Skip to main content
ARS Home » Research » Publications at this Location » Publication #83806


item Van Vleck, Lloyd

Submitted to: World Congress of Genetics Applied in Livestock Production
Publication Type: Proceedings
Publication Acceptance Date: 8/21/1997
Publication Date: N/A
Citation: N/A

Interpretive Summary: The basic measures of yield of dairy cows are taken approximately ten times during the lactations - the test day measurements. Traditionally these have been summed to estimate 305-d performance. Currently much research is being devoted to using the test day measurements directly in what are called "test-day models". Some of these alternatives are 1) to use each test day as a separate trait, which would result in approximately 30 milk, fat, and protein traits per lactation and 2) to use random regression functions based on the test day measurements to predict the lactation curve and hence, on summation, the 305 d yield. Comparison of these alternatives with use of traditional 305 d yields requires knowledge of the genetic relationships among yields at the different test days and across the traits of milk, fat, and protein yields. These genetic correlations were esti- mated from five samples of about 20,000 records each of Holstein cows in the Midwest. As might be expected, the correlations were large for measurements taken close together during the lactation (.97) and decreased to about .40 for those nine tests apart. Genetic correlations between same day milk and fat yields were small especially in mid lactation but were high for milk and protein yields. These estimates from 2,175 analyses of pairs of traits will allow examination of the feasibility of using a multiple trait (30 traits per lactation) approach to use of test day measurements for genetic evaluation.

Technical Abstract: Test day measurements on first lactation of Holstein cows were used to estimate genetic parameters for test day production traits. Traits were milk, fat or protein yields at test day, in 10 intervals based on days in milk. The data set was divided into five subsets and the traits were analyzed with a series of two-trait animal models to obtain REML covariance eestimates. Fixed effects were classes for days in milk within test day interval and age at calving, and herd-test date. Pooled heritabilities for test day traits were small, varying from .14 to .12 for protien. Each test-day yield was highly genetically correlated with the adjacent test day yield throughout the lactation. Among test day milk and protein yields, all correlations were consistently positive and high. Genetic correlations of mid lactation test day milk yields and all test day fat yields were small, and near zero. Genetic correlations of all test day fat yields were small, and near zero. Genetic correlations of all test da fat yields were small, and near zero. Genetic correlations of all test day protein yields with initial fat yields were positive and large and with fat yields in mid lactation were close to zero.