CREAM-RAISING COEFFICIENT. 143 



whether the milk-ring in the drum is more or less strong, whether 

 the drum works regularly and quietly, and whether the machine 

 can be conveniently and simply worked. 



(5) On the special properties of the milk which is to be 

 separated. Under ordinary conditions, milk brought from a dis- 

 tance, or lazy milk, or boiled milk, is less easily creamed than fresh 

 milk of ordinary properties. Perhaps also milk, very rich in fat, is 

 less perfectly creamed than milk containing an average percentage 

 of fat. These conditions are insignificant, and hardly possess any 

 importance in practice. They have a perceptible influence in 

 properly regulated separators only if the creaming takes place at 

 a temperature under 20 C. 



The numerous experiments carried out during the years 1877-1885 

 at Raden, with different separators, were the first which distinctly showed 

 that creaming is more effective the quicker the separator-drum revolves, 

 and the warmer the milk is which is to be creamed, and the smaller the 

 quantity of milk that passes through the drum in a given time. They 

 showed, however, that between the percentage of fat (/) in the skim-milk 

 on the one hand, the rapidity (u) of the drum which determined, on 

 the other hand, the quantity of milk (m) creamed in an hour, and the 

 temperature of creaming (t), a certain regular relation existed. Numerous 

 detailed calculations, which the author has made on the basis of a large 

 number of single experiments, show that the truth is very nearly obtained 

 by assuming that the percentage of fat in skim-milk (/) is inversely 

 proportional to the square of the number (u), denoting the revolutionary 

 speed, and directly proportional to the square root of the number (m), 

 denoting the quantity of milk creamed in an hour. The relation of the 

 number (/) to the temperature of cream -raising (t) was found, if (/), 

 denoting the fat percentage of skim-milk at 40 C., lay between the 

 limits of 13 and 40 C. by the equation 



/-/i-x 



and this yields also 



(c) indicates a constant factor, which has been obtained for each separator 

 by means of exact experiments. If the value of this factor has been care- 

 fully fixed for a definite separator, it is easy, as has been elsewhere shown, 

 by the author, to find the exact value of (/) for all values of (u) between 

 J (u') and 2 into (u'\ for all values of (ra) between (J m') and (2 into, 

 ra'), and for all values of (t) between 20 and 40 C. In the case of some 



