Assessing why two introduced Conyza differ in their ability to invade Mediterranean old fields

Researchers have suggested that species-community interactions determine invasion success. Therefore, it is likely that small biological differences between species interact with habitat characteristics to produce distinct patterns of distribution and abundance throughout a new range. In this study we test the hypothesis that differences in the distribution and abundance of species sharing an identical set of ''ideal weed characteristics'' are explicable in terms of species-specific responses to environmental variation within their new range. Using multifactor experiments, we investigated some of the ecological interactions influencing reproductive success in two very closely related species of annuals having invaded the French Mediterranean region for >150 yr and showing marked differences in their local distribution and abundance patterns. We transplanted seedlings of Conyza canadensis (a species restricted to recently disturbed areas) and C. sumatrensis (a species colonizing early- to mid-successional old fields) at equal densities in three contrasting old fields (6 mo, 4 yr, 17-yr abandonment, respectively) during 1991-1992, a growing season with average rainfall. Individual performances (measured as survivorship, reproductive timing, and reproductive output) were evaluated with respect to: (1) competition with plant neighbors (tested with a weeding treatment), (2) resource availability (tested with nutrient and water addition), and (3) herbivory (tested with chemical limitation). Manipulated factors interacted in a rather complex fashion to influence survivorship and reproduction in both species. However, patterns of relative performance were consistent with relative distribution patterns across Mediterranean landscapes: C. sumatrensis performed better than C. canadensis in all fields, including the youngest one (6 mo old). Herbivory only slightly affected transplant performances. In contrast, competition with plant neighbors had substantial effects on either Conyza species and may be the most important determinant of performance in Mediterranean old fields. The experiment showed unambiguously that the two species differ markedly in their competitive ability, with C. sumatrensis performing better than C. canadensis in the presence of neighboring vegetation. In addition, C. sumatrensis displayed a superior ability to take up and/or to use water and nutrient resources when they become available in competitive environments. We argue that potential physiological or anatomical species differences responsible for this differential susceptibility to local resource reduction by neighbors could involve differences in constructional organization, leaf morphology, and reproductive phenologies.