Abstract
Melanism is one of the more extreme colour polymorphisms and is observed in many taxa. In insects, melanism appears to be positively associated with several physiological and/or ecological factors including crypsis, aposematism, thermoregulation, desiccation resistance and immune function. The New Zealand alpine tree wētā (Hemideina maori) is colour polymorphic with a yellow and a melanic morph. Melanic morphs occur sporadically throughout the distribution of the species. Two hypotheses, thermal melanism (dark-coloured insects can potentially absorb heat more quickly than lighter-coloured insects) and the melanisation-desiccation resistance hypothesis (darker cuticles are associated with decreased cuticular permeability and a reduction in cuticular water loss) have not been tested in H. maori. Melanism may have either evolved convergently in different lineages or melanic wētā may form a monophyletic group with respect to non-melanic wētā.
Some regions of South Island, including regions where H. maori are currently found, were effected by repeated cycles of Pleistocene glaciation. Glacial events influenced the phylogeographic structure of South Island taxa, although response to glaciation varies between co-distributed taxa. For example, many taxa retreated to refugia in the south and north of the island, resulting in an area of low endemicity in central South Island, while other taxa persisted in this area of low endemicity, possibly surviving in microrefugia.
Hemideina maori is distinct from six other typically arboreal Hemideina species in having adapted to a wholly terrestrial, montane life-style. Hemideina maori and H. ricta (the Banks Peninsula tree wētā) share many morphological, behavioural and physiological commonalities, while genetic analysis has also shown a close relationship. Such commonalities have led to the specific status of H. ricta being questioned.
We used phylogeographic analysis using the mtDNA gene cytochrome c oxidase I (COI) and thirty nuclear DNA (nuDNA) markers as well as investigating physiological factors to answer specific questions about H. maori and its relationship with H. ricta. Melanism appears to have evolved repeatedly, as it occurs sporadically in seven of nine mtDNA groups and all three nuDNA groups identified. Hemideina maori likely persisted in microrefugia in the central low-endemicity region as this region exhibits both substantial endemic mtDNA diversity and maintains a distinctive nuclear lineage. We also found evidence for repeated introgression of H. maori mtDNA into H. ricta. However, morphological and nuDNA differences suggest that H. ricta is best treated as a separate but closely related species to H. maori.
Melanic morphs had significantly lower rates of cuticular water loss than yellow morphs, thus supporting the melanisation-desiccation resistance hypothesis. However, melanic H. maori are typically found at lower (and presumably less-desiccating) elevations than yellow morphs, suggesting that desiccation resistance is not the primary adaptive mechanism for melanism in H. maori.
We did not find support for the thermal melanism hypothesis, with heat gain not significantly increased in melanic morphs, compared to yellow either when wētā were exposed to direct radiant heat (basking simulation) or to indirect heat (microhabitat simulation). Instead of colour, weight was the significant covariate, with heat gain slower in heavier wētā.
Neither the adaptive significance of melanism in H. maori, nor the molecular basis for melanism, have been established and identifying what these are may reward future research.