Several quantitative tests of a general hypothesis linking impacts of large asteroids and comets with mass extinctions of life are possible based on astronomical data, impact dynamics, and geological information. The waiting times of large-body impacts on the Earth derived from the flux of Earth-crossing asteroids and comets, and the estimated size of impacts capable of causing, large-scale environmental disasters, predict the impacts of objects > or = 5 km in diameter (> or = 10(7) Mt TNT equivalent) could be sufficient to explain the record of approximately 25 extinction pulses in the last 540 Myr, with the 5 recorded major mass extinctions related to impacts of the largest objects of > or = 10 km in diameter (> or = 10(8) Mt events). Smaller impacts (approximately 10(6) Mt), with significant regional environmental effects, could be responsible for the lesser boundaries in the geologic record. Tests of the "kill curve" relationship for impact-induced extinctions based on new data on extinction intensities, and several well-dated large impact craters, also suggest that major mass extinctions require large impacts, and that a step in the kill curve may exist at impacts that produce craters of approximately 100 km diameter, smaller impacts being capable of only relatively weak extinction pulses. Single impact craters less than approximately 60 km in diameter should not be associated with detectable global extinction pulses (although they may explain stage and zone boundaries marked by lesser faunal turnover), but multiple impacts in that size range may produce significant stepped extinction pulses. Statistical tests of the last occurrences of species at mass-extinction boundaries are generally consistent with predictions for abrupt or stepped extinctions, and several boundaries are known to show "catastrophic" signatures of environmental disasters and biomass crash, impoverished postextinction fauna and flora dominated by stress-tolerant and opportunistic species, and gradual ecological recovery and radiation of new taxa. Isotopic and other geochemical signatures are also generally consistent with the expected after-effects of catastrophic impacts. Seven of the recognized extinction pulses seem to be associated with concurrent (in some cases multiple) stratigraphic impact markers (e.g., layers with high iridium, shocked minerals, microtektites), and/or large, dated impact craters. Other less well-studied crisis intervals show elevated iridium, but well below that of the K/T spike, which might be explained by low-Ir impactors, ejecta blowoff, or sedimentary reworking and dilution of impact signatures. The best explanation for a possible periodic component of approximately 30 Myr in mass extinctions and clusters of impacts is the pulselike modulation of the comet flux associated with the solar system's periodic passage through the plane of the Milky Way Galaxy. The quantitative agreement between paleontologic and astronomical data suggests an important underlying unification of the processes involved.
