We investigated the effects of contemporary and historical factors on the spatial variation of European dragonfly diversity. Specifically, we tested to what extent patterns of endemism and phylogenetic diversity of European dragonfly assemblages are structured by (i) phylogenetic conservatism of thermal adaptations and (ii) differences in the ability of post-glacial recolonization by species adapted to running waters (lotic) and still waters (lentic). We investigated patterns of dragonfly diversity using digital distribution maps and a phylogeny of 122 European dragonfly species, which we constructed by combining taxonomic and molecular data. We calculated total taxonomic distinctiveness and mean pairwise distances across 4,192 50 km × 50 km equal-area grid cells as measures of phylogenetic diversity. We compared species richness with corrected weighted endemism and standardized effect sizes of mean pairwise distances or residuals of total taxonomic distinctiveness to identify areas with higher or lower phylogenetic diversity than expected by chance. Broken-line regression was used to detect breakpoints in diversity–latitude relationships. Dragonfly species richness peaked in central Europe, whereas endemism and phylogenetic diversity decreased from warm areas in the south-west to cold areas in the north-east and with an increasing proportion of lentic species. Except for species richness, all measures of diversity were consistently higher in formerly unglaciated areas south of the 0°C isotherm during the Last Glacial Maximum than in formerly glaciated areas. These results indicate that the distributions of dragonfly species in Europe were shaped by both phylogenetic conservatism of thermal adaptations and differences between lentic and lotic species in the ability of post-glacial recolonization/dispersal in concert with the climatic history of the continent. The complex diversity patterns of European dragonflies provide an example of how integrating climatic and evolutionary history with contemporary ecological data can improve our understanding of the processes driving the geographical variation of biological diversity.