Abstract
A hybrid zone between two distinctive colour forms of the alpine weta, Hemideina maori, was identified on the Rock and Pillar range. The distribution of yellow and black colour forms is clinal, with yellows predominating to the north of the mountain range, switching to blacks in the south. Intermediate coloured weta (forming a continuum between yellows and blacks) are found throughout the mountain range. A pilot survey of weta from well outside the contact area as well as inside the zone was conducted. Using isozyme electrophoresis, little variation was found and there were no marker alleles. RFLP analysis of an amplified mtDNA COII gene revealed two distinctive haplotypes that correlated completely with body colour in allopatry and nearly so in sympatry. Intermediates had either one or other haplotype.
A microsatellite library was generated and four primer pairs used to assess variation between black and yellow weta. Three loci were multiallelic, the fourth dialleleic. No alleles were useful for distinguishing the colour morphs but they have potential uses in examining population structure. Cross-species amplification was tested and was successful in the four species of Hemideina examined. A survey of 100 RAPD primers was unsuccessful, either in finding marker alleles or in repeatably amplifying the same bands. Potential confounding problems are discussed.
Clines for body colour and mtDNA were compared along two transects. In the first transect, both dines coincided and the transition between forms was abrupt. In the second transect, the transition was smoother. The difference is most likely explained by geographic factors. In the first transect, there is a break in suitable habitat (rock tors) - i.e. a density trough, while in the second transect, the distribution of rock tors is more even. An AMOV A revealed mtDNA was highly structured in the first transect, not only between colours, but also among sites (rock tors). Variation in the second transect was only found among sites within colours. Microsatellites also revealed evidence of significant structuring in the first transect, similar to that found in mtDNA but in the second transect there was no evidence of any structuring. Results for microsatellites varied depending on the method used to categorise the two taxa. When individuals were grouped on the basis of mtDNA haplotype, significant structure was found, whereas when body colour was included as a category, there was no significant variation. The criteria defined to allocate individuals to a particular colour category (yellow, intermediate-types, black) may not accurately reflect true taxonomic designations.
A comparison of mtDNA haplotypes found on the Rock and Pillar range with those from other populations revealed that Rock and Pillar black haplotypes cluster together as a monophyletic group.