International evaluation of Holstein-Friesian dairy sires from three countries

Abstract

Official 1985 genetic evaluations of Holstein-Friesian dairy sires for milk yield, fat yield and fat percentage from Canada and the U.S.A. were pooled with unofficial 1985 genetic evaluations of sires from Italy in order to obtain international evaluations of sires for these three countries. There were 1766 sires from Canada, 9746 from the U.S.A. and 4191 from Italy with at least 20 daughters or 5 effective daughters each. The linear model for describing a sire's daughter average, free of all non-genetic fixed effects, included the country of proof, the country of origin and year of birth, sires within country of origin and year of birth and all additive genetic relationships among bulls. Daughter averages were approximated by de-regressing sire proofs and accounting for year of birth within country. Differences in country of proof solutions reflect differences in genetic bases of evaluation; the Italian minus the Canadian and the U.S. solutions were, respectively, 334 and 565 kg for milk yield, 14.4 and 19.6 kg for fat yield, and 0.05 and -0.01% for fat percentage. Between 1955–1959 and 1960–1964, there was very little genetic progress in all countries. Genetic change in sires born between 1965–1969 and 1975–1979 were 121, 198 and 87 kg for milk yield, 4.6, 6.0 and 2.4 kg for fat yield, and 0.01, -0.02 and -0.02% for fat percentage in Canada, the U.S.A. and Italy, respectively. Sires imported to Canada, the U.S.A. and Italy had average evaluations different from those of local sires within countries by 126, -211 and 45 kg for milk yield, 4.2, -6.4 and 2.0 kg for fat yield, and -0.01, 0.02 and 0.01% for fat percentage, respectively. The estimation of international sire breeding values on a routine basis was shown to be feasible, but many problems still remain to be solved. Besides the need for sufficient links between countries, the primary problem is that of accounting for non-random mating of sires within each country.