In our research on the effect on the metabolism of reducing the body-weight by alterations in diet, several plans for reducing the weight of the subjects suggested themselves. The total weight reduction was tentatively set at 10 per cent in the belief that a 10 per cent loss would be of sufficient magnitude to show positively any changes in the metabolism, while a percentage less than this might not be measurable with many of the factors. A greater reduction in weight would prolong the experiment and require greater dietetic control; the amount of discomfort might be taken as roughly proportional to the degree of loss.

Previous experience with fasting men has shown that it is perfectly safe and not productive of great discomfort to fast completely for several days. While the loss of 10 per cent could be secured by complete abstinence from food (judging from our experience with the man fasting for 31 days this loss could be obtained in a period of complete fasting of 14 days), it was recognized that it was impractical to ask a group of men to sacrifice their entire time to a test of this kind and to undergo a complete deprivation of food for 14 days. Consequently it was considered best to produce the loss in weight by the administration of a diet so reduced that there would be each day material drafts upon body fat. It was therefore tentatively proposed that the ingesta should be approximately from 50 to 70 per cent of the actual food requirements.

After the reduction in body-weight of 10 per cent had been reached, the basal ration was then to be supplemented in each case with sufficient energy in food materials to hold the body-weight at the lower level and an attempt made to obtain nitrogen equilibrium as soon as possible. Carbon equilibrium or energy equilibrium would be indicated by constant body-weight over a period of weeks; nitrogen equilibrium would be shown in the usual manner by the balance between the nitrogen in the intake of food and the nitrogen in the feces and urine.

From our experience with fasting men and from the experience of this Laboratory in conjunction with Dr. E. P. Joslin in studies of diabetics with their extreme losses, it appeared perfectly safe to attempt an observation of this kind, since the reduction in body-weight was to be but 10 per cent or slightly more.