Successful plant invasions are often explained by adaptation to novel environments. However, invasive species usually occupy broad niches within their native and introduced ranges. A true understanding of microevolution during invasion therefore requires a wide sampling of ranges, ideally with a knowledge of introduction history. We tested for genetic differentiation in herbivore resistance among 128 introduced (European, North American) and native (Chinese, Japanese) populations of the invasive Japanese knotweed (Reynoutria japonica) in two common gardens in the native range. In both common gardens, we found that the resistance traits of the introduced populations differed from most Chinese native populations but not from native populations in Japan, which are the two putative sources of introduction. Compared to Chinese populations, introduced European populations had thicker leaves in both common gardens with a lower C: N ratio but higher flavonoids content in the Shanghai garden. In the native range, variation in herbivore resistance was much more strongly associated with climate of origin than in introduced populations. Our results support that introducing particular resistance phenotypes from Japan played a key role in the invasion of knotweed into Europe and North America. Our study demonstrates how knowledge of introduction history can avoid misinterpretations of observed biogeographic divergence.