Differential Owl Predation on White and Agouti Mus musculus

DICE (1947) demonstrated selection against conspicuous phenotypes of mice (Peromyscus maniculatus) by owls.His experiments were conducted in a laboratory room with either a Barn Owl (Tyto alba) or a Long-eared Owl (Asio wilsonianus) as predators.Vegetation was simulated by a latticework of boards.Subsequently Kaufman (1974) found that Barn Owls and Screech Owls (Otus asio) selected against conspicuous phenotypes of old-field mice (Peromyscus polionotus) in field enclosures with natural vegetation.It then became of interest to examine the effectiveness of selection against conspicuous prey whose conspicuousness was greatly increased; this was done by using white and agouti house mice (Mus musculus).In field experiments white mice (Mus musculus) disappeared faster than agouti mice during 3-and 6-day periods in a 1acre enclosure (Kaufman and Wagner 1973).Under similar vegetation conditions, wild Loggerhead Shrikes (Lanius ludovicianus), diurnal predators, captured more white than agouti Mus released in pairs near the birds (Kaufman 1973a).Here, I report the results of a study designed to test the effectiveness of nocturnal predators, Barn Owls and Screech Owls, in capturing the conspicuous phenotype with white and agouti Mus used as prey.Effect of the amount of vegetation and changes in illumination on effectiveness of owls are also discussed.MATERIALS AND METIIODS Pairs of house mice (Mus musculus) of the same sex, one white and one agouti, were simultaneously placed in experimental enclosures with a background of either light or dark brown soil with natural vegetation (3.6 X 9.0 X 3.9 m, described in Kaufman 1971).The phenotype of the first mouse captured and time elapsed from release to capture were recorded.The second mouse was removed from the enclosure.Both mice were removed when owls failed to capture a mouse within 15 minutes.Trials were repeated each evening until each owl stopped preying on the mice.Four Barn Owls and seven Screech Owls were used during the experiments.Owls were maintained on a varied diet of live and dead white, agouti, and light brown house mice, light and dark brown old-field mice, and beef heart to reduce complications caused by a specific searching image (concept defined in Tinbergen 1960) for one type of mice.Illumination during trials was measured with an International Light IL 600/IL 660 Low .Level Photometer connected to a single channel recorder.Intensity of night light varied from 2.8 X 10 6 to 8.5 • 10 -3 lm'ft 2. Light intensity was arbitrarily divided into three classes (1) 2.8 X 10 -4 to 8.5 • 10 -a lm'ft 2, (2) 2.8 • 10 -a to 2.8 X 10 -4 lm'ft 2, and (3) 2.8 • 10 6 to 2.8 X 10 -3 lm'ft 2 for analysis of the effect of light intensity on owl predation.In general terms, light class 1 represented 145The Auk 91: 145-150.