Genetic analysis of lactation persistency in the Polish Holstein-Friesian cows*

The objectives of this study were to analyze the genetic properties of three measures of lactation persistency in Polish Holstein-Friesian cows, and possibly to choose one measure which could be used for estimation of breeding value for lactation persistency in the Polish dairy cattle population. Data included 117,327 first three lactations of 110,141 cows calved in 1995-2009. The lactation curve model of Ali and Schaeffer was fitted to test-day milk yields. The first definition of persistency (P2:1) was milk yield in the second 100 days in milk (DIM) divided by yield in the first 100 DIM. The second definition (P3:1) was milk yield in the third 100 DIM divided by yield in the first 100 DIM, and the third definition (Pd) was milk yield at 280 DIM divided by milk yield at 60 DIM. The multipletrait REML method was applied for (co)variance component estimation. Heritabilities for three measures of persistency were very low, and ranged from 0.01 to 0.08. Genetic correlations were highest between P3:1 and Pd (0.96-0.99), and lowest between P2:1 and Pd (0.66-0.81), in the first three lactations. The correlations between 305-d milk yield and P3:1 or Pd in each of the first three lactations, and P2:1 in the second lactation, were negative and moderate. The phenotypic correlations between 305-d milk yield and persistency measures were low in the first three lactations. The phenotypic correlation between milk yield and Pd in each lactation was almost the same (0.140.15); the correlation between milk yield and P3:1 (0.11-0.17) or P2:1 (0.08-0.13) showed little variation in the first three lactations. All three compared measures of persistency were low-heritable and practically uncorrelated with total milk yield of 305-d lactations, so any of them could be used in the breeding program. However, the Pd measure could be recommended for use in practice because it is easy to calculate and interpret.