2019. A structural equation model linking past and present plant diversity in Alaska: a framework for evaluating future change.Abstract. Recent findings of peak plant species diversity occurring in alpine tundra in Alaska may reflect the filtering of the regional species pool during Pleistocene Epoch cold periods. Specifically, herbaceous plant groups and those with far northern geographic distributions centered in Beringia may have persisted coincident with repeated diminution of tall-statured and southern continental plant groups during cold intervals, thus potentially structuring current diversity patterns in relation to elevation. Numerous biotic and abiotic drivers of plant species distributions vary with elevation, raising questions concerning which of these drivers are responsible for these surprising diversity patterns. We used structural equation modeling to investigate relationships among a suite of biotic and abiotic covariates and the richness of species groups based on growth form (GF) and biogeographic affinity across interior Alaska. Our results confirm the primary importance of microclimate in controlling the distribution, abundance, and richness of woody plants, which increased in warm, south-exposed plots. In contrast, the suite of other site factors had considerably stronger combined direct influences on richness of forbs, graminoids, and dwarf shrubs as compared to microclimate. Thus, species groups with negative richness responses to increasing temperature and potential solar radiation may have a competitive disadvantage to woody species, suggesting competitive displacement as one mechanism for these patterns. Our findings reveal that conditions associated with the high plant diversity (including cooler air temperatures, ground disturbance, and weakly acidic soil pH) are spatially restricted as compared to widespread conditions associated with low species diversity (warm lowlands with acidic soil reaction). Our results confirm a strong correspondence between localized richness patterns and variation in the overall fractions of the GF and biogeographic groups within the regional species pool. Our work suggests important links between current and historical drivers of plant diversity that help explain the apparent contradictions suggested by the relatively limited spatial extent of species-rich habitats in Alaska. By elucidating the relationships among biotic and abiotic site characteristics and between these characteristics and richness, our findings provide a useful framework within which to evaluate possible changes to vegetation structure and species richness patterns caused by rapid warming.