208 ReEeporv’ OF THE CHEMIST OF THE 
amido compounds, so that the ratio of monoamido to diamido 
compounds was as 1 to 0.25 or less; in other words, the mono- 
amido compounds were greatly in excess of the diamido com- 
pounds, or just the reverse of what we find to be the case in 
the results embodied in the table above, based on the assump- 
tion that the nitrogen in the tannin-salt filtrate represents the 
total amido compounds and nothing more. The most obvious 
and rational explanation of this discrepancy, observed in the 
ratio of monoamido to diamido compounds, is that it is wrong 
to assume that the tannin-salt filtrate contains only amido 
compounds and not any peptones. Withdrawing that assump- 
tion, then, and allowing that the nitrogen in the tannin-salt 
filtrate represents some peptones as well as the amido com- 
pounds, how can we tell in this particular case the true amount 
of amido compounds in the cheese. Unquestionably, the results 
with phosphotungstic acid more nearly represent the truth in 
regard to the amido compounds, because, under the conditions 
of the experiment, we should expect very small amounts of 
amido compounds, if any; and, in this particular case, .the 
amounts are so small as practically to indicate the absence of 
amido bodies altogether. From this it may be seen that it is 
possible for the tannin-salt reagent to give results that are 
decidedly misleading. 
(3) By bromine with hydrochloric acid.— To the filtrate from 
the zine sulphate precipitate in 7, we add 2 or 3 drops of strong 
hydrochloric acid and then bromine until the liquid becomes 
saturated and there remains after vigorous agitation an undis- 
solved globule of bromine amounting to 0.5 cc. to 1 cc. This 
is allowed to stand over night. The precipitate is then filtered 
and washed with bromine-saturated water. The nitrogen in 
the precipitate is then determined by the Kjeldah] method and 
is called nitrogen in the form of peptones, the filtrate contain- 
ing the amido-acid compounds and ammonia. 
The use of chlorine by Rideal and Stewart" in precipitating 
proteids suggested to Allen and Searle!’ the use of bromine. 
They reported that bromine quantitatively precipitates the prod- 
17 Analyst 22: 228 (1897). 
84 nalyst 22: 259 (1897). 

