Better reproductive success of an invasive thistle than its native relative under pollinator exclusion.

Abstract

Invasive plants often can self-pollinate and have higher reproductive outputs than native counterparts. Pollinator declines may exacerbate disparities in reproductive output by negatively impacting native plants more than invasives. To determine how pollinator availability affects reproductive success of two species (one native, one invasive), we conducted a pollinator exclusion experiment for two functionally similar species: the invasive musk thistle Carduus nutans and native field thistle Cirsium discolour. We manipulated pollinator access to flowers by partially or completely excluding pollinators and evaluated how pollinator visitation rates and community composition influenced reproductive success. Both thistle species received pollinators from all seven functional groups that we considered, but pollinator community composition differed by thistle species and treatment. Sweat bees and butterflies were important drivers in community differences between the Ca. nutans and the Ci. discolour flower heads. Complete pollinator exclusion resulted in a higher probability of total reproductive failure; 55% of total pollinator exclusion flowers failed while <7% failed in other treatments. When flower heads produced at least one viable achene, any level of pollinator exclusion resulted in lower seed output, but germination success did not differ from the control. Overall, Ca. nutans had higher reproductive success per flower head than native Ci. discolour in control and partial pollinator exclusion due to higher seed output rather than higher germination rates. While pollinator loss will be detrimental to reproduction of both species, our study provides evidence that reproductive losses in invasive Ca. nutans will be lower than the native Ci. discolour.