The nature of the purebred-crossbred genetic covariance

Abstract
SUMMARY Epistatic models are examined for conditions which result in a negative covariance between a sire's purebred and test-cross progenies. It is found that heterozygote superiority is not a necessary condition for its occurrence. The frequencies of the same allele in the selected population and in the tester should be divergent. The implications of using the covariance in prediction are discussed. The important result to note is that selection in the crossbred may not be optimum even with a negative covariance.
Keywords