COMPOSITION OF HUMAN MILK IN AUSTRALIA. 175 



In the present work an endeavour was made to comply 

 with the above recommendations by having the mammary 

 gland emptied as completely as possible for each sample. 



It has also been shown by Engel (1910), that the per- 

 centage of fat in human milk is lower the longer the interval 

 of time is which has elapsed since milk was last withdrawn 

 from the gland. Account has been taken of this statement 

 in the present case by noting the time which had elapsed 

 for each sample since the last withdrawal of milk from the 

 same gland, with a view to ascertaining whether any 

 similar effect of time on the percentage of fat in the milk 

 secreted was to be observed. 



In the case of the cow, the fact that the last portions 

 of a milking are richer in fat than the first is well known, 

 and seems to have been first demonstrated by Parmentier 

 and Deyeux as long ago as 1790. It is also well known to 

 dairymen that the shorter the period of rest between two 

 successive milkings of a cow the richer in fat is the milk 

 yielded at the second milking, but the smaller is the volume 

 obtained. 



The total protein in the first fifty-three samples ofjiuman 

 milk was estimated by the method of Sikes (1906). In this 

 method, 5 cc. of milk to which two or three drops of a 

 saturated solution of citric acid have been added to hinder 

 the precipitation of salts, are mixed with 100 cc. of absolute 

 alcohol, and boiled. The proteins are completely precipi- 

 tated, and the fat, sugar, and extractives go into solution 

 in the hot alcohol. The precipitate is spun in a centrifuge 

 while hot and washed twice with 30 cc. of boiling alcohol. 

 In the present case large quantities of absolute alcohol and 

 a centrifuge in which the washing could be done con- 

 veniently were not available. The precipitation was 

 therefore brought about by means of 95% methylated spirit, 

 and the precipitate was washed twice with 50 cc. of the 



