Performance of the herb Verbascum thapsus along environmental gradients in its native and non-native ranges

AimWe evaluated whether the performance of individuals and populations of the invasive plant Verbascum thapsus differs between its native and non-native ranges, across climate gradients, and in response to its position in a global-scaled niche model. LocationIndia (Kashmir) and Switzerland (native range) and Australia and USA (Hawaii, Montana and Oregon) (non-native range). MethodsWe measured population characteristics (density of flowering individuals, population size), plant traits (plant height, number of flowering branches) and seed germination for 50 populations of V. thapsus growing along elevational gradients in six mountain regions around the world (two in the native range, and four in the non-native range). We fitted linear regression models to determine the relationship of plant and population level performance variables to range, region, climate and probability of occurrence from a global niche model. ResultsPlant height, number of flowering branches and population density of V. thapsus did not differ consistently between the native and non-native ranges, but the area covered by populations in the non-native range was on average two orders of magnitude larger than the area of native populations. Within and among regions, individual plant performance traits responded variably to precipitation and mean annual temperature, depending on the climatic range of the observed populations; however, performance was greater for populations that had a greater modelled probability of occurrence. Seed germination under controlled conditions was highest between 20 and 35 degrees C and consistent across populations, although germination at low temperatures was substantially higher for native populations from Kashmir. Main conclusionsThe introduction of V. thapsus to its non-native range is not associated with consistent increases in plant performance. Instead, plant performance is more strongly influenced by position within the climate niche of the species, position along environmental gradients, and climatic or other differences among regions. We demonstrate that individual-level and population-level performance traits can yield different predictors of plant performance. Therefore, future studies comparing plant performance in native and non-native ranges should include both plant and population characteristics and should also sample the target species from multiple regions in both ranges and along environmental gradients that comprehensively represent the niche of the species.