Aim Species' responses to climate change are likely to depend on their ability to overcome abiotic constraints as well as on the suite of species with which they interact. Responses to past climate change leave genetic signatures of range expansions and shifts, allowing inferences to be made about species' distributions in the past, which can improve our ability to predict the future. We tested a hypothesis of ongoing range shifting associated with climate change and involving interactions of two species inferred to exclude each other via competition.Location New Zealand.Methods The distributions of two tree weta species (Hemideina crassidens and H. thoracica) were mapped using locality records. We inferred the likely modern distribution of each species in the absence of congeneric competitors with the software Maxent. Range interaction between the two species on an elevational gradient was quantified by transect sampling. Patterns of genetic diversity were investigated using mitochondrial DNA, and hypotheses of range shifts were tested with population genetic metrics. ResultsThe realized ranges of H. thoracica and H. crassidens were narrower than their potential ranges, probably due to competitive interactions. Upper and lower elevational limits on Mount Taranaki over 15 years revealed expansion up the mountain for H. thoracica and a matching contraction of the low elevation limits of the range of H. crassidens. The observed nucleotide diversity in H. thoracica was consistent with a species that persisted in northern areas during Pleistocene glacial periods, from where it expanded at warmer times. In contrast, a two-tailed distribution of nucleotide diversity in H. crassidens was as expected for a species that expanded northwards during glacials and southwards during interglacials.Main conclusions Range shifts resulting from climate change involve complex species interactions. Competition among related species is an important factor limiting realized ranges. In New Zealand, H. thoracica is likely to continue to displace H. crassidens as human-induced global warming proceeds.

Local adaptation is inferred for many morphological and physiological traits but determining the role of natural selection in shaping geographical variation relies on evidence such as provided by fitness estimates or transplantation experiments. In addition, habitat‐specific convergent (or parallel) evolution provides a powerful means of testing adaptive hypotheses. In the present study, we contrast size, growth rate, and metabolic rate (as inferred by oxygen consumption) in a pair of Orthopteran species collected from high and low altitude locations and raised in identical environments. We find that two related insects (tree wētā: Hemideina crassidens and Hemideina thoracica) have the same (convergent) pattern of larger adults and faster growth rates in populations from a high altitude location compared to conspecifics from low altitude. However, variation in metabolic rate was detected only between species and not among altitudes. The high and low altitude populations of each species were collected from the same location; therefore, selection pressures on the two species are likely to be similar. Thus, the independent detection of larger adults and faster growth rate of wētā derived from high altitude suggests an adaptive role for both these traits in tree wētā. © 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, 113, 123–135.