efferent" cells, and the latter as "sensory or afferent," but it is doubtful if this physiological distinction will hold good. In the spinal cord the cells of the anterior horn, those of Clarke's column, and many of the cells of the lateral cell-column and of the posterior horn belong to the first category, while many of the solitary cells, the small cells of the substance of Rolando, and some of those of the lateral cell-column, belong to the latter.

Fig. 15.-CENTRAL CANAL SHOWING ITS EPITHELIUM. (E. A. S.)

A, from a child of six. Magnified 150 diameters. A', some of the ciliated cells, highly magnified. B, section of the isthmus from the lumbar cord of an adult (æt. 33), showing the central canal in the middle surrounded by the substantia gelatinosa centralis. f.a., anterior median fissure; p.m.c., posterior white column; a. c., anterior white commissure. Magnified 30 diameters.

Commissures.-The anterior commissure (fig. 15, B, a.c.) consists of medullated nerve-fibres which pass on each side, some into the anterior white column, others into the anterior horn of grey matter. Their course is not strictly transverse, many fibres which enter the ventral part of the commissure at one side leave it at the dorsal portion on the other side. There is thus an oblique decussation at the middle line (fig. 15). This decussation is most distinctly seen in the comparatively short and wide commissure of the lumbar region, and in the upper part of the cervical region. In the latter situation it appears as a continuation of the decussation of the pyramids of the medulla oblongata, to be afterwards described. In addition to the transverse there are a few longitudinal bundles of fibres in the region of this commissure. The fibres are often somewhat displaced by vessels which pass into the grey substance from the anterior fissure.

The posterior commissure also contains medullated fibres running transversely or with a slight obliquity, but there is a large amount of neuroglia between them, and this gives the commissure a grey aspect. In this commissure is contained the central canal of the spinal cord surrounded by an area free from medullated nerve-fibres, and occupied by a substance which, like the superficial neuroglia and the substance of Rolando, has in the hardened condition a reticulated structure and in the fresh state a gelatinous aspect. It is termed substantia gelatinosa

centralis. In this gelatinous substance many of the fibres of the posterior commissure appear to end (see fig. 15, B). There is rather more of the posterior commissure behind the central canal than in front. The fibres of the posterior part curve backwards on reaching the crescents of grey matter and pass towards the bundles of the posterior roots, whereas the rest diverge at various angles into the crescent.

The central canal is lined with a layer of ciliated epithelium (fig. 15, A, A'). Each cell is provided with a bunch of cilia on the side which is turned towards the

Fig. 16.-DIAGRAM TO ILLUSTRATE THE RELATIONS OF THE ENTERING NERVE-ROOTS, AND THE FIBRES OF THE WHITE COLUMNS OF THE SPINAL CORD TO THE NERVE-CELLS IN THE GREY MATTER. (E. A.S.)

lumen of the canal: the other end of the cell is prolonged into the reticular substance just mentioned, and there becomes lost to view.

In the adult human subject the lumen of the central canal is not unfrequently obliterated, being filled up by detached cells.

ORIGIN OF THE SPINAL NERVES.

The roots of the spinal nerves are attached along the sides of the cord, opposite to the corresponding horns of the grey matter; the posterior roots entering along a straight line at the postero-lateral groove, and the anterior roots being scattered somewhat irregularly upon the surface (fig. 7, B).

The anterior roots are seen in a transverse section to pass through the white substance and to enter the grey cornu in several bundles, which have a slight upward inclination, so as to be often cut obliquely if the section be exactly in a transverse plane. Some of the fibres on reaching the grey matter are directly connected with the axis-cylinder processes of the large nerve-cells of the cornu. Others pass by the cells without, so far as can be seen, entering immediately into relation with them.

As soon as the bundles enter the grey matter, their fibres diverge from one another, some passing mesially, others laterally, and others straight backwards (fig. 16). Of those which pass mesially some are continued into the axis-cylinder processes of

the mesial group of cells of the anterior cornu, and others to the other side of the cord through the anterior commissure, there perhaps to become connected with the corresponding cells. The outwardly directed fibres of the root are connected with the cells which form the lateral groups of the anterior horn, and partly with some of the cells of the intermediate cell-column. The middle fibres are partly connected to the cells of the anterior horn, and partly pass on to the posterior horn, where they are probably continued into some of the cells there present.

In the cervical region of the cord, from the sixth nerve upwards, the root-fibres of the spinal part of the spinal accessory nerve may also be seen passing out of the anterior horn through the lateral column (fig. 31, p. 40). They arise from a group of somewhat large cells, which in the uppermost region of the cord lies in the antero-lateral part of the horn, close to

A

B

-d

Fig. 17.-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 horn, C.

B, Posterior root-fibres passing from the bipolar cells of the spinal ganglion, into the posterior column of the spinal cord, D, where they bifurcate, d, and become longitudinal; e, f, g, collaterals from these fibres, passing into the grey matter.

its ventral edge, but in the middle and lower cervical regions tends gradually to occupy a position nearer the base of the horn (see p. 16). This group of cells is the direct continuation downwards of the accessory nucleus which will be seen in the lower part of the medulla oblongata.

The posterior roots at their entrance into the cord are seen to separate into two sets. Of these the lateral, which contains most of the smallest fibres of the root, enters opposite the tip of the posterior horn and joins a bundle of fine longitudinal fibres which lies close to the periphery of the cord in this situation. This bundle is known as the marginal bundle or column of Lissauer. Its fibres, which are evidently different from most of those which enter with the posterior root, for they acquire their medullary sheath considerably later (Bechterew), appear, as they pass vertically, gradually to lose themselves in the adjacent gelatinous substance. The mesial set containing the larger and some smaller fibres of the posterior root passes into the

postero-lateral white column, with the longitudi-
nal fibres of which its fibres become continuous;
from this column many of them sweep with
bold curves into the adjacent grey matter. At
their entrance into the cord their direction like
that of the bundles of fibres of the anterior roots
is in most regions of the cord obliquely upwards.

Fig. 18.-LONGITUDINAL SECTION OF THE POSTERIOR COLUMN
OF THE SPINAL CORD OF A CHICK ON THE 8TH DAY OF
INCUBATION (Ramón y Cajal).

This figure shows the course of 5 entering fibres of the posterior root and some of the longitudinal fibres of the posterior column.

A, A, fibres of the posterior root; B, bifurcation of one of them in form of a Y; C, D, origin of collateral branches; E, fibres of Goll's tract, also giving off collaterals.

Collateral fibres of the posterior roots and of the white columns.-It has been definitely shown by Ramón y Cajal, whose researches into this subject have been confirmed. by Kölliker and v. Gehuchten, that the fibres of the posterior roots as they enter the cord join by a Y- or T-shaped junction with longitudinal fibres of the posterior column; in other words that they bifurcate as they enter the cord into two principal branches which run upwards and downwards in the posterior white column or in the adjacent part of the posterior cornu. It has also been proved by the same observers

B

Fig. 19. TRANSVERSE SECTION OF THE CORD OF A CHICK ON THE 9TH DAY (Ramón y Cajal).

This figure shows collaterals passing into the grey matter from all the white columns of the cord. They are seen to form an especially close plexus in the posterior horn, just within the substance of Rolando, and in the posterior commissure.

c, central canal; v, anterior and d, posterior surface of the cord; p, p, posterior roots.

that both from the root-fibre before its bifurcation and from its ascending and descending stems there are given off at tolerably frequent intervals "collateral" branches which are directed inwards towards the grey matter; into which they penetrate for a variable distance and within which they ultimately end by breaking up into a ramification of nerve-fibrils which may frequently be seen to have a close relationship to cells of the grey matter. In this way collateral fibres are given off from the posterior root-fibres and from their longitudinal extensions, some of which penetrate as far as the anterior horn of the same side, others pass through the posterior commissure into the grey matter of the opposite side, others towards the lateral horn and intermediate cell-column, whilst others do not extend beyond the posterior horn, passing towards Clarke's column or the solitary cells of the crescent, while many end in or near the substance of Rolando. These collateral fibres can be seen in sections of the embryonic cord which have been prepared by Golgi's method, to pass into the grey matter in large numbers from the posterior columns, most of the fibres of which are in fact prolongations of posterior roots which have entered the cord. But they are by no means confined to the fibres of the posterior columns, for from all the longitudinal white columns of the cord the same convergence of collateral fibres into the grey matter can be seen (see figs. 16 to 20).

The passage of collateral fibres into the grey matter from the white columns of the cord was first noticed by Golgi, who stated that the fibres of the anterior roots also give off in their passage towards the surface of the cord fine lateral ramuscles which become lost in the adjacent nervous matter. Other observers have for the most part failed to detect these collaterals of the anterior root-fibres.

COURSE OF NERVE-FIBRES WITHIN THE SPINAL CORD: CONDUCTING TRACTS.

It is impossible mechanically to unravel the tracts of nerve-fibres in their passage along the spinal cord, and it is exceedingly difficult to trace the same fibre or fibres for any distance in microscopical sections of the organ. But the task of following out the course of certain sets of fibres has been much facilitated of late years by the application to the subject of certain special developmental and pathological methods of observation. Thus it is found that if the development of the spinal cord is carefully observed, the medullary substance of the nerve-fibres is formed later along certain tracts of the white columns than in the rest of the white matter, appearing first in those tracts which are the immediate prolongations of peripheral nerves, and being longest delayed in those which are connected with the higher centres in the brain, so that in transverse sections of the cord these non-medullated tracts are easily distinguishable by their more transparent grey appearance and by their different behaviour with staining fluids (Flechsig).

The following list (from Kahler) gives the order of formation of the myelin-sheath in the several tracts of the cord:-1. Fibres in anterior column derived from anterior roots. 2. Postero-lateral column (its posterior part somewhat later than its anterior part). 3. The lateral column near grey matter (the anterior part first, then the narrow posterior part between pyramidal tract and posterior horn). 4. Postero-mesial column. 5. Direct cerebellar tract. 6. Antero-lateral ascending tract. 7. Pyramidal tracts (shortly before birth).

Another method by which similar results are arrived at consists in tracing the course which the degeneration of the fibres pursues in consequence of a lesion either in the encephalon, or in the spinal cord itself, or even in the peripheral nerves; the lesions being produced by accidental injury, by pathological changes, or experimentally in animals. The degenerations which follow are either the result of the Wallerian law that separation of a nerve-fibre from the nerve-cell with which it is connected and from which it has grown out is followed by degenerative changes in