230 ON THE EXCRETION OF 



with considerable regularity. The nitrogen was determined 

 in the urine and diet by the Kjeldahl method ; sugar, when 

 present, was estimated by titration with Fehling's or Purdy's 

 solution,* and kynurenic acid was found by the method of 

 Capaldi,f which has proved very satisfactory. The product 

 thus obtained always responded to Jaffa's test J and was crys- 

 talline : in a few urines a very small quantity of an amorphous 

 substance was precipitated, which failed to give the character- 

 istic reaction. As an immediate test for kynurenic acid in 

 the urine the bromine water reaction, first recommended by 

 Baumann, was frequently found useful. Bromine, as is well 

 known, usually gives an insoluble yellow precipitate when 

 added to dog's urine, the composition of the precipitate de- 

 pending upon the presence of phenol bodies, indol, or kynuf- 

 enic acid. With a little experience it becomes easy to make 

 use of the reaction in judging the relative amounts of kynur- 

 enic acid, since the latter ordinarily composes (as tetrabrom- 

 kynurin ||) by far the greater part of the precipitate formed. 



Experiments on dogs. For these experiments commercial 

 cracker-dust containing as an average 1.46 per cent nitrogen 

 was obtained in large quantity and kept in glass-stoppered 

 bottles. This constituted the carbohydrate food fed. The 

 fat used was a good quality of lard practically free from 

 nitrogen. The other food-stuffs used will be referred to in 

 the protocols. 



The following experiments demonstrate the formation of 

 kynurenic acid after the ingestion of various proteids of both 

 vegetable and animal origin. The "dog biscuit" used was 



* Cf . J. Bishop Tingle, American Chemical Journal, 1898, xx, p. 126. 



t Capaldi, Zeitschr. f. physiol. Chemie, 1897, xxiii, p. 92. Solomin (ibid., 

 p. 498 note) recovered by this method 99 per cent of 0.210 gram kynurenic 

 acid added to urine. The following figures show average duplicates obtained 

 by us from a dog's urine containing small quantities : (a) 0.0852 gram, 

 (6) 0.0872 gram. 



J Jaffa", Zeitschr. f. physiol. Chemie, 1883, vii, p. 399. 



Baumann, Zeitschr. f. physiol. Chemie, 1877, i, p. 62. 



|| Brieger, Zeitschr. f. physiol. Chemie, 1881, iv, p. 89. 



