Heritability and repeatability of sexual performance scores of rams

Sexual performance has been subjectively measured with a libido test during screening of rams before public sale and breeding at the U.S. Sheep Experiment Station from 1990 to 2000. The objective of this study was to determine whether sexual performance was genetically influenced. Sexual performance scores ranged from 1 to 6 with scores increasing from sexually inactive to highly sexually active in the presence of estrous ewes. The overall average score was 3.5+/-0.02. Records from four breeds (Columbia, n = 807; Polypay, n = 1,668; Rambouillet, n = 1,208; and Targhee, n = 1,002) were combined into one analysis because breeds had similar phenotypic variances. Total number of records was 4,685, which included a second sexual performance test on 1,212 rams in the following year. Variance components were estimated using a REML algorithm. Fixed effects were breed of ram, selection line within breed, and year by breed. A permanent environmental effect for ram was included to account for repeated observations on individual animals. Age and weight of the rams at time of the libido test were linear covariates and were breed-specific. Adjusted means for sexual performance scores did not differ among breeds (P > 0.05). Age was a significant effect (P < 0.01), with sexual performance score increasing 0.05 units for each month of age. The additive genetic variance was estimated as 0.54. The estimate of variance due to ram permanent environmental effects was 1.19. The residual variance was estimated to be 0.67. The heritability estimate was moderate (0.22+/-0.04) and repeatability was high (0.72). These results imply that one screening for sexual performance provides a reliable measure of sexual performance and that favorable response to selection for ram serving capacity may be expected.