New York AGRICULTURAL EXPERIMENT STATION. 223 
(3) The xanthoproteic reaction— The presence of dissolved 
casein in the filtrate is readily shown by the xanthoproteic reaction. 
For the purpose of making our results comparable, this test was 
applied in the following manner: We boiled 10 cc. of the filtrate 
to be examined with 10 cc. of nitric acid (sp. gr. 1.42) and then 
added about 12.5 cc. (about 2 cc. in excess of neutrality) of a con- 
centrated solution of sodium hydroxide. A blank test was used 
for comparison in cases where the result of the reaction appeared 
in doubt. 
(4) Deviation between results by titration and conductivity 
methods.— When any casein dissolves, titration of the filtrate shows 
more acid than does the measurement of conductivity, because 
titration of the filtrate with alkali measures not only the acid not 
taken up by casein but also that combined with dissolved proteid, 
and, in addition, the acidity of the dissolved proteid itself, which 
in some cases may neutralize even more alkali than the free acid. 
On the other hand, the error caused in conductivity results by par- 
tial solution of proteid can, at most, amount to only a relatively 
small percentage, because the fixation of free hydrochloric acid, for 
instance, to form a soluble proteid compound, in which the light 
and rapidly moving Ht ion of hydrochloric acid is replaced by the 
enormously complex and sluggish proteid radicle, results in the 
loss of most of the conducting power of the acid. Thus, the ad- 
dition of sufficient egg albumin to combine with all the acid in 
N-20 HCl solution, forming soluble albumin hydrochloride, results 
in a decrease of 83 per ct. in the conductivity of the solution.' 
Similarly the combination of hydrochloric acid with globulin results 
in a decrease of 76 per ct. in conductivity.2 When solution of 
casein in acid occurs, the soluble acid proteid has still, undoubtedly, 
a low conductivity of its own, as in the two cases just mentioned, 
and this will render somewhat too low the final results that are 
calculated on the assumption that the combined acid forms only 
a compound which is, like the insoluble one, entirely non-conduct- 
ing. The error will, however, be one of.a small percentage, par- 
ticularly when only a fraction of the casein dissolves and the greater 
part of the acid taken up goes to form the insoluble substance. 
*Sjoqvist. Skand. Archiv. Physiol. 5:344. 1895. 
*Hardy. Jour. Physiol., 33:273. 1905. 
