"Every paleontologist knows that most new species, genera, and families, and that nearly all categories above the level of family appear in the record suddenly and are not led up to by known, gradual, completely continuous transitional sequences."