DAIRY METHODS 139 



for this are probably three. I. The centrifugal machine itself 

 is somewhat complicated and is extremely difficult to get thor- 

 oughly clean. After using, it is, of course, washed, but it is 

 difficult or impossible to wash the bacteria all away; and when 

 the next lot of milk is passed through it, there may be some 

 contamination of milk by bacteria from the machine. Probably, 

 however, this is a small item. 2. In the method of determining 

 bacteria, as shown on page 273, the number is determined from 

 the 'numbers that grow in gelatin, and it is assumed that there 

 was in the original milk one bacterium for each colony. It not 

 infrequently happens that bacteria are just about to divide, and 

 if such a half-divided bacterium is in the milk that has been 

 tested, the two halves being so closely together will make only 

 one colony, and will be counted as one, whereas they will be 

 broken apart by the centrifugalization and appear as two col- 

 .onies. 3. It frequently happens that several bacteria may cling 

 together in the original milk, when the whole bunch would 

 produce a single colony and be counted as one. When, however, 

 milk is passed through the centrifugal machine the agitation 

 it receives is apt to break the bacteria apart and thus increase 

 the number of isolated individuals. The result would be a larger 

 number of colonies in the gelatin plates, and these when counted 

 would give a greater number of bacteria in the milk passed 

 through the machine. Hence, although centrifugalization can- 

 not really increase the number of bacteria, nor does it improve 

 the keeping quality of the milk, it does increase the number 

 counted on a gelatin plate. In spite of this fact, it is beyond 

 doubt a very useful means of cleaning milk, since milk that has 

 been thus centrifugalized is deprived of most of its filth. 1 



Dunbar and Kisler. Milchztg., p. 753, 1899. 

 Rolet. Rev. Gen. d'Lait, I., p. 185, 1901. 

 Eckles and Barnes. Bui. 59, Iowa Exper. Sta., 1901. 

 Severin. Cent. f. Bact., II., xiv., p. 605, 1905. 



