MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 295 



the daily urine in Table IN. Tin- urine was also collected during the preliminary period of 4 

 days and during 12 hours following the experiment. Aliquot portions (from one-half to two- 

 thirds) in those 6-hour periods were taken for the preparation of a composite sample of the urine 

 for the day. and in like manner aliquot portions (about one-eighth of the total weight of urine) 

 of the composite sample of the urine for 24 hours were taken for the preparation of a composite 

 sample for the whole period of the experiment. The nitrogen was determined in the urine for 

 each day and in the composite for 4 days of the experiment. The quantities of nitrogen eliminated 

 each day. as determined from the 6-hour periods and from the composite sample for the day. do 

 not always agree exactly. Such discrepancies may be due in part to small errors in the sampling 

 of the composites, in part to errors in the amount of urine measured out for analysis, and in part 

 to errors in the analyses. Samples were measured out for analyses in a calibrated 5-c. c. pipette, 

 and it is possible that differences in the amount delivered from time to time might introduce 

 slight errors in the results. It is assumed, where discrepancies exist, that the values obtained 

 from the 6-hour periods are the more accurate, and these latter are consequently used in the esti- 

 mation of the nitrogen balance. 



It is difficult to evaporate urine to dryness without more or less decomposition of urea to 

 ammonium carbonate, and consequent loss of energ}*. Accordingly, no attempt was made to 

 determine the solid matter in the urine of individual days, but a portion of the composite sample 

 for the experiment was dried according to the manner described on page 239 and the residue used 

 for the determination of carbon, hydrogen, and heat of combustion. The heat of combustion 

 is also determined in the composite samples of the fresh urine each day, as explained above. 

 The precautions taken to avoid error through loss of nitrogen, carbon, and energy during the 

 process of drying of the urine have been described in the publication referred to on page 239. 



The nitrogen is determined in the fresh urine from day to day, but in order to obtain an 

 approximate measure of the amount of carbon and hydrogen in the urine on the successive clays 

 of the experiment some computations are necessary. In making these computations it is assumed 

 that the ratio of nitrogen to carbon, hydrogen or water-free substance will be the same for each 

 individual day as for the 4 days. Thus, the amount of nitrogen in the urine of the first day 

 of the experiment was 17.62 grams, and that for the whole experiment 71.86 grams. The 

 carbon for the whole experiment was found by actual determinations to be 49.15 grams. The com- 

 putations for the amount of carbon in the urine for the first day would then be as follows: 

 71.86 : 49.15 :: 17.62 : x (= 12.05). This method of estimating the carbon and hydrogen in 

 the urine on the different days is manifestly more accurate than would be the case if the total 

 quantity of carbon and hydrogen in the urine for the experiment were divided by the number of 

 days, as is done in estimating the daily excretion through the feces. We know that the quan- 

 tities of nitrogen and carbon in the urine vary from day to day, and have an accurate measure 

 of the variation of the nitrogen, and, since the variation in the nitrogen must involve variations 

 in the amount of carbon united with this nitrogen in the form of urea and allied compounds, it 

 does not seem inappropriate to take the variations in the nitrogen as a measure of the correspond- 

 ing variations in the carbon. Of course, there may be varying quantities of non-nitrogenous 

 compounds in the urine from day to day, which would render the above method of estimation 

 more or less inaccurate. It is probable, however, that the variations in nitrogen give the fairest 

 measure of the variations in carbon and hydrogen. As a matter of fact, it has been found that 

 the heat of combustion varies in close relation to the nitrogen. Of course, the results for the 

 experiment as a whole are not affected by the subdivisions of the amounts for the individual days. 



