Whatever doubts may have existed in the layman's mind with regard to the advisability of reducing the input of liquids in the diseases we have just mentioned, he has never had any difficulty in believing that such a procedure is absolutely demanded in cases of obesity, and medical men in all ages have encouraged the perpetuation of this idea. Methods based upon this conception have been in existence from time immemorial, but it was in 1884 that Dancel, a French veterinary surgeon, placed them upon a scientific basis by his observations that the ingestion of much water or watery foods had a powerful influence in causing great abdominal development in horses. Mere reduction in the quantity of fluid consumed caused a rapid diminution of weight and an increase of vigour, whereas an additional allowance of water to thin horses soon resulted in augmented weight. He therefore counselled a similar system for human beings, enforcing at the same time abstinence from fatty and farinaceous foods, frequent aperients, and outdoor exercise.

The method of treating obesity associated with Banting's name was modelled on the same lines. The systems of Ebstein and Oertel likewise reduce the consumption of fluids even in normal quantities as a favouring agency in the development of adipose tissue, but the former recommends an increased ration of fat with a minimum of carbohydrates and proteins, and the latter a large allowance of proteins, a diminished quantity of fats, and the almost total exclusion of carbohydrates.

Oertel in addition recommended that the food should be partaken of separately from liquids, contending that in this manner the blood was depleted of fluid by the demands made for the secretion of gastric juice. He was satisfied that, by the measures he instituted, not only was there an increased concentration of the blood and a dehydration of the tissues, but that deposited fat was actually consumed by a process of oxidation.

Another advocate of a similar system to Oertel's was Schweninger, whose explanation of the reduction of weight effected by his treatment was that the thirst engendered by the restriction of liquid caused the body to manufacture water from its own fatty tissues by splitting the fat into simpler molecules.

In each of these systems, as in all others with a similar object, in addition to the deprivation of water, other measures, such as hard exercise, massage, and a reduction in the amount of solid food, were laid under contribution, so that it is difficult to attribute any loss of flesh to the drink restriction alone. Further acquaintance with the subject also elucidates the fact that a great reduction in the allowance of fluids always results in an impairment of the appetite, so that the quantity of food ingested is necessarily diminished.

It is to be noted that in obesity nitrogen equilibrium is usually maintained without a corresponding maintenance of carbon equilibrium. The aim, therefore, in dietetic treatment should always be to maintain nitrogen equilibrium, and to restrict the carbonaceous income, so that the output of carbon may be increased at the expense of the adipose tissue. In this connection it is important to remember that dry starch contains 44.2 per cent, of carbon, whilst dry fat contains 76.5 per cent. of carbon. It is essential, however, not to administer too much protein food, as although protein metabolism will gradually increase and nitrogenous equilibrium be established at continually higher levels, the amount of fat burned correspondingly diminishes, and there is even a risk that some of the carbonaceous moiety of the ingested protein may be retained in the body. For this reason Moritz has recommended an exclusive diet of milk in the treatment of obesity.

An interesting dissertation on the study of thirsting in animals is contributed by von Noorden, and serves as an introduction to observations made on his own patients. He mentions that Giirber, in a thirsting frog, found an increase of the erythrocytes, while in a thirsting dog von Westendorf noted that both the osmotic pressure and the specific gravity of the blood were increased, and that, whereas venesection is in such circumstances followed by a lowering of the specific gravity of the blood, this did not occur, because the dehydrated tissues were unable to yield more water to the blood. Landauer demonstrated on dogs that destruction of albumin is increased by thirsting, and this was corroborated by Straub, who likewise showed that in addition to the increase of urinary nitrogen, the excretion of P2O5 was also increased. Most important of all, however, as bearing upon the oxidation of fatty tissue, the last mentioned observer found that the excretion of carbon dioxide via the lungs and skin was not appreciably augmented.

Dennig verified most of these results, and also noted a gradual daily diminution of the insensible perspiration. He arrived at the conclusion that the end-products of the decomposition of albumins were retained until they were washed out of the tissues by subsequent water-drinking, and that they were formed in greater abundance in subjects who were lean than in those who were fat, It must be borne in mind that this albumin destruction only occurs when drink restriction is carried to great excess, i.e., when the amount of liquid consumed is limited to from 300 c.c. to 500 c.c, and that even Oettel never reduced the quantity below 1 litre, inclusive of the water contained in the food. But it is risky to venture to such extremes, even as this, for despite the careful superintendence of trained medical men, a system permitting the ingestion of less than 1,000 c.c. of liquids, excluding the water contained in the food, is not free from danger, because with such a regimen cases of renal calculus, colic, gout, albuminuria, and even fatal oedema, are on record.

Von Noorden himself could discover no evidence that there was any increase in the oxidation of fat, as the total oxygen metabolism of the organism is not augmented during thirsting. The first effect of continued thirsting, even in a normal case, is a loss of that variable quantity of fluid always contained in the tissues, because of the consumption of water, so that its withdrawal will effect the loss of this fluid. In other words, an actual dehydration of the tissues has taken place, and although there is an unquestionable loss of weight by the means described, there is really no loss of flesh, because the moment water-drinking is resumed in most cases the body-weight again increases. Loss of fat, therefore, is not a primary consequence of fasting, as asserted by Oertel and Schweninger.

It is very different, however, with many cases of obesity, because, especially in the older subjects, one has to reckon with the retention of water due to the accumulation of chlorides, so that many pounds of what amounts to cedematous fluid may thus be excreted in them. There can be little harm in losing this at any time or in any case, but any attempt at further reduction in the weight by possible increased oxidation of the fatty tissues must be undertaken with great caution, especially in the aged.

Von Noorden, whilst noting the fact that the appetite is impaired by limiting the drinking of fluids, was careful to insist that this in no way interfered with the digestive capacity, which is not at all lessened by the dry diet, and he expressed his belief that a normal or even increased quantity of fluid might be consumed between meals without diminishing the loss of weight effected by the limited diet. Other observers have even contended that loss of weight is promoted by this procedure.

In conclusion, therefore, one may recapitulate the good results of the restriction of liquids briefly as follows: relief to an overburdened stomach and cardio-vascular apparatus, dehydration of the tissues and blood, with reduction of the body-weight, diminution of appetite, and, when carried to the danger limit, increased destruction of albumin, but no irrefutable evidence of actual combustion of deposited fat.

As a therapeutical procedure, it is indicated in chlorosis and serious haemorrhage as well as in the conditions already mentioned, and in many such cases, as also in some of severe ascites, has been found of decided value. When, on the other hand, the body has lost an excessive quantity of its tissue and circulatory fluids, prompt measures must be taken to replace them, either by transfusion of blood or the infinitely simpler introduction of normal saline solution, either into the rectum or the subcutaneous tissues (hypodermoclysis).