Genetic Influences on the Behavior of Chickens Associated with Welfare and Productivity - Genetics and the Behavior of Domestic Animals - page 328

of males was strongly associated with fitness in small flocks of non-game
breeds ( Guhl and Warren, 1946 ). However, in somewhat larger flocks with
higher densities, Kratzer and Craig (1980) found only a moderate correlation
of male social status and completed matings. With even larger flocks and
sex ratios similar to those used in commercial breeding flocks, Craig et al.
(1977) found that social status of White Leghorn cockerels appeared to have
little effect on frequency of mating.
Bidirectional selection for high and low social dominance ability, based
primarily on the outcomes of initial pair contests in both sexes, was carried
out over four generations in White Leghorns by Guhl et al. (1960) . In a simi-
lar study Craig et al. (1965) selected on the basis of pair contests between
males only over five generations within White Leghorns and Rhode Island
Reds. Although realized heritabilities were only moderate, very large differ-
ences between high and low strains were produced by the end of the studies
in strains derived from each of the three foundation stocks. In the more
comprehensive study of Craig et al. (1965) , there was no obvious
decrease in additive genetic variation over generations for the selected
trait. The availability of unselected random breeds from the foundation
stocks and testing of both males and females by Craig et al. (1965)
revealed that the responses were essentially symmetrical and were not sex
specific. Peck-order status of selected strain hens in intermingled-strain
flocks within breeds differed significantly, but the differences appeared to
be of reduced magnitude as compared with relative status based on initial
pair contests.
McBride (1962) , noting the results of Tindell and Craig (1959) , postu-
lated that high aggressiveness in a genetic stock would have the same effect
on productivity as he had hypothesized for poor husbandry ( McBride, 1960 ).
Specifically, he predicted that more aggressive strains would be under
greater stress and would have lower mean performance levels and greater
variances. Further, he postulated that strains with higher levels of aggres-
siveness would have a greater negative skew to their frequency distribution
curves for productivity traits than found in strains having lower aggres-
siveness ( McBride, 1968 ).
McBride's hypothesis was tested by Craig and Toth (1969) using hens of
the high and low social dominance strains from the White Leghorn and
Rhode Island Red strains developed by Craig et al. (1965) . Strains were kept
separately in floor pens and social instability was assured in half of the pens
of each strain by randomly redistributing birds among flocks weekly.
Although the unstable flocks had more fights, peck-avoidances, and threat-
avoidances than the stable flocks ( Craig et al., 1969 ), there was no evidence
of loss of productivity in the unstable flocks. A possible explanation
advanced was that the greater frequency of agonistic interactions in
unstable flocks were not necessarily associated with greater stress for all
hens because the frequent changes in group membership would benefit those