The Sahara Desert was recently proposed as the agent of vicariant speciation in the North African elephant-shrew (order Macroscelidea, Elephantulus rozeti). Based on evidence from molecular dating, however, this speciation event was estimated as having occurred in the Miocene, an early date of origin for an extant species, predating the dramatic climatic fluctuations of the Pleistocene. Here, we explore the possibility of more recent connections of populations of this species across the Sahara Desert that would have facilitated faunal exchange through the Last Glacial Maximum. We used ecological niche modelling based on current climatic datasets and known occurrence points to develop models of present-day potential distributions of E. rozeti and related taxa. Given that the Miocene scenario was proposed based on a single set of evidence, we here provide a partial test based on an independent analysis of present and past potential distributions. Our results corroborate the idea of vicariance, but indicate that distributional patterns and ecological characteristics are consistent with a more recent, post-Pleistocene vicariance across an increasingly arid Sahara. We emphasize the importance of multiple sets of evidence to be brought to bear on complex biogeographical questions, and offer models of palaeoclimatic and palaeoecological distributions and independent sources of such information to complement molecular systematic studies.