MILK TESTING. 



The writer does not consider it a part of his duty, in this work, 

 to teach the art and science oi working a Babcock milk or cream 

 tester, although one cannot fail to observe the clumsy manner in 

 which both milk and cream are tested in many of our factories and 

 creameries. For example, who has not seen tests being read off from 

 the test tubes with the fat column blurred with either undissolved 

 or burnt casein showing throughout ? This should not be so. The 

 fat column should be clear, pure, and well defined. 



In submitting the simple cream test chart on the next page to 

 my readers, it is desired they should at once understand It is in- 

 tended for dairy farmers only who have time to add and multiply 

 so as to arrive at an estimate of any given amount o-f cream in 

 pounds not mentioned herein. 



The following rules should be strictly adhered to in testing dairy 

 cows : 



ist. Milk the cow perfectly clean every twelve hours the day pre- 

 vious, and during the test. 



2nd. Sample as soon as milked, first pouring the milk into another 

 bucket, then taking the sample at once, not allowing the cream 

 to rise to the surface. 



3rd. For a composite sample dip out a tablespoonfu'l from each milk- 

 ing for not less than eight consecutive milkings just previous 

 to the time of testing. 



4th. For sample oil single milking to be tested take a little less than 

 one quarter of a pint. 



5th. Add as much corrosive sublimate to each bottle used to hold 

 samples as will lay on the point of a penknife from the point 

 to three-eighths of an inch back from the point. Put the cor- 

 rosive sublimate in the bottle before putting in any milk. Al- 

 ways shake the bottle with every sample added. 



6th. To take a composite sample of a herd, pour the milk together 

 and then pour from one can to another until thoroughly mixed; 

 take the sample at once, and should you have a large number 

 of cans, take a sample from each and put them together. 



/th. The percentage of butter fat fo'~ " in each milking from any one 

 cow varies so much that a c^./ect average cannot be secured 

 from less than eight consecutive milkings; therefore do not ex- 

 pect accurate tests of a cow or a herd from a composite bottle 

 containing less than eight consecutive milkings. 



8tl.. If a dairyman wishes to study the variations in percentage of 

 butter fat that occurs from milking, let him obtain a testing 

 machine and test a cow for eight days in succession ; then wait 

 a month and continue the test, and so on ior six months. He 

 will find it instructive and profitable. 



9th. Label every bottle with the date of month, hour of milking, and 

 name of cow. 



loth. Corrosive sublimate is extremely poisonous. Handle it with 

 great care. 



nth. It will be revealed that the temperature of the atmosphere and 

 the yield of milk have a more or less connection with each other. 



i?th To arrive at the average of a dairy herd each cow's milk shou d 

 t>e worked out separately each day according to weight of milk 

 and butter iat test. Weekly composite milk tests are only valu- 

 able, therefore, as a check on the creamery man when it is found 

 inconvenient to test the herd separately every day. 



261. 



