Many animals are regarded as relatively sedentary and specialized in marginal parts of their geographical distributions. They are expected to be slow at colonizing new habitats. Despite this, the cool margins of many species' distributions have expanded rapidly in association with recent climate warming. We examined four insect species that have expanded their geographical ranges in Britain over the past 20 years. Here we report that two butterfly species have increased the variety of habitat types that they can colonize, and that two bush cricket species show increased fractions of longer-winged (dispersive) individuals in recently founded populations. Both ecological and evolutionary processes are probably responsible for these changes. Increased habitat breadth and dispersal tendencies have resulted in about 3- to 15-fold increases in expansion rates, allowing these insects to cross habitat disjunctions that would have represented major or complete barriers to dispersal before the expansions started. The emergence of dispersive phenotypes will increase the speed at which species invade new environments, and probably underlies the responses of many species to both past and future climate change.

The dispersal patterns of animals are important in metapopulation ecology because they a¡ect the dynamics and survival of populations. Theoretical models assume random dispersal but little is known in practice about the dispersal behaviour of individual animals or the strategy by which dispersers locate distant habitat patches. In the present study, we released individual meadow brown butter£ies (Maniola jurtina) in a non-habitat and investigated their ability to return to a suitable habitat. The results provided three reasons for supposing that meadow brown butter£ies do not seek habitat by means of random £ight. First, when released within the range of their normal dispersal distances, the butter£ies orientated towards suitable habitat at a higher rate than expected at random. Second, when released at larger distances from their habitat, they used a non-random, systematic, search strategy in which they £ew in loops around the release point and returned periodically to it. Third, butter£ies returned to a familiar habitat patch rather than a non-familiar one when given a choice. If dispersers actively orientate towards or search systematically for distant habitat, this may be problematic for existing metapopulation models, including models of the evolution of dispersal rates in metapopulations.

Generalist species and phenotypes are expected to perform best under rapid environmental change. In contrast to this view that generalists will inherit the Earth, we find that increased use of a single host plant is associated with the recent climate-driven range expansion of the UK brown argus butterfly. Field assays of female host plant preference across the UK reveal a diversity of adaptations to host plants in long-established parts of the range, whereas butterflies in recently colonized areas are more specialized, consistently preferring to lay eggs on one host plant species that is geographically widespread throughout the region of expansion, despite being locally rare. By common-garden rearing of females' offspring, we also show an increase in dispersal propensity associated with the colonization of new sites. Range expansion is therefore associated with an increase in the spatial scale of adaptation as dispersive specialists selectively spread into new regions. Major restructuring of patterns of local adaptation is likely to occur across many taxa with climate change, as lineages suited to regional colonization rather than local success emerge and expand.

Abstract. 1. The incidence of parasitism by larvae of the mite species Trombidium breei was reported in one population of the lycaenid butterfly Polyommatus icarus, four populations of the satyrine butterfly Maniola jurtina, one population of the satyrine butterfly Aphantopus hyperanthus, and two populations of the satyrine butterfly Pyronia tithonus, as well as on one specimen of the dipteran Alophorus hemiptera. A considerable proportion of butterflies (11±50%) was infested in all study populations.2. The pattern of infestation was examined in detail in M. jurtina. Males had a significantly higher incidence of infestation than females, and middle-aged butterflies had a higher incidence of infestation than old or young butterflies. The incidence of infestation peaked in the middle of the flight season, and this seasonal effect was independent of the effect of butterfly age.3. Using a model based on capture±recapture data, it was estimated that a hypothetical ideal male M. jurtina that lives exactly the mean expected lifespan of 9±10 days has an approximately 75% chance of becoming infested with mites at least once during its lifetime, a mean time to first infestation of 3±4 days, and an average infestation persistence time of 2±3 days.4. Capture±recapture data failed to show any effect of mite infestation on the lifespan or within-habitat movement rate of M. jurtina.5. In experiments in which individual butterflies were taken out of their normal habitat and released, M. jurtina and P. tithonus that were infested with mite larvae did not differ from uninfested individuals in the efficiency with which they returned to suitable habitat. Thus, parasitism by T. breei larvae had no detectable effects on flight performance or orientation ability.6. The results suggest that trombidiid mite larvae have limited potential in the biological control of insect pests.