Abstract

1. Arbuscular mycorrhizal (AM) symbiosis has been shown to enhance some plant traits to which pollinators are known to respond. Moreover, in gynodioecious species pollinators prefer hermaphrodite flowers over female ones, but the role of fungal symbiosis in sex-specific pollinator attraction is unknown.
2. We examined how plant gender dimorphism and AM symbiosis affect floral visitors and reproductive output in the gynodioecious plant Geranium sylvaticum. Floral visitors were moni- tored in a common garden experiment using mycorrhizal plants inoculated with either Glomus claroideum or Glomus hoi and in non-mycorrhizal condition.
3. We hypothesized that because of the larger flower display and pollen presence in hermaph- rodites, (i) hermaphrodites receive more floral visitors than females. However, as females pro- duce higher numbers of seeds than hermaphrodites, we predicted that (ii) effective pollinators visit both sexes equally and that (iii) AM symbiosis, although affecting floral characteristics and insect behaviour in general, should not have sex-specific effects on insects that transport pollen.
4. Our results confirm the hypotheses as hermaphrodites received 1Æ5 times as many floral visi- tors than females. However, the most important pollinator group, the bumblebees, did not prefer either sex, which may be important for the maintenance of both sexes in gynodioecious popula- tions. We further corroborated that AM symbiosis, although affecting flower size and the amount of pollen, did not affect bumblebee behaviour, whereas other hymenopterans showed a deterrence for G. hoi inoculated plants.
5. This work reports the impact of AM symbiosis on the behaviour of floral visitors in a sexually dimorphic plant species. Positive interactions between plants, floral visitors and AM have been documented previously. However, potential negative tritrophic interactions between plants, floral visitors and AM may also occur. Our results suggest that plant sex and the mycorrhizal status of the plant also influence floral visitor groups that are unlikely to carry out pollination. This may benefit the plants if these floral visitors carry fungal diseases and consume pollen and nectar without effectively vectoring pollen.