Abstract - Estimation of (Co)Variances of Test Day Yields for First Lactation Holsteins in the United States

Estimation of (Co)Variances of Test Day Yields for First Lactation Holsteins in the United States

N. GENGLER,*^† A. TIJANI,^† G. R. WIGGANS,^‡ C. P. VAN TASSELL,^‡and J. C. PHILPOT^‡
*Fonds National Belge de la Recherche Scientifique and
^†Unité de Zootechnie, Faculté Universitaire des Sciences Agronomiques, B-5030 Gembloux, Belgium
^‡Animal Improvement Programs Laboratory, Agricultural Research Service, USDA, Beltsville, MD 20705-2350

ABSTRACT (Co)variance components for milk, fat, and protein yields during first lactation were estimated from data for test days from 23,029 Holstein cows from 37 herds in Pennsylvania and Wisconsin. Four lactation stages of 75 d were defined, and the test day nearest the center of each interval was used. In four analyses, a lactation stage was added, and observations with missing values were deleted; 17,190 observations were available for the final analysis of lactations with test days in all lactation stages. Missing values were deleted because a canonical transformation was used for estimation of (co)variance matrices. Heritability estimates were similar across analyses, which indicated little effect from selection. Heritabilities usually increased with lactation stage and were highest for milk; mean heritability estimates were 0.19 for milk, 0.14 for fat, and 0.16 for protein. Phenotypic and genetic correlations were higher between milk and protein than between milk and fat. Within a yield trait, genetic correlation declined from ≥ 524;0.90 for adjacent stages to 0.75 for milk and protein and to 0.82 for fat between initial and final lactation stages. Within lactation stage, mean genetic correlations were 0.40 between milk and fat, 0.78 between milk and protein, and 0.56 between fat and protein; corresponding mean phenotypic correlations were 0.64, 0.91, and 0.66. The effect of solving the model iteratively was examined with records that had been adjusted using solutions from fitting the full model. Heritabilities for the beginning of lactation increased slightly with the iterative solution, which indicated a better model fit.

Key Words: variance component estimation, test day model, multitrait evaluation, heritability

1999 J. Dairy Sci. 82:225