EXPERIMENTAL,. 19 



when heated to 43 , flocculent at 56 , and no more precipitation on further 

 heating, showing that either a higher-coagulating protein was thus removed 

 or that coagulation of the albumin was more complete in the strong saline 

 solution at the lower temperature. This dialyzed solution likewise gave 

 a considerable precipitate with nitric acid. On heating, a part remained 

 insoluble, and on filtering this off, the filtrate gave a precipitate on cooling, 

 which dissolved again on heating and reappeared as often as the solution 

 was cooled. The filtrate from the salt and acid precipitate did not give 

 this reaction, but the solution of the precipitate in water gave it very 

 strongly. 



This reaction is characteristic of some proteoses, and shows that the salt 

 and acid precipitate contains a proteose together with the albumins. This 

 proteose is likewise precipitated by saturating the extract with salt, for 

 on dissolving the precipitate so produced and separating the albumin con- 

 tained in it by coagulation the filtrate gave a strong red biuret reaction, and 

 a precipitate with nitric acid, which dissolved on warming and precipitated 

 again on cooling. The filtrate from the precipitate caused by saturation 

 with salt gives no reaction with nitric acid, showing that the proteose is thus 

 completely precipitated. 



In order to be sure that the coagulable protein, which was apparently an 

 albumin, was not a globulin held in solution by the small amount of salts 

 contained in the river- water used for dialysis, as was suggested by its partial 

 precipitation by saturation with sodium chloride, the following experiment 

 was tried : 250 cc. of a strong aqueous extract of winter-wheat meal was 

 dialyzed in running distilled water for 48 hours. A small precipitate was 

 then filtered off, the clear solution returned to the dialyzer, and the process 

 continued for five days longer. No more substance separated. The entire 

 solution, which was still found to coagulate at 54 , was then evaporated to 

 dryness, the considerable protein residue burned off, and the total mineral 

 matter found to weigh only 0.0008 gram, thus proving the protein to be an 

 albumin. 



We are able, then, to recognize two distinct protein substances soluble in 

 pure water, namely, a coagulable albumin and a proteose. As it was found 

 that the protein removed from the flour by treatment with alcohol was to a 

 slight extent soluble in pure water, it might be thought that one of these 

 bodies was identical with this alcohol-soluble protein. Its identity with the 

 albumin is excluded by the fact that the latter is precipitated by heat, and 

 with the proteose by the fact that the alcohol-soluble proteose gives a pre- 

 cipitate with hydrochloric acid when it is dissolved in distilled water and also 

 with a little sodium chloride, which the proteose does not. 



