(54 Report of the Department ok Bacteriology of the 



at each of six milkings. These six cows were quite evenly distributed 

 about the stable. They were milked by two men, each man using 

 the same pail while milking his three cows and always milking the 

 same cows in the same order. Millie F., Nora D., and Millie F. B. B. 

 were milked by milker D. and Chloe B., Carey S. F., and Mabel S. F. 

 by milker W. The pails and their protection, the sampling of the 

 milk from each cow, the plating out, incubation and counting of the 

 germs in the milk were all done in accordance with the description 

 given under the head of Technique (page 53). 



After this study of the influence of the earlier barn conditions, the 

 ceiling and walls down to within 3 feet of the floor were covered with 

 wire lath and two coats of cement. The area between the cement 

 and the floor was covered with zinc. 



In putting the barn in order after the plastering, the stanchions, 

 floors and mangers were thoroughly cleaned. 



As soon as the barn was in order the germ content of the milk as it 

 was obtained from the same six cows was again examined on six days. 

 During this second test a strong effort was made to conduct all of 

 the operations connected with the barn management and the bac- 

 terial examination of the milk under conditions identical with those 

 of the earlier test with the single exception that the surface of the 

 interior of the stable had been renewed or thoroughly cleaned. 



In Technical Bulletin 27 8 it was shown that the germ content of 

 the udder could be measured fairly accurately by determining the 

 bacterial content of the strippings. In the present study samples of 

 the strippings were secured each day from each cow by milking a 

 single stream from each teat into a separate sterile test tube. This 

 was the method of sampling found satisfactory in connection with 

 the study of the udder flora referred to above. 



A composite sample was prepared for each cow by mixing together 

 one cc. of the strippings from each teat and the germ content of this 

 composite sample was determined in the same manner as for the 

 samples from the milk pail. It was shown in Technical Bulletin 27 

 that the germ content of the milk in the back quarters of the udder 

 was ordinarily distinctly higher than that from the front quarters. 

 Since more milk is secreted by the back than by the front quarters 

 a composite sample prepared as above shows a number below 

 that of the true udder content. 



' See footnote 4, p .198. 



