Alpine-type mountain belts formed by continental collision are characterised by a strong cross-sectional asymmetry driven by the dominant underthrusting of one plate beneath the other. Such mountain belts are flanked on either side by two peripheral foreland basins, one over the underthrust plate and one over the over-riding plate; these have been termed pro- and retro-foreland basins, respectively. Numerical modelling that incorporates suitable tectonic boundary conditions, and models orogenesis from growth to a steady-state form (i.e. where accretionary influx equals erosional outflux), predicts contrasting basin development to these two end-member basin types. Pro-foreland basins are characterised by: (1) Accelerating tectonic subsidence driven primarily by the translation of the basin fill towards the mountain belt at the convergence rate. (2) Stratigraphic onlap onto the cratonic margin at a rate at least equal to the plate convergence rate. (3) A basin infill that records the most recent development of the mountain belt with a preserved interval determined by the width of the basin divided by the convergence rate. In contrast, retro-foreland basins are relatively stable, are not translated into the mountain belt once steady-state is achieved, and are consequently characterised by: (1) A constant tectonic subsidence rate during growth of the thrust wedge, with zero tectonic subsidence during the steady-state phase (i.e. ongoing accretion-erosion, but constant load). (2) Relatively little stratigraphic onlap driven only by the growth of the retro-wedge. (3) A basin fill that records the entire growth phase of the mountain belt, but only a condensed representation of steady-state conditions. Examples of pro-foreland basins include the Appalachian foredeep, the west Taiwan foreland basin, the North Alpine Foreland Basin and the Ebro Basin (southern Pyrenees). Examples of retro-foreland basins include the South Westland Basin (Southern Alps, New Zealand), the Aquitaine Basin (northern Pyrenees), and the Po Basin (southern European Alps). We discuss how this new insight into the variability of collisional foreland basins can be used to better interpret mountain belt evolution and the hydrocarbon potential of these basins types.