Unbiased readings of fossils are well known to contradict some of the popular molecular groupings among primates, particularly with regard to great apes and tarsiers. The molecular methodologies today are however flawed as they are based on a mistaken theoretical interpretation of the genetic equidistance phenomenon that originally started the field. An improved molecular method the 'slow clock' was here developed based on the Maximum Genetic Diversity hypothesis, a more complete account of the unified changes in genotypes and phenotypes. The method makes use of only slow evolving sequences and requires no uncertain assumptions or mathematical corrections and hence is able to give definitive results. The findings indicate that humans are genetically more distant to orangutans than African apes are and separated from the pongid clade ∼17.6 million years ago. Also, tarsiers are genetically closer to lorises than simian primates are. Finally, the fossil times for the radiation of mammals at the K/T boundary and for the Eutheria-Metatheria split in the Early Cretaceous were independently confirmed from molecular dating calibrated using the fossil split times of gorilla-orangutan, mouse-rat, and opossum-kangaroo. Therefore, the re-established primate phylogeny indicates a remarkable unity between molecules and fossils.