While it may seem unlikely, it is believed that the overall mutation rate within a species does not vary much over long periods of time. This means that the mutation rate serves as a "molecular clock." The clock can be used to determine the time since the evolutionary divergence of two species. Two organisms with very few DNA sequence differences between them diverged more recently than two that display more accumulated differences. The absolute amount of time for these divergences can be determined if the clock is calibrated, that is, if a known number of sequence differences can be correlated with a known time since divergence. This is done by comparing sequence data with data from the fossil record.
Early work in this field concentrated on a small handful of genes and gave conflicting results. Because mutation rates vary among genes, the best results come from analyzing changes in many genes. A 1998 study of the evolution of mammals analyzed 658 nuclear genes from 207 vertebrate species. It showed that the ancestors of most contemporary mammals arose more than 80 to 110 million years ago, long before the extinction of dinosaurs, and demonstrated that the fossil record from that time has some very large gaps in it.
Crow, J. F. "The High Spontaneous Mutation Rate: Is It a Health Risk?" Proceedings of the National Academy of Science 94 (1997): 8380-8386.
Griffiths, Anthony J. F., et al. An Introduction to Genetic Analysis. New York: W. H.Freeman, 2000.
Kumar, Sudhir, and S. Blair Hedges. "A Molecular Timescale for Vertebrate Evolution." Nature 392 (1998): 917-920.