EXPERIMENTAL. 29 



tion with salt, while others are not. As we have shown, these different prop- 

 erties are the result of changes caused by varying the conditions under which 

 the protein exists in the extract, and depend chiefly on the degree of acidity 

 of the extract, whereby the numbers and kinds of acid molecules that com- 

 bine with the protein molecule are altered. 



Whatever may be the true cause of these changes, it is evident from the 

 results here described that the distinctions heretofore made between globulin 

 and albumin, myosin, and vitellin, etc., have very little value as a basis 

 for classifying protein substances. This explains the difference between 

 O'Brien's classification of leucosin as a my osin-like globulin, to which refer- 

 ence was made at the beginning of this paper, and our designation of it as 

 an albumin, because of the ready solubility in water and coagulability by heat 

 of the preparations which we had made. 



Thus preparation 22, weighing 9.17 grams, was insoluble in water and in 

 salt solution, and was not a precipitate of globulin, since in the filtrate from 

 which it had separated on dialysis only 0.87 gram of coagulable albumin 

 was found instead of 9.5 grams, as usually found by direct coagulation of 

 the aqueous extracts ; moreover, the analysis indicates that it is a compound 

 of leucosin, with 20 per cent of nucleic acid. 



On the preceding pages it was shown that a small part of the precipitate, 

 produced by saturating the aqueous extract with sodium chloride, is soluble 

 in dilute salt solution, and can be precipitated from this solution by dialysis, 

 as a globulin-like substance, readily soluble again in salt solution. This 

 globulin-like substance contains little or no nucleic acid, and has very nearly 

 the same elementary composition as leucosin, of which it is probably a com- 

 pound with a small proportion of some body of low molecular weight. 



It is plain from these facts that O'Brien's myosin contains the same pro- 

 tein substance as my leucosin. 



O'Brien's " albumin," coagulating at 75 to 80, is probably more of this 

 same leucosin, as shown by preparation 8, which formed about 25 per cent 

 of the total coagulable protein. The experience of the writer has been that 

 complete coagulation, especially in a solution nearly free from salts, can be 

 effected, if at all, only by heating the solution much above the lower coagu- 

 lation temperature of the protein to be separated. 



From the whole seed leucosin was obtained with the same composition and 

 general properties as from the embryo, but preparations from the whole seed 

 were free from phosphorus. This was probably because the proportion of 

 nucleic acid to protein matter was smaller in the whole seed than in the embryo, 

 so that on extracting with water the nucleic acid did not form soluble com- 

 pounds with the leucosin, but remained undissolved in combination with the 

 other proteins. In table 4, on the following page, is given the average of 

 analyses of albumin from the cereals. 



