Abstract
The interrelation of sire x year interactions (SY) and direct-maternal genetic correlation (rdm) for weaning weight was examined for bias in estimating rdm. Weaning weight records were simulated using models containing SY (D1), rdm (D2), or both (D3). When D1 data were analyzed ignoring SY, a nonzero rdm was observed, and direct and maternal genetic variance estimates were inflated. Analysis of D2 data ignoring rdm did not reveal a spurious SY, and direct and maternal genetic variance estimates were deflated. On application to weaning weights of Simmental cattle, the model ignoring SY resulted in a direct-maternal genetic correlation estimate of -.29. The model using both SY and rdm fit the data better (P < .01). The SY variance represented only 3% of phenotypic variance but explained 62% of the covariance between direct and maternal genetic effects estimated ignoring SY. However, a negative estimate of the genetic correlation (-.14) was still obtained. Copyright © . .