Some bacterial endosymbionts selectively kill male embryos of their host to enhance their own transmitting efficiency. This male-killing strategy results in a large reduction in the number of host hatchlings in a patch. Moreover, in species with trophic egg provisioning (i.e., unhatched eggs in a clutch are consumed by sibling hatchlings), excess amount of maternal resource is reallocated to each hatchling through dead male embryos. Consequently, optimal values in hatchling number and trophic egg proportion are violated. In this case, it is expected that hosts may adjust clutch size and trophic egg proportion in response to male-killer infection. In this talk, I introduce the result of my experiment that evaluates this prediction by using a specialist ladybird Harmonia yedoensis.