04 January 2011

Hybridizing Stem-Humans: Or, Is Everyone Right?

Analysis of mitochondrial DNA shows that the human matrilineage split from the Denisovan matrilineage around a million years ago. The nuclear genome shows a more recent split of humans (Homo sapiens) from both Denisovans (Homo sp. indet.) and Neandertals (Homo neanderthalensis) around 270–440 millennia ago. And yet some modern humans (Melanesians) appear to have inherited a small portion (4–6%) of nuclear DNA from Denisovans. This means that hominin populations can recombine even after being split for hundreds of thousands of years.

A "wholphin" (Tursiops truncatus × Pseudorca crassidens).
Photo by Mark Interrante.
Properly considered, this is not shocking at all. In other placental species, populations that have been split for far longer periods of time can hybridize. Look at "wholphins", hybrids between bottlenose dolphins (Tursiops truncatus) and false killer whales (Pseudorca crassidens). Those parent species have been split for around seven million years, longer even than the split between humans and chimpanzees!

Of course, humans can't (or at least don't) interbreed with chimpanzees, so the length of the split is not a perfect predictor of whether lineages can recombine. But it's interesting to consider how much lineage recombination might have occurred in stem-humans. The most divergent known lineages from our own are probably late Paranthropus (P. robustus and P. boisei). Our common ancestor with them is generally thought to be something like Australopithecus africanus, or perhaps Praeanthropus afarensis. Even opting for the older choice, this would make the length of their split from our putative contemporaneous ancestor, Homo habilis, only around a million years (roughly). That's not a terribly long split.

Did Homo habilis have multiple ancestors?
(Photo by Charles Roffey)
What this says to me is that there is no a priori reason to suppose that any two contemporary populations of hominin could not have interbred. Maybe Paranthropus aethiopicus interbred with early Homocould this explain Australopithecus garhi? Maybe the Denisovans themselves are Homo erectus × neanderthalensis. Maybe Homo floresiensis are pinheaded, pygmy descendants of Homo erectus and an unknown, pre-Homo lineage! (I'm not saying I necessarily support any of these ideas; I'm just throwing them out there.)

Also consider the debates over human ancestry in the field of paleoanthropology. It's a common observation that whenever someone finds a new stem-human (or stem-mangani) species, they declare it a human ancestor, while their rival colleagues pooh-pooh the finding and maintain that their own specimens are the true ancestors. (There are notable exceptions to this, of course, but it does seem to happen again and again.) But what if everyone is right? What if most of these fossil species are ancestral to us, but in varying proportions? I can't see any reason why this would be unimaginable.

A human with partial
Neandertal ancestry
(blog's author).
Again, this is not an uncommon phenomenon in other placental species. Consider coyotes (Canis latrans)the eastern populations have partial ancestry from wolves (Canis lupus). Eastern lowland gorillas (Gorilla beringei graueri) may have partial ancestry from western gorillas (Gorilla gorilla). Our species is not unique in being partially hybridized.

So when people argue whether we are descended from Praeanthropus afarensis vs. Australopithecus africanus vs. Orrorin tugenensis vs. Kenyanthropus platyops—maybe everyone is right! At the very least, it seems to me that future discovery depends on allowing for significant amounts of admixture, and not blindly assuming simple bifurcation.


  • Ackermann & Bishop (2009). Morphological and molecular evidence reveals recent hybridization between gorilla taxa. Evolution 64(1):271–290. doi:10.1111/j.1558-5646.2009.00858.x
  • Green & al. (2010). A draft sequence of the Neandertal genome. Science 328:710722. doi:10.1126/science.1188021
  • Kays & al. (2009). Rapid adaptive evolution of northeastern coyotes via hybridization with wolves. Biol. Lett. 6:89–93. doi:10.1098/rsbl.2009.0575
  • Kim & al. (2009). Evolutionary charactterization of a highly repetitive sequence identified from the false killer whale (Pseudorca crassidens). Genes Genet. Sys. 84:185–189. doi:10.1266/ggs.84.18
  • Reich & al. (2010). Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature  468:1053–1060 doi:10.1038/nature09710
  • Xiong & al. (2009). Seven new dolphin mitochondrial genomes and a time-calibrated phylogeny of whales. BMC Evol. Biol. 9. doi:10.1186/1471-2148-9-20