The average nitrogen intake in the food from October 1 to 10 inclusive (table 40) was 12.71 grams, although on 3 days high values were observed. The average of the first 3 days is 15.47 grams; the average of the last 7 days is 11.53 grams. The irregularity of the intake of nitrogen, even after the curtailment in the diet began, is shown by the fact that on October 7 the average nitrogen intake was 9.33 grams and on the next day it was 4.5 grams larger. Notwithstanding these relatively large variations in the nitrogen intake, the average urinary nitrogen (table 41) remained singularly constant throughout this period.

The nitrogen in the food intake in the period from December 5 to 18 is shown in table 42. Considerable variation is also found in these amounts, the daily averages ranging from the low value of 8 grams on December 7 to a maximum of 13.52 grams on December 17. The minimum average of 8 grams is 3 grams lower than the general average of 10.99 grams, and the maximum average is 2.5 grams higher. Turning again to the urinary output during this period (table 43), it is perhaps surprising that we find a greater degree of uniformity obtaining, for while there are variations, they do not approximate in size the variations found in the food intake. The fact that the average nitrogen intake in the period of December 5 to December 18 is 10.99 grams and the average nitrogen output in the urine is 10.92 grams shows that these men were not in nitrogen equilibrium, for no allowance has been made for fecal nitrogen or for losses through the skin. So while these subjects were in weight equilibrium, they were not as yet in nitrogen equilibrium. Mention is made of this here to emphasize the fact that while nitrogen equilibrium may suggest weight equilibrium, the reverse is by no means true. (See page 353).

The inspection of the fragmentary evidence of the nitrogen intake and output, as shown in these sections of the food and urine tables for the two periods of the experiment, makes it necessary to present in abstract form the average nitrogen in the food and urine per man per day for the entire experimental period, for the relationship between them is an important one. Tables 40 to 43, inclusive, give not only the average values, but likewise the values for each man, and permit a study of the variations found between individuals and between the results for the same individual from day to day. Of greater importance, however, is the picture presented by the squad as a whole. This is shown in table 44, in which is recorded the total nitrogen in the food and in the urine for the entire series, expressed as the average per man per day, for Squad A. An examination of the nitrogen in the urine shows, on the whole, a constant nitrogen output. It is rarely that differences of more than 2 grams are noted from day to day. Values less than 10 grams do not appear until November 9-10. An absolute minimum of 7.90 grams is noted on January 15-16. There is no clear picture of a definite appreciable reduction in the nitrogen output per man per day until approximately the middle of January, but in the latter part of January low figures appear with considerable frequency. In general one may state that in spite of the great alterations in the diet made with these men, the nitrogen excretion on the average was but little affected. If we average the low-diet values by months, we would find that the average excretion for the month of October would be 11.47 grams, for the month of November 10.60 grams, for the 3 weeks in December 10.87 grams, and for the month of January 10.34 grams.

The most important aim in the reduction of diet was to secure a decrease in caloric content and no attempt was made to secure an especially low level of protein. In fact, in certain instances, at least, the protein was measurably increased, since it was found that the low diet was resulting in a somewhat heavy draft upon body nitrogen. The nitrogen intake is also shown in table 44. After the first 3 days in October, the reduction in diet with special reference to the caloric content began. The irregularities of the nitrogen intake, as pointed out in the discussion of the two periods in tables 40 and 42, are likewise found throughout the entire period, practically no two days being alike. Occasionally, the variations are somewhat large as, for example, from October 17-18 to October 18-19, when there was an increase of nearly 4.5 grams. This again illustrates the fact that the protein was not controlled. With such variations in nitrogen intake, and particularly with the low nitrogen intake during a considerable period of the time, it is surprising that the urinary nitrogen was not more affected. Thus, for a period from October 31 to November 12, inclusive, the daily values for the nitrogen intake are all less than 10 grams; in fact most of them are less than 9 grams. Yet the nitrogen excretion in the urine remained at a fairly high level throughout the entire time, this being, of course, positive proof of a loss on the part of the squad as a whole of a considerable amount of body nitrogen.

The most important fact to be drawn from this table, therefore, is that in spite of material reductions in food nitrogen and relatively large variations in the intake of nitrogen, the urinary nitrogen output remained singularly constant for the squad as a whole throughout practically the entire period, aside from the general slight change of 1 gram in level in the periods indicated above. The nitrogen in urine in the last week in October is essentially on the same level as the nitrogen in urine in the last week in January.

Table 44. - Total Nitrogen In Food And Urine1 - Squad A

[Average per man per day].

1 The values in this table, except for January 13-14, represent only days when the diet was controlled and do not include the uncontrolled Sundays.