The ventricles of the olfactory lobes are sometimes obliterated; those of the cerebral hemispheres only in some Pisces. The spinal cord extends primitively throughout the whole length of the spinal canal or backbone, and so persists in Pisces; but in other Vertebrata it shortens, its terminal portion being represented by a filum terminale.

1 In Cephalochorda and Urochorda the neural folds unite over the groove from behind forwards, and a pore opening into the brain-cavity persists anteriorly for some time. In Vertebrata, however, the folds unite from before backwards. Gotte describes in the embryo Bombinator a cord of cells uniting the pineal body with the epidermis, and suggests that in this union is to be seen a remnant of the pore alluded to above. A short slit has been found by Van Wijhe in the same position in the embryo Elasmobranch. Recent researches have shown that in Lacertilia the apex of the pineal gland is transformed into an azygos eye. According to de Graaf (Z. A. ix. 1886), this eye, which is detached from the pineal gland itself, lies below the parietal foramen of the skull, between the membranes of the brain. It is a vesicle with cellular walls. The part of the wall turned towards the foramen is thickened and lens-like; the remaining part is deeply pigmented, and in Anguis fragilis has a complex structure, consisting of three tiers of cells. Of these, the one turned towards the cavity of the vesicle has its cells elongate, pigmented and provided with refractile processes. W. B. Spencer (cf. Nature, xxxiv. 1886) has confirmed and extended de Graaf s results.

He finds that the azygos eye is sometimes embryonic (?), e. g. in Cyclodus, sometimes highly developed, e.g. in Hatteria and Iguana. He finds that it is often connected to a nerve, which may be traced into the pineal gland, and is always accompanied by a blood-vessel; that the pigmented cells are transversely striated, and appear to be connected at their bases with the cells of the two other tiers, the whole forming a complex retina; that the cavity of the vesicle is filled with a coagulable fluid. But he has not observed the refractile processes of de Graaf. Mr. Spencer will publish his results in extenso in the Q. J. M. Ahlborn came to the conclusion, on various grounds, that the pineal gland represents a rudimentary eye (Z. W. Z. xl. 1884), but the view appears to have been also hinted at by Rabl-Ruckhard (see, Z. A. ix. 1886, p. 405). The extinct Stegosauria possess a large parietal foramen, which, as suggested by de Graaf, may possibly indicate the former existence of a pineal gland of great functional importance.

The original cellular walls of the spinal cord develope into an outer layer of medullated nerve-fibres (=white matter), and an inner mass of ganglion cells and fine fibrils ( = grey matter) surrounding a canalis centralis, a remnant only of the original cavity of the cord continuous anteriorly with the ventricles of the brain, which is lined by a ciliated epithelium. The grey matter in the thalami optici, mid-brain, and medulla oblongata, forms ganglionic masses; in the cerebellum an outer layer; in the cerebral hemispheres both ganglionic masses (corpus striatum) and an outer layer or cortex. The superficial layer of white matter in the cerebellum and hemispheres is exceedingly thin; on the contrary white matter is largely developed internally to the grey layer. In Amphibia and higher Vertebrata the hemispheres are united anteriorly by a transverse band of fibres - the anterior commissure. A similar band behind the root of the pineal gland forms the posterior commissure of Elasmobranchii, Sauropsida, and Mammalia. In the last-named class there is a grey or middle commissure uniting the two optic thalami, and two bands, one of transverse fibres - the corpus callosum - uniting the hemispheres across the roof of their ventricles, the other of longitudinal fibres - the fornix1. The brain and spinal cord are invested by three membranes - a pia mater, immediately applied to their surfaces; an arachnoid, covering the pia mater; and a dura mater, which lines the inner surface of the cranium, but in the spinal cord is split into two layers, one lining the spinal canal, the other, or theca spinalis, forming a sheath round the cord.

The cranial or cerebral nerves are (1) olfactory, in connection with the olfactory lobes of the brain; (2) optic, which forms a chiasma or union with its fellow, in which fibres cross from one to the other side, except in Teleostei, where the two nerves cross one another, the right to the left side, and vice versa; (3) the oculo-motor, which supplies all the muscles of the eye except two; (4) the trochlear, which supplies the superior oblique muscle of the eye; (5) the trigeminus, or mixed sensory and motor nerve of the face and jaws; (6) the abducens, or nerve of the external rectus muscle of the eye; (7) the facial, chiefly a motor nerve; (8) the auditory; (9) the glossopharyngeal; (10) the vagus; (11) the spinal accessory; and (12) the hypoglossus. Of these nerves the optic is derived from the brain, the spinal accessory and hypoglossus apparently from spinal nerves, the trochlear and abducens perhaps belong to the trigeminus and facial respectively. The auditory is supposed by some authorities to be derived from the facial, by others it is supposed to be like the remaining cranial nerves (1 (?) 3, 5, 7, 9,10, supra) segmental1. The cranial nerves differ from spinal nerves in having primitively single roots, and in being connected with branchial sense organs (infra) from which the olfactory and auditory sensory epithelia are perhaps derived in the first instance.

The ganglion of the roots of the oculo-motor, i. e. the ciliary; of the trigeminus, i. e. the Gasserian; and of the remaining segmental nerves are supposed to arise in connection with the same sense bodies. Each segmental nerve typically supplies a gill cleft. The main trunk passes down the posterior side of the cleft, a small branch down the anterior side, a third to the pharynx, and a fourth - the so-called dorsal branch - appears in continuity with the branchial sense organ at the upper end of the branchial cleft 2. This primitive arrangement is very well seen in the case of the glossopharyngeal nerve of many fish. The spinal nerves have two roots - a dorsal sensory root with a ganglion, and an anterior motor root which joins the sensory distally to the ganglion. The two roots have different origins. But all nerves, cranial as well as spinal, are outgrowths continuous with the central nervous system. And the more or less extensive series of sympathetic ganglia, which lies at the side of the backbone, and is connected with the nerves and ganglia of the viscera, take their origin from the spinal nerves, to which they are always united by rami communicantes.

1 The corpus callosum appears to be represented in both Reptilia and Amphibia. The anterior commissure of the brain in these classes is divisible into an upper and lower portion. The latter represents the true anterior commissure; the former represents the corpus callosum, 'and contains the fibres of the dorso-medial moiety of the hemispheres.' See Osborn, Z. A. ix. 1886. The corpus callosum is also possibly indicated in Fish.