MOLECULAR AND MORPHOLOGICAL EVOLUTION WITHIN SMALL ISLANDS

Although explicit rules can be used to hypothesise an inter-island col onization sequence from a molecular phylogeny, and statistical methods can be used to assist in partitioning historical factors from current selective factors, this latter process has limited efficacy where the re are a limited number of islands. Studying numerous populations with in small, but heterogenous, islands allows a better understanding of t he factors causing geographic variation. Three main approaches have be en used to study within-island geographic variation using lizards on C anarian and Lesser Antillean archipelagos: (i) matrix correspondence t ests and their partial regression/correlation extensions on morphologi cal and molecular data; (ii) identification of within island patterns of morphological geographic variation paralleled on independent island s; and (iii) large scale field experiments on selection. These studies reveal that, even on small islands, 'island populations' may not be h omogeneous in morphology, or molecular phylogeny, and that natural sel ection for current ecological conditions appears to be a primary force influencing morphological population differentiation, irrespective of phylogenetic history.