Premise of the study: Recent estimates of crown ages for cycad genera (Late Miocene) challenge us to consider what processes have produced the extant diversity of this ancient group in such relatively little time. Pleistocene climate change has driven major shifts in species distributions in Mexico and may have led to speciation in the genus Dioon by forcing populations to migrate up in elevation thereby becoming separated by topography. Methods: We inferred orthologs from transcriptomes of five species and sequenced these in 42 individuals representing all Dioon species. From these data and published plastid sequences we inferred dated species trees and lineage-specific diversification rates. Key results: 84 nuclear orthologs and plastid data confirm four major clades within Dioon, all of Pleistocene age. Gene tree analysis, divergence dates, and an increase in diversification rate support very recent and rapid divergence of extant taxa. Conclusions: This study confirms the Pleistocene age of Dioon species and implicates Pleistocene climate change and established topography in lineage spitting. These results add to our understanding of the cycads as evolutionarily dynamic lineages, not relicts or evolutionary dead ends. We also find that well-supported secondary calibration points can be reliable in the absence of fossils. Our hypothesis of lineage splitting mediated by habitat shifts may be applicable to other taxa that are restricted to elevation specific ecotones.