RESULTS OF ANALYSIS 71 



clusively of di-amino acids, especially arginine, salmine containing over 

 80 per cent. Only small amounts of mono-amino acids are present in 

 them. The mono-amino acids are alanine, aminovalerianic acid, serine, 

 proline, tryptophan and tyrosine. It has not been determined whether 

 the aminovalerianic acid is identical with valine from other proteins, 

 though this is probable. 



The chief features are the presence of arginine and the absence of 

 lysine and histidine in salmine, clupeine, cyclopterine and other 

 protamines. Sturine only contains arginine, histidine and lysine ; 

 cyprinine contains arginine and lysine ; percine contains arginine and 

 histidine. The general composition of the protamines seems to be 

 a^m or (ahl\m or (ah\m where a represents arginine, h histidine, 

 / lysine, and m mono-amino acid [Kossel, 1913]. 



As to the origin of the di-amino acids in the protamine of fish 

 sperm the observations of Miescher suggested their formation from the 

 muscular tissue of the fish. The salmon does not feed during its life 

 in the river and loses weight at the expense of the increase in size of 

 its roe. Kossel [1905] made calculations as to this possibility: the 

 roe of a salmon contains 27 grams of protamine with 22-8 grams of 

 arginine ; if the muscular tissue contain 7*1 per cent, of arginine, 321 

 grams of it would be required : more than this quantity of muscle is 

 decomposed during the life of the salmon in the river, so that, as there 

 is no evidence for its synthesis, its origin is by transference from 

 the muscular tissue. Actual analyses of the protein of salmon muscle 

 and salmon tissue have been made by Weiss [1907] and they confirm 

 the calculations. In the transference of the arginine from the muscle to 

 the roe it seems that it is rendered stable by combination with nucleic 

 acid or some other grouping and thus transported, for under the 

 ordinary conditions of decomposition of tissues arginine is itself de- 

 composed. The transfer appears to occur not as the transfer of a 

 single unit, but as the transfer of a complex containing several units. 

 Such a complex is a histone which contains more mono-amino acid 

 units than a protamine. 



In general, Kossel regards the protamines as the simplest proteins, 

 the more complex or ordinary proteins containing a protamine nucleus 

 (i.e., of di-amino acids) to which is attached the mono-amino acid units, 

 the histones being intermediate substances. This theory is supported 

 by the results of analysis of other proteins, all of which contain 

 arginine and other di-amino acids. The isolation from proteins of 

 complexes containing only di-amino acids will be the only proof of a 

 protamine nucleus in a protein molecule. 



