Chapter two surveys the "cacophony of causes" that have been suggested for the PT extinction, in the context of the other great extinctions. Erwin covers six major theories: an extraterrestrial impact (of the kind now accepted as an explanation of the Cretaceous-Tertiary extinction); volcanism and the eruption of the Siberian flood basalts; continental drift and a reduction in biomes with the formation of Pangaea; a drop in sea level resulting from glaciation; some kind of oceanic anoxia; and the release of methane. And then of course there's the "all or several of the above" option...
Five chapters then present the evidence we have and the science behind it. Some of the best fossil evidence for the extinction comes from Meishan in China; biostratigraphy using ammonoids, conodonts and other widespread species has revealed the structure of the Permian-Triassic boundary. Complementing this, Potassium-Argon and Uranium-Lead radiometric dating, using zircons from volcanic ash beds, has given us absolute dates for the extinction.
In the oceans, the extinction marks the divide between the Paleozoic Fauna and the Modern Evolutionary Fauna. Vulnerability may have depended on ecology (filter feeders), physiology (low metabolic rates), and association with reefs; other possible factors include sea levels and anoxia. On land, Erwin focuses on vertebrate extinctions in the Karoo of South Africa, but touches on other regions and on plants (and coals) and insects. He also considers geological changes, such as a shift from meandering to braided rivers, and a putative "fungal spike" immediately after the Permian-Triassic boundary. Important evidence comes from shifts in carbon isotope ratios, and to a lesser extent from sulfur and strontium isotopes.
All this evidence can be used to rule out some of the suggested causes, but several candidates remain possible. Erwin's conclusion is that "we do not know, or at least I do not know", but he favours the Siberian flood basalts over an impact, and the nexus of "acid raid produced by sulfuric aerosols, massive releases of carbon dioxide from volcanic eruptions and the destruction of coal beds with consequent cooling then warming, and possible impact of the thermal pulse at very high latitudes".
Erwin then turns to events after the extinction. Early Triassic fossils are dominated by a few opportunist species, and "not until the Spathian, 4 million years after the mass extinction, is there a demonstrable increase in speciation". There are also "Lazarus taxa" that seem to disappear only to reappear much later. Erwin looks at some of the key species and explores whether ecological models of "recovery and rebound" are applicable to mass extinctions, when niches may need to be reconstructed rather than simply refilled.
The final chapter sets the Permian-Triassic extinction in broader perspective. Major radiations followed the extinction: all modern sea urchins, for example, are derived from only two species which survived. But the Triassic produced no new phyla or even classes, with an apparent lack of "substantial morphological innovations", and it has been argued that the extinction had little effect on longer-term trends.
Extinction stays focused on the science, with just a bit of biography and travel and no dramatisation. It does a good job of mixing basic science with historical narrative, and details with broader perspective. It is accessible, informative, and recommended to anyone curious about the history of life on Earth.