All along the upper curved margin of the lateral wall, from the pillar of the fornix to the pineal gland, runs a white stria, known as the stria pinealis, stria medullaris, or tania fornicis (fig. 73, Tfo). The floor of the ventricle is formed posteriorly by the tegmenta of the crura cerebri, and where the crura diverge from one another by the following parts, which have been already mentioned as seen at the base of the cerebrum; viz., commencing from behind, the grey matter of the posterior perforated space, the corpora albicantia, seu mamillaria, the tuber cinereum and infundibulum; the lamina cinerea serves to close the ventricle in front. The roof of the cavity is limited before and behind by two commissures, named from their position, anterior and posterior. Of these the anterior will be described with the cerebral hemispheres. The third ventricle is lined, like the other cavities already described, by ciliated epithelium, which is thin and flattened over the roof, i.e., lining the velum and choroid plexuses, but longer and more columnar at the bottom and sides. The floor, which is narrow, is formed, underneath the epithelium, of grey matter continuous with that of the Sylvian aqueduct, and this central grey matter extends a short distance upwards on the wall of the thalamus. The central grey substance rests behind upon the still conjoined part of the tegmenta; but anteriorly, after these have diverged, it comes to the surface at the base of the brain as the posterior perforated lamina and the tuber cinereum. The lateral walls of the ventricle have but a thin covering of neuroglia (ependyma) underneath the lining epithelium; so that the white covering (stratum zonale) of the thalami comes to view through it.

The epithelial covering of the roof of the ventricle is not free but covers the under surface of the median portion of an expansion of pia mater named the velum interpositum, which overlies the third ventricle as well as the larger part of the optic thalami. The epithelium follows all the inequalities of two fringed vascular tracts (choroid plexuses of the third ventricle) which project downward from the membrane, and it becomes torn away when the pia mater is removed. At the pineal stria (fig. 73, Tfo) on either side it is continuous with the epithelium covering the lateral wall This stria therefore represents the limit of the third ventricle so far as the lateral boundaries of the roof are concerned: the upper surface of each optic thalamus is excluded from this cavity.

The central grey matter of the aqueduct (fig. 76, c. gr.) is a layer 2 to 3 millimeters thick which surrounds the aqueduct, and is prolonged from the grey matter of the fourth ventricle. It contains, scattered through its substance, nerve-cells of varying size, the largest being prolonged upwards from the locus cœruleus of the fourth ventricle; the cells are very numerous and small at the dorsal side of the aqueduct. In addition to these scattered cells the grey matter of the aqueduct contains certain more defined groups or columns of cells which are connected with the roots of the third and fourth, and of the fifth cranial nerves. Amongst the cells there is a network of fine medullated nerve fibres, whilst near the aqueduct and immediately under the ependymal layer many fine longitudinal fibres are seen.

The nuclei of the third and fourth nerves (fig. 76, n.III., IV. extend on either side along almost the whole length of the ventral part of the aqueduct, close to the middle line, the nuclei of the two sides being only separated from one another by the raphe; at one part they even meet across this (fig. 78, n.III). The cells of these nuclei are large and irregular in shape, and of a yellowish colour. The nucleus from which the root-bundles of the fourth nerve spring does not begin to show itself until the level of the upper part of the inferior corpora quadrigemina, and it here lies just below and rather to the side of that from which the bundles of the third originate. From here the bundles of the fourth pass obliquely downwards towards the pons, and just before reaching this the nerve turns sharply dorsal-wards, and passes into the superior medullary velum, in which it crosses horizontally, decussating with that of the opposite side (fig. 72).

NUCL. EDINGER-
WESTPHAL

NUCL.MED

ANT

The nucleus from which the third nerve takes origin extends upwards underneath the superior corpora quadrigemina, ventral to the Sylvian aqueduct, and even extends into a corresponding situation in the posterior part of the third ventricle. The anterior (superior) part is composed of smaller cells than the other portion: it extends forwards into the wall of the third ventricle, and from the experiments of Hensen and Voelckers and the observations of Starr, it appears to be subdivided into two portions, of which that which is the more mesial, and lies just above the corpora mamillaria, is connected with the fibres of the third nerve to the ciliary muscle, whilst that which is the more lateral is connected with the fibres concerned with the contraction of the sphincter pupillæ. The main part of the oculomotor nucleus is formed by large cells, which tend to be grouped (see diagram, fig. 74). Thus, there are two distinct groups on each side which are dorso-lateral (dorsal nuclei), and two which are ventro-mesial (ventral nuclei). These four nuclei are grouped around a central nucleus which lies in the middle line. Besides these, there is an elongated nucleus of small cells which inferiorly (caudalwards) lies in close contact with the central nucleus but superiorly curves outwards. This is known as the nucleus of Edinger and Westphal, but it is uncertain whether it gives origin to any fibres of the third nerve. It has not been certainly ascertained from which of the several groups the fibres to particular muscles moving the globe of the eye proceed. From the several groups of cells which constitute the oculomotor nucleus the fibres of the third nerve pass with a curved course through the tegmentum, to emerge at the inner margin of the crusta of the same side; but the fibres from the posterior of the dorsal groups undergo decussation. These fibres are believed to pass to the internal rectus of the opposite side.

NUCL. LAT. ANT

(DARKSCHEWITSCH)

NUCL.DORS.I.(ANT)

HUCL.VENT.I.(ANT.)-—.

NUCL.DORS.11.(POST)
(V. GUDDEN)

NUCL. CENTRALIS-

NUCL.VENT.IL(POST)

NUCL.TROCHL.

NERV. TROCHL.

NERV. OCULO-
MOTORIUS

Fig. 74.-DIAGRAM OF THE GROUPS OF CELLS FORMING

THE NUCLEI OF THE THIRD AND FOURTH NERVES.

(Perlia.)

In a case recorded by Kahler and Pick, in which there was paralysis of the levator palpebræ, the rectus superior and the obliquus inferior, a lesion was found involving the postero-lateral bundle of the nerve-roots. The observations of Starr point to these three muscles being innervated from the dorsal (dorso-lateral) groups, and the rectus internus and rectus inferior from the ventral (ventro-mesial) groups, in the order here given (from above down).

It has been shown by Duval and Laborde that the third nerve receives fibres from the mesial part of the posterior longitudinal bundle of the opposite side (possibly some fibres also pass to it from the posterior longitudinal bundle of the same side). These fibres are derived mainly from the nucleus of the sixth nerve, and pass out along with the fibres of the third nerve to the internal rectus, so that the nucleus of the sixth thus supplies both the external rectus of the same side entirely, and the internal rectus of the opposite side partially (fibres derived from the nucleus of the third also going to the internal rectus). These are, it may be noted, the muscles which are brought together into action in conjugate deviation of the eyes to either

side, and the cases which are sometimes met with of conjugate paralysis involving the internal rectus of one side, and the external rectus of the other side, which are accompanied by atrophy of the nucleus of the sixth, are thus accounted for.

The prolongation of the upper nucleus of the fifth nerve consists of a small number of large globose cells (figs. 75, 76, d.V), which lie at the extreme lateral margin of the grey matter of the aqueduct close to the bundles of the descending root of the fifth nerve, towards which their axis-cylinder processes are directed. This nucleus and root gradually become smaller, and disappear before the superior end of the mesencephalon is reached.

The crura cerebri (fig. 32, P) emerge from the upper border of the pons and diverge from one another, leaving between them the posterior perforated space and the corpora mamillaria and disappearing in the cerebral hemispheres under the optic tract. The triangular interval seen at the base of the brain to be enclosed between the diverging crura has been termed trigonum interpedunculare by Schwalbe. Near the point of the angle of divergence the roots of the third nerve issue in several bundles from a groove along their inner side (fig. 32, III.); and this groove serves to indicate the separation between the more prominent ventral part of the peduncle (pes s. basis s. crusta pedunculi, fig. 75, cr.) and the dorsal and larger part (tegmentum, t.) which is in great measure concealed from view by Fig. 75.-OUTLINE OF TWO SECTIONS ACROSS THE MESENCEPHALON. Natural size. (E. A. S.)

d.v pl.

A

l.g.

CT

Cr

c.

dv

P.l.

t

B

cr:

1.g.

A, through the inferior pair of the corpora quadrigemina; B, through the superior pair.

cr, crusta; s.n., substantia nigra; t, tegmentum; s, Sylvian aqueduct with the central grey matter; c.q., grey matter of quadrigeminal bodies; l.g., lateral groove; p.l., posterior longitudinal bundle; d. V, descend

ing root of fifth nerve; s.c.p., superior cerebellar peduncle; f, fillet. The dotted circle in B indicates the tegmental nucleus.

the pes when viewed from below and in front. A section into the crus cerebri shows the two parts of which it is composed to be separated from one another by a tract of dark coloured grey substance known as the substantia nigra (fig. 75, sn), which comes to the surface on the inner side at the groove above mentioned from which the third nerve issues (sulcus oculomotorii), and on the outer side also along a grooved line-the sulcus lateralis (fig. 75, l.g.).

Of the two main parts of each peduncle the crusta (cr) is formed almost entirely of lamellated bundles of longitudinal fibres, some of which are continuous with the pyramid-fibres of the medulla oblongata and pons, whilst others are superadded; and the tegmentum is a continuation of the formatio reticularis of those parts, with the addition of much grey matter and white fibres, amongst the latter being those of the superior peduncle of the cerebellum. The two ventral portions (crustæ or pedes) are entirely distinct from one another (as shown in the accompanying sections, fig. 75), and each is marked off externally from the tegmentum of the same side by the grooves just mentioned; but the two tegmenta are united in the median plane by a prolongation of the raphe, and extend dorsally at the sides of the aqueduct to become continuous with the bases of the corpora quadrigemina.

Crusta. The crusta is semilunar in section, the substantia nigra projecting into it with an irregular convex border. It is made up of longitudinal white fibres which become arranged in the higher parts of the mid-brain into flattened bundles, with their edges dorsal and ventral, separated from one another by processes of pia mater. The main part is a direct prolongation of the longitudinal bundles of the pons and passes superiorly towards the internal capsule of the cerebral hemisphere.

Close to the substantia nigra, the bundles of white fibres are smaller and somewhat separated by projections of the grey matter extending between them. They have received the name of stratum intermedium. These are usually stated to have a different origin and destination from the other fibres of the crusta, passing, according to Meynert, between the lenticular nucleus of the corpus striatum (see p. 131), and the substantia nigra and reticular formation of the bulb and pons. It is doubtful, however, if this is true for any of these fibres, and is certainly not the case with the majority, for (in monkeys) after a lesion of the Rolandic region, most if not all the fibres of the stratum intermedium undergo degeneration along with the fibres of the pyramidal tract, to which, therefore, they must be regarded as belonging. The pyramidal tract of the mesencephalon, or continuation of the pyramidbundles of the pons, occupies about the middle third of the crusta. Superiorly its fibres pass through the middle part of the internal capsule to the fronto-parietal or Rolandic region of the hemisphere. By far the majority arise from the cortical cells, but a few have their cell-origin in the cord or bulb, and degenerate after lesions of these parts.

It is remarkable that whereas in the lower part of its course (spinal cord and bulb) the fibres of the pyramidal tract acquire a medullary sheath later than the other fibres of the white columns, in the upper part (crus cerebri and cerebrum) it is acquired earlier than in the other fibres.

The outer or lateral third of the crusta is formed of fibres which are traceable downwards to the lateral longitudinal bundles of the pons, and upwards to the posterior part of the internal capsule, but their origin and destination have not yet been satisfactorily made out. They are probably connected superiorly with the occipito-temporal regions of the cerebral cortex, and according to Flechsig they arise below from the cells of the nuclei pontis.

The mesially situated bundles of the crusta are also distinct from the pyramidal tract proper (Flechsig), being developed at a later period. They are perhaps connected through the anterior part of the internal capsule with the prefrontal region of the hemisphere. Finally, one well-marked bundle in the crusta is connected with the fillet (mesial fillet, pp. 66, 67, and 103). This bundle is at the lateral border of the pyramidal tract in the upper part of the crusta, but lower down crosses obliquely over or between the fibres of that tract to attain the mesial border of the crusta, whence it is traceable to the fillet. It contains, according to Spitzka, the afferent cerebral tracts of the cranial nerves. Traced upwards it is lost in the subthalamic region.

Lastly the crusta includes some fibres derived originally from the cerebellum and joining the pyramidal tract in its passage through the pons, such fibres being scattered amongst the fibres of the other tracts.

The substantia nigra is a mass of grey matter which is characterised by the presence of a number of very darkly pigmented irregular nerve-cells, which give the substance in which they are scattered the appearance from which it derives its name. It forms a layer which separates the crusta from the tegmentum. It is thicker near the mesial border of the peduncle than laterally, where the tract of the fillet may be but incompletely separated by it from the longitudinal bundles of the crusta. It commences at the upper margin of the pons, and can be traced as far forwards as the posterior border of the corpora albicantia. At the origin of the third nerve it is traversed in its mesial part by some of the issuing fibres of the nerve-root. The grey matter of the substantia nigra projects here and there between the adjacent bundles of the crusta; one considerable projection in particular in the lower part of the mesencephalon serving to mark off the mesial portion of the crusta from the rest. The cells in this projection are much smaller, and relatively more numerous than in the rest of the substantia nigra.

The tegmentum, like the formatio reticularis of the bulb and pons, of which it is the prolongation upwards, is composed of small longitudinal bundles of white fibres, separated by transversely coursing or arched fibres, together with a considerable amount of grey matter containing scattered nerve-cells.

In addition to these diffused bundles of longitudinal fibres there are others which are collected into more defined tracts. One such tract constitutes the posterior or dorsal longitudinal bundle, which is seen in all sections of this part of the brain as a pyriform area of transversely cut fibres which lies on each side of the middle line between the grey matter underlying the aqueduct, and the formatio reticularis (fig. 76, p.l.b.). The fibres which constitute this bundle below

Fig. 76.-TRANSVERSE SECTION ACROSS THE MID-BRAIN, THROUGH THE INFERIOR CORPORA QUADRIGEMINA. MAGNIFIED ABOUT 3 DIAMETERS. (E. A. S.) From a Photograph.

gr., dorsal quadrigeminal groove (sulcus longitudinalis); c.q.p., corpus quadrigeminum posterius ; str.l., stratum lemnisci; c.gr., central grey matter; n. III, IV, oculo-motor nucleus; d. V, descending root of fifth nerve; p.l.b., posterior longitudinal bundle; f.r.t., formatio reticularis tegmenti; d, d', decussating fibres of tegmentum; 8.c.p., decussating fibres of superior cerebellar peduncles; f, upper fillet f', lower or lateral fillet; p.p., pes pedunculi; s.n., substantia nigra; g.i.p., interpeduncular grey matter; Sy, Sylvian aqueduct.

have already been noticed (see p. 65 and figs. 53, 54); traced upwards its fibres become related to the nuclei of the third and fourth nerves, and mostly pass out with the roots of these nerves. The posterior longitudinal bundle is composed exclusively of large nerve-fibres, which acquire their myelin at an early stage, in fact, as soon as the roots of the nerves themselves. Although its fibres are large, it gives off principally fine nerve-fibres (Köppen). If this is the case they are probably collaterals. It appears to be developed in nearly all vertebrates, and in some is better marked than in mammals; e.g., in the lizard it can be traced right down the cord dorsal to the anterior or ventral commissure. Although mainly related to the