SUMMARYCorrelations and regressions between estimated breeding values for 122 day part-lactation, 1st and 2nd lactation yield (part, 1st, 2nd) were estimated from 106 sire progeny groups, each with at least 60 effective daughters.The genetic correlations between part, 1st and 2nd kg milk were 1·0 and 0·84 respectively, and the regressions of 1st on part and of 2nd on 1st test were 2·40 and 0·82 kg/kg milk, respectively. Selection by one standard deviation between sires on part-lactation tests would improve 1st lactation yields by an average of 204 kg milk with a standard deviation of 84 kg. Selection on 1st lactation yields by one standard deviation would increase 2nd lactation yields by an average of 183 kg milk with a standard deviation of 156 kg, indicating that some bulls so selected could have below average 2nd lactation yields.Two sire selection models were simulated differing in the selection criterion; 1 st yield in model 1, and 1 st followed by 2nd yields in model 2. The contribution of proven sires to the rate of genetic improvement in lifetime production was greater from model 2 than from model 1 by about 15% without any additional costs for bull maintenance.
Sire selection and genetic improvement of dairy cattle assuming pure market competition
