We present the first empirical evidence that mammalian sex-ratio deviations result from variation in adult-weight sexual dimorphism via correlated effects on blastocyst development. Two selection lines of mice exhibiting high and low sexual dimorphism in adult weight showed correlated sexual weight differences at birth and at weaning, caused by relatively decelerated growth of males in the low line from before birth. The sex ratio at birth was significantly female-biased in the low line, and significantly lower than in the highly dimorphic line. Concomitantly, blastomere numbers were at significantly higher variance in the low than in the highly dimorphic line, owing to an increased frequency of slowly growing blastocysts. Since low-dimorphism mice produced more corpora lutea and more female pups than the high-dimorphism mice, but not more males, birth sex-ratio bias most parsimoniously resulted from the loss of slowly growing male blastocysts. This is in agreement with the observation that sex-ratio skews in mammals arise when timing of uterine responsiveness (i.e. its temporally limited capacity for implantation) varies in relation to sex-specific embryonic growth rates. Hence, natural mammalian sex-ratio variation that stems from developmental asynchrony might be a by-product of natural selection for sexual dimorphism in adult weight.