Early Triassic coals are unknown, and Middle Triassic coals are rare and thin. The Early Triassic coal gap began with extinction of peat-forming plants at the end of the Permian (ca. 250 Ma), with no coal known anywhere until Middle Triassic (243 Ma). Permian levels of plant diversity and peat thickness were not recovered until Late Triassic (230 Ma). Tectonic and climatic explanations for the coal gap fail because deposits of fluctuating sea levels and sedimentary facies and paleosols commonly found in coal-bearing sequences are present also in Early Triassic rocks. Nor do we favor explanations involving evolutionary advances in the effectiveness of fungal decomposers, insects or tetrapod herbivores, which became cosmopolitan and much reduced in diversity across the Permian-Triassic boundary. Instead, we favor explanations involving extinction of peat-forming plants at the Permian-Triassic boundary, followed by a hiatus of some 10 m.y. until newly evolved peat-forming plants developed tolerance to the acidic dysaerobic conditions of wetlands. This view is compatible not only with the paleobotanical record of extinction of swamp plants, but also with indications of a terminal Permian productivity crash from δ13Corg and total organic carbon of both nonmarine and shallow marine shales.