The Role of Permutation Tail Probability Tests in Phylogenetic Systematics

Abstract

cladistic structure (Faith and Cranston, 1991) and Karl Popper's (1959) concept of corroboration. In so doing, authors ar? gued that Popper's probabilistic concepts have relevance for cladistics and suggested that PTP-style testing provides a mecha? nism whereby a Popperian justification for cladistic parsimony might be achieved. Faith and Cranston (1992) demonstrated that PTP testing is hypothetico-deductive and consistent with Popperian corrobora? tion. The cladistic structure of most-par? simonious cladograms is evaluated by testing of a null hypothesis that states that degree of cladistic structure, or con? gruence, in most-parsimonious clado? gram can be accounted for by a random? ized data set. The rejection of that null hypothesis results in purported cor? roboration of hypothesis that mostparsimonious cladogram is based on un? derlying hierarchical structure. Popper (1959:269-273) associated logical probabil? ity with a set of statements/' ac? cepted facts, or assumptions. In PTP test? ing, these basic statements are expressed in a null model. Popperian corroboration occurs with rejection of null hy? pothesis based on null model. In this instance the logical probability is low and the observations that are basis for