In house mice, post-partum maternal aggression against a potentially infanticidal male conspecific can be considered a parental investment act. As such, it constitutes a suitable experimental paradigm for testing parental investment allocation. In the present study, 60 nulliparous female albino mice were tested for maternal aggression on day 8 of the lactation period (single 5-min test exposure to a male intruder). Four experimental groups were created by manipulating the litter size. in two groups, litters were culled at birth to 4 and 8 pups respectively, without any further pup removal. In the third group, litters were culled at birth to 8 pups, but 4 further pups were removed 3–4 hours before the test. In the fourth group, litters were manipulated as in the third group except that the 4 pups were returned to the litter after a 10-min removal. Dams with 8 pups at the time of testing showed significantly higher scores of aggressive behaviour than dams with 4 pups. The females of the two groups that had 4 pups at the time of testing did not differ in the intensity of maternal aggression. These results indicate that female house mice defend the investment made in their litters according to expected benefits (i.e., offspring number) and not to cumulative past investment.