Anthony Hallam writes: I was intrigued by Kerr's relating of a marine mass extinction event and associated sea-level rise, oceanic anoxia and global warming to an episode of oceanic volcanism on massive scale. What was especially interesting was his invocation of a possible runaway greenhouse climate developing rapidly as a consequence of the solubility of CO2 in seawater decreasing by 4% for every 1 °C temperature rise, so that the warmer the oceans get, the less CO2 will dissolve in them. The pattern of change envisaged by Kerr for the Cenomanian-Turonian boundary corresponds remarkably well with what Paul Wignall and I envisage for the biggest mass extinction event of all, at the Permian-Triassic boundary (Hallam & Wignall 1997). While the most fully documented extinctions were in the marine realm it is evident that there were also extensive extinctions on land, which are most reasonably explained by an episode of global warming. The best evidence comes from the floras. The cold-adapted glos-sopterids went extinct abruptly, to be replaced by warm temperate floras; cold temperature and polar-type floras are completely unknown at this time. Data from both oxygen and strontium isotopes are consistent with the inference of global warming.
We have speculated that eruption at this time of the Siberian Traps, the largest known continental flood basalt province, had a primary role in generating a significant increase in atmospheric CO2. However, we also need to invoke a mechanism for producing a strikingly rapid and pronounced eustatic sea-level rise, with the associated spread