backwards. Often six strands of nerve run forwards, whilst a dorsal and a ventral trunk pass backwards. The size of these trunks depends on the length of the body. The cephalic ganglion is bilateral and is largely developed. In the Hirudinia and Annelida the cerebral ganglia are connected by commissures with a ventral cord, which, in turn, shows individual ganglia connected by commissures. Each ganglion consists of two equal portions with a transverse commissure, and in the higher forms they are so close as to form almost a single cord. It is also evident that the cerebral ganglia are composed of several ganglia fused together, and acquire functional importance as the sense-organs are more highly developed. In the Echinodermata the nervous system consists of a number of trunks placed ventrally and having a radial arrangement. Each of these trunks corresponds to the ventral ganglionic chain of the Annulata. In Asterida (star-fishes) each radial nerve consists of two bands thickened in the middle, and at the end there is a swelling connected with an optical apparatus placed there. In the Echinus (sea-urchin) the nervous ring lies above the the masticatory apparatus. From this ring lateral branches issue which accompany the branches of the ambulacral vessels. In Holothuroida (sea-cucumbers) the nervous ring lies in front and near the mouth, and is thicker than the five nerves which it gives off, thus differing from the Asteroida and Echinoida. The nervous system of the Arthropoda resembles that of the Annelida. There is a large ganglion above the oesophagus, the cerebral ganglion, united to a ventral ganglion by two commissures so as to form a nervous ring. From the ventral ganglion a series of ganglia united by commissures extends along the ventral surface of the body. The increased size of the cerebrum is the most striking characteristic, and no doubt bears a relation to the higher degree of development of the sense-organs, more especially those of sight. In some Crustacea the optic nerves arise from distinct lobes. As pointed out by Gegenbauer, when the optic organs are reduced or lost the cerebrum becomes so small as to be represented by nothing but a commissure. In the individuals having a large portion of the body composed of similar metameres the ganglia are regular in size, appearing in pairs. On the other hand, in the Thoracostraca (crabs, &c.) the anterior ganglionic masses are fused into larger masses so as to correspond to the concrescence of the anterior metameres into a cephalo-thorax. In the abdominal portion of the body, where the metameres are small, distinct, and more or less regular, the ganglia are also distinct and in pairs. In the Protracheata (Peripatus) the nervous system is simpler, and consists of the œsophageal collar with a double ventral cord having no ganglia or swellings on it, although nerve-cells are distributed through it. In the Myriapoda there is a well-marked ventral cord, with ganglia corresponding to the metameres. In the Arachnida the ventral ganglia are often reduced in number and fused. They are characterized by the close connexion between the cerebral ganglia and the ventral cord, owing to the extreme shortness of the commissures (Gegenbauer). In the Scorpions the nervous system is richly segmented, and remarkable for the large size of the ganglion giving off the pedal nerves. The Spiders have a single large ganglion in the cephalo-thorax, no doubt consisting of several ganglia. In the Acarina (mites) the cerebral ganglion is extremely small, and the other ganglia are fused so as to form one single mass, giving off nerves all round. These minute animals show a remarkable degree of concentration of the nervous system. In Insecta (see fig. 12) the ventral cord traverses the whole length of the body, the ganglia being at equal distances, and all united by commissures. A This condition is well seen in the larval condition, and is like the permanent state of the Myriapoda. When the insect passes into the adult condition changes occur, consisting essentially of the fusion of ganglia and a shortening of the commissures. The cerebral ganglion is composed primitively of three pairs,

floor of the oral cavity, between the oesophagus and the lips of the ossicles of

3

and in most cases does not unite FIG. 12.-Typical forms of nervous system with the rest of the ventral cord. It shows hemispheres and a complicated structure. The first ganglion of the ventral cord supplies the organs of the mouth; the three succeeding send nerves to the appendages, feet, and wings; the remaining ganglia are small, except the last, which supplies the generative organs. There is great variety among the Insecta in the number of ganglia in the ventral cord, but coalescence always indicates a higher type of structure. The nervous system of the Brachiopoda is formed of masses of ganglia near the oesophagus. From these nerve-fibres pass to various parts of the body. There is an resophageal ring, but the superior ganglion is very small, owing to the absence of higher sensory organs. În Mollusca (see vol. xvi. p. 635, fig. 1) the nervous system is divided into a superior ganglionic mass, which lies above the commencement of the oesophagus-the supra-oesophageal or cerebral ganglia and a ventral mass which is connected with the other by commissures, and forms the inferior or pedal ganglia. They are both paired. The cerebral ganglion is connected with the sense-organs. Both the cerebral and the pedal ganglionic masses really consist of ganglia fused together. This is well shown in some of the lower forms, in which the pedal ganglia are divided, and form an arrangement like the ventral cord of the Annulata. The remarkable feature in the nervous system of Mollusca is the great development of the visceral ganglia and nerves supplying the heart, branchial apparatus, and generative organs (see vol. xvi. p. 643, figs. 17, 18; p. 644, figs. 20, 21, 22; p. 647, figs. 34, 35; p. 648, fig. 36). In the Lamellibranchia the cerebral ganglia are very small, owing to the absence of a head and its sense-organs. In some forms they are placed so much to the side as to be united by a long commissure. There are also two pedal ganglia, of a size proportional to the degree of development of the foot. The visceral ganglionic mass is often the largest. It lies behind the posterior adductor muscle, and is united by long commissures to the cerebral ganglion (vol. xvi. p. 693, fig. 144). The nervous system of the Gastropoda is remarkable for the large size of the cerebral ganglia. In the Pteropoda the cerebral ganglia either retain their lateral position or approach the pedal ganglia, with which the visceral ganglia are also fused. The three ganglionic masses, cerebral, pedal, and visceral, are also represented in the Cephalopoda, but they are more approximated by the shortening of the commissures. The ganglionic masses consequently are of great size, and they are more differentiated than any other ganglia in invertebrates. It is possible to distinguish an outer grey layer, formed of ganglionic cells, surrounding a white layer, composed of fibres (vol. xvi. p. 679, figs. 113, 114, 115). Lastly, in Tunicata the nervous system is dorsal, instead of ventral, as in other invertebrates. It is developed from the ectoderm, or outermost layer of the embryo, by an infolding so as to form at first a groove and afterwards a tube. In the Ascidian larvæ this nervous tube reaches throughout the length of the tail, and we have thus the remarkable condition of a-dorsal-inedian nerve-cord, analogous to the cerebro-spinal system of vertebrates. Further, embryologists are of opinion that this rudimentary nervous system is the true central organ, although the greater portion of it

in invertebrates. A, in Serpula, a marine annelid; a, cephalic ganglion. B, in a crab; a, cephalic ganglion; b, ganglia fused under cephalo-thorax. C, in a white ant (Termes); a, cephalic ganglion. (Gegenbauer.)

disappears by the atrophy of the tail in the passage from the larval to the adult state. (Gegenbauer.)

Fig. 13.

Fig. 14.

Cerebral vesicles.

am

Comparative View of Nervous System of Vertebrates.—To understand the structure of the complicated central nervous system of vertebrates, and to appreciate the physiological importance of its various parts, it is necessary to trace its development in the embryo and to note the various forms it presents from the lowest to the highest vertebrates. A consideration of the embryological and morphological aspects of the subject clears up many difficult problems which a study of the human nervous system, by far the most complicated physiological system in the body, fails to do, and in particular it gives an intelligent conception of its architecture, as seen both in simple and complex forms. The cerebro-spinal axis begins in the embryo as a tube of nervous matter produced by an infolding of the epiblast, or outermost embryonic layer. The tube widens at its anterior end, and, by constrictions in its wall, three primary cerebral vesicles are formed, which afterwards become the anterior, middle, and posterior parts of the brain. In the fullydeveloped condition the cavity of the tube remains as the central canal of the spinal cord and the ventricles of the brain, whilst the various parts of the brain and cord are formed by thickenings in its walls. The three cerebral vesicles have been called the forebrain, the mid-brain, and the hind-brain. A protrusion from the anterior cerebral vesicle, at first single, but afterwards divided by a median cleft, becomes the rudiment of the cerebral hemispheres (prosencephala), the cavity remaining in the adult condition as the lateral ventricle on each side. From each cerebral vesicle another hollow process protrudes which constitutes the olfactory lobe (rhinencephalon). What remains of the cavity of the first vesicle becomes the third ventricle (thalamencephalon). In the outer and under walls of the prosencephala a thickening formed which becomes the corpora striata, two large bodies in the floor of the lateral ventricles of the adult brain, whilst the roof is modified into the substance of the cerebral hemispheres. Immediately behind the corpora striata, and in the floor of the thalamencephalon, two similar thickenings occur which become the optic thalami, a thin layer between the two constituting the tænia semicircularis, and the Y-shaped canal passing from the cavity between the thalami to the cavities in the cerebral hemispheres (lateral ventricles) is the foramen of Monro. The floor of the third ventricle is produced into a conical process, the infundibulum, at the blind end of which is the pituitary body, or hypophysis cerebri. The roof of this ventricle is very thin, and in connexion with it is developed the pineal gland, or epiphysis cerebri. Transverse fibres pass from the one corpus striatum to the others, constituting the white commissure, whilst the two optic thalami are connected by two grey commissures. In mammals the two cerebral hemispheres are connected by a large and important set of commissural fibres, forming the corpus callosum. In addition there are certain sets of longitudinal commissural fibres. Thus two sets of fibres arise from the floor of the third ventricle, arch upwards, and form the ante ior pillars of the fornix. These are continued over the roof of the third ventricle and run backwards, constituting the body of the fornix. Behind this the bands diverge so as to form the posterior pillars of the fornix. In the higher vertebrates the upper lip of the foramen of Monro thickens, and becomes converted into a bundle of longitudinal fibres, which is continuous anteriorly with the anterior pillars of the fornix. These are continued back between the inner boundary of the cerebral hemisphere and the margin of the corpora striata and optic thalami, and project into the lateral ventricle, forming the hippocampus major. As in highlyformed brains the corpus callosum passes across considerably above the level of the fornix, a portion of the inner wall of the hemisphere on each side and a space between are intercepted. The two inner walls constitute the septum lucidum, and the space the cavity of

is

FIG. 13.-Outline from above of embryo chick in first half of the second day. 1 to 2, three primary encephalic vesicles enclosed in front and at the sides by the cephalic fold; 3, hinder extremity of medullary canal dilated into a rhomboid space in which is the primitive trace; 4, 4, seven proto-vertebral somites. (Quain's Anatomy.)

FIG. 14.-Embryo of dog, more advanced, medullary canal is now closed in; c, anseen from above (after Bischoff). The terior encephalic vesicle; o, primitive optic vesicle; au, primitive auditory vesicle, opposite third encephalic vesicle; veins entering heart posteriorly; pv, am, cephalic fold of amnion; ov, vitelline proto-vertebral somites. (Quain's Anatomy.)

the fifth ventricle. By a thickening of the floor of the middle cerebral vesicle (mesencephalon) two large bundles of longitudinal fibres, the crura cerebri, are formed, whilst its roof is modified into the optic lobes, corpora bigemina or corpora quadrigemina. The cavity, reduced to a mere tube, is the iter a tertio ad quartum ventriculum, or the aqueduct of Sylvius. The third cerebral vesicle, mylencephalon, undergoes less modification than the others. The upper wall is exceedingly thin before the cerebellum so as to form a lamina, the valve of Vieussens, whilst the part behind is covered only by membrane, and opens into the posterior subarachnoid space. The cerebellum makes its appearance as a thin medullary lamina, forming an arch behind the corpora quadrigemina across the wide primitive medullary tube. The portion forming cerebellum, pons Varolii, and the anterior part of the fourth ventricle is termed the epencephalon, whilst the remaining portion, forming the medulla oblongata and fourth ventricle, is the metencephalon. These facts are briefly summarized as follows (Quain, vol. ii. p. 828). Cerebral hemispheres, corpora striata, corpus callosum, fornix,

1. Anterior cerebral vesicle.

2. Middle cerebral vesicle.

3. Posterior primary vesicle.

(a. Prosencephalon -Fore-brain.

lateral ventricles, olfactory bulb (rhinencephalon).

b. Thalamenceph- Optic thalami, pineal gland, pituitary body, third ventricle. optic nerve (primarily).

alon - Inter

brain.

Jc. Mesencephalon

d. Epencephalon -Hind-brain.

e. Metencephalon -After-brain.

Corpora quadrigemina, crura cerebri, aqueduct of Sylvius, optic nerve (secondarily). Cerebellum, pons Varolii, anterior part of the fourth ventricle. Medulla oblongata, fourth ven

tricle, auditory nerve.

The general architecture of the brain considered in this way will be understood by the diagram in fig. 15, whilst details as to the

FIG. 15.-Diagraminatic longitudinal and vertical section of a vertebrate brain. The lamina terminalis is represented by the strong black line between FM and 3. Mb, mid-brain, what lies in front of this being the fore-brain, and what lies behind the hind-brain; Olf, the olfactory lobes; Hmp, the hemispheres; ThE, the thalamencephalon; Pn, the pineal gland; Py, the pituitary body; FM, the foramen of Monro; CS, the corpus striatum; Th, the optic thalamus; CQ, the corpora quadrigemina; CC, the crura cerebri; Cb, the cerebellum; PV, the pons Varolii; MO, the medulla oblongata; I, olfactorii; II, optici; III, point of exit from the brain of the motores oculorum; IV, of the pathetici; V, of the abducentes; VI-XII, origins of the other cerebral nerves; 1, olfactory ventricle; 2, lateral ventricle; 3, third ventricle; 4, fourth ventricle. (Huxley.)

exact anatomy of the human brain will be found under ANATOMY (vol. i. p. 869 sq.). The complex structure of the brain in the higher animals arises to a large extent from the great development of the cerebral hemispheres. At a very early period these grow forward and project more and more beyond the region of the first primary vesicle, which, as has been noticed, never advances farther forward than the pituitary fossa (lamina terminalis); in expanding upwards they take the place previously occupied by the mid-brain, and fill the most prominent part of the head; and by a downward and lateral enlargement

a

P

A

FIG. 16.-Surface of foetal brain at six months (from R. Wagner). This figure is intended to show the commencement of forination of the principal fissures and convolutions. A, from above; B, from left side. F, frontal lobe; P, parietal; 0, occipital; T, temporal; a, a, a, slight appearance of several frontal convolutions; s, Sylvian fissure; s', its anterior division; within C, central lobe or convolutions of island of Reil; r, fissure of Rolando; P, parieto-occipital fissure. (Quain.)

they form the temporal lobes. Thus frontal, parietal, and temporal lobes come to be distinguishable, and somewhat later, by a arther increase posteriorly, the hindmost lobes ccnstitute the

occipital lobes, and the cerebrum at last covers completely all the lower parts of the brain. The hemispheres, therefore, which are small in the early embryo of all animals, and in adult fishes permanently, attain so large a size in man and in the higher animals as to conceal all the other parts. Whilst this general development is going on the layer of grey matter on the surface of the hemispheres increases to such an extent as to throw the surface into folds or convolutions. The upper surface of the hemispheres is at first smooth (see fig. 16). The first appearance of division into lobes is that of a blunt notch between the frontal and temporal parts below, in what afterwards becomes the Sylvian fissure. In the fourth and fifth months there appear the vertical fissure, separating the parietal and occipital lobes, and the transverse fissure, called the fissure of Rolando, which divides the frontal and parietal lobes superiorly, and which is peculiarly characteristic of the cerebral type of man and of the apes (Allen Thomson). Then the convolutions appear from the formation of secondary grooves or sulci, for even at birth they are not fully perfected; and by the deepening of the grooves and the formation of subordinate ones the process goes on during the first years of infancy. For the convolutions see vol. i. p. 873; also PHRENOLOGY, vol. xviii. p. 847.

The evolution of the brain throughout the animal kingdom shows Devel a graduated series of increasing complication proceeding out of the ment of same fundamental type; so that the forms of brain found perma- brain i nently in fishes, amphibians, reptiles, birds, and in the lower mam- animal mals are repetitions of those shown in the stages of the embryonic series, development of the brain of one of the higher animals.

3

A

B

3

C

In the whole class of fishes the brain retains throughout life more or less of the elementary form,- that is, it consists of a series of enlargements, single or in pairs (see fig. 17, C). The simplest of all forms is 2 in the lancelet (Branchiostoma), in which there is no distinction between brain and cord, there being no anterior enlargement to form an encephalon. In the Cyclostomata, as the lampreys, the form is nearer that of the embryo when the five fundamental parts of the brain can be distinguished. At this stage the cerebrum and cerebellum are extremely small, whilst the ganglia chiefly developed are those connected with the organs of sense, more especially those of vision. sharks and skates (Selachii, In the sharks and skates (Selachii, or cartilaginous fishes) the cerebral portion is considerably larger. In osseous fishes (Teleostei) the thalamencephalon is so fused with the mesencephalon as to make the homology of the parts difficult to trace, but both cerebellum and cerebrum are still small relatively to the rest of the brain. The most important part of the brain of a fish is the part behind the mesencephalon, as from it all the cerebral nerves originate. Thus not only are the optic lobes relatively important as being the centres of vision, but the medulla oblongata is usually very large. In many sharks it forms the largest part of the brain (Gegenbauer). The spinal lobes of the electric fishes are differentiations of this portion of the encephalon.

FIG. 17. Typical forms of brains of lower vertebrates. A. Brain of tortoise (Testudo). 1, olfactory; 2, cerebral lobes; 3, corpora striata; 4, optic lobes; 5, medulla. Part of the surface of the cerebral lobes has been removed to show the cavities in the interior, termed "the ventricles." Immediately behind 4, the optic lobes, is the imperfectly. developed cerebellum. B. Brain of common frog (Rana). a, olfactory; b, cerebral lobes covering corpora striata; c, corpora quadrigemina, or optic lobes; d, cerebellum (rudi mentary); s, back of medulla, showing fossa. cerebral lobes; 3, optic lobes; 4, cerebellum. C. Brain of gurnard (Trigla). 1, olfactory; 2,

In the Amphibia the hemispheres are larger, and are divided into two parts (see fig. 17, B). In the Urodela (siren, proteus, triton, newt) the mesencephalon remains small, and consists of one lobe, but in the Anura (frogs, toads, &c.) there is an advance in this part, it being divided into two. In reptiles there is still an advance in the size of the thalamencephalon and mesencephalon, and the prosencephalon is so large as to pass backwards and overlap the thalam, encephalon. The cerebellum (metencephalon) is still small, especially so in Ophidii (serpents) and Saurii (lizards), but in the Chelonii (tortoises, &c.) and in Crocodilini (crocodiles, alligators) it is larger. In the crocodile there is a transverse grooving of the cerebellum, giving rise to foliation or laminar division, which is carried much farther in birds and mammals, indicating a greater power of co-ordination or regulation of movement.

In birds (fig. 18) the vesicles of the mid-brain are partially hidden by development of the cerebral hemispheres. These are connected by a fine anterior commissure, and they contain a large amount of ganglionic matter bulging into the primitive cavity or ventricles, which are of very small size. The middle portion of the cerebellum shows a distinctly laminated structure and a differentiation into white and grey matter. But there is no pons Varolii, nor corpus callosum, nor fornix, nor hippocampus. In the floor of the lateral

ventricles may be seen a ganglionic mass corresponding to corpus striatum and optic thalamus. The optic lobes are relatively large and show considerable differentiation of structure.

Mammals, even the lower orders, not only show a general enlargement of the cerebral hemispheres, but we find a commissure, the corpus callosum, uniting them. This

B

commissure is of small size, FIG. 18.-Typical brain of bird. A, view and is confined to the fore from above; B, lateral view of a bisected brain. A.-a, olfactory; b, cerebral part of the hemispheres in lobes; c, optic or bigeminal lobes; d, Monotremata (Ornithorhynchus, cerebellum; e, medulla oblongata; and Echidna) and Marsupialia f. spinal cord. B.-a, cerebrum; b, cere(kangaroos, &c.), and in some bellum; c, olfactory; d, optic nerves; e, medulla; f, spinal cord. of the Edentata (ant-eaters, sloths, &c.), but it gradually extends farther and farther back as we ascend to the higher orders. The chief changes thus occur in the prosencephalon. In the lower orders of mammals the hemispheres are comparatively small and simple, and do not present any division into convolutions, and very little distinction even of lobes. The cerebral hemispheres gradually grow backwards, covering mid-brain, cerebellum, and medulla oblongata, as we find in the higher Primates (monkeys, apes, and man). There is also a general enlargement of the brain and of the cranial cavity. The development of a posterior lobe only takes place in the higher orders, and in these also the enlargement of the frontal lobes brings the front of the cerebrum more and more over the nasal cavities, causing a development of forehead. This also explains how the olfactory bulbs in more highly-formed brains are thrown below the frontal part of the hemispheres, instead of originating at their anterior borders. But the internal arrangements of the brain also become more complicated. The fornix, already described, establishes, by its longitudinal commissural fibres, a connexion between the anterior and posterior lobes of the cerebrum. In the Monotremata and Marsupialia the mid-brain retains a bifid form, constituting the optic lobes, or corpora bigemina, but in all higher animals each is divided into two by a transverse groove, forming the corpora quadrigemina, of which the anterior pair is the largest. As we ascend also, we find the surface of the brain becoming more and more convoluted (see figs. 19 and 20). This is the general fact; but whilst the convolutions are most numerous and deepest in the highest orders there is no regular 3gradation, as in each group there are very great variations in the degree of convolution (Allen Thomson). Thus in the Monotremata the Echidna has a more convoluted cerebrum than the Ornithorhynchus, whilst in the Primates the brains of the marmosets show a comparatively smooth non-convoluted surface, in striking contrast to the rich convolutions seen on the brains of the higher monkeys and of the apes. It is important to note that

Fig. 19.

Fig. 20.

FIG. 19.-Rabbit's brain. 1, olfac

tory; 2, surface of cerebral hemisphere; 3, lateral ventricle, on the floor of which is seen the corpus FIG. 20.-Cat's brain, showing constriatum; 4, cerebellum.

voluted surface. Contrast the form

the rabbit. In the cat the central

the cerebellum also becomes more and more complicated as we ascend of the cerebellum in the cat and from the lower to the higher groups. lobe is small, whilst the lateral At first merely a lamina or band, as lobes are largely developed. seen in fishes and amphibia, it is a centrally differentiated body in crocodiles. In birds there is an indication of a division into three portions, a central and two lateral, whilst the central is by far the larger, the two lateral being feebly developed. In Monotremata the central portion is larger than the lateral, but; whilst it is larger in Marsupialia, Edentata, and Cheiroptera (bats, &c.), it is clear that the lateral portions are increasing in size so as to make the disproportion less. But in Carnivora (felines, hyæna, otter, bear, &c.) and in Ungulata (sheep, ox, camel, rhinoceros, horse) the lateral lobes, or hemispheres, of the cerebellum develop to a much greater size; and in most of the Primates they are much larger than the median portion, which is now called the worm or "vermiform process." As regards the development of the spinal cord continuous with the medulla oblongata, it need only be said that it does not show any marked peculiarities of structure in different animals. The grey matter from which nerve-fibres originate and in which they end is found in the centre of the cord, and it is most abundant in the regions associated with the development of limbs. The white matter is external, and, in the cords of the higher animals, can be differentiated by fissures into columns, the special functions of which will be hereafter considered. The size of the cord is influenced by the masses of nerves given off from it, so that it attains its greatest thickness and development in the four higher

divisions of the vertebrates possessing limbs. Thus, too, are formed cervical, dorsal, and lumbar enlargements, contrasting with the more uniform and ribbon-like form of the cord in fishes, although even in these there are special enlargements corresponding to the points of exit of important spinal nerves.

Size and Weight of Brain.-The gradual increase in the size of the brain, as compared with that of the body, which is observed as we rise in the animal scale, has some intimate proportional relation to a corresponding increase of the nervous and mental endowments. Information as to the size of the brain may be obtained by direct measurement of dimensions and weight; but as this is often difficult recourse may be had to the measurement of the capacity of the cranium, which contains, however, not only the brain but its accessories, such as membranes and blood-vessels. Details will be found in vol. i. p. $79. After considering the measurements of several thousand skulls made by different observers, the late Dr Allen Thomson arrived at the conclusion that the cranial capacity is on the whole greater among the highly-civilized than among the savage races, and that there is even a very manifest difference to be found between persons of higher mental cultivation and acknowledged ability and those of the uneducated class and of inferior intellectual powers; and he states further that the amount of this difference may be from 5 to 7 per cent. in persons of the same race, and about double that range in those of different races. Thus, the average adult brain of men in Britain being taken at 3 tb, or, more precisely, at 49 oz. avoir. (women, about 44 to 444 oz.), at an average specific gravity of 1040, would give a bulk of 82.5 cubic inches of brain-substance; 10 per cent. being deducted for loss by membranes, fluid, &c., the cranial

capacity will be about 90 inches. Conversely, the weight of the brain may be calculated from the known cranial capacity. If, therefore, the brain of the

uneducated class falls 2.5 oz. below the average, whilst that of the more cultivated persons rises to the same amount above it, or to 52 oz., we may regard these brain-sizes as corresponding with brain-bulks and cranial capacities of 78 and 87 cubic inches, and of 88 and 97 cubic inches respectively. The average brain-weight of an Australian aboriginal man is about 42 oz., corresponding to a brain-bulk of about 70 cubic inches, and a cranial capacity of about 78 cubic inches. There are, however, great variations in all races. Thus the brain of Cuvier, the great naturalist, weighed 65 oz. avoir., corresponding to a brainbulk of 108 cubic inches and a cranial capacity of 118 cubic inches; whilst, on the other hand, in Europeans the brain-weight has fallen as low as 32 oz., or a brain-bulk of 53 cubic inches and a cranial capacity of 63 cubic inches. The brains of the anthropoid apes-gorilla, chimpanzee, and orang-are all inferior to man in their dimensions. In the gorilla the brain does not attain more than a third of the weight of the average human brain, and in the chimpanzee and orang it does not reach a fourth, so that the ratio of brain-weight to bodyweight in these animals may be as 1 to 100, whilst in man it ranges from 1 to 40 to 1 to 50. It is remarkable that in general among the largest animals of any group the brain does not reach a size proportionate to the greater magnitude of the other organs or of the whole body, so that in the smaller members of the same order a considerably greater proportional size of the brain is observed. Thus in the small marmosets the proportion of the brain-weight to the body-weight may be 1 to 20, or more than double the proportion in man. Similar facts are brought out in comparing the brains of cetaceans, pachydermis, dogs, &c., as shown in the following table.

Table of comparative sizes of Brain and Body.

Although the proportion of brain-weight to body-weight in a male child at birth is 1 to 10, yet so rapidly does the brain continue to grow during the early period of childhood that by the age of three years it has attained more than threefourths of its full size, by the age of seven years it has reached the proportion of nine-tenths, and after this, only by slow and small gradations, it attains the full size between the ages of twenty and twenty-five years.2 See PHRENOLOGY. From this survey of the comparative development of the brain the following general conclusions can be drawn.

1. The first and essential portion of the cerebro-spinal axis i the portion forming the spinal cord and medulla oblongata, inasmuch as it is found throughout the whole range of vertebrate existence, and is connected with the reflex or automatic movements on which locomotion, respiration, and the circulation more or less depend, and with the simple sense of contact, or touch, or pressure. This portion is necessary to mere existence.

2. When higher senses are added, such as those of taste, smell, hearing, vision, portions of the anterior part of the cerebro-spinal axis are differentiated so as to form centres. The earliest and most important of these senses (next to touch) is vision, hence the high degree of development of the optic lobes even in the lowest forms; to these are added the optic thalami, which may be regarded as the centres of tactile sensations involving appreciation of differences of touch as to softness, smoothness, hardness, &c., requiring in the periphery special terminal organs. Special centres for hearing, taste, and smell are not differentiated. It is remarkable that the organs relating to the sense of smell are most anterior and most closely related with the prosencephalon, indicating, apparently, that this sense is one of the earliest in appearance, and probably, along with vision and touch, one of the most necessary to existence. 1 The large cranial bulk in this instance is connected with the enormous size of the roots of the cranial nerves.

2 Many of the facts of this paragraph as to size and weight of brain are derived from an unpublished lecture by the late Dr Allen Thomson.

the cord outwards. Again, section of a number of posterior roots is followed by loss of sensation of a part of the body on the same side, and, if the proximal ends of the divided roots-those next the cord-be irritated, painful sensations are excited. The posterior roots, therefore, contain sensory fibres, carrying impressions into the cord from the periphery. As we have seen, these roots are connected with the grey and white matter of the cord, and it is practically impossible to trace all their ramifications. Recourse must therefore be had to the evidence supplied by experiment (cutting, or by the Wallerian method, p. 26) and by pathological observation. In tracing the path of fibres, what may be called the "developmental method" has been pursued. It has been shown by Flechsig that, "if the development of the cord be carefully observed, the medullary substance of the nerve-fibres is formed later along certain tracts of the white columns than in the rest of the white matter, so that in transverse sections of the cord these tracts are easily distinguishable by their more transparent grey appearance" (Quain, vol. ii. p. 277). If the anterior columns be cut by an incision extending into the grey matter, leaving the posterior columns intact, voluntary movements disappear in the parts below the section. Again, section of the posterior columns and grey matter, leaving the anterior uninjured, enteebles but does not destroy the power of voluntary movement below the section. Finally, section of an antero-lateral column on one side paralyses voluntary motion on the same side. From these facts it is inferred (a) that the motor tracts passing from the brain to the periphery are in the anterolateral columns, and (b) that the fibres forming these tracts are chiefly distributed to the same side of the body. These inferences are supported by pathological observation. In diseases where the anterior horns of grey matter are affected paralysis ensues, with complete flaccidity of the limbs; and if, from hæmorrhage, softening, or the pressure of tumours, the anterior portion of the cord be irritated there are spasmodic twitchings of muscles. Complete transverse section of the posterior columns does not abolish sensibility in the parts below; but there is a loss of the power of making co-ordinated movements. Section of the posterior columns and of the antero-lateral columns, leaving only the grey matter in the centre of the cord intact, does not abolish sensibility. Again, section of the antero-lateral columns and of the whole of the grey matter, leaving only the posterior columns uninjured, is followed by complete loss of sensibility in the parts beneath. The inference therefore is that sensory impressions pass through the grey matter. As already seen, many of the sensory fibres connected with the posterior roots decussate in the grey matter. This explains some of the results obtained by Brown-Séquard, that hemi-section of the cord, involving the grey matter, enfeebled sensibility on the opposite side more and more as the section cut deeply into the grey matter; that a vertical section in the bottom of the posterior median fissure caused loss of sensibility on both sides; and that a lateral section, whilst it caused loss of sensibility (anesthesia) on the opposite side, was followed by increase of sensibility (hyperaesthesia) on the same side, -a curious fact, explained by BrownSéquard as being due to irritation caused by paralysis of the vessels of the cord on the side of the section. It would appear also that tactile impressions travel, for a certain distance at all events, in the posterior columns. This has been inferred chiefly from the fact that in certain cases of paralysis involving the posterior columns, where the sensation of touch was absent, the patient could still feel a painful sensation, as when a needle was thrust into the skin; whilst in other cases, in which these columns were not affected, the converse held good. In the disease known as locomotor ataxia (see ATAXY and PATHOLOGY, vol. xviii. p. 392) the patient first passes through a period in which there are disorders of general sensibility, especially lancinating pains in the limbs and back. By and by there is unsteadiness of gait when the eyes are closed or in the dark, and to a large extent the patient loses the power of co-ordinating movement. Especially he is unable to judge of the position of the limbs without seeing them; in other words, the so-called muscular sense is enfeebled. At last there is a stage before death in which there is almost complete paralysis. A study of this disease has thrown much light on the physiology of the cord. It is known to be caused by a slow disorganization or sclerosis of the posterior root-zones, the posterior columns,-slowly passing on to affect the columns of Goll, the lateral columns, and the anterior grey horns, and ultimately involving the cord. The disordered sensations at an early stage, the staggering gait at a later, show that the posterior part of the cord has to do with the transmission of sensory impressions. The man staggers, not because he is paralysed as regards the power of movement, but because, in consequence of the sensory tracts being involved, he does not receive those peripheral impres sions which excite or indirectly regulate all well-ordered movements of locomotion.

2. As a Reflex Centre.-The grey matter of the lower cervical, dorsal, and lumbar regions of the cord may be regarded as composed of reflex centres associated with the general movements of the body, whilst in the upper cervical region there are more differentiated centres corresponding to special actions. The initial excitation

may commence in any sensory nerve; the effect passes to the cord, and sets up changes in the nerve-cells of the grey matter, involving time, and resulting in the transmission outwards along motor fibres of impulses which excite particular groups of muscles. There is an exact co-ordination, with a given strength of stimulus, between certain areas of skin and certain groups of muscles, and thus movements may be so purpose-like as to simulate those of a conscious or voluntary character. Thus irritation near the anus of a decapitated frog will invariably cause movements of the limbs towards the irritated point. towards the irritated point. The activity of reflex centres may be inhibited, as already shown, by higher centres, or possibly by certain kinds of sensory impressions reaching them directly from the periphery. Hence removal of these higher centres is followed by apparently increased reflex excitability. Strychnia and the alkaloids of opium increase it, whilst aconite, hydrocyanic acid, ether, chloral, and chloroform have an opposite effect. certain pathological conditions also, as in tetanus, or in some slow progressive diseases of the cord, reflex excitability may be much increased. In tetanus the slightest touch, a movement of the bedclothes, the closing of a door, the vibration caused by a footstep, may throw the patient into severe and prolonged convulsions. The earlier formed ganglionic cells are those specially concerned in reflex acts.

In

Special reflex centres have been clearly made out in the cord. (1) A cilio- Special spinal centre, between the sixth cervical and third dorsal nerves, associated reflex with the movements of the iris. The fibres controlling the radiating fibres of centres. the iris, and found in the sympathetic, originate here (see EYE). Hence irritation of this region causes dilatation of the pupil, an effect not produced if the sympathetic fibres have been divided. (2) Accelerating centres, supply. ing fibres to the sympathetic which ultimately reach the heart, and irritation of these centres quickens the movements of that organ. (3) Respiratory centres. The movements of respiration, of a reflex character, involve the action of many thoracic and abdominal muscles. Section of the cord above the eighth dorsal paralyses the abdominal muscles; above the first dorsal, the intercostals;

above the fifth cervical, the serratus magnus and the pectorals; and above the fourth cervical, by paralysing the phrenics, it arrests the action of the diaphragm. (4) Genito-spinal centre. This is in the lumbar region. Irritation causes erection, &c.; destruction or disease is followed by loss of virile power. (5) Ano-spinal and vesiculo-spinal centres. These, connected with the movements of the sphincter ani and of the bladder, exist in the lower portion of the dorsal and upper portion of the lumbar regions. Disease or injury involving these centres causes involuntary evacuation of the bowel and complete paralysis of the bladder, with non-retention of urine. The bladder may be full whilst the urine constantly escapes in small quantity.

3. As a Trophic Centre. -The ganglion-cells in the anterior Cord as cornua undoubtedly have a trophic or nutritive influence upon trophic muscles. This has been determined chiefly on pathological evidence centre. If these cells undergo atrophy or degenerative changes, the muscles, even though they may be kept periodically in a state of activity by galvanism, become soft and fatty changes take place. There is thus a correlation between the nutritive condition of muscle and nerve-centre, and influences affecting the one affect the other also.

It has been supposed that the cells in Clarke's vesicular column may form the centres in visceral innervation. They are bipolar, like those in the sympathetic, and not multipolar as in the rest of the cord, and the columns are absent in the lumbar and cervical enlargements. The cells are found where nerves come off that influence the viscera, and similar cells are found at the roots of the vagus in the medulla,-a nerve also much concerned in the innervation of viscera.

Inhibition of Reflex Actions.-The reflex actions of the spinal Inhibicord may be inhibited or restrained to a greater or less extent by tion of the action of centres in the encephalon, so that pure reflex actions reflex only occur after removal of the cerebrum, or during profound sleep, actions. when the cerebrum is inactive Thus a strong effort of the will may restrain from scratching an irritated part of the skin, whilst the same amount of irritation would certainly cause reflex movements if the will were in abeyance. Such power of voluntary control, however, is limited with respect to most reflex actions, whilst some reflex acts cannot be so influenced. Any movement that may be orginated by the will may be inhibited or restrained to a certain extent when the movement is of a reflex character; but, if the movement be invariably involuntary, it can never be inhibited. Thus the ejaculation of semen cannot be voluntarily induced, whilst the reflex act once provoked cannot be arrested (Hermann). That these inhibitions of reflex actions of the cord depend on mechanisms in the brain is proved by the fact that removal of the brain is followed by an increase in the reflex excitability of the cord, and that even section of the cord permits of increased reflex excitability below the plane of section (Setschenoff). Further, after section of the spinal cord in the cervical region, irritation of the lower end arrests reflex movements dependent on reflex centres in the lower cervical, dorsal, and lumbar regions (M 'Kendrick).

Medulla Oblongata.-This is the prolongation into the cranium of the spinal cord so as to unite it with the brain. Strictly speaking, the medulla spinalis and the medulla oblongata form one organ. The columns of white matter of the cord undergo changes in form, structure, and relative position when they pass into the medulla (see vol. i. p. 870). Without again detailing the minute anatomy, it is necessary to show, as in the following table, the connexions of the cord and of the medulla with the rest of the brain