Traditionally managed meadows and pastures are increasingly abandoned or replaced with intensively cultivated arable fields in the UK and Europe. These habitats are important reservoirs of plants valued for their medicinal properties, cultural significance, and genetic diversity. Whether grassland biodiversity will persist without human management is currently unknown. Here, I combine palaeoecological data and modern vegetation surveys to determine the origin of species-rich grassland in Oxfordshire, England and long-term temporal and spatial changes in grassland biodiversity over time. I also examine whether forage crop wild relatives, a group of plant genetic resources, are dependent upon pastoral activities. In situ and ex situ conservation of crop wild relatives is increasingly important because they provide plant breeders with novel sources of genetic diversity to adapt crops to changing climate conditions. I demonstrate that species-rich grasslands only developed around 2000 years ago due to anthropogenic intervention and would revert to oscillating between open grassland and woodland without human management. Temporally and spatially, high levels of biodiversity are associated with the creation of species-rich grasslands. These areas are still hotspots of vegetation diversity today and wild relatives of forage grasses and legumes are closely linked to these areas. I argue that conservation activities should focus their attention on species-rich grasslands located in areas of long-term biodiversity, allowing other sites to revert, and that future research could focus on monitoring, characterizing, and collecting plant genetic resources from these sites.