The planthopper Prokelisia marginata Van Duzee is native to the eastern coast of North America, where densities on its foodplant, the cordgrass Spartina alterniflora, frequently exceed several thousand per square metre. It has little impact on its host plant in its native range where both species have co-evolved, however where the plant has been introduced and has had no recent exposure to the planthopper, it has a major impact and has been trialled as a biological control agent. P. marginata was recorded for the first time in Britain in 2008, where it feeds primarily on the widespread S. anglica, itself listed as one of the world's 100 worst invasive species, as well as its progenitors S. alterniflora, S. maritima and S. x townsendii. P. marginata appears to be in the successful early stages of invasion in Britain, having already spread extensively. Significantly outnumbering all other saltmarsh arthropod groups, it is benefitting from partial natural enemy escape, and a high proportion of macropterous individuals in all populations indicates strong potential for further range expansion. Utilising both glasshouse and field manipulations, I show that exposure to P. marginata has a significant negative impact on S. anglica, an interaction which has the potential to destabilise Britain's important saltmarsh habitat. I suggest that the four host species that occur in Britain represent a ‘gradient' of shared co-evolutionary history with the planthopper. I show that, whilst all species are negatively impacted by P. marginata exposure, S. alterniflora, the species with which it shares the longest co-evolutionary history, is the least affected. I further show that P. marginata exhibits a preference for, and performs better on, S. anglica. As S. anglica is by far the most abundant of the four Spartina species in Britain, these results suggest P. marginata may be undergoing rapid evolution in its new range to take advantage of this widespread host species, thereby maximising its potential for further range expansion.