There is overwhelming evidence, based on the distribution of distinctive sediments and fossils and oxygen isotope data, that the climate of the Mesozoic world was appreciably more equable than that of today, with no polar ice caps, but precise quantitative data are not available. Except for an episode of late Cretaceous cooling there is no good documentation of any significant change in global temperature distributions through the era. The distribution of coals and evaporites, together with other criteria, indicates a pattern of humid arid and zones appreciably different in important respects from that of today. During the Triassic and Jurassic, western Pangaea in low to middle latitudes was largely arid, but in the early Cretaceous the lands on the margin of the newly opening Central Atlantic and western Tethys experienced a humid climate. By late Cretaceous times arid zones had become very restricted in extent. Because of insufficient suitable data, attempts at climatic modelling have had only modest success, and only to a limited extent can the major long-term changes in climate between the Permian and the present be explained in terms of changing geography. The most probable explanation of Mesozoic equability is an increased atmospheric CO2 content. A number of enigmas remain, such as the existence of flourishing forests in polar palaeolatitudes. Whereas for the late Cenozoic short-term climatic changes can be related successfully to variations in the geometry and mechanics of the earth-sun system, there is a long way to go before comparable success can be claimed for the Mesozoic.