110 Keport of the Department of Bacteriology of the 



serious in the case of fields containing large numbers of bacteria. 

 The use of the micrometer referred to above obviates this difficulty. 



Dead bacteria. — Dead bacteria may be counted. In normal 

 fresh milk however, there is every indication that they are a negli- 

 gible factor, as previously explained. 



Large clumps. — Even though all bacteria in compact clumps 

 cannot be counted, such clumps occur so infrequently that they may 

 be regarded as unimportant. 



Cells and streptococci. — Thus far the sanitary significance of 

 tissue cells and streptococci is still a matter of discussion. They 

 are both readily seen and counted. This point is discussed in detail 

 in Bulletin No. 380 of this Station. 



Loss of bacteria. — The question whether bacteria are lost in the 

 preparation of the slides is discussed on page 111. 



two other direct microscopic methods of milk examination. 



Skar's method. — This method of making milk smears differs from 

 that used in these investigations in that the stain is put directly 

 into the milk as described on page 82 and the smear is dried and 

 studied without washing, or dissolving out the fat. This procedure 

 was planned in order to avoid all washing of the smears which, it 

 was feared, would remove some of the bacteria. The principal 

 aim in making a study of this method was to determine whether 

 this possibility was a real one, because at first thought it appears 

 to be perfectly reasonable. Comparisons were made between 

 various smears prepared from milk having a high bacterial content. 

 The resulting counts are given in Table XIV, and show the total 

 number of single bacteria per cubic centimeter, the clumps having 

 been disregarded. 



An examination of these results shows that while there is some 

 irregularity in the counts, there is ordinarily a markedly lower 

 count obtained from Skar's smears than from Breed's smears made 

 from the same sample of milk. Unquestionably, this is due to the 

 fact that the fat globules in the Skar smear are so numerous as to 

 completely hide many bacteria from view. This does not prove, 

 however, that Skar's fear that the bacteria are carried away in the 

 removal of the fat is groundless. Nevertheless, a study of the 

 two right hand columns of figures, secured from counts made on 

 smears where Breed's method was so modified that 0.02 of a cubic 



