J98 



glass jar. At the end of tlie week these composite sam- 

 ples of milk were tested for butter fat and notes taken, 

 which are herewith presented. 



Three to four grains of potassium bichromate in a pint 

 of milk served fairly well as a preservative, this mate- 

 rial being best in the winter but requiring too frequent 

 duplication of test in the summer when the weather is 

 hot. It causes a more or less leathery condition of the 

 cream which is difficult to re-emulsify, and in hot 

 weather the milk often undergoes a fermentation which 

 causes a loss of butter fat. The milk should not be over 

 one week old before being tested. 



Mercuric chloride proved unsatisfactory in nearly 



