Sexual conflict over shared traits - intralocus sexual conflict (IaSC) - may be common and consequential, but experimental tests of its relative magnitude are challenging and limited in number. We use a sex-limited selection experiment, designed to subject genomic haplotypes of Drosophila melanogaster to selection for male fitness without opposing selection acting on female fitness. Importantly, we use three novel base populations to compare results with those from the LHM population, the sole population investigated using this technique. In contrast with previous studies, we find that male fitness of haplotypes subject to male-limited selection (ML populations) are not consistently better than their matched (MC) controls when tested in the "wildtype" state. Males from ML lines did not outperform controls in competitive fitness assays, mate choice trials, fecundity induction or sperm offense tests. As predicted, genetic variation for male fitness was reduced, with low fitness haplotypes apparently removed by selection, but this was only surveyed in one replicate population pair and included a potential artefact in the protocol. Female fitness was markedly reduced by carriage of ML genomic haplotypes, as predicted by sexual antagonism. Hence, our results are only partially consistent with the IaSC hypothesis, raising questions about the relative contribution of sexual conflict to the standing genetic variation in these populations and the potential role of artefacts in the protocol that may have obscured our ability to detect IaSC.