VARIATION IN TESTS OF MILK. 151 



deemed best, the feed for all being weighed and charged by careful and 

 responsible men employed by the exposition. 



The milk from each cow was brought to the dairy room as soon as 

 drawn and here weighed and recorded and two samples taken for testing. 

 One taken with a milk-thief (a copper tube of about f-inch bore) was added 

 to a composite jar provided for each cow, and the other, taken with a small 

 dipper, was put in another set of pint jars for testing with the lactometer. 



Each mess of milk was poured, after weighing, into a common shot- 

 gun or setting can and at once sampled; the milk-thief thus taking a pro- 

 portional amount of each mess to be added to the composite sample. 



The composite samples, provided with a little corrosive sublimate, were 

 tested with the Babcock test each week on the same day, and from these 

 tests and total milk yield for the week, the pounds of butter fat and value 

 of same for each cow was computed and value of feed consumed by each 

 deducted. Thus the profit on butterfat or "estimated butter" for each 

 cow and herd was determined for each week separately. 



The samples taken for lactometer testing were tested daily and the 

 readings for the week for each cow averaged. To the average lactometer 

 reading and composite Babcock test of each cow a formula was applied 

 which gave the per cent, of total solids in her milk for the week. Thus the 

 data for the third award (profit on total solids) was supplied from week to 

 week and a separate record made for each animal in total solids as well as 

 in butterfat. 



Total solids include all constituents of the milk excepting water and 

 represent its full food value. In average milk the solids are divided about 

 as follows: Fat 3.5, casein and albumen 3.4, milk sugar 5.0 and ash .7 per 

 cent., total, 12.6 per cent. 



The committee who fixed the price of 9c. per Ib. for total solids, evi- 

 dently assumed each element of solids to be of equal nutritive value for 

 human food and based their calculation on the average price and approx- 

 imate composition of milk sold in New York City. 



The award of greatest profit on "churned butter" was a sort of con- 

 ciliatory measure, probably intended to satisfy that faction of the Jersey 

 breeders who contend that no records excepting those of actually churned 

 butter signify anything. 



Owing to the impracticability of churning all the milk from each of 

 the ten breeds separately, the rules were so modified as to provide for the 

 separating and churning of one day's milk from each breed each week. 

 The yield of butter per 100 pounds for the single day's milk churned to be 

 applied to the total yield of milk for the week and this estimate called 

 "churned butter." 



The test of a herd's milk on the day it was churned frequently varied 

 .2 to .3 per cent, from the average per cent, of fat in the herds' milk for the 

 entire week, so that with perfect work in separating and churning, the 



