Australia's western continental margin formed during Gondwana breakup between Australia – Antarctica and India, leaving behind an evolving plate boundary with significant new magmatic crust accreted to the continental fragments. In this context, the margin segment from the outer Exmouth Plateau to the Gascoyne Abyssal Plain has been repeatedly studied as an example of volcanic margin formation. However, many recent analyses based on selective interpretation of geophysical datasets have come to conflicting conclusions regarding both the tectonic architecture and the age of breakup on this margin. These differences profoundly impact on models for both the global plate circuit and for the timing and extent of magmatism implicated in volcanic breakup processes. We present new interpretations of the distribution of magmatic and pre-rift rock packages in this margin, based on the integrated interpretation of two deep crustal transects with existing seismic-reflection, refraction, gravity, magnetic and geochemical data. Interpretations are constrained by data from sparse Ocean Drilling Program and petroleum-exploration drilling, and dredging. We find evidence for significant accumulation of magmatic rocks and their clastic derivatives infilling extensional fault-controlled basins developed in a broad volcanic-margin transition zone between the outer Exmouth Plateau and true oceanic crust. These rocks have distinctive seismic facies in the form of seaward-dipping reflector sequences, and are dense and magnetised. Most significantly, these packages give rise to potential-field anomalies that have previously been interpreted as due to seafloor spreading. Recognition of these packages in a volcanic-margin transition zone has implications for the recognition of the inboard edge of unequivocal oceanic crust, the oceanic – volcanic-margin boundary. The main locus of igneous activity in the volcanic-margin transition zone off the Exmouth Plateau is spatially offset from a previously recognised high-velocity zone, suggesting that these two phenomena may not be temporally related. Seismically imaged differences in total thinning and partitioning of thinning, between upper and lower crust, provide support for models of depth-dependent thinning previously proposed for this margin.