Five Measures of a Sire

SUMMARY Lactation milk and fat records of 6,949 daughters, 6,201 different dams, and the contemporary D.H.I.A. herd averages of 2,420 herds were studied, to compare five measures of a dairy site's transmitting ability. They were the daughter average, daughter-dam difference, equal-parent index, daughter-contemporary herd difference, and daughter-contemporary-herd index. The daughters and dams were included in the D.H.I.A. proofs for 350 Guernsey and 486 Holstein sires used in artificial-breeding associations, and whose proofs were published for 1949-1951. A variance component analysis did not reveal any marked superiority for any of the measures, as determined by the percentage of the total variance comprised by the sire component for each of the measures. Sufficient data for the Holsteins were available for an empirical check of the reliability of these five measures. The simple daughter average appeared to be about as reliable as the equal-parent index or the daughter-herd index to predict future daughters' production for milk or fat. The D.H.I.A. herd average included the daughter in question and reduced the magnitude of the sire component for the herd measures. Use of a contemporary-herd average, which excludes the daughter in question in the two daughter-herd measures, should improve their merit as measures of a dairy sire's breeding value compared to the other measures in this study. The performance of later daughters of sires selected for use in artificial-breeding associations, on the basis of proofs in one or two herds (natural), usually has reflected little of their sire's original apparent superiority (10, 11). Much of the disappointment in the performance of the artificially sired daughters undoubtedly has resulted because conventional methods of evaluating sire proofs have not adequately accounted for the influence of herd differences and changes in intra-herd environment. The use of the contemporary herd average has been suggested to avoid this glaring weakness (2, 8, 21). Most of the differences between herds are environmental (15, 18, 19) and, presumably, most of the intra-herd environmental changes are reflected by the herd average. Inasmuch as the herd average is readily available, could it be substituted for the dam's average in tile daughter-dam difference and ill the equal-parent index ? If so, ulany more daughter-herd comparisons than daughter-dam comparisons would be per-mitred, since about one-half of the tested daughters have tested dams. For these reasons, five measures of a sire's transmitting ability for milk and fat production, namely, the daughter average (0), daughter-dam difference