Parasite-mediated selection can rapidly drive up resistance levels in host populations, but fixation of resistance traits may be prevented by costs of resistance. Black bean aphids (Aphis fabae) benefit from increased resistance to parasitoids when carrying the defensive bacterial endosymbiont Hamiltonella defensa. However, due to fitness costs that come with symbiont infection, symbiont-conferred resistance may result in either a net benefit or a net cost to the aphid host, depending on parasitoid presence as well as on the general ecological context. Balancing selection may therefore explain why in natural aphid populations, H. defensa is often found at intermediate frequencies. Here we present a two-year field study where we set out to look for signatures of balancing selection in natural aphid populations. We collected temporally well-resolved data on the prevalence of H. defensa in A.f. fabae and estimated the risk imposed by parasitoids using sentinel hosts. Despite a marked and consistent early-summer peak in parasitism risk and significant changes in symbiont prevalence over time, we found just a weak correlation between parasitism risk and H. defensa frequency dynamics. H. defensa prevalence in the populations under study was, in fact, better explained by the number of heat days that previous aphid generations were exposed to. Our study grants an unprecedentedly well-resolved insight into the dynamics of endosymbiont and parasitoid communities of A.f. fabae populations, and it adds to a growing body of empirical evidence suggesting that not only parasitism risk but rather multifarious selection is shaping H. defensa prevalence in the wild.

This data submission contains R scripts as well as raw & processed data to analyse the seasonal dynamics of aphid symbionts and parasitoids as described in the paper. See the README file for more information.

Data files are provided as .csv tables; the scripts can be run with R. To run analyses including temperature variables, temperature data has to be requested from MeteoSwiss (Swiss Federal Office of Meteorology and Climatology) as described in the README file.