Ideally, an experiment such as this should be carried out with every meal controlled. Practically, although we enjoyed the fullest cooperation of the subjects, it became psychologically impossible to control every meal throughout the entire period of 4 months. Consequently, the men were allowed an uncontrolled diet on the Sundays following the biweekly experiment in Boston. They were specifically cautioned, however, to control the diet so far as possible and to make reports of what they ate. At the Thanksgiving recess for a few days it again became necessary to allow the men uncontrolled diet. Again, during the Christmas vacation, a number of the men went to their homes. For several of them this was possibly a farewell visit, as they were going into active war service; under the circumstances we could not urge them to remain in Springfield for the holidays. They were, however, requested to curtail the diet in so far as they could and to endeavor to return to Springfield as nearly as possible with no change in body-weight. More than this we could not conscientiously do. Subsequent inspection of the data returned by these men as the record for the uncontrolled meals on Sundays made us regret extremely that we did not urge more strongly the desirability of complete control throughout every meal, even at the sacrifice of shortening the entire experiment. Some of the Sunday meals were excessive in amount in both nitrogen and calorie content. It would seem almost incredible that such quantities of food could be eaten without extreme discomfort. In fact, a certain amount of discomfort was produced in a majority of cases.

An extreme case of an uncontrolled Sunday diet is that given in table 33 for Gul on January 13. On this day the food ingestion began with the standard laboratory breakfast, followed by several light lunches after the subject left the Laboratory, with a very large dinner and a large supper, although the exact distinction between dinner and supper is somewhat in doubt. Suffice it to say that Gul on this day obtained over 28 grams of nitrogen and 5,264 gross calories of energy.

Table 33. - Sunday Diet (Uncontrolled), Typical Of Large Intake Of Nitrogen And Energy

Gul, Squad A, Jan. 13, 1918.

1 Computed.

Although we had to base our computations not only for this subject but for all others upon their personal statements, with no weighings and only approximate estimates as to the composition of the cooked foods, we have computed the probable nitrogen and energy intake of the subjects on the uncontrolled Sundays, including, also, the laboratory breakfast. This is of such general interest, especially in subsequent discussion, that it is deemed important to present it in considerable detail in table 34. In general the amount of food eaten was very large, the caloric intake being on the average not much different from that obtained for the normal group of 12 men from the Y. M. C. A. College studied from November 20 to 24. (See table 32.) The average nitrogen intake on these uncontrolled Sundays was 16.62 grams nitrogen, and the average gross energy was 3,994 calories. Comparing these amounts with the actual food consumption at or about this period, it can be seen that the excess in the nitrogen and energy intake was large during the break in training on these days.

This excess food has been for us a perplexing problem, the sudden ingestion of a large amount of nitrogen interfering considerably in the intelligent interpretation of our results. The sudden ingestion of a large amount of energy has not been so confusing, for the men invariably took considerable exercise on subsequent days or voluntarily curtailed their diet to offset the excess intake. So far as possible, therefore, we have indicated the probable nitrogen and energy intake of the men in Squad A on the uncontrolled days. Squad B, when on diet, had no uncontrolled meals; hence this problem of uncertainty does not enter into the consideration of their results.

Table 34. - Computed Nitrogen And Energy In Sunday Diets1 (Uncontrolled), Squad A

1 The uncontrolled diets of Sunday, October 14, 1917, were not reported. 2 The average nitrogen intake for these Sundays was 16.62 grams and the average gross energy 3,994 calories.

In the last analysis, however, it should be borne in mind that these men were, in a sense, on controlled diet even during these unrestricted periods, for the actual control was the body-weight. They were thoroughly educated in the belief that an increase in body-weight indicated overeating. Judging from the character of the food in the restricted diet, the men had evidently for the most part a relatively low glycogen storage. We have every reason to believe that many of the increases in weight noted with these men were due to the fact that there was on the free days a liberal storage of glycogen, which carried with it a large amount of water. This addition of water would accentuate the increases in weight, a point which has been thoroughly discussed in the section on body-weight. (See pp. 194 and 224.) On the other hand, the men soon found that by rigid training the weight gain could be rapidly lost and the original desired level speedily obtained.