Variation in Milk Fat Production in State-Owned Herds in Iowa

This investigation was of 12,623 records made by 4,487 cows from 1940 through 1956 in 12 state-owned herds in Iowa.IIerd differences and herd-by-year interactions were responsible for about 6 and 10% of the total variance, respectively.Years, seasons, and the other interactions among these three mainly environmental factors were of minor importance individually; together they accounted for 4% of the total variance.The records were corrected for the major environmental effects by taking each record as a deviation from the average of all records made in that herd in that same year.Then, for obtaining sire and dam components~ etc, only the cows whose sires and dams each had at least two daughters in the herd were used.This analysis involved 7,003 records by 2,398 daughters of 295 sires and was made in three different ways: (1) hierarchal classification of sires, dams within sires, etc.; (2) full sister analysis; and (3) cross classification by sires and by dams.Four times the sire component in the first method gave 0.16 as the intraherd heritability of differences in butterfat production.By the second method, twice the full sister component gave 0.12 as the estimate.By the first method 1.6% of the total variance was attributed to dominance deviations and other things, such as maternal effects, that make full sisters resemble each other more than twice as much as paternal half-sisters.In the cross classification, the component for sireby-dam interaction was 1.8%.These results suggest that deviations from the additive scheme of inheritance are unimportant, but these estimates of nonadditive genetic effects were not precise.Butterfat production, as exhibited by an individual cow, is the response of its genotype for this characteristic under the particular environment in which the individual is raised and makes the records.When a population is subjected to selection, its average genotype changes only slowly within short periods of time because: (1) the number of gene pairs involved is large, (2) selection among females can not be intense, and (3) the selections can not be as accurate as they can be with characters less affected by environment and with characters phenotypically expressable ia males, also.During recent decades, milk production per cow in the U.S.A. has increased progressively; for instance, from 4,167 lb. in 1924 (18) to 6,438 lb. in 1959 (19).This advance has surely been brought about by improving both the genetic merit of cattle and their feeding and management.One can not specify certainly the relative role of these factors.