ation between liver and kidney as excretory organs; an association that we may at once characterise as formation by the liver, separation by the kidney. The chief items in the argument are the following:-1. Bilirubin is made by the liver; urobilin is discharged by the kidney. 2. Urea is made by the liver; it is discharged by the kidney. 3. Glycin is made by the liver; hippuric acid (which is glycin + benzoic acid) is discharged by the kidney. 4. Uric acid is made by the liver, discharged by the kidney. 5. Sugar, made by the liver, may be discharged by the kidney.

The data upon which these several dicta are based are as follows:-1. Urobilin is chemically identical with reduced bilirubin, or hydrobilirubin. 2. A freshly-excised liver, submitted to artificial circulation with blood containing ammonium carbonate, yields urea found in the outcoming blood. In acute atrophy of the liver, with abolition of hepatic function as its consequence, leucin and tyrosin are found in the urine in place of urea. 3. Benzoic acid taken by the mouth appears as hippuric acid in the urine, but after excision of the liver it passes unchanged as benzoic acid. 4. After excision of the liver on geese, uric acid ceases to be excreted (Minkowski). 5. Excessive sugar-formation by the liver is accompanied by discharge of sugar by the kidney.

We must not, however, take the association between liver and kidney-viz., formation by the former, separation by the latterin too absolute a sense, as implying a strict division of labour. Urea and hippuric acid are not formed by the liver exclusively; other organs the kidney in particular-submitted to artificial circulation, are capable of forming these substances, this being especially the case as regards the formation of hippuric acid by the kidney. Blood containing benzoic acid and glycin, made to pass through a recently-excised kidney, or left in contact with a minced fresh kidney, is after a few hours found to contain hippuric acid. Thus the living kidney can effect the conjugation of glycin (formed by the liver) with benzoic acid (formed or set free in the course of digestion). The chief evident function of the kidney is the separation of formed materials, but it certainly also contributes to the actual formation of the materials that it separates.

On the other hand, we must recognise that the liver, although it is pre-eminently formative of excreta, such as urea and (in birds) uric acid, is not thus formative to the entire exclusion of other organs and tissues. We have already seen an illustration of this in the case of bile-pigment; although no doubt the principal normal relation is formation of bilirubin by the liver,

and separation of urobilin by the kidney, yet we must admit that, failing the liver, urobilin may take origin otherwise in the body.

Diabetes. The essential phenomena of diabetes are the passage of sugar into the urine, and an excessive discharge of water. The first event is the characteristic of diabetes mellitus, and is found to be associated with the presence of an excessive amount of sugar in the blood; the second event is the characteristic of diabetes insipidus. No conclusive proof has yet been given to show that the urine normally contains any sugar, and it is certain that normal urine does not contain it in considerable amount; in a case of diabetes mellitus the urine may contain as much as 5 per cent. The average percentage of sugar in normal blood is only 1; the percentage of sugar in diabetic blood may reach 5. Comparing these two amounts, 5 per 100 in blood, 5 per 100 in urine, we recognise that mere diffusion will not account for the excretion of sugar, but that, as in the case of urea, an activity of renal epithelium must be invoked. The excess of sugar in the blood, disposed of by the kidney, is attributable to two causes :- -(1) abnormal action of the liver, (2) deficient consumption by the tissues; and it is probable that both causes contribute to the result. The abnormal action of the liver can be regarded in two different lights: either the liver action is defective, sugar coming from the intestine is not stored as glycogen; or the liver action is excessive, more sugar than usual is made. The second alternative is the more probable; diabetes is not produced by diseases in which liver action is certainly defective, and it may persist, though in diminished degree, when all starch and sugar are withheld from the dietary.

The relation between excessive action of the liver and diabetes is further borne out by the effects that follow puncture of the spinal bulb, and several other operations. Rabbits shortly after a puncture of the fourth ventricle become diabetic in so far as they pass sugar in the urine. Frogs similarly treated, especially at the beginning of winter, are similarly affected. This diabetic state is, however, temporary, and an essential condition of its appearance is that the liver shall, at the time of puncture, contain a large store of glycogen. It appears, therefore, that puncture diabetes is due to an accelerated conversion of liver-glycogen into sugar. This view is borne out by further observations; it has been found that several other operations will produce the same result, e.g. section of the cervical sympathetic or of the spinal cord, destruction of the

cervical or of the stellate ganglia, excitation of the central end of the vagus, or indeed of any afferent nerve, ether, morphia, apnoea, and, finally, the injection of arterial blood into the portal vein; the common effect in all these cases, viz. sugar in the urine, being attributable to one common cause, viz. to hyperæmia of the liver. It should be added, however, that the sugar reaction may be simulated by the allied carbohydrate, glycuronic acid, CHO,, which, according to Ashdown, is the substance giving the apparent sugar reaction in the urine of chloroform narcosis. Puncture of the bulb may also give rise to diabetes insipidus; this effect, the mechanism of which is more obscure, is attributable to vasomotor disturbance of the kidney rather than of the liver.

Pancreatic diabetes.-Recent experiments have shown that the pancreas bears some peculiar relation to the sugar function; dogs are rendered diabetic by complete removal of the pancreas, the diabetes so produced remaining permanent until death (Minkowski and v. Mering), contrasting in this respect with the temporary character of puncture diabetes, as well as in the fact that it is not a nerve effect; the nature of the relation between diabetes and loss of the pancreas is as yet entirely obscure; it is comparable with the equally obscure action of the thyroid gland, i.e. in both cases removal of a gland causes a peculiar malnutrition, characterised by excess of sugar if the pancreas is removed, by excess of a mucinoid substance (myxedema) if the thyroid is removed; in both cases, therefore, the gland in question must be of importance to healthy nutrition, possibly by virtue of some as yet unknown process of internal secretion.' This is not the place to discuss at length the possibilities that are suggested by observations of this order; it is enough for us to recognise that a relation-however obscure really subsists between the special functions of particular organs and the general nutrition of the body, and that remedial measures may be looked for by following up this clue; successful results are, in fact, reported subsequent to injections of testicular, pancreatic, and thyroid extracts in cases where the function of the respective glands was presumably deficient; experimentally, dogs rendered diabetic by removal of the pancreas have had life prolonged by injection of pancreatic extract, and by pancreatic grafting.

These are modern examples of the conception long ago formed by Treviranus, to the effect that the several parts of the body are mutually interdependent, the waste of one organ serving as the raw material to another organ, and even the organs themselves being conceivably 'excretory products' in relation to the remainder of the organism. The development of the mammary gland coincident with the presence of a foetus, the arrested development of male characteristics in consequence

of castration, may also be pointed to in token of the obscure interorganic relations which play a part in the internal economy of the processes of nutrition.

Addison's disease,' of which the most obvious sign is a bronzed skin due to excessive pigmentation, is perhaps another case in point; the pigmentation is one among other signs of a malnutrition which, in some cases at least, is attributable to diseased suprarenal glands.

Influence of renal action upon tissue-disintegration.-Recent observations by Bradford may be alluded to here in further illustration of the evident fact that our precise but rough knowledge of physiological mechanics and chemistry includes only a small portion of the actual phenomena occurring in the living organism, and that we must keep our minds open to the idea that the several parts of the body are interdependent. The facts discovered by Bradford are briefly as follows: A large wedge of one kidney, amounting to between and its total weight, is excised, the surfaces of the remaining portion are fastened together; when the animal (dog) has recovered, the entire second kidney is removed, thus leaving only to a single kidney on duty; the consequences of this changed condition are: acute emaciation (e.g. a fall of body-weight amounting to 37 per cent. in fourteen days), in spite of normal appetite and diet; greatly increased discharge of water, and, in less marked degree, increased discharge of urea. These effects seem to suggest that the renal function is not merely to remove degradation products, but also to produce some as yet unknown body or bodies, adjuvant of tissue-integration, or obstructive of tissue-disintegration. We have already recognised that the hepatic gland is concerned in the terminal act of metabolism as well as in its initial phenomena; we are now led to admit that the renal gland plays a part somewhere in the course of metabolism as well as at its final event.

Excretory action of the skin. SWEAT. The skin, by virtue of its glands, is an excretory organ. The substance excreted is almost exclusively water, the amount discharged varying to such an extent with temperature as to cause the urine to be pale and copious in winter, dark and scanty in summer. Sweat, when perfectly fresh, is alkaline, but this reaction is very quickly reversed owing to fatty acids produced in its decomposition.

The connection between the cutaneous and renal channels of excretion is further illustrated by the appearance of urea in the sweat; even normal sweat is said to contain a trace of urea, and the sweat of a uræmic patient may contain the substance in considerable quantity.

The amount of evaporation taking place from the surface of the body depends upon the physiological activity of the sweat

glands, upon the movements of respiration, and upon the physical state of the atmosphere as regards temperature and pressure, saturation with moisture, and movement of the air. Low temperature, saturation, and stillness are unfavourable, the reverse conditions are favourable, to evaporation. Under ordinary conditions air contains about 1 vol. per cent. of water vapour. Air at normal pressure is saturated

e.g. we have seen that the expired air is saturated at 37°, i.e. it contains about 7 per cent. water vapour; if into an atmosphere at 1 per cent. saturation a man expires per diem 10,000 litres air, he gives off 600 litres water vapour (= 480 grammes).

From the skin the water is given off partly by 'insensible perspiration,' partly as visible drops of sweat, and the latter partly evaporates from the surface, partly trickles off or is absorbed by clothing. The daily amount of perspiration may be put at 900 litres of water vapour (=720 grammes). Its rate of discharge in the form of insensible perspiration may be gauged by a hygrometer, consisting of a shallow glass vessel coated with calcium chloride, which is weighed before and after it has been left inverted over the skin for a given period; the increase of weight indicates the amount of water vapour absorbed by the calcium chloride. Tested in this way the palm of the hand is found to give off moisture more rapidly than any other equal area of skin. The constant transpiration of water from the skin and lungs, and the greater weight of CO, output as compared with O, income, cause an appreciable fall of body-weight, i.e. nearly 1 gramme per minute, or 2 ounces per hour.

2

The secretion of sweat is under the control of sympathetic nerve-fibres which leave the spinal cord by the anterior roots and run a similar course to that of vasomotor nerves. If the peri

pheral end of the sciatic nerve of a young cat be excited, beads of sweat break out from the pad, and the effect can still be obtained after arrest of the circulation or after amputation of the limb; thus proving the separate existence of sudo-motor nerves (Luchsinger). As in the case of the salivary gland, pilocarpin stimulates the secretion and atropin stops it. Langley and Sher