Optimal parental sex allocation depends on the balance between the costs of investing into sons vs. daughters and the benefits calculated as fitness returns. The outcome of this equation varies with the life history of the species, as well as the state of the individual and the quality of the environment.We studied maternal allocation and subsequent fecundity costs of bank voles, Myodes glareolus, by manipulating both the postnatal sex ratio (all-male/all-female litters) and the quality of rearing environment (through manipulation of litter size by −2/+2 pups) of their offspring in a laboratory setting.We found that mothers clearly biased their allocation to female rather than male offspring regardless of their own body condition. Male pups had a significantly lower growth rate than female pups, so that at weaning, males from enlarged litters were the smallest. Mothers produced more milk for female litters and also defended them more intensively than male offspring.The results agree with the predictions based on the bank vole life history: there will be selection for greater investment in daughters rather than sons, as a larger size seems to be more influencial for female reproductive success in this species. Our finding could be a general rule in highly polygynous, but weakly dimorphic small mammals where females are territorial.The results disagree with the narrow sense Trivers & Willard hypothesis, which states that in polygynous mammals that show higher variation in male than in female reproductive success, high-quality mothers are expected to invest more in sons than in daughters.