The initial colonization of a host by symbionts, ranging from parasites to mutualists, can generate priority effects that alter within-host interactions and the trajectory of parasite community assembly. At the same time, variation in parasite communities among hosts can also stem from stochastic processes. Community ecology theory posits that multiple processes (e.g. dispersal, selection and drift) interact to generate variation in community structure, but these processes are rarely considered simultaneously during community assembly. To test the role of these processes in a parasite community, we experimentally simulated dispersal of three symbionts by factorially inoculating individual plants of tall fescue with two foliar fungal parasites, Colletotrichum cereale and Rhizoctonia solani, and a hypothesized mutualist endophyte, Epichloë coenophiala. We then tracked parasite infections longitudinally in the field. After the initial inoculations, hosts were exposed to a common pool of parasites in the field, which we expected to cause parasite communities to converge towards a similar community state. To test for convergence, we analyzed individual hosts’ parasite community trajectories in multivariate space. In contrast to our expectation, there was no signal of convergence. Instead, parasite community trajectories generally diverged over time between treatment groups and the magnitude of divergence depended on the symbiont species inoculated. Parasite communities of hosts that were inoculated with only the mutualist, Epichloë, showed significant trends of divergence relative to all other symbiont inoculation treatments. In contrast, hosts inoculated with only Rhizoctonia did not exhibit clear trends of divergence when compared to other parasite inoculations. Further, co-inoculation with both parasite species resulted in faster rates of divergence and greater temporal change in parasite communities relative to hosts inoculated with only the parasite Colletotrichum. As predicted by existing theory, parasite communities showed evidence of drift during the beginning of the experiment, which contributed to among-host divergence in parasite community structure. Overall, these data provide evidence that initial dispersal of symbionts produced persistent changes in parasite community structure via ecological selection, that drift was important during the early stages of parasite community assembly, and together, dispersal, selection and drift resulted in parasite community divergence.Open Research statementThe data and code that support the findings of this study are available through Zenodo at https://doi.org/10.5281/zenodo.5714452

Host individuals are commonly coinfected with multiple parasite species that may interact to shape within-host parasite community structure. In addition to within-host species interactions, parasite communities may also be structured by other processes like dispersal and ecological drift. The timing of dispersal (in particular, the temporal sequence in which parasite species infect a host individual) can alter within-host species interactions, setting the stage for historical contingency by priority effects, but how persistently such effects drive the trajectory of parasite community assembly is unclear, particularly under continued dispersal and ecological drift. We tested the role of species interactions under continued dispersal and ecological drift by simultaneously inoculating individual plants of tall fescue with a factorial combination of three symbionts (two foliar fungal parasites and a mutualistic endophyte), then deploying the plants in the field and tracking parasite communities as they assembled within host individuals. In the field, hosts were exposed to continued dispersal from a common pool of parasites, which should promote convergence in the structure of within-host parasite communities. Yet, analysis of parasite community trajectories found no signal of convergence. Instead, parasite community trajectories generally diverged from each other, and the magnitude of divergence depended on the initial composition of symbionts within each host, indicating historical contingency. Early in assembly, parasite communities also showed evidence of drift, revealing another source of among-host divergence in parasite community structure. Overall, these results show that both historical contingency and ecological drift contributed to divergence in parasite community assembly within hosts.