Host density shapes the relative contribution of vector-based and aerial transmission of a pathogenic fungus.

Pathogen transmission mode is a key determinant of epidemiological outcomes. Theory shows that host density can influence the spread of pathogens differentially depending on their mode of transmission. Host density could therefore play an important role in selecting for pathogen transmission mode. We test theoretical expectations using floral arrays of the alpine carnation Dianthus pavonius in field experiments of spore dispersal of the anther-smut fungus, Microbotryum, by vector (pollinator)-based floral transmission and passive aerial transmission at a range of host densities. Pollinators deposited fewer spores per plant at high host density than at lower density (ranging from 0.2 to 2m spacing between plants), and vector-based spore deposition declined more steeply with distance from diseased plant sources at higher densities. In contrast, while aerial spore deposition declined with distance from the diseased source, the steepness of this decline was independent of host density. Our study indicates that the amount and distance of vector-based transmission are likely to be a non-monotonic function of host density as a result of vector behavior, which is not readily encapsulated by fixed dispersal functions. We conclude that the spatial spread of pathogens by vectors is likely to be greater at lower and intermediate densities, whereas spatial spread of aerially transmitted pathogens would be greater at high densities. These contrasting patterns could lead to a differential importance of each transmission mode in terms of its contribution to subsequent infections across host densities. This article is protected by copyright. All rights reserved.