parts in gold preparations is well exhibited in figs. 411, 412. Applied to the branches of the ramification small granular nuclei (fig. 414, n) are seen at intervals; these nuclei of the arborisation are different from the clear nuclei of the bed (n), and also from the flattened nuclei of the sheath which lie immediately under the sarcolemma covering the end-plate, and which resemble the nuclei of the sheath of Schwann of the nerve. The sarcolemma over the situation of the nerve-ending is slightly raised above the general surface (fig. 410, a). It would appear that in mammals each

n, nerve-fibre; r, terminal ramification of axis-cylinder; m, clear substance surrounding the ramification (matrix); b, granular bed or sole of the end-organ.

Fig. 412.-CROSS-SECTION OF MUSCULAR FIBRE AND END-ORGAN OF LIZARD GOLD PREPARATION.

(Kühne.)

", terminal ramification of axis-cylinder, m, matrix; b, nucleus of bed; t, nucleus of telolemma.

Fig. 413.-MOTOR END-ORGAN OF HUMAN MUSCLE GOLD PREPARATION.

n, medullated nerve-fibre; r, terminal ramification of axis-cylinder.

(Kühne.)

muscular fibre has but one terminal structure, and receives consequently but onenerve-fibre. As, moreover, the fibres of a nerve undergo division, probably repeated division, before ending, it follows that one fibre in a nerve-root or -trunk may supply several muscular fibres. In reptiles the longer muscular fibres may have two or more nerve-endings.

The shape and extent of the terminal ramification of the axis-cylinder within the end-organ varies greatly, not only in different classes of animals, but also in animals belonging to the same class, and there is even some variation in individuals of the same species, as is evident from the various representations of the end-organs of the green lizard, which are here given. On the whole, it may be stated that the terminal ramification is most compact in mammals and reptiles and least so in

amphibia (fig. 415), where there is no continuous granular bed with clear nuclei imbedded in it, and the ramifications of the axis-cylinder are extended over a much larger proportionate area of the fibre than in reptiles, birds, and mammals, so that the termination of the nerve is far less localised. The branches of the axis

Fig. 414. TERMINATION OF A NERVE IN

A MUSCULAR FIBRE OF THE LIZARD

(Lacerta viridis). (Ranvier.) Highly magnified.

h, sheath of the nerve-fibre; b, bifurcation of the fibre; e, node; m, short segment beyond the node; r, terminal ramifications of the axis-cylinder; n, nuclei on the branches of the axiscylinder; n', nuclei in the granular substance of the end-plate. The granular substance lies in the intervals between the branches of the axis-cylinder; it is 4ot seen in this figure.

Fig. 415.-NERVE-ENDING IN MUSCLE OF FROG. (Kühne.)

a, one of the branches of the medullated fibre passing within the sarcolemma; b, b, granular pearshaped nuclei of the arborisation; c, c, nuclei of sheath; e, e, muscle-nuclei.

cylinder run for a short distance parallel with the axis of the fibre between the sarcolemma and muscular substance, terminating abruptly by rounded extremities. They have here and there slight enlargements, connected with which are seen, as in the end-plate of the lizard, granular pear-shaped nuclei (b), entirely different in appearance from the proper nuclei of the muscle (e). In other animals, e.g., in snakes, there is a tendency for the branches of the ramification to become dilated at their termination into bulbous enlargements, which in a well-stained preparation gives an appearance like that of a bunch of berries. Many other variations are met with, but in no case is there a departure from the general rule that the ultimate

termination is in the form of a ramification of the axis-cylinder on the surface of the fibre within the sarcolemma.1

The termination of motor nerves in special granular expansions within the sarcolemma was first noticed by Doyére in insect-muscles. The aborescent termination of the axis-cylinder was discovered in the frog by Kühne in 1862. In the same year the end-plates were recognised by Rouget in the lizard, and in 1863 by W. Krause in mammals. The last named observer was the first to describe the termination of the axis-cylinder as a ramified expansion imbedded in granular substance, but maintained that the whole structure lay outside the sarcolemma. In this opinion Krause is supported by Kölliker, but by far the majority of observers regard the whole end-organ as hypolemmal in position. Engelmann and Foettinger have been led from observations upon insect-muscles to the conclusion that the expansion of the nerve-fibre comes into actual continuity with the isotropous substance, i.e., with the sarcoplasm, of the muscular fibre. But the effect of section of a motor nerve in the living animal-the resulting degeneration extending no further into the muscular fibre than the end-plate itself—is a strong argument against the existence of any such anatomical continuity.

DEVELOPMENT OF THE NERVES.

"he embryonic development of the nerves has already been treated of in the par. of this work devoted to Embryology. It was there shown that all nerve-fibres and nerve-cells, whether belonging to the central nervous system or to the peripheral and sympathetic nerves, are originally derived from the neural or neuro-sensory epiblast, and that in the case of the afferent nerve-fibres, such as those of the posterior roots, the axis-cylinders grow from the cells of origin (neuroblasts of the spinal ganglia), both centripetally into the nerve-centre, and centrifugally towards the peripheral sensory parts, while in the case of the efferent nerve-fibres, those namely of the anterior roots, the axis-cylinders grow only centrifugally from their cells of origin, which here lie within the nerve-centre (neuroblasts of the nervecentres), and thence eventually pass to and unite with the muscular fibres. So far as is known, this is the only mode of development of nerve-fibres, viz., as out-growths from nerve-cells or neuroblasts, and they always, whether in the nerve-centres or in the nerve-trunks, at first appear as pale fibres, destitute both of primitive and of medullated sheath.

It appears somewhat doubtful whether the pale fibres which are thus first formed are single axis-cylinders or bundles of such. However this may be, they early become surrounded by cells from the adjacent mesoblast, which penetrate. also between them, and eventually produce the connective tissue of the nerve-sheath (epineurium, perineurium, endoneurium). In the nerve-centres very little connective tissue passes between the nerve-fibres, which are there supported by the spongioblasts (see Embryology, p. 58). These are cells which have a common (epiblastic) origin with the neuroblasts, although their function, according to His, is early differentiated. From the spongioblasts, neuroglia cells appear ultimately to be produced, and these, within the central nervous system, take on much the same supporting function which is elsewhere fulfilled mainly by connective tissue.

The medullary sheath does not make its appearance until a comparatively late period of embryonic life, and there is much doubt as to its mode of formation. It first appears as a thin layer of myelin, not unfrequently interrupted, which closely ensheaths the axis-cylinder, and is itself ensheathed by the nucleated neurolemma or sheath of Schwann which has been previously formed. Vignal refers the formation

1 For an account of the variations which are met with in different animals, illustrated by a large number of drawings, and for a discussion of many disputed points regarding the details of structure of these organs, the reader is referred to a paper by Prof. W. Kühne in the "Zeitschrift f. Biologie," Bd. 23, 1886.

2 Ramón y Cajal, however, states that many of the neuroblasts are directly derived from cells which are identical with the spongioblasts of His.

of the medullary sheath to the cells which compose the nucleated sheath of Schwann, and the same view has been taken by other observers; but it must nevertheless be regarded as possible that it is actually formed by the axis-cylinder, or by the protoplasm which forms the peripheral layer of the axis-cylinder in its embryonic condition, and for the following reasons, viz. (1) that in the central nervous system the medullated nerve-fibres never at any time possess this nucleated sheath; (2) that in regenerating nerve-fibres, a thin medullary sheath appears around the growing axiscylinders before these are surrounded by their special nucleated sheath. At the same time it is possible to suppose that the cells of Schwann's sheath may influence the deposition of myelin. The neurolemma or sheath of Schwann on the other hand appears certainly to be formed by cells which have applied themselves to and have become flattened out around the preformed axis-cylinder, but whether they are to be regarded as of mesoblastic origin (Vignal, Kölliker), or whether they have passed out from the nerve-centre along with the processes of the neuroblasts, and are therefore like the latter epiblastic in nature, is a question which requires further investigation. On the whole, although we have no certain information on the subject, it is probable that the medullary sheath has an origin in common with the axis-cylinder, but that the primitive sheath is different, and not improbably mesoblastic in origin.

The formation of the medullary sheath occurs, not simultaneously over the whole nervous system, but in regular order along definite tracts, and the knowledge of this in the hands of Flechsig, has proved an important means of tracing the course of certain strands of fibres in the nervous centres, as will be noticed when the subject of the continuity of the fibres in those centres is dealt with.

The fact that the nerve-segments or internodes of the peripheral nerves are considerably shorter in the young animal, points to the existence of an interstitial as well as a terminal growth of nerve-fibres. Besides such expansion of the internodes, Vignal has described another method of growth in length of nerve-fibres; mesoblast cells similar to those which originally produced the nucleated sheath, applying themselves to the axis-cylinder of the nodes, and determining first an increase in length. of the nodal axis-cylinder, and then a formation of myelin upon this, so that a short segment becomes intercalated at the node. These short segments soon grow so as to attain the length of the remaining segments of the nerve-fibre.

Degeneration and regeneration of nerves.-The divided ends of a nerve that has been cut across readily reunite by cicatricial tissue, but the cut ends of the fibres themselves do not thus unite. On the contrary, soon after the section, a process of degeneration begins in the peripheral or severed portion of the nerve. The nuclei become multiplied, and the protoplasm about them largely increased in amount, the segments taking on to some extent their embryonic condition. At the same time the medulla of the white fibres degenerates into a granular mass consisting of fatty molecules, and is then totally removed, and eventually the axial fibre also disappears (fig. 416, A, B, and C).

In regeneration the new fibres grow afresh from the axial fibres of the central end of the divided nerve-trunk (often more than one from each); and, penetrating into the peripheral end of the trunk, grow along this as the axis-cylinders of the new nerves, becoming after a time surrounded with medullary substance (fig. 416, D).

To this brief summary the following details may be added :-In warm-blooded animals the first changes in the peripheral part of the nerve are seen twenty-four hours after the section. The nuclei underneath the primitive sheath are everywhere found hypertrophied, the primitive sheath is distinctly visible, and protoplasm is found to have accumulated at the expense of the medullary sheath, both in the immediate neighbourhood of the nuclei, at the nodes, and also at other points in

the fibre, which correspond, according to Ranvier, with the intervals between the medullary segments. Fifty hours after the section in the rabbit (but not till four days in the dog) the protoplasmic aggregations are found here and there altogether to interrupt the continuity of the medullary sheath, and they contain numerous fatty granules, and sometimes droplets of myelin (fig. 416, A). About the fourth day the nuclei are seen to be multiplied, but not to any great extent (C); and

Fig. 416.-DEGENERATION AND REGENERATION OF NERVE-FIBRES IN THE RABBIT.

(Ranvier.)

A, part of a nerve-fibre in which degeneration is commencing in consequence of section (50 hours previously) of the trunk of the nerve higher up; my, medullary sheath becoming broken up into drops of myelin; p, granular protoplasmic substance which is replacing the myelin; n, nucleus, not yet multiplied; g, primitive sheath. B, another nerve-fibre in which degeneration is proceeding, the nerve having been cut four days previously. This specimen is differently prepared from the others, so as to exhibit the axis-cylinder (cy) also partly broken up into portions of different length, enclosed in the myelin, my. C, more advanced stage of degeneration, the medullary sheath having in great measure disappeared, while several nuclei (n", n") have been formed by division of the single nucleus of the internode. D, commencing regeneration of a nerve-fibre. Several small nerve-fibres (t', t'), have sprouted out from the enlarged cut end (b) of the nerve-fibre (t); a, an axis-cylinder, which has not yet acquired a medullary sheath; s, s', primitive sheath.

the whole of the myelin after four or five days is broken up into drops, some larger, some smaller. The axis-cylinder is also found to be interrupted at numerous places, and remains only in the shape of short fibres, often curled round at their broken ends, enclosed in the large drops of myelin (B). Eventually these portions also may disappear. The myelin at length becomes almost entirely removed, partly through the agency of leucocytes or phagocytes, until nothing remains of it except a few isolated drops, which escape absorption, and all that then remains of the original fibre is the primitive sheath, which is occupied by a protoplasmic mass containing anincreased number of nuclei. During the disappearance of the myelin