|
|
||||||||
University of Guelph1, Guelph, Ontario, Canada
Abstract
Henderson's best linear unbiased prediction procedures were used to evaluate 444 Charolais, 70 Simmental and 45 Limousin beef sires for 200-day weaning weight, gain during 165 days on test and yearling weight. Sire and error components of variance were estimated using an iterative MINQUE procedure. Data were from the Canadian Department of Agriculture and the Alberta Provincial Record of Performance Programs. Loss of data because of inaccurate sire identification, missing information and other edits was 44%. Different models were used for each trait and each breed was analyzed separately. Sire variances for Simmental and Limousin breeds were much smaller than for Charolais, but error variances were not greatly different among breeds. Standard errors of prediction were high indicating a need for 100 to 150 progeny per sire for accurate evaluation of differences among sires. Between-trait correlations of sire proofs were around .6 for weaning weight with yearling weight and .8 for gain on test with yearling weight for the same sex of calf, but between-sexes of calf correlations for the same trait or different traits were low.
1 Department of Animal and Poultry Science.
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |