SEPARATION OF CREAM. 127 



their surface tension when the temperature is lowered. Owing 

 to this increase in surface tension, the liquid drops unite together 

 at a low temperature much more rapidly than they do at a 

 high temperature. For instance, two drops of molten iron 

 unite much more readily just previous to soUdifying than 

 they do while the temperature is higher, and the liquid more 

 fluid. As the fat in milk is present in the form of small liquid 

 globules, as mentioned previously, it seems probable that 

 these fat-globules might have properties similar to those of 

 the liquid mentioned above, and behave similarly in the milk, 

 when set at low temperatures, in accordance with the deep 

 setting method. If the fat-globules act in accordance with 

 this theory, it seems probable that there is no real membrane, 

 other than that resulting from surface tension, enveloping 

 each fat-globule. If there were such a membrane, composed 

 of albuminoid chiefly, then undoubtedly the fat-globules would 

 not assume this property. 



With such a deep layer of milk the lower most fat-globules 

 must evidently encounter a great many other globules as 

 they rise. If the physical force mentioned does not facilitate 

 the process of uniting the globules, they would partly unite 

 without it. The more they unite in small bunches, or masses, 

 the greater would be the tendency for them to rise, as explained 

 previously, and more of the smaller fat-globules would be 

 carried along. The bottom globules would tend to partly 

 unite and form a filter, which passes up through the milk 

 by the buoyant force, or force of levity. 



If this latter explanation holds true, then more of the milk 

 constituents would be present in the cream from the deep- 

 setting system than in the cream from the shallow-pan system. 

 By comparing the cream raised by the shallow-pan system 

 with that raised by the deep-setting system, before the cream 

 has been removed from the milk, it will be noticed that the 

 cream raised by the shallow-pan system appears to be much 

 yellower than is that raised by the deep-setting system. This 

 condition can only be due to the fact that the surface cream, 



