An analysis of the genetic implications of maternal and grandmaternal effects in beef cattle selection programs
Date of Award
Doctor of Philosophy
C.C. Melton, L.M. Josephson
A study utilizing 3,220 performance records of Angus calves dropped over a 19-year period from 1957 to 1975 was undertaken in an attempt to estimate the importance of direct, maternal, and grandmaternal variances and to evaluate their interrelationship as causative factors in creating phenotypic variation in birth weight, gain from birth to weaning, weaning condition, and weaning weights. All of these records were obtained from cattle at the Ames Plantation in Tennessee and were from non-creep-fed calves. The data were adjusted by least squares procedures for the effects of year of birth, season of birth, and age of dam. These adjusted data were used to calculate the various covariances among relatives. The model for maternal effects utilized the covariances of the individual with itself and paternal half-sibs, maternal half-sibs, full-sibs, dam-offspring, and granddam offspring covariances. While the model for assessing grandmaternal genetic influences utilized, in addition to the above six, covariances between cousins, within cousins, and within paternal half-sibs. All of these were equated to their expected biological components, direct genetic variance, maternal genetic variance, grandmaternal genetic variance, covariances between direct and maternal, between direct and grandmaternal, between maternal and grandmaternal, direct environmental variance, maternal environmental variance, and the covariance between direct and maternal environmental effects.
The maternal model yielded positive effects for all estimates of variances with the direct environmental variance contributing the largest fraction of the total phenotypic variances for all traits except adjusted weaning weight (6.6%). The estimates ranged quite high (up to 83.4% for birth weight); however, the heritability estimates are in line with accepted values for these traits. The direct estimates of variance, ranging from a low of 16.1% for weaning condition to a high of 41.5% for weaning weight, were, therefore, considered quite reasonable. Estimates of the maternal variance all tended to be low (from 1.4% to 4.5%); however, they are positive and do exist. The covariance between direct and maternal effects and the environmental covariances between direct and maternal effects exhibited negative signs except for the genetic covariance for birth weight (7.4% and 6.1%) and environmental covariance for weaning condition (16.4% and 17.0%). This negative covariance supports the theory of an antagonism existing between direct and maternal effects for the weaning and preweaning traits.
The grandmaternal model showed the variance estimates for all effects to be positive except adjusted weaning weight and adjusted gain (-1.2 and -1.2) for the maternal environmental variance. These estimates ranged from 20.2% for birth to 42.8% for weaning weight for direct effects, while the maternal variances were in the 6% to 16% range as to their influence upon the total phenotypic variance. The estimates of the grandmaternal variance were in the range of 5% to 10%, thus very evident as to their importance upon the total phenotypic variance. The estimates for grandmaternal genetic variance were all fairly large in magnitude and were thought to play an important role in validating the alternate generation phenomenon.
Brown, Dan Travis, "An analysis of the genetic implications of maternal and grandmaternal effects in beef cattle selection programs. " PhD diss., University of Tennessee, 1977.