Wilson — The Inheritance of Milk-Yield in Cattle. 93 



would be necessary to determine, as closely as such a matter would permit, 

 how far the 3'ield of a cow is increased wlien her lactation exceeds the normal, 

 and how far it is decreased wlien the lactation period is shortened. The 

 normal lactation is, of course, taken to be that in whicli tlie next calf is bom 

 about twelve months after the previous one. It was also necessary to 

 determine how mucli a cow's annual yield increases with her age. 



For these objects the records of the herds kept at tlie Department of 

 Agriculture's farms, as also a number of Dutch and Danish herd-books, in 

 wliich cows' yields are recorded, were examined. But the examination of 

 these books showed also that still other circumstances would liave to be taken 

 into account. 



A cow's yield is exceedingly sensitive aud liable to be affected by many 

 causes. The chief of these were found to be food and weather and time of 

 calving. An examination of the records at Grlasnevin, which have been kept 

 since 1890, showed that the combined yields of the herd rise and fall 

 at certain seasons, sometimes very considerably. From November to 

 January, when the cows are ou the usual winter food of hay, roots, and 

 concentrates, the yield is practically constant ; but it falls in February 

 and rises again in March. Early in May, just at the time the cows 

 are put out on the pastures, it falls again, but only for a week or so. 

 Then it rises quickly, and till about the end of June or the beginning 

 of Jul}^ it remains at from 10 to 20 per cent, above the winter yield. 

 During July aud August, depending on the rainfall, it falls again ; and, 

 in October, until the cows are housed for the winter, it is usually at its 

 lowest. These variations have considerable effect upon the cows' total yields. 

 A cow calving between December and the end of February has the advantage 

 of the rise produced by pasture before Iier yield has dropped far, and, in 

 consequence, if she be a very good one, may give a hundred or even a 

 hundred and fifty gallons more during a normal lactation tlian another verj' 

 good cow calving between May aud September. 



It was also found, of course, that illness and abortion, aud such tilings as 

 injury to the udder, caused the yields to fall; but several causes had results 

 that would not be generally expected. It is a common belief that a cow that 

 fattens up before calving cannot milk well. This, if not disproved directly, 

 is disproved indirectly by the Glasnevin figures ; for the cows that continue 

 to yield up to the end of the previous lactation, aud therefore do not recover 

 " condition " before calving, fall off in yield during their next lactation. It 

 is also disproved by the fact that breeders who exhibit cows at milking trials 

 almost invariably feed their cows well for some time before calving. One 

 well-known breeder believes that the difference between a good cow lean and 



