Observations on the Cranial Nerves of Scyllium

THE investigations recorded here were undertaken in the first instance for the purpose of controlling certain determinations published by one of us in a previous number of this journal ¹ concerning the cranial nerves of Elasmobranchs. To this end we have carefully re-examined the specimens upon the investigation of which the former account was based, and have, in addition, made a large number of new preparations, illustrating more especially the later stages of development—stages M to Q of Balfour’s nomenclature. ²

During the course of our work so many altogether new and unexpected points were brought to light, that we soon found it necessary to widen considerably the scope and limits of our investigations, and have finally been led to attempt a complete account of the development of the cranial nerves from stage K to the adult form, and to endeavour in this way to connect together directly the accounts previously given of the early stages ³ with the descriptions of the nerves of adult Elasmobranchs published by Stannius, ⁴ Gegenbaur, ⁵ and other anatomists. ⁶

Owing to defective supply of materials, our observations on the stages earlier than K are too fragmentary to be reliable; this we greatly regret, inasmuch as many features in the early stages are of extreme importance, and would well repay thorough investigation..

In the present paper we propose to confine ourselves to the consideration of the preauditory nerves, reserving the postauditory, which present many features of peculiar interest, for a future occasion.

Our investigations have been conducted almost exclusively by means of sections of hardened embryos of Scyllium, and, as on former occasions, we have found a mixture of chromic and osmic acids superior to any other hardening agent. For the specimens from which our best preparations have been made we are indebted to the courtesy of the managers of the Southport Aquarium; our best thanks are also due to Mr. A. J. Moss, of Owens College, for his gift of a fine specimen of Musiehis, as well as for valuable assistance in connection with the literature of our subject.

—We do not propose to deal in the present paper with either the olfactory or optic nerves, inasmuch as the former has been already fully described, ¹ while concerning the latter we have nothing new to communicate; we therefore commence with the third or oculomotor nerve.

Concerning the development of this nerve we have very little to add to the account given in the paper already referred to. ³ At stage L it arises from the base of the mid brain, not far from the mid ventral line, by a large posterior ganglionic root and by several smaller anterior ones clearly distinguished from the former by possessing no ganglion cells. The nerve itself runs backwards as a long slender stem to the interval between the first and second head cavities (fig. 11 o., in), where it expands into a ganglionic swelling (fig. 11, o. c.g.) wedged in between the tops of the two cavities. From this ganglion the two main branches of the third arise; of these the upper one (fig. 15, III b), at a rather later stage, supplies the rectus superior and rectus internus muscles, whilst the lower one (fig. 15, III c) runs down behind the rectus inferior, and ends in the obliquus inferior muscle (fig. 15, o. i.). At stage K, at which our observations commence, the third nerve has the same point of origin and the same relation to the head cavities; it differs from the condition described above chiefly in not possessing anterior non-ganglionic roots, and in not having its terminal branches fully developed. At stages later than N the nerve seems to arise from the base of the mid brain by a single large ganglionic root, no smaller non-ganglionic ones being visible (fig. 14, III), though at stage N itself they are very prominent (fig. 13, III).

Besides the branches of the third nerve, mentioned above, there are two others in direct connection with the ganglion c.g.: of these the first, at stage L (fig. 11 o., N. c.), is a short nerve, which lies along the top of the second head cavity and serves to connect directly the ganglion c. g. of the third nerve with the Gasserian ganglion at the root of the fifth.

The second (figs. 10, 11, and 12, N) passes straight forward from the ganglion c.g., running through the walls of the first head cavity under the rectus superior and rectus internus, and through the substance of the sclerotic on the inner surface of the eyeball. Passing out from the orbit, immediately above the obliquus inferior, it still pursues its course straight forward, but becomes more superficial. In the earlier stages it crosses the root of the olfactory nerve, with which it lies in very close contact: in the later stages (figs. 12 and 15, N) it is, rather more dorsally situated, and then crosses the ophthalmic branches of the fifth and seventh nerves at a considerable angle (fig. 12, N), and ends in the skin at the extreme fore part of the head. After careful and repeated examination we have failed to detect any branch given off from the nerve at any point of its length.

At stage K all the above-mentioned branches of the third nerve are developed, except the upper branch, to the rectus superior and rectus internus, which we have failed to detect: the ganglion c.g. is very conspicuous, and the nerves N.c. and N. have the same structure and connections as at stage N, the 1 atter of the two stretching forward to the extreme anterior part of the head, in the skin of which it ends.

In the later stages the modifications which the third nerve undergoes are merely ones of detail, all the principal branches of the nerve being already established, and maintaining their relations practically unaltered in the adult. The most important changes concern the ganglion c.g.; this, which at stages K and L is a large prominent swelling (fig. 10, c.g.), in the later stages becomes far less conspicuous, and the ganglionic cells, instead of being concentrated at one spot, occur in small scattered patches at different parts of the nerve. This change is seen commencing at stage N (fig. 11, c.g.), where the ganglion has divided into two main portions, one part retaining its original position, whilst the other becomes connected with the nerve N. at some little distance from the third nerve; at stage o-p (figs. 14 and 15) ganglion cells appear to be constantly present at two well-marked points in the course of the third, (1) where the nerve N, is given off, and (2) immediately above the rectus superior.

At stages K and L (fig. 10) the angle between the nerves N.c. and N. is very considerable, and this increases in the later stages so much that at stage N (fig. 11) the two nerves are almost at right angles to one another; at stages later than this the nerve N. is much more difficult to define, whilst owing to the close proximity of the third to the fifth nerve it is only with extreme difficulty that the nerve N. c. can be distinguished at all.

We find, therefore, that the main stem and the branches III band in III c of stage N become directly the nerves which have the same course and relations in the adult. The ganglion c.g. becomes the ciliary ganglion of the adult. ¹ The nerves N. and N.c. become directly continuous with one another, and together form the nerve known as the Ramus ophthalmicus profundus. The discussion of the morphological import of these two very remarkable nerves we postpone till after the description of the fifth and seventh nerves.

—Concerning the development of the fourth nerve no description has yet appeared, and though our observations do not yet enable us to give a complete account, still, so far as they go, they are of so definite a character that we think it well to record them here rather than wait for the possibility of completing them at some future time.

The condition of the fourth nerve at stage N is well shown in figures 11 and 13, of which the former shows the greater part of its course, and the latter its root of origin. The nerve arises (fig. 13, iv) from the dorsal surface of the extreme hinder border of the mid-brain, so far back indeed that very careful examination is necessary to determine whether its origin is really from mid and not from hind-brain. The roots of the two nerves are in close contact on the dorsal surface of the brain. From its root the nerve runs at first almost directly outwards, following the curvature of the brain, until it comes to lie a very short distance below the surface; it then runs backwards and downwards as a long, straight, and very slender nerve (fig. 11, iv), which very commonly branches early in its course, and terminates in the obliquus superior muscle. Just before reaching the muscle (fig. 11) the fourth nerve crosses the ophthalmic branches of the fifth and seventh nerves at right angles, lying at a slightly deeper level than these nerves. It also divides into a number of branches, of which one or more appear to become connected with similar branches from the ophthalmic branch of the fifth,the remainder entering the obliquus superior, whilst in the later stages (fig. 16) the main nerve is seen to come into very close contact with the two ophthalmic branches themselves of the fifth and seventh nerves.

It will be seen from the above description that the fourth nerve has by stage N acquired its adult relations and distribution, and at stages o-p (fig. 15) and Q (fig. 16), where it is shown in transverse section, it is still the same. In the adult its course, like that of the other nerves, becomes altered owing to the varying rates of growth of the brain, the skull, and the face ; the rapid growth of the skull relatively to that of the brain causing the fourth nerve to run some distance forward within the brain case before passing out, when, as before stated, it runs immediately beneath and in very close contact with the ophthalmic branches of the fifth and seventh nerve.

In attempting to trace the fourth nerve in stages earlier than N we have met with considerable difficulties, and have hitherto obtained only a moderate amount of success. At M the relations are the same as at N, the sole difference being that the nerve is more slender than at the later stage. At stage L, though We have examined a very considerable number of specimens in excellent histological preservation, we have as yet recognised the fourth nerve in one specimen only, and even in that one not with absolute certainty. In the specimen in question the nerve has the same position and relations as at’ N, but is very much more slender, so thin, in fact, as to be almost unrecognisable. Earlier than L we have failed, after the most careful search, to find any trace of the nerve.

Though our observations are imperfect they yet seem to point to certain conclusions of importance touching the morphology of this important nerve. In the first place the fourth nerve is the only one in the body which, in the adult, arises from the dorsal surface of the brain; it is, therefore, of great importance to notice that from the very earliest stage at which we have seen it the point of origin is that of the adult nerve. ¹ Inasmuch as the majority of the cranial nerves, as well as the dorsal roots of the spinal nerves, arise at first from the dorsal surface of the brain or spinal cord, it seems natural to suggest that the fourth nerve differs from all the rest, not in its mode of origin, but in the fact that, whilst all the other nerves shift their attachment to a greater or less extent, it alone preserves the primitive position of its roots of origin. This shifting of the roots is, in part, ³ due to the rapid growth of the dorsal part of the brain forcing the roots of the two sides from each other; and it becomes of interest to notice that the fourth nerve arises from a portion of the brain where this rapid growth of the roof does not occur, and where, consequently, one cause of the change in the other nerves is absent.

The fact that the direction of the fourth nerve is at first at right angles, or nearly so, to the axis of the part of the head from which it arises is of importance, as showing that the fourth nerve comes under the category of segmental nerves; ¹ and inasmuch as there is no room as regards visceral arches and clefts for a segmental nerve between the third and fifth the fourth would probably be rightly viewed as a separated branch of the third— the only other nerve arising from the mid brain.

On the other hand, certain other facts in connection with the fourth nerve cannot be explained so easily. Thus, segmental nerves not only arise from the neural crest; they also arise early, are from the first large, and have ganglia at or near their roots. Now, so far as our observations go, the fourth nerve, though it may possibly arise from the neural crest, would not appear to arise early, and certainly is not at first a large nerve, being of much greater size at stage o-p (fig. 15) than at stage N (fig. 11); whilst at stage L, in the Single specimen in which it has been detected, it is a nerve of extreme slenderness; moreover, at no part of its length have ganglion cells been found—an important point of difference from segmental nerves.

Another curious feature concerning the fourth nerve is that, in our embryos, it appears almost constantly to divide close to its root into two or even more branches, as is well shown in fig. 11, Iv. These again subdivide near their terminations (figs. 11 and 15), but all the branches, whether primary or secondary, are distributed to the superior oblique muscles, with the possible exception of a few of the smaller ones, which appear to join the ophthalmic branch of the fifth. We would suggest that this branching may possibly be an indication of the fourth nerve having previously had a more extended distribution than its present very limited one.

On the whole, our observations lead us to believe that the fourth nerve is to be regarded as a separated part of that segmental nerve of which the third nerve forms the main portion. A further suggestion concerning the fourth nerve will be made after the seventh nerve has been considered.

—We propose to consider separately the roots and the branches.

—The earliest stage in the development of the fifth nerve that we propose to treat of in the present paper is that represented in fig. 1, taken from a transverse section through the hind brain of an embryo at a stage intermediate between i and K, the plane of section passing on the left side through the roots of the trigeminal (v).

As shown in the figure, the roof of the hind brain is very thin, and passes rather abruptly into the much thickened sides. To the top of the thickened portion the fifth nerve is attached by a long slender root (fig. 1, v a). This root passes down alongside the brain, but not in actual connection with it, widening considerably as it does so. Its inner border follows the curvature of the brain until reaching a point about half way down the sides of the neural canal; it then turns suddenly outwards, leaving the brain altogether, and forming, as it does so, a very conspicuous blunt projection (fig. 1, vβ), which is in rather closer proximity to the brain than is the part of the nerve immediately above it. The trunk of the nerve then passes outwards and downwards, lying just beneath the superficial epiblast, between it and the outer wall of the second or mandibular head cavity (fig. 1, 2). The whole of the nerve, including its root, consist of closely-packed spherical or polygonal cells, which, like all nerve cells in the early stages, stain very deeply with osmic acid, and differ materially in appearance from the much less closely arranged mesoblast cells.

By stage K the root of the fifth nerve has undergone very remarkable changes; as shown in fig. 4, the dorsal attachment (fig. 1, v a) to the top of the thickened side of the brain has disappeared completely, and the nerve is now attached to the brain at a point about half way down the side (fig. 3, v β), i. e. at a point exactly corresponding to the conspicuous projection (fig. 1, v β) already described at the earlier stage. Immediately beyond the root of origin the nerve enlarges suddenly, and presents a distinct dorsal projection at the base of the secondary root of attachment. Although hitherto we have not secondary in following all the intermediate stages, and have not yet obtained satisfactory preparations of the stages earlier than that drawn in fig. 1, yet we feel justified in putting forward the following explanation of the appearances we have just described, relying for our justification partly upon the description given by Balfour, and still more on our own observations on the development of the roots of the seventh nerve, which will be described immediately.

Balfour has described and figured the fifth nerve as arising at “stage G, near the anterior end of the hind brain, as an outgrowth from the extreme dorsal summit of the brain, in identically the same way as the dorsal root of a spinal nerve.” ¹

He has further described ³ how, by the growth of the roof of the brain, the nerves of the two sides, which at first are in contact dorsally, shift their position and become widely separate. His descriptions and our own somewhat fragmentary observations on these stages, when considered in connection with our much more complete series of observations on the seventh nerve, leave no room for doubt that the root of origin (v a) shown in fig. 1 is the primary root, the one which at stage G was situated at the top of the brain, and which has acquired its present position merely in consequence of the rapid growth of the roof of the brain pushing its two lateral halves apart, and so separating the roots of the nerves.

Concerning the root of attachment (V β) shown at stage K in fig. 3 there is more room for dispute. Balfour appears to bold ¹ that this further change in position is due to the same cause as the former one, i. e. to growth of the roof of the brain; but this explanation, while it would fully account for the first change, would in no way explain such a shifting of the root down the thickened sides of the brain, as is clearly seen to have occurred on comparing fig. 3 with fig. 1. We believe that what really happens is that about the commencement of stage K the nerve acquires a new and secondary connection with the brain at the point (v β) opposite the projection already noticed, that the primary attachment (v a) is lost, and that the part of the nerve left above the secondary root rapidly diminishes and ultimately disappears altogether, the slight dorsal projection already noticed in fig. 3 being the last rudiment of it. As our arguments in support of this view depend almost entirely on our own observations on the development of the roots of the seventh nerve, we postpone further consideration of the point till a later portion of this paper.

At the commencement of stage K then, the fifth nerve arises about half way up the sides of the hind brain by a single large root in which ganglion cells are abundant, and opposite to which there is a well-marked external bulging of the walls of the brain ; ² beyond this root the nerve expands suddenly into a very arge ganglionic swelling, the future Gasserian ganglion.

Before the close of stage K additional roots appear; a long, slender process runs forward from the anterior-superior angle of the Gasserian ganglion, and becomes connected with the brain some distance in front of the main root; in addition to which one, two, or more roots of a similar kind appear in intermediate situations.

These anterior roots of the fifth nerve are well shown at the next stage (L) in fig. 10, v γ, which shows that the fifth nerve at this time arises from the brain by three distinct roots, of which the posterior one is much the largest, and is the ganglionic root (v β) of fig. 3, while the two anterior slender non-ganglionic roots are the new ones. We have studied these roots very carefully, but have been unable to determine with certainty whether they are outgrowths from the brain to meet the, ganglion, or from the ganglion towards the brain; our observations, however, though inconclusive, tend very strongly towards the latter of the two alternatives. It has also occurred to us that these new roots may possibly be the original primary root of origin (v a, fig. 1), which, after losing its original attachment, has acquired a new one lower down; however, though the dates of the disappearance of v a and appearance of v γ agree fairly well with this hypothesis, we have yet no actual observations in its favour, and do not wish to lay stress upon it.

These anterior roots which, during stage L, may be three or more in number, appear in the later stages to be very constantly reduced to two, one of which is the most anterior of the original roots while the other appears to be formed by the fusion of the intermediate ones. This condition at stage N is well shown in fig. 11, v γ. At a stage between o and p (fig. 14, v γ) they are rather less conspicuous owing to the interval between them and the secondary root (v β) being filled up by dense tissue. They are clearly recognisable in the adult, and form, as will be noticed more fully further on, the first or anterior root of the fifth nerve of zootomists.

—-The Gasserian ganglion is, from its first appearance, wedged in between the dorsal ends of the second and third head cavities in the same manner as is the ciliary ganglion between the first and second (fig. 11). From the Gasserian ganglion, at stage K, two nerves arise; of these, one, which runs straight down between the second and third head cavities, and then along the anterior border of the mandibular arch in front of the ventral portion of the cavity in the latter, is the mandibular branch, and from this, a considerable distance below the ganglion, a small anterior branch—the maxillary nerve—is given off. The second branch from the Gasserian ganglion arises from its anterior inferior angle, runs along the top of the first head cavity and Joins the ciliary ganglion; it is the communicating branch between the fifth and third nerves already mentioned, and forms the proximal part of the ramus ophthalmicus profundus of zootomists.

At stage L a slender branch arises from the anterior superior angle of the Gasserian ganglion; this is the ophthalmic branch of the fifth (fig. 10, v a) which runs forward through the orbit dorsad of all the eye muscles, giving off branches to the neighbouring parts in its course and terminating in the skin of the fore part of the head. One other nerve in connection with the fifth remains to be noticed: this is the connecting branch (N. ć., fig. 10) between the fifth and seventh nerves : this is present at K, at which stage as well as at L, it forms a very stout; though short nerve; running forward and downwards from the seventh nerve, over the top of the third head cavity, to join the lower part of the Gasserian ganglion.

The. branches of the fifth nerve at stage N are well seen in figs. 11 and 12 which have been constructed so that each of them may show the whole course and distribution of certain nerves ; the outlines of the figures were drawn, with the camera, from individual sections and the branches of the several nerves carefully filled in, again by the aid of the camera, from other sections of the same series. In this way such a view of the nerve is obtained as might be got from a transparent embryo in which the nerves alone stood out as opaque objects. To prevent confusion, from the overlapping of different nerves, two figures have been given of which the first (fig. 11) shows the roots of the fifth and seventh, the connections of these with one another and with the third nerve, the branches of the latter and of the fifth, the fourth nerve, and some of the branches of the seventh ; in the second (fig. 12) the remaining branches of the seventh, with certain of the glossopharyngeal, are shown, and in addition to these, the whole course of the ophthalmics.

The branches of the fifth nerve are seen to be the following :

1. The ophthalmic branch, (v a) which arises by a slightly ganglionic root, runs forward over the obliauus superior (o. s.), crossing, as it does so, the fourth nerve at right angles and giving off branches, some of which appear to be connected with the fourth.

2. The communicating branch (N.c.) between the Gasserian and ciliary ganglia, the position and relations of which nerve are sufficiently well shown in the figure. We reserve the discussion of this branch, merely noticing here that, though we describe it with the fifth nerve, it appears to belong to the third quite as much as to the latter.

3. The main stem of the fifth, running down behind the first head cavity (1) and the rectus externus, receiving the communicating branch (N.ć.) from the seventh, and after passing downwards and forwards for some distance, dividing into two branches, (a) an anterior or maxillary nerve (v b) which again gives off numerous branches to the skin of the upper jaw, and (b) a posterior or mandibular (v c) which runs backwards and downwards, lying in close contact with the outer wall of the lower part of the mandibular head cavity (2), and supplying it with branches. The distribution of ganglion cells is sufficiently shown in the figure; there is a small ganglion at the base of the ophthalmic nerve, and the ganglion cells of the main stem extend some little distance beyond the point of junction with the communicating branch from the seventh.

In figs. 14 and 15 some of these branches (v a, v b, v c) are seen at a stage between o and p : except that the roots of v and vii. are much more closely approximated, there is no difference of importance between this stage and the earlier one which we have more fully described. We have traced all these nerves up to what is practically the adult condition, and have identified them with the branches bearing the same names in the adult. Our observations show that in the fifth, as in the third nerve, all the main branches of the adult nerve are fully established by stage L, and that the after changes are comparatively unimportant.

—Fig. 2 represents a transverse section through the roots of origin of the seventh nerve of an embryo between stages i and K, the same, in fact, of which fig. 1 represents the roots of the fifth nerve. The two nerves (VII) are seen to arise from the extreme dorsal summit of the hind brain, the roots of origin of the two nerves (VII a) being continuous with one another across the top of the brain. It will also be noticed, as contrasted with figure 1, that the two sides of the hind brain are close together, both above and below, and that the cavity of the hind brain is a mere vertical slit; that, in fact, the growth of the roof of the brain, which we have seen is the first cause of the separation of the roots of the fifth, has not yet commenced in the part of the brain from which the seventh nerves arise. The section further shows that the nerve on either side extends down as a somewhat club-shaped mass of compactlv arranged polygonal cells lying between the external epiblast and the neural canal, but distinct from both, its ventral end having a tendency to pass to the outer side of the third head cavity (fig. 2, ³), just as the fifth nerve passed to the outer side of the second cavity (fig. 1, ²).

The next stage is represented in fig. 3, a section through the hind brain and roots of the seventh nerve of an embryo of stage K, of the same age, though not from the same specimen as fig.4. The figure shows that very important changes have occurred; the roof of the hind brain has grown rapidly and considerably, so as to separate widely the two primary roots of the seventh nerves (VII a). On the right side only this dorsal primary root is seen, but on the left side a considerable portion of the nerve is shown, and it is seen that, in addition to the primary root (VII a), which is still present, the nerve has acquired a new or secondary root (VII β), about half way down the sides of the brain. Both roots of attachment are perfectly clear and unmistakable, while between them the nerve and brain are quite distinct from one another, and separated by an appreciable interval.

If fig. 3, showing the condition of the root of the seventh nerve at stage K, be compared with fig. 1, showing the root of the fifth nerve at a rather earlier stage, it will be seen at once that there is a very close resemblance between the two; the sole point of difference being that in fig. 1, though the nerve still retains its primary attachment, the secondary has not yet been actually acquired. Balfour’s figures and description, already referred to, show that at a still earlier stage the fifth nerve has exactly the same appearance and relations which the seventh has in fig. 2; and it is mainly on this fact, coupled with the close similarity between such specimens as those represented in figs. 1 and 3, that we rely in support of the explanation we have given above of the development of the root of the fifth nerve.

Inasmuch as figs. 1 and 2 are taken from the same embryo, it would seem that the fifth nerve appears before the seventh, and is, during the earlier phases of development, just one stage ahead of it in development. At a time (fig. 1) when the primary roots of the fifth have already become widely separated by growth of the brain-roof, and the secondary attachment (v β) is on the point of being acquired, the two seventh nerves (fig. 2) are still in contact with one another across the top of the unexpanded brain-roof; and at stage K the seventh nerve (fig. 3) is in exactly the same condition as the fifth at the end of stage I (fig. 1).

Our observations appear, therefore, to prove conclusively that as concerns the seventh nerve, while the change of position of the dorsal or primary root (vII a) is due solely to rapid growth of the roof of the brain, the lower or ventral root (vII β) is a new and purely secondary attachment.

Whilst these results concerning the roots of the seventh are, we believe, new as applied to Elasmobranchs, they are in perfect accordance with the account previously given by one of us of the development of the seventh nerve in the chick, in which the very same series of changes—the separation of the primary roots by growth of the brain-roof, and the acquiring of new or secondary roots—are shown to occur in a precisely similar manner. ¹ The close correspondence between these two very different types of vertebrates is of much interest, partly as tending to confirm the correctness of the account, and partly as showing that this curious shifting of the nerve roots, though clearly a change of a secondary nature, must yet have been acquired very early by Vertebrates.

The later stages of development of the roots of the seventh also present points of great interest. Fig. 6 represents a transverse section through the roots of the seventh nerve of the same embryo of stage N, of which fig. 5 shows the roots of v. The seventh nerve is seen to rise on either side by two roots, one (vII a) from the top of the sides of the brain at the junction of the thickened side with the thin roof, while the other (VII β) arises about half way down the sides of the brain. Between the two roots the nerve is in contact with the brain, but apparently not connected with it. We have traced the intermediate steps between figs. 3 and 6, and find that the upper root (VII a) of fig. 6 is the original dorsal or primary root, and the lower one (VII β) the secondary root of fig. 3. In other words, there is an important difference between the fifth and the seventh, inasmuch as in the former the primary root is lost and the secondary alone retained, whilst in the latter loth primary and secondary roots are retaimed up to stage N, and, indeed, as we shall see immediately, throughout life. The difference between the roots of the fifth and seventh nerves just noticed, does not occur in the case of the chick, in which the primary root of the seventh is lost as completely as is that of the fifth in Elasmobranchs. ¹

This shifting of the roots of origin and acquiring of a secondary connection with the sides of the brain is not confined to the cranial nerves. It has already been shown to occur in the posterior roots of the spinal nerves of the chick, ² and occurs also in the posterior roots of the spinal nerves of Elasmobranchs. It is a point of much interest to note that the seventh nerve, in the retention of its primary as well as its secondary root, is not only more primitive than the fifth, but more primitive even than the spimal nerves.

The condition of the roots of the seventh at stage o is shown in the left hand side of fig. 9, representing half of a transverse section through the hind brain and roots of origin of this nerve. The two roots, the primary (VII a) and the secondary (VII a), are even more distinct than at the earlier stages. The primary root (vn a) arises as before from the top of the thickened side of the hind brain just before its junction with the thin roof; from this origin the root runs downward, alongside of, and closely applied to the brain, but unconnected with it, to join the secondary root (VII β). This latter is now situated still nearer to the ventral surface than at its first appearance, the distance between the two roots being considerably greater than at the earlier stages, as is evident from a comparison of fig. 9 with fig. 6. The two roots also differ histologically, the dorsal or primary root consisting almost entirely of elongated fusiform cells, whilst the ventral root (vII β) is mainly composed of spherical ganglion cells.

This ventral root, at stage o, has, as shown in the figure, two distinct attachments to the brain, one just below the other. It is shown in longitudinal and vertical section at stage N in fig. li, where the brain presents opposite to its point of origin an external bulging precisely similar to that opposite to the secondary root of the fifth (v β). The dorsal or primary root (vII a) is shown at the same stage in fig. 12.

In fig. 14 the two roots of the seventh are seen in longitudinal and vertical section, at an age intermediate between stages o and p. The dorsal root (VII a) arises very far up the sides of the brain, in fact, as in the earlier stages, from the junction of side and roof; it is of considerable length, is widely separated from the secondary root, and still consists mainly of fusiform cells; the secondary or ventral root, which is overlaid and almost concealed by the primary root, is only seen in part, its most anterior portion alone being visible.

The dorsal or primary root is also well shown at the same stage in fig. 15.

—It will be convenient here to briefly summarise our results concerning the roofs of origin of the fifth and seventh nerves, and to trace their changes up to the adult form.

About the close of stage i the fifth nerve (fig. 1) still retains its primary attachment (v a) to the brain, and is on the point of acquiring its secondary one (vβ) : owing to the growth of ‘the roof of the brain the two primary roots, which were at first continuous across the top of the brain, are now widely separate. The seventh nerve (fig. 2) arises by its primary root from the dorsal summit of the brain, whose roof at this point has not yet commenced its rapid growth, so that the nerves of the two sides are still directly continuous with one another; there is as yet no trace of the secondary root of the seventh.

At stage K the fifth nerve (fig. 4) has completely lost its primary root and is now attached to the brain by the secondary root alone (v β); a slight trace of the former-is still present as a small dorsal projection on the nerve just beyond the root of attachment. Towards the close of stage K, the tertiary or anterior roots have appeared, arising almost certainly as outgrowths of the ganglion towards the brain; but whether these are altogether new developments or merely new attachments of the primary root is uncertain. The seventh nerve (fig. 3) is in the same condition as the fifth at the preceding stage; it is now attached by both primary and secondary roots, the former, owing to the growth of the brain-roof, being widely separate from one another.

At stage N the condition of the roots is much the same as at the end of stage K. The fifth nerve (figs. 5 and 11) is attached by its secondary and tertiary roots, the latter being very constantly two in number, of which the anterior is the larger and attached to the brain some distance in front of the secondary root (v β). The seventh nerve (figs. 6, 11. and 12) is attached by both primary and secondary roots, the nerve between the two being in contact, but not in connection with the brain; the secondary root (fig. 11) is divided into an anterior or facial, and a posterior or auditory division.

At stage o (figs. 8 and 9) the chief differences are—first, that, owing to increased* growth of the brain, the distance between the primary and secondary roots of the seventh nerve is much greater than before; secondly, that the roots of the fifth and seventh nerves, which, from the first, have been quite independent of one another, are now situated much closer together than they were at the earlier stages.

At stage O-P (fig. 14) the two roots of the seventh (vII a and VII β) are still further apart from one another, but are now very close to those of the fifth. The connection between the two nerves which we have already seen is fully established at stage K, and which is shown at stage Lin fig. 10 (iV. ć.) and at stage N in fig. 11 (N. ć.), is, by stage o-p, very much more extensive and intimate than previously. The roots of the nerves are still quite distinct from one another (fig. 14), but immediately beyond these roots the two nerves become so closely and extensively united together that it is impossible to draw a line of separation between them. The connection is rendered still more intimate by the crossing of one of the branches of the seventh, as will be described fully later on, over the main stem of the fifth, so as to lie in front of the branches of this nerve.

The condition of the roots, as of the branches, of the fifth and seventh nerves at stage O-P differs but little from that of the adult, the sole change of importance as concerns the roots being that the ventral roots (v β and vII β) approach still closer together, and come into actual contact.

The primitive distinctness, gradual approximation, and ultimate more or less complete fusion of the roots of the fifth and seventh is of great interest, as proving that the fusion of these two nerves, so characteristic of Pisces and Amphibia, is a purely secondary feature, and that the two are at first as independent of one another in these forms as they are throughout life in the higher Vertebrates.

In adult Elasmobranchs the combined roots of the fifth and seventh nerves are usually described together, and the descriptions of different observers, though not quite in harmony with one another in certain details, yet agree fairly well on the main points. Stannius, whose descriptions are the most elaborate, describes the combined fifth and seventh nerves as arising in Plagiostomes by three roots, ¹ of which one is seen on closer examination to be double, giving four roots in all; of these the first, or most anterior one, arises from the ventral surface of the medulla by two short non-ganglionic roots, which unite together shortly after leaving the brain. This root is in Raja, according to Stannius, mainly motor, supplying the muscles by which the respiratory movements of the anterior wall of the spiracle are effected, and also certain others in connection with the jaws. The second root of Stannius is large, lies posterior to the first, and is in close proximity behind with the auditory nerve; it may be distinguished into an anterior part which belongs to the trigeminal, and a posterior, more ventrally situated portion, which belongs to the facial. The third root is very large and much more dorsally situated than the other; it is connected by its deeper fibres with the second root, whilst, from its superficial fibres are derived, according to Stannius, the nmus superficialls of the fifth, and also, in part, the maxillary and buccal nerves.

Gegenbaur, ² in his account of the cranial nerves of TRexanchus, distinguishes between the roots of the fifth and the seventh. He describes the fifth as arising by the union of two trunks of about equal size, an anterior and a posterior; of these the anterior (a) arises from the ventral surface of the medulla by two roots situated very close together; the posterior (b) has also two distinct roots, a dorsal one (a) arising from the side of the medulla by a large swelling projecting into the fourth ventricle, and a ventral one (B) situated immediately above the root of the facial, and in front of, and above that of the auditory.

The seventh nerve in Hexanehus is described as arising by two roots, a larger one immediately in front of the auditory, and a smaller one passing to it from the fifth.

Jackson and Clarke ¹ describe the combined fifth and seventh nerves in Echinorhinus as arising by three roots; an anterior inferior root (v α), itself with two well-marked rootlets, a second root (v β) arising by a well-marked superior rootlet from the lobus trigeminus and by a smaller inferior one, and a third root (v γ and vII) closely connected with the second one.

Balfour ² describes the fifth nerve in Scylliwm stellare as arising by three roots:—(1) an anterior more or less ventral root; (2) a root rather behind this arising by two strands, a dorsal and a ventral, and closely connected behind with the root of the seventh; and (3) a quite distinct dorsal and posterior root situated slightly behind the dorsal strand of the second root. The seventh nerve is described as arising by a single root close to, and behind, the second root of the fifth.

On comparing these descriptions of adult Elasmobranchs with our own observations on embryos and adults we are led to the following conclusions:

The fifth nerve in the adult arises by two roots :

a. An anterior one arising from the ventral surface of the medulla by two non-ganglionic rootlets, whose distinctness varies much in different adult Elasmobranchs. These rootlets are the tertiary or anterior roots of our embryos (figs. 10, 11, and 14, v γ). This root corresponds to the first root of Stannius, the anterior root (a) of Gegenbaur, the anterior inferior root (v a) of Jackson and Clarke, and the anterior root (1) of Balfour.

b. A posterior larger ganglionic root, the ventral or secondary root of our embryos (figs. 10, 11, and 14, v β). This is at first quite distinct from the root of the seventh, but during the later stages of development gradually approaches this latter, and in the adult cannot be clearly distinguished from it.

This root is the anterior part of the second root of Stannius; the ventral division (β) of the posterior root (b) of the fifth of Gegenbaur; apparently the inferior rootlet of the second root (v β), and possibly part of the third root (v γ and vII) as well, of Jackson and Clarke; and the second root (2) of the fifth of Balfour.

The seventh nerve in the adult arises by two roots :

a. A dorsal root arising far up the side of the medulla, at the junction of the thickened sides and thin roof of the fourth ventricle. This root is the primary or dorsal root of the seventh nerve of our embryos (figs. 2, 3, 6, 9, 12, 14, and 15, vII a). It has by previous observers been almost invariably described as a root of the fifth, and never as a true root of the seventh; our description and figures here given, leave no room for doubt that it belongs to the seventh. As already noticed, the retention of this root marks the seventh as being more primitive than the spinal, and possibly more so than any of the other cranial nerves, all the other nerves apparently retaining their secondary roots only.

This root is the third or dorsal root of Stannius; the dorsal rootlet (a) of the posterior trunk (b) of the fifth of Gegenbaur; the superior rootlet of the second root (v β) of Jackson and Clarke, and the dorsal and posterior root (3) of the fifth of Balfour.

b. A ventral root arising from the side of the medulla at a rather lower level than the posterior root of the fifth. This is the secondary or ventral root of the seventh of our embryos (figs. 3, 6, 9, 10, 11, and 15, VII β). The auditory nerve is at first derived from this root, but in the adult appears to be more distinct from the facial than is the case in the embryo. This root is, at first, some little distance behind the secondary one (v/3) of the fifth nerve (figs. 10 and 11), from which it is perfectly distinct in the later stages the two roots gradually approach one another (fig. 14), and in the adult are usually in close contact.

This root is the posterior part of the second root of Stannius, the root of the seventh of Gegenbaur; part, or possibly the whole of the third root (v γ and vn) of Jackson and Clarke; and the single root of the seventh of Balfour.

It would appear, therefore, that the fifth nerve loses its primary root, retains its secondary, and acquires tertiary roots, while the seventh retains both primary and secondary. Concerning the fourth nerve we would suggest the possibility that it may prove to be the primary root of a nerve of which the third nerve is the secondary root, which has, in this case, acquired complete independence.

—At stage K the seventh nerve, which, as we have already seen, has acquired its secondary as well as its primary roots of origin (fig. 3), expands below the secondary root into a large ganglionic swelling lying immediately behind the third head cavity. From this enlargement three branches arise: (1) from the anterior and superior angle of the ganglion a large nerve with a ganglionic base arises, and runs forward along the dorsal surface of the head, lying just beneath the superficial epiblast; this is the ophthalmic branch of the seventh, and is referred to in our figures as vn a.

(2) The second branch, which is also large, and has a ganglionic base, arises from the front part of the ganglion immediately below the root of the ophthalmic; its deeper portion runs forwards and slightly downwards over tbe top of the third head cavity, and becomes connected with the main stem of the fifth; it is referred to in the figures as N. ć. The more superficial portion passes on further forward in the same direction, crosses the mandibular arch, and enters the maxillary process, lying immediately superficial to the maxillary nerve, and just beneath the external epiblast; it is referred to in the figures as VII d.

(3) The third branch is the direct continuation of the main stem of the facial nerve; it runs downwards and backwards along the anterior border of the hyoidean arch, and is the rudiment of the posterior or hyoidean branch of the seventh in the adult; it is referred to in the figures as vII c.

Of these branches the first, or ophthalmic, is from its earliest appearance connected with the dorsal or primary root of the seventh rather than with the ventral root. The second branch is the most remarkable of the three; its deeper portion forms, as we have seen, a direct connection between the fifth and seventh nerves, a communication which appears to be very early established, inasmuch as by stage K the connecting branch is already a nerve of considerable size; the superficial portion of this branch (VII d) is noteworthy, mainly on account of its very close relations with the maxillary division of the fifth nerve.

At stage L the only changes of importance are, (1) that the several branches have increased in size, and, excepting the branch VII d, which has a very straight course, and ends abruptly in the skin, have divided into secondary branches near their terminations ; and (2), that a small anterior branch has arisen from the hyoidean nerve (VII c), some distance from the brain, which runs forward over the top of the spiracular or hyomandibular cleft, and then down in the anterior wall of the spiracle, i. e. in the posterior portion of the mandibular arch; this branch will be referred to as VII b.

The several branches of the seventh nerve at stage N are well shown in the diagrammatic figures 11 and 12. The ophthalmic branch (VII a) is seen in fig. 12 arising from the base of the primary or dorsal root (VII a) as a stout nerve, which expands very shortly after its origin into a large somewhat fusiform ganglion, beyond which the nerve runs forward as a stout trunk to the extreme anterior part of the head. Throughout its course it lies just beneath the external epiblast, and immediately dorsad of the ophthalmic branch of the fifth (v a), with which it is in very close relation; like this latter nerve it gives off branches along its whole course to the integument of the neighbouring parts, the branches being few in number at the proximal end, and much more numerous distally. A short distance before its termination this nerve, like the ophthalmic branch of the fifth, is crossed at a considerable angle by the nerve A (fig. 12).

The connecting branch (N. ć.) between the seventh and fifth nerves is well seen in figs. 11 and 12; it is now shorter and wider than at stage L (fig. 10), and contains very numerous ganglion cells along its whole length.

The superficial portion of this nerve (VII d) is not shown in fig. 11, but is represented along its whole length in fig. 12; it is a stout nerve with a remarkably straight course; it gives off no branches at all along the greater part of its length, but near its distal termination divides rather suddenly into a number of branches, which end in the integument of the maxillary process, the most anterior of the branches extending forwards almost as far as the hinder border of the olfactory pit. The relations of this nerve to the maxillary branch of the fifth are very curious; the two nerves are very close together, the branch of the seventh lying, as at the earlier stage, immediately superficial to that of the fifth. These relations are well seen in the transverse section drawn in fig. 6. This shows, as already noticed, the primary and secondary roots of the seventh, and also the proximal portion of the nerve vn d. This nerve is seen to be a direct continuation of the primary root; its inner, or deeper, portion is seen on both sides to become continuous with the maxillary branch of the fifth, (v b), the junction of the two forming the connection between the fifth and seventh nerves already noticed. Beyond this point of union the nerve vn d is continued downwards, lying immediately superficial to the maxillary nerve (v b). The two nerves preserve this relation up to their terminal distribution, two of the ultimate branches being represented in figure 5 (v b and VII d). This very remarkable branch of the seventh puzzled us greatly for a long time, and it was only after working out the whole history of its development up to stage Q that we succeeded in determining its import. The nerve has already been described and figured by one of us, and named tentatively the palatine. This determination now proves to be erroneous; the palatine is a deep-lying nerve, whereas the nerve VII d retains its superficial position in the adult.

This nerve (VII d) we have now identified as the buccal nerve, the proof of this determination, consisting in our having traced the nerve directly up to the adult. The buccal nerve has always hitherto been regarded as a branch of the fifth, and is described as such by Stannius, ¹ Gegenbaur, Jackson and Clarke, Balfour, and others. Stannius points out that the buccal nerve in fishes is very variable; that it may either arise from the main stem of the fifth, from the maxillary or the mandibular, or from both these nerves, or, finally, may be absent altogether.

Up to stage N, as shown in figures 6, 11, and 12, the buccal nerve is clearly a branch of the seventh, and could not possibly be taken for a branch of the fifth. After stage N, however, the connection between the roots of the fifth and seventh nerves becomes, as we have seen, very much, more intimate; and at the stage between o and p, represented in fig. 14, the buccal nerve (VII d), which is now situated completely in front of the maxillary (v b), might very easily be taken for a branch of the fifth rather than of the seventh; careful examination shows, however, that the buccal, which is, as before, the most superficial of all the ventral branches, can be traced up to the dorsal root of the seventh from which it arises, as in the earlier stages.

This origin of the buccal nerve from the root VH a has already been noticed by Stannius, ⁶ who, however, as we have seen, did not refer the root in question to the seventh. Stannius’ figure of the nerve in Chinuerd⁷ shows clearly the very superficial position of the buccal nerve and its independence of both maxillary and mandibular nerves.

Of the remaining branches of the seventh the anterior one (VII b) is shown in fig. 11 at its origin arising from a large ganglionic swelling on the main or hyoidean branch of the seventh, and running forward in close contact with the top of the spiracle (sp.), in front of which it divides almost at once into two branches, the distribution of which is shown in fig. 12; of these the anterior one (VII PA) runs downwards, forwards, and inwards, giving off numerous branches to the roof of the mouth. In fig. 12 the anterior branches of this nerve appear to cross the posterior branches of the buccal, but it must be borne in mind that at this point the two nerves are at very different levels, the buccal being very superficial and the nerve (VII pa) lying very deep. This latter is seen in transverse section in fig. 7, which shows, on the right side, its origin from the ganglion, and, on the left, its distribution to the mucous membrane of the mouth. By comparing this figure with fig. 6, the difference of levels between the two nerves will be at once apparent. This anterior branch (vn pa) is the palatine nerve ; it has already acquired by stage N its characteristic distribution, and undergoes no further change of importance from this period up to the adult stage.

The. second or posterior division (fig. 12, VII sp) of the nerve (VII b) runs downwards and slightly backwards along the anterior border of the spiracular cleft; it gives off branches along the whole of its length, the great majority of which run backwards to the mucous membrane of the border of the cleft and to the spiracular branchia. This nerve, which at this stage is of about equal size with the palatine, is the spiracular or prae-spiracular nerve of zootomists.

The only branch of the seventh still left for description is the main trunk or hyoidean branch (fig. 11, VIIc), which forms the direct continuation of the main stem of the nerve. This, as is seen from fig. 11, arises from the ventral or secondary root of the seventh, and is at its origin closely connected with the auditory nerve (VIII). Immediately after the auditory nerve leaves it, the facial forms a ganglionic swelling from which the communicating branch (N. ć.) to the fifth nerve is given off; beyond this point it is continued for a short distance as a stout nerve with comparatively few ganglion cells; this very speedily dilates into the large ganglionic swelling on the top of the spiracular cleft, from which the anterior branch (VII b) is given off. The main stem of the seventh (vn c) continues its course downwards, running along the anterior border of the hyoid arch and very close to the posterior border of the spiracular cleft; during this part of its course it contains few or no ganglion cells, it gives off a number of branches, of which the first is the largest, from its posterior border which supply the muscles derived from the wall of the third head cavity (3). A short distance below the lower edge of the spiracular cleft the nerve divides into two branches, of which the anterior (VII c¹.) runs forward along the lower border of the mandibular arch, sending numerous branches to the integument of this part and extending forward so as to come into very close relation with the posterior branches of the maxillary division of the fifth (v b). The posterior of the two branches (VII c, 2) into which the seventh divides continues the direction of the main stem, and runs down in the hyoid arch just in front of the third head cavity, in the terminal dilatation of which it ends. Of these two terminal branches of the seventh, the anterior, sensory, and superficial one is the ramus mandibularis exlernus of Stannius ¹ and Gegenbaur, ² while the posterior, muscular, and deep branch is the ramus mandilmlaris internus v. profundus of the same authors. Both these branches have already acquired, by stage N, not only the characteristic distribution of the adult nerves, but nearly all the minor branches as well.

To recapitulate; we find that by stage N the seventh nerve has acquired all the important branches of the adult nerve, the main trunks and many of the branches being fully developed at a much earlier period—stage L. The seventh nerve at stage N has two roots, a dorsal or primary, and a ventral or secondary. From the dorsal root (VII a) arise two branches : (1) the ophthalmic (VII a) and (2) the buccal (VII d), both of which appear to be purely sensory nerves. The connecting branch (N. ć.) to the fifth nerve, though it appears in longitudinal section (fig. 11) to be a distinct branch, in transverse sections (fig. 6) seems to be only the deeper portion of the buccal nerve. Brom the ventral root arises the main or hyoidean branch (VII c), from which the branch vn6 runs forward over the top of the spiracle, dividing, almost immediately, into the palatine (pa) and spiracular (sp) nerves, whilst the hyoidean itself divides distally into the sensory ramus mandibularis externus (v c, 1) and the motor ramus mandi-bularis internus (v c, 2).

—Concerning the development of the sixth nerve our observations simply confirm the account already given by one of us. ¹

The whole length of the nerve is shown in longitudinal and vertical section in fig. 13 (vI), where it is seen arising from the base of the brain by a number of small non-ganglionic roots which unite to form a slender nerve; this nerve runs forwards a short distance, then turns downwards, pierces the investing mass (i. v.), and ends in the posterior extremity of the rectus externus muscle (r. e.).

Fig. 7 shows the sixth nerve in transverse section at the same stage (N) : on the left side of the figure the termination of the nerve in the rectus externus is seen; while on the right side, which is taken from a more posterior section, one of the roots of origin is seen. This figure shows that the roots of the sixth are considerably nearer the mid ventral line than are the secondary roots of the seventh nerves; and also that the sixth and seventh nerves are quite independent of one another. At stage N the sixth nerve appears to be altogether behind the seventh, but in stage o it is situated rather further forwards, so that the same section may pass through the roots of both nerves.

We have not yet detected the sixth nerve in embryos younger than stage L : concerning the morphological value of this nerve we adhere to the opinion already expressed that it is to be viewed as bearing the same relation to the seventh that the anterior root of a spinal nerve does to its posterior root.

—This nerve also we can dispose of briefly: at stage K it appears as a large ganglionic posterior branch of the seventh nerve, given off immediately beyond the root of origin. It is from the first connected with the ventral or secondary root(VII β). The condition at stage L is shown in fig. 10 (VIII). At stage N (fig. 11) its root, though still intimately connected with that of the facial, shows a very evident line of separation from it j the ganglionic character of the auditory nerve placing it in marked contrast with the non-ganglionic root of the facial. This distinction between the two roots becomes more marked in the later stages.

—Several questions of a more general character arise out of the facts we have recorded above, and we propose to conclude the present paper with a brief notice of the more important of them. The problems in connection with the roots of origin of the nerves have been already sufficiently discussed, so that we turn at once to the consideration of their branches, concerning which the most important points are the determinations of the equivalence of the branches of the different nerves to one another.

We commence with the ophthalmic branches of the fifth and seventh nerves, the branches named v a and VII a in our figures. These two nerves, whose courses and relations are well shown in figs. 11 and 12, accompany one another very closely along their whole length; they appear to be both sensory nerves, their branches being distributed exclusively to the skin of the top and front of the head, and more especially to the mucous canals of these parts. Of the two nerves the branch of the fifth nerve (v a) is the smaller and the more ventrally placed of the two : though the smaller, its branches are, especially in the earlier part of its course, more numerous than those of the seventh. The two branches in their course through the orbit lie dorsad qf all the other contents of the orbit. They are at first quite distinct from one another (figs. 11 and 12) and lie close beneath the external epiblast (fig. 5, VII a); the branch of the seventh being the more superficial of the two. In the later stages of development, as in the adult, the two nerves lie in very close contact with one another (fig. 16, v a and VII a), the branch of the seventh lying immediately dorsad of the branch of the fifth ; they also, as shown in fig. 16, lie at a deeper level than at the earlier stages.

The fourth nerve bears, as already noticed, a very close relation to these branches. As shown in figs. 11 (Iv) and 16 (Iv) it crosses the ophthalmic branches at right angles, lying at a slightly deeper level but turning outwards immediately beneath them, to end in the superior oblique muscle (o.s.). At the point of crossing the branches of the two nerves are in very close relation with one another, and we are inclined to believe that a communication exists between the fourth nerve and the ophthalmic branch of the fifth at this point, though we have failed to determine this with certainty.

In determining the morphological value of these ophthalmic branches of the fifth and seventh nerve, very valuable evidence, by which we have been much influenced, is afforded by the condition of the glossopharyngeal nerve. This nerve, at stage L, gives off, just beyond its root of origin, a slender dorsal branch (fig. 10, Ix a), which, at first passing upwards and backwards, soon curves round the hinder end of the auditory vesicle (aud.), and reaching the dorsal surface of the head, runs forward a short distance, lying immediately beneath the superficial epiblast. It gives off branches along its whole course, which are distributed, as shown in the figure, to the integument of the top of the head. At stage N this nerve (fig. 12, Ix a) has the same course and appearance, the only differences being that it extends rather further forwards than at the earlier stage, so as to reach some distance in front of the middle of the auditory vesicle; and, secondly, that its branches are now seen to be in connection with the commencing mucous canals of this region. This branch of the ninth nerve is clearly the ramus dorsalis, and an examination of the figures 11 and 12 will, we think, leave no doubt that the nerves VII a and v a, which have a similarly superficial course and a like distribution to mucous canals, must be viewed as the equivalent branches of the seventh and fifth nerves. We are, therefore, led to adopt the view put forward by Balfour, that the ophthalmic branches v a and VII a of the fifth and seventh nerves are the rami dorsales of these nerves. ¹ Stannius and Gegenbaur speak of the ophthahnics as rami dorsales, but refer them entirely to the fifth.

What the causes are which have led to the very marked extension forwards of the rami dorsales of these nerves is not very evident; we would suggest that it is due mainly to an extension forwards and accumulation at the anterior end of the head of the special tegumentary sense organs—the mucous canals—this extension forwards involving a corresponding extension of the nerves supplying these organs; in connection with this suggestion it is of interest to note that no one of the nerves in front of the fifth sends any branches to these organs. Whether there is any trace of a ramus dorsalis to the third is very doubtful ; at any rate the fourth nerve cannot be the ramus dorsalis of the third, as its course is, at first, at right angles (fig. 11) to the rami dorsales of the fifth and seventh nerves; and, secondly, it is a motor and not a sensory nerve.

We now come to a far more intricate problem, viz. the import of the connecting branches between the third, fifth, and seventh nerves, with which it will be convenient to consider the nerve N. (figs. 10, 11, 12, and 15).

These three nerves, N. c., N. ć. and N. all appear very early; we have failed to determine the date of their first origin, but by stage K they are fully established. The posterior one (W. ć.), connecting the fifth and seventh nerves together, is the most difficult to investigate, owing to its appearing from the first as merely the deeper portion of the buccal nerve (fig. 6); in longitudinal sections, however, it appears very distinct (vide fig. 10). It is from the first much shorter than either of the other two nerves we are considering, and in the later stages (fig. 14) and the adult condition, owing to the close approximation of the fifth and seventh nerves, ceases to be visible as a distinct trunk.

The second of the three nerves (N. c., figs. 10 and 11) forms, as already noticed, a direct connection between the Gasserian ganglion of the fifth and the ciliary ganglion (c. g.) of the third nerve, and is much more slender than N. ć. Concerning the nerve in question, it is of the utmost importance to notice that not only is it fully established at the stage at which our observations commence, but that it is from the very first a connecting nerve, and that there is no reason whatever in the early stages for considering it as belonging to the fifth rather than to the third nerve. We have, therefore, in this paper given it a perfectly neutral name.

The last of these nerves, W., is still more remarkable; like the others it is present at K. Starting at this stage from the ciliary ganglion it runs an almost perfectly straight course to the anterior end of the head, ending abruptly in the external epiblast, and giving off no branches whatever. At stage L it is in very close relation to the olfactory nerve, and in some specimens seems to be connected with it, though we cannot speak with certainty on this point.

As soon as the eye muscles are established they have very definite relations to this nerve; the rectus superior and internus, and the obliquus superior lying above it, and the remaining three muscles below it (fig. 11). It also passes through the substance of the sclerotic, as noticed in a former-portion of this paper.

Concerning the morphological significance of these nerves, bearing in mind their very early appearance and the total absence of evidence for regarding them as branches of either of the nerves they serve to connect, we are disposed to view the nerves N. c. and N. ć. as persistent portions of the neural ridge between the outgrowths to form the third, fifth, and seventh nerves, and as being, therefore, homologous with the primitive commissure connecting the posterior roots of the spinal nerves together. ¹ As to the nerve N. we are much more in doubt; its apparent connection with the olfactory nerve at L, if confirmed, would tell in favour of its being regarded as a similar commissure between the third and olfactory nerves, and would greatly support views previously advanced by one of us concerning the morphological value of the olfactory nerve. ² On the other hand, the extension forwards of the nerve N. beyond the olfactory nerve to the extreme anterior end of the head must, for the present, be regarded as almost conclusive against its commissural nature. In this case it can only be a branch of the third nerve, for the only other nerve with which it is in direct, or indirect, connection is the connecting nerve (N. c.) between the third and fifth, which, if it does not belong to the third, there is at any rate no reason for referring to the fifth.

In the adult Scyllium this nerve retains the relation to other nerves which it has clearly acquired by stage N ; it is described in the adult by Schwalbe ³ as “dieser scheinbare Zweig des Oculo-motorius.” In Mustelus Schwalbe describes it as a branch of the fifth. We much regret that we have had no opportunity of studying the development of this nerve in Mustelus; should it prove to arise as in Seyllium, then it must definitely be regarded as a branch ofthe third.

As we have already pointed out, the nerves N. c. and N. together make up the ramus ophthalmicus profundus of zootomists, a nerve which seems to have escaped Balfour’s notice both in the adult and in the embryo. Balfour does, indeed, in his description of the nerves of the adult Scyllium, speak of a ramus ophthalmicus profundus, but inasmuch as he says concerning it that “this latter nerve arises from the anterior root of the fifth, separately pierces the wall of the orbit, and takes a course slightly ventral to the superior ophthalmic nerve, but does not (as is usual in Elasmobranchs) run below the superior rectus and superior oblique muscle of the eye,”¹ itis clear that he is describing the ophthalmic branch of the fifth and not the trice profundus, whose existence he has overlooked. There appears to be considerable confusion in the use of the terms ramus ophthalmicus superficialis and ramus ophthalmicus profundus by different writers, a confusion which our observations on Scyllium may help to remove. We find, as already stated, three perfectly distinct nerves to which the term ophthalmic nerve can be, and is, applied; of these the two dorsal ones (v a and VII a of our figures) are the rami dorsales of the fifth and seventh nerves, and may be spoken of as the ophthalmic branches of the fifth and seventh nerves respectively. Both these nerves are very superficial along their whole course, and both lie dorsad of all the eye muscles and other contents of the orbit. The two nerves are at first perfectly distinct, but in the adult unite more or less closely together, the extent of their union varying much in different Elasmobranchs; the two together constitute the ramus ophthalmicus superficialis.

The third of the ophthalmic nerves, the ramus ophthalmicus profundus, has a very different course, and is of a totally different nature; it is formed in ScyUium by the union of the connecting branch between the fifth and third nerve (N. c.) with the branch N of the third nerve. It is very definitely characterised by its course ventrad of the superior rectus, superior oblique, and internal rectus muscles, by its close relation with the inner wall of the eyeball, by the fact that the ciliary ganglion is either in its trunk or is connected with it directly, by its having at first no branches and by its close connection with the olfactory nerve.

We believe that the ophthalmicus superficialis and ophthalmicus profundus always maintain these relations; that the profundus, which is clearly the nasal nerve of Mammalia, is a primitive and very constant nerve, and that it never shifts its position so as to lie dorsad of all the eye muscles, as supposed by Balfour.

The two divisions of the ophthalmicus superficialis, on the other hand, appear to be very variable indeed in different Vertebrates ; they attain their maximum development in the Elasmobranchs, probably on account of the great development and importance of the organs they supply—the mucous canals. In Mammalia the ophthalmic branch of the fifth is represented by the frontal and lachrymal nerves, while the seventh nerve has no ophthalmic branch.

It would hardly be profitable to discuss the various descriptions of these nerves by different writers; we will merely point out here that Schwalbe clearly distinguishes the three ophthalmic nerves. He calls, as we have done, the ophthalmic branches of the seventh and fifth nerves together the ramus ophthalmicus superficialis, distinguishing the component parts ns portio major (vu a) and portio minor (v a) respectively. He also employs the term ramus ophthalmicus profundus in the same senseas we have done. Balfour, wlio was the first to clearly recognise the double nature of the ophthalmicus superficialis, is in error in calling the lower portion of it (v a) the ophthalmicus profundus.

Concerning the other branches of the nerves in question, there can be little doubt that the hyoidean branch (VII c) of the seventh and the mandibular branch (v c) of the fifth are homologous nerves, supplying respectively the anterior walls of the hyoidean and mandibular arches; and there appear to be good reasons for viewing the nerve (III c) as the corresponding branch of the third. ² All the three nerves in question are either mainly or exclusively motor in function.

The seventh, like the hinder cranial nerves, forks over a visceral cleft—the spiracle. As shown in figs. 11 and 12, there are two branches of the seventh which run down in front of the spiracular cleft, viz. the buccal (VII d) and the mandibular (VII b), which latter divides almost at once into the palatine and spiracular nerves. Of these two a history of their development and a comparison of the branches of the seventh with those of the glossopharyngeal (ix, fig. 12), leave no possible room for doubt that the mandibular branch (VII b) is the homologue of the anterior branch (Ix b) of the glossopharyngeal. This latter nerve (Ix b, fig. 12) extends very far forwards in the hyoidean arch, being in this respect very closely imitated by the palatine nerve (VII, pa), so that we are disposed to regard the whole of the mandibular division (VII b) of the seventh, i. e. both palatine and spiracular nerves, as together equivalent to the anterior or hyoidean branch (Ix b) of the glossopharyngeal.

Balfour ¹ describes the mandibular branch of the seventh as being large in the embryo, so large in fact that he feels difficulty about identifying it with the adult spiracular nerve. His figures, ¹ however, show perfectly clearly that what he describes as the mandibular branch of the seventh is really the nerve we have shown to be the buccal. ²

The maxillary nerve (v b) is, from its time and mode of development, almost certainly to be regarded as the true anterior branch of the fifth corresponding to the mandibular branch of the seventh, although in the absence of a visceral cleft in this region this determination cannot be considered absolutely proved. Whether there is any equivalent branch of the third nerve is very doubtful; at any rate no such branch can be pointed out with certainty.

There now remains for consideration the buccal nerve, the determination of which, as a branch of the seventh, is one of the most striking points we have brought to light. Whether this remarkable nerve has any homologue among the branches of the fifth is a point our investigations have not yet enabled us to determine. It is perhaps worth while pointing out that there are many points of resemblance between this nerve and the ramus ophthalmicus profundus, points of sufficient importance to render a comparison between the two nerves at any rate a possible and suggestive one. In each case the proximal portion of the nerves in question connects together directly the ganglion of one segmental nerve with that of the nerve next in front, while the distal portion passes forward into the segment anterior to that in which the main branches of the nerve are contained. The early origin, the curiously straight course, and the absence of branches until close to their termination, are features common to the two nerves, and ones in which they stand in marked contrast to most other branches. The deep course of the profundus as contrasted with the very superficial one of the buccal nerve may perhaps be attributed to the great development of the eye: in front of the orbit the profundus is a superficial nerve, and like the buccal, is purely sensory in its distribution.

On the other hand, it must be noticed that, as already pointed out, the evidence is distinctly in favour of the distal portion of the profundus (beyond the ciliary ganglion) being a branch of the third rather than of the fifth nerve. Another point of distinction between them lies in the fact, that the former (the profundus’} is distributed to what is, morphologically, the dorsal surface, the buccal to the ventral.

In the present paper we have purposely refrained from attempting to determine the homologies between the nerves of Scyllium, and those of other vertebrates, preferring to wait until by the study of the development of other types we are enabled to bring forward positive evidence in support of our determinations.

In conclusion, we would express the hope that, by working out the development of the roots and branches of the cranial nerves in a very typical vertebrate, and following these roots and branches through their subsequent changes up to their adult condition, we have rendered more practicable than has hitherto been the case comparisons between the descriptions of embryos and of the corresponding adult animals, and have contributed something towards the establishment upon a firm basis of comparative neurology.