ILLINOIS STATE DAIRYMEN'S ASiSOOIATION. 149 



be found that tests of the top layer of this milk will be per- 

 ceptibly higher than those made of the last inch of milk left 

 in the pail after the bulk of it his been poured out. 



Neglect of a thorough mixing of the sample before test- 

 ing, as well as failing to remember that some of the fat 

 globules will rise to the surface very quickly, is the cause of 

 many surprises in milk testing. 



Any one familiar with milk will also understand the 

 necessity of pouring it from one vessel to another in order to 

 evenly mix the fat globules throughout the whole sample. If 

 milk is stirred with a dipper or put into a covered vessel and 

 shaken, for the purpose of evenly mixing the fat, it very often 

 happens that some of the fat is separated by this churning 

 process, and an accurate test of such milk is impossible, as any 

 amount of pouring will fail to evenly distribute this churned 

 fat throughout the milk. 



This property of the fat, to separate by agitation, should 

 always be remembered by persons sending samples of milk by 

 mail or express to be tested at some other place. We have 

 received many such samples and almost always find a lump 

 of butter floating on its surface when the samples arrives. 



If the test of such a churned sample is very important, 

 and another one cannot be obtained, this lump of butter can 

 sometimes be dissolved and mixed with the milk by adding a 

 teaspoonful of ether to the milk; then by corking the bottle 

 and shaking it until the butter dissolves in the ether, this 

 ether solution of the fat will mix fairly well with the milk, 

 and it will probably represent more nearly the original mix- 

 ture of the fat in the milk than the churned sample with a 

 lump of butter floating on its surface. The dilution of the 

 milk by the ether introduces an error in the testing, and only 

 the smallest quantity of ether necessary to dissolve the lump 

 of fat should be used. 



All this trouble of churning samples in bottles could be 

 avoided if sender would fill the bottle full of milk. The agita- 

 tion by transportation will not churn out the fat when the 

 bottle is full. This simple precaution of completely filling a 

 bottle when milk is sent to other parties for testing, will pre- 

 vent its churning and save many a disappointment to the 

 sender. 



