Abstract
Two parental strains of laboratory mice, NZB and CBA/H, were chosen for their differences in attack behavior. NZB have higher scores than CBA/H. An effect of the Y chromosome on attack behavior was determined for two maternal environments. Each male was tested once in a dyadic encounter with an A/J male as a standard opponent. The two reciprocal F1s and the four reciprocal backcrosses were used. In each group, the proportion of attacking males was used as the dependent variable. In the first experiment, the ovarian graft method was used to test for an effect of variation of the overall maternal environment: parental vs. F1. The results demonstrated an interaction between the Y chromosome and the maternal environment. By use of the adoption method, it was shown in the second experiment that this maternal effect was probably postnatal (and not prenatal).