Although the distribution of impact craters on Venus is indistinguishable from a completely spatially random population on the basis of crater data alone, the addition of geologic information indicates that the craters are not random with respect to geology. Areas of low crater density correlate with concentrations of tectonic and volcanic features, as well as high proportions of faulted and embayed craters. High-resolution mapping of large volcanoes, flood-type lava flow fields, rifts, and coronae shows that these features have low crater densities that are unlikely to have occurred merely by chance. Because these features also appear young based on stratigraphic data, difference in age is the most likely reason for differences in crater density between geologically defined terrains. Crater densities may be used to determine reliable relative ages for sufficiently large geologic terrains containing at least eight to nine craters, as long as the mapping criteria are independent of impact crater locations. This result provides a tool to begin identifying and dating major geologic provinces, and to begin developing a geologic history of Venus. Using a simple data model, and calculating crater ages relative to an assumed global mean age of 300 Ma, we show that the ages of large volcanoes (72¿45 Ma), flow fields (128¿91 Ma), rifts (130¿145 Ma), and coronae (120¿115 Ma) are all substantially younger than the mean plains age, and probably represent ongoing volcanic and tectonic activity, rather than the end of a brief global resurfacing event with a mean age of 300 Ma. ¿ American Geophysical Union 1996