ERRATA. Page 122, line 14, for Entomasome read Entomosome 125, line 28, /or lamina read laminae ... line 29, delete " the " before the word superadded ... 127, line 16, for a principle read the principle ... 129, line 22, delete " at " before the word present ... 164, line 6 from bottom, /or fossa, read fossas, . . . 165, line 18 from bottom, for their read these ... 166, line 11 from bottom, read Edentata, among the Mammalia. The great . . . line 4 from bottom, read extruded from the walls 178, lines 22 and 21 from bottom, for neuractinopophyses read neu- ractinapophyses ... 181, lines 15 and 17, for element read segment DETAILED ABSTRACTS OF PAPERS ON 1. — THE MORPHOLOGICAL RELATIONS OF THE NERVOUS SYSTEMS IN THE ANNULOSE AND VERTEBRATE TYPES OF ORGANIZA- TION. 2. — THE MORPHOLOGICAL CONSTITUTION OF THE SKELETON OF THE VERTEBRATE HEAD. 3. — THE MORPHOLOGICAL CONSTITUTION OF LIMBS. COMMUNICATED TO THE CHELTENHAM MEETING OF THE BRITISH ASSOCLVTION. AUGUST 1856. BY PROFESSOR GOODSIR. From the Edinburgh New Philosophical Journal, Netu Seriet, vol. v., Jan. 1857. 118 Proceedings of Societies. PROCEEDINGS OF SOCIETIES. British Association for the Advancement of Science, Cheltenham, August 5-12, 1856. The following detailed abstracts of tlie papers submitted by Professor Goodsir to Section D. at the Cheltenham Meeting of the British Associa- tion in August last, were prepared by the author for our last number : — 1. On the Morphological relations of the nervous systems in the An- nulose and Vertebrate types of organization. 2. On the Morphological constitution of the skeleton of the Vertebrate Head. 3. On the Morphological constitution of Limbs. 1. On the MorpJiological relations of the Nervous System in the Annulose and Vertebrate types of Organization. The object of this communication is to indicate the morphological character of the apparently ditFerent anatomical relations of the central portions of the Annulose and Vertebrate nervous systems. The term Annulose is employed provisionally, and in a morphological sense, as including all animals possessing a ganglionic nervous collar and axis, and presenting, at the same time, more or less distinct indications of a segmented structure of body. Physiologists appear generally inclined to consider the central portions of the Annulose and Vertebrate nervous systems as modified forms of the same arrangement. These forms are held to possess a general similarity of structure, and correspondence in function ; and the ganglionic collar and axis of the Annulose are assumed to be homologous either with the cerebro-spinal axis, or with the series of ganglions on the posterior roots of the spinal nerves, or with the system of sympathetic ganglions of the Vertebrate animal. In my own examination of this subject I have been strongly impressed with the necessity of determining the morphological character of the oesophageal collar, and the opposite positions of the so-called brain and abdominal ganglionic cord, before any satisfactory advance could be made in ascertaining the relations of the two forms of nervous system. The apparent morphological difference between them does not appear, in the estimation of physiologists generally, to present that obstacle to a satis- factory comi3arison which its essentially fundamental character would lead us to expect. The difficulty has, however, been clearly stated by Professor Owen, who, in discussing the relations of the endo- and exo- skeletons in his Lectures on Fishes, page 21, says, — " Geoffroy StHilaire thought it needed but to reverse the position of the Crustacean — to turn what had been wrongly deemed the belly upwards — in order to demon- strate the unity of organization between the Articulate and Vertebrate animal. But the position of the brain is thereby reversed, and the ali- Proceedings of Societies. 119 mentary canal still intervenes iu the Invertebrate between the aortic trunk and the neural canal." I must here premise, that while I hold the general morphological rela- tions of the Annulose and Vertebrate nervous systems to be identical. I do not consider these two types of organization to be mutually reducible. On the contrary, they are fundamentally distinct, presenting differences which demand careful consideration. It is, nevertheless, incumbent on the morphologist to ascertain in what respects they correspond, so as to determine their distinctive limits. My earlier conception of the morphology of the Annulose nervous system was based on that of Cams. I conceived that each segment of the An- nulose animal contains potentially an annular nervous arrangement, set in a plane at right angles to the axis of the segment, or longitudinal axis of the animal ; that the only complete nervous ring is that one through which the oesophagus passes ; that the ganglions on this ring are arranged in the various forms of superior, lateral, and inferior oesophageal masses ; that the nervous rings in the post-cephalic segments are aU incomplete above, and have their ganglions united into a single or double mass below ; and that all the rings are united by a series of longitudinal abdominal commissures. According to this view, the oesophageal collar, with its superior, lateral, and inferior ganglions, is homologous with each pair of segmental nerves, and the corresponding abdominal ganglionic centre ; the oesophageal collar being in a plane parallel to those in which the post-cephalic ganglions and their pairs of nerves are situated, but at right angles to the line of the series of abdominal ganglions. I first recognised what I believe to be the real morphological relations of the Annulose nervous system during the delivery of a course of lectures on Invertebrate Anatomy in 1849 ; but more fully and completely during courses on the Anatomy of the Mollusca in 1850, and on the Anatomy of the Crustacea in 1851. I now perceived that the fundamental difference between the morpho- logical relations of the Annulose and Vertebrate nervous systems, con- sists in the position of the mouth. I saw that the entire axis or central portion of the nervous system extends along the neural aspect of the body in both types of organiza- tion ; but that while, as is well known — although its morphological importance does not appear to have been perceived — the Vertebrate mouth opens into the haemal, the Annulose mouth passes through the neural aspect of the body. In the Annulose animal, therefore, the buccal entrance interferes with the nervous axis — passing up between the two lateral halves of one of its longitudinal commissural or inter-ganglionic cords, so as morphologi- cally to divide the continuous axis into a,pre-stomal and a post-stomal portion. These relations are most satisfactorily seen in the Crustacea, in which the so-called brain, or supra-oesophageal ganglion or nervous mass is actually in front of the mouth, and not above it. In Insects, Annelids, and Mollusca, the bulk of the buccal mass, and other necessary modifications of the oral apparatus, elevate the so-called brain, curving upwards the morphological axis of the body of the animal. By comparing the indications of segments in front of the mouth, and their corresponding diverging appendages, with the arrangement and dis- tribution of the nerves given off from the so-called brain, it appears very evident that this brain is the aggregate of the segmental nervous centres in front of the mouth. In like manner indications afforded by the segments, and their appen- dages immediately behind the mouth, enable us to determine whether the 120 Proceedings of Societies. so-called sub-oesophagoal ganglionic mass is a single segiaental ganglion, or an aggregate of antero-posteriorly united segmental ganglions. In this way I was enabled to perceive that the axis of the nervous system of the Annulose animal does not consist of a supra-cesophageal mass, of an oesophageal collar, of a sub-cesophageal mass, and a continuous sub-intestinal ganglionic chain ; but of a continuous line of connected and serially homologous ganglions situated in the mesial line of the neural aspect of the body. The Annulose, like the Vertebrate animal, is dereloped with its ner- vous axis turned away from, and its ha;mal axis applied against, the vi- tellary mass,* But, in the course of development, the mouth of the Vertebrate opens through the surface applied against the vitellary mass, whilst that of the Annulose animal passes through the aspect turned away from it. The Vertebrate mouth is haemal, the Anntilose mouth neural. _ Rathke formerly described the pituitary body as originating in a diver- ticulum passing up from the pharyngeal mucous membrane through the basis of the embryo skull. I at one time conceived it to be probable that the pituitary body, and the mucous tube, in which, according to Rathke, it originates, might be indications in theVertebrateof a structure which, in the Annulose animal, is converted into the mouth. This presumed neural alimentary passage may be conceived as passing up between the bodies of the anterior and posterior sphenoid bones into the sella turcica, along the course of the infundibulum to the third ventricle of the brain, and through the cavity of that organ to its upper surface behind the cerebellum, thus leaving the origins of the nerves of smell and vision in the pre-stomal portion of the organ, while the origin of the nerve of hearing would re- main in the medulla oblongata or post-stomal portion of the cephalic ner- vous mass. The arterial circle of Willis, and other peculiar arrangements at the base of the skull and brain, appeared to support the view taken. I shall not, however, pursue this hypothesis further, because, from the observations of Reichart, we know that the base of the cranium is not perforated in the embryo, and that the supposed canal or diverticulum was an incorrect interpretation of the peculiar appearances produced by the curvature downwards of the early Mammalian head.f If I have determined aright the morphological relations of these two forms of nervous system, we shall have advanced a step in our concep- tions of the anatomico-physiological relations of the Annulose and Verte- brate animals, and this without losing sight of the fundamental differences, • From the passage in his Lectures already quoted, Professor Owen would appear to consider the dorsal heart, with its anterior and posterior arterial trunks in the decapod Crustacean, and consequently the dorsal vessel in the Insect, Arachnidan, and Annelid, as corresponding to the thoracic, abdominal, and caudal aortic trunk of the Vertebrate animal. On this supposition only can we understand his assertion, that when the so-called belly of the Crustacean is turned upwards, its alimentary canal is still interposed between the aortic trunk and the neural canal. Embryology, Comparative Anatomy, and Physiology, appear to me, however, to afford ample proof that the cardiac-arterial dorsal trunk of the Annelid, Crustacean, Insect, or Arachni- dan, is homologous not with the sub-spinal aorta of the Vertebrate, but with the pri- mordial cardiac-arterial tube in all the forms of the embryo Vertebrate, and, conse- quently, with the heart and trunk of the branchial artery of the Fish. If this, then, is the real homology of the " aortic trunk" of the Crustacean, and if its " brain" is in fact only a pre-stomal portion of its nervous a.^is, the French anatomist was quite correct in his general morphological statement, although he was not legitimately en- titled at the time to employ the illustration. t I have introduced the hypothesis of a Vertebrate neural mouth (cast aside in the course of my examination of the subject), because I believe it will be found to involve relations of importance in the anatomico-physiological investigation of the pre-stomal and post-fitomal portions of the Vertebrate and Annulose cephalic nervous masses. Proceedings of Societies. 121 developmental and structural, between them. The researches of Milne - Edwards, and of Newport and others, on the Annulose nervous axis may thus be physiologically associated with those of Wagner, Schroeder Van der Kolk, Owsjannikow, Jaeobowitsch, and Kupffer, on the cerebro-spinal axis ; and we may now legitimately employ the Annulose animal in the morphological investigation of the Vertebrate skeleton. Omitting, for the present, the consideration of the mode in whicb the nervous systems in the Tunicata, Rotifera, and Entozoa, are reducible to the typical Annulose form, I proceed to make some general morphological statements, based to a certain extent on the principle indicated in this, and introductory to the two following communications : — 1. The morphology of any one organic system in the Annulose or Ver- tebrate animal, cannot be safely or satisfactorily investigated, without constant reference to the others. That it must be so is evident from the fact, that all the organic systems are dependent on one another, in the constitution of the organism. 2. All sound morphological inquiry demands constant reference to the series of embryo, as well as of adult forms. 3. As morphology deals with forms and relations of position, it de- mands a careful selection of terms, and a methodized nomenclature. All terms involving more or less than their morphological application demands, must be avoided. Terms derived from other departments of the science, and having therefore an established technical meaning, have invariably produced misconception, when transferred for morphological purposes. Influenced by these considerations, and satisfied that the Annulose and Vertebrate types of organization, although fundamentally distinct, present parallel forms of structure, and must consequently be closely linked to- gether in morphological inquiry, I have to suggest a more extended and precise system of nomenclature for this department of the science, In the Annulose and Vertebrate types of organization, the body of the animal consists of a linear series of segments. To the constituent seg- ment, with its diverging appendages, I apply the term Somatome {vuftei. For the purpose of avoiding circumlocution, and of supplying a term for a generalized conception, and thereby facilitating morphological de- scription, without encroaching on zoological nomenclature, I denominate a segmented animal, whether Annulose or Vertebrate, an Entomosome — an entomosomatous animal {ivro/^o?. oo>fA.a,). As the constituent somatomes are invariably arranged in groups, in each of which they are more or less modified in form, or fused together, I find syssomatome {aw. aoifict. rifii/a) a convenient designation for such a group. A typical Crustacean presents a cephalic, a thoracic, and a caudal syssomatome, in each of which there are seven somatomes — twenty- one in all. The constituent somatomes lie in planes at right angles to the morpho- logical axix of the body, and are symmetrical in the transverse, but un- symmetrical in the perpendicular direction. They are, however, not only unsymmetrical in their upper and under surfaces, but the surfaces so named in the Annulose are morphologically distinct from those similarly designated in the Vertebrate animal. The Annulose animal moves on the surface which was turned away from the vitellary mass during develop- ment ; the Vertebrate animal moves on the surface which was applied to it during development. As the axis of the nervous system is formed at the surface turned away from the viteUary mass, and the axis of the vas- cular system is formed at the surface applied to it in both types of organi- zation, I employ, as morphological designations, the term Neuropod {viv^ov. 122 Proceedings of Societies. Toyy) for an Annulose, and Haemapod {aif^ct. woyr) for a Vertebrate animal. The moutli of the entomosomatous animal is invariably situated be- tween two somatomes, and so that a certain number of somatomes are interposed between it and the anterior termination of the body. As the mouth is only one of a number of openings situated between somatomes, I find such openings conveniently distinguished as metasomatomic. The mouth of the Neuropod is a neural, that of the Ha;mapod a haemal metasomatomic opening. As the somatome exhibits in its structure corresponding segments of certain or of all the organic systems, I have found the following morpho- logical terms extremely convenient in referring from the segment of one organic system, to the corresponding segments of the others. For tlio entire framework of an Entomasome, whether this framework be developed in its integument or in its interior, whether it be fibrous, carti- laginous or osseous, I employ the term Sclerome { In the Mammal the primordial cartilaginous streak in the second visce- ral lamina, and which is attached superiorly to the auditory region, di- vides into segments, the uppermost of which becomes the stapes ; while the succeeding become, in succession with the intermediate soft portions, the " stapedius muscle," the pyramid and its prolongation downwards, the styloid process, the stylo-hyoid ligament, and the series of sclerous ele- ments which terminates below in the anterior horn of the hyoid. The primordial cartilaginous streak in the third visceral lamina is at- tached to the occipital region, breaking up into four segments ; the two upper disappear ; tiie two lower become respectively the posterior horn and corresponding half of the body of the hyoid. In the second visceral lamina of the Bird, in like manner, the auditory columella is developed superiorly, and the feeble anterior horn of the hyoid below, while the elements of the suspensory or posterior horn of the hyoid are formed in the tliird visceral lamina. The fibrous septum of the tongue and the epiglottis of the Mammal make their appearance in the line of junction of the second and third visceral laminaj. The respective share taken by these two laminae in the formation of the so-called basi- glosso- and uro-hyals in the Bird remains to be determined. The precise observations of Rathke have shown that the lateral halves of the feebly-developed hyoid of the Ophidian are formed by the lower portions of the primordial cartilaginous streaks of the second pair of visceral laminae, while the auditory columella: are formed in their upper portions. Rathke also found that the primordial cartilaginous streaks of the third pair of visceral laminae, and which are attached to the occipital region, disappear altogether. There are no embryological observations in sufficient detail to indicate the morphological relations of the more or less complex hyoid apparatus in the Chelonian and Lacertian. The so-called hyoid, or suspensory arch of the branchial apparatus in the Ampliibia is developed in the second pair of visceral lamina;. The corresponding arch in the Tadpole, and the anterior or suspensory horn of the so-called "hyoid" of the Frog, are also developed in this pair of visceral lamina;. The suspensory arch of the branchial apparatus is attached to the quadrate, or so-called Proceedings of Societies. Ill "tympanic" piece of the mandibular arch, and not to the base of the cranium. Rathke had observed a filament extending between the auditory region of the cranium and the quadrate cartilage of the Tadpole. He found that the so-called " malleus and incus" are developed in this filament. According to Reichart, this filament appears to be the upper part of the second primordial cartilaginous streak, which, in consequence of the peculiar manner in which it curves forward superiorly towards the quadrate cartilage (a curvature of the same kind towards the quadrate bone has been observed by Rathke in the Adder), becomes attached to it. In consequence of this attachment, the hyoidean arch becomes sus- pended to the quadrate portion of the mandibular ; and the upper portion, between the quadrate cartilage and the auditory region of the skull, be- comes converted into those elements in the Frog, which have their homo- logues in the stapes of the Mammal, and the columella, with its cartilagi- nous extremities, in the Bird and Reptile. As the cartilaginous branchial arches of the Tadpole, and of the other Amphibia, are formed in the succeeding visceral laminae, it would appear to follow, as a necessary consequence, that the suspensory or hyoidean arch of the Amphibian, with its inferior mesial element, and along with the auditory ossicles, is homologous with the anterior or suspensory part of the hyoid, along with the stirrup-bones in the Mammal, and with the cor- responding structures in the Bird and Serpent ; and that the first branchial arch of the Amphibian, with its corresponding inferior mesial elements, are homologous with the posterior horns and body of the hyoid in the Mammal, and with the posterior or suspensory horns, with the correspond- ing inferior mesial elements of the hyoid in the Bird. The so-called posterior horns of the hyoid of the Frog cannot, therefore, be the homo- logues, as Professor Owen's statements might lead us to infer, of the pos- terior horns of the hyoid of the Mammal or Bird. The posterior horns of the hyoid of the Frog are the remains of its posterior pair of branchial arches, or enlargements of the posterior angles of its basi-hyals. They are developed therefore in its posterior visceral laminae ; while the pos- terior hyoidean horns of the Mammal and Bird are developed in the third pair of visceral laminaj. As the skeleton of the hyoidean and branchial apparatus of the Fish is developed in the form of a series of inverted arches in the corresponding visceral laminae, from the second inclusive, we are obliged to conclude that its hyoidean arch is the homologue of the stylo-hyoidean arch, with the stirrup-bones, — or second post-stomal arch — in the Mammal ; and of the corresponding portion of the hyoidean apparatus in the Bird, with the columella! ; and of the entire hyoid in the Serpent, with the columellae ; and that the first branchial arch in the Fish is the homologue of the corresponding arch iu the Amphibian ; of the posterior horns of the hyoid, and their associated elements in the Bird ; and of the posterior horns and body of the hyoid in the Mammal. It has not yet been determined upon what developmental change the suspension of the hyoidean arch of the Fish to its mandibular arch de- pends. It is probably of the same nature as that which occurs in the Tadpole, with this difterence, that the upper portion of the hyoidean arch disappears in the Fish, without developing a stapedial ossicle ; while its lower portion remains permanently connected to the mandibular arch, instead of regaining an attachment to the cranium. The hyoidean and branchial arches of the Fish are provided, as has been already stated, with a well-developed double series of actinapophyses, for the support of the branchiostegal membrane, and the branchial laminiu. These actinapophyses ui the Fish are foreshadowed in the Tadpole by the tubercular margins of its branchial styles. M 178 Proceedings of Societies. The question may now be put — if we are brought by reference to the development of the parts to allocate to the three post-stomal sclerotomes, haemal arches, consisting respectively of the sclerous parts developed in the three anterior post-stomal visceral laminae, to what sclerotomes are we to refer the potential or actual hsemal arches in the remaining visceral laminae? For reasons already stated, they cannot be disposed of by referring them to a splanchno-skeleton, because in that case the hyoidean arch or arches, and apparently the mandibular arch also, must be referred to the same category. Neither can they be referred to any of the cervical, or trunk sclerotomes ; because it would appear that the visceral walls of the head are alone perforated by clefts. We are not yet prepared to answer the question. It involves, as it appears to me, the investigation of 9. residual quantity, the solution of which will require some information in reference to certain points, regarding which we cannot at present be said to possess any. First, the development of the Cyclostomes, but more especially of Branchiostoma ; secondly, the mode in which the trunk sclerotomes increase in number and become arranged in groups ; thirdly, the mode in which the same changes proceed in the cranium ; fourthly, the determination of the series of cephalic nervous centres, with their corresponding nerves (neurotomes), more especially in the medulla oblongata, with the causes which determine the gi-ouping and order in which the cerebral nerves pass through the walls of the cranium. If there appears to be no sutRcient developmental grounds for making a distinction between the branchial arches of the Amphibian and Fish, as belonging to a splanchno-skeleton, and the hyoidean and mandibular as referable to the neuro- or endo -skeleton, it becomes important to deter- mine the signification of the sclerous elements of the larynx, trachea, and bronchial tubes. Without presuming to anticipate the minute observation of the development of the parts themselves necessary for the solution of a question of this kind, I would venture to suggest that the proper carti- lages of the larynx are developed from the inferior or mesial extremities of certain of the visceral laminae ; and that the cartUages of the trachea and bronchial tubes are a pair of highly developed actinapophyseal systems, referable to one of the posterior visceral arches. Post-stomal neuractinopophyses. — In addition to the auditory capsules, I recognise as post-stomal neuractinopophyses more particularly those ossicles attached to the post-frontals, mastoids, and external occipitals of Fishes. Those attached to the post-frontals may enter into the formation of the infra-ocular bony arch. Those, again, which are developed on the temporal and occipital sclerotomes are modified so as to co-operate in the cranial suspension of the scapular girdle. In conclusion, Goethe was the first to indicate the intermaxillaries, the maxillaries, and palatals, as elements of three distinct cranial segments. In the course of my investigation of the development of the teeth I became early aware of the correctness of Goethe's views on this subject, and have found myself, therefore, imable to coincide with the doctrine of Professer Owen as to the constitution of his palato-maxillary or nasal hsemal arch. 3. On the Morphological Constitution of Limbs. Carus, maintaining generally the doctrine of cephalic limbs, originally propounded by Oken, has at the same time given much greater precision to the conception of the skeleton of a limb, by viewing it as a system of elements radiating from the exterior of a costiform arch. Professor Owen, wliile he rejects with British and the majority of Foreign anato- mists, the fantastic doctrine of Oken and his immediate followers with regard to cephalic limbs, has adopted the general doctrine of the skeleton Proceedings of Societies. 179 of the limb as propounded by Carus, and has developed and applied it withmuch ingenuity to the illustration of actual structure. Professor Owen has, however, at the same time, by his allocation of the scapular girdle to the occipital segment of the cranium, as its haemal arch, and by the view which betakes of the opercular and brancheostegal elements, ac- tually reproduced the doctrine of cephalic limbs in another form. I do not propose in this communication to examine in detail the grounds on which Professor Owen's general doctrine of limbs is based, but shall merely state categorically those considerations which appear to me to render it untenable. 1. It is highly improbable that the sclerous elements of a limb should be derived from one, or at most two, sclerotomes, while its other elements, and more especially its nerves, are supplied by a greater number of somatomes. 2. It appears to be highly improbable that the bones which enter into the structure of an arm or leg, or that the corresponding sclerous parts in the lower animals should be the result of teleological subdivision of a single " diverging appendage" or " archetypal element." Professor Owen virtu- ally admits that these " teleological " elements have a morphological value when he institutes an inquiry into their "special" and "serial homologies." 3. It appears to me that the scapular girdle cannot be the haemal arch of the occipital segment of the head — firstly, because that segment is already provided with a haemal arch in the series of transitory and per- sistent sclerous elements developed in the third pair of visceral laminae ; secondly, because the scapular girdle is invariably found to be developed at or in the immediate neighbourhood of that part of the trunk of the ani- mal where it is ultimately situated ; and, thirdly, because it is impro- bable that the exceptions to a general law should be more numerous than the instances in which it is adhered to.* The germs of the limbs make their appearance when the ventral laminas of the primordial vertebral system are passing down towards the haemal margin. At first they resemble lappet-like projections of the inferior margins of these laminae ; they extend along at least four or five of their segments, and are situated in those regions of the body to which the future limb is attached, viz. , in the pelvic and posterior region of the neck, except in the Fish, in which the pectoral lappets are situated close behind the head. As the ventral laminae extend downwards, the lappets retain a position more or less elevated on the side of the trunk. At this stage they abo begin to exhibit a change in their form and position. They become first sessile, then pedunculated, and the peduncle then indicates by an angle at its centre the formation of the central joint of the shaft of the future limb — the elbow or knee-joint. At the same time, what I term the plane of the limb is changed. The lappet was originally developed in a plane, which is coincident with the axis of the corda dorsalis. This is the pri- mary or fundamental plane of the limb ; and when in this plane the lap- pet presents its radial or tibial margin forwards towards the head, and its ulnar or fibular margin backwards. When the limb leaves its primary position, it lies in its secondary plane, which cuts the corda dorsalis more or less obliquely, so that the radial or tibial margins of the limb are di- rected more or less forwards and inwards, and the ulnar and fibular back- wards and outwards. The permanently sessile pectoral lappets or fins of the osseous Fish exhibit a peculiar modification of the same movement ; • It is somewhat remarkable that the only embryologioal evidence which Professor Owen adduces in support of that portion of his Doctrine of Limbs, in which the anterior limb is assumed to be developed at or close to the head, is a reference to a passage in Rathke's Entwiokolung dor Schildkrbten, in which the author adduces the fundamental position of the bones of the shoulder — viz., the posterior region of the neck — as a circumstance tending to explain their ultimate passage into th« thoracic cavity. 180 Proceedings of Societies. they rotate on a transverse axis, so that their anterior or radial margins are directed downwards and their ulnar margins upwards. In the Sharks and Hays the pectoral and abdominal fins continue permanently in the primary plane. _ While the lappet is still in its primary plane, the rudiments of the girdle of the future limb may be detected under the integumentary cover- ing, and therefore external to the proper mass of the visceral wall of the body. In the primordial condition of the lappet of the wing of the Chick, Remak has detected four parallel streaks running to its outer margin, and continuous internally with the rudimentary nervous structures of the four primordial vertebrae, with which the attached margin of the lappet is con- nected. Guided by embryological facts and conclusions, to the more important of which I have just alluded, I have endeavoured to detect, more particu- larly in the osseous Fishes, Plagiostomes, Amphibians, and Reptiles, the principle which lies at the basis of the morphology of limbs. The view which this inquiry has induced me to take of the subject I shall, in con- clusion, state very briefly. 1. A limb does not necessarily derive its elements from one somatome — about fifty segments of the trunk appear to contribute towards the structure of the great pectoral fin in the Ray. 2. The nervous elements of the limbs appear, as in other parts of the vertebrate animal, to indicate most distinctly the morphological consti- tution of the sclerous elements. About fifty spinal nerves contribute the greater part of their hsmal divisions to the pectoral fin of the Ray ; and there are about one hundred fin-rays — a pair of fin-rays to each nerve, and derived from each sclerotome. This correspondence does not apparently exist between the fin-rays and nerves of the osseous fish ; but it may be fairly assumed that when we have detected the develop- mental circumstances which induce the attachment of the pectoral girdle of the osseous Fish to its cranium, as well as those peculiarities ex- hibited by its anterior trunk sclerotomes, this discrepancy will be explained. A more careful analysis than we yet possess of the num- ber of spinal nerves which supply branches to the limbs of the higher Vertebrata is still a desideratum in this department of the subject ; but it appears to be extremely probable, that in the Mammalia, at least five spinal nerves transmit filaments to the five distal divisions of the limb. It would appear, too, that, notwithstanding their plexiform arrangement at the attached end of the limb, the greater number of the filaments of each nerve reach their own morphological district at the distal part of the limb. The radial and the ulnar nerves are formed principally by the upper and lower roots of the human brachial plexus, that is, from the nerves of the upper and lower primordial segments with which the embryo limb was connected, and from which it derived its various elements. 3. The nerves supplied to a limb are not the inferior or ha;mal divisions of the spinal nerves, but radiating on actinal branches of these divisions. The intercostal nerves are not the nerves serially homologous with the roots of the brachial plexus. The thoracic nerves, serially homologous with these roots, are the intercosto-humeral and the succeeding middle intercosto-cutaneous. 4. Each sclerotome supplying elements to the structure of a limb sup- plies as a sclerous element a single actinapophysis ; or, as in the Rays, an anterior and a posterior — that is, a pair of actinapophyses. 5. From the structure of the mesial and lateral fin- rays of the Fish, the actinapophyseal elements of a limb may be assumed as primordially segmented. 6. The fin-rays in the Fish, and the phalangeal, metacarpal, and me- tatarsal bones of the higher vertebrata, are more or less persistent con- Proceedings of Societies. 181 ditions of the distal segments of the primordial actinapophyseal elements of a limb. 7. By atrophy, or otherwise, one or more of the segments in the successive transverse rows of actinapopbyseal elements disappear, so as to leave in Man, e.g., four elements in each carpal row ; two in the fore-arm ; one in the arm ; two in the next row for the coracoid and clavicle ; one in the proximal row for the scapula. 8. The nature of the subsequent changes, which the elements of the limb undergo, up as far as the shoulder or hip, may be inferred from an examination of the paddle of the Enaliosaur or Cetacean. 9. A careful application of the hypothesis to the linib girdles of the cartilaginous Fishes, Amphibia, and Reptiles, leaves me strongly inclined to believe that the coracoid is an actinapopbyseal segment between the humerus and scapula, prolonged downwards, towards the haemal margin of the body ; that the scapula is a proximal element, elongated towards the neural margin of the body ; that the clavicle is the only other retained element in the same transverse row as the coracoid, in front of it, and elongated liie it in the hajmal direction ; and that the corresponding elements of the posterior limb have a similar morphological signification.