The functional significance of communal nesting and nursing is poorly
understood. Female house mice often communally nest, and within these
communal nests females appear to indiscriminately nurse all pups, a ra
re trait for any mammal. In this study, the hypothesis that communal n
esting provides protection from conspecific infanticide was tested and
supported in semi-natural populations of house mice. Conspecific infa
nticide in single-mother nests (69%, N=412) was twice that in communal
nests (33%, N=508). Because this major benefit of communal nesting do
es not require communal nursing, direct benefits to communal nursing i
tself were tested. Most proposed benefits should result in heavier wea
ning weights, but no differences were found between communal and singl
e nests in the semi-natural populations. If communal nursing is to be
avoided in communal nests, dams must recognize their own pups. Retriev
al tests conducted in the laboratory produced equivocal results. Dams
discriminated between pups that differed in age, but not between their
own and other age-matched pups. The major survival advantage of commu
nal nesting, coupled with the failure to find nutritional advantages f
or communally nursed pups, supports a recent suggestion that communal
nursing is an unavoidable consequence of communal nesting. This hypoth
esis is further strengthened by data indicating that communal nesting
partners tend to be kin, thereby providing inclusive fitness benefits
to communal nursing. Although costs of communal nursing were proposed
and tested, no such costs were found. We also show from 15 observation
s of infanticide that all classes of adults (territorial and non-terri
torial males, pregnant and non-pregnant females) are infanticidal. The
se observations are in conflict with previous laboratory studies. (C)
1995 The Association for the Study of Animal Behaviour