Group selection for adaptation to multiple-hen cages: beak-related mortality, feathering, and body weight responses.

The hypothesis was tested that selection on the basis of family means for increased survival and hen-housed egg production, when sisters with intact beaks were kept together in multiple-bird cage, would cause adaptive changes in behavior. Specifically, it was posited that beak-inflicted injuries causing cannibalistic mortality and feather loss and damage would be reduced. Body weight effects were not predicted, but were examined. Three stocks were compared; the Selected (S), representing the seventh generation of selection, the Randombred Control (C) from which S was derived, and a commercial stock (X), known to be highly productive and peak-trimmed by commercial producers. Pullets were placed in single-bird (1H) as well as in 12-hen (12H) cages using a completely randomized block experimental design. Mortality from beak-inflicted injuries differed among stocks in total hens lost (P < 0.005). Of 576 per stock in 12H cages 287, 128, and 46 replacements were used from 17 to 44 wk in X, C, and S, respectively, to maintain group size. The C and S hens also differed from 44 to 59 wk and 17 to 59 wk. X hens were not included in comparisons of mortality beyond 44 wk. Relative incidence of mortality caused by vent-cloacal injuries differed with X > C = S (P < 0.005 for X vs C and S). For cages with > or = 1 cannibalistic death, X had twice (P < 0.025) and C 1.6 times (P < 0.10) as many with repeated losses as S. Means and variances of feather scores were different for 1H vs 12H cages, ages, and genetic stocks. Greater variances were observed in 12H cages and among older birds. Within 1H units, genetic stocks did not differ in general, but in 12H cages X and C were always more variable than S. In 12H cages, mean feather scores and body weights were decreased and S hens had better feathering than either C or X. The evidence supported the hypothesis.