PRODUCTS OF HYDROLYSIS 71 



The figures in the above table show a degree of uniformity that is 

 highly suggestive, but they must not be understood to actually repre- 

 sent the true molecular weights or empirical formulas of any of these 

 proteins. The number of sulphur atoms in some of these formulas 

 does not correspond with that which would be required by the figures 

 given in the preceding table for the probable amount of cystine 

 sulphur which they contain. Thus the cystine sulphur in gliadin, 

 some of which is known to be present in this protein, requires an even 

 number of atoms if all the sulphur is cystine sulphur. The formula in 

 the above table gives five atoms of sulphur in this protein. For other 

 proteins, however, the calculated cystine sulphur is in full harmony 

 with the requirements of the formula. It is possible, and in fact not 

 improbable, that the actual molecular weights are at least twice as 

 great as those given in the table, as the proportion of sulphur found in 

 vicilin is so small that, if this element is a constituent of all the mole- 

 cules of the preparations of this protein, the molecular weight must be 

 much in excess of 30,000. The extremely small proportion of sulphide 

 sulphur obtained from phaseolin and casein also require a molecular 

 weight in excess of 30,000. It is therefore evident that definite conclu- 

 sions cannot be obtained from calculations based on the sulphur content 

 of even the most carefully prepared and carefully analysed proteins, and 

 such calculations are to be taken as simply showing that the molecular 

 weights are probably very high if the preparations of these proteins re- 

 present a single substance. 



