METHODS OF REDUCING THE GERM CONTENT 



527 



Undesirable flavors were never decreased, and in some instances 

 intensified, but the reduction in the numbers of bacteria was 

 satisfactory. The average percentage of reduction was 99.56 and 

 was always above 99. 



In a later paper Hammer and Hauser draw attention to the 

 fact that in pasteurizing milk in bottles the heating may not be 

 uniform. As a result the bacterial content and the creaming abil- 

 ity of the milk in different bottles when pasteurized by this process 

 may vary widely. This was found to be true when a certain type 



Pulley = 



Fig. 215. Apparatus for pasteurizing milk in bulk. (S. H. Ayers and 

 W. T. Johnson, Jr., from Dairy Division, U. S. Dept. of Agriculture, Wash- 

 ington, D. C., Jour, of Inf. Dis.,1914, vol. 14, p. 224.) 



of pasteurizer was used, while in another type the differences were 

 slight, although, as a rule, the bottles from the top tier of cases 

 had the lowest and the bottles from the bottom tier the highest 

 bacterial counts. The authors noted a distinct, though not 

 precise, relation between the bacterial count and the creaming 

 ability. The more the creaming ability suffered, the smaller was 

 the bacterial content. The test for creaming ability, according 

 to Hammer and Hauser, "affords a simpler and quicker method of 

 detecting the uniformity or lack of uniformity in heating with the 



