hard, who showed that the stimulation of certain branches of the first and second, and occasionally the third, sacral nerves (dog) caused a dilatation of the blood-vessels of the penis and erection of that organ, and with Goltz, who found an erection centre in the lumbo-sacral cord. Numerous researches in recent years, among which the reader is referred especially to the work of Langley and Langley and Anderson, have shown that the vaso-motor nerves of the external generative organs of both sexes may be divided into a lumbar and a sacral group.

The lumbar fibres pass out of the cord in the anterior roots of the second, third, fourth, and fifth lumbar nerves, and run in the white rami communicantes to the sympathetic chain, from which they reach the periphery either by way of the pudic nerves or by the pelvic plexus. The greater number take the former course, running down the sympathetic chain to the sacral ganglia, and passing from these ganglia through the gray rami communicantes to the sacral nerves. None of the fibres thus derived enter the nervi erigentes of Eckhard. Of the various branches of the pudic nerves (rabbit), the nervus dorsalis causes constriction of the blood-vessels of the penis and the perineal nerve contraction of the blood-vessels of the scrotum. The course by way of the pelvic plexus is taken by relatively few fibres. They run for the most part in the hypogastric nerves, a few sometimes joining the plexus from the lower lumbar or upper sacral sympathetic chain, or from the aortic plexus. The presence of vaso-dilator fibres in the lumbar group is disputed.

The sacral group of nerves leave the spinal cord in the sacral nerve roots. Their stimulation causes dilatation of the vessels of the penis and vulva.

Internal Generative Organs (those developed from the Müllerian or the Wolffian ducts).—Langley and Anderson find vaso-constrictor fibres for the Fallopian tubes, uterus, and vagina in the female, and the vasa deferentia and seminal vesicles in the male, in the second, third, fourth, and fifth lumbar nerves. The internal generative organs receive no afferent, and probably no efferent, fibres from the sacral nerves.

The position of the sympathetic ganglion-cells, the processes of which carry to their peripheral distribution the efferent impulses brought to them by the efferent vaso-motor fibres of the spinal cord, may be determined by the nicotin method of Langley. About 10 milligrams of nicotin injected into a vein of a cat prevent for a time, according to Langley,' any passage of nerve-impulses through a sympathetic cell. Painting the ganglion with a brush dipped in nicotin solution has a similar effect. The fibres peripheral to the cell, on the contrary, are not paralyzed by nicotin. Now, after the injection of nicotin the stimulation of the lumbar nerves in the spinal canal has no effect on the vessels of the generative organs. Hence all the vaso-motor fibres of the lumbar

nerves must be connected with nerve-cells somewhere on their course. The lumbar fibres which run outward to the inferior mesenteric ganglia are for the most part connected with the cells of these ganglia. A lesser number is con'Langley and Anderson: Journal of Physiology, 1894, xvi. p. 420.

nected with small ganglia lying as a rule near the organs to which the nerves are distributed. The remaining division of lumbar fibres running downward in the sympathetic chain, and including the majority of the nerve-fibres to the external generative organs are connected with nerve-cells in the sacral ganglia of the sympathetic.

The sacral group of nerves enter ganglion-cells scattered on their course, most of the nerve-cells for any one organ being in ganglia near that organ. Bladder.-Neither lumbar nor sacral nerves send vaso-motor fibres to the vessels of the bladder.

Portal System.-It has already been said that vaso-constrictor fibres for the portal vein were discovered by Mall in the splanchnic nerve. Constrictor fibres have been found by Bayliss and Starling in the nerve-roots from the third to the eleventh dorsal inclusive. Most of the constrictor nerves pass out from the fifth to the ninth dorsal.

Back. The dorsal branches of the lumbar and intercostal arteries, issuing from the dorsal muscles to supply the skin of the back,2 can be seen to contract when the gray ramus of the corresponding sympathetic ganglia are stimulated.

Limbs.-The vaso-motor nerves of the limbs in the dog leave the spinal cord from the second dorsal to the third lumbar nerves. The area for the hind limb, according to Bayliss and Bradford, is less extensive than that for the fore limb, the former receiving constrictor fibres from nine roots, namely the third to the eleventh dorsal, the latter from six roots, the eleventh dorsal to third lumbar. Langley finds that the sympathetic constrictor and dilator fibres for the fore foot are connected with nerve-cells in the ganglion stellatum; while those for the hind foot are connected with nerve-cells in the sixth and seventh lumbar, and the first, and possibly the second, sacral ganglia.

Thompson and Bancroft have studied the nerves to the superficial veins of the hind limb. The latter finds that in general the arrangement of the vaso-motor nerves corresponds to that of the arterial vaso-motor nerves and the sweat fibres. The fibres to the superficial veins originate from the lower end (first to fourth lumbar nerves) of the region of the spinal cord supplying all the vaso-motor nerves for the hind limbs.

Tail.—Stimulation of any part of the sympathetic from about the third lumbar ganglion downward almost completely stops the flow of blood from wounds in the tail. The vaso-motor fibres for the tail leave the cord chiefly in the third and fourth lumbar nerves. Their stimulation may cause primary dilatation followed by constriction.

Muscles. According to Gaskell, the section of the nerve belonging to

1 Bayliss and Starling: Journal of Physiology, 1894, xvii. p. 125.

'Langley: Journal of Physiology, 1894, xvii. p. 314.

3

Thompson Archiv für Physiologie, 1893, p. 104; Wertheimer: Archives de Physiologie, 1894, p. 724; Bancroft: American Journal of Physiology, 1898, i. p. 477; Bayliss and Bradford: Journal of Physiology, 1894, xvi. p. 16; Langley: Journal of Physiology, 1894, xvii. p. 307; Piotrowski: Archiv für die gesammte Physiologie, 1893, lv. p. 258.

Langley: Journal of Physiology, 1894, xvii. p. 311.

VOL. I.-14

any particular muscle or group of muscles causes a temporary increase in the amount of blood which flows from the muscle vein. The stimulation of the peripheral end of the nerve also increases the rate of flow through the muscle. The same increase is seen on stimulation of the nerve when the muscle is kept from contracting by curare, provided the drug is not used in amounts sufficient to paralyze the vaso-dilator nerves. Mechanical stimulation by crimping the peripheral end of the nerve gives also an increase. The existence of vaso

dilator nerves to muscles must therefore be conceded. The presence of vaso-constrictor fibres is shown by the diminution in outflow from the left femoral vein which followed Gaskell's stimulation of the peripheral end of the abdominal sympathetic in a thoroughly curarized dog, but the supply of constrictor fibres is comparatively small. In curarized animals reflex dilatation apparently follows the stimulation of the nerves the excitation of which would have caused the contraction of the muscles observed, had not the occurrence of actual contraction been prevented by the curare. The stimulation of the central end of nerves not capable of calling forth reflex contractions in the muscles observed --for example, the vagus-seems to cause constriction of the muscle-vessels.

IV. SECRETION.

A. GENERAL CONSIDERATIONS.

THE term secretion is meant ordinarily to apply to the liquid or semiliquid products formed by glandular organs. On careful consideration it becomes evident that the term gland itself is widely applied to a variety of structures differing greatly in their anatomical organization-so much so, in fact, that a general definition of the term covering all cases becomes very indefinite, and as a consequence the conception of what is meant by a secretion becomes correspondingly extended.

Considered from the most general standpoint we might define a gland as a structure composed of one or more gland-cells, epithelial in character, which forms a product, the secretion, that is discharged either upon a free epithelial surface such as the skin or mucous membrane, or upon the closed epithelial surface of the blood- and lymph-cavities. In the former case —that is, when the secretion appears upon a free epithelial surface communicating with the exterior, the product forms what is ordinarily known as a secretion; for the sake of contrast it might be called an external secretion. In the latter case the secretion according to modern nomenclature is designated as an internal secretion. The best-known organs furnishing internal secretions are the liver, the thyroid, and the pancreas. It remains possible, however, that any organ, even those not possessing an epithelial structure, such as the muscles, may give off substances to the blood comparable to the internal secretions—a possibility that indicates how indefinite the distinction between the processes of secretion and of general cell-metabolism may become if the analysis is carried sufficiently far. If we consider only the external secretions definition and generalization become much easier, for in these cases the secreting surface is always an epithelial structure which, when it possesses a certain organization, is designated as

a gland. The type upon which these secreting surfaces are constructed is illustrated in Figure 46. The type consists of an

FIG. 46.-Plan of a secreting membrane.

epithelium placed upon a basement membrane, while upon the other side of the membrane are blood-capillaries and lymph-spaces. The secretion is derived ultimately from the blood and is discharged upon the free epithelial surface, which is supposed to communicate with the exterior. The mucous membrane of the alimentary canal from stomach to rectum may be considered,

if we neglect the existence of the villi and crypts, as representing a secreting surface constructed on this type. If we suppose such a membrane to become

URS

FIG. 47. To illustrate the simplest form of a tubular and a racemose or acinous gland.

invaginated to form a tube or a sac possessing a definite lumen (see Fig. 47), we have then what may be designated technically as a gland.

It is obvious that in this case the gland may be a simple pouch, tubular or saccular in shape (Fig. 48), or it may attain a varying degree of complexity by the elongation of the involuted portion and the development of side branches

(Fig. 49). The more complex structures of this character are known sometimes as compound glands, and are further described as tubular, or racemose (saccular), or tubulo-racemose, according as the terminations of the invaginations are tubular, or saccular, or intermediate in shape. As a matter of fact we find the greatest variety in the structure of the glands imbedded in the cutaneous. and mucous surfaces, a variety extending from the simplest form of crypts or tubes to very complicated organs possessing an anatomical independence and definite vascular and nerve-supplies as in the case of the salivary glands or the kidney. In compound glands it is generally assumed that the terminal portions of the tubes alone form the secretions, and these are designated as the the acini or alveoli, while the tubes connecting the alveoli with the exterior are known as the ducts, and it is supposed that their lining epithelium is devoid of secretory activity.

The secretions formed by these glands are as varied in composition as the glands are in structure. If we neglect the case of the so-called reproductive

1 Flemming has called attention to the fact that most of the so-called compound racemose glands, salivary glands, pancreas, etc., do not contain terminal sacs or acini at the ends of the system of ducts; on the contrary, the final secreting portions are cylindrical tubes, and such glands are better designated as compound tubular glands.