Sexual-selection theories generally assume that mating preferences are heritable traits. However, there is substantial evidence that the rearing environment may be important for the development of mating preferences, indicating that they may be learnt, or modified by experience. The relative importance of such sexual imprinting across species remains largely unexplored. Here, we report results of a large-scale cross-fostering experiment in the wild in which nestling birds were raised by parents of a different species. We show that resulting sexual imprinting may have a negative effect on pairing success in one species (the great tit, Parus major), but not in two other species (the blue tit, P. caeruleus and the pied flycatcher, Ficedula hypoleuca). A remarkable variation thus seems to exist, even between species that are congeneric and have similar breeding ecologies. The cross-fostering resulted in heterospecific pairings between the two tit species (female blue tit breeding with male great tit), which has never, to our knowledge, been previously documented. However, the chicks fledging from these nests were all blue tit.