The subfamily Carabinae is a diverse clade distributed across all biogeographical regions except Antarctica. In a seminal work, René Jeannel hypothesized a Gondwanan origin for this group, but this has hitherto remained untested with molecular data. We test this hypothesis by using a supermatrix approach. We also infer the most comprehensive phylogeny of the genus Calosoma, the only lineage within Carabinae comprising predominantly flying species. We use a recent timetree of Coleoptera to infer divergence time estimates in Carabinae. Our results identify four main lineages within Calosoma and reject the monophyly of several species groups erected by Jeannel. The subfamily Carabinae is estimated to have arisen in the Jurassic as suggested by Jeannel, and this dating is congruent, to some extent, with a vicariant hypothesis linked to the timing of the fragmentation of Gondwana. The main lineages of Calosoma are suggested to have diverged from each other in the Palaeogene, suggesting a dynamic biogeography, possibly shaped by dispersal rather than vicariance. This pattern could have resulted from the unique morphological evolution in Calosoma, allowing certain lineages to actively fly. Our divergence times within Carabinae are markedly inconsistent with previous studies, therefore reiterating the need for a fine-scale, fossil-based timetree of Adephaga.