298 PHYSIOLOGICAL CHEMISTRY. 



several points of difficulty and the published results do not 

 show very good agreement. The separation of the proteins 

 offers the greatest difficulty, as the simple and accurate meth- 

 ods employed in the analysis of cow's milk fail to give equally 

 satisfactory results when applied to mother's milk. The ex- 

 planation of this will be given below. There are variations 

 in the composition of human milk as in that from other species, 

 but average values are about as given below : 



Water 87.5 



Fat 3.8 



Casein 1.6 



Albumin 0.5 



Sugar 6.2 



Salts 0.4 



100.0 



This analysis must be accepted as representing the facts 

 only in a general way. Indeed, some authors go so far as to 

 assert that no mean value for woman's milk is possible, as the 

 variations from individual to individual are too great to per- 

 mit an average result to have any legitimate meaning. This 

 much, however, is well established : the fat in woman's milk 

 is not greatly different in amount from that in cow's milk; 

 the sugar is about fifty per cent higher in the mean ; the salts 

 are lower, sometimes as little as 0.2 or 0.3 per cent of ash 

 being found; the total proteins are about half as much as in 

 cow's milk. But as to the relation of the casein to the albu- 

 min and as to the nature of the casein itself, the greatest diver- 

 gence of views exists. Some analysts, have actually found 

 more albumin than casein as a result of experiments. This 

 is probably due to the employment of a faulty method for the 

 precipitation of the casein; it has been pretty well established 

 that the conditions of precipitation or coagulation are entirely 

 different from those obtaining for cow's milk It is indeed 

 likely that the protein called casein in woman's milk is quite 

 distinct from that of cow's milk. Under the action of rennet 

 the former coagulates in fine flakes while the curd of cow's 



