Carbon in Urine. 385 



Thus it appears that the influences which affect the nitrogen elimination 

 on the first day also affect the carbon elimination, and, indeed, in an even 

 more striking manner. 



The ratio of the carbon to the nitrogen presents wide variations, the minimum 

 being observed on the third day of experiment No. 71, namely, 0.660, and the 

 maximum on the second day of experiment No. 77, 1.293. While in experi- 

 ments Nos. 73 and 75, there appears to be a tendency for the ratio to increase 

 as the fast progresses, in experiment No. 77, the ratio remains almost constant 

 after the first day. Practically the same can be stated regarding experiments 

 Nos. 69 and 71. In the series of 2-day experiments, the ratios exhibit rather 

 wide fluctuations on the different days of different experiments, but the averages 

 for the two days of each experiment agree quite closely, the extremes being 

 0.909 and 0.701. There seems to be no general rule regarding the fluctu- 

 ations seen in the ratios for the first and second days of the different experi- 

 ments. Of special interest is the fact that with the same subject, S. A. B., the 

 average ratio varied from 0.725 in experiment No. 73 to 1.193 in experiment 

 No. 77. In fact, in the four experiments with 8. A. B., there is a constancy in 

 the ratio in the first two experiments and a marked increase in experiment No. 

 75, with a still further increase in experiment No. 77. 



If the high ratios obtained in the Middletown experiments are the results of 

 errors in determinations, they may be accounted for in only two ways first, 

 the determination of nitrogen may be low, and second, the determination of 

 carbon may be high. Eegarding the former point, it is hardly to be considered 

 that the results as determined by the Kjeldahl method are too low. In fact, 

 check tests with organic materials of known nitrogen content, which are 

 frequently made, show the method and apparatus to be unusually accurate. 

 Furthermore, in all these cases, the nitrogen was determined in all of the four 

 periods as well as in the composite and the agreement was invariably very 

 satisfactory. 



On the other hand, a critical inspection of the method for determining 

 carbon renders it almost impossible to believe that the carbon results are too 

 high. While Munk (7), in discussing his determination of carbon in Cetti's 

 urine, was of the opinion that the excessive amount of water in the dried urine 

 might have escaped absorption in the first calcium chloride tube of the absorp- 

 tion system and thus be subsequently held in the potash tube, here the urine 

 was completely dried in a vacuum desiccator and sulphuric acid was used for 

 the first tube to absorb the moisture resulting from the combustion. Further- 

 more, it is the general experience in all laboratories that in organic elementary 

 analysis, the determination of carbon is somewhat lower than the theoretical. 



Considering, then, the possibilities of loss of carbon dioxide by the decom- 

 position of urea or ammonium carbonate during the process of drying, it would 

 appear that the results for carbon, if incorrect, are somewhat too low, and 



25 



