New Light on the Dates of Primate Origins and Divergence

Living primates (excluding tree-shrews) form a monophyletic unit containing 5 subgroups: (1) lemurs, (2) lorises, (3) tarsiers, (4) New World monkeys, (5) Old World monkeys + apes + humans. Early divergence between these subgroups is suggested by chromosomal and molecular evidence. If, indeed, divergence occurred early, continental drift in the late Cretaceous / early Tertiary may have been a contributory factor. However, the earliest unequivocally recognized fossil primates are from the basal Eocene (about 55 Mya), and the standard view is that primates originated no earlier than 65 Mya. Similar conclusions have been reached for most orders of placental mammals, and it is widely accepted that their origin and diversification followed the extinction of dinosaurs at the end of the Cretaceous. A parallel explanation applies to the adaptive radiation of modern birds. (It is important here to distinguish between the time at which a group diverged and the time at which its diversification began, as defining characters could theoretically emerge some time after initial divergence.) The view that recognizable primates first appeared about 65 Mya reflects the common procedure of dating the origin of a group by the first known fossil representative and possibly adding a few million years. Direct dating from the fossil record is beset by 2 problems: (1) If there are many gaps in the record, the first known fossil representative is likely to be much younger than the actual first emergence. (2) Bias in the fossil record may entail additional error. These problems have direct implications for the common practice of calibrating molecular trees with a single date for the first known fossil of a group. This includes inferences regarding the timing of human evolution. Major gaps undoubtedly exist notonly for primates but also in many other parts of the mammalian fossil record. The most striking primate example is the record for Malagasy lemurs, currently limited to subfossils a few thousand years old although they must have existed at the very least for 20 My, as the sister-group (lorises) is documented by fossils of that age. Poor representation of early placental mammals in the fossil record is strikingly illustrated by 2 cases: (1) bats, (2) anteaters. Sampling inadequacies are further indicated bya  still-accelerating discovery rate for new fossil species. A simple approach used by Martin (1993) indicated that only 3% of extinct primate species have so far been documented. Rough correction for underestimation of the time of origin led to the proposal  that ancestral primates existed about 80 Mya. This has now been confirmed by a newly developed statistical approach. Although Gingerich & Uhen (1994) calculated that the probability of primates originating 80 Mya was less than 5 in a billion, their calculation is demonstrably flawed. Many recent analyses of molecular data using various calibration dates external to primates have confirmed an early date for the origin of primates. For instance, inference of divergence times for bird and mammal orders fromnuclear gene divergence, calibrated with the well-documented split between synapsid and diapsid reptiles, set the origin of primates at about 90 Mya (Hedges et al., 1996; Kumar et al., 1998). Demonstration of an African clade of placentals (Springer et al.,  1997) provided further support for early divergence between mammal orders. Calibration of complete mtDNA-sequence trees with dates for the earliest known (Palaeocene) cetaceans also set the origin of primates at about 90 Mya (Arnason et al., 1996, 1998). The problem of bias in the fossil record also needs attention. Modern primates are essentially confined to tropical/subtropical forests on the southern continents. Yet the earliest known primates occur in the northern continents and there is little overlap  in distribution. The most plausible explanation is that the likelihood of fossil preservation/discovery has been far greater in the north and that the record simply reflects a transitory northward expansion of tropical/subtropical primates when average temperatures were notably higher during the Eocene. The early history of primates in the southern hemisphere remains very poorly documented.