Abstract
Counts of Trichostrongylid eggs were made at weekly intervals for 12, weeks on fecal samples from 103 lambs constituting most of the 1939 lamb crop at the University of California, Davis. The mean square roots of counts were used as estimates of resistance to infestation. The distributional variance agreed with a theoretical Poisson variance, confirming the validity of using the square root transformation for these data. The experimental error, Sheep × Weeks, was significantly greater than the distributional error. Weeks of sampling and individual sheep differences were highly significant. Sheep, adjusted for other variables, was the appropriate error term for effects not involving time. The main effects of Sex, Breed and Sire showed significant differences while Class (mutton or wool) was not significant. Sex × Weeks and Breed-sire × Weeks, within class, were non-significant. Sex × Breed and Sex × Breed-sire were not significant within the mutton breeds. All other interactions were highly significant. Sex × Breed-sire lost significance when adjusted for disproportionality. The interaction of Sex and Class was significant. Females in the wool class were significantly more resistant than either the males in the wool class or the females in the mutton class. Males and females were not significantly different in the mutton class. Heritability by the half-sib correlation was .22 ± .41 for two Hampshire sires and .25±.2,1 for two Merino sires. Heritabilities for only the males were .74±1.25 and 2.54±1.67 respectively. The Hampshire females gave a negative value and no value was available for the Merinos. Hampshire ewes from the same sires in 1938 gave a figure of 2.24±0.30.

This publication has 0 references indexed in Scilit: