Seedling establishment of Asteraceae forbs along altitudinal gradients: a comparison of transplant experiments in the native and introduced ranges

Aim Since ecological and evolutionary context changes when a plant species is introduced to a new area, it can be assumed that responses of alien plants to changing conditions along environmental gradients differ from those in their native range. Even if seed availability is not limited, the distribution of alien plants along such a gradient might still be restricted by their ability to germinate and establish as seedlings. In the present study, we aim at testing what factors promote or limit plant invasions during early establishment by using altitude as a model gradient. Location Altitudinal gradients in the Wallowa Mountains (Oregon, USA) and the Swiss Alps (Valais, Switzerland). Methods In transplant experiments along altitudinal gradients, we investigated the early establishment success of eight invasive alien Asteraceae species in their native and introduced ranges in the Wallowa Mountains and the Swiss Alps. Results Seedling recruitment was not restricted to relatively lower altitudes in the introduced range. In addition, we found no evidence for genetic adaptation along the altitudinal gradient in the introduced range, highlighting the importance of phenotypic flexibility for invasions. Furthermore, seedling recruitment was only enhanced by disturbance in the native range where vegetation was comparably dense but not in the introduced range. However, plant development was strongly delayed in the introduced range, probably due to low seasonal water availability. Main Conclusions We conclude that introduced plants, due to their ability to tolerate a wide range of environmental conditions, are not necessarily more restricted in their altitudinal limits than they are in their native range. Furthermore, due to other interacting factors (e.g. different competition situations among ranges), attempts to predict distributions of alien plants in the introduced range that are based on their distributions in the native range may be misleading.