Teeth on trial: What can dental morphology really tell us about hominin phylogeny?

Results of studies on cusp homology, experimental studies of dental growth and development and the dental morphology of new fossil hominins like Homo floresiensis and H. naledi force us to re-examine to what extent we can rely on dental morphological data to reconstruct evolutionary relationships. H. floresiensis has primitive deciduous lower canines and primitive permanent lower third and fourth premolars. However, its small, four-cusped lower molars are morphologically derived towards H. sapiens. Rather than indicating a unique phylogenetic link with H. sapiens, it is possible that the simplified molars of H. floresiensis are a result of diminutive tooth size, similar to that seen in the Middle Pleistocene hominins from Sima de los Huesos. But what about the opposite end of the spectrum? The talonid expansion observed in Paranthropus lower deciduous molars and lower permanent premolars and molars has been traditionally understood as derived characters that link P. robustus and P. boisei into a monophyletic clade. However, certain dental morphological characteristics of H. naledi force us to question this interpretation. The permanent lower third premolar and six out of nine deciduous teeth represented by that sample show greatest morphological similarity to P. robustus and/or A. africanus (ui1, li2, lc, udm1, ldm1 and ldm2). Yet, a number of other morphological traits are derived towards later Homo (e.g., lack of upper and lower molar accessory cusps) or unique within the hominin clade (upper molar cusp height and spacing). Here we present alternative ways to interpret these conflicting signals