A striking feature of hymenopteran societies is the absence of male workers. Nevertheless, there are some reports of males feeding larvae in natural colonies of some primitively eusocial wasps. In the extensively studied primitively eusocial wasp Ropalidia marginata however, males have never been observed to feed larvae in natural colonies. This may be either because male R. marginata are incapable of feeding larvae or because they do not have opportunities to do so due to limited access to food and/or superior efficiency of females in feeding larvae. Here we have investigated the possible reasons for males not to feed larvae. By providing excess food and by removing the females, we have eliminated the potential reasons for males not to feed larvae. We find that under these conditions males feed larvae with probabilities and rates more than, or comparable to, those of females, suggesting, that lack of preadaptation to feed larvae does not explain the absence of male workers in social Hymenoptera. Although males are not as efficient as females in feeding larvae, they seem to be capable of doing enough for natural selection to have promoted the evolution of male workers if there were not other factors to prevent such evolution. Genetic relatedness asymmetry may be one such factor. A recent hypothesis concerning the increased susceptibility of haploid hymenopteran males to infection, is not supported because males handle and masticate prey for their own consumption and otherwise interact with the larvae. Since explanations based on relatedness asymmetries are believed to be inadequate, the absence of male workers remains a challenging unsolved problem in insect sociobiology.