104 BUTTER-MAKING. 
Average Sample.—It is sometimes desirable to obtain an 
average test of the milk from a whole day’s delivery. This 
can be obtained in two ways: First, by taking a sample from 
each patron’s milk with a sampling-tube, and putting it all 
together in one jar. The result represents an average test, pro- 
viding the samples have been correctly taken. Second, an aver- 
age test can be had by boring a small hole in the conductor-head. 
When the milk passes over this hole, a small portion of it 
drops through. A vessel of some kind can be put underneath 
to catch the drops. Such a drip-sample will represent very 
accurately the average quality of the milk received at the 
creamery. If it is desirable to keep this sample, a preservative 
can be added to it. 
Composite Sampling without the Use of Preservatives.— 
Pipettes can be obtained holding 5.87 ¢.c. of milk. These are 
one-third the size of the ordinary 17.6 c.c. pipette used for 
the Babcock test. With this small pipette a sample may be 
taken every day from each patron’s milk, during three suc- 
cessive days, and emptied into the same test-bottle each 
day. At the end of three days the samples may be tested 
and the bottles cleaned, ready for use again. 
Accurate composite samples may be obtained in this way, 
providing the sample in the pipette is correctly taken each 
day. No preservative is needed. The preservatives are added 
to the composite samples to prevent curdling. The test-bottles 
may be placed on a shelf, or preferably in a rack made to hold 
them. They should be marked in such a way as to identify 
them. A good way is to mark them the same as the com- 
posite jars, the number on the jar corresponding to the number 
on the milk-sheet for each patron. 
