Measuring genetic change in a dairy herd using a control population

A selection experiment was conducted to assess the change in production resulting from selection with artificial insemination using a randomly maintained control. The experiment continued for 16 yr with approximately 20 lactations in each of the two breeding groups annually. Selection and control production was compared using 1) intraseason herdmate comparisons (weighted) and 2) mixed model maximum likelihood estimates of year-season effects (adjusted). The regressions of differences in milk yield on years for the two approaches were 110 kg for milk (weighted) and 108 kg for milk (adjusted), corrected for inbreeding. The regressions for fat were 3.9 kg for fat (weighted) and 4.2 kg for fat (adjusted). These regression coefficients were 1.6 and 1.5% of the least squares means for milk (6987 kg) and fat (259 kg), respectively. Definitive trends were not evident for differences between the two groups in percent fat, percent SNF, days open, final type classification score, or heart girth. Even with the limited numbers in the closed control population, inbreeding, and nonrandomness in culling of females or in the choice of dams of bulls appeared to have little impact on control mean during the 16 yr.