924 EXPERIMENT STATION RECORD. 



first of these v^as with 17 cows and extended from September 15 to 

 early in IStoveniber. The cows were at pasture but were stabled dur- 

 ing the night. The cows were grouped according to the quality of their 

 milk, and samples of the morning's and night's milk of each grouj) 

 were tested daily with the Babcock test and lactometer. Opportunity 

 was afforded for 72 comparisons of the effect of changes of temijera- 

 ture on percentages of total solids and fat in the milk. 



"In cases of general rising quality no connection with weather could be traced. 

 The amount of change was as great one way as another. 



"During stationary temperatures, more or less quality fluctuations occurred, but 

 they were found to be equally divided in direction. 



" In brief, it may be said that more than half (55 per cent) of the changes in fat 

 percentage, and nearly two thirds (65 per cent) of the changes in per cents of total 

 solids were opposite to those of the thermometer, but that for periods of more than 

 a day the inverse character of the qualitj^ change was less marked. 



"Coincident with a warm spell [October 15 to 25] the milk flow fell for the first 

 time below the line of calcuhitod natural shrinkage. On the other hand, 5 days 

 averaging as warm (October 3 to 7) were accompanied with a little more milk than 

 usual. There was a marked decrease in quantity early in November, as the weather 

 grew colder." 



The other test was with 7 cows stabled in a poorly constructed barn 

 on a bleak hillside, and lasted from February 25 to March 30, 1894. 

 The morniDg's and night's milk of each cow was tested by the Babcock 

 test. Omitting the times when the temperature varied less than 5*^ 

 from day to day, there were 40 opportunities for observing the eflect of 

 changes of temperature on the milk. 



"These show that 21 times the quality change was inverse to temperature change, 

 13 times in the same direction, and 6 times no quality change followed thermometric 

 fluctuation. Nearly two thirds (62 per cent) of the changes were in the inverse 

 direction to temperature change. They were also more pi'ouounced than the changes 

 in the same direction (0.275 per cent inverse, 0.215 per cent same). 



"There were 12 cases of rather decided temperature changes extending over 3 days 

 or more (using both sets of temperatures separately). The general quality of the 

 milk changed inversely 6 times, in the same direction twice, and fluctuated irreg- 

 ularly 4 times. Three quarters of these changes were in the inverse direction." 



From a study of the records of the World's Fair test the statement 

 is made that "over half the changes in per cent of total solids, and 

 three quarters of the changes in fat percentages were inverse to ther- 

 mometric variations. . . . No connection could be traced between tem- 

 perature and quantity changes (save as high temperature induced 

 lighter feeding and consequent falling off) or between storm, and either 

 quantity or quality change." 



The author summarizes the results of the 2 tests previously reported, 

 and taking these in connection with those rei)orted above, draws the 

 following inference: 



"These 5 separate tests, covering practically the entire year, and the conditions 

 of pasture, summer soiling, and winter barn feeding, point directly to the conclusloa 

 that the tendency of cows is to give from day to day richer milk when the temper- 

 ature falls and poorer milk as it rises; or, in other words, the quality of the milk 

 (solids and fat) varies inversely to temperature changes." 



