Is parsimony a justified criterion in the assessment of possible phylogenetic reconstructions