| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 |

Oxygen and fire
Why did oxygen remain at 21 per cent, and not rise higher? I think the answer is fire. The correlation between oxygen abundance and flammability is steep. Below 15 per cent, nothing will burn: above 26 per cent combustion is instant and awesome fires would rage, destroying all forests. Charcoal layers in the geological record show that oxygen has long been above 15 per cent, and remains of ancient forests show that it has not exceeded 25 per cent. But how could the oxygen-fire relationship in practice act as a Gaian regulatory mechanism? An answer could lie in the fire ecology of forests: certain species, the conifers and eucalypts, do include fire in their evolutionary strategy; others do not. As with the dark and light daisies in Daisyworld, the competition for space between the trees could provide a feedback control on oxygen and fire. (see variation 35)

Oxygen—polluter or eutrophicator?
In the early stages of the Gaia hypothesis, Lynn Margulis and I often talked of the rise of free oxygen
in the atmosphere as if it were a disaster. We called it the first and greatest air pollution incident, where vast ranges of species of organisms were killed off by oxygen poisoning.

I now think that the opposite is true. Oxygen is poisonous, it is mutagenic and probably carcinogenic, and it thus sets a limit to life spans. But its presence also opens abundant new opportunities for organisms. At the end of the Archean, the appearance of a little free oxygen would have worked wonders for those early ecosystems. Oxygen was good for two reasons: first, its presence enabled consumers to recycle the organic matter of the photosynthesizers much more rapidly than could be done by the methanogens: and second, oxygen would have changed the environmental chemistry. The oxidation of atmospheric nitrogen to nitrates would have increased, as would the weathering of many rocks, particularly on the land surfaces. This would have made available nutrients that were previously scarce, and so allowed an increase in the abundance of life.

Geophysiological models of the change suggest that even the denizens of the anoxic regions would have benefited. This is simply because oxygen is, in a sense, a eutrophicator - that is, by increasing nutrient levels it encourages surface "blooms" of algae, and these deny light and oxygen to life below, which soon dies. The increased supply of dead plants in its presence would have enriched the methanogens until consumers took over this role.

(see variation 19 and note early form of the theory)