5 Comments

I got a little bit starry eyed writing about the Neanderthal genome the other day. I chose to retrace the arc of scientific progress that links the initial description of Neanderthal man as something different than modern humans to the point reached last month, where we are able to tag some of those differences to a single gene. Most of the news stories about the Neanderthal genome focused not on the genes that made us different from them, but a small percentage of the genome that reinforced the continuity been them and us. Genetic evidence that Neanderthals interbred with the ancestors of some modern humans. The revelation of these ancient assignations has caused some quite sensible people to say some quite silly things about what species are and what Neanderthals were. So, perhaps I can compliment my slightly hazy earlier piece with a more hardheaded take on why Neanderthals remain a species unto themselves.

Let’s start with the evidence that Neanderthals interbred with the ancestors of modern humans. Modern humans (Homo sapiens) arose in Africa about two hundred thousand years ago, all modern human populations outside of Africa descend from a relatively small number of migrants who left that continent between eighty and fifty thousand years ago. When those migrants first left Africa and entered the Middle East they would have met other humans. The ancestors of the Neanderthal had moved out of Africa and established themselves in Europe and Central Asia thousands of years before. Until now we haven’t known which of the four ‘F’s (fighting, fleeing, feeding or reproduction) followed that first contact, the Neanderthal genome has given us a clue.

When you compare individual DNA bases that are variable within modern human genomes to the corresponding sequences in the Neanderthal genome you find that non-African sequences match the Neanderthal sequence slightly (but significantly) more often than African sequences do. It’s possible that this pattern is an artifact of our poor sampling of African genomic diversity (that observant nerd Christie does a good job of explaining how here) but for the sake of argument let’s take it for granted that his pattern is the result of ancient interbreeding. The authors of the paper describing the Neanderthal genome estimate people with no recent African ancestry inherited between one and four percent of their genome from Neanderthals. That number is the same for Papuan and East Asian populations as it is for Europeans despite Neanderthals having lived alongside Europeans for thousands of years, suggesting any interbreeding that contributed to modern human genomes was limited to that first period of contact.

This is where the problems start. Having heard the news that Neanderthals and some of our ancestors might have once swapped genes some people remember that nice easy test of species-status from high-school biology. Something like “if two animals can interbreed then they’re part the same species.” So, are we Neanderthals; or are Neanderthals us? No. In fact, the Neanderthal genome serves to highlight some the mistakes we commonly make when start trying to define species.

Biologists have spent a lot of time arguing about just what a species is and how can delimit species from the creatures that we study, too often we’ve forgotten that those are two different arguments. DeLene from Wild Muse has a thoughtful overview of some of the factors that contribute to the “species problem” in her review of Jody Hey’s book on the same topic. You should read her piece because the species problem really is a fascinating philosophical question, but I think most of the fights that erupt around competing definitions of species come from a failure to understand that defining species and organising critters into species are two different tasks. We’ve been studying speciation, the process by which new species arise, for a while now and we’ve developed a pretty good idea of how it works. Two populations stop interbreeding with each other, during that period of “reproductive isolation” genetic changes in one population can’t effect the other so natural selection and random changes (called genetic drift) change each population independently. Species are populations which are on independent evolutionary trajectories.

Reproductive isolation drives the independence that is at the heart of what species are, but it’s not the sine qua non of a species. James Mallet from University College London has made a special study of hybridisation, and he reckons 10% of animal species and a whopping 25% of plants interbreed with other species from time to time. As molecular tools have been applied to non-model organisms it’s become increasingly clear that the “species barrier” is more porous than we’d thought, and species can maintain their independence even in the face of the occasional injection of genes from other species.(If you’re interested in the wider question, I’ve written a bit on the species problem here. The short version is we should see competing “species concepts” as operational tools that might be used to help delimit species, but not as definitions).

Now, think about the results from Neanderthal genome. Most sequences in that genome are separated from their human counterpart by a split that happened over five hundred thousand years ago. There is pretty good evidence that Neanderthals and the ancestors of non-Africans interbred when they met each other in the Middle East about four hundred and fifty thousand years after that initial split. That gene flow had the potential to homogenise the two populations into one, but it didn’t. Each lineage maintained its identity. For the twenty or so thousand years that Neanderthals continued to exist they retained identifiable morphological traits. There are fossils in Europe that some argue show a mixture of characters, but any interbreeding in that continent left no mark on modern European genomes, which have no more Neanderthal DNA than Papuan and Chinese genomes do. At the same time, the authors didn’t detect any flow of modern human genes into Neanderthal genomes (so it’s not a case of of modern humans swamping Neanderthal populations and erasing any trace of genetic admixture in the process). The available evidence seems to point o Neanderthals and modern humans as separately evolving populations, and a little bit of gene flow between them wasn’t enough to upset that pattern.

I should stress, by saying H. neanderthalensis and H. sapiens are different species we aren’t saying very much about how different Neanderthals were from us. Species are not defined by a degree of difference, or an essence that was missing in Neanderthals but is present in us, they’re just another human population that was moving in a different direction (and eventually extinction). If some of us do have Neanderthal genes, then it only goes to show how fuzzy the line between our species and the rest of the biological world is.


Green RE, and many, many others (2010). A draft sequence of the Neandertal genome. Science (New York, N.Y.), 328 (5979), 710-22 PMID: 20448178

James Mallet’s bit on the frequency of hybridisation is taken form here:

Mallet, J. (2005). Hybridization as an invasion of the genome Trends in Ecology & Evolution, 20 (5), 229-237 DOI: 10.1016/j.tree.2005.02.010

The ideas about species and species delimitation presented above are pretty similar to Kevin de Quieroz’s take:

De Queiroz, K. (2007). Species Concepts and Species Delimitation Systematic Biology, 56 (6), 879-886 DOI: 10.1080/10635150701701083