Pathogen dilution in experimental plant communities

For a generalist pathogen, understanding disease dynamics within any one host requires considering the community context in which the host and pathogen are embedded.  Since host species vary in epidemiological traits such as resistance, tolerance, and vector preference, the species richness and species composition of the community can influence disease spread.  The dilution hypothesis suggests that the effect of changing host community species richness on the prevalence of generalist pathogens depends on the identity of the species lost from or added to the community and the number of vectors acquiring the pathogen from that species.  In mesocosm experiments, we examined the role of host identity in controlling the prevalence of the vector-transmitted barley yellow dwarf virus in simple grass mixtures.  All grasss species were hosts of the virus, but they differed significantly in attractiveness to vectors and host competence (probability of an infected host individual transmitting the pathogen to a vector individual feeding on it).  We found that virus prevalence in highly competent hosts was reduced in mixtures with less competent hosts, but the degree of reduction was not always predicted by virus prevalence in host monocultures. Virus infection reduced the biomass of most species, but mixtures with less competent hosts buffered these effects for highly competent hosts.  These results illustrate the importance of community context in determining the impacts of generalist pathogens on plants.