Impact of a social parasite on ant host populations depends on host species,habitat and year

Parasites often affect the abundance and life‐history traits of their hosts. We studied the impact of a social parasite – a slavemaking ant – on host ant communities using two complementary field manipulations. In the first experiment, we analysed the effect of social parasite presence on host populations in one habitat. In a second experiment, conducted in two habitats, we used a cross‐fostering design, analysing the effect of sympatric and allopatric social parasites. In the first experiment, host colonies benefited to some extent from residing in parasite‐free areas, showing increased total production. Yet, in the second experiment, host colonies in plots containing social parasites were more productive, and this effect was most evident in response to allopatric social parasites. We propose several explanations for these inconsistent results, which are related to environmental variability. The discrepancies between the two habitats can be explained well by ecological variation as a result of differences in altitudes and climate. For example, ant colonies in the colder habitat were larger and, for one host species, colonies were more often polygynous. In addition, our long‐term documentation – a total of four measurements of community structure in 6 years – showed temporal variation in abundance and life‐history traits of ant colonies, unrelated to the manipulations. © 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, 103 , 559–570.