A randomisation test of the null hypothesis that two cladograms are sample estimates of a parametric phylogenetic tree

Cladograms for the same group of taxa derived using different datasets often agree extensively but are seldom identical. This disagreement may be due to the fact that cladograms are sampling estimates of the true phylogeny and as a result may differ only because of sampling error. A protocol is proposed to test the null hypothesis that two trees estimate the true or parametric phylogeny and are no more different than would be expected due to sampling error. In the event that the null hypothesis is rejected, the datasets are pruned to remove potentially confounding information, and the test of the null hypothesis is repeated. If the null hypothesis cannot be rejected, a method for combining the cladistic information from both datasets is proposed that takes account of the variability of the cladistic structure. The procedure is illustrated using morphological and molecular data from genera of the sponge Order Hadromerida (Porifera: Demospongiae).