270 MILK 



certified milk. The variation in the milk from each one of the 

 3 cows was also material. The maximum count from one cow was 

 305,000, the minimum 1700; from the second cow the figures were 

 3,500,000 and 2500; and from the third one, 4250 and 50. It is 

 clear that the variation in numbers of bacteria in the udder at 

 different periods may be considerable. 



This fact is emphasized by the work of Harding and Wilson, 

 who made the following counts in milk derived from the different 

 quarters of the same udder: the average content of the samples 

 from the front quarters was 191 and 249 per cubic centimeter 

 respectively, while those from the back quarters gave averages 

 of 625 and 635 per cubic centimeter. The authors state further 

 that this is not an accidental relationship, because in each of the 

 three herds studied the average germ content of the back quarters 

 was from two to four times that of the corresponding front quar- 

 ters. This study showed furthermore that the maximum number 

 of bacteria found in the front quarters was 4632 per cubic centi- 

 meter, while in the back quarters more than 16,000 were found. 

 In some cases the reverse was found, namely, a higher number in 

 the milk from the front quarters than from the back quarters. 

 However, the authors state that in these cases the milk was ob- 

 tained largely from cows during the early part of the lactation 

 period, or in cases where the front quarters contained large num- 

 bers of a particular type which was absent from the correspond- 

 ing back quarters. No explanation is offered for the cause or 

 significance of this phenomenon. 



The distribution of bacteria in the udder is not uniform, as 

 has been pointed out before. The first streams of milk are in- 

 variably richer in bacteria than succeeding portions. Schulz 

 found the following numbers of bacteria in the first and last 

 milk: 



BACTERIA IN FIRST AND LAST MILK 



First milk. Last milk. Average of whole product. 



55,566 2070 



97,240 500 2590 



74,088 ^ 9261 



In this work but 0.1 c.c. of the milk sample was plated, so that 

 the absence of bacteria in the last milk does not necessarily im- 

 ply that none were present in larger amounts. 



Russell counted 2800 bacteria per cubic centimeter in fore- 

 milk, against 330 as an average for the whole product. 



There is no difference of opinion about the germ content of 

 fore-milk compared to that of later portions, it being generally 

 agreed that the first streams usually contain many more bacteria 

 than later ones. But whether bacterial numbers gradually dimin- 

 ish, or whether there are fluctuations, or whether finally there is 



