M.V.Lomonosov Moscow University, Zoological Muzeum, Bolshaya Nikitskaya ul. 6, Moscow 103009, Russia

Abstract

The author presents a new method of cladogram analysis that allow to reveal the distributions of character states reflecting the parallel (or convergent) origin of these states. As a result, concrete statistic distributions are distinguished. Gradistic distributions are characterized by power (=the number of independent events that can explain the given distribution) and intensity. The latter one is equal to the ratio of number of knots (including terminal groups) that possess the given state of a character to the total number of knots and terminal groups and to the number of independent events that must be postulated to explain the given distribution. The author compares and classifies discrete cladistic distributions using the ideas of cis- and trans- distribution and expression. The classification of concrete gradistic distributions allows to reveal the system of parallelism (=grads). Using this idea the author discusses the degree of progressivity and the rate of taxa evolution.