Potential for coevolution in a host plant : holoparasitic plant interaction

Abstract

Host - parasite interaction is one of the most common ecological interactions. The aim of this thesis was to examine the potential for coevolution in a host plant (Urtica dioica) - holoparasitic plant (Cuscuta europaea) interaction. An infection experiment suggested that previous parasitism may have selected for quantitative resistance in the host as indicated by the lower parasite performance on hosts from parasitized populations compared to that on hosts from unparasitized populations. In a common garden experiment, previous parasitism was associated with lower allocation to asexual reproduction in the host, indicating that selection by the parasite may have modified this trait. According to a reciprocal cross-infection experiment, parasites had overall higher infectivity on their sympatric hosts than on allopatric hosts, but the level of local adaptation varied among populations. An infection experiment among different host families demonstrated genetic variation in resistance and tolerance as well as costs of resistance and tolerance. Based on allozyme data, gene flow among populations was high. However, local selection seems to be strong since I found evidence for parasite local adaptation, and divergence in host resistance and tolerance among unparasitized and parasitized populations. In summary, coevolution seems possible in this system. I found differences in host and/ or parasite traits between "novel" and "familiar" host – parasite combinations (i.e. between unparasitized and parasitized populations, and between allopatric and sympatric combinations). These results indicate effects of past selection by parasitism. Further, I found genetic variation in host's resistance and tolerance, which is a prerequisite for the evolution of these traits.