Were there Cretaceous anthropoids? Part 1. The problem in a nutshell

5 minute read

The evolution of early primates is a field that has developed rapidly in the last fifteen years. Many of the central issues were reviewed earlier this year by Blythe Williams, Richard Kay and E. Christopher Kirk (“New perspectives on anthropoid origins”). I want to touch on some issues, in a series of posts that may seem like a bit of a grab-bag.

I got started with a fairly simple question – the one from the title – were there anthropoid primates in the Cretaceous? Since this has expanded into a series, I think I’d better lead with the answer: We really don’t know.

From the fossil perspective, fifteen years ago you’d have thought I was crazy to even ask the question. It was common knowledge that the living orders of mammals diversified after the extinction of the dinosaurs 65 million years ago. Even today, there are no widely-accepted anthropoid fossils earlier than the Middle Eocene. A few specimens argued to represent anthropoids are earlier, one as early as the Late Paleocene. But that’s it. No positive evidence of earlier primate diversification, nothing that even looks like a primate before the Paleocene. Many paleontologists concerned with primate origins have assumed that the common ancestors of today’s primates lived in the Late Paleocene, and that the primates had diverged from their closest relatives – tree shrews, colugos, or bats – sometime after the end of the Cretaceous.

Opposed to this traditional view, molecular comparisons of living primates and other mammals have suggested an earlier diversification of primates, as early as 90 million years ago. For example – and I’ll review many others during the course of the series – Steiper and Young (2006) estimated the ages of primate divergences from long sequences homologous to an area around the CFTR gene on human chromosome 7. Like most studies, they assumed some “calibration points” with dates based on fossil evidence. One of these was the human-chimpanzee divergence (7 million years, based on Sahelanthropus), the other was the macaque-baboon divergence (8 million years). Steiper and Young did not have a tarsier sequence, so they reported the estimated strepsirrhine/anthropoid divergence – necessarily older than the first anthropoids, since tarsiers are the sister group of the anthropoid clade. Their estimate: between 88.2 and 110.2 million years ago. In addition to these “old” estimates, they reported a range of “young” estimates based on lower calibration times, only 60.8 to 75 million years ago. That’s a mere 20 to 30 million years older than the oldest uncontroversial anthropoids.

Primates are only one skirmish point of a much larger battle about the timing of diversification of mammal orders. DNA comparisons have consistently sketched out a long pre-Tertiary history for placental mammals. Many paleontologists favor a hybrid view, in which some superordinal groups of mammals – like Afrotheria or Archonta – existed in the Late Cretaceous, while the modern orders themselves got started after the extinction of the dinosaurs. That’s a softer view of mammal diversification than the traditional idea of a rapid origination of all these groups after the K-T boundary. Even the hybrid hypothesis hides an apparent problem: It means that the K-T impactor must have spared dozens of distinct lineages of mammals, even as it wiped out every kind of dinosaur, large marine reptile, and 75% of the rest of species.

For a brief review of this issue as applied to mammals generally, I can suggest a 2009 article by Jennifer Evans, “The disputed rise of mammals.” She reports on the largest molecular phylogeny yet constructed for mammals, called the “supertree,” and its apparent conflict with the fossil record:

Although the supertree dated the origin of mammals at 93 million years ago and showed 43 placental lineages surviving the K/T boundary, Wible's analysis of more than 400 morphological characters in Cretaceous fossils across 69 taxa placed the oldest placentals at 63 million years ago. "There was no evidence in the fossil record that any of Cretaceous forms previously identified [by molecular biologists] as placentals were in fact placentals," says Wible. Although fossils were used to date divergence points on the supertree, they could only date back to around 55-65 million years ago, where paleontologists have fossil evidence of modern mammals. "Until there's [fossil evidence of] a Cretaceous primate that everyone agrees upon there will be conflict between molecular and paleontological evidence," says Ross MacPhee, from the American Museum of Natural History.

Primates take center stage as one of the best-documented early mammalian lineages. Some expect to find anthropoids as early as 90 million years ago, yet the oldest well-established anthropoids are only around 45 million years old.

You might think that the solution is simple. For example, we might posit a different mutation rate on early branches of the primate phylogeny. Or we might simply give up on proposed Miocene hominins like Ardipithecus. Move those calibration points, and you totally eliminate the conflict between molecular and fossil information. But there’s uncertainty on the fossil side as well. We expect the known fossils to underestimate the ages of branches on the primate phylogeny – missing information can do nothing else. Some scientists have suggested that the known primate record is fully consistent with a Cretaceous origin and diversification, given what we know about which lineages of living primates are missing from the fossil record.

So understanding this problem will require some examination of both the fossil record and molecular evidence. Today this molecular side can be expanded to whole-genome comparisons of various kinds, and the data have expanded faster than anybody’s analysis of them. That makes it a great topic – there’s work to do here, for those who understand the connections between the fossil and genetic records.

Next: “What is an anthropoid?”


Evans J. 2009. The disputed rise of mammals. The Scientist 23:47. Online.

Steiper ME, Young NM. 2006. Primate molecular divergence dates. Mol Phylogenet Evol 41:384-394. doi:10.1016/j.ympev.2006.05.021

Williams BA, Kay RF, Kirk EC. 2010. New perspectives on anthropoid origins. Proc Nat Acad Sci USA 107:4797-4804. doi:10.1073/pnas.0908320107