1909.] PUBLIC DOCUMENT — No. 31. 83 



Sufficient information will be given in the proper place to enable 

 the reader to have a clear understanding of the processes em- 

 ployed. 



The butter from which the samples for analyses were prepared 

 was taken shortly after churning, melted, filtered through a 

 jacketed funnel into clean bottles, stoppered, and set away at an 

 approximate temi^erature of 40° to 45° F. until used. The 

 samples of fat were examined as soon as possible after the close 

 of the period ; it is evident, however, that some four to six weeks 

 elapsed between the preparation and examination of the first 

 samples. This, in some resjDects was unfortunate, inasmuch as 

 slight changes would occur. Because of the large number of 

 samples to be examined the condition could not well have been 

 avoided. It is believed that the low temperature and the ex- 

 clusion of air prevented any serious changes. Each analysis 

 represents a composite of two weekly samples. 



