Valentine and Moores [Valentine JW, Moores EM (1970) Nature 228:657-659] hypothesized that plate tectonics regulates global biodiversity by changing the geographic arrangement of continental crust, but the data required to fully test the hypothesis were not available. Here, we use a global database of marine animal fossil occurrences and a paleogeographic reconstruction model to test the hypothesis that temporal patterns of continental fragmentation have impacted global Phanerozoic biodiversity. We find a positive correlation between global marine invertebrate genus richness and an independently derived quantitative index describing the fragmentation of continental crust during supercontinental coalescence-breakup cycles. The observed positive correlation between global biodiversity and continental fragmentation is not readily attributable to commonly cited vagaries of the fossil record, including changing quantities of marine rock or time-variable sampling effort. Because many different environmental and biotic factors may covary with changes in the geographic arrangement of continental crust, it is difficult to identify a specific causal mechanism. However, cross-correlation indicates that the state of continental fragmentation at a given time is positively correlated with the state of global biodiversity for tens of millions of years afterward. There is also evidence to suggest that continental fragmentation promotes increasing marine richness, but that coalescence alone has only a small negative or stabilizing effect. Together, these results suggest that continental fragmentation, particularly during the Mesozoic breakup of the supercontinent Pangaea, has exerted a first-order control on the long-term trajectory of Phanerozoic marine animal diversity.paleogeography | paleobiology | biodiversity | biogeography