An evaluation of clade-shape statistics using simulations and extinct families of mammals

The use of clade-shape statistics, particularly center of gravity, on paraphyletic groups has been questioned since the introduction of these descriptive statistics. In addition, previous studies have found that early-arising groups of organisms are bottom-heavy in center of gravity relative to later-arising groups, leading to macroevolutionary hypotheses about the nature of diversification. Paraphyletic groups have been asserted to be inherently bottom heavy because of preconceived ideas as to how paraphyletic group formation affects center of gravity. In turn, early-arising groups of organisms have been asserted to be bottom-heavy because they also tend to be paraphyletic, thus calling into question the biological significance of the bottom-heaviness of early-arising groups. In this study, both computer simulations and the evolutionary history of families of mammals show that paraphyletic groups are inherently top-heavy. Simulated monophyletic clades and monophyletic families of mammals have centers of gravity near their expected positions at the midpoint of clade duration, while paraphyletic groups are significantly top-heavy. Early-arising families of mammals are found to be inherently top-heavy, rather than bottom-heavy as found by earlier studies. This new result is likely to be due to the use of updated stratigraphic range data and different taxonomic assignments rather than to bias due to inclusion of paraphyletic groups or bias due to unequal time-unit lengths in the Cenozoic time scale.