THEORY OF CREAM SEPARATIOX. 259 



must pass, as in the cream separator, where the layer of milk is 

 always of constant thickness ; the space is determined by the 

 depth of the layer of milk set. 

 The formula 



_ s ir'' 

 t ~ c 



may be transformed into 



t = — , where « is a constant. 

 r' 



Taking the diameter of the largest globules as O'Ol mm. and 

 the smallest as O'OOIG mm., we calculate that the smallest globules 

 will take about fifty times as long to pass through a given space 

 as the largest ; the author deduces from his experiments that 

 the largest fat globules move at the rate of 15 mm. per hour. 

 If we assume that the total weight of fat globules of any size, 

 is equal to the total weight of fat globules of any other size, in 

 an ordinary cream tube we may expect roughly the following 

 figures : — 



In ') hours about 35 per cent, of total fat will be found in the cream. 

 „ 10 „ ,, 65 



„ 24 „ „ 85 



while from three to four days should elapse before the whole of 

 the fat is found in the cream. 

 From the equation 



\' k . i 7! (dg — dj)g . r- 



V = > 



c 



it will be readily seen that if the density of the fat varies the 

 time will be considerably affected. The density of sohd fat at 

 60^ F. (15-5° C.) is about 0'93 ; the density of liquid fat is about 

 0'92 at the same temperature ; and, as has been shown by the 

 author and S. 0. Richmond, it is highly probable that the soUdi- 

 fication of the fat is a process which takes time. The difierence 

 between the specific gravity of milk serum and milk fat is also 

 accentuated at temperatures above 60° F. ; it is probable that 

 when milk is rapidly cooled, the fat globules do not so easily 

 attain the lower temperature as the serum. It would appear, 

 theoretically, that there is a considerable advantage in setting 

 milk for cream immediately after milking, and that the fat 

 globules will rise at a much more rapid rate than if the milk be 

 cooled and kept for some time. The experiments of Babcock 

 completely substantiate this view ; he finds that delaying the 

 setting for even a short time materially afiects the percentage of 

 fat in the skim milk. 



