20 N. H. Agricultural Experiment Station [Bulletin 302 



methods. The percentage of cows raised that reacted to the test tended 

 to increase more with the advance of years than did the per cent of cows 

 purchased. Proportionately more of the cows tested in the earlier years 



* Some of these herds were purchased in entirety at some time previous 

 to test and were purchased as clean herds. The results are therefore not com- 

 parable with other groups. In ordinarj^ practice it was customary for most 

 of this group to raise their replacements. 



were raised. Of the 36 herds tested before 1925, 71 per cent of the cows 

 were raised and only 22 per cent reacted. Cows raised constituted only 

 46 per cent of the herds tested after 1930 and 37 per cent reacted. Thus 

 it would appear that farmers by delaying the test suffered greater loss- 

 es among cows raised and that losses among purchased cows were approx- 

 imately the same regardless of whether or not the test was delayed. 



The per cent of cows purchased is also an important factor in ac- 

 counting for difference in severity of losses between counties. Rock- 

 ingham and Merrimack Counties afford the two extremes. Only 44 per 

 cent of the cows tested in Rockingham County were raised and 50 per 

 cent were reactors. The losses among cows raised were much greater 

 than for any other county. In Merrimack County where nearly three- 

 fourths of the cows were raised, the reactors amounted to only 33 per 

 cent of the cows tested, and only 22 per cent of cows raised reacted, but 

 the losses among purchased cows were much greater than for other 

 counties. 



Greater losses were undergone among mixed herds than in those 

 herds which ju-edominated in some breed. More than two-thirds of 

 the cows in mixed herds were reactors, 57 per cent of the cows raised 

 and 80 per cent of those purchased. This heavy loss is associated with 

 a comparatively large percentage of purchased cows. Next to mixed 

 herds those wliich were predominantly Holstein consisting of only 55 

 per cent of cows raised had a 42 per cent loss from the test, much higher 

 than for other breeds. Four dual purpose herds in which only eight 

 per cent of the cows were purchased all passed a clean test. 



