The first animals to eat plants on land appeared later than you might expect. Vertebrates colonized land by the late Devonian, roughly 375 million years ago. Plants were already there, providing food, shelter, and a three-dimensional habitat. But for tens of millions of years, the earliest land vertebrates ate each other, ate insects, ate whatever moved. Nobody ate the plants.
Published in Nature Ecology and Evolution, researchers described Tyrannoroter heberti, a Late Carboniferous microsaur from Cape Breton Island, Nova Scotia, approximately 307 million years old. Its palatal and coronoid dental batteries show wear patterns consistent with both shearing and grinding plant matter. Historical fossils of related pantylid microsaurs push similar herbivorous adaptations back to approximately 318 million years ago — less than 60 million years after the first terrestrial vertebrates.
The structural insight is about the gap between opportunity and exploitation. Plants were the most abundant food source on land for tens of millions of years before any vertebrate evolved to eat them. The opportunity was not the constraint — the adaptation was. Herbivory requires specialized dentition (grinding surfaces, not piercing teeth), specialized digestion (fermenting cellulose, which no vertebrate enzyme can break down), and behavioral changes (stationary feeding instead of pursuit). Each adaptation is costly. The cost was not worth paying when animal prey was abundant.
When herbivory did appear, it appeared rapidly and independently in multiple lineages almost simultaneously. This suggests it was not a difficult innovation waiting for a lucky mutation but a latent possibility held back by ecological conditions that changed. Once conditions favored plant-eating — perhaps increased plant biomass, perhaps increased competition for animal prey — multiple lineages converged on the same solution within a few million years. The constraint was not capability. It was incentive.