem to be different from the epipodial gills. Fissurella and Emarginula also have their two gills placed in the anterior region and below the mantle. In Haliotis also, they are distinctly paired, but they are placed more to the left hand of the animal. They are also characteristically modified in the Zeugobranchia. In the Anisobranchia the left gill is smaller, and the right one more largely developed ; this arrangement, which is z 338 COMPARATIVE ANATOMY. distinctly allied to the arrangement in the Zeugobranchia, is adapted to the asymmetry of the branchial cavity, which, again, is dependent on the characters of the shell. The smaller gill is generally approxi- mated to the other, and becomes asymmetrical in position ; in some Prosobranchiata it disappears altogether (Janthina, Neritaceas, Heteropoda). The right gill is generally developed on one side only, so that it is semi-pinnate, owing to the disappearance of the second row of lamellae. Although, as a general rule, we find the lamellar structure to be the most common, a few (Calyptra3a, Crepidula) have fila- mentous gills, and so call to mind the primitive form of the Lamelli- branch gill. The gills become modified, and may disappear altogether when the mantle and the branchial cavity are atrophied. This happens in various divisions ; thus, in the Heteropoda, among the Proso- branchiata, the gill of Carinaria is not covered over by the mantle ; in Pterotrachea, where there is no mantle at all, the gill is quite free, while in Firoloi'des, the gill, as well as the mantle, disappears. Among the Opisthobranchiata the characters of the gills are equally dependent on the condition of the mantle. There is a gill on either side, between the mantle and the foot (Pleurophyllidia), or there is only a single gill in the gill-chamber, or, finally, the gill is only partly covered over by the mantle (Tectibranchiata) . When the shell and mantle disappear, gill-like struc- tures may be developed on the dorsal surface of the body, as in some of the Nudibranchiata. Lamellar, or tufted and branched appen- dages, are sometimes developed in the anal region (Doris), sometimes in rows over the whole body (Tritonia, Scyllasa). If we are right in regarding the possession of a shell by the larvae of all the Opisthobranchiata as a fact, which proves conclusively that these Gastro- poda are derived from shelled forms, and if wo must suppose that the primitive position of tlio gills was within the mantle- cavity, then we must regard the arrangement of the gills in the Dorididae as having been inherited in its essential features from this condition, for we must remember that the anus also is placed in the mantle -cavity. There are many inter- mediate steps between this and the more general distribution of gills over the back of the body ; and further, these gills, howsoever modified, and howsoever peculiar their form, are never anything more than mere processes of the integument. Their varied external form is due to their superficial position, which is due to the loss of the mantle which invested them ; and it is because of its absence that they Tig. 177. Ancula (Polycera) cristata; dorsal view. a Anus. Ir Branchia?. t Ten- tacles (after Alder and Hancock). BEANCHI/E OF MOLLUSCA. 339 lose their apparently specific structure, and get to be more and more like the surrounding integument, of which they form processes varying greatly in character. Their relations to the circulatory apparatus are of great significance as bearing on this view, for they so far agree completely in character with the true gills. Lastly, when most differentiated, the gills are seen to bo distributed over the whole of the dorsal region of the body, where they form one or more rows of papilla3, or villous processes on either side ; these again may bo branched (^Eolidise). The loss of the shell is the cause of the wider distribution of the gills, just as, on the other hand, the forma- tion and development of this organ of defence was the cause of the gills being more limited in extent. These gills are atrophied in many Opisthobranchiata, when the whole of the integument takes on the respiratory function (Phyllirhoe, Elysia, Pontolimax) . § 262. Another arrangement of the respiratory apparatus, which is a modification of the one first described, is due to the development of the respiratory canal-system in the walls of the mantle -cavity. In many of the branchiate Gastropoda this network of canals extends beyond the gills into the "neighbouring parts of the branchial cavity, which are thereby enabled to take part in the respiratory function. In this way the mantle-cavity is adapted to taking in air, and becomes a lung. An organ of this kind — which is not at all adapted for those Molluscs that are so organised as to be fit for an aquatic life — is found in various forms, which belong to very different divisions, and it is to be regarded as due to a change in their mode of life. A lung is present, in addition to a gill, in Ampullaria ; in this animal it forms a sac, which is placed by the side of the gill, and is pro- vided with a contractile orifice. In the terrestrial genus, Cyclostoma, the gill has disappeared altogether. There is a lung in Onchidium, but it is also a renal organ. A similar cavity is found close to, and has the same orifice as, the renal organ in the Helicinae and Limacinaa ; this functions as a lung. In the LymnaBidoe and Planorbidre, however, the mantle-cavity itself is adapted to the reception of air. But in these forms the abran- chiate mantle-cavity also serves as a water-breathing organ, for many Lymnasidas are known to live always in deep water. § 263. In the Gymnosomatous Pteropoda either the whole of the integu- ment (Clio) serves as an organ of respiration, or processes are developed from its surface which function as gills (Pneumodermon) . In the thecosomatous forms only do we meet with plaited folds (Hyalea), which are placed in the mantle-cavity (Fig. 171, A br), and so far resemble the arrangements which obtain in the rest of the Mollusca ; their position is the same as that of the gills in the 340 COMPARATIVE ANATOMY. Cephalopoda. Iu these latter tlie gills arise in just the same way — ■ between the mantle and the foot (Fig. 171, B br) — as they do per- manently in many Gastropoda. When, however, the mantle is developed they sink downwards, and are then placed in a mantle- cavity, which, as compared with the Gastropoda, appears to open on the hinder surface. The gills are arranged symmetrically in all of .V.lj7 Fig. 178. Mantle-cavity and funnel of Sepia officinalis. The mantle-cavity has been opened by an incision along the middle line. In it is seen the visceral sac projecting, while posteriorly to it two muscular branches are given off (m.) to the funnel and head. Br Branchiae, v br Branchial vein, v br' Its bulbous enlargement. t Ink-bag. r Orifice of the excretory organ, opened on the right side, and displaj'ing at R the venous appendage, g Genital papilla, a Anus. J Funnel, opened by an in- cision along the middle line, i Tongue-shaped organ, c Depression for the process at the edge of the mantle (paHial hinge) c'. C Head. P Fins. them ; there are two pairs in the Nautilus, but only one in all the other extant Cephalopoda. As a rule, each gill is pyramidal in form, with the apex directed towards the side, and its base towards the middle Hue (Fig. 178, Br). It either consists of closely-approximated lamella1, which gradually increase in number at the tip (Nautilus, and most Loliginidse), or of several much-coiled dermal folds, which arise between the two branchial vessels which pass to the edge of the mantle (Octopoda). INTERNAL SKELETON OF MOLLUSCA. 341 In this group the mechanism of respiration is combined with the locomotion of the animal. Each time that the muscles of the edge of the mantle relax, water passes into the branchial cavity by its orifice, that is, at each side of the funnel; after it Las bathed the gills it is driven out again by the contractions of the mantle. At this moment the cleft of the branchial cavity is closed, so that the water cannot get out except by the funnel, and this serves not only as the passage by which the water reaches the exterior, but also takes an active share in driving it out. Internal Skeleton. § 264. In most Mollusca the absence of an internal skeleton is compen- sated by the shells and tests described in § 258; for these serve as supports for the internal parts. Independent in- ternal organs of sup- port are, however, found in the Gastro- poda. Two, or some- times four, small plates of cartilage are found in the head of these animals ; they are surrounded by the muscles of the pha- rynx, and arc more or less closely connected with one another. They form the sup- porting apparatus of the radula and the parts connected with it, and also afford points to which some of the pharyngeal muscles, and especi- ally those of the ra- dula, are inserted. Cartilaginous or- Fig. 179. Section through the head of Sepia offici- nalis. K K' Cephalic cartilages. G Cerebrum, go Gan- glion of the optic nerve, w White body. I Lens, ci Ciliary body, e Cornea, p Eyelid. P Buccal mass, m External, n Internal labial membrane. e/Jaws. r Radula. oe (Eso- phagus, t Arms. gans of support are much more highly de- veloped in the Cepha- lopoda. The most important one lies in the head, where it serves as an investment of the nerve-centres, a support for the optic 342 COMPARATIVE AXATOMY. and auditory organs, as well as the point of origin of a large number of muscles. In Nautilus this cephalic cartilage is formed of two pieces, united along the middle line, and drawn out into anterior as well as posterior processes; these surround the com- mencement of the oesophagus. In the Dibranchiata the cephalic cartilage is much better developed. It consists of a median portion, which is traversed by the oesophagus (Fig. 179, K), and of two lateral processes, which are sometimes mere flat enlargements, in which case accessory cartilaginous plates are added on to them to form the orbits ; at other times they are more highly developed, are then continuous with superior processes (K% and completely enclose the orbits. The central nervous system (G) is placed on that portion of the cephalic cartilage which is traversed by the oesophagus. The Dibranchiata are provided with additional cartilaginous skeletal pieces. A dorsal cartilage is the most common. In the Sepiadas this forms a semi-lunar piece, which lies in the anterior dorsal region of the mantle, and is continued into two small lateral cornua ; in Octopus, where there is no median pieces, we find the cornua only. There is a cartilaginous plate in the neck also, in addition to two cartilages at the base of the funnel — the hinge-cartilages. These are less constantly present than those which lie at the base of the fins, and which are found in all Dibranchiata provided with fins, for they serve as the point of attachment for the muscles of these organs. Muscular System. § 265. We can understand how it is that separate groups of muscles are so feebly developed, if we bear in mind that there is a dermo- muscular tube united with the integument and external organs of support, and that these, notwithstanding the great modifications which they undergo, have very much the same character in all cases. With this may be correlated the absence of internal organs of support in the lower divisions, and their relatively slight develop- ment in the higher ones. The muscular system is made up of band- like fibres, which, not unfrequently, give indications of their greater differentiation by the possession of transverse strias. In the Lamellibranchiata the adductor muscles, which pass either inmsversely or obliquely through the body from one valve to the other, are those that are best developed. There are either two of them, which form bundles separated by some distance from one another, one anterior (Fig. 167, ma), and the other posterior (m p), as in Unio or Anodonta; or there is but one muscle, which corre- sponds to the hinder one of the Dimyaria, and occupies the middle of MUSCLES OF MOLLUSCA. 343 the shell (Pecten, Ostrea). Special muscles, which are Interwoven with the integument, serve as retractors of the foot ; these arise from the dorsal portion of the shell, and are sometimes broken up into several pairs. These retractors are also found again in the shelled Gastropoda. They generally form a single, but sometimes a double muscle, which arises from the base of the shell, and which passes to the anterior regions of the body, increasing in size as it does so. It supplies the foot as well as the head, and the anterior region of the digestive tube (pharynx); while further it gives off special bundles to the other protractile regions, that is, to the tentacles and copulatory organ. The muscle which arises from the columella of the shell, and accompanies it, is known as the columella muscle. In the Heteropoda it has a wide origin in the carinate foot. In the Ptero- poda it spreads out into the fins given off from the foot. In addition to these muscles other bundles are given off to the viscera. The muscular system of the Cephalopoda is much more differen- tiated, in correlation with the formation of an internal skeleton. Two powerful retractors are attached to the cephalic cartilage in Nautilus ; these arise from the sides of the shell-chamber occupied by the animal. In those Decapoda that have an internal shell these muscles take their origin from the wall of the outer wall of the shell; and in the Octopoda, from a cartilage found at that spot. Two branches are given off from these two muscles, which pass to the funnel. Another and much larger pair of muscles arises in the neck of the animal, and broadens out towards the ventral sui'face, where they pass into the funnel. The muscles in the mantle are also arranged in separate layers, as are also the fin-muscles. Lastly, there is the greatly-developed muscular system of the arms, which partly arises from the cephalic cartilage, and surrounds a canal which passes along the axis of the arm. Nervous System. Central Organs and Nerves of the Body. § 2G6. This system of organs also has points in which it resembles that of the Vermes. The whole of the central apparatus, that is, is divided into a superior ganglionic mass, which lies above the commencement of the oesophagus, the supra-cesophageal or cerebral ganglia, and a ventral mass which is connected with the other by commissures, and forms the inferior or pedal ganglia. They are both paired. The earliest rudiment of the cerebral ganglia is seen as a differen- tiation of the ectoderm, the form-elements of which grow inwards, and are accompanied by the rudiments of the eye (Gastropoda). The relations between the cerebral ganglia, and the higher sensory 344 COMPARATIVE ANATOMY. organs which are placed in the head, prove that these ganglia are homologous with the cerebral ganglia of Vermes (and of Arthropoda). The pedal ganglia may also be derived from a more simple condition, for in many of the lower Mollusca we find them replaced by an arrangement which corresponds to the ventral chord of the Annulata. Longitudinal trunks are given off from the pedal ganglia, and are distributed along the foot ; since they are con- nected together by transverse chords, they are arranged in the same manner as a ventral nerve-chain. Although there may be nothing really fresh in this arrangement of the nervous system, inasmuch as the two ventral (or pedal) ganglia must be regarded as a concentrated nervous mass, which is broken up in lower forms, and constitutes a ventral ganglionic chain; yet the greatly-developed visceral ganglia form an arrangement which is nothing like so well marked in the Vermes as it is here. In the Mollusca the visceral ganglia are not only of importance as forming a part of the general nervous system, but they may also fuse with the cerebral ganglia, owing to the gradual shortening of their commissures. New, and primitively peripherally-placed parts, are thereby added on to these central organs ; and it becomes a matter of doubt whether or no these ganglia, which formerly belonged to the visceral nervous system, should still be regarded as belonging to it. These parts of the nervous system which supply the viscera (heart, branchial apparatus, and generative organs) are the cause of great complica- tions of the whole system ; owing to the way in which they vary in number in different divisions, they make compainson very difficult, as indeed also do the great modifica- tions in position undergone by the primitive ganglia, in consequence of the abbreviation or elongation of their commissures. The nervous system of the Placo- phora is one of the lowest found. A nervous band formed of two chords (Fig. 180, 0) surrounds the oesophagus, but there arc no superior enlargements on it; this is probably due to the absence of eyes and tentacles. The inner of the two chords is continued separately below the oesophagus ; part meets its fellow of the other side in the subpha- ryngeal ganglion and part passes on to a pedal ganglion (P). Each of these bilateral ganglia gives off a thick nerve- trunk, which passes backwards, and which, like the ganglia themselves, is connected with the trunk of the op- posite side by transverse anastomoses set at regular distances; nerves are given off to the foot from corresponding points. The outer chord, Fig. 180. Chiton cinereus. nerve-chord. P A probe passed through pylorus, c Commence- ment of the caecum. e e Its spiral portion. i Hind-gut. a Ink-bag. b Its opening iuto the rectum (after Home). of glands are to be Organs appended to the Enteric Canal. 1) Appendages of the fore-gut. § 280. Of those glandular organs which are connected with the enteric caual, the salivary are found only in those forms in which the pharynx is developed ; it is possible therefore to make out a certain connection between these structures. In the Gastropoda they are always placed on each side of the fore-gut, and open into the pharynx. Sometimes they form short cascal tubes (Pteropoda), which are sometimes hidden in the very substance of the pharynx 364 COMPARATIVE ANATOMY. (many Opisthobranchiata) . When they are more highly developed their duct is elongated, so that the secreting portion comes to lie some way further back ; sometimes it is placed on the oesophagus, and at others on the stomach itself. In this case the glands are rounded, elongated, and generally flattened tubes (Prosobranchiatn, many Pulmonata), which may be broken up into several smaller parts, or have the form of ramified organs ; the glands found on the stomach of Pleurobranchus are examples of this latter kind. We not unfrequently meet with two pairs, the efferent ducts of which are either separate all along their course, or the ducts of the hinder pair unite with one another. Even when there is but a single pair of glands they may be often observed to fuse into one mass, the double nature of which is shown by the presence of two efferent ducts. The salivary glands of many Prosobranchiata are differen- tiated functionally (Dolium, Cassis, Cassidaria, Tritonium), for part of the gland secretes free sulphuric acid. The glands of some. Opis- thobranchiata (Pleurobranchus, Doris) are differentiated in the same way. Among the Cephalopoda, Nautilus is provided with a paired glandular mass, which is placed inside the pharynx. These glands are also present in many Dibranchiata (Octopus, Eledone), as are others, which are short and lie just behind the pharynx; these have an efferent duct which penetrates the wall of the pharynx, and unites with its fellow of the opposite side immediately in the orifice of the duct (Fig. 199, gls .), while the trunk is continued on as an aorta cephalica (aa). This latter passes directly to the anterior 2 b 2 372 COMPARATIVE ANATOMY. parts of the body, and generally gives off a large branch to tlie foot ; this sometimes looks as if it were a continuation of the chief trunk. In addition to this, it gives off on its course branches which pass to the stomach, salivary glands, and so on ; it either ends simply, or after several ramifications, in the region of the pharynx. In the Pteropoda it is more widely distributed, for in them the pedal branch has the characters of a continuation of the chief trunk, and in the head it divides into two large terminal branches which ramify largely in the fins. The visceral artery, which corresponds to the posterior artery of the Lamellibranchiata, is but slightly broken up iu the Pteropoda and lower Gastropoda. Like the cephalic artery, it loses itself in larger hasnial spaces. In most of the Prosobranchiata it is greatly developed, and much broken up (ap) ; this is the case also in many of the Tectibranchiata (Pleurobranchus), but in others it is replaced by several smaller arteries (Aplysia). The visceral arteries of the Nudibranchiata are branches of the principal trunk (Doris). The afferent vessels differ according to the number, form, and disposition of the respiratory organs. In many of the Nudibran- chiata the blood is collected from the body-cavity near the auricle. In others, which have distinct respiratory organs, there are definite canals, or even vessels with special walls, which convey the blood from the venous passages to the respiratory organs. When these are most simple in character, the blood passes to the auricle without going through branchial veins. This is the case, too, in many of the Pteropoda. When the gills are more highly developed the returning blood is collected into special venous trunks, which open, either separately or together, into the auricle. These branchial veins are always so arranged as to be adapted to the size as well as to the posi- tion of the branchias. In many of the Nudi- branchiata (.ZEolidia, Scyllasa, Tritonia), large vessels pass off from the gills and unite together to form larger trunks, which give rise to a median or two lateral branchial venous trunks, connected with the auricle of the heart. When the bran- chias are scattered over a larger portion of the surface of the body, this afferent system of bran- chial vessels is well developed, but when they are limited in extent it is reduced in size (Doris, Polycera). Tritonia is an example of the former arrangement (Fig. 198), for in it two lateral bran- chial venous trunks (ss) pass by means of a trans- verse trunk to the heart. The transverse canal forms a kind of double auricle (a), although indeed it has only one opening into the ventricle (v). The paths by which the blood goes to the branchias are always more or less lacunar. In many of the Opisthobranchiata the blood Fig. 198. Part of the circulatory or- gans of Tritonia. s Venous sinuses laid open. The wall is perforated by the openings of the branchial veins. v Ventricle. VASCULAK SYSTEM OF MOLLUSCA. 373 o©\© from the ccclom is collected into canals, which run in the integu- ment; and thence it is distributed to the gills. All the blood, how- ever, does not pass to them, for some is returned to the heart after having passed through the integument. In Helix and Limax the haemal spaces, which pass into the wall of their branchial cavity, and which form a system of canals which carries blood to the res- piratory organs, are de- veloped in a vascular manner. They break up into a rich network in the integument, and this network gives off a number of large trunks with distinct walls, which unite to form a " pulmonary vein ; " and this vein passes into the auricle. The network of pulmonary vessels may be regarded as a large blood sinus, which extends into the walls of the lungs, and which is broken through at various points by islets of firm substance. § 287. The heart of the Cephalopoda is placed at the base of the vis- ceral sac, and is formed of a rounded or trans- versely - oval ventricle (Fig. 195, BG; Fig. 199, c), which receives blood from as many bran- chial veins as there are Fig. 199. Anatomy of Octopus. Mantle-cavity and visceral sac, opened from the ventral surface, ph Pha- rynx, gls s Superior salivary glands, gls i Inferior salivary glands, o Eye. i Funnel, br Branchiae. ov Ovary, od Oviduct, c Heart, v Ir Branchial veins. a Arteria cephalica. vc Arenas cavas. av Venous appendages (after Milne-Edwards.) branchiae. That is to say, in Nautilus there are four, and in the rest of the Cephalopoda two branchial veins opening into the ventricle. These veins are generally considerably widened out in front of their opening into the ventricle (Fig. 199, v br), and this enlargement is homologous with the auricle in the Gastropoda and Lamellibranchiata. Two arterial trunks always arise from the heart; one, the larger, passes straight forwards ; this is the arteria cephalica (Fig. 199, a) ; at some dis- tance from it a smaller trunk, the arteria abdominalis («') is given 374 COMPARATIVE ANATOMY. off, and this generally passes backwards. This, which is the general mode of arrangement, shows distinctly how the Cephalopoda agree with the two other classes, while it points to their having a close affinity to those Mollusca which are distinguished by the possession of two auricles. The cephalic artery first of all gives off large branches to the mantle, and some to the intestinal tract and to the funnel ; when it reaches the head it gives off the optic arteries and supplies the oral regions, and divides into larger branches, the number of which depends on that of the arms. In some of the Cephalopoda the brachial arteries spring from a circular vessel, which is developed around the commencement of the oesophagus. The abdominal artery is more varied in character; in some of the Sepiadas (Fig 202, a') and Loliginidas, it arises opposite to the cephalic artery, and has, therefore, the same relations as the visceral artery in the Lamelli- branchiata ; but, in the Octopoda, it arises from the anterior region of the heart, close to the aorta (Fig. 199); it very soon breaks up into several branches for the enteric tube and generative organs. Iu the Loliginida3, however, it gives off two additional branches for the fins ; in Ommastrephes a special enlargement (which is perhaps an accessory organ of circulation) may be observed on these vessels. All over the body the terminal branches of the arteries communi- cate with the veins by means of a well-developed system of capil- laries. In the greater part of the body, at any rate, this system takes the place of the lacunar blood-passages which were found in the other Mollusca, and of which it seems to be a differentiation. The venous roots from the capillaries are collected into larger trunks, which have sometimes the characters of true veins, and are sometimes widened out into large spaces, so that they are inter- mediate between true vessels and mere lacuna?. Of the more special characters of the venous system we have to note that the brachial veins are united into a circular sinus, placed in the head ; this is supplied also by neighbouring smaller venous trunks, while it gives off a large haemal canal (vena cephalica, or great vena cava) (Fig. 202, vc), which passes backwards to the branchial region. At the gills it breaks up into two (Dibranchiata) or four (Tetrabran- chiata) venous trunks ; these take up the other veins which come from the mantle and viscera (vc") and pass to the base of the gills. In the Dibranchiata the branchial arteries acquire a muscular in- vestment and form a contractile portion, or branchial heart (Fig. 202, vc ), which pulsates rapidly, and acts as an accessory organ of the circulation. Special appendages are attached to the branchial artery in front of these branchial hearts ; these, which are diverti- cula of the walls of the vessels, arc bathed by the venous blood, which passes into the branchiae, in just the same way as are the organs of Bojanus in the Lamellibranchiata (vide Excretory Organs, § 289). Although the venous blood-receptacles of the Cephalopoda which we have described may be regarded as a venous system, provided with closed walls, true blood lacuna) are not absent. They are dis- tributed in just the same way as in the other classes of the Mollusca. EXCRETORY ORGANS OF MOLLUSCA. 375 The cceloin is such a blood-space ; all the organs in it are bathed by venous blood. Various veins open into it, and it is also connected by two canals with the large vena cava (vena cephalica). Milne-Edwards et Valenciennes, Nouv. obs. sur la constit. de l'appareil de la circulation chez les Moll. Mem. Acad, des Sc. T. XX. Milne-Edwards, Voyage en Sicile. T. I. § 288. The blood fluid of the Mollusca is, as a rule, colourless, and often has a bluish or opalescent appearance. In many Cephalopoda, however, it is violet or green, and in some Gastropoda (Planorbis) the blood is red — the plasma being coloured. The form-elements of the blood are always colourless, and have the character of indifferent cells, which give off all kinds of pseudo- podia-like processes during their amoeboid movements; this has been observed in the Lainellibranchiata and Gastropoda. A rounded organ extends alongside the branchial arteries of the Cephalopoda; its function is not known, but it may perhaps be an organ which is of importance in the development of the form- elements of the blood. [Lankester, E. Rat, Distribution of Haemoglobin in the Animal Kingdom (red corpuscles of Area and Solen). Proc. Royal Society, 1873.] Excretory Organs. § 289. In addition to the various organs which have been already noted as present in the integument, and which serve as excretory organs, there are others which open on to the surface of the body, and which are much more important from this point of view. In the Placophora there is a glandular organ which is placed close to the anus; but it is not certain that it is comparable to the excretory organ of the Conchifera. Its internal orifices have not been observed. For the present, therefore, we must regard this organ as not belonging to the same series as the excretory organs of the rest of the Mollusca. These typical excretory organs of the Mollusca are homologous with the organs found throughout the Vermes, and there called renal; which organs form the looped canals (nephridia) of the Annulata. In the Mollusca also they com- mence by an external orifice, and open into the cceloin after a longer or shorter course. The internal opening is distinguished by special arrangements; most commonly, perhaps, indeed, generally, by an in- vestment of cilia, whereby they call to mind the ciliated funnels of the looped canals of Vermes. Owing to the presence of these organs, the internal cavity of the body communicates with the surrounding medium. They are able, therefore, to bring water into the body, while, like their homologues among the Vermes, they may preside 376 COMPARATIVE ANATOMY. over o tli or functions also. Thus they may have relations to the generative organs, as is clearly the case in some of the Lamellibran- chiata. There is good reason also for thinking that the efferent ducts for the generative products in the Cephalopoda are derived from these excretory organs. They have not, therefore, any exclusive relation to excretion. When they are excretory in function, the walls of the canals, which are otherwise simple, undergo a certain amount of metamorphosis, and may then be seen to have a glandular structure. In these cases they may be regarded as " kidneys," on account of the chemical constitution of their products. In this case, if examined under the microscope, they are seen to be provided with secreting cells, the contents of which are formed by granular or con- centrically striated concretions, similar to those which are of such importance in the renal secretions of other groups of animals. Where the internal orifice has been observed, it has been seen to lead into the pericardial sinus, through the wall of which the duct passes. If it be true that the excretory organ is derived from a looped canal, it is highly probable that the wall of this pericardial sinus is derived from a dissepiment, such as those which carry the openings of the looped canals in the Annelides. Many facts, however, are wanting to confirm this supposition, and particularly those which would explain how the change in the position of this dissepiment has been brought about. § 290. In the Lamellibranchiata the excretory organ is known as the Organ of Bojanus; it is always paired, although sometimes it is fused into one mass along the middle line ; it lies on the dorsal side of the body, close to the base of the gills. Its substance is made up of a yellowish or brownish coloured spongy tissue, the interspaces in which often run together, and generally form a large central cavity. From this cavity a pore, on either side, leads into the peri- cardium, while another leads into the efferent duct. The latter either lies close to the genital pore, or is confluent with it, or, lastly, the generative organs open into the organ of Bojanus, so that the generative products are passed out to the exterior through it (Pecten, Lima, Spondylus). Area and Pinna have the efferent ducts united; Cardium, Chama, Mactra, Pectunculus, Anodonta, Unio, etc., have the orifices of the excretory and generative organs separate. The walls, which rise up in folds, or the meshwork-like tissue of the organ, is thickly invested with secreting cells, which secrete the already-mentioned concretions, in many of which the characteristic excretion — uric acid — has, of course, not yet been observed. As to its relations to the vascular system, see p. 370. The Scaphopoda resemble the Lamellibranchiata in the possession of a paired excretory organ. § 291. In the Gastropoda the excretory organ varies still more in character. A paired excretory organ — the predecessor of the per- EXCKETOET OKGANS OF MOLLUSCA. 377 niaiient kidney — lias been made ont in the Pulmonata. In adult animals the organ is, as a rule, unilateral. This double rudiment poiuts to its being the same as the paired organ of the Lamelli- branchiata. The recent discovery of a paired excretory organ in Haliotis, Fissurella, and Patella, in which forms the left organ is more or less rudimentary, seems to be decisive on the matter. The degeneration of one of the two organs appears to be correlated with the degeneration of other paired organs, as, for example, the gills. So far as exact observations show, it opeus by one pore into the pericardial sinus, and by the other to the exterior. In the majority of the Gastropoda, uric acid has been detected in the organ. This is especially true of the Pulmonata, in which the kidney, which is placed between the heart and pulmonary veins, can be easily recog- nised on account of its whitish or yellowish colour. It is lamellar or spongy in structure, and the lamellae, or bands which make it up, are covered by large secreting cells, in which firm concretions of various forms can be made out. The long efferent duct, which commences some way back, opens into the pulmonary cavity, which in Helix appears to be a widened terminal portion of this duct. In the Prosobranchiata the kidney lies between the gills and the heart ; it has the same position in some of the Opisthobranchiata. As a rule an efferent duct passes forwards and accompanies the hind- gut, below which, and often not far behind the anus, we find its orifice. In many of the Opisthobranchiata it appears to have lost its excretory function (as in Polycera), or the excretion is fluid. In this case the kidney (as in Phyllirhoe, Actason, etc.) has the form of a long transparent tube, which extends some way back behind the heart along the middle line of the body, and is placed close to the Fig. 200. Diagram of Doris, to show the excretory organ 11. o Mouth. I Lips. B Buccal mass, ce (Esophagus, v Stomach, i Intestine, a Anus. at Auricle. v Ventricle (after A. Hancock). back. It opens into the pericardial sinus by a ciliated orifice, and by a contractile one on to the surface of the body. In many of the Opisthobranchiata a greatly-developed csecal-sac is given off by this organ, which gives off secondary diverticula (Fig. 200, B), and so gradually forms a ramified tube. Structures of this kind, which resemble a ramified gland, have been observed in Doris, Dendronotus, Scyllsea, etc. A canal (R') is continued from the 378 COMPARATIVE ANATOMY. way, e.g. in Chreseis pericardial orifice (>•") into the interior of the tube, where it opens (r), so that it only communicates with the exterior in a round- about manner. In the Thecosomatous Pteropoda, and also in the Heteropoda, the organ which is regarded as the kidney, because of the similarity between its two orifices and those of the Prosobranchiata, is also remarkable for its spongy character. In Carinaria, among the Heteropoda, it is provided with a distinct investment of secreting cells ; in all the rest there is a clear cellular layer instead. The framework of the kidney is stiff; but in Atlanta and the Firolida3 it is contractile, and performs en- ergetic and spasmodic contrac- tions. In the Thecosomatous Ptcropoda also, the kidney can act in this (Fig. 201, re). As the glandular nature of this organ is doubtful when the secreting cells do not contain concretions, greater weight may be laid on its relations to the in- gestion of water, which has been best observed in these cases. The movements of the organ are not limited to merely open- ing and closing its external orifice, but they also drive the water inwards and mix it with the blood which is returning to the respiratory organs from the general circulation ; this organ is always placed in the course taken by this current. Although the ingestion of water by the excre- tory organ has not been directly observed except in these divi- sions, there is no reason for concluding that it does not also obtain in the rest of the aquatic Gastropoda. It is in the Nephro- pneusta only that the pai'ts concerned have other relations, but even in them the kidney has the same relations to the system of blood-canals, for the fluid of the blood maybe observed to pass out by the renal orifice. Fig. 201. Organisation of Chreseis. P)> The fins, ce (Esophagus, v Stomach. r Hind-gut. //Liver, a Auricle, c Ven- tricle, re Kidney, x Its internal orifice. / External orifice. h Ciliated organ. (j Hermaphrodite gland, g' Hermaphro- dite duct. i - ■Si I?1'':)'*' J" ;■ ^ Fig. 213. Organisa- tion of an Ascidian (Amarcecium pro- Kferum). sb Bran- chial sac. v Stomach. i Intestine, c Heart. t Testis, vd Efferent duct of the testis, o Ovary, d Eggs in the body - cavity. The arrows indicate the direction of the stream of water at the orifices of the body (after Milne- Edwards). Organs of excretion have as yet been recognised in the Tunicata to a limited extent. In many Ascidians (Molgula, A. conchilega, com- planata) a tubular organ is found in the neigh- bourhood of the branchial chamber, or placed farther back in the body, which exhibits, besides other cells, some containing concretions. In one species the murexide reaction has been obtained. No openings have been discovered in the orgau, so that the arrangement appears to represent that condition in which excretionary matters arc separated in the organism, and form concretions which are not removed to the exterior. Sexual Organs. § oil. Only one division of the Tunicata is provided with sexual organs universally — the Copelata. In the others, in consequence of the elaborate asexual reproduction, a large proportion of indi* viduals are devoid of sexual organs ; the absence of which is explained by the production of germs, a modification of the process of multiplication by budding (cf. p. 391). The hermaphrodite arrangement, common among Tunicata, is often found at a very low stage of development. The Appen- SEXUAL ORGANS OF TUNTCATA. 407 dicularia) have no efferent ducts for their sexual glands, which are sometimes paired and sometimes single. In the Acopa the sexual elements are discharged into the cloaca. The male organ has the form of a sperm-producing cascum, which in Doliolum, and also in many Ascidians, retains this simple form ; whilst in Pyrosoma it acquires a rosette-like shape, and in most Ascidians, and also in the Salpa?, is produced into branches, and forms a kind of multilobular gland. In many Ascidians (Molgula) the testes surround each of the two ovaries as a number of separate glands, and open to the exterior with separate efferent ducts. The ovaries too have often a multifid character, at least in many Ascidians ; in others they are only represented by a group of eggs of different stages of develop- ment, each of which is enclosed in a kind of capsule. In many only a few such eggs, joined together by a common stalk, are present ; and in the Salpas and Pyrosoma we have only a single egg, the stalk of which exists in the early stages of growth, but gradually shortens. The development of the sexual products occurs here at different periods for the two sexes ; the male organ only reaches its maturity after the egg has commenced to develop as an embryo. The presence of an efferent duct for the sexual products appears to depend upon the greater or less distance of the sexual glands from the cloaca. The whole apparatus, however, requires a very much more careful investigation than it has yet received. Ninth Section. Vertebrata. General Review of the Group. § 315. The most essential characters of the Vertebrata are the possession of a skeleton traversing the longitudinal axis of the body, and the division of the body into a number of metameres (primitive ver- tebra)). They differ from the Tunicata, with which alone of all the divisions of the Invertebrata they have any well-marked rela- tions, by this metamerism. They have more distant relations to the Vermes, but then this group obviously has relations to most of the other phyla. The axial skeleton separates a dorsal from a ventral division of the body. The former contains the central nervous system, the latter the digestive canal, which is continued on from a respiratory chamber, and Yv^hich, with the organs differentiated from it, is embedded in a coclom. Two regions therefore can be made out, which extend along the body; the upper one is neural, and the lower enteric; the chief trunks of the system of canals for the nutrient fluid belong to the latter region. The various divisions are set in order in the following review : A. Acrania. Lcptocardii. Amphioxus. B. Craniota. I. Cyclostomata.* Myxinoidea. Bdellostoma, MyxinO. Petromy zontes. Petroiuyzon. * The Cyclostomata should be placed apart from all the rest of the Craniota, for their whole organisation shows that they branched off very early from the Craniota. VEKTEBRATA. 409 II. Gnathostoniata. a) Ananinia. 1) Pisces. Selachii. Squali. Hexanclms, Heptanchus, Acanthias, Scynmus, Galeus, Scyllium, Squatina. Raja1. Raja, Torpedo, Trygon. Holocephali. Chimera. Dipnoi. Honopneumones. Ceratodus. Dipneumones. Protopterus, Lepidosiren. Ganoi'dei.* Sturiones. Acipeuser, Spatularia. Polypterini. Polypterus. Lepidosteini. Lepidosteus. Arniadini. Amia. Teleostei. Physostomi. Abdoniinales. Clupea, Salmo, Esox, Cyprinus, Silurus, Mormyrus. A p o d e s. ]\Iura3iia, Conger, Gymnotus. Pbysoclysti. Anacanthini. Gadus, Pleuronectes. Pharyngognathi. Belone, Hcmirhamphus, Cliromis, Labrus. Acantliopteri, Perca, Labrax, Trigla, Scropama, Anabas, Mugil, Scomber, Zeus, Trachypterus, Gobius, Cyclopterus, Blennius, Lophius. Plectognathi. Ostracion, Diodon, Ortliagoriscus. Lophobrancbii. Byiignatlius, Hippocampus. 2) Amphibia. t Urodela. Perennibranchiata. Siredon, Menobranchiis, Proteus. Caducibrancbiata. * I regard each of these divisions of the Gano'idei as highly independent. They represent the last shoots of very divergent series of forms, of which that of the Polyp- terini has many points of relationship to the Dipno'i ; the Amiadse, on the other hand, are the nearest allies of the Teleostei (Clupeida'). It would, perhaps, be best to separate them completely from the Gano'idei. The Sturiones show the greatest resemblance to the Selachii. I must regard the Selachii as being nearest to the ancestral form of the Gnathosto- matous Vertebrata. The Holocephali, as well as the Dipnoi and Ganoidei, appear to have brauched off from them, while the Teleostei again are a branching off from the Gano'idei. f The living Amphibia form only a very small group, which in many parts indi- cate considerable retrogression ; but few fossil forms can be safely placed in it. The palasontological records of the Amphibian phylum arc fragmentary in the highest degree. There are many reasons for placing the Archegosauria with them, but yet there are many points in which these forms resemble the Reptilia. 410 COMPARATIVE ANATOMY, u) Anauiuia (continued). Derotremata. Cryptobranchus, Menoponia. SSalamandrina. Triton, Salamandra. Anura. Pelobates, Bonibinator, Hyla, Ceratophrys, Rana, Bufo. Gymnophiona. Coecilia. b) Aniniota. 1) Sauropsida. 1. Reptilia." Chelonii. Sphargis, Trionyx, Chelonia, Chelys, Chelydra, Emy*, Testudu. Saurii. Ascalabota. Platydactylus, Henndactylus. Rhynchocephala. Sphenodon. Lacertina. Iguana, Calotes, Draco, Phrynosonia, Uromabtix, Lacerta, Ameiva. Monitores. Monitor, Psamtnosaurus. Scincoi'dea. Scincus, Seps, Anguis. Chalcidea (Ptychopleura). Chalcis, Zonurus. Chamasleonida. Chamseleo. Amphisbsenida ( Annulata) . Amphisbrena, Lepidosternum. Opbidii.t Eurystomata. Python, Boa, Coluber, Tropidonotus, Dryopbis, Dipsas, Hydro- phis, Crotalus, Trigonocephaly, Vipera. Stenostomata. Typhlops, Uropeltis. Ci'OCOdilini. Alligator, Crocodilus, Ramphostoma. 2. Aves.t Ratita;. Struthio, Droinreus, Apteryx. Carinatse, Gallinacea?. Megopodius, Penelope, Crax, Crypttirus, Lngopus, Tctrao, Tavo, Numida, Callus, Phasianus. * The various divisions of this class appear to be the very divergent terminal twigs of a branch of the Vertebrata, which was in former times largely represented. Many of the fossil divisions which are ascribed to the Reptilia, such as the Enaliosauria, apparently, however, branched off from the Vertebrate phylum before the Amphibia. In another group of fossil Saurii we find forms intermediate between the Eeptiliaand Birds ; and that most markedly in the characters of the skeleton of the foot. These are the Ornithoscelida. By uniting Reptiles and Birds into one division of the Sauropsida, as Huxley has done, we put these relations in their proper light. f The Ophidii form a division nearest to that of the Saurii, and derived from it, and, With it, equivalent to the Chelonii or Crocodilini ; and they were thus put together by Stannius as Streptostylica. % The class of Birds which arose from reptilian forms is one which is divided by the most important points in its organisation into groups which diverge very slightly from one another, for the characters of these subdivisions present distinguishing points of very slight importance in comparison with those of the other divisions of the groups of the Vertebrata, " Through the Saururi (Archa-optci vx) they arc directly connected with the Ornithoscelida. VEETEBEATA. 411 b) Aniniota (continued). Columbse. Colurnba. Grallatores. Otis, Dicbolophus, Grus, Ardea, Ciconia, Vanellus, CliaraiU'ius, Scolopax, Fulica, Gallinula, Rallus. Natatorca (Palmipedes). Procellaria, Sterna, Larus, Phaeton, Plotus, Pelecanus, Carbo, Anser, Anas, Cygnus, Pbcenicopterus, Mormon, Uria, Alca, Aptenodytes. Pas seres (Insessores) . FringiUa, Alauda, Tardus, Sylvia, Motaeilla, Parus, Muscieapa, JLanius, Sturnus, Oorvus, Hirundo, Uertbia, Troebilus, Upupa, Merops, Coracias, Alcedo, Buceros. Picides. Picus, Yunx. Psittacides. Psittacus, Strygops, Nestor. Rapaces. Gypogeranus, Falco, Buteo, Aquila, Gypaetus, Vultur, Catkartes, Harpyia, Surnia, Strix. 2) Mammalia. Ornithodelphia(Monotremata). Ornithorhyncbus, Ecbidna. Didelphia- (Marsupialia). Botanopbaga. Halmaturus, Dendrolagus, Phascolomys, Phascolarctus, Pbalan- gista. Zoopbaga. Perameles, Dasyurus, Thylacinus, Didelpbys, Cbironectes. Monodelphia (Placentalia). Edentata.t Myrmecopbaga, Manis, Cblamydopborus, Dasypus, Bradypus. Ungulata. Artiodactyla. Sus, Dicotyles, Moschus, Camelopardalis, Cervus, Antilope, Capra, Ovis, Bos. Tylopoda. Camelus, Aucbenia. Perissodactyla. Tapirus, Rbinoceros, Equus. Sirenia. Manatus, Halicore* Prosimii.J Stenops, Lenrar, Otobcnus, Titrsius, Galeopitbccus, CMromys; Rodentia. Sciurus, Arctomys j Mus, Hypud;eus, Crieetus, Georbychus, Spalax, Pedetes, Dipus, Lagostomus, Myopotamus, Castor^ Hystrix, Ccelogenys, Cavia, Lagomys, Lepus. Proboscidea. § Elepbas. Lamnungia.H * I regard the division of the Marsupialia as equivalent to the monodelphous Mammalia, for not only are there found in it representatives of most of the orders of Monodelphia, but, further, there are in the Monodelphia many indications which jjoint to their having arisen from a didelphous form. The Marsupialia. or uniting with them the Monotremata, the Implacentalia, consecpiently represent the ancestors of the Placentalia. f The great variations which the relations of the placenta present in various Edentata weaken somewhat the value of the placental classification, although the various orders are generally distinguished by the similar characters of their placenta. X The Prosimii form a stem-group, in some divisions of which peculiarities are retained which are found in various other of the following orders. Thus there are characters which we meet with in Insectivora, Eodentia, Carnivora, and Primates. § and || The Proboscidea and Lamnungia are representatives of orders which it is very difficult to associate with the rest. They have genetic affinities to the Kodentia; Hyrax has also relations to the Ungulata. 412 COMPARATIVE ANATOMY. b) Amniota (continued). Hyrax. Fera. Carnivora. Felis, Hy.Tun, Proteles, Oanis, Herpestes, Vivcrra, Lutra, Mustela, Meles, Nasua, Procyon, Ursus. Pinnipedia. Phoca, Otaria, Trichechus. Cetacea/1 Delphinus, Physeter, Balamoptcra, Balajna. Insectivora. Chrysocliloris, Talpa, Sorex, Myogale, Erinaceus. Chiroptera. Pteropus, Ehinolophus, Glossopliaga, Vespertilio, Vesperugo. Primates. Hapale, Callithrix ; Ateles, Mycetes, Cebus ; Cynoeephalus, Inuus, Cercopithecus; Troglodytes, Hylobates, Pithecus; Homo. Bibliography. Owen, R,, On the Anatomy of Vertebrates. Vols. I.— III. London, 18CG-G8.— Huxley, T. H., A Manual of the Anatomy of Vertebra ted Animals. London, 1871. Leptocardii : Muller, J., Ueber den Ban und die Lebenserscheinungen des Branchiostoma lubri- cum. Abhandl. d. Berl. Acad. 18 41. — Goodsir, Transact. Royal Soc. of Edinburgh. T. XV. i. — Quatrefages, Ann. sc. nat. III. rv.- — Kowalevsky, A., Entwickl. des Amphioxus. Mem. Acad, des Sc. de St. Petersbourg. Sc'r. VII. T. XI.— Rolph, Unters. iiber d. Ban d. Amphioxus. Morphol. Jahrb. II. Cyclostomata : Muller, J., Vergl. Anatomie der Myxinoiden. Abhandl. d. Berl. Acad. 1835-45. — The same, Ueber den Bau und die Gronzen der Ganoi'den. Ibid. 1846. — Rathke, Bemerkungen iiber den inneren Bau der Pricke. Danzig, 1825. — The same, Ueber den Bau des Querders. Bcitr. z. Gesch. der Thierwelt. IV. Halle, 1827.— Langerhans, Unters. iiber Petromyzon Planeri. Freib. 1873. Pisces : Monro, A., The structure and physiology of Fishes. Edinburgh, 1785.— Cuvier et Valenciennes, Hist. nat. des poissons. 'XXII. vols. 1828-18.— Agassiz, Rechcrchcs sur les poissons fossiles. 5 vols. av. Atlas. 1833-13.— Agassiz et Vogt, Anatomie des Salmones. Neufchatel, 1845.— Leydig, Beitiiige zur mikroskop. Anatomie und Entwickelungsgeschichte der Rochen und Haie. Leipzig, 1852.— Owen, Description of Lepidosiren annectens. Transact. Linn. Soc. XVIII.— Bischoff, Lepidosiren paradoxa. Leipzig, 1840.— Hyrtl, Lepidos. parad. Abhandl. der bohm. Ges. d. Wiss. 1845.— Peters, Lepidosiren. Arch. f. Anat. u. Phys. 1845.— Gunther, Ceratodus. Phil. Trans. 1871.— Balfour, F. M., Development of Elasmobranch Fishes. Journal of Anat. Vols. X. XL Amphibia : Cuvier, in Recueil d'observations de Zoologie et d'Auat. comp. I. 1805.— Ruscont et Configliachi, Del Proteo anguineo di Laurenti monografia. Pavia, 1818.— The same. Hist, naturelle, developpement et metamorphose de la Salamandre tcrrestre. Pavie, 1854.— Muller, J., Beitriige zur Anatomie der Amphibien. Zeitschr. f. Physiol. IV. 1832.— Duges, Recherches sur l'ostoologie et la myologie des Batraciens. Paris, 1834.— Mayer, Zur Anatomie der Amphi- bien Analecten f. vergleichende Anatomie. Bonn, 1835.— Calori, Sulla Anatonna del Axolotl. Mem. della Accademia dclle sc. dell' istituto di Bologna. III. 1851.— Rathke (Coecilia annnlata). Arch. f. Anatom. u. Phys. 1852. p. 331.— Leydig, Untersnchungen iiber Fische und Reptilicn. Berlin, 1853.— Vaillant', L. (Siren lacertina). Ann. sc. nat. IV. xviii.— Fischer, J. G., Percnni- branchiaten u. Derotremen. Hamburg, 18(54.— Hyrtl, J., Cryptobranchus japonicus. Vindo- bonffi 1805.— Van der Hoeven, Ontleed-en deerkundige Bijdragen tot de Kenniss von Mcnobran- chus.' Leyden, 1807.— Wiedersheim, R., Salamnndra perspicillata, etc. Genua, 1875.— Gotte, Entwickelungsgeschichte der Unke. Leipzig, 1875. Beptilia : Bojanus, Anatomia testudinis europseee. Vilnse, 1819.— Dumeril et Bibeon, Erpt'tologie generate. Paris, 1834-54.— Duvf.rnoy (Serpens). Ann. sc. nat. I. xxx.— Rathke, Entwicke- lungsgesch. der Natter. Konigsberg, 1837.— The same, Entwick. der Schildkroten. Braunschw. 1848— The same, Ueber die Entwick. und den Korperbau der Krokodile. Braunschw. lsoo. — Calori (Urornastix) . Mem. della Accad. delle sc. dell'ist. di Bologna. III. ii. 1863.— Gunther, (Hatteria), Phil. Tr. 1867. II.— Leydig, Die in Deutschland lebenden Arten der Saurier. Tiibingcn,' 1872.— Wiedersheim, P., Z. Anat. v. l'liyllodactyhis europ. Morph. Jahrb. I. Aves- Tiedemann, Anatomie und Naturgesch. der Vogel. Heidelberg, 1810-14.— Owen, On the anatomy of the southern aptcryx. Transact. Zool. Soc. Vols. II. III.— The same, Art. Aves in * The Cetacea arc apparently of the same phylum as the Pinnipedia, which phylum is derived from the Carnivora ; this is showu by certain fossil forms (Zenglodon). But the peculiarities of the Cetacean organisation are too great for them to bo put under the Fera ; they form the end of a series. FOEM OF THE EODY OF VEETEBEATA. 413 Todd's Cyclopaedia. I. — Milne-Edwards, ALPH.,-Recli. sur les ossemens fossiles dcs oiseaux Paris, 1806.— Alix, E., Appareil locomotcur des oiseaux. Paris, 1374. Mammalia : Meckel, J. Fr., Omithorhyncbi paradoxi descriptio anatomica. Lips. 182G.— Veolik (Dendrolagus), Verbandel. d. Kon. Acad. Amsterd. V.— Guklt, Handb. d. vergl. Anat. dcr Haussiiugethiere. 4 Aufl. Berlin, I860.— Fbanck, L., Anatomie der Hausthiere. Stuttgart, 1871. — Brandt (Lama), M^m. Acad. St. Petersbourg, 1841.— Owen (Giraffe), Transact. Zool. Soc. II.— The same (Rhinoceros), Transact. Zool. Soc. IV. n.— Milne-Edwabds, Alph. (Moschiaxl), Ann. sc. nat. V. ii.— Murie, J. (Manatus). Tr. Zool. Soc. VIII.— Camper, Observations sur la structure intime et le squelette de Cetac6es. Paris, 1820.— Rapp, Die Cetaceen. Stuttgart u. Tubingen, 1837. — Vrolik, W., Natuur- en ontleedkund. Beschouwing van d. Hyperoodon. Haarlem, 1848. Eschricht, Untersucb. iiber die nordischen Waltbiere. Leipzig, 1849.— Murie, J. (Globio- cephalus, Otaria, Trichechus), Tr. Zool. Soc. VII. VIII.— Rapp, Anatom. Untersucbungen iiber die Edentaten. 2 Aufl. Tubingen, 1852.— Owen (Myrmecophaga jubata), Tr. Zool. Soc. IV.— Hybtl (Chlamydopborus truncatus), Denkschr. d. Wien. Acad. IX. 1855.— Pouchet, G., M), in which the glands open, raised up into a papilla or nipple ; at the tip of which a number of gland-ducts always open. In the other arrangement the mammary pouch is persistent. Owing to the continued elevation of the periphery of the gland (a) the glandular area is more and more depressed, the edge of the mammary pouch is developed into a pseudo-nipple, from the tip of which a single canal passes to the glandular area (G). This arrangement has been observed in some of the Ungulata. Intermediate stages between the two arrangements can be made out in the Marsupialia (Halmaturus) andRoclentia (Murina). The number of mammary glands which are distinguished by their nipples varies in different divisions. They generally correspond to the average number, or to the maximum number of young produced at one birth. They vary even within the limits of the same order ; and they also vary in position. As a rule they form two rows, which, when there is a large number present, extend from the inguinal to the pectoral region (Carnivora, Suina). In many of the Didelphia they are arranged in a circle on the abdomen. When the number is not so large, they either occupy an abdominal position, as in many Didelphia, or they are only found in the lumbar region (Perisso- dactyla, Ruminantia, Cetacea), or, finally, they are limited to the pectoral region (Elephant, Sirenia, many Prosimii, Chiroptera, and Primates). When more than one pair is present some glands are sometimes aborted, so that there are rudimentary organs present, together with well-developed and functionally active ones ; and these may be recognised by their rudimentary nipples. In a similar way the whole apparatus is atrophied in the male. The most important adaptation of the integument to the function performed by the mammary glands, is the formation of the folds of integument found in the Marsupialia; these form a sac, the marsupium, which encloses the mammiferous region of the ab- domen. The extent of its development appears to vary inversely with the extent to which the young are developed at birth. Dermal Skeleton. § 324. The function of the integument as a defensive organ for the body is increased in value by the formation of hard structures. When these parts are of some size they give rise to a dermal INTEGUMENT OF VERTEBRATA. 423 skeleton. In many cases we know but little as to tlie development of these structures, but they may all be reckoned among osseous £ Fig. 218. Vertical section through the skin of an Embryonic Shark. C Corium. c, c, c Layers of the corium. d Uppermost layer, p Papilla. E Epidermis, e Its layer of columnar cells, o Enamel layer. formations, to which indeed they completely correspond in the higher divisions. The dermal denticles (placoid scales), which are distri- buted over the whole of the in- tegument in the Selachii, may be regarded as the structures from which the various forms have been derived. In them we distin- guish a basis, which is inserted into the corium, and is ordinarily rhom- boidal in shape, and a portion, which stands out from it, and which ordinarily has its apex directed obliquely. This is covered over by the epidermis. In some parts, as, for example, on the head, they often have a bombous surface, and are set irregularly ; while on the trunk they are generally set in per- fectly regular and obliquely running rows (Fig.219). They are developed on papillas of the corium (Fig. 218, p), which are covered over by a layer developed from the epider- mis : this secretes an enamel-like substance on the projecting por- tion of the papilla, while the body of the papilla is ossified from the tip downwards. The epidermis and corium, therefore, both share in the formation of these structures. There is a central cavity in the Fig. 219. Dermal denticles of Centrophorus calceus (a little magnified). 424 COMPARATIVE ANATOMY. papilla, whence fine branched canals radiate out to the surface. The placoid scale has therefore the structure of dentine, is covered by enamel, and is continued at its base into a plate formed of osseous tissue ; as they agree with teeth in structure they may be spoken of as dermal denticles. In the Rays these structures have altogether disappeared (Electric Ray), or are replaced by larger structures, which are grouped together in the form of spines or larger bony teeth, or are separate from one another (Spiny Rays). The dermal denticles of the Shark are very generally converted into larger bony plates in the Gano'idei ; in the Rhombifera they have not only the same arrangement on the body, but have essen- tially the same minute structure. In the Sturiones larger bony plates alternate with smaller ones. They generally retain exactly the rhomb-form, which is lost in the rest of the Gano'idei (the Cyclifera). The common flat and thin scales of the Teleostei follow on here. They appear to differ in many points from the ganoid scales, and re- present an offshoot from the type, which obtains in the Gano'idei, and which may be derived from the Selachii ; this offshoot is characterised by its variety of form. In many Teleostei the scales undergo complete degeneration. On the other hand they give rise to parts which differ somewhat from scales, and which are formed by the fusion of the dermal denticles ; such are the plates and spines of the Plectognathi, where the plates may become more firmly united together and form a connected carapace (Ostraeion, Lophobranchii). Parts which are likewise formed from the concrescence of dermal denticles are found in the integument which covers the appendages of the GanoTdei and Teleostei. To compensate for the atrophy of the internal or primary skeleton of the limbs, these bony plates form a number of rays, which often branch dichotomously at their ends, and unite to form an organ of support for the fins (secondary skeleton of the fin). The ray which occupies the anterior edge of the fin is frequently massive, or gives rise to a strong spiny ray, which may be connected with the internal skeleton. This ray may not only be larger than the rest of the rays, but it may even, as in certain Siluroids, represent the whole of the pectoral fin. Hertwk;, 0., Ueber d. Baa u. die Enfcw. der Placoi'dsclmppen u. der Ziihnc der Selacliicr. Jon. Zeitschr. Bd. VIII. — The same, Uebcr das Hautskelet der Fische. Morph. Jahrb. II. § ;325. The ossifications of the integument are of special importance in those regions of the body where parts of the internal skeleton come to the surface. These ossifications are developed in just the same way as the bony plates on other regions of the surface of the body, and may likewise be derived from the indifferent stage represented by the dermal denticles. Although the various kinds of dermal bones which are found on the trunk have an importance which is limited to INTEGUMENT OF VERTEBRATA. 425 the fishes, there are others which are of more importance ; these are the bony plates which are definitely arranged, and constantly present, on the head, where they form the earliest rudiments of the bony skull, or, at first, of the roof of the skull (cf. Fig. 220). These dermal bones are inherited by all Vertebrata that are pro- vided with a bony skull, and are connected with other ossifications, which do not appear till later, in the carti- laginous skull. This arrangement is first seen in the Sturiones. There are a number of smaller bony plates in addition to the large ones, but most of these have no general sig- nificance. On account of these rela- tions to the internal skeleton, their more special characters will be ex- pounded when we come to treat of it. Other skeletal parts besides the bones of the skull are derived from ossifica- tions of the integument ; the clavicle, for example, has a similar origin. Lastly, there is another category of bones which are likewise derived from placoid scales; the bones around the mouth have been recognised as having their origin in tooth-bear- ing plates derived from fused placoid scales. § 320. We meet with dermal bones in the higher classes ; in the Amphibia, and also in the fossil Archegosaurii, in which there were dermal bones in the form of scutiform plates. We find only scattered dermal bones in a ru- dimentary form in extant Amphibia. In Ceratophrys there is an osseous shield in the skin of the back; in Brachycephalus there are three which are united to several vertebra?. The bony scales which are pretty generally found in the Coccilia?, and which are set in pouch-shaped depressions, do not appai'ently belong to this set of structures. They are more common in the Reptilia, which so far approach the old Amphibian phylum. In the fossil Teleosaurii, as in the living Crocodilini, there are dermal bones distributed over the whole integument, which form a kind of carapace ; in the Scincoidea we generally meet with interlocking bony plates in the integument. Similar kinds of dermal ossifications in the Chelonii form a special, though well-developed form of, dermal skeleton, in consequence Fig. 220. Head of Acipenser sturio; seen from above, to show the osseous plates covering the cartilaginous cranium, which is shaded dark. I2G COMPARATIVE ANATOMY. of their connection with the internal skeletal parts. They not only form a dorsal shield on the dorsal surface, but a ventral one on the ventral surface (plastron). In the dorsal shield we can make out a median row of bones, which are fused with the spines of the vertebras, and project from them. At the sides there are larger plates, which are fused with rib-like processes, and in addition to these there are special marginal plates around the edge of the shield. These are wanting in Trionyx. Four paired pieces and one unpaired piece can be made out, as a rule, in the plastron. All of these parts are variously developed in different families of the Chelonii. Although the dermal bones of the Eeptilia may probably be rightly regarded as derived from the bony carapace of Fishes, we must regard the ossifications which are found in the Edentata as independent arrangements, which have had their origin in fresh adaptive modifications. <^J Internal Skeleton. § 327. The internal skeleton is of greater morphological importance than the skeletal structures formed from the integument ; it is connected, on the one hand, with arrangements which are found in the Iuver- tebrata, and on the other, and by a long- series of very varied arrangements, it can be followed out through all divisions of the Vertebrata. At first the internal skeleton has the form of a rod-like structure which tra- verses the whole length of the body, and is, when simplest, made up of indifferent cells, and surrounded by a cuticular structure which is formed from a secre- tion of these cells. This primitive organ of support is the chorda dorsalis or notochord ; we have already met with it in the Tunicata (cf. §303). The invest- ment formed by it is the chordal sheath (cs). The earliest rudiment of the noto- chord is placed just below the central nervous system; it has not always the same relations to the germ-layers, al- though it must be derived either directly or indirectly from the mesoderm. The compact, and, in all cases, primitively unjointed condition of the notochord speaks to its having been inherited from an unjointed condition of the organism, and this is what might be supposed from its early appearance in the embryo. Fig. 221 a. Section through the vertebral column of Am- moccctes. Ch Chorda. cs Chordal sheath. vi Spinal chord, a Aorta, v Veins. INTERNAL SKELETON OF VERTEBRATA. 427 Tlie notochord lias always tlie same topographical relations to the most important of the other organs. Above it, is the central nervous system, and below it, the respiratory and nutrient apparatus. Processes are given off from the connective tissue surrounding the chord which enclose the so-called dorsal and ventral cavities ; these processes pass into the musculature of the body, which is thereby broken up into a number of segments, set one behind the other. In Amphioxus these segments are so far asymmetrical that they are found alternately on either side. The low condition of the axial skeleton, which is represented by the chord, is permanent in the Leptocardii, where it merely presents special histological modifications. In all the rest of the Vertebrata the chord is the sole axial skeleton in the earliest stages of development only ; new differentiations appear, and it becomes of less physiological importance. These differ- entiations affect the notochord as well as the tissue which surrounds it, and which has been called the "skeletogenous layer'"' or "skele- togenous tissue/' on account of its relations to the future skeleton. The cells of the chord form a tissue resembling cartilage, and the sheath becomes a more independent por- tion— it forms a supporting organ — owing to the thickening of its layers. Cartilaginous tissue forms around the chord (Fig. 221, b /.■), and that segmentation of the axial skeleton into separate segments, the so-called vertebras, which was before merely indicated, now be- comes obvious. This segmentation of the axial skeleton is an expression of the metamerism of the whole body; a series of these vertebras make up the vertebral column. In each vertebra we call that por- tion which surrounds the notochord the cent- rum, and the outgrowing portions which enclose the dorsal and ventral cavities of the body, and which are given off directly or indirectly from the centrum, the arches. These again are distinguished as upper or lower arches, according to their relations to these two cavities. As the axial skeleton becomes segmented, a well-defined portion in the most anterior segment forms the primitive Cranium in the Craniota. An inferior system of arches, which encloses the most anterior portion of the intestinal tract, which functions as a respiratory organ, is distinguished as the branchial or visceral skeleton. The cranium and visceral skeleton make up the most anterior portion of the whole skeleton — the skeleton of the head. The other skeletal structures which are connected with it are represented by the more throiu Section spinal young Chorda. sheath. Jc Supe- 221 b. gh the column of a Salmon. Ch cs Chordal m Spinal chord, rior, k' Inferior arch (in rudiment), a Aorta. v Veins. 428 COMPARATIVE ANATOMY. or less homogeneous vertebral column, which extends to the caudal end of the body. The upper arches remain in close connection with the centra. Movable girder-like pieces are, however, separated off from the lower arches in the region which encloses the ccelom ; these are the ribs. Lastly, there are the skeletal portions of the appendages which are connected by special organs — the pectoral and pelvic girdles — with the skeleton of the trunk. The cartilaginous stage of the primitive skeleton is found in all of the higher divisions, but in them it has no function after a short time, for it is gradually replaced by osseous tissue, whereby the skeletal parts come to have a greater physiological importance. In correlation with this we note a greater differen- tiation in morphological points. Even in the osseous skeleton, however, the cartilage is of great importance. A modified form of cartilage, which is characterised by the deposit of calcareous matter in it, is also of importance. This form is not only antecedent to the ossification of the parts of the skeleton, which are laid down in cartilage, but — as is seen in that superficial calcification of the car- tilaginous skeleton of the lower Griiathostomata — is also sometimes a permanent arrangement. Vertebral Column. § 328. The separation of the rachis into skull and vertebral column is not completely effected in Amphioxus ; the whole axial skeleton is represented by the notochord. In the Crauiota they begin to be separated. The lowest characters of the spinal column obtain in the Cyclostomata, where the more highly-developed notochord, with its sheath, forms the chief portion of it. Around the sheath there is cartilaginous tissue, which is continued into lateral ridges as well as into the wall of the dorsal canal. This tissue has its origin in the continuous differentiation of the skeletogenous layer, and must not be confounded with the cartilages, which ordinarily form the vertebral segments. Speaking exactly, therefore, the spinal column is not here separated into distinct vertebras ; of which, indeed, there are indications only in Petromyzon, where cartilaginous pieces, which correspond to the superior arches, are enclosed in the wall of the more anterior division of the dorsal canal. We meet also with indications of inferior arches. The notochord also retains its primitive characters in Chimera and the Dipnoi. In the Chimajras circular calcifications of the large chordal sheath point to a segmentation of the notochordal tube, but they arc more numerous than the primitive vertebra), which are merely represented by the arches on the chordal sheath. In the most anterior region they grow round the chord and give rise by fusion to a larger undivided piece. In the Dipnoi a strong tube, VEETEBE/E OF VEETEBBATA. -129 derived from tlie skeletogenous layer, is developed around the primitive sheath, and on this the cartilaginous and superficially ossified arches are set. The axial skeleton of the Selachii is much more highly developed. The rudiments of the superior and inferior cartilaginous arches appear around the notochord; these grow around it and so form cartilaginous circular centra. That part of the cartilage which encloses the chord is marked off from the peripheral part, which is continued into the arches, and the former represents, just as in the Dipnoi, a kind of cartilaginous sheath (skeletogenous chordal sheath), which is deposited on the cuticular sheath. The vertebral column of the Selachii varies greatly in structure according to the mode of growth of tho notochord and its skeleto- genous sheath. B I) cs~ Fig. 222. Diagram of the changes produced in the notochord by the skeletogenous layer (longitudinal sections), c Chorda, cs Chordal sheath, s Skeletogenous layer. v Bodie3 of the vertebra). iv Intervertebral portion. g Intervertebral joint. A The chordal tube, when all its parts are equally well developed (Fishes). B Interver- tebral growth of the chorda. Formation of amphicoelous vertebras (Fishes). 0 In- tervertebral constriction of the chorda by cartilage ; while the rest of the chorda is retained in the vertebrae (Amphibia). D Intervertebral constriction of the chorda (Reptilia, Aves). E Vertebral constriction of the chorda, where part of tho interver- tebral portion is retained (Mammalia). The cartilage sometimes forms a cylindrical tube, in which the vertebras are merely represented by the arches and circular parts of the skeletogenous sheath. The notochord is sometimes developed between the vertebras (Fig. 222, B), and retains its earlier size at tho points where the vertebra (y) and arches were first laid down around it. This arrangement gives rise to biconcave (amphicoelous) vertebras (B), the depressions in which are filled up by the intervertebral chord. This is the way in which the vertebras of nearly all other fishes are formed. § 329. In the Gano'i'dei the vertebral column, when simplest in organisa- tion, resembles that of the Selachii. Just as in the Selachii and 130 COMPARATIVE ANATOMY. Chimaerae special cartilages are intercalated, which aid the superior arches, which are connected with the bodies of the vertebrae, in closing1 the vertebral canal. In the.Sturiones the skeletogenous sheath forms a considerable tube, and the separation of the column into vertebras is only indicated by the superjacent arches. The vertebral column of the other Ganoidei is sharply marked off from this, its lowest form. Amia resembles the Teleostei. A small portion of cartilage is retained at the point where the arches are connected with the centra of the vertebras ; but this is absent in Polypterus, so that in it the arches and the centra are united together by bone. Lepidosteus is the most divergent form, for in it the cartilage becomes constricted between the vertebras. In the cartilage which forms the constrictions, an intervertebral articular cavity is formed, so that the opisthoccelous vertebras articulate with one another. So far they resemble the Amphibia, but, later on, the remnant of the vertebral por- tion of the notochord disappears, and a bony centrum is developed, which is connected, and continuous, with the upper arches. The vertebral column of the Teleostei is cha- racterised by the reduction of the cartilaginous rudiment. This reduction may be seen to be gradual, and may indeed be seen in one and the same vertebral column in certain stages of de- velopment ; where, that is, the cartilage may be seen to diminish in quantity as we go from before backwards. As a rule, four cartilaginous pieces, belonging to the superior and inferior arches (Fig. 221 b, hh1), may be seen around the chord, and these take a certain share in the formation of the arches. They very rarely form complete superior arches. When osseous substance is developed, these cartilages are generally retained in the middle of the centrum, so that on making a vertical section through it we get an obliquely set cross (cf. Fig. 223, Jc7c')} the arms of which are directed towards the bony arches. The notochord is always well developed between the vertebras, so that the centra are amphiccelous. Fig. 223. Vertical section, through the middle of a vertebra of Esox lucius. ch Notochord. cs Chordal sheath. Iclc' Cartilaginous cross. A' Corresponds to the tipper, and 7/ to the lower arches (in rudi- ment), h Osseous transverse process. n Spinal canal. 330. The vertebral column of Fishes can only be divided into two regions, the body and the tail. They are distinguished from each other by the characters of the inferior processes of the vertebra?, while the upper arches arc connected with the vertebras in the same manner throughout; and are generally distinguished by the possession of median (spinous) processes. In the region of the VERTEBRA OF VERTEBRATA. 431 trunk, the lower arches are divided into ribs, and supporting' transverse processes (parapopliyses). In the tail of the Selachii and Gano'idei they are continuously connected with the centrum, and run out into spinous processes, just like the upper arches. In the Teleostei the costiferous transverse processes gradually converge, in the caudal region, and form inferior arches, which are not homologous with those of the Selachii and Gano'idei, although they also form spinous processes. In the Chirnasras, Dipnoi', and many Teleostei, the caudal portion of the vertebral column ends by gradually diminishing in size, but in most fishes it presents great modifications, which are correlated with the development of the caudal fin. These modifications first affect the lower arches, which, in the Sharks, form spinous processes, which are greatly widened out at their ends, and with which the caudal fin, which is most developed in its ventral region, is connected. In many Sharks, and still more in the 1^%/ Fig. 22i. End of the caudal portion of the vertebral column of a young C y pri- ll oid. v Centrum, n Superior; h In- ferior arches (tho cartilaginous parts are dotted), c End of the notochord. d Covering bony lamella, r Bony rays of the caudal fin. Sturiones, this caudal skeleton is differentiated in an unequal fashion. The inferior spinous processes are more largely de- veloped; this is correlated with the degeneration of the superior spinous processes, and of the superior arches of the terminal caudal vertebras; this of course produces an up-turning of the caudal end of the vertebral column; and in this way tho inferior lobes of the caudal fin of the shark get to be terminal in position. In the Teleostei this up-turn- ing affects also the axial portion of the vertebral column. As a number of the terminal centra of this column are generally fused together, and, like their upper arches, feebly or not at all developed, while their inferior arches still persist, the up -turning must be the more marked in proportion as the inferior arches become more numerous and larger than the superior ones. This condition (Fig. 224) is carried still farther by the atrophy of a large number of vertebras, so that nothing remains of them but their inferior arches (Physostomi) . Finally, the vertebrae completely disappear, and the remains of the inferior arches of the caudal region are connected, in the form of vertical plates, with a single vertebra which represents the end of the vertebral column; a style-shaped process (urostyle) of the column is directed upwards, and contains the end of the notochord (Acanthopteri). Supporting organs, formed from the integument, are connected with the parts thus formed by the vertebral column, and they 432 COMPARATIVE ANATOMY. are continued into the caudal fin. In the Selachii the fin-rays are formed by the so-called horny filaments, and in the Ganoi'dei and Teleostei by ossifications. Like the caudal fin, the other unpaired ones have their supporting organs formed partly by the axial skeleton and partly by the integu- meut. In the Selachii jointed pieces of cartilage pass from the spinous processes into these fins, and gradually acquire an inde- pendent significance. In the Ganoi'dei and Teleostei they become distinct bony pieces, which are known as "supports for the fin- rays;" these are quite separate from the spinous processes. They are connected with the fin-rays; these are jointed structures, which' are sometimes made up of separate bony plates, and are sometimes represented by solid bony rods (spinous rays). §331. In the Amphibian vertebras the cartilaginous rudiment of the arches grows around the notochord, and forms constrictions in it by means of intervertebral enlargements (Fig. 222, C). In many of them the notochord is destroyed at these points. In the Anura the notochord remains persistent in the middle of the centrum ; the only exception to this rule is to be found in those forms in which the centrum is developed on the surface of the notochord (Hyla., Bombinator, Pelobates, etc.) ; when the articulating cavities are developed, the articulating ends of the centra are developed from the intervertebral cartilage. These intervertebral articulations are incom- plete in most of the Urodela, where we find the articulating processes, derived from the centra at every stage of development. In the rest of the Urodela the intervertebral cartilage is only feebly developed, so that the notochord is but slightly or not at all constricted by it. It persists all along the vertebral column, and is alternately constricted and widened out in Menobranchus, Siredon, and Menopoma. In the latter the cartilage takes a markedly small share in the formation of the vertebra; indeed, a series, in which the intervertebral cartilage may be seen to undergo gradual degenera- tion, can be followed out from the Salamandrina up to Proteus. In proportion to this degeneration the vertebra is formed by deposits of bony layers, which may even be laid down directly on the chordal sheath itself. No separate rudiments of superior and inferior arches can be seen in the trunk ; they seem to have been fused into a common mass of cartilage. Henceforward, therefore, that arrangement which we saw in the Fishes disappears, and the rudiment of the cartilaginous vertebra is formed of a single piece early in life. The shortening of the hinder end of the vertebral column in the Anura is the cause of the development of a small number of ver- tebrae. When the tail disappears, a long dagger-shaped bony piece, VEKTEI5ILE OF VERTEIJKATA. 433 which is ordinarily known as tlie urostyle (Fig from the rudiments of a few vertebras ; counting than ten vertebral segments can be made out. more in the Urodela ; Amphiuma has as many as 100; Menopoma, 48; Salamandra, 42; and the Ccecilias, about 230. The transverse processes (tr) are small in the Salamandrina ; the anterior ones are generally divided into two segments ; in the Anura they are larger, but not divided. The superior spinous processes are always rudimentary. Articulations between the arches of the vertebras are very common, and are effected by the formation of paired articular processes. The connection between the pelvic girdle and the vertebral column does not only more dis- tinctly mark off the caudal portion from the region of the trunk, but a sacral portion is thus represented by a vertebra, which is generally distinguished (and especially in Pipa) by the size of its transverse processes. Gegenbauk, Unters. uber die Wirbels&ule der Amphibien. 225, c), is formed this then, no more There are many Fie-. Verte- Leipzig, 1861. § 332. 225. bral coluuiu and pel- vis of the Frog. tr Transverse pro- cesses, s Sacral ver- tebra, c Urostyle. il Ilium, is Ischium. / Femur. The rudiments of the vertebral column are developed around the chorda dorsalis of the Sau- ropsida, as of the Amphibia. Arches, which enclose the spinal canal, are given off by the cartilaginous centra. The notochord is also constricted between the vertebras (cf. Fig. 222, B), but the whole of it eventually disappears (except in the Ascalobota). The continuous rudiment is separated into centra in just the same way as in the anourous Amphibia; in the Saurii and Ophidii, the centra are proccelous. In the Crocodilini and Aves the cartilaginous portions of the rudiment, which lie between the centra in the cervical region, are converted into a special intervertebral apparatus. Articular processes extend from the superior arches to the vertebra next in front and behind. They are greatly developed in the cervical region of the Chelonii. The superior spinous processes vary in size, especially in the dorsal region ; in the Crocodilini and many Saurii they are present on the caudal ver- tebras. Transverse processes are either given off from the centrum itself, or quite close to it. They are greatly developed in the dorsal and caudal region of the Crocodile, but much more so in the Chelonii, where they are surrounded by the bony plates of the dorsal shield, which have been developed in the integument. They are seen to be divided into an upper and a lower portion in the Ophidii. In the 2 F 131 COMPARATIVE ANATOMY. Fig. 226. Cervical vertebra of Vultur cinereus. c Centrum.

), which is connected with the lower jaw by a ligament (Ug). Accessory bones are formed from other pieces, which are developed from parts of the dermal skeleton ; the most important of these are the infraorbital bones (cf. Fig. 245, i i i i). They form a curved series around the lower edge of the orbit ; the hindermost piece is attached to the postfrontal, and the foremost to the lateral ethmoid. Some of these bones acquire a considerable size in the Cataphracta (Trigla) . The nasals also belong to this series of bones, in consequence of their inconstant presence ; there are also many other pieces that are connected with the so-called mucous canal system, and which are modifications of scales. Vkolik, A. J., Ueber die Verknocherung u. die Kuochen des Schadels del' Teleostei. Niederland. Archiv f. Zoologie. I. — Parker, W. K. Develop* meiit of the Skull in the Salmon. Philos. Transact. 1873. § 340. In the skull of the Amphibia the primordial cranium is some- times greatly developed. It very frequently, however, loses its roof, and also its floor, owing to the formation of spaces in the cartilage. The palato- quadrate is directly connected with the primordial cranium ; it is attached posteriorly to the auditory capsule of the skull, while anteriorly it forms an arch around the orbits, and either projects freely forwards (as in the Urodela), or is connected with the cranium in the ethmoidal region. Behind, and at the sides, it carries the glenoid cavity. It presents therefore those relations which were seen in the Chimeerae, and in the Dipnoi ; and there are^ 456 COMPARATIVE ANATOMY. indeed, many ossifications in the cranium of the Amphibia which resemble those in the Dipnoi. The primordial cranium gives rise to a few bones only. In the posterior primary region the exoccipitals alone are present (Fig. 246); each of these gives rise to a condyle (co). The next region of the auditory capsule presents large lateral processes, which are attached more exteriorly to the hinder portion of the palato-quadrate. The anterior portion of this segment is provided with an ossification, the prootic. This contains the anterior portion only of the labyrinth, the hinder portion of which is contained in the exoccipital ; it also forms a foramen for the trigeminus. There are sometimes indica- tor!? Fig. 246. Skull of the Frog. A from above. B from below. C from behind. D from the side. In A and B the covering bones are removed from the right half of the cranium, so that the whole of the primordial cranium and its ossifications can be seen, and, in A, the spaces in the roof of the cranial cavity. Pa Fr. Fronto-parietals. Na Nasal. Bs Parasphcnoid. Ty Tympanic. Ft Pterygoid. Bl Palatine. Vo Vomer. J Jugal. Mx Maxilla. Pas Premaxilla. o Exoccipital. Be Prootic. co Occipital condyle. Co Columella, fo Fenestra ovalis. Exits of the nerves : 0 Optic. Tr Tri- geminus. Y'j Vagus. In the lower jaw : da Dentary. a Angular. Art Articular. tions of an epiotic. The fenestra ovalis is a cleft in the region of the labyrinth, which is covered over by a small piece of bone. In the anterior portion of the orbital region there are sometimes more or less extensive ossifications. They either embrace the lateral wall only of the cranium (Siredon), or form a circular piece of bone, to which Cuvier gave the name of " girdle-bone." This bone may extend into the ethmoidal region, and even reach to the base of the nasal capsules. The paired parietal s and frontal s are covering bones. In the Anura these fuse together on either side to form a fronto-parietal (Pa Fr). In front of these, and separated from one another by the frontals, are the nasals, which we now meet with as constant SKULL OF VERTEBRATA. 457 bones for the first time. At the base of the skull is the parasphenoid (Ps), which still retains very much the same characters as it had in Fishes, and in front of this, and in the ethmoidal region, there is a paired bone (vo), which is regarded as the vomer. The palato-quadrate is more simple in character than in Fishes. The whole piece sometimes remains for the most part in a cartilaginous condition. An ossification at the point where' it articulates with the lower jaw corresponds to the quadrate of Fishes. In many, the palato-quadrate is divided into an anterior and a posterior portion (Triton). It is not completely united with the cranium, for there is a distinct articular surface on its lower portion, between it and the cranial capsule (Rana). There are two covering bones on the palato-quadrate cartilage ; the upper one (Ty) is distinguished, in the Frog, by the possession of a strong process, which is directed forwards, and which probably, though by no means certainly, corresponds to the squamosal of Fishes. As it aids in supporting the tympanum, it may be called the tympanic. The lower bone is the pterygoid (Ft), and it extends forwards along the cartilaginous arch. Its anterior end reaches as far as the palatine, which lies transversely behind the vomer (PI). In some of the Amphibia another bone is continued forwards in front of the glenoid cavity ; this is the so-called jugal (quadratojugal). The premaxillas (Px) and maxilla (Mx) appear as covering bones of the primordial cranium ; relations which obtain in many Fishes lead to this condition. The maxilla varies greatly in the extent to which it is developed laterally ; in the Anura it ordinarily extends as far back as the jugal. The premaxilla is connected with the primordial cranium by a process which projects forwards in the middle of the nasal region. These maxillary bones did not primitively bound the opening of the mouth, as the presence of special cartilages (rostral and adrostral) in front of the primordial cranium of the larvae of Anura distinctly shows. In the lower jaw the primordial cartilage is present, as in Fishes, and bony parts are developed in connection with it, which essen- tially correspond to those of Fishes. Parker, W. K., Development of the Skull in the Frog. Philos. Trans. 1871.— Wiedersheim, E., Das Kopfskelet der Urodelen. Morphol. Jahrb. III. § 317. The skulls of the Sauropsida have much in common with one another, whilst they are far removed from the skulls of the Am- phibia, and from those of the Mammalia. The primordial cranium, as a rule, has the roof incomplete, but it is much more completely ossified than in the Amphibia, while the great size of the bones which are developed in, and from, the palato- quadrate cartilage leaves but a small portion of the true cranium 408 COMPARATIVE ANATOMY. Fig. 247. Skull of a Chelonian seen from behind. 1 Basi-occipital. 2 Exoccipital. 3 Supra- occipital. 5 Basi-sphe- noid. 15 Prootic. 17 Quadrate. visible. The greater development of the cranial capsule in Birds is the cause of its parts being more distinctly visible than they are in Reptiles. In the occipital region we can make out the four bones which were present already in Fishes. Of these, the basi-occipital and the exoccipitals take part in the formation of a single occipital condyle. The relation that the bones have to the foramen magnum varies a good deal. In the Chelonii the supra -occipital is continued into a large crest. As in the Amphibia there is a fenestra ovalis in the osseous auditory cap- sule ; and in addition to it there is a fenestra rotunda-, which is closed by membrane. In all Birds and Reptiles the petrosal (prootic) lies in front of the exoccipital; its anterior edge is distinguished by the foramen for the third branch of the trigeminal. Another bone (opis- thotic) unites with the prootic to form the boundary of the hinder portion of the fenestra ovalis ; but it does not remain separate in any but the Chelonii, as in other forms it fuses with the exoccipital. In addition to these, there are several, and in Birds numerous, ossifications which are independent for a short time ; these cannot be definitely regarded as similar to any distinct cranial bones in the rest of the Vertebrata. In Birds all the constituents of the auditory capsule fuse with the neighbouring bones, as well as with one another. In the Ophidii the squamosal (Fig. 249, G Sq) is a projecting bone, which carries the quadrate. In other Reptiles, and in Birds it has the same position, but is more embedded between the auditory capsule, parietals, and post-frontal, and is partly set in the roof of the tympanic cavity. The sphenoidal segment varies in size according to the extent of the cranial cavity. A basi -sphenoid is generally present, as is also the presphenoid, which is ordinarily small ; the parasphenoid is no longer developed. Two bones, however, which are seen at the base of the temporal region in Birds, and which fuse with one another (basi- temporals), may be regarded as parts of a parasphenoid. In Birds there is an alisphenoid, and also an orbito-sphenoid — in the Ratitaa, at least — which represent the side-walls of the skull. The Crocodilini, also, are provided with an alisphenoid. In most of the Saurii, however, there is an interorbital membranous septum, in which indications merely of this bone can be made out. In the Saurii (Lacerta, Varanus, Podinema) there is a bone which extends from the parietal to the jitcrygoid (columella) (Fig. 248, A co) ; in the Chelonii this is represented by a broad plate of bone, which passes straight down from the parietal, and which takes a share in the limitation of the cranial cavity; in Hie Ophidii there is a similar process, which embraces the cranial cavity, and extends on to the frontal. SKULL OF VERTEBRATA. i.VJ The covering bones are : the and Aves), or unpaired (Ophi- dii, Saurii, Orocodilini) (Fig. 248, Fa). The frontal also is unpaired in most of the Saurii, and in the Orocodilini (Fig. 248, B Fr). It is paired in Lacerta and Monitor (Fig. 248, A Fr), and in the Ophi- dii, Ohelonii, and Aves. Iu the Reptilia the postfrontals limit the posterior edge of the orbit (Figs. 248, Ff; 249, BCPf). In the middle of the ethmoidal region there is a considerable remnant of the primordial cranium (Ohelonii). The lateral ethmoids (pre- frontals) bound the anterior edge of the orbit in Reptiles ; in Birds they appear to be connected with the median portion of the ethmoid. The vomer is paired in the Ophidii and Saurii (Fig. 250, Vo). The nasals are almost always absent from its upper surface in the Ohelonii, and in some of the Saurii. A new covering bone which is seen on the outer face of the ethmoidal capsule is Saurii, Orocodilini, and Aves (Fi parietals, which are paired (Ohelonii Pig. 248. Skull of Reptilia; seen from above. A Monitor. B Crocodile. Os Supra- occipital. C Occipital Condyle. Pa Parietal. Pf Postfrontal. Fr Frontal. Prf Pre-frontal. L Lachrymal. N Nasal. Sq Scmamosal. Qj Quadratojugal. Jit Jugal. Q Quadrate. Mx Maxilla. Px Premaxilla. co Columella. the lachrymal ; it is found in most gs. 248, 240, L). § 348. The anterior portion of the primitive palato- quadrate cartilage undergoes atrophy very early, so that the bones which belong to it are partly "developed on the skull itself. The hinder portion of the palato- quadrate persists as the quadrate (Fig. 249, Q). The quadrate is movable in the Saurii, Ophidii, and Aves, while in the Ohelonii and Orocodilini it is firmly united to the skull. The whole complex of bones, which is differentiated in the palato- quadrate cartilage, is intimately and immovably connected with the cranium, while, when the quadrate bone is movable, some, at least, of these bones are also so. Another character is due to the development of the nasal cavity (see also § 413). The bones, which in Fishes are placed at the sides of the base of the skull, extend to the middle 4G0 COMPARATIVE ANATOMY. lino, so tliat the base of the skull is more or less shut off from taking any part in bounding the buccal cavity. The nasal cavities, which in the Amphibia lead into the buccal cavity at the very anterior edge of the skull, have their internal orifices always placed farther back in the Reptilia, owing to the union of the horizontal processes of the maxilhe, palatines, and pterygoids in the middle line, and in front of them. In this way the nasal is more completely shut off from the buccal cavity, and forms an upper cavity, the base of which is the roof of the mouth — that is, is the "hard palate/' These Fig. 249. Side views of Skulls. A Struthio. B Orocodilus. C Python. 01 Kx- occipital. Os Supra- occipital. Pi Pterygoid. Pa? Palatine. Tr Transverse. Col Colu- mella,, fov Fenestra ovalis. S Foramen for the trigeminal nerve. The other letters as in the preceding figures. changes are least marked in the Saurii, Ophidii, and Aves, more so in the Chelonii, and most completely so in the Crocodilini. The parts that form the suspensorium in Fishes (hyomandibular and symplectic) have undergone just the same fate as in tlic Amphibia; or, in other words, they have no longer any relation to the cephalic skeleton. The "columella auris " and the parts con- nected with it are developed from its rudiments, partly a bony, and partly a cartilaginous skeletal bit which has entered into the service of the auditory organ. When the quadrate is movably united with the skull (Ophidii, SKULL OF VERTEBRATA. 461 Saurii, Aves), articulations, more or less well developed, appear in those parts of the maxillo-palatine apparatus which are attached to it. These are not present in the Crocodilini and Ohelonii, where the quadrate is jammed between the squamosal and the bones of the auditory capsule. Sphenodon presents an intermediate arrangement; its skull is on the Saurian type, but the quadrate is firmly united with the pterygoid and squamosal. B § 349. Two sets of bones, which extend f orwards, are attached to the quad- rate, just as in the Amphibia. Nearer the middle line we find — in Birds, Serpents, and Lizards — the pterygoid (Fig. 250, Pt) articulated with the base of the cranium. The two bones are united suturally in the middle line in the Ohelonii and Crocodilini, and are also firmly attached to the base of the skull (Fig. 251, Pt) ; in the latter they bound the posterior nares. In the Ophidii, Saurii, and Crocodilini, there is an ex- ternal pterygoid (Os trans- versum, Figs. 250, A Tr; 251, B Tr) which connects the pterygoid with the maxilla; it is uncertain whether this corresponds to the ectopterygoid of fishes. In front of the ptery- goid lie the palatines (Pal) ; in the Chelonii and Croco- dilini they are united su- turally in the middle line, but in the Ophidii, Saurii, and Aves, they are separated from one another and form the lateral boundaries of the posterior nares (Fig. 250, Pal). In the Chelonian skull the vomer (Fig. 251, A Vo) reaches to the roof of the cavity of tbe mouth and lies between the palatines, while above the nasal cavity the palatines unite together at the base of the cranium. In Birds the palatines are generally long and flat bones (Fig. 250, B Pal), which are connected at their anterior end with a process of the maxilla (Mx)} or fused with a process of the premaxilla. C Fig. 250. View of the base of the skull. A Of Monitor. B Of Struthio. Ob Basi-occipital. C Occipital condyle. Ol Exoccipital. Spb Basi- sphenoid. Q Quadrate. Pt Pterygoid. Tr Trans, verse. Pal Palatine. V Vomer. Qj Quadrato- jugal. Ju Jugal. Mx Maxilla. Mb' Its median process. Px Premaxilla. 462 C0MPAKAT1VE ANATOMY. In most Saurii (and in Chelys among the Ckelonii), as in the Birds, tke premaxillaB are fused together; in the latter they are distinguished by the possession of a long frontal process. Their size is correlated with that of the beak, the form of which is largely due to the share - B that they take in it. Tkey are rudimen- tary in tke Opkidii (Fig. 249, 0 Px) and small in tke Ckelonii. Tke max- illa therefore forms the greater portion of the boundary of the edge of the lower jaw (Mx) ; this bone is of great size in tke Croco- dilini and Saurii, and very large in tke Opkidii, in wkick also it is capable of a large amount of move- ment. There is a lateral series of bones at- tacked to tke quad- rate. Tke first of tkese is tke quad- rato-jugal ; tkis is absent in tke Opki- dii. In tke Saurii it is connected with tke quadrate at tke point wkere tkis bone is united to tke skull. It is continued forwards into a second piece, wkick is partly connected witk a post-frontal, and partly with a jugal wkick passes round tke lower edge of tke orbit. In Birds tke quadrato-jugal (Fig. 250, B Qj) is a slender piece of bone, wkick arises from tke side of tke mandibular joint of the quadrate. In tke Ckelonii and Crocodilini it is connected witk a larger portion of tke quadrate, and supports tke jugal, wkick aids in bounding tke orbit. In all cases tke lower jaw is articulated to tke quadrate, and is made up of tke same parts as in Fiskes, to wkick, kowever, a supra- angular and a complementary bone are added on. In tke Ckelonii and Aves tke two dentaries fuse very early; in Birds tke otker bones retain indications merely of their primitive Fig. 251. View of the base of the skull. A Of Chelonia. B Of Crocodilus. Ob Basi-oceipital. 01 Exoecipital. C Occipital condyle. Spb Basi-sphenoid. Qpo Opisthotic. Pi Pterygoid. Pal Palatine. Vo Vomer. Q Quadrate. Qj Quadratojngal. Tr Transverse. Mx Maxilla. Pa Pre- inaxilla. Pa Parietal. Pfr Post-frontal. Fr Frontal. Ch Posterior nares. E Eustachian tube. SKULL OF VEETEBKATA. 4G3 separation. The two halves of the jaw are movably connected together in the eurystomatous Ophidii. Parker, W. K., Structure and development of the skull iu the Ostrich tribe. Philos. Trans. 1866. — The same, On the structure and development of the skull of the common Fowl. Phil. Traus. 1869. § 350. In the Mammalian skull the cartilaginous primordial cranium is ordinarily developed in the basal region only, and limited to the early stages of development. That part of the skull which is derived from the cartilaginous cranium is in this group also to be distinguished from the parts which are developed from other elements, but it becomes intimately con- nected with them. As a capsule for the brain it is itself larger, and is enclosed by a larger number of bones. It is more distinctly divided into different segments than it is in the lower divisions, but this must be regarded as a secon- dary adaptation (§ 340). In the occipital seg- ment the lateral pieces (Fig. 252, 01) always unite with a part of the basi-occipital (Fig. 253, Ob) to form the posterior occipital condyles, by which they bound the foramen magnum, while superiorly they enclose between them the supra-occipital (Os). This latter may be shut out from the edge of the foramen magnum. The four pieces almost always fuse together, but they may remain long- separate (Marsupialia). In many Mammals (several Marsupials, Ungulates, etc.) the exoccipitals send down long processes (pm) (Paramastoid processes). In the region of the auditory capsule separate ossifications in the cartilaginous portion can be seen in the earliest stages only. They form centres of ossification which partly correspond to the bones in Fishes and Reptiles; these soon fuse into a single piece, the petrosal (Pe), the greater part of which gets to be placed at the base of the cranium as the skull grows out laterally. The lateral portion of the petrosal is overlaid by other bones, which are developed from the metamorphosis of the branchial skeleton, and becomes converted into the middle wall of the tympanic cavity, in which there is a '/JOT Fig. 252. Lateral view of the cerebral portion of the Skull of a Goat. 01 Exoccipital. Os Supra-occipital. Jp Interparietal. Pa Parietal. Pe Petrosal. Sq Squamosal. Ty Tympanic. Spb Basi-sphenoid. As Alisphenoid. Ors Orbito-sphenoid. Fr Frontal. Na Nasal. L Lachrymal. Ju Jugal. Mm Maxillare superius. Pal Palatine. Pb Pterygoid, pm Para- mastoid process, si Styloid process. 464 COMPARATIVE ANATOMY. fenestra rotunda as well as a fenestra ovalis. The posterior portion of the petrosal, which is likewise ossified from an independent centre, is attached to the sides of the exoccipitals, and is distinguished as the mastoid portion, in consequence of its carrying the mastoid process in Man. Superiorly, the squamosal (Sq) is attached to the petrosal, and is sometimes fused with it to form the temporal bone, of which it then forms the "squamous portion." In some forms it is completely shut out from the cranial cavity; in others a small portion only can be seen from the inner surface of the skull (Cetacea, Ruminants). It is in the Primates only that this portion is of any great size, and this leads to the well-known arrangement which is seen in Man. A process (zygomatic process) of the squamosal which is directed forwards, takes part in the formation of the jugal arch. In front of the temporal region is the sphenoidal region, which is made up of two perfectly developed segments. The basal piece of the hinder one (basi-sphenoid) (Fig. 253, Sj>h) abuts directly on the basi- occipital, and carries at its sides the alao tempdrales (ali- sphe- noids). Thepresphe- noid with its alae orbi- tales (orbito-sphenoid) (Ps) lies in front of the basi - sphenoid. The two median pieces are separate through- out life, or for a long- time, in Mammals. In Man they fuse very early to form the so- called body of the sphenoid. In the roof of the skull we again find the well-known covering pieces, which increase in extent as the cranial cavity grows larger. The parietals (Figs. 252, 253, Pa) are often fused together (in the Monotremata, many Marsupialia, the Ruminantia, and Solidungula). Between them, a special bone, which marks off the supra- occipital, projects from behind forwards; this interparietal is generally, as in the Primates, fused with the supra-occipital (Figs. 252, 253, Jp), but, at times, with the parietals (Rodentia and Ruminantia). The frontals (Fr) are attached to the ala3 orbitales; they are always paired ; they are fused in some, as in Elephas, Rhinoceros, and also in the Prosimia), Insectivora, Chiroptera, and Primates. The most anterior segment of the primordial cranium presents the most important modifications. It is developed into the wall of the nasal cavity owing to the formation of various laterally and Fig. 253. Vertical median section through the same skull. Ob Basi-occipital. Ps Presphenoid. Eth Eth- moid (vertical plate of the cribriform bone, the anterior edge of which is continued into the cartilaginous par- tittons between the nares, which is not seen in this figure). Eth Turbinate bones. Vo Vomer. s/Frontal sinus. The other letters as in the preceding figure. , SKULL OF VERTEBRATA. 465 inwardly projecting processes. Below it, lie tlie skeletal parts of the maxillo-palatine apparatus, and to these a median lamella of cartilage, which forms the wall of partition of the nasal cavity, is sent down. The vomer is developed as an investing bone on this plate (Fig. 253, Vo). Two ethmoidal pieces are formed by the ossification of the two lateral halves of the ethmoidal cartilage, and of the lamellar processes (turbinate bones), which are given off from it. They bound a portion of the cranial cavity in front of the pre- sphenoid, and are fenestrated to allow for the passage of the olfactory nerve. In Ornithorhynchus there are only two orifices in this region, but they are many more in all other forms, so that this portion is converted into the cribriform plate. An unpaired bone is formed by the fusion of the two lateral halves with the median piece (lamina perpendicularis) (Fig. 253, ffih). The turbinals vary greatly in character, and by the development of multi-ramified lamella? aid in increasing the size of the nasal cavity. The ethmoidal segment is, as a rule, covered by other bones, and especially by those which form the maxillo-palatine apparatus ; and that to such an extent that no part of it at all can be seen from the outside. Except in some Edentata, it is in the Primates only that a portion of the lateral surface reaches to the median boundary of the orbit, where it forms the " lamina papyracea." The lachrymals and the nasals form the investing bones of the outer surface of the ethmoidal region. The former (L) are less constant, and often seem to pass into the neighbouring bones, so that they, as for example in the Pinnipedia, are absent as separate pieces. They are wanting, also, in the Delphinoidea. As in Reptiles and Birds they form part of the anterior wall of the orbit, and also appear on the facial surface of the skull, from which they extend backwards to the median wall of the orbit, in the Primates. As to the nasals (No), they exhibit merely subordinate variations, which are expressed partly by degeneration (Cetacea), partly by a considerable increase in size. Their size is proportional to that of the nasal cavity, and is correlated with an elongation of the facial portion of the skull. They are small in the Primates. § 351. The most important peculiarities in the Mammalian skull are seen in the part which is developed out of the primitive branchial skeleton. A bone which corresponds to the quadrate lies on the outer surface of the auditory capsule. It forms an auditory ossicle, the incus. The skeletal parts, which are developed in front of the quadrate, and along the base of the skull, are intimately connected with the cranium. The pterygoids (Fig. 253, Pt) are generally flat pieces of bone, which are placed on the inner surface of the large processes which 2 H m COMPAEATIVE ANATOMY. are developed from the basi-sphenoid. They form the side-walls of the posterior nares, and may also limit these orifices below by uniting together in the roof of the palate (in Echidna, Dasypus, and some Oetacea). In most Mammals they ai*e permanently distinct, as they are also, for a very long time, in the Primates, before they unite with the above-mentioned processes of the sphenoid to form the medial lamella3 of the descending pterygoid processes. The palatines generally form the inferior boundary of the posterior nares, and the hinder portion of the hard palate. The ruaxillre vary in length according to the extent of the facial region, and always form the largest portions of the upper jaws. The premaxilla3 vary more considerably ; as a rule, they also take part in the forma- tion of the side walls of the nasal cavity. They are rudimentary, or, as compared with the maxillas, feebly developed in many Chiroptera and Edentata. They bound the foramen incisivum. In the Apes they fuse with the maxillas ; this union takes place so early in Man that their existence was justifiably doubted for a long time. The outer series of bones, which is present in the Sauropsida, and which extends from the quadrate to the maxilla, is reduced in Mammalia to the jugal; this bone unites the jugal process of the squamosal with the maxilla, and so forms the jugal arch. In a few forms the jugal is absent (Sorex), or, though united with the maxilla, does not reach the jugal process (Myrmecophaga, Bradypus). When it is united with a process of the frontal it gives rise to a posterior wall for the orbit, and so separates this region from the temporal fossa ; there are various stages of this arrangement. This process is most com- plete in the Primates, where the orbital fissure represents the re- mains of the wide communication which exists between the orbit and the temporal fossa in other Mammals. In the Mammalia a tympanic bone is developed on the outer face of the petrosal ; this serves as a support for the tympanic membrane. It is not certain that this bone is homologous with the one of the same name which we found in the Amphibia. At first it always forms a bony incompletely-closed ring (annulus tympanicus) (Fig. 254, at), which grows out into very various forms. In the Monotremata and Marsupialia, as well as many Insectivora, etc., it is never more than a simple ring. In many forms it is never united to the petrosal; in the Oetacea it is very loosely so. In many it forms a bony capsule which is continued into the external auditory meatus. A bulla of this kind is most common in the Marsupialia, Rodentia, Ferai, and in the Artiodactyla. In some Marsupials, where the tympanic does not pass beyond the annular condition, there is an apparently similar bulla, but this is formed by an extension of the bases of the alas temporales (Dasyurus, Petaurista, Perameles). When the tympanic is fused with the petrosal and squamosal it takes part in the formation of the temporal bone (Primates). SKULL OF VEETEBKATA. 4G7 in vesting § 352. Very early in development, the primitive cartilage of the lower jaw turns off from that line of differentiation which obtains in the rest of the Vertebrata. The part, which, in others, forms the articular bone is converted into one of the auditory ossicles, the malleus (Fig. 254, m); the Meckelian cartilage (p), which is never developed beyond the cartilaginous stage, is continuous with it. The dentary forms an bone on the outer surface of this car- tilage. It abuts in the middle line on its fellow of the opposite side, and unites with it to form the lower jaw ; this is articulated to the skull on the lower surface of the jugal process of the squamosal. It repre- sents therefore a new formation, though the primitive one has not disappeared, but per- sists in other relations. Meckel's cartilage (p) is retained for some time longer on the inner sur- face of the lower jaw, but then disappears; the only portion of it which persists is the part which is placed within the tympanic cavity, and which extends to the Glaserian fissure, where it is ossified to form the processus folianus mallei. The early differentiation, and the, at first, relatively large size of the auditory ossicles, shows that they must be regarded as skeletal parts, which in a lower stage were much more developed in size. The two halves of the lower jaw are permanently separate in a large number of Mammals ; in others they unite early (Perissodactyla, Chiroptera, Primates). Low morphological conditions are implied by the straight mandibles of the Monotremata, in which there is no distinct coronoid process ; in some others, also, this process is merely indicated (Cetacea). The piece which is developed from the upper portion of the primitive hyoid arch (hyomandibular of Fishes) appears to form the rudiment of a third auditory ossicle — the stapes. ma Fig. 254. Lateral view of the skull of a human foetus, with its auditory ossicles. Part of the upper wall of the tympanic cavity and the tympanic mem- brane have been removed, at Tympanic ring from which a piece has been removed superiorly, m Mai- leus. ma Manubrium of the malleus, p Meckel's process extending along the inner side of the lower jaw. i Incus, s Stapes, st Styloid process. 1st Stylohyoid ligament extending to the anterior cornu of the hyoid. t Mastoid foramen. 2 n 2 4G3 COMPAEATIVE ANATOMY. Branchial Skeleton. § 353. A ventral system of arches is connected with the most anterior portion of the axial skeleton, and forms the organs of support for that portion of the alimentary canal which functions as a respiratory cavity. The number of arches, and the backward extension of this apparatus, depends on the size of this respiratory cavity. We meet with two very different types of these structures. The first type is found in the Acrania (Amphioxus). In this framework there is a cartilaginous arch around the mouth — that is, in its most anterior portion ; this arch is beset with cartilaginous rods which are directed forwards. The rest of the apparatus is formed of a homogeneous substance, which forms, as in Balano- glossus (cf. § 112), a complicated lattice-work. The branchial bars of either side are independent of those on the other ; that is to say, they are not united along the ventral line. We cannot derive the second type, which obtains in the Craniota, directly from this. In its earliest stage it is made up of cartilaginous pieces only ; these do not form so lai'ge a number of arches as exist in Amphioxus, and are, while completely symmetrical as regards their arrangement, united ventrally by a copula. In the Cyclostomata the branchial skeleton is made up of com- plicated cartilaginous bars, which are connected inferiorly with one another, as well as with either side of the spinal column superiorly ; owing to their superficial position they may be spoken of as forming an external branchial framework. Very evident signs of this are retained by the Selachii, but in them there is another, or internal, organ of sup- port; and this is found in all the rest of the Verte- brata. The various arches pre- sent indications of their primitive similarity ; this disappears in consequence of the gradual change in their functional relation, which is due to a divi- sion of labour. We obliged to speak of of these arches in with the Fig-. 255. Skull and branchial skeleton of a Selachian (Diagrammatic), a b c Labial cartilages. I Mandibular arch, o Upper, u Lower portion. II llyoid arch. Ill — VIII Branchial arches. mg were some deal- cranium ; so that now they need be but briefly considered. The first of them surrounds the entrance to the alimentary canal, and is divided into two pieces ; one, superior, the palato- quadrate (Fig. 255, o), and the other inferior, the primitive lower jaw (//). The succeeding pairs BRANCHIAL SKELETON OF VERTEBRATA. 4G9 of arches either retain their primitive function of being supports for the branchial arches, or undergo a number of modifications. All these arches evidently had the same original function. Their relation to the respiratory apparatus has not disappeared in the first pair only, which has been converted into jaws ; the hinder arches also have gradually lost their functional and anatomical characters ; it is, therefore, reasonable to suppose that this is but the last of a series of reductions, which first commenced on a much larger number of arches. If this is so, the branchial skeleton of the Craniota is the remnant of an apparatus, in which there were primitively a far larger number of arches. This view is supported by a comparison with Amphioxus, as well as by the considerations to which we are led by a study of the branchial apparatus and of the peripheral nervous system. As we pass through the Fishes to the Amphibia we may note how this apparatus gradually loses its primitive relations ; while in the Reptilia and all higher forms it has no relation at all to the respiratory organs. § 354. All the branchial arches are united ventrally by azygos pieces— the copula?. The various arches are always segmented into a number of pieces, which are generally movably united with one another. The upper portion of the hyoid, as above described, as well as of the mandibular arch, enters into relations with the cranium; these arches thus lose all connection with the other arches, with which the lower portion only of the hyoid arch is still connected. The succeeding arches are either slightly connected with, or only indirectly united to, the cranium ; this is effected by their being attached to the base of the skull, or, when their point of attachment is more extended, to the commencement of the vertebral column. In many Selachii the hyoid arch has the same conformation as the branchial ones (Fig. 255, II). As a rule its copula is in- creased in size, and affords a suppoi^t for the tongue. In the Selachii and Chimaeras this arch retains its primitive function of a branchiferous portion of the skeleton. This relation disappears in the Ganoi'dei and Teleostei, where this gill is rudimentary, and the rays of the upper portion, which is converted into the hyomandi- bular and symplectic, are represented by the opercular apparatus (§ 345). The lower portion of the hyoid arch, or true hyoid, has then bony instead of cartilaginous rays (Fig. 256, I r, branchiostegal rays), and a membrane extends between them which covers over the whole of the branchial apparatus. The hyoid arch thus develops an organ of defence for the respiratory apparatus. There are five pairs of arches connected with the respiratory apparatus; occasionally there are six or seven (Notidani). No more than five are ever found in the Osseous Fishes. While the anterior arches (I II III) are always provided with copulas (/ g), 470 COMPARATIVE ANATOMY. the hinder ones (IV V) are united to a single piece (a), and are always degenerated, both in size and number. The last pair of all (VI), which merely consists of a single piece on either side, carries no gills; in the fifth arch also there are often gill-lamellas on one side only ; in Fig. 256. Hyoicl and branchial arches of Perca flnviatilis. I — VI Arches; the first (I) is converted into an organ for the support of the hyoid; the next four (II — V) are branchial arches, and the last (VI) forms the infra-pharyngeal bone, abed Segments of the arches. The uppermost piece (d) forms the supra-pharyngeal bones. r Branchiostegal rays, f y h Copula? (after Cuvier). the last, however, dental structures are more completely developed, so that this piece is often capable of functioning as a masticatory organ. In the Pharyngognathi the rudiments of the last arch, on either side, are fused into one piece. We meet with other modifications of the posterior branchial arches in the Labyrinthobranchiata, and in various Clupeidse ; these are due to the conversion of various segments of the arches into the walls of spaces into which water is received. Just as the hyoid arch of the Selachii is provided with cartila- ginous appendages, so also the succeeding arches are beset with cartilaginous rays which support the walls of the branchial pouch. Even these structures are rudimentary in the Ganoi'dei and Teleostei, where they form fine cartilaginous lamellae, placed between the rows of the branchial folds. § 355. The branchial skeleton of the Amphibia is considerably reduced ; such forms as undergo a metamorphosis have the gills reduced, and BBANCHIAL SKELETON OE VEETEBEATA. 471 4- * Fig. 257. Hyoid, and bran- chial arches of a larva of Salamandra maculosa. b Hyoid arch, c c Supports of the branchial arches. d Appendage of the Copula. present at the same a gradual change in this apparatus. It is retained in the Perennibranchiata, and undergoes slight changes only in the Derotremata. It is made up of four or five pairs of arches ; the first of which, as in Fishes, forms a hyoid arch (Fig. 257, 6). The suc- ceeding arches are united to a com- mon copula. The posterior ones do not severally ex- tend as far as it, but are connected together on either side. In correlation with the reduction of thearches,theco- pulas are increased in size. The only portion which re- mains complete after metamorpho- sis is the hyoid (Fig. 258, b). It is united with the copula («), which is generally of some size, and which is converted into the body of the hyoid. A larger piece of the second arch is retained in the Salamandrina, and a small portion of the third arch ; in the Anura, however, there is a cartilaginous plate, which is made up of all the branchial arches on either side, and which fuses with the copula into one piece. Rod-shaped pieces (columellaB), which are developed from the ends of the primitively paired plates, are attached to this (Fig. 258, c). The changes in the branchial skeleton, which are perceptible when it changes its function, afford a striking example of the great influence which adaptation to external conditions of life exercises on Fig. 258. Hyoid of Bufo cinereus. a Copula, b Cor- nua of the hyoid. c Rem- nants of the branchial arches (after Duges). the internal organisation. § 356. The degeneration which is seen in individuals among some of the Amphibia is an inherited arrangement in the higher classes. Except those parts which enter into the composition of the auditory organ, all the parts which were developed at one time from the large branchial skeleton of Fishes, are converted into that support for the tongue, which is known as the hyoid bone. The copula forms its " body," and to this the rest of the arches are attached, under the form of " cornua." As a rule the remains of two arches — the hyoid portion of the primitive hyoid, and parts of the first branchial arch — are used for this purpose. The simple body, which rarely consists of several pieces, is beset, in Eeptiles, with portions of two or three arches; these are often very rudimentary* They are either single, or divided 472 COMPAEATIVE ANATOMY. into two pieces. The arches are most numerous in the Chelonii, where there are as many as three, and next to these come the Saurii; in the Crocodilini the broad curved body of the hyoid has but a single pair of arches. In the Ophidii the apparatus is reduced to a cartilaginous remnant, and even these remnants of an arch are lost in various forms (Tortrix, Typhlops, etc.). Two pairs of arches can be made out in Birds. The rudimentary first arch fuses to form the so-called entoglossal bone (Fig. 259, 2), pos- teriorly to which lies the true body of the hyoid. The second arch, however, is well developed, and gives rise to the cornua (4 5), which are formed of two large pieces, which generally curve backwards behind the skull, without being directly connected with it. Be- hind the copula there is the remnant of a second one, which forms the hyoid process (3). In the Mammalia two arches are per- manently connected with the single body of the hyoid. The anterior cornua are the largest, and are connected with the petrosal ; they are made of several (three) pieces. When the median piece is merely connected by a ligament, this portion is divided in such a way that the uppermost piece, if connected with the petrosal, as it is in the Orang and in Man, forms the styloid process of this bone ; when this is the case, the remaining portion is formed by the stylo-hyoid ligament, and the rest of the arch is attached to the body of the hyoid as a small, and sometimes even unossified piece. In most Mammals, the posterior cornua, which are always formed of a single piece, are the smaller ; occasionally, as in various Eodents and Edentates, they are altogether wanting. In the Primates they are larger than the remnants of the anterior cornua. They are connected with the larynx, the thyroid cartilage of which is attached to them by ligaments. Fig. 259. Hyoid appara- tus of the domestic fowl. 1 Body of the hyoid (copula). 2 Entoglossal. 3 Hyoid process. 4 An- terior, 5 Posterior por- tion of the cornu of the hyoid. Skeleton of the Appendages. §357. The two pairs of appendages in the Vertebrata, however much they vary in the extent to which they are developed, have their skeleton arranged in very much the same way ; this points to their being homodynamous structures. In this skeleton we may dis- tinguish an arched piece, which lies in the trunk, and which in its lowest condition forms a band of cartilage ; according to the position APPENDICULAR SKELETON OF VEETEBKATA. 473 which, it occupies, it is known as the thoracic (shoulder), or pelvic (hip) girdle. The skeleton of the free appendage is attached to the extremity of the girdle. When simplest, this is made up of cartilaginous rods (rays), which differ in their size, segmentation, and relation to one another. One of these rays is larger than the rest, and has a number of other rays attached to its sides. I have given the name of Archipterygium to the ground-form of the skeleton, which extends from the limb-bearing girdle into the free appendage. The primary ray is the stem of this archipterygium, the characters of which enable us to follow out the lines of development of the skeleton of the appendage. Cartilaginous arches beset with rays form the branchial skeleton. The form of skeleton of the appen- dages may be compared with them ; and we are led to the conclusion that it is possible that they may have been derived from such forms. In the branchial skeleton of the Selachii the cartilaginous bars are beset with simple rays (Fig. 260, a b). In many, a median one is a Fig. 2G0. Diagrams to illustrate the homodynamy of the appendicular skeleton with that of the branchiae, abed Branchial arches of Selachii. e Archipterygium. developed to a greater size. As the surrounding rays become smaller, and approach the larger one (c), we get an intermediate step towards that arrangement in which the larger median ray carries a few smaller ones (d). This differentiation of one ray, which is thereby raised to a higher grade, may be connected with the primi- tive form of the appendicular skeleton; and, as we compare the girdle with a branchial arch, so we may compare the median ray and its secondary investment of rays with the skeleton of the free appendage. We meet with greater difficulties when we come to examine the topographical relations of the appendages. If the comparison of the skeleton of the appendages shows that it is similar to the branchial skeleton, and that therefore it is possible to derive the appendages from branchial arches, we must further suppose that the two appendages were primitively branchial arches, which carried rays, and that they have been differentiated in a different way to the other branchial arches, and have been separated off from the branchial apparatus. The hinder one altered its position more than the anterior one, and this, of course, happened during changes which affected the rest of the organism. The anterior 474 COMPARATIVE ANATOMY. appendage lias still some relations to the head, as is shown by the muscles which are supplied by cerebral nerves, while, in Fishes, it, and its arch, lie just behind the branchial arches. In this respect the hinder appendages are quite independent. They must be supposed to have travelled to a greater distance, if we are right as to the homodynamy, which a comparison of the skeletons leads us to infer. The anterior appendage has, however, clearly also undergone great changes in position; this is evident when we note how, owing to the continual increase in the number of the cervical vertebras, it moves farther and farther back as we pass from Fishes to Birds. As there are no facts known to us which point to the formation of new vertebra), which could only be brought about by the intercalation of new metameres of the body, this distinct change in position must be explained as due to the continual retrogression of the appendages ; in other words, we are led to postulate just the same process in its case, as in that of the hind limbs. These considerations merely point to the manner in which it is possible that the appendages were developed, and there are still many questions which cannot be safely answered until a compara- tive examination of the muscles and nerves which belong to the appendages has been made. Gegenbaur, C, Zur Morphol. der Gliedmassen der Wirbelthiere. Morphol. Jahrb. II. Anterior Appendages. Shoulder- Girdle. § 358. In its simplest form the shoulder- girdle is a piece of cartilage, which, in the Selachii, forms an arch on each side united to its fellow along the ventral line, and placed just behind the branchial apparatus. Owing to its attachments to the muscles of the appen- dages, definite sculpturing may be made out on the arch; this is most distinct in the Rays. In the Ganoidei the two halves of the cartilaginous arch are completely divided; a new apparatus is connected with the primary shoulder-girdle, represented by the cartilage; this non-carti- laginous part is formed of bones which primitively belonged to the integument, and in the course of its differentiation up to the Mammalia it plays an important part. We must therefore distinguish the secondary from the primary shoulder-girdle. The latter is always cartilaginous in the Sturiones, though various bony plates of the integument are developed on it ; the two lower ones I have shown to be the clavicle and infra-clavicle, and the two upper ones the supra-clavicles. In the primary thoracic cartilage wider spaces are developed from the canals which are found SHOULDER-GIRDLE OF VERTEBRATA. 475 Fig. 261. Eight half of the shoulder, girdle, and thoracic fin of Gadiis. c Clavicle. a b Supra-clavicles. d Accessory piece, e Coracoid. / Scapula, g Basalia of the fin. h Eays of the secondary skeleton of the fin. iii Selachians. In the rest of the Ganoidei and Teleostei a part alone ordinarily remains cartila- ginous, and the rest is ossi- fied, but the whole piece appears to decrease in size. As a rale two bones (/ e) are developed from it in the Teleostei, and with these, parts of the skeleton of the fin may be closely connected. The clavicle, however, which is small in the Sturiones, has increased in size (Fig. 261, c). It is connected along the ventral median line with that of the opposite side, and by the supra-clavicles (a b) with the skull. The primary shoulder-girdle, in fact, undergoes degenera- tion, and forms a mere appendage to the clavicle, which becomes the chief support of the anterior extremity. A § 359. The clavicle, developed on the carti- laginous shoulder-girdle of Fishes, is reduced in the higher Vertebrata. The primary apparatus, however, becomes of greater importance owing to its con- nection with the sternum, and the greater power of movement possessed by its uppermost (dorsal) portion, which is no longer firmly connected with the axial skeleton. That region of the girdle at which the free limb is connected with it, is distinguished by the formation of a cavity which receives the articular head of the humerus, and divides the primary shoulder- girdle into two parts. The dorsal portion forms the scapula ; the ventral is divided into a hinder piece, the coracoid, and an anterior piece, which is ossified from the scapula, when it is ossified — the precoracoid. Among the Amphibia, the shoulder- girdle of the Urodela forms a skeletal piece on either side ; it is largely carti- laginous, and is only ossified in the region of the glenoid cavity. The widened dorsal end of the scapula, the supra-scapula, is almost always cartilaginous, or has an Fig. 262. Shoulder-girdle; A of a Frog, B of a Chelonian. C of a Saurian, s Scapula. s' Supra-scapula. co Precora- coid. co' Coracoid. d Clavicle, c Episternum. st Sternum. The cartilaginous portions are dotted. 47G COMPARATIVE ANATOMY. independent periosteal ossification. The ossification sometimes ex- tends from the scapula on to the precoracoid. In the Anura the two ventral processes (Fig. 262, A co cd) on either side of the shoulder-girdle are united by their cartilaginous ends, and may also become united in the middle line (liana). In this case there is a foramen on either side, in the ventral portion of the shoulder-girdle. The coracoid (co') is ossified independently, while the precoracoid becomes closely related to the clavicle (d). In the Reptilia, likewise, each half of the shoulder-girdle forms a single piece, which closely resembles in form the same part in the Amphibia. The coracoid, which is generally broad, is not unfre- quently fenestrated (Saurii). A process of the scapula, which is merely indicated in the Amphibia, is converted into the acromion, and unites the scapula with the clavicle (Fig. 202, C d). In the Chelonii the scapula is generally a cylindrical bone (B s), which forms an angle with, and is directly continuous with, the precoracoid (B co) at the glenoid cavity. The end of this precoracoid is connected by a ligament with the cartilaginous end of the coracoid. In the Crocodilini the precoracoid has completely disappeared, so that the scapula and coracoid alone make up the shoulder-girdle. In Birds there is a somewhat similar arrangement ; the small and slightly curved scapula is united to the strong coracoid at the glenoid cavity ; the coracoid itself is, as in Reptiles, attached to the plate of the sternum. The Ratitaj indicate their closer affinity to the Saurii by the presence of a rudiment of the precoracoid. Among Mammals the coracoid is complete in the Monotremata only. In the rest, the only sign of it is the process (coracoid process) which is given off from the scapula, and lies in front of the glenoid cavity ; it is in rare cases only that the sternal end of the coracoid persists. I have discovered it, however, in Sorex and Mus, where it forms a piece of cartilage attached -to either side of the manubrium sterni. The scapular remnant of the coracoid still continues to take part in the formation of the glenoid cavity, but this share decreases as that of the scapula increases, so that at last this latter bone alone forms the support for the anterior extremity, which thereby acquires a greater power of free movement. The primitive independence of the remnant of the coracoid is implied by the presence in it of a special centre of ossification, which persists so long as it is not completely fused with the scapula. In form, the scapula of Mammals resembles that of the Reptiles, but owing to the presence in it of new constituents it differs from the latter in some essential points. In the Monotremata there are indi- cations of a spine, the end of which forms the acromion. In the rest of the Mammalia the lateral edge of this broad piece is developed into a larger ridge, which now, owing to the development of the median ridge also, into a projecting plate of bone, or spina scapulas, marks off a superior and an inferior fossa. The anterior end of the spine is always developed into an acromial process. The most important of the other changes which occur in ANTERIOR EXTREMITY OF VERTEBRATA. 477 it is the enlargement of the base of the scapula, which obtains iu the Ohiroptera and Primates. § 360. Owing to the development of the primary shoulder-girdle the secondary apparatus, which forms the clavicle (§ 358), is either placed completely iu the background, or used for purposes other than those which it had in Fishes. The Auura only among the Amphibia are provided with a clavicle (Fig. 262, A d), which forms an investing bone for the precoracoid. It is seldom separated from the shoulder- girdle, and this separation is never complete in any forms below the Reptilia (B d). In them it forms a bone which connects the acromial process of the scapula with the episternum (B c). In Birds the clavicle has the same relations ; it is small in Dromasus, and absent in all other Ratita3 ; iu the Cariiiatoe, however, the clavicles soon unite into an unpaired bone, the f urcula, and are connected with the keel of the sternum by ligaments (Fig. 234,/). The independent appearance of this portion of the skeleton, the primitive origin of which was that of an investing bone for a piece of cartilage, leads to a histological change in the Mammalia ; the clavicle is in them largely formed from a cartilaginous rudiment, which is similar in many points to all other bones, which are preformed in cartilage. This bone, however, is retained in some Mammals only ; — ■ in those, namely, in which the anterior limbs are capable of a large amount of movement. It disappears so far as to leave no signs of its presence in the Ungulata ; in other forms there are only rudiments of it which are sometimes merely formed by ligamentous bands (Carnivora). Anterior Extremity. § 361. All the varied forms, which the skeleton of the free appendages exhibits, may be derived from a ground-form which persists in a few cases only, and which represents the first, and consequently the lowest, stage of the skeleton of the fin — the Archipterygium. This is made up of a stem, which consists of jointed pieces of cartilage, which is articulated to the shoulder-girdle, and is beset on either side with rays, which are likewise jointed. In addition to the rays on the stem there are others which are directly attached to the limb-girdle (cf. Fig. 260, d). Ceratodus has a fin-skeleton of this form ; in it there is a stem beset with two rows of rays. But there are no rays on the shoulder- girdle. This biserial investment of rays on the stem of the fin may also undergo various kinds of modifications. Among the Dipnoi', Pro- topterus retains the medial row of rays only, which have the form of fine rods of cartilage ; in the Selachii, on the other hand, the 478 COMPARATIVE ANATOMY. lateral rays are considerably developed. The remains of the medial row are ordinarily quite small (Fig. 263, R'), but they are always sufficiently distinct to justify us in sup- posing that in higher forms the two sets of rays might be better developed. Rays are still attached to the stem, and are connected with the shoulder-girdle by means of larger plates (p ins). The joints of the rays are sometimes broken up into polygonal plates, which may, further, fuse with one another ; concres- cence of this kind may also affect the pieces which form the base of the fin (p ms). By regarding the free rays, which are attached to these basal pieces, as belonging to these basal portions, we are able to divide the entire skeleton of the fin into three segments — pro-, meso-, and metapterygium. The metapterygium (mt) represents the stem of the archipterygium and the rays on it. The propterygium (pi) and the mesopterygium (ms) are evidently derived from rays which still remain attached to the shoulder-girdle. The peculiar form of the fin in the Eay is due to the great development of the propterygium ; the arrangement in Squatina leads towards this. One ray is here converted into a support for rays, and forms, by gradually reaching forwards, a stem for the propterygium, just as the metapterygium in the stem of the archipterygium possesses one. The Chimasraa agree in all essential points with the Sharks. Fig. 2G3. Skeleton of the thoracic fin of Acanthias vulgaris, p Basale of tho protopterygium. mt Of the metapterygium. B Median edge of the fin. The line drawn through mt indicates the series which formed the stem of the archipterygium. The dotted lines correspond to the rays, which are mostly arranged at the sides (R B), and are rudimentary only on the medial side (B'). § 362. The skeleton of the thoracic fin in the G-ano'idei may be derived from a condition which is similar to that which obtains in the Shark ; it is the same fin, with the peripheral parts reduced (cf. Fig. 264). In correspondence with this, a few rays only are attached to the stem of the fin (B), and those which are set on the shoulder-girdle are also rudimentary. In the Teleostei the peripheral portion of the skeleton of the fin is still further reduced, and as a rule nothing remains of the primary fin-skeleton except four or five elements which are very similar to one another (Fig. 261, g) ; a very variable number of small and always cartilaginous pieces are attached peripherally to them. These, then, serve as supports for the APPENDICULAR SKELETON" OF VEETEBEATA. 479 Fig. 264. skeleton Primary of the thoracic fin of Aci- penser rnthenus, after the removal of a portion of the secondary skeleton. B Basale of the nieta- pterygium. R Bony marginal ray of the secondary skeleton of the fin, only figured in part. secondary skeleton of the fin-rays (h). Basal pieces can be seen in a few only, and it is difficult to refer these even to their primitive significance. The arrangements which obtain in the Ganoi'dei would lead us to regard the basale of the nietapterygiuin, and the basalia of some of the rays, as being the most constant constituents of these pieces. In consequence of their having the same function they have the same form, so that it is impossible to show that they have any connection with the primary stage, except by referring them back to the skeleton of the Ganoid fin. In many divisions of the Teleostei these pieces undergo great changes, in addition to being diminished in number. They are, for instance, intimately attached to the shoulder- girdle, and immovably connected with the parts of which it is made up (Cataphracti). In this way we are able to make out a con- tinuous series from the well-developed skeleton of the fin in the Selachii to that which is found in the Teleostei ; the most important changes consist in the gradual reduction of smaller or larger parts. Keduction first affects the periphery, and then the base, so that the latter is the most constant portion. The decrease in size which the primary skeleton suffers is made up for by the appearauce of ossifications of the integument, which consist, as in the unpaired fins, of jointed or firm bony rays, and are developed on both surfaces of the fin. Gegenbaur, C, Untersuchungen zur vergleich. Anatomie der Wirbelthiere. II. Leipzig, 1865. § 363. In the skeleton of the fore-limb of the higher Vertebrata we are able to recognise the stem of the archipterygium, with rays attached to one side of it ; no rays are now attached to the shoulder-girdle, the stem only is so attached. The arrangement of the joints of the rays in rows set obliquely to the stem — which is just the same arrangement as that of the primitive rays — is obscured by subsequent transverse jointing, but it can be recognised without difficulty in the lower forms. This jointing gives rise to new pieces ; transverse rows of the joints of the rays, as well as the corresponding joints of the stem, being developed into longer pieces. This change is due to a change in function, in consequence of which the ap- pendage is converted from a swimming organ into a compound system of levers. In Ichthyosaurus among the Enaliosaurii the basale of the archipterygium is first of all differentiated from the rest of the 480 COMPARATIVE ANATOMY. appendage as a large bone, which forms a piece of about the same size as the rest of the appendage ; this may be called the humerus. In Plesiosaurus two succeeding pieces, which in Ichthyosaurus are still indifferent, are also increased in size ; these correspond to the fore- arm : radius and ulna; these ai*e succeeded by two transverse rows of smaller pieces, which form a carpus, and these again by longer rows of bones, which represent the metacarpus and the phalanges of the fingers. The segmentation which affects the appendage after the stem and rays are broken up into several pieces, may here be seen in its different stages. The arrangements which are presented by the Amphibia are similar in character; for although one finger is atrophied, we can fill up the void by the aid of the arrange- ments seen in the hinder limb, where they are complete. The stem of the archiptery- gium must, therefore, be sought for in a lateral series of skeletal pieces, which ex- tends from the humerus, through the ulna to the fifth finger, and in the carpus con- sists of two pieces. The other skeletal pieces are arranged on these rays. One ray begins with the radius, and extends into the first finger. A second, third, and fourth begin in the carpus, and end in the second, third, and fourth fingers. The primitive carpus is therefore composed of ten pieces; five carpals carry the fingers, three are attached to the bones of the fore- arm; these are the radial, intermedium, and ulnare ; two centralia (c c) are enclosed by these two sets. The change in the function of this appendage is connected with a rotation of the humerus on its own axis, and this rotation may be observed in the individual development of higher forms. It brings about a difference in the position of the limb as compared with that of lower forms. Fig. 265. Diagram of the fore-limb of an Amphi- bian. The dotted lines indicate the rays, which remain attached to the stem of the Archipterygium. § 364. A more or less complete copy of the typical form of limb derived from the archipterygium is retained in all divisions of the Verte- brata. In all there are often unmistakable traces of the characteristic relations, in opposition to which numerous deviations, due chiefly to reduction and concrescence, make their appearance. These modifi- cations are clearly due to the varied uses to which the limbs are put, just as the complete atrophy of some parts, or even of the whole limb, are due to their being no longer required. ANTERIOR EXTREMITY OF VERTEBRATA. 481 In the Amphibia the two upper portions are greatly developed, but, except that the radius and ulna are fused in the Anura, they present no such striking differences as those which are seen in the carpus. Some of the primitive carpalia disappear in the distal row ; with this is generally correlated a shortening of the fingers, which are commonly limited to four ; or, again, two or three carpalia may be fused together (Frogs, etc.). Concrescence may likewise be seen to affect the proximal series of carpalia. In the Reptilia, the various portions of the skeleton of the arm are least altered in the Ohelonii, which have not only nine carpal bones, but all five fingers. In the Saurii two of the three carpalia of the first row are fused together ; those of the second row are also greatly modified, and are reduced in number when any of the fingers disappear. The carpus is still more altered in the Crocodilini. The radiale has become much lai^ger than the ulnare, and the second row of carpalia is merely represented by a few elements, which are always partly cartilaginous. The two ulnar fingers are conse- Fig. 266. Skeleton of the arm of Ciconia alba. It Humerus, u Ulna, r Radius. c c' Carpus, m Metacarpus, p p' p" Phalanges of the first three fiugers. quently shortened in comparison with the three radial ones. In the limbs of the snake-like Saurii there are all stages of reduction. The Ophidii are distinguished by the complete absence of these parts. In birds, where the whole of the fore-limb is converted into an organ of flight, the reduction of the manus is still more marked. Two bones only (Fig. 266, cc') are well developed in the carpus, while a piece of cartilage, which corresponds to the second row in the carpus, soon fuses with the base of the metacarpus. Three fingers are always more or less developed in the manus ; in the Saururas these were permanently separate, but in the Ratita? and Carinataa the metacarpals (m) of the second and third, and generally also that of the first, are fused into one piece of bone. On the third finger there is a rudiment of a fourth one. As compared with the Saurii, the number of phalanges is reduced iu Bii'ds. In the Saurii, starting from the first finger of the radial side, which has two phalanges, we find one more in each finger as far as the fourth, which has five ; but the fifth finger has not so many. In the Crocodilini this increase stops at the third; in most Birds the second finger has only two phalanges (p')} the 2 i 482 COMPARATIVE ANATOMY. first and third one only (p p"); the first and second fingers rarely have an extra phalanx. Furbringer, M., Die Knocheu nnd Muskeln der Extremitaten bci den schlan- genartigen Sauriern. Leipzig, 1870. § 365. The greater variation in adaptive relations to various conditions is implied by the greater variations in the structure of the skeleton of the Mammalian forelinib. Its elements somewhat resemble the lower condition, such as is seen in the Ckelonii, with regard to the number of the carpal bones. Although the manus is often modified by the atrophy of certain fingers, the extremity, even in the lower divisions of the Mammalia, has very various uses. Owing to the greater power of movement possessed by the two bones of the fore- arm, and the connection between one of them (the radius) and the manus, the anterior extremity loses its lower function of an organ of support, and is converted into a prehensile organ. This pheno- menon is seen in the Didelphia, as well as in the Monodelphia ; it is most complete in the Primates. The carpus has the three primitive pieces of the proximal row. A centrale, also, is not unfrequently present (Rodentia, Insectivora, Lemurs, Orang, and, for a short time, Man). The distal row of the carpus have the two ulnar bones fused into an uncinate (cf. Fig. 268, I II). The pisiform is a special bone, which is attached to the ulnar edge of the carpus ; it is very large in many forms. It is also found in the Reptilia, and may be shown to be the soli- tary remnant of a numerous series, which was possessed by the Enaliosaurii. The modifications derived from this series of forms are very closely correlated with the function of these parts. When the arm is used as an organ of flight (Chiroptera), we find that its different portions are considerably elongated; and so, again, they are shortened, and various parts be- come very large in those numerous cases in which the arm acquires a special function, as in digging and so on; the Monotremata, many Edentata, Talpa, etc., are examples of this. In- stead of this great increase in size, which is seen in various parts of^the skeleton of the arm, there may be atrophy, as is the case with the fore-limb of the Cetacea. It forms a paddle, the separate parts of which have but little power of move- ment, and the various bones of these parts may lose all their articulations, and become united into an unjointed fin-like mass (Fig. 267). In another series, several of the fingers are atrophied, and the Fig. 267. Anterior extremity of a young Dolphin, s Scapula. h Humerus, r Ra- dius, u Ulna, c Car- pus, m Metacarpus. X> Phalanges. ANTEEIOE EXTEEMITY OE VEETEBEATA. 483 fore -limb becomes a mere organ of support and locomotion. It is clear that tliis is not a primary condition from the relative position of the bones of the fore-arm, which requires us to presuppose a condition in which they were capable of pronation and supination. As the limb ceases to have more than one function, this power is lost ; the radius and ulna are connected immovably, and this may lead to the atrophy of various parts of these bones, or to their more complete fusion with one another. This is the case in the Artio- dactyla, where the distal end of the ulua is rudimentary in the Ruminant forms. In the Tylopoda and Solidungula this end of the iilua has quite disappeared, while the upper end is united with the radius into one bone. The fingers may take on one of two sets of characters. In Pig. 268. Skeleton of the manus of various Mammals. I Man. II Dog. Ill Pig. IV Ox. V Tapir. VI Horse. r Radius, u Ulna, a Scaphoid, b Lunar. c Cuneiform, d Trapezium, e Trapezoid. / Magnum, g Uncinate, p Pisiform. either case the pollex is absent, and it is not functional even in the digitigrade carnivora (Fig. 268, II). Of the remaining digits, however, the third and fourth are so greatly developed in the Artiodactyla (III IV), that the other two (2 and 5) often do not touch the ground (Suina, Moschidas). The fifth finger is next lost, so that the third and fourth only are well-developed, and the second forms a mere appendage (Anoplotherium). The third and fourth fingers become still larger when their two metacarpals are fused together (IV), while the second and fifth fingers become rudi- mentary (Oxen, Sheep, Deer, etc.). The Perissodactyle series also begins with the four-fingered form, but in them one finger ouly (the third) is markedly larger (Tapir) (V). When the fifth, which is already the smallest, disappears (Paiaeotherium), the second and fourth are attached to the third in the form of appendages 2 i 2 484 COMPARATIVE ANATOMY. (Hipparion), and when the two lateral fingers are reduced to their metacarpals alone, these are attached to the large metacarpus of the third finger as mere "splint-bones" (VI), and the third finger becomes the sole support of the limb (Eqinis). The number of phalanges in the different fingers is increased in the Cetacea only ; all other Mammals have two in the pollex, and three in all the other fingers. Posterior Appendages. Pelvic Girdle. § 366. The relations of the pelvic girdle are also correlated with differences in the functions of the extremity. The homology between the two skeletal portions is consequently more fully recog- nisable as the functions of the two extremities are more nearly the same, and the extent to which they are differentiated from one another less. A single piece of cartilage forms the groundwork of the pelvic girdle. In the Selachii this is rarely enlarged in a dorsal direction. In the Ganoi'dei and Teleostei the two halves of the ossified portion are connected in the middle line. They undergo considerable varia- tions in position, for they may be placed more or less anteriorly and close to the shoulder-girdle (Pisces thoracici), or may even be united with it (Pisces jugulares). In the Amphibia the two bones of the pelvis are connected with the vertebral column ; at the same time they may be seen to be divided into two pieces at the point where they are connected with the femur ; the dorsal one, which is attached to a transverse pro- cess (that is, to a rudimentary rib), forms the ilium ; the ventral one, which is connected along the middle line with its fellow of the opposite side, is known as the ischio-pubic bone (Urodela). There is reason, however, for supposing that it merely corresponds to an ischium. This arrangement is modified in the Anura (cf. Fig. 225), for the long and slender ilia (11) are uuited with the ischio-pubic bones (is), which are converted into a vertical disc, and fused with one another. The ilium of the Reptilia is greatly developed ; in Chamasleo it resembles a scapula, and is continued into a process, which is com- parable to a supra-scapula. In the Saurii it is elongated (Fig. 269, Jl) ; in the Crocodilini it is shorter and broader (Fig. 270, J7). The bone is directed forwards, so that it is connected with the sacrum behind the acetabulum. In the Saurii and Ohe- lonii the ventral portion of the pelvis is continued from the aceta- PELVIC GIRDLE OF VERTEBRATA. 485 buluni into two divergent pieces (Fig. 269), which enclose a large opening (foramen obturatum). The anterior process is called the pubis (P), the posterior one the ischium (Js). The two bones of either side are more or less connected together along the middle line, but this connection may disappear. The pelvis of the Crocodilini Fig. 269. View of the left side of the pelvis of Monitor. Jl Ilium. Js Ischium. P Pubis, a Hinder cud of the ilium, b Its anterior process. Fig. 270. View of the left side of the pelvis of Alligator lucius. x y Two limbs of the ischium, which unite with r s, two processes of the ilium, to enclose a foramen (o) at the base of the acetabulum. The other letters as iu Fig. 269. (Fig. 270) differs from this in many points, for a single bone (Js) is given off ventrally from the acetabulum, and is connected by means of two processes with the ilium (x y). It appears to represent an ischium only, while a bone, which takes no part in the acetabulum, but articulates with the ischium (p), and converges, like its fellow, towards the anterior wall of the abdomen, represents the pubis. The pelvis in the fossil Dinosaurii was of the same character ; the ilium was distinguished by a process which was directed forwards, and of which there is an indication only in the extant Saurii and Crocodilini (b). The acetabulum was similarly incomplete, and was connected with a long ischium, which was directed obliquely back- wards and downwards, and was not united with its fellow of the opposite side. A long pubis, which also ended freely, was given off from the anterior margin of the acetabulum, and ran parallel to the ischium. This relation of parts is the same as that which characterises the Avian pelvis (Fig. 271). In them the ilium (J7) does not only extend a long way back (a a), but its anterior process is converted into a broad plate (b b). This extends along the lumbar region of vertebral column, and even into the thoracic region, and so presses a very large number of vertebras into the pelvic region. The ischium (J*) runs backwards from the incomplete acetabulum, and in a direction which is nearly parallel to that of the hinder portion of the ilium ; the small pubis, which has a slight share in the formation 486 COMPARATIVE ANATOMY. Fig. 271. View of the left side of a Bird's pelvis. The dotted portion represents that part of the three pieces of the pelvis, which extends backwards by the development of cartilage. The dotted line marks off that portion of the ilium (bb) which grows forwards without the addition of any cartilage. The letters as in the preceding figures. of the acetabulum,, takes the same course ; its ends project farther back than those of the ischium, and generally converge ; in Struthio they even form a symphy- sis. There are various kinds of connections between the ilium and ischium, and between these and the pubis. The pelvis in the Mammalia is very dif- ferent. The primi- tive connection with the sacrum is always in front of the acetabulum. The ilium, however, is directed from be- fore backwards, and the hinder edge of the Bird's ilium corresponds to the anterior edge of the Mammal's ilium. Two different positions therefore for the ilium are derived from the Amphibia. In the Amphibia it is directed laterally and inferiorly away from its connection with the sacrum, in Keptilia and Aves obliquely forwards, and in Mammalia obliquely back- wards. The ventral portion of the pelvis encloses an obturator foramen, and is united ven- trally with that of the other side. The primitive pelvic carti- lage gives rise to the ilium and ischium ; the pubis is derived from a separate rudiment, which is united with the ischio-iliac rudiment in the acetabulum (Man). This leads us to think that the pubis is an independent piece of the skeleton, which has retained its independence in the Cro- codilini. The ilium of the Mammalia is connected with a few vertebras. The ischium also may be united with the false sacral vertebras (Dasy- pus, Bradypus). When the two ventral pieces are united at the ischio-pubic symphysis, as they are in the Marsupialia, many Rodents, Artiodactyla, and Peris- sodactyla, the pelvis is elongated in form. In the Insectivora and Carnivora the greater part of the symphysis is formed by Fig. 272. View of the left side of the pelvis of a Dog. il Ilium, is Ischium. j> Os pubis. vl Penultimate lumbar vertebra. vc Caudal vertebra. POSTEEIOE EXTEEMITY OF VEBTEBEATA. 187 the two pubic bones, and this is still more marked in the higher orders. An independent adaptation, which is seen in various Mammals (Iusectivora and Chiroptera), is the presence of a ligamentous con- nection instead of the pubic symphysis ; this may be very wide in the female (Erinaceus). When the posterior extremities are absent, the pelvic girdle also undergoes atrophy. There are rudiments of it in the Cetacea. In the Monotremata and Marsupialia there are two bones in front of the pubes ; these marsupial bones are directed forwards ; in Thylacinus they are reduced to small rudiments in cartilage. Gegenbaur, C, Beitriige zur Kenntniss des Beckeus der Vogel. Jen. Zeitschr. VI. — Hoffmann, C. K., Beitnige zur Kenutniss des Beckens der Amphibian u. Reptilieu. Niederland. Arch. III. Posterior Extremity. § 367. Wo find just the same arrangements in the hind-limb as we have described as existing in the fore-limb. In Fishes the hind limb forms the ventral fin. In the Selachii its skeleton has the same characters as that of the thoracic fin ; the most striking difference is that the rays are arranged in a simpler manner. The basale of the stem is generally greatly elongated. The joints which succeed the basal piece undergo a special metamorphosis in the male, where they are converted into copulatory organs. The skeleton of the ventral fin in the Ganoidei may be derived from this by supposing that there has been a peripheral reduction, very similar to that which we saw in the skeleton of the thoracic fin ; and the Teleostean fin can be derived from the Ganoid. This is generally much simplified, both as regards the size and the number of its separate pieces, in consequence of the feebler develop- ment of the whole ventral fin. In both divisions the dermal skeleton takes part in increasing the surface of the ventral fin, just as it has been shown to do in the thoracic one. When we come to compare the hinder extremity of the higher Vertebrata with the ventral fin of Fishes, we must again begin with the archipterygium, which seems to be the lowest stage of this extremity also. The segmentation of the extremity into successive pieces is a repetition of the arrangement which we met with in the skeleton of the arm. We distinguish the femur, tibia, and fibula ; and lastly, in the foot, a tarsus, metatarsus, and phalanges. The four inner toes, and the parts that carry them, may be again regarded as joints of the rays which are given off from a row of bones extending from the femur, through the fibula, to the outermost toe. The tarsus is made up of ten pieces, three of which 488 COMPARATIVE ANATOMY. arc attached to the leg; these are the fibulare, intermedium, and tibiale. There are two centralia ; and five distal tarsalia carry the bones of the metatarsus (cf. Fig. 265). In the Enaliosaurii the skeletal portions of the hinder extremity are an exact repetition of those of the anterior one ; and even in some of the Amphibia (Urodela) we meet with an arrangement which is the same in all essential points, so that we need not describe them specially. In most Urodela, all the five terminal pieces, or toes, are retained in the hind-limb ; this is more distinctly like the primitive form than is the skeleton of the fore-limb. In Cry ptobran elms, Menopoma, and others, the two centralia even are per- sistent. But in the Anura there is a very great change ; the tibia and fibula are fused. In the place of the three proximal tarsal bones there are two long bones, which are, however, often fused at their ends ; they are ordinarily known as the astragalus and calcaneum. The distal row of tarsal bones is also greatly reduced. Finally, we must note the presence of a rudiment of a sixth toe. § 368. In the Chelonii there are unimportant modifications in the larger pieces of the extremities ; in addition to this we must note the gradual concrescence of some of the bones of the tarsus, which is of great importance as explaining the ske- leton of the foot in Birds, as well as in other Reptiles. An intermedium is united with a tibiale to form an astragalus ; and the centrale is attached to, or even com- pletely fused with, this bone. The fourth and fifth tarsalia similarly form a single bone, the cuboid. Owing to the forma- tion of a single piece out of the bones of the first tarsal row, and the firm union that is effected between this piece, and the tibia and fibula, the foot gets to be articulated in a peculiar manner. It moves on an intertarsal joint. The skeleton of the Crocodile's foot is somewhat different. The tibia and fibula articulate with two bones, of which the fibulare has the greater power of movement. The larger bone, connected with the tibia, corresponds to the similar bone in the Chelonii. A piece of Fig. 273. Hinder extremity of a larva of Salamandra macu- losa. The dotted lines arc drawn through the rays, to which the different pieces he- loner. POSTERIOE EXTREMITY OF VERTEBRATA. 489 cartilage, which is more closely connected with the metatarsus, is articulated to it ; while a cuboid is articulated to the fibula. Owing to the independence of the fibula, we have here a peculiarity, which is only seen again in the Mammalia. In the Saurii, the tarsal bone developed out of four primary elements (Fig. 274, A ts) has no sigus of its constituent parts even in the embryo. It is immovably connected with the tibia and fibula, while the distal bones of the tarsus (tl) are more or less connected with the metatarsus. This appears to have been most complete in the fossil Saurii (Ornithoscelida). In these arrangements we may per- ceive an outline of what obtains in the foot of the Bird, which, in its em- bryonic condition (Fig. 274, B), pre- sents us with those characters which are permanent in many Reptiles. The fibula (f)) extends to the tarsus. This is formed of two pieces of cartilage ; the upper one (ts) is undoubtedly homo- logous with the bone, which is made up of four elements in the Eeptilia ; the lower one (ti) corresponds to the distal series of tarsal bones. The metatarsus is made up of five bones, which were primitively separate, but only four of these (B, I — IV) carry toes, while the fifth is very small and completely fused with the distal portion of the tarsus. The difference between the adult and embryonic arrangements consists in the degeneration of the fibula (Fig. 275, b'), which later on is attached to the tibia, as a small appendage (¥), and which never reaches the tarsus (b) . The proximal tarsal cartilage fuses with the tibia, and forms its articular condyle; the distal one unites with the single piece (c), which is formed from the fusion of the three longer metatarsal bones, and in which no permanent signs of separation can be made out except what is implied by the separate condyles at its distal end (Fig. 275, c'). The metatarsal of the hallux remains dis- tinct, and generally forms a small appendage of the large tarso- metatarsus. The arrangements, therefore, which are seen in the foot of the Reptile are still further developed in that of the Bird, for the parts which in the former are merely united firmly together, are fused in the latter ; the foot still moves on the same intertarsal joint. Fig. 274. Skeleton of the foot of aEeptile (Lizard) (A) and a Bird (B) ; the latter is in its embryonic condition. / Femur. t Tibia. p Fibula. ts Upper, tl Lower piece of the tarsus, m Metatarsus. I — V Metatarsalia of the toes. 490 COMPARATIVE ANATOMY. Witli regard to the toes, we find five to be the dominant number in the Reptilia ; it is in Birds only that they fall to four, or three, or even to two (Struthio). The phalanges of the toes generally increase in number from within outwards ; there are two on the hallux and five on the fourth toe. This holds for the Saurii, Crocodilini, and Aves. There are not so many in the Amphibia or Chelonii. Amongst the Reptilia the limbs are reduced in the snake-like Lizards, and in all Ophidii, among which the Peropoda only are provided with any rudiments of them at all. Gegenbaue, C, Untersuchuugen zur vergleick. Leipzig, 1864. Anat. I. 369. Fig. 275. Hinder ex- tremity of Buteo vulgaris, a Femur. 6 Tibia. V Fibula. c Tarso-metatarsus. c' The same pieco isolated, and seen from in front, d d' d" d'" Four toes. The special differentiations in the skeleton of the hind-limb of Birds and Reptiles do not re- semble those which are seen in the Mammalia. As a rule it is less altered than the fore-limb. In the Perissodactyla, many Rodents, etc., the femur is distinguished by the possession of a third trochanter. The tibia is the most im- portant bone of the leg ; the fibula is often rudi- mentary, especially in the Ungulata. In the Artiodactyla the distal end remains ; it is articu- lated to the tibia and to the tarsus (astragalus), and appears to enter into the composition of the latter. In some (as in Rodents and Insectivora) the tibia and fibula are complete, and are fused together. The tarsus is the most characteristic part ; it is attached by two pieces to the leg, but, as a rule, only one of these forms the ankle- joint. The process on the second bone (calcaneum), of which there were indications in the Crocodilini, is still more developed. The centrale remains separate, but passes to the inner edge of the foot, where it forms the navicular. In some of the Prosimioa it unites with the calcaneum to form a long process (Macrotarsi). Of the five distal bones the two outer ones are always replaced by the cuboid, while the three inner ones generally remain distinct (cunei- form). When the number of toes is diminished, these latter bones are often reduced ; they may even fuse with the metatarsus, as in Bradypus. The cuboid also may bo united to the navicular (Ruminantia). In addition to its primitive function as an organ of support and of movement, the foot may be developed into a grasping organ; when this happens, the foot comes to resemble in many points the end ); the Mid-brain or Mesencephalon (B C c) forms a third swelling ; and this is succeeded by the Hind-brain or Metencephalon (d), and the After-brain or Myelencephalon (e), which is directly continuous with the spinal chord, and with the metencephalon. The metencephalon forms the most anterior por- tion of the roof of the myelencephalon, and is not therefore as distinct as the rest of the cerebral vesicles. At first, the vesicles are placed one behind the other, and lie in the line of the longitudinal axis of the spinal chord, bat they soon come to be set at an angle to one another. This is due to the unequal growth of the upper and lower portions, for the upper ones increase greatly in size. Those parts which are least developed become covered over by the growth of some of the upper parts. Between the prosencephalon and thalamen- cephalon the wall is thinned out, and a fissure- like portion developed (primitive cerebral cleft, Fig. 280, a), into the interior of which a process from the envelopes of the brain is continued. This is not a time lacuna, but is merely due to the gradual thinning-out of the wall of this portion. The epiphysis (pineal gland) is developed from a part of this roof. The lower portion of the thalamencephalon forms the floor of the Fig. 280. Vertical aud median sections through a Vertebrate brain. A Of a young Selachian (Heptanchus). B Of the embryo of an A d d e r. C Of the embryo of a Goat, a Prosencephalon. l> Thalamencephalon. c Mesencephalon (in A it is marked by d). d Meten- cephalon. e Myelence- phalon. s Primitive cere- bral cleft, h Hypophysis. 501 COMPARATIVE ANATOMY. second cerebral vesicle, and gives rise to a diverticulum, which is found in all Craniota, and is known as the infundibulum. From the lower side of the head a depression of the ectoderm grows towards this diverticulum; later on, the ingrowth becoming pinched off, forms a portion of the cerebral appendix attached to the infundibulum (hypophysis). The range of the position of the depression for the hypophysis as far forward as the entrance into the cavity of the mouth enables us to recognise in this structure an organ, which primitively did not belong to the nervous system at all, and the function of which is still a matter for speculation. Just as the upper wall between the fore- and twixt-brains gets thinned out, so too the roof of the myelencephalon is thinned out, in such a way that no roof remains but such as is formed by the outer- most vascular layer of the nerve-centre, the pia mater. The large cavity which is thus roofed over forms the fourth ventricle. The ventricles, or cavities in the portions derived from the primary cerebral vesicles, are connected with one another in just the same way as the cavities of the cerebral vesicles. The brain of the Cyclostomata is the simplest in form ; among them the lowest grade is occupied by the Myxinoidea, where the various segments have very nearly the same characters. A portion, which is developed from the fore-brain, and which gives off the olfactory nerves (bulbus or lobus olfactorius), generally forms large lobes, which, in the Selachii, are connected with the brain by a more or less long tractus olfactorius (Fig. 281, h). The ventricle of the prosencephalon is continued into them. They may also be fused with the prosencephalon, which is larger than the other divisions, in the Selachii (r/), and gives indica- tions of a separation into two, four, or more paired pieces. It is large also in the Ganoidei (Fig. 282, sn gepfcam (after ^ v of Stenson. Lacerta). Visual Organs. § 397. The eye in the Vertebrata appears to have essentially the same structure as in the more highly developed groups of lower animals ; but the ontogeny of the organ shows that it belongs to another type, and this is also obvious from its minute structure. We cannot, therefore, connect it directly with the relatively well- developed stages of the eye in other animal phyla ; tho only indi- cations of any connection are to be seen in the Tunicata. In the larvae of the Ascidiae, as in Vertebrata, the eye is not directly de- veloped from the ectoderm, but from the anterior portion of the central nervous system. What is known as the eye in Amphioxus is of a much lower grade ; it is a spot of pigment which varies in character, and is attached to the anterior end of the central nervous system. The central nervous system chiefly, and secondarily the integu- ment, are concerned in the constitution of the Vertebrate eye. The former gives rise to the apparatus which perceives, the latter to the apparatus which refracts, the light. The earliest rudiment of the eye is a diverticulum, which is developed from the sides of the prosen- cephalon (Fig. 294, A a), and which has the form of a vesicle, connected by a stalk (b) with the rudiments of the brain (c). The "primitive optic vesicle " lies below the ectoderm ; the ectoderm next gives rise to a thickening (B), which pushes in the anterior wall of the vesicle towards the posterior one. Below this thickening a process of the mesoderm grows towards the optic vesicle, and puts the side-walls also of that vesicle in continuity with the epidermic thickening. The effect of these processes is to bring the anterior and posterior 528 COMPARATIVE ANATOMY. I e walls of the primitive optic vesicle close to one another ; the whole then forms the se- condary optic ves- icle, and is cup- shaped; the month of the cup is filled by the ec- todermal thicken- ing. This latter forms the rudiment of the lens (I). While the stalk of the primary vesicle is being converted into the optic nerve, the tissue behind, which is enclosed by these parts, is converted into Fig. 291. A Vertical section through the rudimentary head of a Fish. c Brain, a Primitive optic vesicle. b Its stalk, d Integumentary layer. B Formation of the secondary optic vesicle, p Outer, r Inner layer of the primitive optic vesicle, e Epidermis pushing the lens (I) into the secondary optic vesicle. The vitreous body is seen behind (after S. Schenk). t substance, which gradually fills up the greater part of the secondary optic vesicle, and forms the vitreous body. The innermost layer of tissue around the secondary optic vesicle is converted into a vas- cular membrane, the choroid, while a firm fibrous layer outside it forms the sclerotic, and invests the secondary optic vesicle ; this grows out anteriorly as far as the connection between the lens and the ectoderm. As a result of the extension of this process, the lens becomes cut off, and a trans- parent portion of the sclerotic now intercalated in front of it forms the cornea, which at the same time becomes connected with the rudi- mentary piece of integument (conjunctiva) which lays in front of the lens. The eye then is a rounded capsule (bulbus oculi), the investment of which (sclerotic) ex- tends over the optic nerve, and is thence con- tinued into the dura mater, while anteriorly it is continued into the cornea. Within this capsule is the secondary optic vesicle, which is developed from the invaginated primary one, and which is separated from the sclerotic by the choroid. The secondary vesicle, in which there is a lateral cleft owing to the ingrowth of the "vitreous body," embraces the lens anteriorly. These two, layers (a b), which pass into one another at this anterior margin, and at the lateral fissure (Fig. 295, s), are not similarly differentiated ; the inner one (l>), which is greatly thick- ened at a very early stage, has its hinder portion converted into the retina, while the outer, and thin one {a), forms the tapetum nigrum. When the tapetum nigrum becomes pigmented, a pale Fig. 295. Section through the secondary optic vesicle of a Fish's embryo, taken vertically to the "choroidal fis- sure s." a Outer, b Inner lamella of the optic vesicle, c Vitreous body. d Lens (after S. Schenk). VISUAL OEGAN OF VEETEBEATA. 529 line can be made out on the lower and inner side of the rudimentary bulb ; this extends from the optic nerve to the free anterior edge of the choroid. It corresponds to the fissure (choroidal fissure) which was formed when the rudiment of the vitreous body grew into the secondary optic vesicle (s)} and which must therefore affect the retina and the pigmented layer of the choroid (tapetum nigrum). A large number of changes subsequently affect this rudiment of the eye. The anterior edge of the secondary optic vesicle grows out, together with the tissue that forms the rudimentary choroid, and gives rise to the iris, which bounds the pupil. When the pro- cess of the cutis pushes its way into the secondary optic vesicle, blood-vessels pass (in the Mammalia) into the cavity ; these are dis- tributed in the periphery of the rudiment of the vitreous body, so that they must have a large share in the nutrition and growth of this structure. The lens, also, of the Mammalia, is invested by a vascular capsule of connective tissue, which disappears again before birth ; in some, however, it does not disappear so early. Mullee, W., Die Staniniesentwickelung des Auges der Wirbelthiere. Leipzig, 1875. — Kesslee, L., Zur Entwickelung des Auges der Wirbelthiere. Leipzig, 1877. § 398. As to the form of the bulb, its anterior segment is much flat- tened in Fishes (Fig. 296). The aquatic Amphibia have the bulb flattened anteriorly ; the Ophidii and Crocodilini, among the Reptilia, are characterised by a more considerable curvature of the cornea. In most Birds (Fig. 298) the bulb is divided into an anterior and a posterior segment; the former carries the very convex cornea. This form of eye is most marked in the Eaptores, but the cornea is flattened in the Natatores and Grallatores. Among Mammals, also, the spherical form of bulb may undergo great variation in form. The Sclerotic may be formed of various kinds of connective substances ; in fact, it may be made up of connective tissue, of bony parts, or of cartilage. This latter is found in the Selachii, Chima3ra3, Fig. 296. Eye of Esox lucius. Horizontal sec- tion, c Cornea, p Pro- cessus falcif ormis. s' s' Os- sifications in the sclerotic. Fig. 297. Eye of Moni- tor. Horizontal section. c Cornea. p Processus falciformis. Fig. 29S. Eye of Falco chrysaetos. Horizontal section, p Pectcn (after W. Sommering). and Ganoidei, and also in the Amphibia, the most varied in the Osseous Fishes. These arrangements are 2 M 530 COMPAEATIVE ANATOMY. In tlie Saurii, Chelonii, and Aves, the anterior portion of the sclerotic, which abuts on the cornea, is supported (Fig. 296, s) by a circlet of flat pieces of bone (sclerotic ring). In all Mammals except the Monotremata the sclerotic is formed of connective tissue; it is very thick in the Cetacea (Fig. 299, s). The choroid is made up of several layers, which, as a rule, have the same characters as in Man. Anteriorly it gives rise to the folded ciliary processes ; these are feebly developed in the Selachii and <» Ganoi'dei (Sturio), and absent in most of Fig. 299. Eye of Baiaena th Teleostei . tho ch0roid is then continued mysticetns. Horizontal , . . . .. , , ... section (after W. Summer. ou as the iris, which bounds by its inner iug). margin the pupil, which varies in form. The tapetum lucidum is a special modi- fication of the choroid; this forms a spot of varying size, which is generally greenish or bluish in colour, and has a metallic lustre ; it is sometimes produced by groups of spicular crystals placed in the cells of the tapetum (Selachii), or by a fibrous tissue (Carnivorous Mammals and Ruminants). It is owing to its presence that the eye can be seen in the dark. A vascular plexus, which lies outside the choroid of Fishes, forms the so-called choroid gland. In the anterior portion of the choroid there is a muscular layer, which forms the ring known as the ciliary ligament. The musculature is continued hence into the iris, in which there are radial and circular fibres. In Fishes, Amphibia, and Mammals, this musculature is composed of smooth fibres; in Reptiles and Birds, of transversely striated ones. The retina, which is placed on the choroid, extends forwards as far as the commencement of the ciliary body, where it ceases to be developed. The optic nerve is distributed, and ends, in it. The optic fibres occupy the innermost layer of the retina, which is merely sepa- rated from the vitreous body by a thin membrane. It is followed by a number of layers, of varying structure, the last and outermost of which is made up of rod-like and cone-like structures, the bacillar layer. These end-organs, which are similar to the rods of the invertebrate eye, are, therefore, turned away from the opening of the eye in the Vertebrata; the Vertebrate eye is therefore distinguished from the optic organs of the Invertebrata by a very essential point, which must not be left out of con- sideration when we are discussing their genetic relations. Connected with the development of the secondary optic vesicle is the formation of a special organ, which makes its way into the vitreous body at the point at which the optic nerve passes into the retina; it has no connection with the choroid, but forms a vascular, darkly-pigmented, process. A structure of this kind is found in the eyes of many Teleostei, and is known as the processus falciformis (Fig. 296, p). Its end, which in many Fishes is distinguished by a layer VISUAL OEGAN OF VEETEEEATA. 531 of smooth muscular fibres, is provided with a swelling (campanula llalleri), which is attached to the hinder part of the capsule of the lens. These processes are also found, in a somewhat modified condi- tion, in the eyes of Reptiles and Birds. In the Saurii there is a curved aud thickened fold which extends to the margin of the capsule of the lens, at the side of which there may be several other folds (Fig. 297, p). This structure is feebly developed in the eye of the Crocodilini. In Birds it is remarkable for the increase in the number of its folds, and is distinguished as the " pecten" (Fig. 298,^). Iu many Natatores and Grallatores it reaches as far as the capsule of the lens. In the Struthiones the end of the pecten is widened out into a pouch (marsupium). In Apteryx, as in the Mammalia, it is absent. The point, at which the optic nerve enters, varies with the characters of this process, for when it is widened out at its base the nerve is placed more to the side of the eye. With regard to the lens, the difference in its form, in accordance with the surrounding media, is a noteworthy point. In Fishes it is very large, and quite spherical, as it is also in the Amphibia, and in aquatic Mammalia ; while in others, and in Birds and Reptiles, we meet with more flattened forms of lenses; the amount of flattening is, of course, very various. The internal cavity of the eye is divided into an anterior and a posterior space by the attachment of the lens to the ciliary portion of the choroid. The vitreous body fills the hinder one; the anterior one, which lies between the anterior surface of the lens and the cornea, is often but a very small portion of the whole eye. It is filled by the aqueous humour. § 399. Accessory organs, which partly serve to move, and partly to protect the bulb, are connected with the eye. The movements of the eye are generally effected by six muscles, of which four are straight and two oblique. They are atrophied in the Myxinoi'dea. In many Teleostei the straight ones are embedded in a canal at the base of the skull ; this is in adaptation to their length, which again is due to the large size of the bulb. They take their origin from a point which is placed some way behind that at which the optic nerve passes out ; it is in the higher divisions only that they acquire relations to this point. In the Amphibia and Reptilia there is a retractor of the bulb, in addition to the four straight muscles. This is retained also by most of the Mammalia, and breaks up into several portions (in Camivora into four), which pass to the bulb from the point at which the optic nerve enters the orbit. In the Mam- malia the superior oblique, which, like the inferior oblique, ordi- narily arises from the median wall of the orbit, is altered in character. It has the same origin as the straight muscles of the eye, and the tendon of insertion passes to the bulb through a pulley, and at an angle. Of the protective organs of the eye, the eyelids are folds of the 2 m 2 532 COMPARATIVE ANATOMY. integument. The inner lamella of these folds is a continuation of the conjunctiva which extends on to the bulb, and which is continuous with the integument at the margin of the lid. Eyelids of this kind are found even in Fishes. In the Selachii there are two slightly projecting and movable folds, which appear to be indications of an upper and lower eyelid ; in many Sharks there is also a third fold at the anterior angle of the eye, which can be drawn over the outer surface of the bulb (nictitating membrane). In the Granoidei and Teleostei the immovable folds are alone present, or there may be merely indications of them ; they are ordinarily distinguished as the anterior and posterior eyelids. Most commonly the integument passes at once into the cornea. In the Perenni- branchiata and Derotremata there is a connection of this kind. Many Salamanders, and the majority of the anourous Amphibia, are provided with horizontal eyelids, of which the lower, and more movable one, functions as a nictitating membrane. In the Reptilia and Aves there is an upper and a lower movable eyelid in addition to the nictitating membrane. In some Saurii (Ascalabotas) and in the Ophidii, the eyelids are developed as an annular fold, which continues to grow until at last it forms a pellucid membrane which lies in front of the eye, and which completely sepa- rates the cornea from the external medium. The circular rudiment of this structure corresponds to the circular eyelid of the Chainseleons. There is a muscular apparatus for the horizontal eyelids, as well as for the nictitating membrane. Whilst the two horizontal eyelids persist in the Mammalia, the nictitating membrane undergoes degeneration. It is supported, as are the two other eyelids, by a cartilaginous lamella. It is generally reduced to a fold, which is placed at the anterior (inner) angle of the eye; in the Primates it has lost its primitive significance, and forms the plica semilunaris. A glandular apparatus for the eyelids is first differentiated in the Amphibia and Reptilia. In Reptiles and Birds, and also in Mammals, there is a gland which opens below the nictitating membrane (Harderian gland, or gland of the nictitating membrane), Avhich is placed at the inner angle of the orbit ; it is not present in the Primates. Its secretion is diffei'ent from that of the lachrymal gland. The Lachrymal glands, which are placed at the outer angle of the eye, are first seen in Reptiles, where they are smaller than the Harderian gland ; they have the same characters in Birds. They are larger in the Ohelonii and Mammalia (except the Cetacea), where the lachrymal gland consists of a complex of separate glands, which are generally united into larger masses. A special efferent duct into the nasal cavity is formed for the secretion of these glands, which is passed out below the upper eye- lid. A canal of this kind, formed by an epithelial thickening on the surface of the head, is present even in the Amphibia. In the Amniota the development of the lachrymal duct is connected with that of the face. The groove, which is formed between the AUDITORY ORGANS OF VERTEBEATA. 533 processes of the upper jaw, and the external nasal processes by the differentiation of these parts, and which leads from the region of the inner angle of the eye towards the edge of the nasal pit, sinks deeper down as these processes are developed (lachrymal groove) ; it now becomes grown over by their edges so as to form a canal, which, when the nasal cavity is developed, opens into it just below the inferior turbinated bones. In the Reptilia (Lacerta) it opens near the posterior nares. At the inner angle of the eye this lachrymal canal is divided into several smaller canals ; there are a larger number (3-8) of such canals on the lower eyelid of the Crocodilini, but a smaller number (2) in Birds and Mammals. Auditory Organs. § 400. The Auditory Organ of the Vertebrata, which has been ob- served in all, except the Acrania, is derived from the ectoderm. During the earliest embryonic period it is laid down in the form of a thickening above the myelencephalon, which extends inwards. A superficial organ of this kind, which must have carried the endings of an integumentary nerve, must be regarded as the starting-point of the great differentiation, which commences so early. The earliest rudiment gives rise to a vesicle, which communicates with Fig. 300. Development of the labyrinth of a Fowl. Vertical sections of the rudimentary skull, fl Fit of the labyrinth. Iv Vesicle of the labyrinth, c Kudimeut of the cochlea. Ir Eecessns labyrinthi. csp Posterior semicircular canal, cse Ex. ternal semicircular canal, jv Jugular vein (after Reissner). the exterior, and which is gradually cut off (Fig. 300), and enclosed by the hinder lateral portion of the cartilaginous cranial capsule, when that is differentiated. The primitive otocyst is the foundation of a complicated cavitary system, in the walls of which the auditory nerve is connected with its end-organs. From this is developed the membranous labyrinth. The surrounding portions of the skull form the cartilaginous, or osseous labyrinth. The simplest condition of the labyrinth is found in the Cyclo- stomata. In the Myxinoidea a tract, which remains connected at two 534 COMPARATIVE ANATOMY. points with the primitive vesicle, is differentiated from it, and forms a semicircular canal, so that the whole labyrinth has a circular form. In the Petromyzontes there are two of these canals, each of which commences with an ampulla-like enlargement, while the rest of the vesicle of the labyrinth forms the "membranous vestibule ; " in this there is a special diverticulum, which is the rudiment of a new differentiation. In the Gnathostomata a third canal is developed, so that henceforward three semicircular canals open into the vestibule. When the vesicle of the labyrinth sinks beneath the surface, its stalk-like basal piece remains open on the roof of the skull, in the Selachii, and swells out below the integument into a saccus endo- lymphaticus. This corresponds to the recessus labyrinthi (ductus endolymphaticus), which passes up as far as the roof of the skull in the Teleostei, and may undergo various metamorphoses. One of these metamorphoses has been regarded as leading to the growth of this portion into a tube which covers the brain (Urodela), or extends to the base of it (Anura). In the Ophidii and Saurii it reaches to the roof of the skull, being filled in the embryo with crystals of lime, and widened out. In Phyllodactylus it extends beyond the skull, and may even pass into the cervical region, being swollen out in parts. The connection between these structures and the primitive stalk of the otocyst is denied, so that the recessus labyrinthi is regarded as an independent structure. Most of its relations, however, require to be more carefully investigated. In Birds it is an open cavity (r /), for a short time only ; so, too, in Mammals, where later on it forms the aqueductus vestibuli. The vestibule and semicircular canals are very large in all Fishes ; in the Selachii and Dipnoi they are com- Fig. 301. Auditory organ of Cyprinus carpio. a Membranous vestibule, b Ampulla of the posterior and external semicircular canal, c United anterior and posterior canal. d Posterior. e Anterior. / Canalis sinus imparis. l> 1/ r s Spinous processes of the anterior vertebra). The numbers indicate the different bones of the skull. 1 Basi-occipital. 2 Exoccipital. 3 4 Supra-occipital. 6 Petrosal. 7 Parietal. 10 Alisphenoid. 11 Frontal (aftor E. II. Weber). plctely surrounded by the walls of the skull, while in the Teleostei the median portion projects freely into the cranial cavity (Fig. 301). Of the three semicircular canals, two — an anterior (e) and a pos- terior one (<1) — arc placed in the direction of two planes, which cross AUDITOEY ORGANS OF VERTEBRATA. 535 one another more or less perpendicularly; the third, and outer one, lies in a more horizontal plane, and is provided with an ampulla on its posterior limb. The two vertical canals have a common piece (c) which opens into the vestibule, and ampullae at the two other ends. Even in Fishes the vestibule of the labyrinth is divided into several portions. An upper one is directly connected with the semicircular canals (utriculus, alveus communis), and with the subjacent sacculus. The sacculus aud utriculus contain otoliths, which are constant in the same, but different for different, divisions ; they are often very large. The branches of the auditory nerve pass into the end-organs which are to be found in the walls of both cavities, as well as in the ampullae of the semicircular canals ; in the ampullae they are placed on a transverse ridge (crista acustica) ; in the saccules they form the maculae acusticae. Of the numerous modifications which may be observed, the connections between the membranous vestibule and the air-bladder are worthy of remark ; the arrangement is effected in various ways ; it is simplest in some of the Percoidea, and Sparoidea, where the vestibule is continued into spaces in the skull, which are merely covered by membrane ; to these spaces processes of the air-bladder are attached. The relations are more complicated in many families of the Physostomi. In the Cyprinoids the sacculus (a) extends back- wards, and is connected with that of the other side by a sinus impar. This gives off a membranous saccule (atrium sinus imparis) on either side, which passes to an opening on the posterior portion of the skull, which is partly closed by a small bone. This is connected by masses of ligament with a series of bony pieces ( i Id) of various forms, the last and largest of which is attached to the anterior end of the air-bladder (m). These ossicles are modifications of ribs, and form a continuous chain between the vestibule and the air-bladder. In the Siluroidea and Clupeidea connections with the air-bladder are effected in a different manner. § 401. In and above the Amphibia, the labyrinth is greatly diminished in size from what it is in Fishes. It is still of a relatively large size in the Amphibia, and is smallest in the Mammalia. The dif- ferences which are seen in it are partly due to the way in which the two cavities of the vestibule, the utriculus and sacculus, are connected together, and to the course taken by the semicircular canals which spring from the former. The posterior canal may sometimes be set at an angle to the external one (Birds). There is a great difference between the portion of the labyrinth, just described, and which is very similar in all forms, and that part which is only developed as an independent structure in the higher divisions; this, which is known in Mammals as the cochlea, on account of its form, presents a continuous series of differentiations 536 COMPARATIVE ANATOMY. from the lower divisions upwards. In Fishes there is an indication of it in the form of a diverticulum, generally a small one, of the sacculus. In the Selachii it contains a number of small otoliths; in the Teleostei one larger one (asteriscus). In the Amphibia this diverticulum of the sacculus is moi'e independent, but it is still connected as before, and is still directed backwards. This portion, which carries the end of a branch of the auditory nerve, is still further differentiated in the Reptilia and Aves, where the diverticulum, which forms it (Fig. 300, 0 D E c), is a short conical piece, which is directed downwards from the median wall of the labyrinth, and converges towards its fellow of the opposite side. Its end is somewhat bent, and it forms the " lagena." Among Mammals this stage of the organ is seen in the Monotremata only j in the rest this stage is not the permanent one, for the organ is converted into a spirally-coiled canal. At first it is only formed by a prolongation of the sacculus, but special differentiations appear in it, and this cochlear canal, which is formed from the sacculus, is permanently connected with it by a narrower portion only (canalis rouniens, Fig. 302). The organ, which thus becomes more indepen- dent, is sur- E rounded on two sides of its course by lymphatic ca- vities, which accompany it in its coils, and pass into one another at the apex of the cochlea. One cavity is connected with the os- seous vesti- bule, the other is shut off from it at its commence- ment, and is only con- nected with the cavity of the vestibule indirectly ; that is by the communication at the apex of the cochlea. Three cavities, therefore, can be dis- tinguished in the Mammalian cochlea; but ono only, the ductus cochlearis, is connected with the vestibular labyrinth. The other two form the scalre — the sc. vestibuli and sc. tympani. The two scalse occupy the periphery of the coils of the ductus cochlearis, at the base of which the end-organs of the cochlear nerves (organ «£. c- -~zsj Fig. 302. Diagrams in explanation of the labyrinth. / Fish. // Bird. /// Mammal. 17 Utriculus. S Sacculus. US Utricuhis and Sacculus. Cr Canalis reunions. R Eocessus labyrinthi. UC Commencement of the cochlea. C Cochlea. L Lagena. K Ca?cal sac at apex. C Caccal sac of the vestibulum of the cochlear canal (after Waldeyer). AUDITORY ORGANS OF VERTEBRATA. 537 of Corti) are spread out. As the scalae arise as spaces in the tissue which accompanies the ductus cochlearis, they are similar to the cavities between the membranous semicircular canals and their bony wall, or between the membranous and osseous vestibules, and are filled with perilymph. In the Amphibia, and all higher forms, spaces appear in that part of the walls of the bony labyrinth which lies on the outer surface of the cranium ; these in a different fashion effect a communication between it and other arrangements that are connected with the auditory organ. The fenestra ovalis, which is always closed by a plate-like piece of bone, is a hole of this kind in the osseous vesti- bule. A second opening, which appears first in Reptiles, and which is correlated with the further development of the cochlea (fenestra rotunda), lies in the wall of the scala tympani, and is closed by a membrane. Both these arrangements are related to the development of external conducting organs. Retzius, G., Anatom. Untersuch. I. Stockholm, 18/2. — Hasse, C, Anatomische Studien. Leipzig, 1870-73. § 402. Other parts are gradually added on, as accessory organs, to the auditory organ, although primitively having no relation to it. The first branchial cleft, which in the Selachii and Granoi'dei persists as the " spiracle," enters into close relation with the wall of the labyrinth in the Amphibia. As it grows over this wall it is converted into a cavity, the wider portion of which forms the tympanic cavity; this is bounded in the middle line by the wall of the labyrinth ; the portion which leads into the primi- tive buccal cavity forms the Eustachian tube. It reminds us of its primitive (spiracular) condition by at first communicating freely with the exterior. The cleft is, however, soon closed, which leads to different arrangements. In the Ccecilias and Urodela the cleft is closed by the superjacent muscles, so that there is no tympanic cavity. One division of the Anura (Pelobatidao) presents the same arrangement, for in it there are only slight indications of the outgrowth of the mucous membrane of the pharynx into this cleft. In most Anura, however, that membrane does form an outgrowth, and leads into a tympanic cavity, which is closed externally by a tympanic membrane. Among the Reptilia, the Ophidii, and Amphisbasnoidea have no tympanic cavity ; in Chamaileo there is no tympanic membrane ; but both these parts are present in all other Reptiles, and in Birds. In the Crocodilini and Aves, the inner openings of the Eustachian tubes are united into a common canal, as is the case also in Pipa among the Amphibia. Those parts of the visceral skeleton which are connected with the 538 COMPARATIVE ANATOMY. bony labyrinth, unite to form the apparatus of the auditory ossicles; the homologies of which have not yet been definitely made out for the different classes. The first portion is formed by an ossicle (operculum) which closes the fenestra ovalis ; in the Urodela this is either flat, or provided with a stalk-like process. Sometimes it is cartilaginous and its stalk ossified (Siredon) ; sometimes the reverse is the case (Menopoma). In the Ccecilias they are both ossified. The same arrangement obtains in the Ophidii (Eurystomata), where a small jnece of bone (columella) reaches to the quadrate. When there is a tympanic membrane present the columella is connected with it; for its cartilaginous end, which has often a peculiar form given to it by processes, sinks into that membrane. The lining tissue of the tympanic cavity then surrounds part of the columella, and causes this bone to appear to be more or less placed in the tympanic cavity. These arrangements are first seen in the Anura, and are still further developed in the Saurii, Chelonii, Cro- codilini, and Aves. The process of the columella is in some Birds (Dromams) connected with its plate by two limbs; in other cases it is simple, or is connected with the plate by one enlargement only. Just the same relations are seen to obtain in the columella of the Mammalia ; with this modification, however, that it is never directly attached to the tympanic membrane. It is converted into the stapes, the form of which, in the Monotremata and many Marsupials, calls to mind the columella. In the monodelphous Mammalia it is ordinarily divided into two limbs, which carry the plate. The other auditory ossicles are the incus, which is connected with the stapes, and the malleus, which is attached to the tympanic mem- brane by a styliform process. A connection between the tympanum and the fenestra ovalis, which was previously effected by a single bone — by the columella alone — is now effected by it, and two other bones. This " chain " of auditory ossicles is, for the most part, at any rate, placed in the tympanic cavity, for it is covered by the mucous membrane which is continued into that cavity from the pharynx, through the Eustachian tube. The tympanic cavity itself has, however, another relation, for it is principally formed by the tympanic bone, in addition to the boundaries provided by the wall of the labyrinth ; this tympanic bone commences as the framework of the tympanum. § 403. The external ear is derived from a prolongation of the edges of the first branchial cleft. In the Amphibia, Reptilia, and Aves, these parts are either altogether absent, or are only present in individual cases, where they have been developed in consequence of various kinds of adaptive changes. Thus, in the Crocodilini a fold of the integument forms an operculum above the tympanic mem- brane, and in the Owls there is a movable membranous valve. As early as the Saurii the tympanic membrane is removed some way from the surface, so that there is a short " external ALIMENTAEY CANAL OF VEBTEBEATA. * i 539 & auditory meatus." The external auditory meatus in the Mammalia is different, for its deeper portion is formed By the tympanic bone. The external ear, the cartilaginous support of which is continuous with a narrow auditory meatus, is attached to this. There is no external ear in the Monotremata. The "external ear" may be much modified, either in form, or in its relations to the muscular apparatus, which moves it. In addition to the muscles, which move the whole of the external ear, and which are sometimes of much power even in Man, there are others which are placed in the cartilage of the ear itself ; these are partly represented, though of course as rudi- mentary organs, in the human ear. This external ear is still more atrophied in aquatic Mammalia. Reduced in Otaria, it is alto- gether absent in the rest of the Pinnipedia, as it is also in the Sirenia and Cetacea. Alimentary Canal. §404. The alimentary or enteric canal of the Vertebrata forms a tube, which runs be- low the axial skeleton, and in which two chief portions can be distinguished mor- phologically, as well as physiologically, at a very early epoch. The most anterior portion is directly connected with the body-wall, and, as it is perforated by branchial slits, it functions as a respira- tory organ, for respiratory apparatuses are developed in the vascular arches between the clefts. This portion does not, therefore, belong exclusively to the digestive organs, although it is used in the ingestion of food. It forms the respiratory cavity, at the end of which the nutrient canal, in the strict sense, commences; this is separated from the body-wall by the pleuro-peritoneal cavity. The Vertebrata have these two portions of the enteric tube in common with the Tunicata. In the Acrania the respiratory chamber of the enteric tube occupies a very large portion, which, as in the Ascidias, represents a large r Fig. 303. Amphioxus lauceolatus (x2^). a Mouth,surrounded by cirri. b Anus, c Abdominal pore. d Branchial sac. e Gastric portion of the enteron. / Cajcuni. g Hind-gut. /i, Coelom. i Notochord, be- low which is the aorta, which accompanies it for nearly its whole length. 1c Aortic arches. I Aortic heart. m Enlargements of the branchial arteries, n Heart of the vena cava, o Heart of the portal vein (after Quatrefages). 540 COMPARATIVE ANATOMY. part of the body. This space is gradually reduced in size in the Craniota ; it still, however, retains its respiratory function, but many other organs are also differentiated in it ; these are, largely, accessory organs for the ingestion of food. Respiratory Ante-chamber (Cephalic enteron). §405. In Amphioxus this portion is bounded, in its most anterior region, which is close to the cavity which carries the mouth, by a ciliated apparatus ; there are a number of movable processes also at that point, which are directed towards the lumen of the tube, and so prevent the entrance of foreign bodies. The ante-chamber (Fig. 303, d), which occupies about two-fifths of the whole length of the body, has its walls broken through by a large number of obliquely-set clefts ; these form a complicated framework, the sup- ports of which have been already (§ 353) mentioned. The water, which is taken in by the mouth (a), passes through the clefts, and so to the exterior. But as two lateral dermal folds are gradually con- tinued ventrally over the surfaces on which the clefts are placed, and become united below, a peribranchial cavity is formed, which opens by a special pore (c). It should here be remembered that there was something similar to this in the Ascidias (§ 310). But it would not be correct to suppose that the two structures are morphologically identical. A vascular plexus is distributed in the walls of the clefts, the water that streams past effects respiration, the clefts function as branchial clefts, and the whole cavity represents functionally a branchial cavity. There are many special points in this arrangement in Amphioxus, such as the want of symmetry in the branchial frame, and its inde- Fig. 30-t. Vertical median section of a larva of Petromyzon. o Mouth, v Velum. h Ilypobranchial groove. n Spinal chord, ch Notochord. a Otocyst. c Heart (after a drawing by Calberla). pendence of the metamerism of the body; so that there is altogether a great difference between this apparatus and that of the Craniota. BEANCHLE OF VEETEBEATA. 541 The region of the body which is occupied by the branchial cavity corresponds to a head, for the nerves which go to it arise from the myelencephalon in the Craniota. Viewed thus the branchial cavity represents a cephalic enteron. Its nutrient and respiratory signifi- cance is the cause of various differentiations in it, which are partly arrangements which are peculiar to the Vertebrata, and partly arrangements which have been inherited from a lower condition. In addition to the branchial clefts, the ventral groove (hypobranchial groove), which is developed on the ventral surface of the branchial cavity, belongs to the latter series ; this has just the same relations as in the Tunicata (cf. p. 402) ; it is found in the larvae of the Petromyzontes, where it forms a grooved depression, enclosed by ridge-like edges (Fig. 304, h). In Amphioxus this structure is also present. Its presence in various stages of metamorphosis in all Craniota, not only brings these forms into closer connection, but is an indication of their genetic relations to the Tunicata, which must not be forgotten (cf. § 416). Branchia3. § 406. In the Craniota the branchial clefts are universally much reduced in number, as are also, in correspondence with this, the arches of the branchial skeleton. This phenomenon must be regarded as the degeneration of a primitively larger number of these struc- tures, such as is found in Amphioxus; it is compensated for by the increased size of the surfaces which carry the respiratory vascular plexus. This increase in size is implied by the development of the gills, whereby the blood-vessels, which, in the Acrania, are distributed over a large number of arches, are limited to a less extensive region, and are therefore arranged on a smaller number of arches. The essential point in the formation of branchias in these animals is the increase of the surface which is directed towards the respired medium, and this increase may be effected by means either of lamellre or of cylindrical processes. The branchial arches are provided with various forms of these organs, which enclose the well-developed respiratory vascular network. We find that in the Cyclostomata these organs have special characters, which have but little resemblance to what is found in Amphioxus ; their earliest condition is most like what is seen in the Gnatho- stomata, for the branchial clefts are simply spaces in the body-wall (Fig. 304). They are differentiated into tubes, the median portion of which has its lumen widened, and forms a branchial pouch (Fig. 305, br). Branchial lamellae are raised up from the wall of the branchial pouches in the form of leaf -like folds, in which the respiratory vascular plexus is spread out. Each branchial pouch is connected, by an "internal branchial duct," with the anterior 542 COMPARATIVE ANATOMY. section of the enteric tube. An external branchial duct (br) leads to the exterior. There are several variations in the characters of the tAvo canals which spring from each branchial pouch. The inner ones either open each separately into the digestive tube (Bdellostoma, Myxine) (Fig. 305), or they all unite into a median respiratory tube which runs below the diges- tive tube, and being connected in front with the digestive tube carries water to each of the branchial pouches (Petromyzon). The external branchial ducts either open sepa- rately on the sides of the body (Bdello- stoma, Petromyzon), or all the ducts of one side are united into a branchial pore (s) which lies behind the branchial apparatus ; on the left side a special canal (c), which comes from the oesophagus (ductus ceso- phago-cutaneus), also opens into the same pore (Myxine). These different forms may be derived from one another; in the case both of the inner and the outer branchial ducts that condition should be regarded as the primitive one, in which there is a direct connection between the respiratory chamber and the surface ; while, on the other hand, the formation of the respiratory tube, and the union of the external branchial ducts, is the result of a subsequent differ- entiation. § 407. Fig. 305. Kespiratory organ of Myxine g 1 u t i n o s a, seen from the ventral sur- face, o (Esophagus, i In- ner branchial clucts. br Bran- chial pouches, br' External branchial ducts, which unite on either side into a common branchial duct, which opens at s. c Ductus cosophago- cutaneus. a Auricle, v Ven- tricle, ah Branchial artery, giving off a branch to each gill, d Lateral wall of the body turned outwards and backwards (after Joh. Miiller). In Fishes, the branchial pouches are more closely related to the skeleton. The phasnomena seen in them lead to the con- clusion that each arch of the primitive branchial skeleton carried gills. The upper part of the first (mandibular) arch is not excluded from this ; as is clear from the frequent presence of a gill in the opening, which is found in many Selachii — the so- called spiracular cl_eft — between the first and second arches (mandibular and hyoid arches). The spiracular canal, which repre- sents a degenerate branchial pouch, is suc- ceeded by the true branchial pouches, of which there are, as a rule, five ; and rarely six or seven (Notidani). The wall of the first pouch is supported iu front by the hyoid arch, and behind by the first (i.e. by the third primitive) branchial arch ; the other pouches have just the same characters. In each of them a septum (. Ganoidei fibbbbbbbbb and Teleostei : — fi' Bl B2 B* B* By the degeneration of the septa between the branchial pouches, the whole gill apparatus is made more compact, and no longer therefore extends back into the region of the trunk, as it does in the Selachii ; it is confined to the base of the skull. Whilst in the Selachii the projecting septum (A s) forms an organ of protection for the succeeding branchial pouch, a similar organ is formed in the Chimaera?, Ganoidei, and Teleostei, from a single arch — namely, from the hyoid ; the integument on this arch grows backwards and covers all the gills, and is developed, in the Ganoidei and Teleostei, into the opercular apparatus and the branchiostegal membrane, with their various skeletal pieces (§ 354) (B op). § 408. In the Teleostei four arches are ordinarily beset with branchial lamella?, the fourth arch having a single row only, or there are but three arches which carry lamella?. When the lamella? of the fourth arch, and the posterior row on the third arch disappear, the fourth branchial cleft is ordinarily closed. Perhaps one of the most im- portant of the modifications which affect the lamella? themselves is seen in the villous gills of the Lophobranchiata. In some divisions of the Teleostei, the branchial arches seem to be so metamorphosed as to be able to retain the water in the branchial apparatus. The organs of the Labyrinthobranchiata arc of this kind; separate branchial arches or parts of such, are modified, to form coiled lamella- like processes, which give rise to a portion which is placed above the gills (Anabas, Poly acanthus). Another apparatus, which is found in various Clupeida?, consists of a spirally-coiled tube (branchial coil), which is formed by a diverticulum of the superior pharyngeal mucous membrane. This tube is generally connected with the superior segment of the fourth branchial arch, and has processes of its skeletal parts in its walls (Heterotis, Lutodeira, Meletta, etc.). The BEAXCHLE OF VEETEBEATA. 545 arborescent processes of the branchial arches, which are placed in special prolongations of the branchial cavity, where they support a respiratory vascular plexus, also belong to this series (Hetero- branchus, Clarias). Diverticula of its investing mucous membrane have the same respiratory function as the cavity itself. Thus, in Saccobranchus, a long tube extends from the branchial cavity, on either side, as far as the lateral trunk muscles ; in Amphipnous there is a similar sac behind the head, which opens just above the first branchial cleft. Both these organs contain respiratory vascular plexuses. ■ § 409. External gills in the form of integumentary structures were not primitively possessed by the Vertebrata, for the so-called external gills of the Selachian embryo are nothing more than filaments of the internal gills which protrude through the branchial cleft. Gills, how- ever, may come to the surface, and take the form of tegumentary pro- cesses; such gills may be seen in the young stages of Polypterus; cer- tain gills of Protopterus, and the gills of the Amphibia generally are of this character. In the Amphibia the gills have the appearance of two or three pairs of branched processes, which spring from as many branchial arches. In the Perennibranchiata this apparatus is permanently functional. In the rest of the Amphibia (Caducibran- chiata) these external gills disappear ; in the Anourous forms, where they are found for a short time only, they are replaced by shorter internal gills. A membrane which grows from before backwards covers the gills, so that there is only one efferent orifice. The orifices on either side may continue to grow out, and get nearer to one another, so as to unite into a single ventral orifice. When the larval stage ceases, the inner and outer gills of the Derotremata and Salamander are atrophied ; in the latter, as in the Anura, the branchial clefts are completely closed, but in the Derotremata a cleft is left on either side. When the gills disappear, the branchial cavity, which constitutes the respiratory antechamber, is converted into the primitive buccal cavity, which is limited by essentially the same parts as it was before. Branchial Clefts, and Palate of the Amniota. § 410. In the Amniota, also, the arrangement which has been trans- mitted from their branchiferous ancestors is retained during certain stages of embryonic life, in the form of clefts in the wall of the pharynx. These branchial or visceral clefts are never more than four in number, and they appear in such a way from before 2 N 54.6 COMPARATIVE ANATOMY. backwards, that when the last has appeared the anterior ones have already undergone certain changes. They are all gradually atrophied, and completely disappear, except the first, part of which is converted into the middle and outer ear (cf. supra, § 402). The degeneration of the embryonic branchial clefts is an impor- tant point of difference between the Amniota and the Anamnia, but, in addition to this, there is a new peculiarity which is due to a differentiation of the primitive buccal cavity. This leads to the formation of the secondary nasal cavity, and of the secondary buccal cavity. The remnant of the primitive buccal cavity, which lies behind, and is not affected by this process, forms the Pharynx. The cartilaginous portion of the ethmoid, which separates the two nasal cavities, and is broad in the Amphibia, is developed in the Amniota into a thin vertical lamella (Fig. 307, e) — the internasal septum. It remains partly cartilaginous, and is partly converted into, and develops bony structures, which were treated of under the cephalic skeleton. A second change is brought about by hori- zontal ridges or processes, which are given off from the maxillary process of the first arch, and which gradually form a plate (Fig. 307, p), the palate, which divides the primitive buccal cavity into two compartments. This plate Fig. 307. Diagram of forms the floor of the UPP^ or nasal cavity («), the differentiation of the and the roof of the lower one (m). When the primitive buccal cavity internasal septum reaches this palatine plate into nasal cavities (»»), i{. separates the nasal cavity into two por- and a secondary buccal . r . , „ , . , J ,r cavity (m). p Palatine tions, into each oi which the nasal canal plate, c Internasal wall, now opens, while its external orifice is coin- cident with that of the bifid nasal cavity. The posterior orifices of the nasal cavity, the choanre, which are separated by the palatine plate from the buccal cavity, and by the vertical internasal wall from one another, open into the pharynx. Very various stages in the arrangement of these palatine plates may be observed. In the Ophidii, Saurii, and Aves, the process of separation is less complete, the posterior nares form a longitudinal cleft, owing to the palatine processes uniting anteriorly, but being separated from one another posteriorly. They are sometimes separate in Birds, in which case they are exceedingly small. In the Crocodilini they are placed farther back than in any other forms, while in the Mammalia they do not open into the secondary buccal cavity, but into the pharynx. This latter region is thereby — as also by the opening into it of the Eustachian tube, which is developed from the first visceral cleft — shown to be a portion of the primitive respiratory antechamber. In Eeptiles and Birds the palate is supported by -pieces of the Skeleton (vide supra) ; in Mammals the hinder portion is formed of soft parts, which form the " velum palatinum." NASAL CAVITY OF VERTEBRATA. 547 Nasal Cavity. § 411. While the nasal cavities are increased in length, owing to their being shut off from the buccal cavity by the palate, the increase in the size of the facial portion of the head also affects them ; they increase both in length and height, and thus become large spaces. The olfactory nerve ends in their superior and posterior portion only (regio olfactoria), while the inferior and anterior portion principally serves as an ff air-passage/'' and consequently comes into relation with the respiratory organs (regio respiratoria). The whole differentiation therefore of the nasal cavity is seen to be connected with the development of the lungs, and their increased physiological importance. The increase in the extent of the internal cavity is effected in various ways. The lateral wall of the nasal cavity, which is developed from the primordial cranium, always takes part in this process; the turbinate bones are lamellar, folded, and coiled processes of this wall. In the Reptilia there is only one turbinate bone; this extends backwards from a cavity, which commences at the external nasal orifice, and is generally horizontal in position ; it is feebly developed in the Chelonii, and best developed in the Crocodilini. It is very varied in character in Birds. Sometimes it is simple (Columbae), sometimes complicated by coils (Raptores), or it may be cleft into several lamella? (Struthio). A turbinated structure is connected with the internasal septum in front of, and below this bone, and is by this connection distinguished from the turbinate bones, which are always lateral in position. This pseudo-concha separates the vestibule of the nose from the internal nasal cavity. Above the turbinate bone, and, as a rule, at the upper blind end of the nasal cavity, there is another process which corresponds to a depression formed in the Avail of the nasal cavity by an air sinus. Part of the olfactory nerve ends on this process, which is not found iu the Columbidre. In the Mammalia three turbinate bones may be distinguished. The lower one corresponds to the single bone in the Reptilia and Aves ; it varies very greatly owing to the way in which its lamella? are ramified and variously coiled, e.g. in the Carnivora (it is most complicated in Lutra and Phoca). These bones are least developed in various Marsupials (Macropus, Phascolomys), in the Apes (they are simplest in the Platyrrhini), and iu Man. In the Cetacea the cavity has undergone degeneration in consequence of the loss of its olfactory function. The orifice on the upper surface of the skull leads into a vertical canal, which is divided by the inter- nasal septum, and which can be shut off from the pharyngeal cavity by an occlusor muscle ; there are no signs of any turbinate bones in it. 2 n 2 548 COMPARATIVE ANATOMY. §412. There are accessory organs belonging to the nasal cavity. These are : 1) Accessory cavities of the nose. These are formed by the sinking of the mucous membrane of the nose into parts of its firm wall. They are first seen in the Crocodilini, where there is a cavity in the side walls of the nasal cavity, which communicates with it. In Birds we frequently meet with connections between the nasal cavity and the spaces in the neighbouring bones. In the Mam- malia the nasal cavity communicates with a number of cavities in different bones of the skull, the most important of which are the sinus frontales. These are cavities which are placed in the frontal bone, and which are either single, or divided into smaller portions ; they are very greatly developed in the Ruminantia. There are other communications with the sphenoid ; these are greatly developed in the Elephant, for example, where the cavities extend through the parietal and temporal bones as far as the occipital condyles. Lastly, there are connections between the nasal cavity and the maxilla ; these form the sinus maxillaris, which is developed in Marsupials and Ruminants, and very largely in the Solidungula. In Pi-iniates they are less extensive, and they are not present in most Carnivora, Edentata, or Rodentia. 2) Glands. There are larger glands connected with the nasal cavity -in addition to the glandular structures which are ordinarily found on the mucous membrane of the nose. When they are more developed they may also extend outside the nasal cavity. Such nasal glands are found in the Amphibia and in the Ophidii, as also in the Saurii and Crocodilini; in the former they lie outside the upper jaw, and in the latter they are enclosed in a maxillary sinus. In Birds also there is an external nasal gland, which is sometimes placed on the frontals, and sometimes on the nasal bones. Among the Mammalia also we find a gland on the sides of the face, but it is absent in several orders. 3) Organ of Jacobson. This is a canal placed at the base of the nasal cavity ; it is generally attached to the nasal septum, and communicates at the palate with the buccal, though it is shut off from the nasal, cavity; its walls, which form various kinds of pro- cesses, carry the ends of a branch of the olfactory nerve, which passes down the sides of the septum. In the Ophidii and Saurii the canal is partly enclosed by the vomer; in the Mammalia these organs are elongated, and are continued, as the ducts of Stenson, through the incisor canals, to the surface of the palate ; they are best developed in the Ruminantia and Rodentia (§ 396). Buccal Cavity. § 413. When the primitive antechamber of the enteric tube is divided into the nasal and buccal cavities, by the formation of a palate, a BUCCAL CAVITY OF VERTEBEATA. 54 Appendices pyloricue. r Hind-gut. FOEE-GUT OF VEETEBEATA. 557 once into the stomach, which can only be distinguished from it by the differences in the characters of its mucous membrane. As a rule, the stomach (Fig. 313) forms a cnecal sac, which is directed back- wards, and from which a narrow portion (pyloric tube) which bends forwards, can be distinguished ; this leads to the mid-gut (/). This is the case in all Selachii and Ganoi'dei, and in many Teleostei, while the rest vary greatly in the absence, or the great development backwards, of the cascal sac. Among the Amphibia we find a lower stage in Proteus, for the enteric tube, which has a perfectly straight course, has no stomachal enlargement at all. In the other Urodela, however, the stomach forms a wider portion of the enteron ; and this is the case also in the Anura, where the stomach is sometimes, indeed, placed transversely (Bufo). Among the Reptilia, the fore-gut is of a lower stage in the Ophidii and Saurii, owing* to the greater width of the ceso- phagus and the straight course of the stomach. However, there is an arrange- ment in the Saurii which calls to mind the pyloric tube of the Selachii, and from this the stomach might gradually acquire a transverse position. In the Chelonii and Crocodilini the oesophagus is more sharply separated from the stomach, which in the former has a large and a small curvature, owing to the great elevation of the pyloric portion. Owing to the approximation of the cardiac end of the stomach to the pylorus, this portion is rounded in the Crocodile, and is also distinguished by a tendinous disc on each face of its muscular wall ; in this point it resembles the stomach of Birds. In the fore-gut of Birds there is a greater division of labour. The influence of adaptation to the mode of life, and here especially to the mode of nutrition, is most clearly shown by the varia- tions in the different arrangements. The oesophagus, which is of the same length as the neck, is either of equal calibre along its whole course, or is provided with a widened portion (Fig. 314, A), or with a cascal diverticulum (B), which looks like an appendage. Portions (/) of this kind, which are characterised by modifications of the glandular organs of the mucous membrane, form a crop (ingluvies). This is best developed in carnivorous and graminivorous Birds ; in the former, iudeed, it generally forms a spindle-shaped enlargement, while in the latter it forms a unilateral diverticulum, which is dif- ferentiated into a caecal appendage, in many provided with a narrow connecting piece. Fig. 314. A Fore-gut of a Eaptorial Bird (Buteo). B Of a Fowl, oe (Esophagus. i Crop, pv Glandular stomach. v Muscular stomach, d Duode- num. 558 COMPARATIVE ANATOMY. The next portion of the oesophagus, which is generally narrower, passes into the stomach, in which two divisions can be made out ; the first is known as the proventriculus (A B pr) ; its walls are greatly thickened by a glandular layer. The second portion is characterised by the great development of its muscular layer, the strength of which varies with the mode of life of the animal. Where it is greatly developed we may observe a tendinous disc on either side {A B). In the Raptores, as also in many Natatores that live on animal food, the muscular layer is feebly developed. It is very strong in the graminivorous forms (Gallmas, Anatinte, Columba?, Pas seres). This portion, which serves for the comminution of food, aud compensates for the absence of masticatory organs, may be provided with other arrangements also which serve the same purpose; its inner surface may be covered by a firm horny layer, which is often of considerable thickness, and functions as a radula. It is produced by a glandular layer, the secretion of which passes into this firm stiff condition. In the Mammalia the fore-gut is more completely divided, owing to the sharper delimitation of the oesophagus from the stomach, than it is in almost any other division. In many cases the shape of the stomach is of a low type. In the Phocidas it retains its position parallel to the long axis of the body, while in other Mammals a position transverse to this axis is the common one. We must regard a number of peculiarities, which sometimes consist in an enlargement of the internal space, at others of a differentiation of the primitively single, and, as we must suppose, uniformly functional stomach, into several portions of different function, as the results of adaptation to the material of nutrition. The first relation is implied by the transverse position of the stomach, in consequence of which the great curvature gets to be much tho larger, and, forming a swelling behind the cardiac portion, gives rise to the fundus of the stomach. This is absent in most Carnivora, but is developed in the Monotreinata, Marsupialia, Ro- dentia, and Edentata, and is found also in most of the Primates. When the fundus is more largely developed the stomach may be divided into several portions, but this division is not unfrequently implied by the characters of the mucous membrane only (Equus). This arrangement is carried farther by the development of a trans- verse constriction ; thus, in many Rodents, the stomach is divided into a cardiac and a pyloric portion, to which smaller diverticula may be added on. Similar stomachs of a more complicated character may be seen in many Marsupials (Halmaturus), and in the Cetacea. The fundus is always a considerable enlargement, which, in the Cetacea, is succeeded by a number of diverticula, which are attached to the pyloric portion ; these give the stomach the appearance of being made up of from four to seven spaces which communicate with one another by connecting pieces of varying width. In the Ruminantia the complication is due to the share taken by the oesophagus, the cardiac end of which bulges out on one side and FORE-GUT OF VERTEBRATA. 559 fuses with the stomach, of which it forms two divisions. The first has the character of an enlarged fundus, and is known as the rumen or paunch (Fig. 315, I) ; it functions essentially as an organ for the reception of the large quantity of food that is ingested. Just below the cardia it is connected with the second division, the reticulum (II), which is succeeded by the psalterium (omasus) ; this third portion is wanting in the Tragulidas and Tylopoda. The last portion, which is formed from the pyloric part, is attached to this ; it forms the abomasus, in the mucous membrane of which the rennet glands are placed. A groove (oesophageal groove) which leads from the oesophagus into the reticulum, and is shut off by a valvular pro- cess (Fig. 315, B s) from the first two divisions of the stomach, represents that portion of the oesophagus which has entered into the formation of the stomach, andformedthe first twopor- tions of that organ by bul- ging out on one side. Thanks to its presence the food that has passed from the reti- culum into the oesophagus, and from thence into the mouth, can be directly returned, after it has been sufficiently masticated, into the psalterium and abomasus, while, when the groove is open, the fodder passes easily into the paunch and reticulum. The influence of the food in determining the size of the various portions may be seen from the differences between the paunch and the abomasus at different periods of life. The abomasus is relatively large in the calf, while later on the paunch may be as much as ten times larger than the abomasus, and even more than that. Fig. 315. Stomach of an Antelope. A From in front. B Opened from behind, oe (Esopha- gus. /Rumen. //Reticulum. ///Psalterium. IV Abomasus. p Pylorus. s (Esophageal groove. Mid-gut. § 419. The mid-gut (small intestine) which is generally separated from the stomach by a circular fold, the pyloric valve, is characterised at its commencement by having glandular organs (liver and pancreas) connected with it. With regard to length it is the most variable portion of the enteric tube. It is straight in the Gyclostomata, some Teleostei, and in Chimsera. In the last it is distinguished by 560 COMPARATIVE ANATOMY. reduced to almost nothing in a spiral fold, which is greatly developed in the Selachii, and divides the greater part of the mid-gut into a number of more or less closely applied coils (Fig. 31 G, G vs). In Carcharias this fold has the form of a rolled-up sheet of paper. This spiral valve is retained in the Ganoidei : it Lepidosteus. It is not present in the Teleostei An enlargement may be observed at the commencement of the mid-gut of the Selachii ; in the Sturiones there is a large, and externally much diverticulated glandular organ at the same point ; it is divided internally into a number of spaces corresponding to the diverticula. In Lepidosteus the various portions are more sharply separated from one another, and have the appearance of groups of short csecal tubes which beset the pyloric portion of the mid-gut, and, as in many Teleostei, form the appendices pyloricas (Fig. 31 G, A B ap). They beset a certain portion of the mid-gut and vary in oe Fig. 31G. Enteric canal of Fishes. A Of Salmo salvelinus. B Of TrachinnS radiatus. C Of Squatina vulgaris, oe CEsophagus. v Stomach, dp End of the air-duct, p Pylorus, op Appendices pyloricas. d Ductus choledochus. vs Spiral valve, i Mid-gut. c Hind-gut, ts Its appendage. number and size. They sometimes open separately into the gut, sometimes are united into larger trunks, and. give rise to branched structures. They are most numerous in the Gadidas and Scom- beroi'da?. In many Fishes the different creca are held together by connective tissue and united at a common efferent duct, in which case they have the appearance of a compact gland (Scoruberoi'das), while their affinity to the gland in the Sturgeon is implied by the frequent union of their orifices. In many Teleostei the mid-gut is much longer than the tract of the coelom which is given up to it, and it is then arranged in coils (Fig. 31G, B i), or in several ascending and descending loops. This implies an adaptation to the cavity of the ccelom, whilst the elongation of the tract, which is always derived from a straight MID-GUT OF VEETEBEATA. 5G1 rudiment, is an adaptation to the functions required of it by the ingesta. In the Amphibia the simple condition of the mid-gut is very rarely permanent (Proteus). It generally forms, as it "does also in Reptiles, a longer tube, and, consequently, a number of coils. (Fig. 317, i.) In the Ophidii these are least, in the Chelonii they are con- siderably, and in the Crocodilini they are still more developed. The mid-gut is very greatly elongated in the larvas of the anourous Amphibia, where this portion forms a long loop arranged in spiral coils. It is reduced when the mode of feeding is changed during the final stages of larval life, and this leads to an abbreviation of the length of the enteron. The length of the mid-gut in Birds also varies very greatly according to the characters of their food. It is arranged in loops, the first of which (duodenal loop) is the best developed, and always contains the pancreas. The mid-gut of Mammals is seen no less dis- tinctly to vary in length according to the kind of food that is eaten ; so that there are different conditions in Carnivorous and Herbivorous forms. The surface of the mid-gut is increased by various arrangements of its mucous membrane, as well as by its increase in length. In the lower groups there are coarser folds (spiral valve of the Selachii), but in the Amphibia and Reptilia by far the most common arrangements are fine longitudinal folds of the mucous membrane. These obtain also in the Birds, but in them they generally form unequal elevations, and may be united by transverse lines. Fine folds arranged in zigzag liues are seen in the Amphibia and Reptilia, and are found also on the mid-gut of Birds. In Mammals, these longitudinal folds of the mucous membrane are commonly found in the Cetacea ; but in most of the other Mammalia the mucous membrane is smooth, or raised up into transverse folds, which are very generally beset with villi. When the folds are feeble we find that these villi are greatly developed in Birds also, while when the folds are present the villi are merely smaller elevations. Fig. 317. Enteric canal of Menobran- chiis lateralis. p Commencement of the fore-gut with the Pharynx, oe (Eso- phagus, v Stomach. i Mid-gut. r Hind- gut. Hind-gut. § 420. The end- or hind-gut is the smallest of all in the lower divisions, and is merely represented by a short and somewhat wider 2 o 562 COMPARATIVE ANATOMY. piece (Fig. 313, r; 316, G c). In the Selachii it is provided with a special glandular appendage (Fig. 31 6, Ox) . It is only in the Amphibia that, owing to its greater length and width, it becomes of some im- portance, but in them, as in the Reptilia, it retains its straight course in correspondence with its shortness. In consequence of this straight course it has got the name of " rectum." It is generally separated from the mid-gut by a transverse fold or valve. Many Reptiles are provided with a cascal appendage, which, in the Ophidii is feebly, and in the Saurii is better developed. The ca3ca in Birds are much more independent. In this group, also, the hind-gut is short and straight (Fig. 320). The cascum is generally paired, and is absent in a few families only (e.g. Woodpecker, Psittacus, etc.). They vary greatly in the extent to which they are developed, so that they may form short papilliform appendages, or very long tubes (Apteryx, Gallinas, Anseres). The hind-gut is longest in the Mammalia, where it forms the large intestine, and is distinguished, as such, from the mid-gut, or small intestine. Owing to its greater length it is arranged in coils, so that the terminal portion, only, has the straight course taken by the hind-gut of other Vertebrata. The anterior portion ordinarily forms a loop which bends from the right side of the abdominal cavity forwards, and then to the left, and then again backwards to be continued into the rectum. This loop is sometimes broken up into secondary loops. At the boundary between it and the small intestine crecal struc- tures are likewise developed, but these are rarely arranged in pairs (Fig. 318, c d), and are commonly single. The size of this caecum may be shown to depend on the food. In the Carnivora it is short, and sometimes com- pletely absent (Ursina, Muste- lina) : it is very large in the Herbivora, where its length is compensated for by that of the colon. The coscuni itself may be affected by differentiations. Its terminal portion is frequently diminished in size (e.g. in various Prosimias and many Rodents) (Fig. 318, c). In various Primates, and in Man, the terminal portion, which, at first, is as wide as the rest, is not developed in proportion to it ; it thus becomes more and more distinct from the other portion, which continues to grow wider, until at last it forms a mere appendage to it — the appendix vermiformis. The hind-gut primitively opens into the same space as the urinary and generative ducts, the cloaca. This arrangement, which ob- tains in the Selachii, Amphibia, Reptilia, and Aves, is permanent in Fig. 318. Crecum and colon of Lago- mys pusillus. a Small intestine. b Opening of the larger (c), and of the smaller (d) ca3cum. e f g Diverticula of the colon (after Pallas). APPENDAGES OF THE MID-GUT OF VERTEBPATA. 563 the Monotremata only, among Mammals, in the rest of which it is confined to the embryonic stage, and subsequently the hind-gut opens to the exterior by means of an anus, Organs appended to the mid-gut. §421. Two large glandular organs, the liver and pancreas, are connected with the beginning of the mid-gut ; they are both dif- ferentiated from the walls of the rudimentary enteron. In Amphioxus an organ, which must be regarded as the liver, has the form of a caacal tube (Fig. 303,/), which arises close to the commencement of the alimentary canal, and is directed forwards (Fig. 303,/). It is provided with an epithelial investment of a greenish colour. A similar condition is seen in the Craniota during the earliest stages of development, in which the rudi- ment of the liver has the appearance of a paired diverticulum (//) of the enteric tube, lying behind the rudimentary stomach (Fig. 319, J). It is partly formed by the epithelial layer of the rudimentary enteron (endoderm), and partly by the external layer developed from mesoderm. As Reptiles, Birds, and Mammals agree in this point, this condition must be regarded as a funda- mental one, while at the same time it calls to mind the morphological characters of the hepatic organ in Amphioxus and many Iuvertebrata (Vermes, Mollusca). Owing to the thickening of the splanch- nopleure and its large connection with the venous portion of the vascular system, together with the simultaneous thickening of the endoderm, relations are produced, which distinguish the liver of the Craniota from that of the Acrania, as well as from that of the Invertebrata. While the first rudiment of the liver appears as a diver- ticulum, the later differentiations are brought about by the thickening of the endoderm, and give rise to solid chords of cells which grow into the layer of mesoderm, and the vascular apparatus embedded in it ; these give off new buds, and are finally connected together in a retiform manner. The parenchyma of the liver is formed by these primitively solid chords, and their secondary o 0 o a O -i Fig. 319. Rudiment of the enteric canal and its ap- pendages in an embryo of the Dog, seen from the ventral surface, a Diver- ticula of the enteric tube towards the visceral clefts. b Rudiment of the pharynx and larynx, c Of the lungs. d Of the stomach. / Of the liver, g Walls of the yolk- sac in connection -with the mid-gut. h Hind-gut (after Bischoff). 561 COMPARATIVE AXATOMY. and otlier processes, while they give rise to the bile-ducts by the formation of intercellular passages, which run in the axes of the epithelial chords. The hepatic lobes, which are formed on either side, fuse with one another into a single organ. The two primitive diverticula, after they have formed the bile-ducts in the parenchyma of the liver, and have been continued into the network of cellular chords, form the efferent ducts of the liver. The liver, which is thus differentiated from the enteron, forms a compact, and ordinarily, very large organ; it is embedded in a fold of the peritoneum, which extends from the anterior portion of the enteric tube to the anterior wall of the abdomen. In Fishes, the liver generally forms a single, undivided mass, but sometimes it consists of two, or more lobes. There are two large portions in the Amphibia ; it is gene- rally simple in the Ophidii, and is merely notched at the margin in the Saurii ; in the Crocodilini and Chelonii it is again divided into two lobes, which in the latter are widely sepa- rated from one another, and united by a slender transverse bridge. Ordi- narily two lobes are, sometimes more, sometimes less, indicated in the Mam- malia. In the Carnivora, Rodentia, some Marsupialia, Simias, and others we find, indeed, multilobate forms, but these may be referred to two larger primary lobes. There are various modifications in the character of the efferent ducts (ductus hepato-enterici) in relation to their primitively double character ; for either the first condition persists, or the two ducts are gradually fused to- gether, that is to say, the diverticulum of the enteron is converted into a single duct, or, lastly, the primitive ducts are atrophied, and secondary canals are converted into efferent ducts: in this case there is a largfe number of ducts (in the Saurii and Ophidii). A unilateral csecal diverti- culum, the gall-bladder (Fig. 320,/) is placed on these ducts; it has very various relations, and is by no means a constant structure. The pancreas is developed in the same way as the liver — from a diverticulum of the wall of the enteron, which is developed behind the rudiment of the liver. The Fig. 320. cinerea Enteric canal of Ardea i (Esophagus and crop. pv Proventriculns. v Gizzard. v' Antrum pylori. d Duodenal loop, it Mid-gut. b Hind-gnt. c Part of the single caecum. cl Cloaca and Bursa Fabricii. h Liver, dh Ductus hepato-cn- tericus. /Gall-bladder.

ement in the Gano'idei and Teleostei. The carotids arise from the first branchial vein, or from the anterior end of the paired arterial trunk, which collects the branchial veins on either side to form the roots of the aorta, and then unites with its fellow of the opposite side to form the aorta, or enters anteriorly into a transverse anastomosis which marks off an 586 COMPARATIVE ANATOMY. • arterial circulus eephalicus at the base of the skull. A special optic artery is given off from the vessels of the rudimentary gill, with which a direct branch of the first branchial vein (Selachii), or a branch of the same vessel which surrounds the body of the liyoid (Teleostei), is connected. There are numerous modifications in the mode of origin and arrangement of the different vessels, the most important of which are found in the carotid and optic arteries. This portion of the vascular system is arranged in a similar manner in the Amphibia. In the Perennibranchiata the cephalic arteries arise from the anterior portion of the roots of the aorta; in the Caducibranchiata, from the first permanent aortic arch, or, they are continuations of the anterior arch itself (Pig. 331, c). In this case, an artery which goes to the tongue (?) represents an external carotid. After this vessel is given off, there is a swelling (c) on the carotid trunk in the Frogs and in the Sala- manders ; this is the so-called carotid gland. The lumen of the vessel is here traversed by a network of bands, which break it up into a number of narrower passages, just as if a capillary network had been intercalated in the course of an artery. The carotid gland appears to be de- rived from an arrangement of this kind, the branchial vascular network not having been com- pletely atrophied. The next pair form the aortic arches (ad as), which converge backwards, and finally unite into an unpaired aortic trunk (a). Each aortic arch gives off a subclavian (sd ss). Just before they unite, a large visceral artery (rn) is given off from the left aorta. The pulmonary artery represents a last aortic arch. Before it goes to the lungs (p) it gives off a large cutaneous branch (cut), which ramifies on the back and neck as far as the posterior region of the head, and affords a distinct proof of the respiratory function of the integument. _ In the earliest stages of the Amniota we meet with many similar arrangements of the arterial system. The internal carotid, which supplies the brain and the eye (Fig. 332, A B, c), is seen to be a forward prolongation of the roots of the aorta on either side. The external carotid is a branch of the third primitive aortic arch. If this arch loses its connection with the fourth, the two carotids are Fig. 331. Arterial system of the Frog. ha Bulbns arteriosus. c Carotid. C Carotid gland. I Lingual _ artery. ad Eight, as Left aorta, a Aortic trunk. m Visceral artery. sd Eight, ss Left subclavian, oes Oesophageal branches. p Pulmonary artery, cut Its cutaneous branches, occ Posterior cephalic branch. HEART AND ARTERIES OF VERTEBRATA. 587 given off on either side of a common trunk (G) . They generally appear as two trunks, which pass along the sides of the neck in company with the vagus. In the Saurii they do not lose their connection with the next arterial arch, and retain, therefore, their primitive relations. In many Ophidii the right common carotid is atrophied, and may even completely disappear. Fig. 332. Development of the great arterial vascular trunks, as seen in the embryoes, A Of a Eeptile (Lizard), B Of a Bird (Fowl), and C Of a Mammal (Pig). The two first pairs of arterial arches have disappeared in all three ; in A and B the third, fourth, and fifth are still persistent; in C the two last are alone complete, and the third is no longer connected with the fourth pair. A branch (p) of the fifth forms the pulmonary artery. Its trunk from this point to the aorta forms the ductus Botalli. c Carotis externa, c' Carotis interna. In A and B this forms the anterior prolongation of the root of the aorta, and in C a common trunk with the external carotid. a Auricle, v Ventricle, ad Aorta descendens. s Branchial clefts, m Rudiment of the fore-limbs, n Nasal pit (after H. Bathke). In Birds also this artery arises in company with a subclavian from a common trunk (art. brachiocephalica), hut it leaves its primitive course, and lies in the middle line of the inferior surface of the cervical vertebras ; on the left, however, it retains its original course. In others, again, the two carotids both leave their old course, and this leads on towards that third form, in which the two closely-approximated vessels are fused together. In this case no part of the right carotid runs by itself, but a vascular trunk arises on the left side, which runs in the middle line, and which passes to the head, under the name of the primary carotid. Many Birds have this character in common with the Crocodilini. A single carotid trunk (Fig. 329, c), which obtains in the Ophidii and various 'Saurii, must be regarded as differing from this, although it also passes into two cephalic arteries anteriorly. This arrangement is due to the approximation of the points of origin of the two carotids on the right aortic arch. A common arterial trunk is given off from the united point of origin. Another peculiarity is the presence of an unpaired subvertebral artery, which runs forward from the right aortic arch along the vertebral column (Fig. 329, sv). 588 COMPARATIVE ANATOMY. There are various kinds of modifications in the Mammalia, which are due to similar changes in the vascular trunk during em- bryonic life. Among others these specially affect the ter- minal branches of the carotids ; the internal carotid, as in various Lizards and Birds, is not sent to the cranial cavity and sensory organs only. The arteries of the fore- limbs arise from various and very different points, so that transmission appears to be of less importance than adapta- tion in the development of these vessels. The trunk of the aorta is continued along the vertebral column, without any change in character, as far as the por- tion which is set apart for the caudal region — caudal artery ; when the tail is shortened, it forms the median sacral artery. When the so-called chevron bones are present, the terminal portion of the aorta always lies in the canal formed by them. But in various Fishes it may be enclosed in a canal, formed by processes from the centra of the vertebras, in the region of the trunk also ; this is the case in the Sturgeon, and in some Teleostei. The aorta gives off arteries (arteriso intercostales) for the metameres in regular succes- sion, as well as the vessels which go to the viscera, and also those which go to the hind- limbs, when these are well developed. In Fishes there is ordinarily only one large arterial trunk for the viscera (A. cceliaco-mesenterica) ; but in some there is a mesenteric artery also. The aorta gives off a larg-er number of arteries for the renal and generative organs. As Fig. 333. Heart and great vessels of Buteo vulgaris, tr Trachea, i Crop, ae Com- munication between the air-sacs and the lungs. b Bursa Fabricii. ao Aortic arch. aad Art. anonyma dextra. aas Art. anonyma sinistra. ps Art. pulmonalis sinistra, c Carotid. am Visceral artery. vci Com- mencement of the inferior vena cava. vcm Vena coccygeo-mescntcrica. VEINS OF VERTEBEATA. 589 in tho Amphibia, the Art. coeliaco-mesenterica arises from the end of the left aortic arch in the Reptilia (Saurii, Chelonii) ; this arch is only connected by a narrow tract with the right one ; or there are several visceral arteries (some Sanrii) ; these are especially nnmerons in the Ophidii, in consequence of the elongated form of their body. In the Crocodilinii, also, independent mesenteric arteries are given off from the unpaired aorta in company with the arteries from the left aortic arch. In Birds, where the left aorta disappears, the aortic trunk is the sole vessel from which the visceral arteries are given off. In Mammals the coeliac and superior mesenteric artery are tho chief arteries of the enteric canal. In the placental forms the inferior mesenteric is also a large vessel. The large number of renal arteries found in Fishes are found also in the Amphibia and in most Reptiles ; in Birds also there are several renal arteries, the middle one being given off from the ischiac artery. It is very rarely that there are a number of these arteries in the Mammalia. The arteries of the hind-limbs are not direct branches of the posterior aorta until these parts are very largely developed. Tho two chief trunks (iliac arteries) of this region are not always tho same. As is clear from their topographical relations to the pelvis, different branches may supply the area of these arteries. In tho Sauropsida the ischiac are the chief trunks of the hinder extremities. In the Mammalia the chief trunks are formed by the crural artery, and there are numerous modifications in its more special characters which are of less importance. Venous System. § 437. Tho venous system of the Vertebrata exhibits no less important phamomena in the various modifications that obtain in it, as we pass from the Fishes to the Mammalia, than does the arterial portion of the circulatory system. Our knowledge is in many points as yet incomplete. The blood which returns to the heart is, in Fishes, collected into four longitudinal trunks, two anterior and two posterior. Those of either side pass into a transverse trunk (ductus Cuvieri, Fig 334, dc), which opens with that of the opposite side into a sinus (sv) which is placed behind the auricle of the heart. The anterior pair, which chiefly collects the venous blood of the head, forms the jugular veins (J), which are placed above the branchial arches ; the hinder pair, which receives the blood from the walls of the trunk and from the renal and generative organs, forms the cardinal veins (c); an unpaired caudal vein runs below the artery in the caudal canal ; in the Cyclostomata, Selachii, and some Teleostei, this divides into two branches which are continued into the cardinal 590 COMPARATIVE ANATOMY. veins of their own side. In many Teleostei, this caudal vein is continued into the right cardinal by a large, and into the left cardinal by a smaller branch ; in this case the left cardinal vein is also smaller than the right one. This leads to the condition in which the whole of the caudal vein passes into the right cardinal ; this has been observed in a number of Teleosteans. As the caudal vein sends off branches into the kidneys, which either break up completely or partially in this organ, these branches form the venas renales advehentes, and send their blood into the cardinal veins through the vena3 reve- hentes. In this way the renal portal system is developed. A second vascular apparatus of similar character has its roots on the digestive canal ; its venous blood is carried to the liver by a trunk which is known as the portal vein. It is distributed in this organ, and is carried to the common venous sinus by hepatic veins, which are generally united into several trunks. In this arrangement of the venous system in Fishes we may distinguish the paired, and ordinarily symmetrical, portion from the unpaired portion, which is solely represented by the hepatic veins. We will first follow out the former in its changes throughout the Vertebrate series, for in all of them its essential characters, at any rate, may be observed in the early stages of development, as a transmitted arrangement, and, since it is the groundwork of the embry- onic venous system, it furnishes the starting- point for all later metamorphoses. Fig. 334. Diagram of the primitive venous system. j Jugular, c Cardinal vein. dc Ductus Cuvieri. h Vena3 hepaticee. sv Sinus venosus. § 438. Fig. 335. Anterior por- tion of the venous sys- tem of an embryonic Ophidian, v Ventricle. ha Bulbus arteriosus, c Auricle. DO Ductus Cuvieri. vc Cardinal vein, vj Jugular, vu Um- bilical vein. {/Primitive kidney. I Rudiment of the labyrinth (after H. Rathke). In the Amphibia and Reptilia the venous sinus receives the two jugular veins, which have the same area of origin as in Fishes. They persist in all the higher Vertebrata, while the hinder pah- — the cardinal veins (Fig. 335, vc) — have only the same characters as in Fishes, during the earliest stages of em- bryonic life. They are the veins of the primi- tive kidneys (U). Their anterior portion is obliterated, while their posterior portion re- ceives veins from other regions, and forms the venas renales adve- hentes. Even before the disappearance of that part of the cardinal veins which opens into the ductus Cuvieri, four other trunks are developed in the Reptilia, which chiefly receive the intercostal veins, VEINS OF VEKTEBEATA. 591 and are known as the vena? vertebrales. The anterior and posterior ones on either side are united, and open into the jugular vein of their own side. Their connection with the left jugular disappears later on, when the left vertebral veins develop transverse ana- stomoses, and become connected with the right ones, and, like them, open into the right jugular. When the cardinal veins cease to be connected with the ductus Cuvieri they form prolongations of the jugular veins, which receive the subclavians which come from the fore-limbs, and are known as the superior vena? cava?. The vertebral veins, which collect the blood from the walls of the body, are not large after the embryonic period, and they are generally considerably atrophied. They cease to have a paired arrangement (Ophidii), and the greater part of their area is occupied by the vena cava inferior. We meet with similar arrangements in the Birds. A pair of jugular veins, which are often unequally developed, form the chief trunks for the blood returned from the anterior parts of the body. At the base of the skull they are generally connected with one another by a transverse trunk, into which the veins from the cervical vertebral column, as well as from the head, may enter. When the left jugular is atrophied, this transverse trunk forms the vessel by which the blood is conveyed into the right jugular. The vertebral veins are now inconsiderable vessels. The jugulars unite with the veins of the anterior extremities, which form the subclavians, and the trunks thus formed are again known as the superior vena? cava?. As these still receive the posterior vertebral veins, a portion is separated off from them, which may be seen to be derived from the transverse trunks (ductus Cuvieri), which are persistent in Fishes. These vena? cava?, however, open separately into the right auricle, for the sinus, which is persistent in the Reptilia, here forms a portion Fig. 336. Relations of the great venous trunks on the heart. I Reptile (Python). II Bird (Sarcorhainphus). Ill Marsupial (Halrnaturus). IV Pig. They are all seen from behind, i Vena cava inferior, s Vena cava superior sinistra. d Vena cava superior dextra. ap Pulmonary artery, a Aorta, sv Sinus venosus. of the auricle (Fig. 336, I, sv). The vertebral veins in Birds pass along a canal which is enclosed by the ribs, by which point they are seen not to be the same vessels as the cardinal veins. 592 COMPARATIVE ANATOMY § 439. The embryonic venous system of the Mammalia is completely similar to that of the lower Vertebrata. Two jugular veins (Fig. 331) receive the cardinal veins, and the common trunks on either side pass into a venous sinus, which is connected with the auricle, and, later on, forms a part of the right one. Two distinct venous trunks then open into this auricle, each of which is continued into an anterior and larger, and a posterior and smaller, trunk. When the anterior extremities are developed, the subclavian veins (s) fall into the anterior ones (Fig. 337, A), and the two venous trunks thus formed are distinguished as the superior venas cavse. When the system of the inferior venas cavas is developed, the area of the cardinal veins is diminished, for part of the blood which was col- lected by the cardinal veins is now carried to the inferior venaa cava). The cardinal veins also undergo de- generation, owing to some of their roots passing into the new longitudinal trunks, which, as in the Reptilia, represent the vertebral veins, and are continued into the end of the cardinal veins, which open into the ductus Cuvieri. Owing to the decrease in the size of their Fig. 337. Diagram of the primitive paired veins in Mammals. A The vertebral have taken the place of part of the cardinal veins, which are indicated by dotted lines. B The left jugular has its lower portion atrophied, and its area is united with that of the right jugular by a transverse trunk. C The left jugular vein has completely disappeared, with the exception of a rudiment on the heart, j Jugular. s Subclavian. cs Vena cava superior, c Cardinal vein, v Vertebral vein, cor Coronary vein, cms Vena azygos. area, these vertebral veins (Fig. 337,^-7?,^) appear to be branches of the trunks derived from the ductus Cuvieri and the jugular veins, or superior venas cavas. These are found in the Monotremata, Marsupialia, many Rodentia, and Insectivora. In others, part of the area of the left superior cava (li) is handed over to the right one (cs), owing to the development of a transverse anastomosis; the loft superior vena cava is now atrophied (Rodentia, Ruminantia, Solidungula). When this arrange- ment is complete, the greater part of the trunk of this vein disap- pears, and the only part that remains is the terminal portion, which primitively formed the left ductus Cuvieri, and which is placed between the left ventricle and auricle {0,cor); the cardiac veins open into it, and it forms the sinus of the coronary vein of the heart. In Man, even, a semilunar fold separates this siuus from the true VEINS OF VEETEBEATA. 593 coronary vein, while the valvula Thebesii, which is placed at its opening into the right auricle, forms, for a long time, the valve of the left superior vena cava. The right superior cava is now the sole anterior trunk (Cetacea, Carnivora, Primates). When the trunk of the left superior vena cava is reduced, the cardinal veins, or the vertebral veins developed in their area, undergo great changes. While in the first case they open into the vena cava of their own side (A), and while in the second they pass separately into the left side of the right auricle, owing to the development of a right vena cava (B); when the vascular passage which leads directly to the heart is reduced, they become connected with the right vertebral vein. The left vertebral vein is connected with the right one by transverse anastomoses, and when the connection between its upper end and the left superior cava is broken, it is converted into a vena hemiazygos, while the right one, which still retains its primitive position, becomes the vena azygos (Fig. 339). When the two superior cavte persist, the two vertebral veins are not, in all cases, unchanged; one trunk often becomes much larger than the other, which may even be so far reduced as to disappear altogether. In this case a vena azygos receives the intercostal veins of both sides ; and this sometimes opens into the left, and sometimes into the right superior vena cava, or even into the only one that is present, as, for example, in the Carnivora (Fig. 337, G, az). In most Mammals the roots of the jugulars are formed of a number of veins from the external and internal cephalic regions, one of which conducts a part of the blood from the cranial cavity through the jugular foramen. It forms a small vessel only, for the greater part of this blood passes out in a canal (canalis temporalis), which is either placed between the petrosal and squamosal, or in the latter only. When the foramen jugulare is enlarged, the vein, which in other cases is a small one, increases in size, and gradually becomes the most important of all the vessels which come from the skull ; in this case it forms, as it does in the Primates, the internal jugular vein. The other veins gradually unite to form the external jugular, which is the most important one in most Mammals. § 440. The second large venous tract is very small in Fishes, for in them it is merely represented by the hepatic veins, which are united into one or more trunks, and open into the common venous sinus. When the tract of the cardinal veins is diminished in extent, a new tract is formed in connection with the hepatic veins — that of the inferior vena cava (Amphibia). This venous trunk collects blood from the kidneys, and is, therefore, the vena renalis revehens (Fig. 338, A, ci). The blood from the hinder extremities passes into an iliac vein (A, /), which, in the Urodela, receives on either side a branch of the divided caudal vein. It breaks up in the liver, and forms a veua renalis advehens. A branch of the iliac vein passes to the middle 2 q 594 COMPARATIVE ANATOMY. line of the abdomen, and receives veins from the so-called urinary bladder (A, o), after which it is united with its fellow of the opposite side to form a single trunk which passes to the liver, and which is therefore connected with the portal system (a) ; this is the epigastric (abdominal) vein. The veins of the digestive canal and of the spleen are united into a portal trunk, which breaks up in the liver. In the Reptilia also the hepatic and the efferent renal veins form an inferior vena cava (B, ci), which opens into the common venous sinus below the right superior cava. But there are various modifi- cations in the different divisions of the Reptilia, and it is in the Saurii and Ophidii only that there is any close similarity to the arrangement of the venous system which obtains in the Amphibia. The caudal vein divides into two trunks, which receive the veins from the hinder extremities in the Saurii, and form the vena3 renaj.es Fig. 338. Posterior portion of the venous system. A Of the Frog. B Alligator. C Bird. R Kidneys, c (azygos trunk) Caudal vein, c Crural, i Ischiac vein. v Venae vesicales. a Epigastric (abdominal) vein, m Vena coceygeo-mesenterica. ra Vena renalis advehens. rr Vena renalis revehens. ci Vena cava inferior, h (in A and C) Vena hypogastrica, (in B) End of the epigastric vein in the liver. advehentes. The veins of the vertebral column are connected with these. Similar arrangements obtain in the Crocodilini, where the caudal vein (B, c) is also divided, but this vessel then forms a transverse trunk which gives off the venae renales advehentes (ra). In all these forms the venas renales revehentes form a trunk which runs in front of the vertebral column, and there is a renal portal system in the kidneys ; this appears to be absent in the Chelonii only. Another venous tract in the Reptilia is represented by the vena3 epigastricse sive abdominales. When the allantois is developed, a pair of veins is developed from the vascular network that ac- companies it, and this primitively opens at the same point as the VEINS OF VERTEBRATA. 595 ends of the ductus Cuvieri (Ratlike: Ring Snake). These umbilical veins receive veins from the abdominal wall, and are also connected with the formation of the hepatic portal circulation. In the Ophidii this umbilical vein disappears after the veins of the abdominal wall which open into it have broken up into a plexus, but in the Saurii the terminal portion of one umbilical vein persists, and. unites with the abdominal veins that open into it to form an epigastric vein ; this receives veins from the urinary bladder, and passes forwards to the liver. In the Crocodilini and Chelonii the ends of the umbilical venous trunks persist, and form, as they are continued into the veins of the abdominal wall, a part of the epigastric veins. Like the single veins in the Amphibia and Saurii they also go to the liver, and, in the Crocodilini, they are connected with branches of the portal vein. In the Chelonii, those of either side unite into a transverse trunk, which receives the various venge intestinales, which are not, in them, united into a portal venous trunk. In both cases they are distributed in the liver, and belong therefore to the portal venous system. In the Crocodilini, as in the Chelonii, the epigastric veins (B, a) are given off from the two branches of the caudal vein (c), and receive the crui^als (c) ; they also receive the ischiac veins more anteriorly. But, since in the Crocodilini, the venae renales advehentes also arise from the caudal vein and its connection with the ischiac vein, part of the venous blood from the hinder portion of the body is carried into the renal portal circulation, and the rest into that of the liver. But in the Chelonii, where there are no advehent renal veins, all the blood from the hinder end of the body is carried into the liver, for the vertebral veins, in these forms, also open into the epigastric veins. § 441. Several of the veins which are found in the Reptilia are not permanent structures in Birds. The inferior vena cava (Fig. 338, (7, ci) is indeed still made up of two trunks from the kidneys, but these receive the veins of the hind-limbs (c), and might from their size be taken to be the continuation of these veins. Two hypo- gastric veins (h) are connected with these trunks in addition to the vessels which arise from the kidneys. They are united by a trans- verse anastomosis at the root of the sacrum ; this anastomosis receives the caudal veins (c) from behind, and gives off in front a vena coccygeo-mesenterica (m), which goes to the mesenteric vein. This vena coccygeo-mesenterica is a wide trunk in the Crocodilini also, where it anastomoses with the transverse trunk, which unites the two branches of the caudal vein ; part of the venous blood from the tail or hinder extremities is conducted away from the renal portal circulation by it. In the Mammalia there are no indications whatever of a renal portal system. The umbilical and omphalo-mesenteric veins have the same relations as in Reptiles, though there are several variations 2 q 2 59 G COMPARATIVE ANATOMY, in some, even of the larger, trunks. The inferior vena cava (Fig. 339, ci), which collects the blood from the kidneys and generative glands, is developed very ea:4y ; it accompanies the united umbilical veins, and, when the right one disappears, it receives the left. After the cardinal veins (c) disappear, the veins of the pelvis (Jiy) are connected with the end of the trunk of the vena cava, as are also the veins of the hinder extremities (il), and the caudals. At the time when the umbilical is the largest of the veins, the inferior cava appears to be merely a branch of it. Where the umbilical vein enters the liver, the hepatic vessels are formed, while at the same time similar branches from the liver pass to the point where the umbilical unites with the vena cava inferior ; these form the hepatic veins. As the blood which is returned to the heart from the umbilical veins passes through the liver, that portion of them which lies between the afferent and efferent veins is atrophied, and forms the ductus venosus Arantii. That portion of the omphalo- mesenteric vein which receives the mesenteric veins is then converted into the trunk of the portal vein, while the hepatic branches of the umbilical form the branches of the portal vein after the obliteration of the ductus Arantii. The inferior vena cava is thus converted into the chief hinder trunk, into which open the veins of the pelvis, of the hinder extremities, of the renal and generative organs, while the veins of the digestive canal and spleen form the portal vein. Fig. 339. Diagram of the chief trunks of the venous system of Man. cs Yena cava superior, s Vena sub- clavia. je Jugularis ex- terna. ji Jugularis in- terna, az Vena azygos. ha Vena hemiazygos. c Indication of the car- dinal veins. ci Vena cava inf. h Venoo hepa- ticre. r Vena) renales. il Vena iliaca. hy Vena hypogastrica. § 442. The blood-vessels are ordinarily distri- buted in the body by the gradual branching of the different trunks, until at last the finest branches of the arteries and veins give rise to the capillary system, which connects the two kinds of blood-vessels with one another. To say nothing of the various special arrangements in certain organs, a somewhat different method of distribution obtains in the blood- vascular apparatus of several regions of the body. A vein or artery suddenly breaks up into a tuft of fine branches, which either do, or do not anastomose, and which lose themselves in the capillary system, or are again soon collected into one trunk. This distribution of the vessels has been long known as a rete mirabile. Its function is clearly to slacken the blood-current, and to increase the surface of the walls of the vessels, so that there must bo a change in the amount of nutrient fluid diffused by osmosis. If from this rete a vascular LYMPHATICS OF VEBTEBBATA. 597 trunk is given off similar to the one that was broken up, it is called bipolar or amphicentric ; if the rete remains broken up, then it is known as a diffuse, unipolar, or monocentric rete mirabile. Some- times arteries only, sometimes veins only (rete mirabile simplex), sometimes both kinds of vessels are united with one another (rete mirabile geminum seu conjugatum) to form the rete. Arterial retia are found in the pseudobranchia, in the choroid of the eye of Fishes, and, in very various forms, on the air-bladder. In Birds and Mammals, retia are not unfrequently found in the area of the carotids and their branches. They are very common on the limbs of the Mammalia (Monotremata, Edentata). In the area of the visceral arteries there are both arterial and venous retia ; thus, in the Pig, the mesenteric artery forms an arterial rete. Arterial retia are very common on the terminal branches of the renal arteries, where they form the Malpighian glomeruli, from which, as we all know, another artery is given off to bo distributed in the capillaries on the urinary tubules (cf. Fig. 343, B). Lymphatic System. § 44 o O. The presence of a system of canals connected with the blood- vascular system — in which the nutrient fluid, which has passed out from the capillary portion t)f the hasinal system, is conveyed again to the blood stream as lymph, after having filtered through the tissues — is an arrangement which is peculiar to the organisation of the Craniota. It appears to be correlated with a high develop- ment of the body, for it is wanting in Amphioxus, and in embryo- logical development it only begins to appear at a relatively late period, and not until the blood-vascular system has been differen- tiated into its arterial and venous portion, and is in full function. That portion of the lymphatic system which has its root on the digestive canal is of especial importance, for it receives the chyle, or nutrient material, which has been prepared from the chyme by the process of digestion, and conveys it to the blood-vessels. In addition to the function of conveying the lymph, this system of canals has yet another duty, which complicates its anatomical relations. The points at which the form-elements of the lymphatic fluid, or lymph-cells, are developed, are embedded in its vessels ; these lymph-cells are carried to the blood, and are gradually con- verted into its form-elements. In the lower divisions of the Vertebrata this lymphatic system has not much independence, for its vessels are chiefly formed of wide spaces, which enclose other organs, and especially arteries. The sheath of connective tissue of the artery also encloses the lymphatic vessel. The veins also may be surrounded by wide lymph-spaces ; 593 COMPARATIVE ANATOMY. thus, for example, the abdominal vein of the Salamander is enclosed in a lymphatic vessel. There are, however, other vessels in the lower divisions besides those which accompany the blood-vessels — in the skin, or on portions even of the digestive canal, or other viscera. Peripherally, the lymphatic vessels anastomose largely, and form capillary net- works or similar spaces. These gradually give off wider spaces, either in the form of canals, or of sinuses with irregular boundaries, the place of which is taken, in the higher Verte- brata, by vessels allied in structure to veins. Although we may note that, as we pass from the lower to the higher Vertebrata, there is a gradual differentiation from spaces, similar to those of the lacunar system of the Inverte- brata, to a definitely developed canalicular system, so that the interstitial nature of the lymphatic ducts is well marked at the periphery only; nevertheless, in the coeloni we have an arrangement which indicates the origin of the lymphatic vessels from a lower condition — for the coelom is a lymphatic cavity. In this point the coelom of the Vertebrata resembles closely that of the Invertebrata. The communications between the coelom and the pericardial cavity, which obtain in various Fishes (Sturio, Selachii), are indications of the same thing ; as are also the pleural cavities of the Mammalia, which are merely differentiations of the general coelom. Fig. 3 10. Portion of tbo Aortaof a Chelonian (Chelydra) surrounded by a lymphatic space, a Aorta, b Outer wall of the lymph sjmce, which is removed at b' so as to expose the blood- vessel, c Trabeculae. § 444. In Fishes the chief trunks have the form of lymph- sinuses. There are generally two pairs of them, or one unpaired one is placed below the vertebral column. The unpaired trunk divides into two branches. Smaller sinuses, and narrower canals, form the lymphatic vessels which are collected into these trunks. They are generally connected with the venous system at two points. A cranial lym- phatic sinus opens into the jugular vein of its own side, and two sinuses, which receive lateral trunks, are connected by a trans- verse anastomosis with the caudal vein near the last caudal vertebra. The subvertebral lymphatic cavity of the Amphibia forms a portion of the system which is about the same size as a very large subcutaneous series of lymphatic cavities, which are present in these animals ; in the anourous Amphibia especially, this latter series extends over a large portion of the surface of the body. The lymphatic vessels of the digestive canal (chyle-vessels) open into it, as do those of the other viscera, while it is connected also with the lymphatics of the LYMPHATICS OF VEETEBEATA. 599 extremities. In the Eeptilia the subcutaneous lymphatic cavities are more varied and numerous, and the system is more intimately related to the arteries ; the lymphatic vessels are sometimes wide spaces (Fig. 340), which surround the arteries and are traversed by trabecular ; sometimes they form plexuses which accompany these vessels. When their trabecular are more largely developed, the lymphatic cavity is broken up into several anastomosing canals. The space which surrounds the aorta is broken up, in the Crocodilini and Chelonii, into two trunks which surround the veins of the anterior extremity ; and lymphatic vessels from the head, neck, and extremities open into them. The lymphatic trunks of Birds have the same characters, but, in them, both the large trunk in front of the aorta (thoracic duct), and the small vessels are more inde- pendent. As in the Eeptilia, the thoracic duct opens into the superior venas cavae (venee brachiocephalics) . At the commence- ment of the tail the lymphatic system is also connected with the ischiac veins, or with the afferent renals, in which point they resemble the Amphibia and Reptilia. In the Mammalia the walls of the lymphatic system are still more differentiated, although it often happens that in them also the sheath of the arteries bounds the course of part of the lymphatic current. Where they do not accompany the blood-vessels they form frequent anastomoses, or wide-meshed plexuses, and are dis- tinguished, by valves, as are the same parts in Birds. The lymphatic vessels of the hinder extremities, as well as the chyle-ducts, unite into a chief trunk in the abdomen, which is rarely paired, and the origin of which is frequently distinguished by a considerable enlarge- ment (cisterna chyli). Thence they are continued into a thoracic duct, which opens into the commencement of the left brachio-cephalic vein ; the trunks of the lym- phatics of the anterior parts of the body (of the head and anterior extremities), and of the wall of the thorax, open into, and on either side of the same vein. The lymphatic trunks are generally widened out near their opening into the veins, and the wall of these enlargements is distinguished by its muscular investment; it executes rhythmic contractions. These arrangements are known as lymphatic hearts. They have been occa- sionally observed on the caudal sinus of Fishes, but they are more accurately known in the Am- phibia (Ranidae) and Eeptilia (Chelonii) ; in the former they are found on both the anterior and posterior openings into the veins, but in the urodelous Amphibia and in the Reptilia only the posterior lymphatic hearts have been made out. found also in the Ratitse (Struthio, Casuarius), and some Natatores, but in other Birds they have no muscular investment, and form mere Fig. sinus. 341. a a Caudal Anas- tomosing transverse trunk. b Lateral vessels c, and origin of the caudal vein d. Of S i 1 u r u s glanis (after Hyrtl). These latter are GOO COMPARATIVE ANATOMY. vesicular enlargements. Finally, in the Mammalia such structures do not seem to be developed. § 445. The apparatuses that produce the lymph-cells are simple in Fishes, where they are placed in the course of the various lymphatics; the cells are produced in the meshes of reticular connective tissue. Where more largely developed, this arrangement gives rise to local enlargements, which accompany the arteries, in consequence of the relation between these vessels and the lymphatics. This arrangement obtains even in the higher Vertebrata, although the cells are not always developed in the sheaths of the arteries. Follicular enlarge- ments are formed beneath the mucous membrane of the enteric canal, the lymphatics of which are connected with these cell-forming regions. They are either scattered, or variously grouped together (closed glandular follicles). At the commencement of the wall of the enteron these structures form the tonsils already mentioned ; in different parts of the mucous membrane of the mid-gut they are placed closer to one another, and form the so-called Peyerian Glands, which are present in the Reptilia, but are only found to auy great extent in the Mammalia. When a number of these lymphatic follicles are united together they form larger structures, lymphatic glands, which are also placed on the course of the lymphatic vessels. In Fishes, Amphibia, and Reptilia there are not, so far as we know, any true lymphatic glands. In Birds, also, they seem to be confined to the neck; it is in the Mammalia only that they are generally present, and in them they are found in other parts of the body, as well as in the chylif erous portion of the lymphatic system of the mesentery. In some Mammals (e.g. Phoca, Canis, Delphinus) the mesenteric glands are united into a single mass, the so-called pancreas Asclli. The Spleen is also one of the organs that form lymph-cells; in its histological structure it only differs from the lymphatic glands by the fact that the cells formed in it pass directly into the blood- vessels. Essentially it is formed of a fine lacunar system inter- posed between the efferent and afferent blood-vessels ; this forms the greater part of the so-called parenchyma of the spleen. The spleen is found in all Vertebrata save Amphioxus, and is always placed in the region of the stomach, and generally close to the fundus. It forms an elongated or rounded organ of a dark- red colour, which is sometimes, as in various Selachii, broken up into a number of smaller lobules, some of which are, in other cases, converted into secondary spleens. § 446. An organ which is very generally present, and which resembles the lymphatic glands in various points of structure, cannot be LYMPHATICS OF VEBTEBEATA. G01 passed over, although its relations to the lymphatic system are still very uncertain ; this is the thymus. This is an organ which is also made up of glandular follicles, and which is divided into larger and smaller lobes ; its smallest vesicles are filled with cells. In the Selachii this organ is placed on the branchial sacs, and between them and the dorsal muscles. In the Sturgeon, and some Teleostei, the similar follicles that are found on the superior hinder boundary of the branchial cavity are regarded as the same organ. In the Amphibia the thymus is a small swelling, placed behind the angle of the lower jaw. It has the same characters in the Keptilia. In the Ophidii and Chelonii it is placed on the carotid, and above the heart. In the Crocodilini, as in Birds (Fig. 312, th), it extends from the pericardium to the lower jaw. The lower portion is the larger in the Mammalia, so that it rarely passes beyond the thoracic cavity. In all cases it is best developed in early life, after which it undergoes atrophy, and it is very rarely that it retains its earlier size in the adult stage (Pinnipedia). In the higher divisions of the Vertebrata there is an organ which lies in front of the kidneys and on either side of the body ; it is con- sequently called the supra-renal gland, but we have no information at all as to its function. In the Anamnia these structures are re- placed by the investment of the sympathetic ganglia by means of a cortical layer made up of cell-containing tubes ; these form yellowish or whitish bodies, and are scattered over a larger portion of the body, whereas in the Amniota they form a mass on either side, while nerve-elements can be made out in their medullaiy substance. The relatively large size of this organ during foetal life is a note- worthy point. The function of these organs cannot be regarded as at all definitely known ; nor are we aided in our inquiry by classing it as one of the " blood-vascular glands " — a term which is altogether obscure, and consequently objectionable. Excretory Organs. § 447. The arrangements which are found subserving the purpose of excretory organs among the Invertebrata obtain also, in their most essential relations, in the Vertebrata, and are therefore indications of the affinity between the vertebrate phylum and lower forms, which in other morphological details are very remote. Organs of this kind have as yet been vainly sought for in Amphioxus ; but in all the Craniota they are found to exist, and to be formed on the same type. The type has been obliterated by gradual differentiation, but it is revealed by the study of individual development. The simplest stage is represented by a canal which runs in the dorsal wall of the coelom, and opens to the exterior posteriorly, and in the G02 COMPARATIVE ANATOMY. region of the anus, while anteriorly it opens into the ccelom by an internal orifice. Although it is clear that this arrangement has much in common with, the excretory organs of the Vermes, yet the pecu- liarity must not be overlooked, that although the vertebrate body is a metameric one, this archinephric duct is not a metamerised organ; it is not therefore completely homologous with the metameric looped canals of the Annulate Vermes. It must consequently be derived from a still lower condition, that is, from one in which the organism was not divided into metameres ; so that it represents, as does also the unsegmented chorda dorsalis, one of the, phylogenetically, oldest organs of the Vertebrate body. This archinephric duct has been observed to be derived from the mesoderm ; in its rudimentary condition it has the form of a solid chord of cells, or is differentiated as a groove from the epithelium of the peritoneal cavity (Teleostei). The rudiments of the canals (Fig. 342, t) are derived from the same parts ; these, either permanently, or for a time only, open by an infundibular orifice into the abdominal cavity, while they are also connected with the archi- nephric duct (Selachii, Amphibia). They develop coiled glands on their course, and so form the secreting portion of the primitive kidneys. At definite points a coil of arterial vessels (glomerulus) s\\ " A (\ )i pushes its way into a dilatation of these ^^r metamerically arranged canals, and gives rise to a Malpighian body, lying in a cap- sular enlargement. This last arrangement obtains in all forms of the renal organ, however much it may be modified in various members of the vertebrate group. The fundamental form' of this primi- tive kidney must be regarded as being a longitudinal canal, which receives trans- verse canaliculi, which open by ciliated inf undibula into the abdominal cavity ; this is the form which the rudimentary apparatus really has in the Selachii. The connection with the coeloin, the epithelial investment of which always gives rise to a large portion of this system, allows us to compare it with the excretory organs of many Vermes, and points to those distant forms in which these organs are the sole cavitary organs that are developed from the mesoderm (Platyhelminthes). The metameric arrangement of the open transverse canals is due to the general metamerism of the vertebrate organism. It must not, therefore, be regarded as the same as that of the looped canals of the Annelids, or even as derived from it, for those canals open to the exterior on the Fig. 342. Section of an Embryo of Pristiurus. ug Archi- nephric duct, t Rudiment of a funnel-shaped organ. cZEnteron. m Medullary tube. ch Noto- chord. a Aorta, v Veins. EXCRETORY ORGANS OF VERTEBRATA. 603 surface of the metameres themselves (§ 145), and not into a longitu- dinal canal. This canal is the part which, by being the first part to appear in the Vertebrata, defines the type of the whole apparatus. But, just as in a large number of Invertebrata the excretory organs partly lose their function, and serve as efferent ducts for the generative products, so too among the Vertebrata do we meet with a process of this kind, by which great changes are effected in the primitive excretory apparatus. It loses, and that often very early in life, its primitive arrangement. And when this is not seen in the embryo, it must be regarded as due to new, acquired, relations. 1 § 448. In the Cyclostomata, Teleostei, and Amphibia, a special portion of the primi- tive kidney appears at the most anterior end of the archinephric duct, which de- serves especial mention, as it does not only appear earlier than the rest of the primitive kidney, but is' generally sepa- rated by some distance from it. This portion is made up of a small number of canaliculi, which commence by ciliated infundibula, and are generally set in a coil. There may be only one canaliculus. A Malpighian body may sometimes be observed on the canaliculi. In the Amphibia this pronephron undergoes atrophy, while in the Amniota it does not seem to be even rudimentarily de- veloped. It persists, however, in the Cyclostomata, where it is provided with a tuft of ciliated infundibula, which project into the abdominal cavity. Among the Cyclostomata the primi- tive kidney is seen at its simplest in Bdellostoma. An elongated canal (Fig. 343, A B, a) gives off short transverse canaliculi at various points (b) ; the blind end of which (c) is constricted off, and encloses a glomerulus (B). The transverse canaliculi form the secretory apparatus (urinary canaliculi) ; the archi- nephric duct is the collecting tube, and functions as the ureter. In the Myxinoidea and Petromyzontes, the kidneys, which are set along the posterior third of the ccelom, are larger, but the urinary canaliculi have exactly the same relations. In both forms the ureter takes a lateral course to the abdominal pore ; but in the Fig. 313. A Portion of the kidney of Bdellostoma. a Urinary duct, b Urinary canaliculi. c Terminal capsule. B Portion of the same more highly magnified, a c As before. In c there is a glomerulus. d Afferent, e Efferent artery (after J. Midler). 604 COMPARATIVE ANATOMY. Petroinyzontes it is first connected with the oue on the opposite side, to form an unpaired and wider portion. We do not yet know its relation to the metameric ciliated infundibula. In the Selachii the primitive arrangement is limited to the early stages of development. The primitive kidney extends along the dorsal wall of the coelom, and is made up of separate canaliculi, which commence by ciliated iufundibula (Fig. 314, i), which open into the abdominal cavity. Each canal, after having broken up so as to en- close a glomerulus (m), is continued on into the archinephric duct. These canals increase in length, so that each of them forms a coiled lobule (r) ; each kidney, therefore, is composed of a series of these coils, which are col- lected together into the archinephric duct (u). This duct opens into the cloaca. Changes may occur in the glandular portion of these kidneys, as well as in their efferent ducts. The anterior portion, which is made up of a number of lobules, does not undergo any very great development, as does part of the hinder portion. This, which is made up of a varying, but large number of primitive lobules (13-14 in Acanthias), is converted into a larger organ, the canaliculi in which may be seen to increase in number by budding off new ones. This portion retains its renal function, while the anterior part is atrophied, and, in the male, enters into con- nection with the generative gland. The ciliated funnels (nephrostomata) are retained in some Sharks only ; they disappear in all the Rays, and many of the Sharks. Where they are retained they are reduced in number. Of the changes which obtain in the primary archinephric duct the most important is its division into two parts. This commences at its anterior end, and extends backwards, so that there come to be two canals. One commences at the anterior abdominal orifice of the primary duct, and has no further relations to the kidney. This is the Mullerian duct. The other canal retains its connection with the primitive kidney, and forms the secondary archinephric d dot. But even this portion may undergo certain changes, inasmuch as in the male it is converted into the seminal duct. The efferent ducts from the posterior portion of the kidney are then collected into a common ureter, which opens into the sinus urogenitalis, into Fig. 344. Portion of the kidney of an Embryo of Acanthias (dia- gram), i Ciliated funnel, m Mal- pighian body, r Renal lobules. u Archinephric duct. EXCRETORY ORGANS OF VERTEBRATA. 605 which, however, several ureters may open separately. In the females, also, the efferent ducts from the anterior and aborted portion of the primitive kidney are connected with the ureter. In the Ganoi'dei and Teleostei the kidneys have the same posi- tion. The primitive kidney appears to be considerably increased in size, while the efferent ducts are not so completely differentiated as in the Selachii, where they were the cause of much complication ; in the Ganoi'dei, however, the presence of a nephrostome, with a wide abdominal orifice on the efferent duct, speaks to the commencement of the process by which the Mullerian duct is differentiated; the ureter, therefore, no longer corresponds to the archinephric duct. In the Teleostei the secondary portion of the gland first appears on the anterior division of the archinephric duct, and forms that portion which, in many, extends as far as the head (head-kidney). The hinder portion, which is developed later, becomes connected with this. The whole forms a compact glandular organ, which is covered by the peritoneum, and extends along the vertebral column; it varies in size in different regions. Its differentiation into lobes is generally implied by the greater development of certain regions. The efferent ducts (Fig. 345, u) either pass along the anterior surface, or more to the sides ; they generally unite into an unpaired portion, which opens behind, or below the generative orifice. The ducts are widened at different points, either in the common, or in the separate portions ; these struc- tures function as " urinary bladders," but they have no morphological connection with the urinary bladder of the higher Vertebrata. The renal organs of the Amphibia have many points in common with those of the Selachii. The rudimentary ducts are always provided with func- tionally active nephrostomata. The primary ureters form lobules by becoming arranged in coils. In the Coecilia) they are all of much the same size, but in the Urodela and Anura the hinder ones are increased in size and number, so that this portion becomes much larger than the anterior part. The nephrostomata, also, are greatly increased in number in this region, and are persistent. In the Urodela the anterior portion of the kidneys receives the efferent ducts of the testes, while in the Ccecilia3 and Anura different parts of the kidneys are connected with these organs. The primary archinephric duct is differentiated so as to give rise Fig. 3-15. Kidneys of Salmo fario. R Kidneys, u Ureter. v Vesicular enlargement. ur Its efferent duct, rr Venaa renalesrevehentes. d Ductus Cuvieri. s Vena subclavia (after Hyrtl). 606 COMPARATIVE ANATOMY. to a MuUeriaHj and a secondary archinephric duct (Fig. 348). The latter serves as the efferent duct of the kidney, or ureter, in the Ccecilia3, Urodela, and female Anura, while in the males of many of these latter the primary archinephric duct appears to retain its original function. They open independently into the cloaca. Muller, W., Das Urogenitalsystem der Cyclostomen. Jen. Zeitschr. IX.— Semper, C, Das Urogenitalsystem der Plagiostomen. Arbeiten aus deni zool. Institut zu Wiirzburg, II.— Spengel, J. W., Das Urogenitalsystem der Amphibien. Ibid. III. [Balfour, F. M., A Monograph of the Development of the Elasmobranch Fishes. London, 1878.] § 449. The primitive kidney is likewise developed in the Amniota. For some time in development it extends through the ccelom, and projects into it from the dorsal wall of this cavity. The archi- nephric duct is again (Fig. 346,ng) the first part to be developed. The urinary canaliculi (u), which form the glandular portion of the organ, open into it. The hinder portion of the primitive kidney, which has always the same function, is well developed even in the Selachii, but still more so in some of the Amphibia ; this is effected by the increase in the number of the urinary canals, and by the formation of special efferent ducts. These processes indicate prophetically the relations of these parts in the Amniota. In the Reptilia the additional por- tion of the urinary canals is directly connected with the hinder portion of the primitive kidney (Lacerta), but it is not connected with it to form the same, but a new organ — the per- manent kidney. For a long time it is present in company with the primitive kidney, but it has its own ducts (ureters), and it takes on the function of the primitive kidney, in proportion as the latter is atrophied, or converted to the purposes of the generative system. In Birds the rudiment of the permanent kidney appears to be formed independently ; and this is still more the case in the Mam- malia. We see, therefore, that the so-called permanent kidney of the Amniota is at first an organ which is connected with and forms part of the primitive kidney, and that it is gradually separated from Fig. 346. Section through the embyro of a Bird (Fowl). A Amniotic cavity. am Amnion. ch Notochord. a Aorta. v Cardinal veins. u Primitive kidney. ug Archinephric duct. e Germinal epithelium. P Pleuroperitoneal cavity D Enteric groove. EXCRETORY ORGANS OF VERTEBRATA. G07 it botli in space and time. No rudiment of the neplirostomata has been observed, nor is the archinephric duct divided as in the Anamnia; the Mullerian duct has a separate rudiment. The kidneys of Reptiles and Birds somewhat resemble those of Fishes in their size and position. They are placed far back and close to the cloaca; in the Snakes only are they placed farther forwards, while at the same time they are longer. They vary very greatly in form, in consequence of the development of lobes. In Birds they are placed in depressions between the transverse processes of the sacral vertebrae, and are generally divided into three portions, which are sometimes connected with one another, and which may vary greatly in size. The ureters (Fig. 349, to) are generally placed on the inner edge of the kidney, and receive at various points larger urinary canals (Ophidii, Chelonii) ; or these canals are enclosed in the parenchyma of the kidney, and do not leave the organ except at its termination (Saurii, Crocodilini). In Birds a large part of the canal is outside the kidney. In all cases they open sepai*ately into the cloaca, or into a sinus urogenitalis, into which the genital ducts also open. The kidneys of the Mammalia vary in several points, and espe- cially as to the characters of the orifice of the ureters, after the differentiation of the rudiment which is known as the "renal canal. " The kidneys, which are developed at the blind end of the " urinary canal," are, after they are differentiated, placed behind the primitive kidneys. At first they appear to have a smooth surface, which becomes uneven when the glandular parenchyma is developed into separate lobes. In either lobe the urinary canaliculi are united together at a papilliform process, with which the common efferent duct of the lobe is connected. It forms the pyramid, and a number of these unite to form the pelvis of the kidney, from which the ureter is given off. The permanently distinct lobes are very numerous (about 200) in the Cetacea. There is a smaller number in the Pinnipedia. In many Carnivora, also, the lobes are separate (Ursus, Lutra), while in others they are fused. This gives a knobbed appearance to the surface of the kidney (e.g. in Hyaena, Bos, Elephas). In others there is a condition of this kind for some time, but when the cortical substance of the lobes is completely fused, the surface of the kidney becomes smooth, although the grooves that remain indicate its primitive division into lobes. Within the organs, however, the division is more or less completely retained, and the number of primitive lobes is implied by the greater or less extent to which the papillae are fused together. This fusion, further, may affect some, or all the lobes, so that the number of renal papillae may be much reduced ; at last, indeed, they may all unite into one (Marsupialia, Edentata, Rodentia, several Carnivora and Primates). The ureters formed from the renal canals, after they are separated from the archinephric duct, primitively pass into that portion of the allantois which runs in the abdominal cavity of 608 COMPARATIVE ANATOMY. the embryo, and is connected with the primitive cavity of the pelvic portion of the enteron (urachus). This is gradually con- verted into a fusiform widened organ — the urinary bladder, while the continuation of the urachus into the umbilicus, and from thence into the umbilical chord, is obliterated. The former portion forms the ligamentum vesico-umbilicale medium. The primitive (fusiform) form of the urinary bladder is retained in some Mammals (Seals), while in others it is gradually modified, and with these modifications there are correlated differences in the way in which the ureters open. Thus in many Rodents they open high up on the posterior wall of the bladder (Fig. 354, C, 11). The other characters of the efferent ducts are common to them and the generative apparatus, with which, therefore, they will be treated. Generative Organs. § 450. In the Vertebrata, the reproductive organs are shared by different individuals ; the separation of the sexes is the rule, although there are various exceptions to it in the class Pisces. In the higher divisions also there are various arrangements which are indications of hermaphroditism. But it seems to me that the point which is of real importance in this matter is the repro- ductive material, and that the characters of the efferent ducts are of no importance, for they were not primitively part of the gene- rative apparatus. Our knowledge of the earliest development of the male generative matter is not quite definite, but we know certainly that the female elements are derived from the epithelial layer which invests the abdominal cavity. In this there are points of likeness between the Vertebrata, and the Vermes among the Invertebrata. In Amphioxus follicular structures, covered by a layer of epithelium, and forming diverticula of it, are developed at various points in the coelom, or in the cavities connected with it; these structures are the germ-glands. The ova are formed in them, between indifferent and flattened cells, which form the stroma of the organ. In this character Amphioxus is very different to the Craniota, where the germ-glands are always developed at a sharply defined and less extensive region. The epithelial investment of the abdominal cavity retains its primitive character along a tract which corre- sponds to the rudiment of the primitive kidney longer than it does in other regions ; and this epithelial layer may be distinguished as the germinal epithelium (Fig. 346, e). At the side of the mesentery in this region there is an elevation of varying length, which is formed by a thickening of the connective tissue — the genital ridge. The epithelium dips into this, and forms the rudi- ments of the ova. Of a group of cells which grows inwards, one GENERATIVE ORGANS OF VEETEBEATA. GG9 cell becomes an egg, while the rest form a cellular layer around it — the follicular epithelium, which unites with the surrounding con- nective tissue to form the ovarian follicle. Each invagination of the germinal epithelium either forms a single follicle, as in the Anamnia (Selachii), or these groups of cells grow out and form the rudiments of a number of follicles, as in the Craniota. The cells of the ovarian follicle that are set around the egg generally remain indifferent, and aid in the nutrition of the egg as well as in the formation of the yolk-sac, which surrounds it. The egg itself, and the cells of the follicle which surround it, undergo more or less considerable modifications. When the egg and the follicle increase equally in size, the follicle-cells form a simple epi- thelial layer, as in Fishes, Amphibia, Reptiles, and Birds. But in the Mammalia they multiply while the egg-cell remains relatively small, and for a long time they fill up by far the greater part of the follicle. As this follicle grows a cavity is gradually formed in its interior which is filled with fluid ; this causes the cellular layer of the follicle to be extended around its wall (membrana granulosa), while at one point, which is somewhat thickened, it encloses the egg. The changes which obtain in the egg-cell relate to the yolk, and they are accompanied by an increase in the size of the egg. This may be seen in the Teleostei, where the granules of the yolk often undergo great metamorphoses. The same happens to the eggs of the Amphibia. In the Selachii, Reptilia, and Aves the yolk- granules are greatly increased in number, and are specially differ- entiated. Owing to the number present the ripe egg is of a considerable size. The region invested by the germinal epithelium is the point at which the male germ-glands are also developed, but it seems that this epithelium does not take any direct share in the formation of the testes. The earliest differentiation of the glandular tubes (seminal canals), which make up the testes, has not yet been observed ; the view that they are formed from a portion of the primitive kidneys is beset by the difficulty of their having no relationship of any kind with these organs. The form-elements of the sperm are developed by the dif- ferentiation of the epithelium of the seminal canals. In all Verte- brata these are movable filaments which are given off from a thicker portion of varying form — the so-called head. This head is dis- coidal or elliptical, as in many Mammals and Fishes, or elongated, as in the Selachii, Amphibia, and Aves. In the latter it is fre- quently coiled in a corkscrew fashion. The seminal filament of some Amphibia (S alamandrina and Toads) is distinguished by an undulating membrane. § 451. The germ-glands are developed from the structures known as genital ridges. Sometimes more and sometimes less of this ridge is converted into the ovary or testis. The simplest condition is seen 2 R 610 COMPAEATIVE ANATOMY. in the Cyclostomata. The ovaries of the Petromyzontes have the form of paired lamellas, which extend along the ccelom, and are thrown into a large number of folds, in which the ova are developed. The testes are similar in character. In the Myxinoidea the germ- glands are unpaired, and arise from the right side of the mesentery. Both sets of generative products are passed into the cosloin, whence they reach the exterior through the abdominal pore. The ovaries of some Teleostei have almost the same characters ; thus, in the Salmonidas the eggs are passed into the abdominal cavity, and are evacuated through the abdominal pore. The same is the case in Lcemargus borealis among the Selachii, where the ovaries contain much smaller eggs, and are themselves much larger. In the rest of the Fishes there are efferent ducts in both sexes, which are largely — perhaps always — due to the differentiations which affect the primitive kidneys (cf. § 448). In this relation the Ganoi'dei are of a low grade, for their germ-glands have no direct ducts, and their products are passed into the ccelom. In both sexes the products escape by an apparatus which is homologous with the Mullerian duct, consisting of a canal of varying length, and provided with an infundibular orifice, which is attached to the ureter (secondary archinephric duct) ; this takes up the generative products. This fact must be regarded as one of special importance, for we learn from it that the Mullerian duct may be turned to use in the male. The presence of this duct in both sexes leads to a correct apprehension of the real facts of the case, and renders it unnecessary to regard the presence of the rudiments of these organs in the male as due to a primitive hermaphroditism, which, cannot be shown to have obtained at the required stage of development. Two different arrangements can be derived from that which is dominant in the Ganoidei. One is seen in the majority of the ajv Fig. 347. Generative organs and enteric canal of Clupea harengus. oc (Esophagus. Stomach. ap Appendices pyloricce. i Enteron. a Anus, vn Air bladder. pn Air duct, s Spleen, tt Testes, vd Their efferent duct, g Genital pore. br Branchiae (after Brandt) . Teleostei, and the other in the Selachii, and from thence in the Amphibia and all Amniota. The male organs in all Teleostei and the female organs in all GENEEATIVE OEGANS OF VEETEBEATA. 611 except those above mentioned, have the tubular form. The germinal region is often limited to one portion of the tube, whence it extends more or less considerably, according to the degree to which its products are developed. The lateral efferent ducts of these genital tubes (Fig. 347, tt), are united into a common duct, which opens by the genital pore. In this arrangement the germ-gland is not ordinarily represented by the whole apparatus, but by the germi- nal region only, which projects on the inner wall of the tube, and is often lobed or branched. The germinal region is probably invested by the Mullerian duct, which is converted into a tubular form, but this has still to be established by embryological observation. In a number of Teleostei hermaphrodite arrangements have been observed, a testicular as well as an ovarian tube being developed ; this is best known in species of the genus Serranus. § 452. In the Selachii the arrangement which obtains in the female Ganoi'dei is retained and further developed. The germ-glands are generally developed on a certain portion only of the genital ridge, while the rest of the organ has its stroma increased in thickness, and converted into a special tissue (epigonal organ). As a rule, the ovaries are paired, and lie some way forward. In many the left one is rudimentary (Mustelus, Galeus, Scyllium, Pristiurus, Carcharias). The long oviducts, which are developed from the Mullerian ducts, unite with their fused abdominal mouths to form a wide infundibular opening; this is correlated with the great size of the eggs which it has to take up. The hinder end of each ovi- duct is differentiated into a portion which is distinguished by its greater width, and often by its thicker walls ; this generally func- tions as a uterus, and opens into the cloaca. In the Selachii, as well as in the Chimasras, a glandular portion is differentiated close to the abdominal end of the oviduct. The generative organs of these two groups, and of the Dipnoi, agree in the most essential points. In these divisions the male organs are generally represented by small testes, the ducts of which are connected with the anterior portion of the excretory organs, so that this portion of the primitive kidney, with its efferent ducts, is adapted to the service of the generative apparatus. After several coils the vas deferens passes to the cloaca; in Chimasra it first unites with its fellow of the opposite side ; it generally opens with the ureter into a sinus urogenitalis, which opens by a papilliform process into the cloaca. Part of the Mullerian duct remains connected with the infundibular ostium, at the same point as that at which it is found in the female. At the binder end also a portion remains connected with the cloaca, in the males. The Mullerian duct in Chimasra is retained in the same way. In the males of the Selachii and Chima3ra3 certain parts of the posterior appendages are converted into copulatory organs (p. 487). 2 r 2 612 COMPARATIVE ANATOMY. The generative apparatus of the Amphibia is very similar to that of the Selachii. The ovaries (Fig. 348, A, ov) form lamelke, which project into the abdominal cavity, and vary in size according to the number of the eggs which are being developed. In the Urodela they enclose a cavity, which is broken up into several spaces in the Anura. The Mullerian duct forms the oviduct (od), which com- mences a long way anteriorly, by an infundibular orifice, and which always opens separately into the cloaca. It is generally greatly increased in size at the breeding season; this results in its being thrown into a number of coils. In the oviparous species (Salamandra) the terminal por- tion of the oviduct functions as a uterus. The testes are placed in the same region as the ovaries. They either form a compact organ, or consist of a series of larger or smaller (and consequently more numerous) bodies. The latter is the case in many Ccecilias, while in others there are intermediate stages towards the more compact form. A longitudinal collecting duct receives the efferent ducts of the different portions of the testis, and gives off again trans- verse canals, which correspond in number to the primary divi- sions of the kidney and are con- nected with them. The kidney, therefore, is the efferent duct for the sperm, which is passed out through the ureter (secondary arehiuephric duct). In the Anura, the sperm from the testis is passed to the kidney by a net- work of fine canals which are The canals, however, which pass into the kidneys from the longitudinal collecting duct, traverse the kidneys, without being connected with the Malpighian corpuscles, and open directly into the ureter. Bufo is the sole exception ; in it there is a connection between the vasa efferentia and the Malpighian corpuscles. In tho Urodela the anterior portion of the kidney (genital portion) is connected with the generative organs. Trans- verse canals (u c), are given off from a collecting duct, which is Fig. 318. Urogenital system of tho Amphibia (Triton). Diagrammatic. A Female. B Male, r Kidneys ; on the surface of which the nephrostomata are indicated. sug Ureter. od Oviduct. m Mullerian duct. ve Efferent duct of the testes, t Testes, ov Ovary, u Uro- genital orifice (partly after Spengel). placed between these two organs. GENERATIVE ORGANS OF VERTEBRATA. 613 placed in, or on the testes (B, t) ; these pass through the niesor- chium to a longitudinal canal, from which canals are again given off, and these pass into the so-called renal portion. The sperm, there- fore, passes through a certain portion only of the kidney, and only passes to the common ureter by the ducts which are given off from this portion ; this ureter is developed from the secondary archi- nephric duct. In proportion as this portion of the primitive kidney is freed of renal secretion it is converted to the uses of the generative apparatus, so that the two secretions are not commingled except in the ureter. In the males, the Mullerian duct remains free anteriorly, but it is generally closely connected with the secondary archi- nephric duct. It is either complete (m), and even has a ccelomic ostium, or parts only are canalicular, and the rest is converted into a solid chord at various points. This is most commonly the case in the Anura, but in Bufo it is very well developed. In the Ccecilia) the hinder portion has its walls provided with well- developed glands, in consequence of which this portion is still functional. In many Anura (Bufo) there is a peculiar large organ contain- ing ova-like cells on the testis ; this was formerly regarded as a rudimentary ovary. We do not know what function it has, any more than we know that of the so-called fatty bodies which are found attached to the anterior end of the germinal gland of the Anura. Since the generative organs open into the cloaca this organ func- tions as part of the generative system. In the female Urodela (Salamandra) the cloacal glands take up the sperm, and function as receptacula seminis. In the Ccecilias the cloaca of the male can be everted, and serves as a copulatory organ. Semper, C, Urogenitalsystem tier Selachier. — Spexgel, Urogeoitulaysteui tier Amphibieu, 1. c. § 453. The generative apparatus of the Sauropsida resembles that of the Amphibia in the more important points, and has, therefore, some of their arrangements more highly developed. The ovaries are racemose organs, which are placed in front of, or at the sides of, the vertebral column, and form large organs, which vary in size accord- ing to the extent to which the eggs, which are very large in this division, are developed. In the Ophidii the ovaries are placed at different levels. The right one is the larger, and is generally placed in front of the left one. In Birds the right ovary is atrophied. In the embryo it is as large as the left one, but while the left is developed it remains at a lower stage, and may at last disappear completely. Rudiments of it are found in the diurnal Raptores. The oviducts are again developed from the Mullerian ducts, and when fully developed are large, and ordinarily coiled canals which 014 COMPARATIVE ANATOMY. commence by a wide abdominal month. The mucous membrane which invests them is set in a number of longitudinal folds ; the lower portion, in addition to the greater thickness of its muscular wall, is distinguished from the other and longer portion by the larger size of its folds and villi; this is especially the case in Birds. This differentiation of the oviduct corresponds to the different function of the different parts ; the longer and more anterior portion secretes the albumen, and the thick- walled terminal part forms the shell. This portion is connected by a short and narrower piece with the cloaca. In correlation with the atrophy of the right ovary the oviduct of the same side is also atrophied in Birds ; not unfrequently, however, remnants of it are found near the cloaca. While the oviducts open by one orifice in the Ophidii and Saurii, as well as in Birds, in the Chelonii they open into the neck of the so-called urinary bladder ; this foreshadows the relation which is the typical one among the Mammalia. In many Ophidii a diverticulum of the posterior wall of the cloaca receives the openings of the oviducts. A remnant of the primitive kidney is retained behind the ovaries (this has been observed in the Saurii and Aves). The testes, which are gene- rally oval, are attached to the vertebral column by a fold of the mesenteries ; this is either effected in front of, or between the kidneys. Their size is closely correlated with their physio- logical activity ; this is especially the case in Birds. In the Ophidii they are arranged in the same way as the ovaries. The vasa efferentia pass to a parorchis, which generally consists of a few canals only, and from which a vas deferens extends to the cloaca. In the Crocodilini it is straight, in the Ophidii, Saurii, and Aves it is arranged in a number of smaller coils, while in the Chelonii (Fig. 349, e) it forms a complex of coils. In many Saurii arid Aves, as well as in the Crocodilini, its hinder portion is widened out. In the Saurii the vasa defe- rentia still unite with the ureters to open into the cloaca; in the Chelonii they open into a sinus uro- genitalis, which is formed by the neck of the urinary bladder. Fig. 319. Urinary and generative organs of a Chelonian (Chelydra serpentina). r Kidney, u Ureter, v Bladder, t Testes. e Secondary testes and vas deferens. ug Opening of the urogenital sinus into the cloaca, cl Cloaca, opened from be- hind, p Penis, s Groove of the penis, re Hind gut. c c' Cloacal csecal sacs. GENERATIVE ORGANS OF VERTEBRATA. 615 Sometimes each spermatic duct opens on a papillif orm process (Saurii, Aves). A rudiment of the Mullerian duct may sometimes be seen in the form of a filament passing forwards from the anterior end of the secondary kidney (Saurii), while further remnants of the anterior portion of the primitive kidney which are not converted into the secondary testis may be recognised. § 454. In the Mammalia the generative apparatus undergoes great metamorphoses, owing to the farther development of various portions of the efferent ducts and the formation of a number of accessory organs. In the female apparatus these are largely correlated with the relations that obtain between the embryo and the maternal organism. As this is least marked in the Monotreinata, they under- go the least amount of modification, and have therefore direct relations to the lower divisions of the Vertebrata, and especially to the Sauropsida. The oviducts (Fig. 350, t) open separately into a Fig. 350. Female generative organs of Ornithorhynchus. o End of the ovi- duct and ovary, t Oviduct, u Uterus. u' Point at which the orifice of the uterus projects upwards, close below the openiug of the ureter. vu Urinary bladder. sug Sinus urogenitalis. cl Cloaca. Fig. 351. Female generative organs of Halmaturus. ov Ovary, od Oviduct. u Uterus, cv Vaginal canals, cug Sinus urogenitalis. vu Urinary bladder. ur Ureter, * Opening of the bladder. sinus urogenitalis, which communicates with the cloaca (cl). The lower end of the oviduct, which is distinguished by the greater thickness of its muscular wall, forms a uterus (u) ; but this merely corresponds to the structures which likewise function as a uterus in many Anamnia and Sauropsida. In the Marsupialia the efferent ducts of the female are connected together on the outer surface, and each of them gives rise to an oviduct and uterus, as well as to a new portion, or vagina, which G16 COMPARATIVE ANATOMY. opens into the sinus urogenitals. The upper portion, which commences by a very wide ccelomic orifice, forms an oviduct (Fig 351, oc7), while the next and thicker- walled portion forms a uterus (u) . Each of the two uteri open by a papillif orm process into a portion, which from the exterior appears to be common to them both, and which is formed by the union of the two Mullerian ducts. A curved vagina is given off from this on either side (Didelphys), or the commencement of the tube is replaced by a caecal vaginal sac which is pushed out backwards, and is usually, though not always, divided internally by a median partition ; from this sac the distinct "vaginal canals" (cv) pass in a curved direction to the urogenital sinus (cue/) (Halmaturus). In the monodelphous Mammalia the archincphric ducts are united with the Mullerian ducts to form a common chord (genital chord). The connection between the two Mullerian ducts, which is well marked in Halmaturus, is effected in them at about the middle point of the duct, and thus they become connected during embryonic life. A portion of these ducts have their cavities fused, while they are separate in front of, and behind this point ; this is an indication of the common sac, which gives off the vaginal canals in the Marsupialia. But in the Monodelphia the lumina are fused as far as the end of the genital chord, and so form a single canal (genital canal) which opens into the sinus urogenitalis. There are, therefore, two canals, which are separated from one another at their commencement, but which unite into an unpaired portion of varying length ; these canals are derived from the Mullerian ducts, which are separate in the early stage of the embryo. The parts, which are distinguishable even in the Marsu- pialia, are due to the differences in the extent to which the walls of the different parts are differentiated, and the modifications in them are essentially due to the greater or less extent of the two tubes. The uterus undergoes a number of changes, most of which are due to adaptations to its relations to the foetus. Two completely separated uteri open into a vagina in many Ro- dentia (Lepus, Sciurus, Hydrochoerus, etc.), and in Orycteropus (Fig. 352, A). In other Rodentia the two uteri are only united for a short distance into a common open- ing into the vagina (e.g. Cavia, Coelogenys, Mus). This leads to the arrangements seen in the uterus of the Insectivora, Carnivora, Cetacea, and ITngulata, where a single uterus is continued into two separate cornua (B), which are continued into the oviducts. When the common portion of the uterus is elongated, the cornua are shortened ; this is the case in the Chiroptera and Prosimiae ; in the Simiaa, as in Man, there is a single uterus (0), which receives an oviduct on either side. Ji Fig. 352. Various forms of the uterus A 11 C u Uterus od Oviduct, v Vagina. GENERATIVE ORGANS OF VERTEBRATA. 617 The cornua of the uterus, and the common, uterus itself, vary very greatly in length ; so, too, does the vagina, the mucous membrane of which may be variously modified. In many Rodents (Lagostoinus) a certain portion retains its original double nature. Its opening into the urogenital sinus is sometimes distinguished by a temporary fold of mucous membrane, which is known as the hymen. This has been observed in the Ruminantia, Carnivora, etc. ; but it is in the Simire only that it has the same relations as in Man. The primitive Mullerian duct, which only served for the passage of the generative products, is therefore differentiated into three parts, owing to the great physiological changes that happen to it ; and of these parts the first, or Fallopian tube, alone retains its primitive relations. The ovaries, which are generally small, vary greatly according to the relation that obtains between the follicles and the stroma of the ovary. In a large number of Mammals they are racemose in form. They seldom retain their primitive position, and generally travel towards the pelvic basin, or, with their oviducts, are com- pletely enclosed in it. They are always in close relation to the oviduct, or rather to its infundibular ccelomic mouth, for a process of the margin of the ostium extends to the ovary. The mesenteric folds (ligamenta uteri lata), which support the ovaries and oviducts, not unfrequently unite with the pouch that encloses the ovary to form the mouth of the oviduct (as in the Carnivora). Remnants of the primitive kidneys and their ducts, which are enclosed with them in the genital chord, are retained at the sides of the uterus, or in the folds of the peritoneum, which connect the ovaries with the uterus. The so-called cauals of Gartner are formed by remnants of the archinephric ducts, which accompany the uteri in Echidna, and open into the urogenital siuus; in other forms, portions only of these canals persist. A rudiment of the primitive kidney, which is placed near the ovaries, is known as the parovarium. § 455. In the male generative apparatus of the Mammalia the testes have, at first, the same position as the ovaries — that is, they are placed at the inner edge of the primitive kidneys. A chord extends from the archinephric duct to the inguinal region of the abdominal wall (n). The primitive kidneys are partly united with the testes, and there form the parorchids (epididymes). As in the female, the archinephric duct unites with the Mullerian duct to form a genital chord, which passes to the urogenital sinus, developed from the lowest portion of the allantois. It forms the vas deferens ; the Mullerian duct is atrophied, its terminal portion only being, as a rule, converted into a permanent organ, corresponding to a sinus genitalis, the so-called uterus masculinus ; this generally opens into the urogenital canal between the orifices of the seminal ducts. G18 COMPARATIVE ANATOMY. The apparatus thus formed is variously modified in different parts. The testes do not retain their primitive position anteriorly to the kidneys in any Mammals except the Monotremata. In the Cetacea, Hyrax, Elephas, and various Edentata, they are placed a little to the side of, or below the kidneys. In others they are found in the inguinal region of the ab- dominal wall, which they pass through (many Rodents, the Camelida3, and various Car- nivora [Lutra, Viverra]). In others, finally, they travel still further by means of the ingui- nal canal, descending through the wall of the abdomen iuto a diverticulum, the scrotum, which is formed from the in- tegument. The space which is formed (canalis vaginalis) when the testis passes into the scro- tum, by the peritoneum which grows out with the descending testis, is permanently open in most Mammals, so that the cavity around the testis is in communication with the ab- dominal cavity. As the testes pass down the inguinal canal the abdominal wall is driven in front of them. When the vaginal canal remains open the testes may return again to the abdominal cavity; this ordinarily happens in many Mammals during the breeding season (e.g. Marsupialia, Ro- dentia, Chiroptera, Insectivora, etc.). The scrotum of the Marsupialia is remarkable for its position in front of the genital orifice. It is a special structure, while in the Mono- delphia the scrotum is de- veloped from the boundary of the primitive urogenital orifice. The lower end of the vas deferens is always simple in the Monotremata and Marsupialia, Carnivora and Cetacea. In the rest it gives rise to glandular structures, which are known as "vesiculas seminales," as the sperm may be collected in them (yl). These organs are greatly develoj^ed in the Insectivora and many Rodents; Fig. 353. I Urinary and generative organs of Cricetus vulgaris. B Kidneys. u Ureter, v Urinary bladder. T Testes. Sp Vasa spermatica. il Vas deferens, gl Vesicular seminales. gV gl" Prostatic glands. m Muscular portion of the urogenital sinus. ic Corpus cavernosum penis, be Corp. cav. urethras. c Cowper's glands. t Tyson's glands, p Prepuce, g Glans penis. II Neck of the bladder, and commencement of the urogenital sinus, opened in front. * Opening of the ductus ejaculatorii. Ill Glans penis seen from in front. GENERATIVE ORGANS OF VERTEBEATA. 619 in the former tliey are often broken up into several largo lobes, while in the latter they are distinguished by their length and by the diverticula which are found on them. The terminal portion also of the vas deferens often has a glandular structure. Besides the seminal ducts, the short terminal portion of which receives the vesiculas seininales, aud is known as the ductus ejacu- latorius, rudiments of the Mullerian ducts open into the urogenital sinus in many Mammals. They either consist of a single or of a paired diverticulum, which corresponds to a rudimentary sinus genitalis of the female, or, rather, to its vaginal portion, so that it is not very exact to call it a uterus masculinus. Part of it some- times forms a portion of the male genital sinus, for the seminal ducts open into it. These organs are largest in the Eodentia (Fig. 354, g), although, indeed, they are not altogether wanting in other forms ; in Man they are represented by the prostatic vesicle. Lastly, the urogenital canal is provided with yet another set of glandular organs, the prostatic glands. These may be of a considerable size, and form paired lobate structures (Rodentia, Elephas, Insectivora) (Fig. 353, gl' gl"), or they are formed of a number of smaller tubes, which are connected by layers of smooth muscular fibres to a mass which is attached to the wall of the urogenital canal. By the further develop- ment of the musculature, which is found on these glands in other forms also, the prostate is converted into a circular body. B § 456. In the lower divisions the efferent ducts of the urinary and generative apparatus unite with the terminal portion of the enteric canal to open into the cavity which has already (p. 562) been called the "cloaca;" but it is doubtful whether this should be regarded as the primitive condition, for we might take the arrangement which obtains in the Cyclostomata, Ganoi'dei, and Teleostei, to be such, where the urogenital organs and the tractus intestinalis open separately. In them, the anus is in front of the urogenital orifices, although, and especially in the Ganoi'dei, there is a depres- sion into which both these orifices open ; this depression is an early indication of a cloaca. The cloaca is well developed in the Selachii, and the orifices of the urogenital apparatus, which lie, in other forms, behind the anus, are there placed on the dorsal wall of the cloaca. Fig. 354. Urogenital canal and urinary bladder of Lepus Cunicnlus. J. From behind. B The posterior wall of the uterus masculinus is laid open. 0 Side view. v Bladder. u Ureter, d Seminal duct, g Sinus geni- talis, ug Urogenital canal. 620 COMPARATIVE ANATOMY. This relation is henceforward the common one ; in the Amphibia, Reptilia, and Aves, there is a cloaca of pretty much the same kind ; in Birds it is provided with a diverticulum, the bursa Fabricii (Fig. 333, h), which is attached to its hinder wall. The cloaca must be regarded as being inherited by all the Mammalia, although it is in the Monotremata only that it persists without any great modifi- cation ; in the rest it undergoes considerable changes. The most important of these is the share which it takes in the differentiation of a copulatory organ, and which was faintly indicated in the Am- phibia ; these changes end by giving rise to a urogenital orifice distinct from the anus. The allantois is one of the most im- portant of the organs which are differentiated from the cloaca ; it is developed from the anterior wall of the cloaca, that is, from the part of the primitive cavity of the hind-gut that represents it. In Lepidosiren and in the Amphibia this organ forms a body which springs from the anterior wall of the cloaca by a short stalk ; in the latter it is continued into two anteriorly placed diverticula ; it lies freely in the ccelom. It is known as the urinary bladder, and seems indeed to function as such, although the ureters open some way from it. Blood-vessels are distributed on its thin wall ; the arteries come from the pelvic vessels, and the veins pass to the portal vein. In the Amniota this organ is very greatly developed during the embryonic stages, and becomes a large sac which grows out far beyond the embryo, and is provided with a large number of vessels ; it envelops the embryo, already covered by the amnion. In Keptiles and Birds it gradually atrophies as the abdominal wall is closed in, and disappears altogether, In the Saurii and Chelonii only the portion of the allantois within the abdominal cavity is retained ; in them it is widened out into a sac, which is provided with diverticula on either side (Fig. 349, z>). In the Mammalia this organ has different relations to the develop- ing organism. As in the Reptilia and Aves it grows out into a vesicle, which communicates with the cavity of the hind-gut by a stalk which runs inside the umbilical chord. That portion of the chord, which passes into the ccelom (urachus) is partly converted into a ligament (Lig.vesico-umbilicale medium), partly into the urinary bladder, and partly into a sinus urogenitalis, where the orifices of the generative ducts pass into it. In the Monotremata and Marsupialia the peripheral portion appears to have the same relations as in the Sauropsida., while in other Mammals it aids in the formation of the "chorion," which is connected by villous elevations with the mucous membrane of the uterus. When these vascular villi of the chorion are further developed, the foetal blood passing along the vessels of the allantois, acquires a distribution in the peripheral regions of the sac. This effects exchanges with the blood which is distributed in the mucous membrane of the uterus. As it becomes more intimately connected with the uterine mucous membrane a placenta is developed ; this varies greatly in character according to the way GENERATIVE ORGANS OF VERTEBRATA. 621 in which, and the extent to which, the chorion is connected with the raucous membrane of the uterus, and according to the modifications undergone by the latter organ. 457. The copulatory organs form another series of parts formed by the differentiation of the wall of the cloaca. In the Selachii, indeed, organs which did not belong to the generative apparatus — • parts of the hinder appendages — are used as organs of copulation and modified accordingly, but new organs begin to be differentiated, which in the Amphibia are faintly indicated by the presence of a papilla which projects into the cloaca. These belong to one of two typical forms ; in one the organs are connected with the posterior, and in the other with the anterior wall of the cloaca. One of them is dominant in the Saurii and Ophidii. The copu- latory organs first appear as external appendages, placed just behind the cloaca ; later on these are invaginated in a tubular fashion (Fig. 355, p), and are only protruded during copulation. When protruded, each of these organs is continued into two more or less blunt ends, which vary in form. At the sides there is a groove, which is continued on from the cloaca, and which has a spiral course posteriorly, and is then directed towards the middle line; this serves to convey the sperm. The largest of the muscles supplied to it are the retractors, which are inserted into the blind end of the tubes. Glands open near the root of the tubes ( Penial tubes, ono of which is laid open longitudinally. 622 COMPARATIVE ANATOMY. placed below tlio two fibrous ones) it forms an erectile welt, wliich represents a penis. Special muscles, which, are inserted into the fibrous bodies, act as retractors of the penis ; in Struthio this organ is provided with special elevator muscles, and is hidden in a diverticulum of the cloaca. The copulatory organs of the Mammalia also belong to the second type ; those of the Monotremata differ markedly from the organs in other Mammals. Their copulatory organs consist of a short penis, which is formed of two erectile bodies, and which lies in a pouch which opens into the cloaca. By means of a muscle this can be approximated to the urogenital canals, and so takes up the sperm through an orifice which is placed at its root, near the opening of the urogenital sinus. Owing to the special mode by which a portion of the wall of the cloaca is differentiated, this organ comes to be exclusively related to the generative apparatus, while the urine passes out through the cloaca. When the cloacal aperture is differentiated into two orifices, the copulatory organ becomes more closely related to the urogenital sinus. During the embryonic stage a fold begins to be raised up around the cloacal orifice, and a process is developed on the anterior wall of the cloaca, which carries on its posterior surface a groove which leads to the opening of the urogenital canal. As the embryo continues to grow, the cloaca becomes shallower, and the wall of partition between the orifice of the hind-gut and the urogenital canal, which is formed from the lower end of the urachus, becomes more distinct. At last the orifices which were formerly placed on the floor of the cloaca come to the surface. The anterior fissure at the base of the genital protuberance forms the opening of the uro- genital sinus, while the hinder orifice forms the anus. In many Mammalia the two orifices are always close to one another, and may even be enclosed by the same fold of integument; in the female sex the two orifices are ordinarily close together. This is most markedly the case in the Marsupialia (where there is even a common sphincter for the anus and urogenital orifice) and Rodentia ; in which forms, indeed, it obtains in the males also. § 458. The urogenital sinus is developed to a different extent in the two sexes, and this is due to the difference in their functions. In the male the urogenital sinus and genital protuberance grow out into a narrower, but ordinarily long canal (the so-called urethra), with the walls of which erectile organs are connected. They form the penis. In the female there are parts which are similar to, though less largely developed than, this organ and its erectile bodies ; they form the clitoris, an organ which corresponds to the penis. The erectile organs of the Marsupialia arc formed of two bodies GENERATIVE ORGANS OF VERTEBRATA. G23 Fig. 356. Divided penis of Didelpkys philander. a b Halves of the glans. s Groove on its inner surface. x Region of the anus which is placed behind the orifice of the prepuco (after Otto). which are derived from tlie genital protuberance and surround the urogenital canal ; in some they are divided at their free end (Fig. 356, a b) and form the glans penis. The urogenital canal is continued on to each half in the form of a groove (s), and these grooves may unite together to form a canal. In others (Halmaturus) these erectile bodies are connected with two others, with which they unite to form a cylindrical penis, and bound the urogenital canal. The first- mentioned erectile bodies generally fuse very early in other Mammalia to form a corpus caver- nosum urethras which surrounds the urogenital canal (urethra), and of which the most anterior end, which varies greatly in form, forms the glans penis. The two other erectile bodies (corpora cavernosa penis), which in the Marsupialia are not firmly connected with the pelvis, are con- nected with the ischium; they pass above the corpus cavernosum urethras, but do not extend into the wall of the urogenital canal. In most Mammals the penis thus formed extends for- wards from the symphysis pubis along the median line of the abdomen, and ends at a varying distance from the umbilicus ; in others (Chiroptera, Primates), it is free and hangs down from the symphysis pubis. In either case, the integument covers it and forms a fold in front of, and around the glans — the prepuce. In the female, the genital protuberance is never developed to the same extent as in the male; it forms the clitoris, which carries on its lower surface the opening of the urogenital sinus, which is bounded by lateral folds. The clitoris is generally more largely developed in the embryo than in the adult, as it projects from the pubic fissure and is afterwards withdrawn into it. In some Apes, however (Ateles), the clitoris continues to be developed and become an organ of some size. Two erectile bodies (corpora cavernosa urethras) lie in the walls of the urogenital sinus and surround it as far as the clitoris, at the base of which there is another pair of erectile bodies. The end of the clitoris is generally provided with a gland, and is also covered by a prepuce. Sometimes this organ is provided with special muscles, which are mostly differentiated, as are also those of the erectile bodies, from a common occludor of the cloaca, such as is seen in the Marsupialia. Iu addition to these, many Mammals have muscles which raise, or retract the penis. Glandular organs open into the urogenital sinus of both sexes. There are others besides the prostatic glands already men- tioned (p. 619) ; there may be one or more, or as many as four pairs (Marsupialia); they lie at the root of the penis (Fig. 353, c). 624 COMPARATIVE ANATOMY. Cowper's glands are connected with the portion which is enclosed by the erectile bodies. These are not always present (Cetacea, Carnivora). In the female they open into the vestibulum vaginas (glands of Duverney or Bartholin). The glands of the prepuce (Tyson's glands) are developed into large organs in many Mam- mals, and especially in the Rodentia, among which they are best developed in Castor (Fig. 353, t). INDEX. AcalephjB — Tentacles, 101 ; Ectoderm, 103; Urti eating capsules, 163; Skele- ton, 106 ; Nervous system, 108 ; Gastric filaments, 118; Sexual or- gans, 119. Acanthias — Fin, 477 ; Kidneys, 603. Acanthocephali — Integument, 136 ; Mus- cular system, 143 ; Muscular fibres, 144 ; Nervous system, 147 ; Enteron, 159 ; Lemnisci, 174 ; Generative organs, 176. Acanthometridre — Skeleton, 82. Acanthopteri — Urostyle, 431. Acarina — Metameres, 237 ; Cerebral ganglia, 256; Ca3ca, 269; Hind-gut, 270 ; Malpighian vessels, 276 ; Germ- glands, 298. Acera — Ganglia, 348. Acervulina? — Supporting organs, 81. A cheta— Testis, 304. Achetida — Auditory organ, 262. Acineta — Figure, 88. Acipenser — Dermal bones, 425; Eibs, 439; Cartilaginous cranium, 450 ; Dermal denticles, 450 ; Thoracic fin, 479 ; Spiracular cleft, 543 ; Pseudobran. chia, 543 ; Air-bladder, 547. Acopa— Form of body, 390 ; Gemmation in, 391 ; Ganglia, 397 ; Branchial sac, 399; Branchial slits, 402; Sexual organs, 407. Acrania — Auditory organs, 533 ; Respi- ratory cavity, 541. Acridida — Auditory organ, 262. Acrodont Lizards — Teeth, 557. Acrocladia — Spines, 206. Actason — Excretory organ, 377. Actiuospha?rium — Figure, 84. Adder — Epigastric veins, 595. ^ginidaj— Tentacles, 102, 107; Mar- ginal vesicle, 110 ; Gastrovascular system, 115 ; Generative organs, 122. iEginopsis — Tentacles, 102. - 642 INDEX. trunks, 149; Ganglia, 150; Vascular system, 168; Excretory organs, 176. Scolopendra — Salivary glands, 274; Liver, 274; Malpighian vessels, 276; Cir- culatory system, 281; Generative or- gans, 299." Scomber — Trunk muscles, 494; Appen- dices pyloric??, 560. Scorpionea — Metameres, 237; Chelicerae, 244; Integument, 24S; Poison glands, 250; Nervous system, 256 ; Eve, 265; Enteric canal, 269; Hind-gut, 270; Salivary glands, 274; Hepatic tubes, 275 ; Malpighian vessels, 276 ; Cir- culatory system, 28 1 ; Lungs, 291 ; Generative organs, 297. Scufcigera — Trachea), 288. Scylkea — Branchiae, 338 ; Stomach, 362 ; Vessels, 372 ; Excretory organs, 377. Scyllium — Brain, 501; Olfactory organs, 525; Branchial cavity, 543 ; Ovaries, 611. Scymnus — Arachnoid, 513. Scyphostoma — Gastrovascular system, 116. Segestria — Tracheae, 291 ; Ovaries, 297. Selachii — Cartilage cells, 26 ; Anterior appendages, 414 ; Dermal denticles, 423; Vertebral column, 429; Trans- verse processes, 431 ; Fin-rays, 432 ; Eibs, 439; Cranium, 447; Palato- quadrate, 448 ; Hyoid, 448 ; Mandi- bular apparatus, 153 ; Branchial skeleton, 468 ; Hyoid arch, 469 ; Copulse of hyoid, 469 ; Archip- terygium, 473; Anterior appendages, 474 ; Shoulder-girdle, 474 ; Fin, 477 ; Pelvic-girdle, 484 ; Ventral fin, 487 ; Dermal muscles, 494 ; Muscles of branchial skeleton, 497 ; Olfac- tory lobes, 504; Thalamencephalon, 501; Mesencephalon, 505; Sinus rhomboidalis, 505 ; Medulla oblon- gata, 505 ; Optic uerve, 515 ; Ce- phalic nerves, 516 ; Vagus, 517 ; Facial nerve, 517; Glossophar- yngeal, 518 ; Gelatinous tubes, 52 1 ; Olfactory organs, 525 ; Sclerotic, 529; Ciliary processes, 530 ; Tapetum, 530 ; Eyelids, 532 ; Vestibule, 53 ! ; Otoliths, 536; Spiracle, 537; Bran- chial pouches, 542; Spiracular cleft, 512; Branchial lamella"-, 544; Teeth, 550; Enteron, 555; Stomach, 557; Spiral valve, 560; Hind-gut, 562; Air-bladder, 567 ; Abdominal pore, 574; Conus arteriosus, 578; Bran- chial arteries, 579 ; Aorta, 585 ; Spleen, 600; Thymus, 6C0; Archinc- phric duct, 602 ; Kidney, 603 ; Nephrostoinata, 603 ; Mullerian duct, 603; Eggs, 609; Yolk, 6C9; Sperm, 609; Ovaries, 610, 611; Germinal glauds, 611; Epigonal Organ, 611 Copulatory organs, 611; Cloaca, 610. Sepia — Mantle, 325 ; Copulatory organs, 327 ; Kespiratory organs, 340 ; In- ternal skeleton, 341 ; Buccal ganglia, 351 ; Eye, 355 ; Auditory plate, 357 ; Caeca, 363; Liver, 366; Ponchos of Needham, 357. Scpiada?— Fins, 325 ; Arms, 326 ; Os sepia?, 334 ; Arteries, 374. Sepiola — -Liver, 366. Sergestes — Auditory vesicles, 261. Serpula— Tentacles, 133 ; Stalk of oper- culum, 134; Ganglionic chain, 149. Serranus — Hermaphrodite arrangements, 611. Sertularia— Colonies, 93 ; Tests, 161. Shad— Beard, 521. Shark — Skin, 423 ; Dermal denticles, 423 ; Caudal region, 431. Sialida — Salivary glands, 274. Siredon— Vertebra?, 432 ; Auditory ossi- cles, 538. Siphoniata — Siphons, 320. Siphonophora — Division of labour in, 95 ; Colonies, 95 ; Nectocalyces, 96 ; Figure, 96 ; Nutritive persons, 95 ; Protective persons, 97 ; Tentacular pprsons, 97; Generative persons, 97; Cilia, 104 : Muscular system, 108 ; Gastric system, 114; Pigmented in- vestment of stomach, 118 ; Separa- tion of sexes, 120 ; Generative products, 121. Silurus — Lymph sinuses, 599. Simia? — Liver, 564; Uterus, 616; Hymen 616, Singing Birds — Syrinx, 572. Siphonostoma — Tentacles, 133, 135 ; Degeneration in, 236 ; Ovaries, 293. Sipuuculns — Cerebral mass, 148; Gan- glion cells, 148 ; Alimentary canal, 161, 162 ; Vascular system, 171 ; Excretory organs, 176. Siren — Blood-corpuscles, 576. Sirenia — Nipples, 422 ; Caudal vertebra1, 436. Smyuthurus — Trachea?, 288. Solastcr — Arms, 196; Alimentary caual, 213. Solen — Ganglia, 345. Solenogastres — Ventral surface, 130; Aciculi, 139; Nervous system, 151; Groove, 318. Solidungula — Fore-limb, 483 ; Vena? cava?, 592. Sorex— Jugal, 466; Coracoid, 476. Spatangida? — Scmita?, 201; Dermal skele- ton, 205 ; Spines, 206 ; Muscular system, 207 ; Blood-vessels, 218 ; Water-vascular ring, 220 ; Stone, canal, 222; Generative organs, 226. INDEX. 643 SphBerodornm — Integument, 138. Sphinx — Digestive organs, 270. Sphenodon — Bibs, 440 ; Quadrate, 461. Spiders — Chelicera?, 214; Spines or seta?, 250 ; Spinning glands, 250 ; Nervous system, 256: Cerebral ganglia, 256; Eye, 265 ; Digestive organs, 269. Spio — Dermal glands, 141. Spirochona — Carapace, 83. Spirorbis — Stalk of operculum, 134. Spirostomum — Contractile bands, 80 ; Nucleolus, 87. Spirula— Shell, 331. Spondylus— Shell, 336. Spongia? — Gastrula, 92 ; Tentacles, 101 ; Integument, 103 ; Alimentary canal, 111 ; Gastric system, 112 ; Lipo- gastria, 113; Sexual organs, 119; Ova, 124. Spongilla — Amphidiscs, 106. Squat ina — Protopterygium, 478; Arach- noid, 513 ; Enteric canal, 560 ; Heart, 578. Squilla — Branchiae, 241 ; Nervous system, 253. Stauridium — Tentacles, 93. Stent or — Contractile bands, 30; Shells, 83 ; Anal opening, 85. Stephanoceros — Wheel organ, 138. Sternaspis — Bespiration in, 136 ; Cere- bral mass, 1 18 ; Ganglion cells, 148 ; Vascular system, 171 ; Excretory organs, 176. Stomapoda — Liver, 275; Heart, 281. Strcpsiptera — Wings, 218; Male organs, 304. Strombus — Siphon, 323. Struthiones — Sacrum, 135; Skull, 160; Toes, 490; Pecten, 530; Marsupium, 531 ; Turbinate bones, 547 ; Lym- phatic hearts, 599; Copulatory or- gans, 621. Sturiones — Dermal bones, 425; Vertebral column, 430; Bibs, 439; Cartilagi- nous cranium, 450; Mandibular apparatus, 453; Operculum, 455; Goblet-shaped organs, 523 ; Ciliary processes, 530; Branchial lamellae, 544 ; Thymus, 600. Suina — Nipples, 422 ; Manus, 483. Suctoria — Pseudopodia, 83. Swans — Trachea, 572. Sycaltis — Skeleton, 105. Sycones — Ectoderm, 103 ; Gastric sys- tem, 113. Syllida? — Parapodia, 134; Visual organs, 155. Synaptic — Ambulacra, 198; Tentacles, 200 ; Calcareous anchor, 206 ; Inte- gument, 206; Internal skeleton, 207; Muscular system, 208 ; Alimentary canal, 214 ; Figure of, 223 ; Polian vesicles, 223 ; Water-vessels, 223 ; Hermaphrodite organs, 226. Syncoryne — Tentacles, 93 ; Figure of, 93. Syncorynida3— Cormi, 94. Taenia — Segmentation, 129 ; Cystic form, 131; Aciculi, 140; Female organs, 182. Tamoya — Marginal filaments, 102. Tauais — Auditory vesicles, 261. Tapir — Manus, 482. Tardigracfca — Eye, 266; Generative organs, 298. Tal pa— Manus, 482. Teleosaurii — Dermal bones, 425. Tegeueria — Trachea?, 291. Teleostei — Bony plates, 424 ; Vertebral column, 430 ; Transverse processes, 431 ; Caudal region, 431 ; Fin rays, 432 ; Articular processes, 437 ; Bibs, 439 ; Cranium (cartilaginous) 450 ; Sphenoids, 452 ; Mandibular appa- ratus, 453 ; Operculum, 455 ; Bran- chial arches, 469 ; Shoulder-girdle, 474 ; Thoracic fin, 478 ; Pelvic- girdle, 484 ; Ventral fin, 487 ; Dermal muscles, 492 ; Muscles of branchial skeleton, 497 ; Olfactory lobes, 504 ; Thalarnenccphalon, 504 ; Mesen- cephalon, 505; Sinus rhomboidalis, 505 ; Optic nerves, 515 ; Facial nerves, 517; Glossopharyngeal, 518; Vagus, 521 ; Goblet-shaped organs, 523 ; Mucous canals, 524 ; Lateral line, 521; Olfactory organs, 525; Sclerotic, 529 ; Optic muscles, 531 ; Eyelids, |532 ; Ductus endolympha- ticus, 534 ; Asteriscus, 536 ; Bran- chial clefts, 543 ; Spiracular cleft, 543 ; Branchial lamella?, 544 ; Teeth, 550 ; Euteron, 555 ; Fore- gut, 556 ; Stomach, 557 ; Mid- gut, 559 ; Air-bladder, 567 ; Bulbus arteriosus, 578 ; Branchial arteries, 579 ; Aorta, 585 ; Thymus, 600 ; Kidney, 603, 604; Yolk, 609; Ovaries, 610 ; Cloaca, 619. Tenthredineae — Feet, 246; Cement glands, 303. Terebella — Tentacles, 133; Branchial do., 135 ; Vascular system, 169. Terebratula — Circlet of cilia, 307 ; Mus- cular system, 307 ; Nervous system, 310 ; Excretory organs, 313. Termes — Nervous system, 257. Termites — Malpighian vessels, 277. Testicardines— Stalk, 308; Shell, 308; Skeleton of arms, 308 ; Alimentary canal, 311 ; Generative organs, 313. Tetractinia — Tentacles, 99. Tetrarhynchus — Aciculi, 140. Thalassema — Excretory organs, 176. Thaliada?— Stolo prolifer, 391. Thauuiantias — Gastrovascular system, 115. 644 INDEX. Thecidium— Metameres, 307; Generative organs, 313. Thecomedusa? — Organisation, 98 ; Scy- pliostoma form, 98 ; Strobila, 99 ; Ephyra form, 99 ; Discophora of, 99. Thecosomata — Velum, 318, 321 ; Head, 321. Thelyphonus — Nervous system, 256. Thomisus — Trachea, 291. Thoracostraca — Optic nerves, 253 ; Ven- tral chord, 251; Heart, 281; Female organs, 291. Thylacinus — Marsupial bones, 407. Thysanopoda — Branchiae, 212. Thysanozoon — Tentacles, 133; Alimentary canal, 158. Thysannra— Gnathites, 245; Feet, 246; Ventral chord, 258 ; Eye, 265; Enteric canal, 272 ; Trachea?, 288. Tinea — Goblet-shaped organs, 523. Tintinnns— Shells, 83. Tipulida?— Eye, 267. Toads — Sperm, 669. Tomopteris — Generative organs, 189. Torpedo — Electric organs, 500, 505. Tortrix— Hyoid, 472. Toucan — Tongue, 552. Tracheata — Development, 234 ; Appen- dages, 238, 243 ; Antenna), 244 ; Ner- vous system, 255 ; Salivary glands, 273 ; Liver, 275 ; Malpighiau vessels, 276, 286 ; Fat body, 278 ; Renal con- cretions, 278 ; Circulatory system, 282 ; Tracheae, 286. Trachelius — Contractile vesicles, 86. Trachynema— Tentacles, 102, 107 ; Mar. ginal vesicles, 110. Tragulida — Stomach, 559. Trematoda — Sporocvst, 131 ; Cercaria?, 131 ; Cilia, 137 ; Aciculi, 139; Stylets, 141 ; Dermal glands, 141 ; Suckers, 143 ; Nervous system, 146 ; Visceral nerves, 151 ; Visual organs, 153 ; Alimentary canal, 157 ; Glandular organ of enteron, 164 ; Liver, 165 ; Excretory organs, 173 ; Generative organs, 179 ; Pore, 183. Tremoctopus — Hcctocotyliscd arms, 327. Tricocephalus — Muscular system, 113. Trionyx— Plastron, 426. Tristoma — Excretory organs, 173. Triton— Lungs, 573 ; Heart, 581. Tubicola;— Tube, 134 ; Seta?, 140 ; Visual organs, 155 ; Hind-gut, 163 ; Vascular system, 169. Tubifex — Generative organs, 189 ; Sper- matophores, 191. Tubipora — Skeleton, 106. Tubularia — Buds, 95 ; Tentacles, 101 ; Tests, 104; Supporting lamella, 107; Pigmented investment of stomach, 108. Tunicata — General review, 388 ; Biblio- graphy, 389 ; Classification, 389 ; Form of body, 390 ; Integument, 393 ; Mantle, 393 ; Skeleton, 39 1 ; Muscular system, 391 ; Nervous sys- tem, 395 ; Sensory organs, 397 ; Ali- mentary canal, 398 ; Ventral groove, 402 ; Sexual organs, 406. Turbellaria— Mouth, 129 ; Tentacles, 132 ; Epidermis, 137 ; Rod-liko bodies, 140 ; Muscular system, 112 ; Nervous system, 146 ; Visceral nerves, 151 ; Tactile seta?, 152 ; Visual organs, 153 ; Auditory organs, 156 ; Alimentary canal, 157 ; Glandular organs of en. teron, 161; Excretory organs, 173; Generative organs, 179 ; Pore, 184. Tylopoda— Fore-limb, 483 ; Stomach, 559 ; Blood-corpuscles, 576. Ungulata — Nipples, 422 ; Vertebral column, 435 ; Pai'amastoids, 463 ; Clavicle, 477 ; Fibula, 494 ; Hallux, 491 ; Cerebellum, 510 ; Uterus, 616. Unio — Muscles, 342 ; Nervous system, 345 ; Efferent renal ducts, 376, Geni- tal canal, 381. Urodela— Vertebra?, 432, 433 ; Shoulder, girdle, 475 ; Pelvic-girdle, 484 ; Hind- limb, 488 ; Brain, 505 ; Ductus en- dolymphaticus, 534; Auditory os- sicles, 538 ; Stomach, 537 ; Iliac vein, 593 ; Lymphatic hearts, 599 ; Kidneys, 604 ; Ureter, 601 ; Cloaca, 613. Uromastix — Sternum, 412. Ursina — Ca?cum, 562. Ursus — Kenal organs, 605. Vaginicola — Shells, 83. Varanus — Columella,458 ; Brachial plexus, 514; Tongue, 552. Vegetable Kingdom — Cells, 15 ; Tissues, 21. Velella — Generative persons, 97 ; Air-sac, 98 ; Hepatic canals, 118 ; Generative products, 121. Vermes — Gastrula,35 ; Dermal branchiae 36 ; Excretory organs, 46 ; Alimen- tary canal, 47, 48 ; Inspiratory or- gans of the enteron, 49 ; Vascular system, 50 ; Gemmation in, 61 ; Ho- mology in, 64 ; Classification, 125 ; Bibliography, 127 ; Form of body 128 ; Segmentation, 129 ; Muscular system, 142 ; Visceral organs, 153 ; Auditory organs, 156 ; Alimentary canal, 156 ; Ccelom, 165 ; Generative products, 190. Vertebrata — Branchial clefts, 7; Lungs, 10; Connective tissue, 24; Osseous tissue, 28; Blood-corpuscles, 29; Cartilaginous skeleton, 38; Muscu- lature, 40 ; Excretory organs, 47 ; Respiratory organs of the enteron, 48 ; Homology in, 64J; General INDEX. G45 review, 403; Classification, 409; Bib- liography, 412 ; Form of body, 413 ; Head, 113 ; Appendages, 414 ; Limbs, -115; Integument, 417; Ectoderm, 417 ; Dermal papillae, 417 ; Dermal bones, 425 ; Internal skeleton, 426 ; Ribs, 43S ; Thoracic-girdle, 473 ; Pelvic-girdle, 473 ; Archipteryginm, 473 ; Fore-limb, 479 ; Muscular sys- tem, 491; Nervous system, 501; Brain, 503 ; Pineal gland, 503 ; In- fundibulum, 501; Pia rnater, 513; Arachnoid, 513 ; Spinal nerves, 51 4 ; Vagus, 516; Olfactory organs, 525; Eye, 527; Iris, 530; Fenestra ovalis, 537; F. rotunda, 537; Tympanic cavity, 537 ; Auditory ossicles, 538 ; Enteric canal, 539; Gall bladder, 561; Pancreas, 565; Coelom, 574; Vascular system, 575 ; Systemic arteries, 585; Spleen, 600; Excre- tory organs, 601; Generative organs, 606 ; Germinal epithelium, 606 ; Male germinal glands, 609 ; Sperm, 6)9; Genital ridges, 610. Virgularia — Generative organs, 123. Viverra — Testes, 618. Voluta — Proboscis, 361. Volvocincaj — Cells, 19. Vortex — Generative organs, ISO. Vorticellinao — Stalk, 83 ; Anal opening, 83 ; Nucleus, 89. Vultur — Cervical vertebra?, 431. Waldheimia — Nervous system, 310 ; Vas- cular system, 382. Wasps — Mid-gut, 272 ; Parthenogenesis, 302. Whales — -Whalebone, 549. Woodpeckers — Tongue, 552; Crecum, 562. Zoothamnium — Contractile bands, 80. CHARLES DICKENS AND EVANS, CBTSTAL PALACE PRESS. mmm £ dm). ' '