A former theory of combustion in which all flammable objects were supposed to contain a substance called phlogiston, which was released when the object burned. The existence of this hypothetical substance was proposed in 1669 by Johann Becher, who called it `combustible earth' (terra pinguis: literally ‘fat earth’). For example, according to Becher, the conversion of wood to ashes by burning was explained on the assumption that the original wood consisted of ash and terra pinguis, which was released on burning. In the early 18th century Georg Stahl renamed the substance phlogiston (from the Greek for ‘burned’) and extended the theory to include the calcination (and corrosion) of metals. Thus, metals were thought to be composed of calx (a powdery residue) and phlogiston; when a metal was heated, phlogiston was set free and the calx remained. The process could be reversed by heating the metal over charcoal (a substance believed to be rich in phlogiston, because combustion almost totally consumed it). The calx would absorb the phlogiston released by the burning charcoal and become metallic again.
The theory was finally demolished by Antoine Lavoisier, who showed by careful experiments with reactions in closed containers that there was no absolute gain in mass – the gain in mass of the substance was matched by a corresponding loss in mass of the air used in combustion. After experiments with Priestley's dephlogisticated air, Lavoisier realized that this gas, which he named oxygen, was taken up to form a calx (now called an oxide). The role of oxygen in the new theory was almost exactly the opposite of phlogiston's role in the old. In combustion and corrosion phlogiston was released; in the modern theory, oxygen is taken up to form an oxide.