This study examined how the size of host pupae of Musca domestica L. and exposure of host spiracles affected parasitism and sex ratio of Muscidifurax zaraptor Kogan & Legner (Hymenoptera: Pteromalidae). Based on a model by Charnov et al. (Nature [Lond.] 289: 27–33, 1981), we predicted that a lower proportion of female offspring would be produced on large hosts when large hosts were presented alone than when both small and large hosts were presented simultaneously. This prediction was supported, suggesting that females used relative cues for determining host size. We also predicted a greater proportion of female eggs would be produced on small hosts when only small hosts were presented than when both small and large hosts were presented simultaneously. This prediction was not supported, suggesting that females also used absolute cues for assessing host size. Host size had no significant effect on the thorax widths of offspring. However, the head widths of offspring varied with pupal size, with wasps emerging from large pupae having larger thoraces. There was a difference in parasitism depending on pupal orientation; house fly pupae that had posterior spiracles exposed (i.e., anterior spiracles concealed) were less often parasitized than when the anterior spiracles were exposed. Pupal position and sex ratio were not related. Finally, females spent less time drilling through posterior spiracles than through anterior spiracles.