BRAIN.] ANATOMY 877 calcareous particle.3, constituting the brain sand, are also found in it. Usually it is hollowed out into two or more small cavities. The function of the pineal body is not understood, but both it and the pituitary body, which possess a certain structural correspondence, are usually referred to the type of the ductless glands. The corpora quadrigemina or optic lobes are situated behind and between the two optic thalami, and rest upon the posterior surface of the crura cerebri. The division into two lateral halves is marked by a shallow longitu dinal fissure, and the subdivision of each half into an an terior and a posterior eminence, by a shallow transverse fissure. The anterior pair of eminences are called nates; the posterior, testes. From each testis a strong white band, the superior peduncle of the cerebellum, passes back wards to the cerebellum, and stretching between the pair of peduncles is the valve of Vieussens or anterior medullary velum. The corpora quadrigemina are tunnelled in the antero-posterior direction by the aqueduct of Sylvius, which opens anteriorly into the third ventricle immediately below the posterior commissure, and posteriorly into the fourth ventricle under cover of the valve of Vieussens. It is lined by a cylindrical ciliated endothelium. INTERNAL STRUCTURE OF THE CEREBRUM. The cere brum is composed both of grey and white matter, the general relations of these two forms of nerve matter to each other may be seen by making sections through the cerebrum. The determination, however, of their minute structure, and of the relations and connections of the nerve fibres to the nerve cells is, owing to the delicacy of the organ, one of the most difficult departments of ana tomical study. Several anatomists have endeavoured to trace out the course of the nerve fibres in the organ, and though our knowledge is by no means complete, yet many important facts have undoubtedly been ascertained. These facts have been summarised, and numerous valuable addi tions made to them by Meynert in a recent elaborate memoir, which has been frequently consulted and made use of in writing the following description. The Grey Matter of the cerebrum is disposed in three great groups : a, The grey matter of the cortex of the hemi spheres ; 6, the grey matter of the great ganglia of the base of the cerebrum ; c, the central grey matter which forms the wall of the cerebral end of the cerebro-spinal tube. a, The grey matter of the cortex of the hemisphere forms the superficial part of the convolutions, and is known as the great hemispherical ganglion, but in some localities, as at the loci perforati antici and the septum lucidum, it has received distinctive names. When a convolution is divided vertically the grey matter is seen to be confined to the surface and to enclose a white core. The grey matter presents a laminated appearance, and as a rule consists of five or six layers, which are composed of the characteristic pyramidal nerve cells of the cortex of the cerebrum, of nerve fibres, of matrix or neuroglia, and of blood-vessels. The most superficial layer consists of neuroglia, in which nerve fibres extend parallel to the surface of the convolutions. In the deeper layers are found the pyramidal nerve cells, which lie with their long axes vertical to the surface of the convolutions, and which contain angular nuclei. From the observations of Lockhart Clarke, Arndt, Cleland, and Meynert, there can be no doubt that the pyramidal nerve cells vary in relative size and in numbers in the different layers of the grey cortex, and that the largest sized pyra midal cells Lie in the third and fourth layers. L. Clarke stated that the cells of all the layers of the posterior or occipital lobe were small and of nearly uniform size, whilst in the convolutions anterior to it numerous cells of a much larger kind were found ; but though it is undoubtedly true that large pyramidal cells are found in the frontal lobe in considerable numbers, and that the greater number of the cells of the occipital lobe are small and nearly uniform in size, there is no difficulty in recognising in the occipital lobe a small proportion of cells, quite equal in magnitude to the largest cells of the frontal lobe, interspersed amongst the smaller pyramidal cells. The nerve fibres which ascend into the grey matter from the white core of the convolution radiate into its several layers, and are apparently continuous with the basal axis-cylinder processes of the nerve cells. According to Cleland, the elongated apices of the cells, which are directed to the surface of the convolution, are continuous with the nerve fibres situated in the superficial layer of horizontal fibres. Immediately subjacent to the large pyramidal cells numerous small, irregularly shaped nerve corpuscles, Like those of the internal granule layer of the retina, form the so-called granule layer ot the grey matter. Fusiform cells, which give off lateral processes, are found in the deepest layer of the grey matter, and form the daustral layer of Meynert. Gerlach has described here, as in the spinal cord, a network of extremely minute nerve fibres, with which the branched lateral processes of the nerve cells are apparently continuous. The neuroglia con tains ^multitudes of small rounded corpuscles. In it also are found small stellate cells, provided with minute branched processes, which cells, as Meynert states, are so pellucid, that in the healthy brain they seem to be only free nuclei ; it is difficult to say whether these cells belong to the neuroglia, or are nerve cell elements. The grey cortex of the cerebrum is much more vascular than the white matter. The arteries derived from the pia mater pass vertically into it, and end in a close polygonal network of capillaries ; but it is also traversed by the arteries, which terminate in the capillary network of supply for the white matter. In the grey matter of the cortex of the occipital lobe eight layers have been described by Clarke and Meynert. The increase in number is due to the intercalation of two additional granule layers, which coalesce and form a dis tinct white band in the grey matter, owing, as Meynert states, to the absence of pigment in the cells of the granule layers. The grey matter of the cortex of the island of Reil and of the convolutions bounding the Sylvian fissure contains a very large proportion of fusiform cells. They form the chief constituent of the grey claustrum, situated deeper than the grey matter of the island, and separated from the outer part of the corpus striatum by a thin layer of white matter. Fusiform cells also occur abundantly in the nucleus amyg- dalce, a grey mass situated below the corpus striatum, which in some sections seems as if isolated, but in reality is continuous with the grey matter of the inferior temporo- sphenoidal convolution. The grey matter of the cortex of the gyrus hippocampi and of the hippocampus major is apparently destitute of both the granule and claustral layers of cells. Its super ficial layer has been named the nuclear lamina, and contains small and scattered nerve corpuscles. Next this lamina lies the striatum reticulare, in which the apices of the numerous pyramidal cells of the third layer branch and again unite to form a delicate network. Deeper than the pyramidal cells is a thick layer of so-called "granules," which A. B. Stirling recognised some years ago as like the granules of the rust coloured layer of the cerebellum ; like them they consist of a well-defined nucleus invested by delicate branched protoplasm. The grey matter of the two layers of the septum lucidum, though included be tween the corpus callosum and fornix, is yet in the same plane as the grey matter of the cortex of the inner surface of the hemispheres, but is cut off from it by the develop ment of the transverse fibres of the corpus callosum.
The grey matter of the locus perforatus anticus contains
