New York Agricultural Experiment Station. 175 



milked cows. The average of 56 samples of milk drawn by hand is 

 381,000 cells per cubic centimeter while the average of 113 samples 

 of machine-drawn milk is 309,000 cells per cubic centimeter. Of 

 the 56 hand-drawn samples, 38 gave counts under 500,000 per cubic 

 centimeter, 14 gave counts between 500,000 and 1,000,000 cells per 

 cubic centimeter, and 4 gave counts over 1,000,000 per cubic centi- 

 meter while 82 of the cell counts from the machine-drawn milk 

 were under 500,000 per cubic centimeter, 17 were between 500,000 

 and 1,000,000 per cubic centimeter and 14 were higher than 1,000,000 

 per cubic centimeter. These results show a somewhat lower cell 

 count for machine-drawn milk than hand-drawn milk which, with 

 the other evidence given which bears on this point, makes it probable 

 that the lower cell count of the machine-drawn milk was due to the 

 method of milking used. 



CONCLUSIONS. 



The results obtained in these investigations confirm the conclu- 

 sions formulated by Breed and Stidger 44 in their earlier paper. 

 More definite statements in regard to some points can be made. 



Apparently the largest average number of the cells present in 

 milk occurs in colostral milk but equally large numbers of cells 

 occasionally occur in milk drawn at any portion of the lactation 

 period. 



Several very high cell counts have been obtained from milk of 

 animals nearing the end of their lactation periods and the evidence 

 here given indicates that such high counts are more common during 

 the latter part of the lactation period than during the height of 

 lactation, but the average cell counts for the latter part of the lacta- 

 tion period do not seem to be markedly higher than the average 

 cell counts of earlier parts of the lactation period. This indicates 

 that the average total number of cells discharged per milking is 

 less during the latter part of the lactation period than during the 

 height of lactation. 



There are marked daily variations in the number of cells dis- 

 charged which do not show a close correlation with any of the sug- 

 gested causes for such variations. 



No constant relation between the number of cells discharged in 

 the foremilk and the number discharged later in the milking process 



44 See footnote 1. 



