The aim of this paper is to investigate the longevity and diversity of the Amazonian rainforest and to assess its likely future. Palaeoclimate and palaeoecological records suggest that the Amazon rainforest originated in the late Cretaceous and has been a permanent feature of South America for at least the last 55 million years. The Amazon rainforest has survived the high temperatures of the Early Eocene climate optimum, the gradual Cenozoic cooling, and the drier and lower carbon dioxide levels of the Quaternary glacial periods. Two new theories for the great diversity of the Amazon rainforest are discussed – the canopy density hypothesis and the precessional-forced seasonality hypothesis. We suggest the Amazon rainforest should not be viewed as a geologically ephemeral feature of South America, but rather as a constant feature of the global Cenozoic biosphere. The forest is now, however, entering a set of climatic conditions with no past analogue. The predicted future hotter and more arid tropical climates may have a disastrous effect on the Amazon rainforest.