Abstract
Biogeographical distribution is seen as a product of small- and large-scale factors, of which only the latter are important in palaeogeographical reconstruction. We review here previous methods for assessing the historical sequence of major events that have caused biogeographical distributions, and we argue that the most rigorous of these is cladistic biogeography. We also advocate our own new method, which uses parsimony analysis of taxa shared between sample localities. In not being dependent on prior phylogenetic analysis, this method is especially suitable for taxonomic groups which lack an adequate phylogeny.