Several sedimentary basins in west Australia contain petroleum reservoirs of Late Permian or older age that are overlain by thick shaly sequences (400–2,000 m) that have been assigned an Early Triassic age. The age of the base of the Triassic shales has been, and continues to be, contentious with strata being variously ascribed to the latest Permian (Changhsingian Stage) or wholly within the earliest Triassic (Induan Stage). In the Perth Basin the Permian‑Triassic boundary appears to be located somewhere in the Hovea Member of the Kockatea Shale. In the Bonaparte Basin, the boundary would appear to be either in the uppermost Penguin Formation or at the boundary between the Penguin and Mairmull formations. The uncertainty of the boundary placement relates to the interpretation of the sedimentological, biostratigraphic and geochemical record in individual sections and basins. Major problems relate to the recognition, or even the presence of unconformities, complications related to the presence of reworked sediments and paleontological material (both conodonts and spore-pollen) and to the significance of geochemical shifts. The age of the basal Kockatea Shale (northern Perth Basin) and the basal Mt Goodwin Sub-group (Bonaparte Basin) is reassessed using palaeontological data, augmented by carbon isotopic measurements and geochemical analyses, supported by wireline log correlations and seismic profiles. The stratigraphy of the latest Permian to Early Triassic succession in the Bonaparte Basin is also revised, as is the nomenclature for the Early Triassic Arranoo Member of the Kockatea Shale in the northern Perth Basin. The Mt Goodwin Sub-group (new rank) is composed of the latest Permian Penguin Formation overlain by the Early Triassic Mairmull, Ascalon and Fishburn formations (all new).