Fig. 12

Evolutionary trends cannot be identified by reading across the tips. In addition to resulting in incorrect interpretations of relatedness (Fig. 11), reading across the tips can engender a false impression of evolutionary trends. For example, many readers confronted with the tree in a might be tempted to infer an evolutionary trend toward increased body size in snail species over time (or, in Fig. 11a, an increase in complexity or intelligence over time). Unfortunately, misinterpretations such as this can be found even in the primary scientific literature. Once again, this can be corrected simply by rotating a few internal nodes, as has been done in b, in which the topology is the same but where the supposed trend is no longer apparent. c shows evidence of a real evolutionary trend toward increased body size. The important consideration is internal branching: In this case, there is information about ancestral states (e.g., from fossils), and it is evident that in every branching event, the two descendant species have been larger than their shared ancestor. Despite this being a clear evolutionary trend, there is no pattern evident across the terminal nodes. Thus, reading across the tips can create apparent trends where there are none and can mask real trends that are strongly supported by historical information