EFFECTIVENESS OF SELECTION FOR GROWTH TRAITS IN BEEF CATTLE

Data derived over a period of 21 years in which growth traits of animals in a herd of Angus cattle were recorded, were utilized in an analysis of the effects of selection. The primary selection criterion was weight at one year of age. In the early phase of the experiment, sire selection was wholly based on own performance and, in the later phase, some sires were selected for reuse based on the performance of their progeny. The selection objectives of the experiment were fairly constant over the 21 years. Genetic trends and parameters were estimated from data on 2576 calves. Data on calves from parents repeated in consecutive years were used to estimate genetic changes due to male and female selection. Annual genetic changes due to male selection were estimated to be .00 Kg, 2.09 Kg, 1.46 Kg and 2 gr for birth (BWT), weaning (WWT) and yearling (YWT) weights and postweaning daily gain (PWDG), respectively. Corresponding changes were .10 Kg, 1.66 Kg, 2.06 Kg and 2 gr due to female selection. Environmental trends were large and negative (-.13 Kg, -2.6 Kg, -1.85 Kg and -3 gr). Progeny mean deviations from year means were calculated for 38 sires with two or more progeny sets of twelve or more offspring and were used to calculate another estimate of genetic trend due to male selection of 0.04 Kg, 0.90 Kg, 2.00 Kg and 12 gr). Theoretical annual genetic progress was .15 Kg, 1.38 Kg, 2.87 Kg and 12 gr due to male selection and .07 Kg, .56 Kg, 1.18 Kg and 5 gr due to female selection. Heritabilities estimated by the paternal half-sib method were .58, .39, .51 and .51 BWT, WWT, YWT and PWDG in the male population and .46, .29, .28 and .45 in the female population. Estimated heritabilities by regression of offspring on parent were small. Genetic correlations of BWT with WWT, YWT and PWDG were .47, .51 and .64, of WWT with YWT and PWDG were .91 and .37 and PWDG with YWT was .93. Repeatabilities of BWT, WWT, YWT and PWDG as a trait of the dam (calculated by correlation between early and later performance) ranged from .20 to .36, .13 to .35, -.07 to .09 and .05 to .13, respectively. Repeatabilities estimated by dam variance component were .21, .24, .17 and .13 for the corresponding traits. Sire progeny sets were expressed as a deviation from the year mean. Correlations between first and later progeny sets were high for BWT. Only a few of the correlations for WWT, YWT and PWDG were statistically significant. With exception of the correlation between first and second progeny sets all correlation of first with later progeny sets were negative for WWT and YWT. Correlations of own performance and breeding value with progeny mean performance were high for BWT and small or negative for WWT, YWT and PWDG. A theoretical study was made for one and two-stage selection using own performance and progeny test with h(\u272) = .37 and genetic standard deviation = 15. With the same mating rate for bulls selected under both criteria of selection (5 to 105 cows/bull) the progeny test produced little improvement in annual genetic change (3.6% in the best case) over selection based on own performance alone. The most important factor contributing to optimal annual genetic progess is a high selection intensity for young bulls based on own performance. Low levels of bull reuse (highest own performance bulls used a second time and highest first progeny test bulls used a third time) resulted in only small increases in genetic change. Estimates of observed genetic change agreed well with theoretical expectations