New York Agricultural Experiment Station. 65 



The test of the barn after the plastering extended from July 11 to 

 July 18, 1910. As soon as the services of painters could be obtained 

 the woodwork and the iron work of the stable were painted. When 

 this had become thoroughly dry the cattle were returned to the stable 

 and the germ content of the milk as obtained from the same cows 

 under this new condition was again determined. This determination 

 was made August 10-17 under the same general conditions which 

 had accompanied the other tests. 



The numerical results from these three comparative tests are given 

 in Table VII. 



A comparison of the germ content as determined in the strippings 

 and in the milk taken from the pail at the close of the milking process 

 shows a close relationship between the two. With Millie F. and 

 Nora D., where the germ content of the whole milk is relatively high, 

 the germs in the strippings are also numerous. With the remaining 

 cows where the total germ content was lower the bacteria in the 

 strippings were less abundant, being particularly low in numbers 

 with Millie F. B. B. In all of these comparisons there were only 

 two cases in which the samples from the strippings gave the larger 

 numbers and these samples were from Mabel S. F. on July 14 and 

 August 16. The high content of the strippings from this cow before 

 and after these particular high counts suggests that this was not a 

 mere error in technique but was due to some unusual udder con- 

 dition. The colonies which appeared so abundantly upon the plates 

 at this time seemed to be identical with those previously observed in 

 connection with this udder. While the numbers found in the other 

 udder samples are not constant from day to day they ordinarily vary 

 within comparatively narrow limits. 



Under the head " Difference " in Table VII is given the 

 number obtained by subtracting the udder content from that of the 

 milk in the pail. If the numbers given under this heading were an 

 accurate measure of barn conditions all of the numbers in each 

 column would be the same. It is manifest from an inspection of the 

 above numbers that the individualities of the cows have affected 

 these results. The work of two different milkers has also had an 

 influence. In short there are a number of factors which have in- 

 fluenced these results which could be balanced only by keeping them 

 as uniform as possible throughout all of the comparisons. Notwith- 

 standing these variations in conditions the results are surprisingly 



5 



