346 COLLOIDS IN BIOLOGY AND MEDICINE 



water and the crystalloid content is practically uniform. G. CORN- 

 ALBA* showed this by extensive investigations upon large dairy 

 herds. The widest limits for the content of dissolved substances 

 was only from 5.9 to 6.6 per cent, whereas the variations were usually 

 from 6.05 to 6.25 per cent. From this we must conclude that milk 

 is the product of at least two processes. One is the result of an 

 ultrafiltration of the blood which yields an ultrafiltrate of uniform 

 water and crystalloid content. The colloids and fat are mixed with 

 this solution by a second process. 



The fat globules of the milk (milk globules) have a diameter of 

 from 0.1 to 22 /* averaging about 3 //, yet it is possible mechanically 

 so to break them up, that they are no longer visible microscopically; 

 according to WIEGNER their average diameter is 0.27 /z. Such so- 

 called homogenized milk is recommended as being very easily digested 

 and it has the advantage that the cream cannot be removed either 

 by gravity or by centrifuging. 



It is known by dairymen that milk may separate spontaneously 

 to a certain extent; on standing, the cream rises to the top. The 

 result is produced more rapidly and completely by centrifugation; 

 though a complete separation is not obtained even in this way, we 

 have an opaque milk which contains the finest globules. Nor can a 

 separation be obtained by filtering through the least porous Cham- 

 berland filter. All the fat globules can be held back by an ultra- 

 filter which still permits the complete passage of albumin. 



If we wish to make butter from cream, in other words to make the 

 milk globules unite, the cream must be churned, since between the 

 individual globules aqueous and water-soluble constituents of the 

 milk occur as partitions, and these partitioning walls must be broken 

 down. It is a remarkable fact that milk globules do not dissolve in 

 ether if milk is shaken with it. 1 If milk fat were an ordinary emulsion 

 such as oil in water, the fat would be completely removed by shaking 

 it with ether. From this we conclude that the fat globules in the 

 milk are surrounded by a pellicle impermeable for ether. If potas- 

 sium hydrate or acetic acid are added, this interference is removed; 

 moreover, the fat may be removed from the dried milk globules by 

 treatment with ether, and the pellicles will be left behind. It is 

 absolutely impossible to extract the fat completely from homog- 

 enized milk by shaking with ether. 



There is quite an extensive but at present practically useless 

 bibliography (see VOELTZ *) on the pellicles of fat globules. Especially 

 erroneous were the experiments directed to splitting off the pellicles 



1 The fat of human milk, with its much larger quantity of albumin, is readily 

 shaken out with ether. 



