Most regions on earth have a long history of human land use. Throughout Europe and eastern North America a large proportion of forests grow on land abandoned from agriculture during the last few centuries, while other forests have persisted for much longer. What are the effects of past human land use on present-day plants and animals, and how long can we expect such legacies to persist? These are critical questions for understanding how organisms are distributed across landscapes, and for guiding future land use decisions with the goal of maintaining biodiversity, or at least minimizing its loss. By comparing the distributions and diversity of plants in forests with different histories of land use across landscapes in both Europe and North America, we have revealed ecological legacies of land use that may persist for decades or even centuries. For example, when a patch of forest becomes isolated because forest in the surrounding landscape is cleared, we expect species to be lost because small isolated populations are prone to local extinction. This is indeed the case, but we have found that even after 100 years following a dramatic decline in forest cover, the process of species loss is still ongoing. The process appears to take so long because perennial plants can persist for decades even after conditions for their growth have become detrimental. We have also found that such "extinction-resistant" plants are also quite slow to colonize new forests, such that it may require decades or centuries for a post-agricultural forest to accumulate the diversity observed in older forests.