216 Messrs. G. Graham and E. P. Poulton. The Alleged 



In the examination of any urine which contains aceto-acetic acid, and 

 which, is thought to contain creatine, two estimations are required, namely, 

 that of the total creatinine + creatine by the original Folin method, and that 

 of the creatinine alone by the method given in this paper. 



III. The Alleged Excretion of Creatine on a Carbohydrate-free Diet. 



It is well known that the consumption of a diet containing no carbo- 

 hydrates produces acidosis, with the excretion of /3-oxybutyric acid, aceto-acetic 

 acid, and acetone. The aceto-acetic acid will cause an error in estimating 

 creatinine and creatine, and must be removed to get accurate results. 



We have performed three diet experiments on three separate individuals, 

 and have investigated the creatinine and creatine excretion, taking this 

 precaution. The experiments were begun about 12 hours after the last 

 ordinary meal in Experiments I and III, and six hours after, in Experiment II. 

 Tables VII, VIII, and IX show the various determinations made. The 

 creatinine was first of all estimated directly by Eolin's method without 

 removing the aceto-acetic acid, and the results are referred to as " apparent 

 creatinine." The true creatinine was then obtained after removing the 

 aceto-acetic acid by the distillation method. The creatinine + creatine was 

 determined by heating the urine for three and a half hours on the water- 

 bath with hydrochloric acid. By subtracting the apparent creatinine values 

 from the creatinine + creatine output, the apparent creatine was obtained, 

 and, by subtracting the true creatinine from the creatinine + creatine, the 

 true creatine output was obtained. Duplicate determinations were performed 

 in each case, and in each determination the mean of six to eight readings of 

 the scale was taken. 



In Experiment I (E. P. P.), cream alone was eaten on the first two days ; 

 on the third day protein was added to the diet. The calorie value of the 

 diet was low. The effect of the withdrawal of carbohydrates was shown by 

 the prompt appearance of aceto-acetic acid in the urine. On the first day 

 the nitroprusside reaction (Eothera's) was faint, but on the second and third 

 days it was well marked, and 0872 and - 874 grm. of aceto-acetic acid were 

 excreted. On the first day the apparent creatinine was T82 grm., while 

 the true creatinine and creatinine + creatine was 1*78 and T80 respectively, 

 so that no creatine was excreted in the urine, as the difference is within the 

 limits of experimental error. On the 2nd day the apparent creatinine was 

 diminished to l - 58 grm., while the true creatinine and creatinine + creatine 

 were practically the same as on the previous day, i.e., 1-8 L and 1"82 grm. The 

 apparent creatine was, therefore, 024 grm., while no true creatine was excreted. 

 On the third day the apparent creatinine had fallen to 142 grm., while the 



