In 1974 Mario Molina and Sherwood Rowland calculated that chlorofluorocarbons (CFCs), used in refrigeration and aerosol cans, could break apart in the stratosphere and destroy ozone. The chemical industry called the calculation theoretical; there was no evidence.

The evidence arrived in 1985 from Joe Farman and colleagues at the British Antarctic Survey: measurements kept for thirty years at Halley Station showed springtime ozone falling by more than forty per cent. The value was so extreme that NASA's satellite processing software had been discarding those readings as errors — reprocessed, the data showed the hole from orbit too. Chemistry operating on the surfaces of polar stratospheric clouds explained the mechanism.

The response was unprecedented in environmental politics for its speed: the Montreal Protocol, signed in 1987, phased out CFC production; the timetable was tightened through the 1990s, and all 198 countries ratified it. The ozone layer is now in a recovery expected to reach 1980 levels around the 2060s. A further gain is that the banned gases are also potent greenhouse gases, so the protocol's climate effect exceeds that of many agreements made for the climate. Why it worked is also debated: the problem was confined to one industry, substitutes were available, and the effect was dramatically visible in one region. Climate change offers none of those three conveniences.