We demonstrate that much of the morphological and metrical data Lehtonen et al. (Journal of Biogeography, 2011, 38, 805–808) present in support of a closer relationship between humans and chimpanzees than between humans and orangutans are faulty. When the numerous invalid features and suggested character states are excluded, as they should be, the most robust theory of relationship that can be generated is between humans and orangutans. With regard to the direct optimization method (DOM) resolving problems in analysing molecular (sequence) data, the method, if valid, requires that demonstration of sequence similarity is per force a demonstration of synapomorphy. In brief, DOM fails both to test theories of relatedness and to take into account the fact that identification of shared similarity does not translate into demonstration of synapomorphy.