Creationists have often argued, rather qualitatively, that if evolution had really happened, the many gaps in the "fossil record" would be most improbable. That is true. The purpose of this paper is twofold: to obtain somewhat more quantitative estimates of the improbability, and to calculate from the presence of gaps in the fossil record being given empirically, the probability that evolution occurred. It is shown that that probability is vanishingly small; one may as well say that, the gaps being admitted, it is impossible that evolution occurred.