salts (acid sodium and acid calcium phosphate), and an average specific gravity of 1017 to 1020. The quantity formed in twenty-four hours is about 1200 to 1700 cubic centimeters. In the very young the amount of urine formed is proportionately much greater than in the adult. The normal urine contains about 3.4 to 4 per cent. of solid matter, of which over half is organic material. Among the important organic constituents of the urine are the following: urea, uric acid, hippuric acid, xanthin, hypoxanthin, guanin, creatinin and aromatic oxy- acids (para-oxyphenyl propionic acid and para-oxyphenyl acetic acid, as simple salts or combined with sulphuric acid); phenol, paracresol, pyrocatechin and hydrochinon, these four substances being combined with sulphuric or glycuronic acid; indican or indoxyl sulphuric acid; skatol sulphuric acid; oxalic acid; sulphocyanides, etc. These and other organic constituents occurring under certain conditions of health or disease in various animals, are of the greatest importance in enabling us to follow the metabolism of the body. Something as to their origin and significance will be found in the section on Nutrition, while their purely chemical relations are described in the section on Chemistry.

Among the inorganic constituents of the urine may be mentioned sodium chloride, sulphates, phosphates of the alkalies and alkaline earths, nitrates, and carbon dioxide gas partly in solution and partly as carbonate. In this section we are concerned only with the general mechanism of the secretion of urine, and in this connection have to consider mainly the water and soluble inorganic salts and the typical nitrogenous excreta, namely, urea and uric acid.

The Secretion of Urine.-The kidneys receive a rich supply of nervefibres, but most histologists have been unable to trace any connection between these fibres and the epithelial cells of the kidney tubules. Berkley has, however, described nerve-fibres passing through the basement membrane and ending between the secretory cells.

The majority of purely physiological experiments upon direct stimulation of the nerves going to the kidney are adverse to the theory of secretory fibres, the marked effects obtained in these experiments being all explicable by the changes produced in the blood-flow through the organ. Two general theories of urinary secretion have been proposed. Ludwig held that the urine is formed by the simple physical processes of filtration and diffusion. In the glomeruli the conditions are most favorable to filtration, and he supposed that in these structures water filtered through from the blood, carrying with it not only the inorganic salts, but also the specific elements (urea) of the secretion. There was thus formed at the beginning of the uriniferous tubules a complete but diluted urine, and in the subsequent passage of this liquid along the convoluted tubes it became concentrated by diffusion with the lymph surrounding the outside of the tubules. So far as the latter part of this theory is concerned it has not been supported by actual experiments; recent histological work (see below) seems to indicate that the epithelial cells of the convoluted tubules have a

The Johns Hopkins Hospital Bulletin, vol. iv., No. 28, p. 1.

distinct secretory function, and that they give material to the secretion rather than take from it.

Bowman's theory of urinary secretion, which has since been vigorously supported and extended by Heidenhain, was based apparently mainly on histological grounds. It assumes that in the glomeruli water and inorganic salts are produced, while the urea and related bodies are eliminated through the activity of the epithelial cells in the convoluted tubes.

Elimination of Urea and Related Bodies.-Numerous facts have been discovered which tend to support the latter part of Bowman's theory-namely, the participation of the cells of the convoluted tubules in the secretion of the specific nitrogenous elements. In birds the main nitrogenous element of the secretion is uric acid instead of urea, and it is possible, owing to the small solubility of the urates, to see them as solid deposits in microscopic sections of the kidney. When the ureters are ligated the deposition of the urates in the kidney may become so great as to give the entire organ a whitish appearance. Nevertheless histological examination of a kidney in this condition shows that the urates are found always in the tubes and never in the Malpighian corpuscles. From this result we may conclude that the uric acid is eliminated through the epithelial cells of the tubes. Heidenhain has shown by a striking series of experiments that the cells of the tubes possess an active secretory power. In these experiments a solution of indigo-carmine was injected into the circulation of a living animal after its spinal cord had been cut to reduce the blood-pressure and therefore the rapidity of the secretion. After a certain interval the kidneys were removed and the indigo-carmine precipitated in situ in the kidney by injecting alcohol into the blood-vessels. It was found that the pigment granules were deposited in the convoluted tubes, but never in the Malpighian corpuscles.

Still further proof of definite secretory functions on the part of the cells of the tubules is given by the results of recent histological work upon the changes in the cells during activity. Van der Stricht,' Disse, and Trambusti 3 describe definite morphological changes in the epithelial cells of the convoluted tubes and ascending loop of Henle which they connect with the functional activity of the cells. The details of the descriptions differ, but the authors agree in finding that the material of the secretion collects in the interior of the cell to form a vesicle which is afterward discharged into the lumen of the cell. According to Disse the inactive cells are small and granular, and toward the lumen show a striated border of minute processes, while the lumen of the tube is relatively wide. As the fluid secretion accumulates in the cells it may be distinguished as a clear vesicular area near the nucleus. The cells enlarge and project toward the lumen, which becomes smaller; the striated border disappears. Finally the swollen cells fill the entire canal, and the liquid secre

2

1 Comptes rendus, 1891, and Travail du Laboratoire d'Histologie de l'Université de Gand, 1892. Referate und Beiträge zur Anatomie und Entwickelungsgeschichte (anatomische Hefte, Merkel and Bonnet, 1893.

Archives italiennes de Biologie, 1898, t. 30, p. 426.

tion is emptied from the cells by filtration. Van der Stricht believes that the vesicles rupture and thus empty into the lumen. In longitudinal sections various stages in the process may be seen scattered along the length of a single tubule.

Secretion of the Water and Salts.-There seems to be no question that the elimination of water together with inorganic salts, and probably still other soluble constituents, takes place chiefly through the glomerular epithelium. This supposition is made in both the general theories that have been mentioned. It has, however, long been a matter of controversy, in this as in other glands, whether the water is produced by simple filtration or whether the glomerular epithelium takes an active part of some character in setting up the stream of water. The problem is perhaps simpler in this case than in the salivary glands, since the direct participation of secretory nerves in the process is excluded. On the filtration theory the quantity of urine should vary directly with the blood-pressure in the glomerulus. This relationship has been accepted as a crucial test of the validity of the filtration theory, and numerous experiments have been made to ascertain whether it invariably exists. Speaking broadly, any general rise of blood-pressure in the aorta will occasion a greater blood-flow and greater pressure in the glomerular vessels provided the kidney arteries themselves are not simultaneously constricted to a sufficient extent to counteract this favorable influence; whereas a general fall of pressure should have the opposite influence both on pressure and velocity of flow. It has been shown experimentally that if the general arterial pressure falls below 40 or 50 millimeters of mercury, as may happen after section of the spinal cord in the cervical region, the secretion of the urine will be greatly slowed, or suspended completely. Constriction of the small arteries in the kidney, whether effected through its proper vaso-constrictor nerves or by partially clamping its arteries, causes a diminution in the secretion and at the same time in all probability a fall of pressure within the glomeruli and a diminution in the total flow of blood. On the other hand, dilatation of the arteries of the kidney, whether produced through its vaso-dilator fibres or by section or inhibition of its constrictor fibres, augments the flow of urine and at the same time probably increases the pressure within the glomerular capillaries, and also the total quantity of blood flowing through them in a unit of time. From these and other experimental facts it is evident that the amount of secretion and the amount of pressure within the glomerular vessels do often vary together, and this relationship has been used to prove that the water of the secretion is obtained by filtration from the blood-plasma. But it will be observed that the quantity of secretion varies not only with the pressure of the blood within the glomeruli, but also with the quantity of blood flowing through them. Heidenhain has insisted that it is this latter factor and not the intracapillary pressure which determines the quantity of water secreted. He believes that the glomerular epithelial cells possess the property of actively secreting water, and that they are not simply passive filters; that the formation, in other words, is not a simple mechanical process, but a more comp'

one depending upon the living structure and properties of the epithelial cells. In support of this view he quotes the fact that partial compression of the renal veins quickly slows or stops altogether the flow of urine. Compression of the veins should raise the pressure within the vessels of the glomeruli, and upon the filtration hypothesis should increase rather than diminish the secretion. It has been shown also that if the renal artery is compressed for a short time so as to completely shut off the blood-flow to the kidney the secretion is not only suspended during the closure of the arteries but for a long time after the circulation is re-established. According to Tiegerstedt, if the renal artery is ligated for only half a minute the activity of the kidney is suspended for three-quarters of an hour. This fact is difficult to understand if the glomerular epithelium is regarded simply as a filtering membrane, but it is explicable upon the hypothesis that the epithelial cells are actively concerned in the production of the water.

The uncertainty as to the mechanism of production of the water and salts renders it difficult to give a theoretical explanation of the action of diuretics. Various saline substances, such as NaCl and KNO3, increase the flow of urine. According to Starling,' these substances increase the bulk of water in the blood by drawing water from the tissues. A condition of hydramic plethora ensues, causing a greater volume of blood in the kidney capillaries and a rise of capillary pressure, conditions that favor greater filtration and account in part for the increased amount of urine. Experiments seem to show, however, that the condition of hydræmic plethora passes off before the increased secretion of urine ceases, so that the diuretic action of the salts is not due to this factor alone. The adherents of the filtration theory assume that in addition the salts cause a vaso-dilatation in the kidney, and thus produce a rise in blood-pressure in the glomeruli. According to the other point of view, these substances may be considered as having a specific stimulating effect upon the glomerular epithelium. So the action of caffein may be referred either to a specific action on the secreting cells or possibly to an indirect effect exerted through the circulation of the kidney. It seems clear that at present we are not justified in asserting more than that the glomeruli control in some way the production of the water and salts of the secretion. The extent of the activity seems to be correlated with the quantity of blood flowing through the glomeruli.

2

It must be borne in mind, however, that some water and probably also some of the inorganic salts are secreted at other parts of the tubule along with the nitrogenous wastes. It is of interest to add that the most important of the abnormal constituents of the urine under pathological conditions, such as the albumin in albuminuria, the hæmoglobin in hæmoglobinuria, and the sugar in glycosuria, seem likewise to escape from the blood into the kidney tubules through the glomerular epithelium.

1Journal of Physiology, 1899, vol. 24, p. 317.

2 See Von Schroeder: Archiv. für exper. Pathologie und Pharmakol., Bd. xxiv. S. 85; and Dreser, Ibid., 1892, Bd. xxix. S. 303.

The normal stimulus to the epithelial cells of the convoluted tubules, using the term convoluted to include the actively secreting parts, seems to be the presence of urea and related substances in the blood (lymph). That the elimination of the urea is not a simple act of diffusion seems to be clearly shown by the fact that its percentage in the blood is much less than in the urine. In some way the urea is selected from the blood and passed into the lumen of the tubule, and although we have microscopic evidence that this process involves active changes in the substance of the cells, there is no adequate theory of the nature of the force which attracts the urea from the surrounding lymph. The whole process must be rapidly effected by the cell, since there is normally no heaping up of urea in the kidney-cells; the material is eliminated into the tubules as quickly as it is received from the blood. The rate of elimination increases normally with the increase in the urea in the blood, as would be expected upon the assumption that the urea itself acts as the physiological stimulus to the epithelial cells.

The Blood-flow through the Kidneys.-It will be seen from the discussion above that, other conditions remaining the same, the secretion of the kidney varies with the quantity of blood flowing through it. It is therefore important at this point to refer briefly to the nature and especially the regulation of the blood-flow through this organ, although the same subject is referred to in connection with the general description of vaso-motor regulation (sce Circulation). It has been shown by Landergren and Tiegerstedt that the kidney is a very vascular organ, at least when it is in strong functional activity such as may be produced by the action of diuretics. They estimate that in a minute's time, under the action of diuretics, an amount of blood flows through the kidney equal to the weight of the organ; this is an amount from four to nineteen times as great as occurs in the average supply of the other organs in the systemic circulation. Taking both kidneys into account, their figures show that (in strong diuresis) 5.6 per cent. of the total quantity of blood sent out of the left heart in a minute may pass through the kidneys, although the combined weight of these organs makes only 0.56 per cent. of that of the body.

The nature of the supply of vaso-motor nerves to the kidney and the conditions which bring them into activity are fairly well known, owing to the useful invention of the oncometer by Roy. This instrument is in principle a plethysmograph especially modified for use upon the kidney of the living animal. It is a kidney-shaped box of thin brass made in two parts, hinged at the back, and with a clasp in front to hold them together. In the interior of the box thin peritoneal membrane is so fastened to each half that a layer of olive oil may be placed between it and the brass walls. There is thus formed in each half a soft pad of oil upon which the kidney rests. When the kidney, freed as far as possible from fat and surrounding connective tissue, but with the blood-vessels and nerves entering at the hilus entirely uninjured, is laid in 1 Skandinavisches Archiv für Physiologie, 1892, Bd. iv. S. 241.

2 See Cohnheim and Roy: Virchow's Archiv, 1883, Bd. xcii. S. 424.