G. W. Shaw in Hoard’s Dairyman of March 10, 1916 says:
“It is a well-known fact that the percentage of butter fat in the milk of cows increases very materially toward the end of a period of lactation. There are also other slight changes in that period. During the first month the fat generally averages higher than during the second month. Under normal conditions of feed, etc., the fat increases from the third or fourth month to the end of lactation.
“Although it is a fact that cows cannot be fed to give beyond a certain percentage of butterfat, yet it has been proven many times that if poorly fed for a considerable length of time, the average test will decrease. This is especially true if cows become thin and poor in flesh. Many times this will account for a farmer’s average herd test dropping from one period to the next. It is also noticeable that the quantity of water taken, whether as water or succulent feed, affects the herd. This is particularly noticeable when cows are changed from a diet of dry hay to green feed or vice versa.
“It has been noted that the change of weather affects the test. A sudden cold period coming will usually decrease the quantity of milk, but increase the percentage of fat. If the cold period continues, this change will tend to right itself. It would seem that there is a connection between the question of heat and cold and the amount of water taken.
“It is a well-known fact that the first milk drawn from a cow’s udder is very low in butterfat, not over 1%, whereas the last drawn is quite high, sometimes reaching 10%. The importance of exhaustive milking is evident. By carefully milking to the fullest extent each time, the test will undoubtedly be higher than if milking were not exhaustive. This, continued over a period of time, would have its effect on the 15-day test.
“Another very important point we wish to make is this, unless a man who does his testing at home understands how to do it thoroughly and is very careful in taking his sample, he will not check with the factory test. There are several reasons for differences between tests made on samples taken at the farm and those made on samples taken at the factory.
“Many farmers have a habit of taking a little cream or top milk for family use, and think that it will not materially affect the average test. As a matter of fact it will affect materially. For instance, if a farmer were producing 100 pounds of milk testing 3.5% and he used one quart of top milk, testing 10%, his average test would be reduced .2 of 1%; that is, instead of delivering milk testing 3.5% it would actually test 3.3%.
“Some farmers adopt another method; they use, for family purposes the milk from a cow which gives the richest milk, so that the result is always the same, the average test being lower.
“Another cause of difference in tests, and we think this is a very important one, is found in the condition of the milk when received at the factory. Some farmer’s milk, when brought in, is smooth and homogeneous; some bring in milk which is slightly churned; that is, there are small particles of butter, which is separated butterfat floating on the surface. This latter milk is very hard to sample; the sampler is plunged into the milk and is likely to miss a due proportion of these floating particles. In addition, some of the separated butterfat is sure to be left behind, both on the sides of can and on the cover. Butterfat adheres to any surface much more rapidly than any other of the milk solids. It is quite evident that milk which is partially churned will get a lower test at the factory than it did at the farm before it became churned.
“In order to prevent this churning, it is most important that the milk be quickly and thoroughly cooled after milking. If milk is poured into cans and stirred and handled in a half warm condition, it is sure to separate to some extent. While the particles of butterfat are not large enough to be particularly noticeable, they are there and adhere to the surfaces as described.