Estimates of genetic parameters for milkability from automatic milking

Genetic parameters were estimated by restricted maximum likelihood with a multi-trait animal model for three milkability traits with serial data from an automatic milking system from a research farm (401 dairy cows) collected between September 2000 and June 2003 (320834 milkings). Furthermore, daily values for milk yield and milkability were formed from all single milkings resulting in 104 132 records and, subsequently, an estimation of genetic parameters was carried out based on these daily values.The resulting estimated heritability coefficients (based on daily values) are h² = 0.55, h² = 0.55 and h² = 0.39 for average milk flow, maximum milk flow and milking time, respectively. The heritabilities are at a high level and thus breeding for good milkability makes sense. The genetic correlations between the three milkability traits are near unity with r_g = 0.98 between average and maximum milk flow, r_g = − 0.89 between average milk flow and milking time and r_g = − 0.86 between maximum milk flow and milking time. Thus it may be sufficient to record only one of these traits in performance tests. The genetic correlations between milk yield and average milk flow, maximum milk flow and milking time are r_g = 0.51, r_g = 0.44 and r_g = − 0.23, respectively.In future, serial data on milkability, already existing on many farms with automatic milk yield recording, should be used to greater extent for selective breeding with the aim of achieving good milkability.