Microgeographic variation in the colour pattern of the lizard Gallotia galloti within the island of Tenerife: distribution, pattern and hypothesis testing

This study of the microgeographic variation in the colour pattern of the lizard, Gallotia galloti, within Tenerife has three facets. One, an analysis of the population distribution and density in relation to physical conditions; two, univariate and multivariate descriptions of the pattern of geographic variation; and three, formal testing of a range of causal hypotheses for these geographic patterns. The range is not divided into separate allotheses high- and low-altitude populations by a inaptitude ring of ground-level cloud although there is a drop in population-density in the middle altitudes. Multiple regression indicates that, or the factors measured, this is primarily due to a decrease in insolation. The pattern of geographic variation in the six independent colour pattern characters is portrayed by contouring the 67 locality means. There is generally good congruence among these patterns and a canonical analysis indicates that the generalized pattern of geographic variation is largely unidimensional and can therefore be represented adequately by (and portrayed by contouring) the first canonical variate. Seven causal hypotheses, from a range of possible and previously suggested causes, are considered. The procedure for testing these hypotheses should depend on their dimensionality and that of the observed pattern. Since all the observed and hypothesized patterns are basically unidimensional the seven hypothesized patterns were simultaneously tested against each observed pattern (individual characters and canonical variate) by partial correlation. All the previously suggested 'phylogenetic' hypotheses based on divergence in allopatry are rejected. The hypotheses that the geographic variation in the colour pattern is caused by the topographically determined climate cannot be rejected. The character state changes can be explained by a balance between sexual selection for 'attractive' colouration and natural selection, via predation, for cryptic colouration.