This common sheath has received the name of epineurium (Key and Retzius); it was formerly termed the "cellular sheath."

The special sheath of a funiculus, termed the perineurium 1 (fig. 379, per, 380, P.), is also formed of connective tissue, but is far more distinctly of a lamellar nature, and indeed may be stripped off in the form of a tube from the little bundle of nerve-fibres of which the funiculus consists. The perineurium is not formed of a single lamella but of several, which are separated from one another by inter

Fig. 380.-PART OF A SECTION OF ONE OF THE FUNICULI OF THE SCIATIC NERVE OF MAN.
MAGNIFIED (after Key and Retzius).

P, perineurium, consisting of a number of closely arranged lamellæ. En, processes from the perineurium, passing into the interior of the funiculus, and becoming continuous with the endoneurium, or delicate connective tissue between the nerve-fibres. The connective tissue fibrils of the endoneurium are seen cut across as fine points, often appearing to ensheath the nerve-fibres with a circle of minute dots (fibril-sheath of Key and Retzius). Numerous nuclei of connective tissue cells are imbedded in the endoneurium; v, section of a blood-vessel.

lamellar clefts moistened with lymph. The separation is not everywhere complete, for here and there bundles pass across, connecting the several lamellæ. Moreover, the outermost lamella is joined by connective tissue bundles and laminæ of the

Fig. 381.-A PORTION OF A SMALL NERVE-TRUNK FROM THE THORAX

OF A MOUSE, TREATED WITH NITRATE OF SILVER (Ranvier).
MAGNIFIED.

Cross markings are seen at the nodes, and the layer of flattened epithelioid cells which covers the surface is brought into view by the silver deposit.

epineurium, and the innermost gives off flattened prolongations (fig. 379, end.), to form imperfect septa between the groups of nerve-fibres within the funiculus.

Although the lamellæ of the perineurium are very thin, each is formed of at least three strata. Thus the main substance of the lamella is composed of a connective tissue, in which both white fibres and elastic elements are found, the white fibres having for the most part a transverse disposition. The elastic elements lie in greater abundance nearer the surfaces

of the lamella, and often occur in the form of patches or incomplete membranes of elastic substance (fig. 282, p. 240), as well as in the form of a fine network of fibres. On both its surfaces each lamella is entirely covered with a layer of delicate flattened endothelial cells, which thus serve also to bound the clefts between the lamellæ. The outlines of the cells are brought into view by the silver treatment (fig. 381).

The funiculi of a nerve, although not all of one size, are all sufficiently large to

1 Formerly known as the neurilemma.

be readily seen with the naked eye, and easily dissected out from each other. In a nerve so dissected into its component funiculi, it is seen that these do not run along the nerve as parallel insulated cords, but join together obliquely at short distances as they proceed in their course, the cords resulting from such union dividing in their further progress to form junctions again with collateral cords; so that in fact the funiculi composing a single nervous trunk have an arrangement with respect to each other similar to that which is found to hold in a plexus formed by the branches of different nerves. It must be distinctly understood, however, that in these communications the medullated nerve-fibres do not join together or coalesce. They pass off from one nervous cord to enter another, with whose fibres they become intermixed, and part of them thus intermixed may again pass off to a third funiculus, or go through a series of funiculi and undergo still further intermixture; but throughout all these successive associations the fibres remain individually distinct, like the threads in a rope.

The nerve-fibres are separated from one another, and supported within the funiculus by de..cate connective tissue, the fibrils of which run for the most part longitudinally, appearing in section as fine points (fig. 380). This tissue has been distinguished as the endoneurium by Key and Retzius. It is continuous with the septa which pass in as above mentioned from the innermost lamella of the perineurium, and it serves to support also the capillary blood-vessels which are distributed to the nerve.

Lying alongside each other, the fibres of a funiculus form a little skein or bundle, which runs in a waving or serpentine manner within its sheath; and the alternate lights and shadows caused by the successive bendings being seen through the sheath, give rise to the appearance of alternate light and dark cross stripes on the funiculi, or even on larger cords consisting of several funiculi. On stretching the nerve, the fibres are straightened and the striped appearance is lost. Both the perineurium and endoneurium accompany the nerves in all their divisions, in some cases as far as their peripheral terminations. In the finest branches the perineurium generally becomes reduced to a single connective tissue lamella, covered on both surfaces by endothelial cells. In this condition it is known as the sheath of Henle.

Both the cerebro-spinal and the sympathetic nerve-trunks are constructed in the manner above described, but the fibres of the cerebro-spinal nerves are chiefly of the white or medullated kind, and contain for the most part fibres of large size, while in nerves belonging to the sympathetic system non-medullated fibres or medullated fibres of very small size greatly preponderate. But very few nerves are composed exclusively of one or the other kind of fibre.

Vessels and Lymphatics.-The blood-vessels of a nerve after dividing into small branches in the epineurium and giving offsets to the groups of fat-vesicles which are there met with, pierce the layers of the perineurium obliquely, being supported by the connective tissue bundles which unite the lamellæ, and conducted into the interior of the funiculus along the septa before mentioned. Here they break up into fine capillaries which for the most part run parallel with the fibres, but are connected at intervals by short transverse branches, thus forming a network with long narrow meshes. Some of the capillaries may be observed to form loops. Lymphatic vessels are found in the epineurium, but within the funiculi there are no distinct vessels for the conveyance of lymph. It is found, however, that coloured fluid which is injected by means of a fine cannula into the interior of a funiculus finds its way into the lymphatics of the sheath after passing through the clefts between the lamellæ of the perineurium, so that undoubtedly a connection exists between these perineural clefts and the lymphatic system.

Course of the nerve-fibres in the nerve-trunks.-Neither in their course along the nervous cords, nor in the white part of the nerve-centres, do the

medullated fibres anastomose together, nor are they observed except in rare instances to divide into branches until they approach their termination. But the nerve-trunks themselves continually ramify, and the branches of different nerves not unfrequently join with one another. The branches are of course formed by collections of nerve-fibres, and it follows therefore that when two branches of nerves join, fibres pass from the one nerve-trunk to become associated with the other in their further progress, or the communication may be reciprocal, so that after the junction each nerve-trunk contains fibres derived from two originally distinct sources. In other cases the branches of a nerve, or branches derived from two or from several different nerves, are connected in a more complicated manner, and form what is termed a plexus. In plexuses-of which the one named "brachial" or "axillary," formed by the great nerves of the arm, and the "lumbar" and "sacral," formed by those of the lower limb and pelvis, are appropriate examples-the nerves or their branches join and divide again and again, interchanging and intermixing their fibres so thoroughly that, by the time a branch leaves the plexus it may contain fibres from several or even from all the nerves entering the plexus. Still, as in the more simple communications already spoken of, the fibres, so far as is known, remain individually distinct throughout.

In some instances of nervous conjunctions certain collections of fibres, after passing from one nerve to another, take a retrograde course in that second nerve, and, in place of being distributed peripherally with its branches, turn back to its root towards the cerebro-spinal centre. Instances of this occur, according to Volkmann, in the connection between the second and third cervical nerves of the cat, in that of the fourth cranial nerve with the first branch of the fifth, and of the cervical nerves with the spinal accessory and the descending branch of the hypoglossal.

Origins or roots of the nerves.-The cerebro-spinal nerves, as already said, are connected by one extremity to the brain or to the spinal cord, and this central extremity of a nerve is, in the language of anatomy, named its origin or root. In

Fig. 382.-RooTS OF ONE OF THE SPINAL

NERVES ISSUING FROM THE SPINAL

CORD. (Allen Thomson.)

A, from before; B, from the side; C, from above; D, the roots separated; 5, 5, anterior root; 6, 6, posterior root with ganglion, 6'. The full description of this figure will be found in the chapter on the cerebro-spinal nervous axis.

some cases the root is single, that is, the funiculi or fibres by which the nerve arises, are all attached

at one spot or along one line or tract; in other nerves, on the contrary, they form two or more separate collections, which arise apart from each other and are connected with different parts of the nervous centre, and such nerves are accordingly said to have two or more origins or roots. In the latter case, moreover, the

different roots of a nerve may differ not only in their anatomical characters and connections, but also in function, as is well exemplified in the spinal nerves, each of which arises by two roots, an anterior and a posterior-the former containing the efferent fibres of the nerve, the latter the afferent.

The fibres of a nerve may be traced to some depth in the substance of the brain or spinal cord, and hence the term "apparent or superficial origin" has been employed to denote the place where the root of a nerve is attached to the surface, in order to distinguish it from the "real or deep origin" which is beneath the surface and concealed from view. If the deep origin be traced out, it will usually be found that the nerve-fibres arise from portions of the grey substance of the nerve-centre : such a portion of grey substance is termed the "nucleus of origin" of the nerve.

In the case of the efferent nerves it would appear that the individual nerve-fibres originate as prolongations of the nerve-cells in the grey substance. In fact, as the researches of His have shown, these nerve-fibres have grown out from the nerve-cells (neuroblasts) within the embryonic nerve-centre (see Embryology, Development of

P

B

Fig. 383.-TRANSVERSE SECTION OF THE SPINAL CORD OF A CHICK ON THE 9TH DAY OF INCUBATION, PREPARED BY GOLGI'S METHOD (Ramón y Cajal).

A. Axis-cylinders of anterior root-fibres issuing from large cells of the anterior cornu, C.

B. Posterior root-fibres passing from the bi-polar cells of the spinal ganglion, E, into the posterior column of the spinal cord (D), where they bifurcate (d) and become longitudinal.

e, f, g, collateral branches from these fibres, passing into the grey matter.

Nerves). In the case of afferent nerves it seems to be clear that these in most if not all cases have grown into the nerve-centre from the cells of the ganglia, or from nervecells in the peripheral organs of special sense. Having entered the nerve-centre (fig. 383, D) the afferent fibres appear usually to bifurcate-at least this is the case with the fibres of the posterior spinal roots-and the two resulting branches become longitudinal, sending off lateral ramuscles (e, f, g) into the grey matter, in which they appear to break up into fine ramifications, without being directly continuous with nerve-cells of the grey substance.

In the nerve-roots the fibres are bound up together, as they pass towards the foramina of exit from the cranio-vertebral canal, by a stout external sheath continuous with the pia mater and receiving an accession from the arachnoid and dura mater as the roots pass through those membranes. This sheath sends in strong septa which branch and unite with one another in the nerve-root, and thus divide it up irregularly into bundles of fibres, which have not the same cylindrical character, with special lamellated sheaths of perineurium, as the funiculi of the peripheral nerves,

but are supported by a kind of framework formed by the dividing septa above mentioned. Only after passing the ganglia do the nerves acquire a true funiculated structure such as has been described for the peripheral nerves. The optic nerve has throughout its whole course the structure of a nerve-root.

The cerebro-spinal nerve-trunks and their branches always present a brilliant whitish aspect, whereas the sympathetic nerves vary in appearance, some being whiter, others grey or reddish in colour. The more grey-looking branches or bundles consist of a large number of the pale fibres mixed with a few of the medullated kind; the whiter cords, on the other hand, contain a proportionally large amount of

Fig. 384.-SECTIONS ACROSS PARTS OF THE ROOTS OF VARIOUS NERVES OF THE DOG, TO SHOW THE VARIATIONS IN SIZE OF THEIR CONSTITUENT FIBRES (Gaskell).

The nerves were stained with osmic acid, and the sections are all drawn to one scale.

A, from one of the upper roots of the spinal accessory.

B, a rootlet of the hypoglossal.

C, from the 1st cervical anterior root.

D, from the 2nd thoracic anterior root.

fine medullated fibres, and fewer of the grey; and in some parts of the nerve grey fasciculi and white fasciculi, respectively constituted as above described, run alongside of each other in the same cords for a considerable space without mixing. This arrangement may be seen in some of the branches of communication with the spinal nerves, in the trunk or cord which connects together the principal chain of sympathetic ganglia, and in the primary branches proceeding thence to the viscera. In the last-mentioned case the different fasciculi get more mixed as they advance,