In reply: is the air in amber ancient?

reports that have appeared in Science (3, 4), these results merit further investigation. Bemer and Landis indicate that during the Late Cretaceous [75 to 95 million years ago (Ma)] the 02 content of the atmosphere was "greater than 30 percent," whereas in the Eocene-Oligocene values were 21%, as today. Lovelock (2) asserts that "at 25% oxygen even damp twigs and grass of a rainforest would ignite," and at above 30% 02 the danger would be even greater. Wildfires at the Cretaceous-Tertiary (KT) boundary (65 to 66 Ma) are suggested by Wolbach et al. (3) on the basis of clay samples from this boundary in Denmark, New Zealand, the North Pacific, and Spain, all ofwhich contain graphitic carbon. These workers suggest that the amount ofbiomass burned to produce the graphite in the boundary clay represents 10% of present biomass of carbon, or 4% of precivilization values. If atmospheric 02 values in the entire Late Cretaceous (95 to 65 Ma) were greater than 25 to 30%, then widespread fires should have been common; graphitic carbon should be found throughout this time, particularly in areas of low rates of sedimentation. If this is not found, then perhaps the data of Bemer and Landis (1) should be questioned. If abundant graphite is found, then perhaps the interpretation of Wolbach et al. (3), and for that matter, that ofAlvarez et al. (4), should be questioned. Alternatively, perhaps the high 02 level found by Bemer and Landis (1) and the threshold that Lovelock (2) cites for highly probable fires suggest that the globe was "poised" for massive conflagration during the Late Cretaceous. Either meteor impactinduced fires (3, 4), or some less dramatic event, might have touched off such fires. Obviously, a large meteor impact or other similar phenomenon occurring today would have grave consequences. However, one interesting negative consequence of considering the Alvarez et al. theory (4), the Bemer and Landis (1) and Wolbach et al. (3) data, and the Lovelock (2) assertion together is the realization that the present time may lack one important component (02 > 25%) to trigger a recurrence of the events that have been hypothesized to have caused the massivc extinctions at the K-T boundary. KEITH E. CHAVE Department ofOceanography, and Hawaii Institute ofGepAyics, Univiity ofHawaii, Honolulu, I-1 96822 STEPHEN V. SMITH Hawaii Institute ofMarine Biogy Univeitzy ofHawaii REFERENCES AND NOTES