An alternative patterning mechanism for vertebrate tooth complexity

How development reliably generates complex patterns remains a central question in biology. Teeth provide an ideal system for addressing this problem, linking cusp-patterning mechanisms to diversification and the fossil record, yet the developmental basis of multicuspid crown patterning beyond mammals remains unresolved. In mammals, cusp initiation and spacing are controlled by transient epithelial signaling centers called enamel knots (EKs). Here, we combine comparative developmental analyses, 3D morphometrics, mutant perturbation, and Bayesian inverse modeling (BITES; Bayesian Inference for Tooth Emergence Simulation) to examine tooth morphogenesis across six squamate species spanning independent origins of multicuspid dentitions. We find that cusp patterning in these taxa does not rely on discrete EK organizers. Instead, cusps arise from a broad, persistent epithelial signaling field that deploys canonical tooth-patterning pathways but lacks the spatial restriction, apoptosis, and secondary organizers characteristic of the mammalian system. Elevated epithelial proliferation drives disproportionate expansion of this field during bud growth, enabling cusp addition. Quantitative variation in field extent and dynamics is sufficient to generate crown morphologies within and across species, providing a developmental basis for positional heterodonty and repeated gains and losses of multicuspidity in squamates. Our results identify an alternative epithelial patterning architecture for vertebrate cusp formation and suggest that the mammalian EK represents a derived, constrained implementation. Together, these findings revise how developmental complexity is encoded and show how shifts in patterning architecture can relax constraints and enable evolutionary convergence.