In most areas of polydeformed crystalline basement, interpretation of structure in the third dimension is based solely on structural observations of incomplete surface exposures. One way to alleviate this problem is to use geophysical data and rock-property data to place additional constraints on structural interpretation. Geophysical data can provide information on the surface distribution of rock units, and can be used to constrain forward models of structural interpretations. Most available geophysical modelling routines, however, do not lend themselves to specification of complex structure in a geologically meaningful way. The Duchess area in the Mount Isa Inlier has been used as a test area for a structural and geophysical modelling package which allows specification and forward modelling of a multiphase deformation history. This study combines the input of field-based structural mapping, rock-property measurements, three-dimensional structural interpretation, and interpretation of a high-resolution aeromagnetic survey. The aim of the study was to use magnetic data to place constraints on the detailed three-dimensional structural interpretation of the area. The deformation history of the Duchess area consists of an early layer-parallel foliation and associated high-strain zones, which are overprinted by a regional east-west shortening event which resulted in the formation of upright folds and a steep foliation, with north-trending axial surfaces. These structures are overprinted by a series of northwest, northeast and north-trending faults which are also associated with east-west shortening. The approach in this study was to produce a preliminary three-dimensional interpretation based on surface mapping and magnetic interpretation. The results of this stage were then used, with constraints from magnetic susceptibility data, in a forward model of the magnetic anomaly field of the interpreted structure. The results of the model were then used to modify iteratively the interpretation until it was consistent with both surface observations and magnetic data. This approach has allowed additional constraints to be placed on: a) the subsurface geometry of folded magnetic units; b) absolute sense of displacement on faults; c) the surface and subsurface geometry of granitic bodies; and c) the nature and extent of deep structural features not exposed at the surface.