Wilson — The Inheritance of Milk- Yield in Cattle. 91 



was inherited through the sire as well as through the dam. Bveatually it 

 became widely accepted that tlie bull's pedigree as regards milk was of vital 

 importance, and, when the selection of a particular sire was under considera- 

 tion, his milk-pedigree, so to speak, was examined as far back as it could be 

 given. 



At no time, however, was there any suggestion from Denmark, or from 

 Holland or Sweden, where records had also been kept systematically, that 

 there was any reason to question the theory that improvement was a slow and 

 gradual process. But it was clear that, as time went on, the confidence of 

 stock-breeders in the system of selecting the progeny of good milkers for 

 stock purposes rose higher and higher. Tlie visit to Denmark, therefore, did 

 not result in any change of view with regard to tlie inheritance of milk-yield 

 further than to emphasize the importance of keeping milk records in order 

 to know which stock should be eliminated and which should be bred from. 



Three years ago, when it became apparent that Mendel's laws applied to 

 colour, horns, length of limb, and other less important characters in cattle, one 

 of the first questions that suggested themselves was, Is milk-yield also inherited 

 in some Mendelian manner ? and the inquiry was begun at once. But the great 

 difficulty was to find data. Two or three years before that time, a number 

 of Ayrshire breeders had commenced to keep records, and an appeal was made 

 to the late Mr. John Speir, who had been at the head of the movement. 

 On Mr. Speir's recommendation, a number of Ayrshire breeders were visited, 

 but their records were not old enough to yield sufficient data. One very 

 important point, however, was brought out in conversation with Mr. Speir, 

 namely, that in a number of cases, where a daughter was an improvement or 

 the reverse upon her dam, the difference between their yields was not small, 

 but large. Cows giving four or five hundred gallons might have daughters 

 giving six or seven ; cows giving sis or seven might have daughters giving 

 nine; andf/ce versa. Poor milking cows usually had poor milking daughters, 

 and good milking cows good milking daughters; but the rule was by no 

 means invariable. At the same time Mr. Speir had no doubt of the power 

 of a bull to improve or damage a herd, and to leave daughters of very 

 unequal capacity. Unfortunately no note was taken of the cases Mr. Speir 

 had in his mind, as it was never thought but that he would be alive to be 

 referred to again and again. Some of his cases are referred to in his Milk 

 Record Reports, published in the Transactions of the Highland and Agricul- 

 tural Society of Scotland during the years 1904 to 1910, and others in a 

 lecture on " Milk Records," delivered in December, 1907, and published by 

 the Scottish Agricultural Publishing Company. 



In dairy literature of recent years much had been written of the great 



p2 



