Aim: Our aim was to test hypothesized biogeographical barriers using small-bodied terrestrial Uperoleia frogs, to identify Pleistocene refugia and to define biogeographical units. Location: The Australian Monsoonal Tropics, defined as the open woodlands and savanna north of the inland arid and eastern winter-rainfall zones. Methods: A multi-locus molecular phylogeny of the Uperoleia lithomoda, U. trachyderma and U. minima species complex, with supporting morphological and acoustic data, was generated to test species boundaries and clarify distributions. Ecological niche modelling with current climate and Last Glacial Maximum climate data was used to identify biogeographical units, barriers to dispersal, and regions of stability that may have served as Pleistocene refugia. Results: Our combined data supported five clades that comprise four allopatric species. Ecological niche models of the resolved species suggest that the Kimberley Plateau represents a distinct bioregion, the Top End extends from the edge of the Kimberley Plateau to the Carpentarian Gap, and the transition from sandstone escarpments to flat, sandy soils represents a major barrier to dispersal between the Top End and the Northern Deserts. The Northern Deserts were found to comprise two distinct subregions. Population- and species-level divergences were evident in a north-south line in the Northern Territory, representing a newly identified biogeographical break. Putative Pleistocene refugia were predicted in the north-west Kimberley Plateau, the western half of the Top End, the Selwyn Range and western Cape York. Main conclusions: By combining detailed genetic, morphological and acoustic data with newly developed statistical methods, we have delineated species boundaries, identified cryptic species and provided a region-wide assessment of the biogeography of northern Australia. We have identified previously unrecognized biogeographical barriers, better defined biogeographical regions, and proposed new hypotheses about the effects of Pleistocene climate cycles on the present-day diversity of northern Australia. Our work provides a solid foundation for the investigation of biogeographical patterns in other taxa.