We have reexamined the population of craters on Venus believed to be embayed by lava flows and discuss the reliability of criteria for determining the volcanic embayment state of a crater, presenting nonvolcanic processes which may produce similar crater morphologies. We conclude that 29 craters have been embayed by lava flows from the exterior, and an additional 27 craters are ambiguous cases (3--6% of the total crater population). All of the embayed craters are embayed by the stratigraphically youngest lava flows except for five to 11 craters embayed by stratigraphically older regional plains deposits which are typically deformed by wrinkle ridges. Through geologic mapping of an area covering over 8% of the surface, we examined the spatial relationships between wrinkle-ridged plains and stratigraphically older plains units, and estimate that a large portion of the wrinkle-ridged plains are thinner than 500 m, which is thin enough for the rims of the largest half of preexisting craters to be preserved. There are not enough craters embayed by the wrinkle-ridged plains to support a model in which the plains were erupted directly onto an ancient surface. Given the number of large craters embayed by thin deposits of wrinkle-ridged plains, we develop a simple statistical model to overestimate the number of craters on the preplains surface. This model suggests that the average age of the surface beneath the plains plus the average length of time over which the plains were emplaced was of the order of tens of millions of years. The short exposure age for the preplains surface indicates that some event or events must have erased the ancient population of craters from the surface before the emplacement of the wrinkle-ridged plains. ¿ 1999 American Geophysical Union