sixth, fourth and third nerve-roots, fibres also pass from it to the auditory nucleus, and others to the cerebellum. After giving off fibres to the root of the third nerve the posterior longitudinal bundle is continued into the posterior commissure (see p. 109), and partly upwards into the subthalamic region (? to the substantia interansalis, see p. 112). The posterior longitudinal bundles come close together at the raphé, and fibres pass from one to the other. These are probably the fibres which effect a crossed connection between the abducens nucleus of the one side and the fibres passing to the internal rectus by the third nerve of the other side (see pp. 63 and 99). Perhaps, also, a connection is established through the posterior longitudinal bundle between the facial and the oculomotor nucleus (Mendel). The posterior longitudinal bundle is very small in the mole (Forel), large in reptiles and amphibia (Spitzka).

Brachium conjunctivum; superior cerebellar peduncle.-Another tract of longitudinal and decussating fibres is derived from the superior peduncle of the cerebellum, which we have already traced as it passes forwards over the superior end of the fourth ventricle. Reaching the sides of the aqueduct as a well-marked bundle, of semilunar shape in section (fig. 72, s.c.p), it gradually takes a more ventral position as it is traced upwards in the mesencephalon, and its fibres soon begin to pass across the raphe, decussating with those of the other side (fig. 75 A, and fig. 76, s.c.p.), the decussation extending as far upwards as the superior pair of corpora quadrigemina. Having thus crossed to the opposite side the tract in question pursues its course longitudinally upwards, appearing at first as a white bundle, but higher up enclosing in its passage a tract of grey matter with numerous large pigmented cells, known as the nucleus of the tegmentum or red nucleus (fig. 75, B, and fig. 78, r.n.), and probably receiving an accession of fibres from these cells. Above, the tract passes into the ventral part of the optic thalamus. Some fibres. do not cross, but enter the red nucleus of the same side.

Between the two red nuclei a small white bundle (Meynert's bundle) passes backwards on either side of the raphé from the ganglion of the habenula near the roof of the third ventricle to a small mass of grey matter which lies between the crura (ganglion interpedunculare, fig. 76, g.i.p.).

Tract of the fillet.-The fillet, which, in sections across the upper part of the pons, forms a considerable flattened bundle of longitudinal fibres at the ventral border of the formatio reticularis, is traceable upwards into the ventral part of the tegmentum. Soon, however, the large laterally situated part of this tract is seen to pass obliquely outwards and emerge at the side of the crus cerebri, curving obliquely over the outer side of the prolongation of the cerebellar peduncle (fig. 75, A, ƒ), and tending for the most part towards the inferior corpora quadrigemina. It is seen on the surface as a band of obliquely curved fibres, occupying a triangular area at the side of the tegmentum (fig. 76, f'), and it was to this band that the name of fillet was originally applied by Reil. It is now known as the lower or lateral fillet. It is reinforced by fibres from the superior medullary velum which also curve round the superior cerebellar peduncle, and which are probably derived from the anterolateral ascending tract of the cord, for they undergo degeneration after section of the cord (see below). The fillet is covered externally by a thin layer of grey matter containing nerve-cells.

But all the fibres of the tract of the fillet do not take the course above indicated. Those nearest the middle line (mesial fillet) separate themselves from the rest, and pass at the lower part of the mesencephalon into the crusta (see p. 101), where they form a mesial bundle (Wernicke), which is traceable up into the subthalamic region, where it joins the ansa lenticularis. Those next in order (middle portion) are for the most part, according to Forel, continued upwards in the formatio reticularis of the tegmentum, but many of the fibres become lost amongst its cells, and are not traceable further as a distinct tract. According to Edinger, they have a cell

station in a special group of nerve-cells (upper nucleus of the fillet) at the level of the inferior corpora quadrigemina. Some of the lateral fibres of this middle portion, however, pass to the upper corpora quadrigemina (upper fillet), and even extend beyond the superior quadrigeminal region to the subthalamic region (Flechsig), eventually reaching the parieto-occipital part of the cerebral hemispheres by the posterior part of the internal capsule (v. Gudden). Both the upper and lower fillet receive fibres from the antero-lateral columns of the spinal cord, for after hemisection of the cord degenerated fibres are seen on the same side, both in the lateral fillet of the mesencephalon, where they are mingled with the mass of undegenerated fibres which have been derived from the nucleus gracilis and nucleus cuneatus (see p. 53 and below), and in the bundle of the upper fillet, which enters the superior corpora quadrigemina; a few degenerated fibres are also seen in corresponding positions on the opposite side. Traced downwards the fibres of the lower fillet pass, as we have seen, a few by means of the trapezium towards the auditory nucleus of the opposite side, a few to the antero-lateral column of the medulla oblongata and cord (antero-lateral ascending tract), but most are traceable to the anterior column of the medulla oblongata dorsal to the pyramids, and passing across the raphe, proceed as internal arched fibres to the nuclei of the posterior columns, from the cells of which they in all probability arise.

To sum up: The fillet as a whole is composed of ascending fibres, most of which are derived from the nuclei of the opposite posterior columns of the medulla oblongata, but a few from other sources. The fibres of the antero-lateral ascending tract which enter the fillet may, according to Edinger, be derived from cells of the opposite posterior horn of the spinal cord; if so, all the fillet-fibres may be regarded as having crossed over from the opposite side, some in the cord, others at the superior or sensory decussation in the medulla oblongata. The fact that they are thus derived is however by no means certain. Superiorly the fillet may be described as being prolonged upwards above the pons in three parts, of which one-the lower fillet-goes to the inferior of the corpora quadrigemina of the same side, a secondupper fillet to the superior corpora quadrigemina and occipital region of the cerebral hemisphere, and a third, mesial fillet, to the base of the cerebrum through the crusta.

DORSAL PART OF THE MID-BRAIN: CORPORA QUADRIGEMINA.

As before stated the Sylvian aqueduct is covered on its dorsal aspect by the quadrigeminal lamina, bearing the bodies of the same name. The median part of the lamina is marked by a comparatively wide groove (sulcus longitudinalis, s. sagittalis), shallower inferiorly, which serves to separate the corpora quadrigemina of opposite sides (fig. 77). This grooved surface, which is raised above the level of the upper medullary velum, is connected with the velum by a small median strand of longitudinal fibres termed the frænulum veli (fig. 77, fr). In front of the upper (anterior) pair of corpora quadrigemina the groove is interrupted by a transverse white prominence-the posterior commissure; but both this and the upper end of the median groove are in the natural condition concealed by the pineal body (p), which projects backwards and downwards from the posterior wall of the third ventricle and rests between the upper pair of quadrigeminal bodies. A well-marked narrow transverse groove (sulcus transversus) which commences a short distance from the middle line, and is curved round the lower border of the upper tubercle, separates this from the lower tubercle of the same side.

The corpora quadrigemina are two pairs of rounded eminences which are mainly composed of grey matter, although covered externally by and containing in their interior many white fibres. The upper or anterior tubercles (fig. 77, c.q.s.)

are broader and longer and also darker in colour, but slightly less prominent than the lower or posterior (c.q.i)'. Laterally the corpora quadrigemina are not bounded by a distinct groove, but each appears to be prolonged obliquely upwards and forwards into a prominent white tract, known as the brachium of the corresponding tubercle. The lower (posterior) brachium (fig. 77, br.i.) loses itself underneath an oval prominence which is seen at the side of the upper end of the crus cerebri, and is termed the inner geniculate body (fig. 77, c.g.i.; fig. 78, c.g.m.). The upper (anterior) brachium passes between the same geniculate body, and the prominent posterior extremity of the optic thalamus into the external geniculate body (fig. 78, c.g.l.) and the optic tract, of which it may be regarded as the main prolongation (see also fig. 86, P, 117). The continuity is much better seen externally in some animals than in man. The connection of the superior quadrigeminal bodies with the optic tract and the sense of sight is far more intimate than that of the inferior. For if in a young animal the eye on the one side be extirpated, the operation is found to be followed

Fig. 77.-VIEW OF THE MEDULLA, PONS, AND

MESENCEPHALON FROM THE RIGHT SIDE AND

BEHIND. (E. A. S.)

The cerebellum, the inferior medullary velum, and the right half of the superior medullary velum, have been cut away, so as to display the fourth ventricle.

obl

c.q.8, c.q.i, superior and inferior quadrigeminal bodies of the left side; the pineal gland, p, is seen projecting backwards between The superior bodies, and the frænulum. fr, passes up from the superior medullary velum, 8.m.v, to the interval between the posterior quadrigeminal bodies; th, right thalamus opticus; br.i, brachium of the inferior quadrigeminal body passing underneath the inner geniculate body, c.g.i.; f, superficial stratum of fibres of the fillet, covering the tegmentum of the crus cerebri; c, crusta of the crus cerebri, separated from the tegmentum by the lateral groove, l.g. ; P, upper part of the pons; III, IV, &c., the r corresponding cranial nerves. The rest of this figure will be found described at p. 43.

after some time by atrophy of the superior quadrigeminal body and of its brachium, whereas the inferior quadrigeminal body and brachium is unaltered (Gudden). Moreover, in the mole the inferior quadrigeminal body is well developed, whereas the superior is atrophied.

nd

p.n

TV

ab

XI

The lower or posterior quadrigeminal bodies are composed almost entirely of grey substance (the so-called nucleus of these bodies (fig. 76, c.q.p.) which is separated by a thin layer of the fillet from the central grey matter of the aqueduct and contains numerous small and some larger nerve-cells. The connections of these have not yet been satisfactorily worked out, but they are believed to be closely related to the termination of the fibres of the lower fillet. The nuclei are united across the middle line by a commissural portion of grey matter which is bounded superficially and deeply by transverse white fibres derived from the fillet.

1 The term nates is often applied to the superior or anterior corpora quadrigemina, and testes to the inferior or posterior. These names were used by Vesalius, but are somewhat misleading, and have fallen

into disuse.

The superficial fibres are continuous laterally and above with the fibres of the brachium of the inferior quadrigeminal body, and laterally and below with those of the lower fillet. On the other hand, if the fibres of the lower fillet are traced downwards, they are partly found as we have already seen (p. 104) to be continuous with fibres of the trapezium; which again is connected with the principal nucleus of the auditory nerve (cochlear division) of the opposite side (Flechsig). This would appear to indicate a close connection between these posterior or inferior quadrigeminal bodies and the auditory nerve, and in conformity with this view it is noticeable that it is only those animals (mammals) which have a well-developed spirally wound cochlea that show the inferior corpora quadrigemina as distinct prominences. In nearly all vertebrates below mammals there are merely corpora bigemina, and these seem mainly to represent the anterior pair of the quadrigemina of mammals. As

Fig. 78.-SECTION ACROSS THE MID-BRAIN, THROUGH THE SUPERIOR CORPORA QUADRIGEMINA. Magnified about 3 diameters. (E. A. S.) From a Photograph.

Sy., aqueductus Sylvii; c.p., commissura posterior; gl.pi., glandula pinealis; c.q.a., grey matter of one of the superior corpora quadrigemina; c.g.m., corpus geniculatum mesiale; c.g.l., corpus geniculatum laterale; tr. opt., tractus opticus; p.p., pes pedunculi; p.l.b., posterior longitudinal bundle; fi., upper fillet; r.n., red nucleus; n. III, nucleus of third nerve; III, issuing fibres of third nerve; l.p.p., locus perforatus posticus.

already mentioned, this anterior pair is chiefly connected with the optic nerves and therefore with the visual sense.

The upper or anterior quadrigeminal bodies of man have been carefully investigated by Tartuferi. Most externally or uppermost is a thin layer of superficial neuroglia, containing no nerve-cells or fibres, but only fine stellate glia-cells and the ends of glia-fibres which radiate from the central canal (aqueduct) towards the periphery. Excluding this neuroglia-layer, and also the central grey matter around the Sylvian aqueduct, Tartuferi distinguishes four strata in vertical sections.

1. Stratum zonale; superficial white layer.-This is a relatively thin stratum of transversely coursing fine nerve-fibres, which are derived through the superior brachium from the optic tract and retina. Many of them dip down and lose themselves in the next layer, but some appear to be continued towards the middle line and to decussate with others from the same stratum of the opposite side.

2. Stratum cinereum; grey cap.-A layer of grey matter, semilunar in section, being of considerable thickness opposite the most prominent part of the tubercle, but thinning off at its margins. Its nerve-cells are numerous but vary in size, the smaller ones being near the stratum zonale, the larger in the deeper part of the layer. They mostly send their dendrites or protoplasmic processes outwards, i.e., towards the stratum zonale, while their nerve-processes (axis-cylinder processes) are directed inwards towards the deeper layers.

3. Stratum albo-cinereum superius; upper grey-white layer; stratum opticum.The grey matter of this layer is largely interrupted by nerve-fibres, which are continued like those of the stratum zonale from the optic tract, which enters by the superior brachium at the antero-lateral aspect. The fibres differ in calibre in the different parts of the layer, and it may be subdivided accordingly into (a) a mediodorsal zone of coarse medullated fibres, (B) an intermediate zone of fine medullated fibres, and (y) a central zone containing much grey matter interspersed with bundles of white fibres.

These are described by Ganser as three distinct layers, and termed the third, or superficial medullated layer, the fourth, or middle medullated, and the fifth, or middle grey layer; while the fourth layer of Tartuferi, immediately to be mentioned, becomes the sixth and seventh of Ganser.

The whole stratum is richly beset with large nerve-cells, which send their axiscylinder processes mostly into the next or fourth stratum. Of the nerve-fibres, those of the intermediate zone (B) are retinal fibres according to Tartuferi; those of the central zone (y) are probably derived from the corona radiata of the occipital region of the brain, whilst the coarse fibres of the medio-dorsal zone are perhaps derived from the opposite side.

4. Stratum albo-cinereum inferius: deep grey-white layer: stratum lemnisci.— This, although composed of grey matter, is also traversed by many nerve-fibres which appear to be connected with the upper fillet, and probably end in the layer. Some of the fibres, however, are derived from the large cells of the fourth layer, and others from the nerve-cells of this stratum itself (which contains many cells of large size). It is possible that some fibres are derived (over the aqueduct) from the fillet of the opposite side.

Ganser subdivides this layer into two, which he terms respectively the deep white (sixth), and the deep grey (seventh) layers.

Structure of the optic lobes of birds.-The relations of the cells and fibres have not been very satisfactorily made out in the corpora quadrigemina of mammals, but in the optic lobes of the bird, which correspond, as we have seen, to the anterior or superior quadrigeminal bodies of the mammal, these relations have recently been considerably elucidated by the investigations of Ramón y Cajal.

All who have specially worked at this subject distinguish more strata in the optic lobes of birds than in the corresponding bodies of mammals. Thus Bellonci makes the number of layers (exclusive of the central grey matter of the aqueduct) nine, Stieda, twelve, whilst Cajal distinguishes as many as fourteen strata. Of these the most superficial (1.) is a layer of thick medullated fibres coming directly through the optic tract and chiasma from the retina of the opposite side. (In birds all the optic nerve-fibres cross at the chiasma.) They pass in from the side (so that they are cut across in a sagittal section, fig. 79), and after a variable course turn downwards into the deeper layers, where they end at four different levels (as far as the seventh layer) in non-medullated terminal arborisations. Of these terminal

Monakow states that in the dog the anterior brachium contains some fibres from the occipital cortex.