New York Agricultural Experiment Station. 85 



Three cows suffered severely from indigestion and their yields 

 dropped 1,200 to 1,900 lbs. below their preceding yields. A 

 fair treatment of the data deanands that the yields during these 

 years be dropped. This will eliminate four comparisons all of 

 which were favorable to hand milking. 



Discarding the questionable results as indicated above leaves 

 43 comparisons of the yields of which 24 are favorable to hand 

 and 19 to the machine method of milking. 



An inspection of the yields shows that the length of time dur- 

 ing which a cow was dry exerted a marked influence upon the 

 next period of lactation. In seven cases there was no interval be- 

 tween lactation periods. Three of these have been dropped from 

 the comparison on account of abortion or indigestion. There re- 

 mains a total of 15 instances where the interval between 'lactation 

 periods was less than 21 days. Such short intervals are not ideal 

 and might be considered as abnormal if they did not constitute 

 approximately 20 per ct. of the total. Since the number of 

 instances is considerable and is quite evenly divided between the 

 two methods of milking it seems allowable to use the data in this 

 comparison. Discarding this portion of the data would change 

 the final balance by 7,806 lbs. in favor of the hand method. 



In considering the relative merits of these two methods atten- 

 tion should be drawn to cows ISTos. 6 and 29. The former had 

 such small teats that hand milking was a slow and unsatisfactory 

 process while the latter had large teats but was so hard to milk 

 that she was drying up rapidly when the machine was substituted 

 for the hand method. ISTeither of these cows would have been 

 a desirable member of a hand-milked herd while both were satis- 

 factory when milked by machine. 



The above comparison of the yields of cows suggests that they 

 have a tendency to give a larger flow when milked by hand than 

 when milked by machine. A measurement of this tendency can 

 be obtained in pounds of milk by striking a balance at the close 

 of each year of the experiment. The balance for each of the 

 four years as well as the balance for the entire experiment is 

 given in Table III. 



