A global charcoal and soot layer, coinciding with the Ir layer, is present at the Cretaceous/Tertiary (K/T) boundary, and apparently comes from a global fire. Soot is present even in the basal layer of the boundary clay, implying that the fire started soon after the impact. No comparable soot enrichments, let alone of global extent, occur in the latest Maastrichtian or in a wide range of other marine sediments.

Much or all of the fuel was biomass, as indicated by the presence of retene (a hydrocarbon diagnostic of resinous wood fires) and by the carbon isotopic composition (δ¹³C = −25.8 ‰ ± 0.6 ‰), which resembles that of natural charcoal and atmospheric carbon particles from biomass fires. The mean amount of elemental C at 11 K/T boundary sites (11 ± 3 mg/cm²) is very large, and requires that much of the Cretaceous biomass burned down and yielded a larger mass fraction of soot and charcoal than small fires.

At one undisturbed site (Woodside Creek, New Zealand), soot in the boundary clay correlates tightly with Ir, As, Sb, and Zn. A possible reason for this correlation is that soot and Ir-bearing ejecta particles—containing some volatile chalcophiles from the target rock—coagulated in the stratosphere, and then scavenged additional chalcophiles from the hydrosphere. In view of this coagulation, the K/T fire would only slightly prolong the period of darkness and cold caused by impact ejecta. However, it would contribute other environmental stresses, e.g., a CO₂ greenhouse and a variety of pyrotoxins and mutagens. Because the total of recognized stresses has risen to 12, there is no basis for the contention that an impact cannot explain the observed selectivity of extinctions.