Differences in the herbivore community between a plant's native (specialists and generalists) and introduced range (almost exclusively generalists) may lead to the evolution of reduced allocation to defences against specialist herbivores in the introduced range, allowing for increased allocation to competitive ability and to defences against generalist herbivores. Following this logic, the introduction of biological control agents should reverse this evolutionary shift and select for plants with life-history traits that are more similar to those of plants in the native range than those of plants in the introduced range that have not been exposed to biological control. In a common garden experiment, we compared performance and resistance traits of tansy ragwort, Jacobaea vulgaris, among populations from the introduced range (New Zealand and North America) that have either been exposed to or grown free from the biological control agent Longitarsus jacobaeae. For comparison, we included populations from the native European range. We found lower levels of generalist-deterrent pyrrolizidine alkaloids (PAs) and of soluble phenolics in New Zealand populations with than in populations without exposure to L. jacobaeae, while the opposite pattern was detected among North American populations. Contrary to expectation, populations with exposure to L. jacobaeae revealed more feeding damage by L. jacobaeae than populations without exposure. Introduced populations had higher levels of PAs and reproductive output than native J. vulgaris populations. Jacobaea vulgaris was introduced in different parts of the world some 100-130 years ago, while L. jacobaeae was introduced only some 20-40 years ago. Hence, the larger differences observed between native and introduced populations, as compared to introduced C. Rapo Á U.