ANTHER-SMUT DISEASE IN SILENE DIOICA: VARIATION IN SUSCEPTIBILITY AMONG GENOTYPES AND POPULATIONS, AND PATTERNS OF DISEASE WITHIN POPULATIONS.

In an archipelago in northern Sweden, populations of the perennial, dioecious, and insect-pollinated herb Silene dioica are commonly infected by the sterilising anther-smut fungus Microbotryum violaceum. The results from transplantation and inoculation experiments in this study show that variation between populations in the incidence of disease may partly be due to variation in resistance among populations. In the transplantation experiment in which plants were naturally exposed to the fungus, disease levels varied greatly among transplants from three healthy populations when transplanted to three diseased populations in the archipelago. Clear genotypic differences in susceptibility among 25 replicated genotypes of the host plants were found when inoculated manually with two different isolates. Susceptibility varied between 0-90%, but the two isolates used did not differ in inoculation success. The results also suggest a geographical structuring in resistance of the host and virulence of the fungus. First, disease levels among experimental plants from two of the disease-free populations of S. dioica (originating from inner and outer archipelago, respectively), were high when transplanted to a diseased population nearby, but low when transplanted far away. Second, regardless of origin, plants from all healthy populations became diseased in the diseased experimental populations located in the middle part of the archipelago. Due to isostatic land upheaval in the studied archipelago, there is a vertical age-axis within islands such that the highest point on an island is the oldest. Since this may affect the demography of the host, disease spread, and the dynamics of disease, spatial patterns in adult and seedling densities, and disease and spore deposition along the age-axis were studied within three diseased populations. A low incidence of disease was found in the young, low and old, high parts of the populations and a high incidence in the vertical, middle parts of the populations. The higher disease incidence in the middle part compared with the lower part of the population may reflect less disturbance and an increased probability of disease with age.