When did humans leave Africa?

Fossil evidence indicates that modern humans evolved in Africa around 195,000 years ago (McDougall et al., 2005). Genetic data confirms this, showing that all living people share an African ancestor that lived  around this time as well (Soares et al., 2009). However, if we’re all African at heart this raises the question of when did some of us leave our homeland and migrate across the world? New research appears to have uncovered the most accurate date for the out of Africa migration yet, and it’s much older than we previously thought.

A reconstruction of the first modern human

Trying to figure out when some of us left Africa has always been tricky. You could look into the archaeological record and try and figure out when the earliest evidence of humans living outside of Africa is. Such investigations typically give dates of ~50,000 years ago (Turney et al., 2001). However, the archaeological record is incomplete and so there’s always the possibility that older examples of humans living outside of Africa will be found. As such scientists have started to focus on  DNA evidence. After all, everyone has a complete genome inside them! Early research into DNA evidence for the move out of Africa produced similar dates, suggesting that we left ~50,000 years ago (Quintana-Murci et al, 1999).

Such genetic dates are calculated by looking at how many mutations separate two groups and working out how many mutations occur in each generation.  This gives you a figure of how many generations have passed since the two lineages split.   For example, say that 100 mutations separate Africans and non-Africans and we know that 1 mutation occurs in each generation.  That means 100 generations have passed. Say the average generation length is 50 years, so we just do 100 x 50, giving us 5,000 years since we split.

However, many of the figures used in the early research were based on early work and so not completely accurate. Since then we’ve sequenced the genomes of several thousand people, providing a more reliable figure of the number of mutations that occur in each generation.  As this new data has been made available there’s been a slurry of researchers using it to produce more accurate dates for various splits that have occurred over the course of human evolution. For example, the date of the human/chimp split has been pushed back from 5 – 8 million years ago to 7 – 13; whilst the date for the human/Neanderthal divergence has been updated to between 400 – 800 thousand years ago (Langergraber et al., 2012).

The recalculated divergence of humans and chimps

The recalculated divergence of humans and chimps

This is also what the new research into the out of Africa migration has done. Alywyn Scalley and Richard Durbin from the University of Cambridge used these new, more accurate figures to recalculate the split between African and non-African populations. Their results show that non-Africans first left Africa between 100 and 120 thousand years ago; which is more than double the earlier estimates (Scalley and Durbin, 2012). This spread out of Africa would’ve taken a long time. People weren’t exploring to find new worlds but just heading to new pastures as areas became overpopulated.

This date is more consistent with other archaeological evidence, which shows that modern humans were living in the Middle East by 115,000 years ago (Schwarcz, 1988). Under the old dates these early sites were viewed as a failed migration, where people tried to leave Africa but were driven back (Quintana-Murci et al., 1999).  Under this new approach these early sites were the first port of call as our species spread throughout the world. Since Neanderthals were also living in the area, this may have been where we the famous human/Neanderthal interbreeding occurred (Scalley and Durbin, 2012).

Our redated journey of of Africa

Our redated journey of of Africa

They also redated the human/Neanderthal split and produced figures very similar to those other researchers have arrived at. The fact that multiple teams from across the world are getting the same results suggests that these dates are accurate. However, all is not peachy in the world of genetics. Another recent study by an international team of researchers has produced a contradictory date, suggesting that humans actually left Africa 90 – 60 thousand years ago (Fu et al., 2013).

This other date was produced by examining mitochondrial DNA, rather than the DNA from the nucleus the Cambridge pair studied.  Both approaches have their advantages and disadvantages so it’s difficult to say whose correct at this juncture. Hopefully it will become apparent with future research.

For the moment we can say that some humans left Africa some point between 60 and 120 thousand years ago. As they spread throughout the world the met other members of our family, such as the Neanderthals, and interbred with them. Now we live on every continent, and even in space. Yet we’re the last of our kind, as the other members of Homo we met as we travelled the globe have since gone extinct. Let’s hope our family doesn’t disappear entirely, eh?

Humans are everywhere!


Fu, Q., Mittnik, A., Johnson, P. L., Bos, K., Lari, M., Bollongino, R., … & Krause, J. (2013). A Revised Timescale for Human Evolution Based on Ancient Mitochondrial Genomes. Current Biology.

Langergraber KE, Prüfer K, Rowney C, Boesch C, Crockford C, Fawcett K, Inoue E, Inoue-Muruyama M, Mitani JC, Muller MN, Robbins MM, Schubert G, Stoinski TS, Viola B, Watts D, Wittig RM, Wrangham RW, Zuberbühler K, Pääbo S, & Vigilant L (2012). Generation times in wild chimpanzees and gorillas suggest earlier divergence times in great ape and human evolution.Proceedings of the National Academy of Sciences of the United States of America, 109 (39), 15716-21

McDougall, I., Brown, F. H., & Fleagle, J. G. (2005). Stratigraphic placement and age of modern humans from Kibish, Ethiopia. Nature433(7027), 733-736.

Quintana-Murci, L., Semino, O., Bandelt, H. J., Passarino, G., McElreavey, K., & Santachiara-Benerecetti, A. S. (1999). Genetic evidence of an early exit of Homo sapiens sapiens from Africa through eastern Africa. Nature genetics,23(4), 437-441.

Schwarcz, H. P., Grün, R., Vandermeersch, B., Bar-Yosef, O., Valladas, H., & Tchernov, E. (1988). ESR dates for the hominid burial site of Qafzeh in Israel.Journal of Human Evolution17(8), 733-737.

Soares, P., Ermini, L., Thomson, N., Mormina, M., Rito, T., Röhl, A., … & Richards, M. B. (2009). Correcting for purifying selection: an improved human mitochondrial molecular clock. The American Journal of Human Genetics,84(6), 740-759.

Turney, C. S., Bird, M. I., Fifield, L. K., Roberts, R. G., Smith, M., Dortch, C. E., … & Cresswell, R. G. (2001). Early human occupation at Devil’s Lair, southwestern Australia 50,000 years ago. Quaternary Research55(1), 3-13.

Help stave off the disappearance of humans by following me on Twitter or liking EvoAnth on Facebook.

28 thoughts on “When did humans leave Africa?

  1. Wouldn’t it be more likely that there was a more or less continual leakage of people from Africa? The normal vision of an individual migration or two seems a bit simplistic.

    • Non-African populations have a few unique haplotypes so they must have been isolated enough for this divergence to occur. However, even taking this into account there were likely multiple waves of migration.

      • And yet all Non-Africans have some Neanderthal DNA. Do you know if there’s evidence that different waves of migration have picked up different Neanderthal genes or maybe all the “migrants” hung around together in the Middle East for some time breeding with each-other and Neanderthals before spreading out?

        • Neanderthal interbreeding is always a tricky issue since we don’t have any human DNA from that period, it’s all based on modern genomes. We could sequence a 100,000 year old person and find they were 25% Neanderthal!

          Based on what we do know it seems that the interbreeding occurred on a relatively small scale in the original out of Africa population from which all others are descended. In other words, a group left Africa, interbred with Neanderthals and then spread around the world.

  2. Pingback: From prehistoric hominids to modern humans, video | Dear Kitty. Some blog

  3. Pingback: Our Past (and Future) With Supervolcanoes | The Advanced Apes

  4. Pingback: Our Past (and Future) With Supervolcanoes

  5. Pingback: The Sixth Extinction | virtuallyfarmer

  6. Pingback: Space Junk – Listening Fluency #95 – Deep English

  7. It’s hard to believe humans left Africa in one big wave, at a certain time. Wouldn’t they have migrated out slowly, over many generations, and back and forth, and left and right, and then intermixing with one population 10,000 years later, or 60,000 years later over there, and then a little over here, and so forth, in a great big giant mess like we have recently in the last 4,000 years or so?

    You always see pictures of arrows showing exodus waves as though there was a big stink in Africa and everyone felt the sudden urge to flee. Makes no sense. That’s not how people have sex or follow the hunt.

    • That’s probably true of this (in fact there’s some evidence modern humans went in and out of Africa a couple of times); and likely true of a lot of other things to. For example, you always here people talking about humans and chimps splitting 7 million years ago; as though the two species just went “that’s it, no more mating after today.” Really it was probably more gradual, as was the migration.

  8. I like the way you have drawn the phylogeny with divergence times (gradually fading blue hybridisation). Did you come up with this yourself?

  9. Is it still tenable that there never happened something that one would call a “textbook” interbreeding between sapiens and neanderthals, or denisovans, at the “peak” of divergence? I wonder if it’s possible something more like a “ring species”, with crossings only happening between more similar “neighbors”, but, given the Saharan barrier, the original “out of Africa” sapiens stock(s) happened to have “neanderthal” and “denisovan” genes that remained from “pre-sapiens”/”archaic sapiens” populations. These genes could have been originally North African, Middle Eastern, European, or Asian, but not necessarily originally “classic neanderthal”/denisovan, but from “heildelbergensis”, “rhodesiensis” or “helmei”. More or less like a more geographical than chronological take on the “phases” of multiregional evolution. And posterior “assimilation” not of exactly archaic hominids or sapiens, but of those sapiens who were previously deemed “failed” migrations.

    • I’m only passingly familiar with ring species; so can’t really comment on how well the evidence matches that. However wouldn’t one expect there to be greater diversity in the proportion of genetic relatedness amongst different populations, depending on their location. Instead we have all non-Africans having a relatively similar rate of Neanderthal relatedness, all sub-Saharan Africans having none, all sub-Siberian people having an relatively similar rate of Denisovan relatedness and so forth

      • I don’t know. I think that the allelic frequencies could be the same, regardless of the exact taxonomic status of the source lineage. Here’s a totally bogus made-up graph:

        The divergence of lineages is somewhat analogous to that blurry branching from the first graph, only instead of blurring, there are made up branchings of varying density.

        The “neanderthal” and “denisovan” alleles could have arisen almost anytime since the first divergences, but could have re-entered the sapiens line “at the edges” not necessarily from direct through the outer ends, but from intermediate lines. I guess.

        This graph is plagiarized and adulterated from this one:

        Borrowing concepts from these ones:

        A more “accurate” one would even be “3d” like this one, not related to humans:

        And ideally have representations of the genetic lineages inside population lineages:

        • It sounds like that If that there were the case there would have to be that intermediate line you refer to, shuffling genes back and forth between the humans and other groups. However, whilst our knowledge of ancient genetics is far from complete I’ve yet to encounter any evidence of such a group.

          • These “intermediates” could have been ancestral heidelbergensis/rhodesiensis and early sapiens, I guess. But not literally bridges between contemporary highly divergent lineages, but just as a larger/more structured population from which sapiens evolved “at the center”, and from which geographical “archaic” local variation could persist at the peripheries. More or less like a more geographically restricted and timid version of multiregional evolution, which barely can be labeled as such, with more emphasis on the continuity of some local variation than on the notions of genes flowing more significantly from all directions and natural selection filtering “sapiens” from these worldwide genes. I guess it has some resemblance with things that Gunter Bräuer and others have said much earlier (an “Afro-European sapiens” and things like that), I’ll check it out and spare everyone further vague speculations from me.

  10. I know the argument of why havent chimps evolved, or any other animal for millions of years, but humans did is an old one. However. It blows my mind that, in a mere 50,000 years or so, even 100,000 years, humans evolved into all races we see today. I have never heard a reputable example of how such drastic mutations (I believe we are different enough from another race) occurred over only a few thousand generations. If mutations were SO quick to take place, and so easily over such a relatively small number of life forms, then it would basically be guaranteed that any other species of animal would go through gradual evolutions no matter WHAT. It’s not like we choose to evolve or not, whether it’s a change of setting or not. Think about it.. 50 thousand years to see significant mutations in humans, while virtually 0 mutations in virtually all other organisms for millions of years? It’s for this reason (and I’m not trying to go there, or change the topic, as my main point is the haste of our evolution compared to anything else) that I have to assume there was some sort of extraterrestrial or intelligent design modifications to our DNA, as quite a few of our “evolutions” are really a downgrade compared to some traits our primate ancestors possess. The evolution of our brain, tripling in size over just a few million years, is quite a miracle even compared to all previous examples of evolution.

      • If the genetic differences are marginal to non-existant, then why is it that no other species seem to go through marginal evolutionary changes in millions of years? This answer is devoid of any reasonable explanation to everything I stated.

        • Yes it was. You’re entire comment revolved around your assertion that you “believe we’re different enough from another race.” My point was that you were significantly over-estimating the differences between races; and once you take that into account they mystery of rapid human change disappears because there hasn’t been rapid change. There’s been very little change.

          And even if you want to take issue with any of that we’re still left with the fact that your comment is based entirely around your belief. Your assertion. You’d have to provide evidence of that before progressing further. Can you point me in the direction of any studies showing, for example, that the genetic differences between chimp groups is smaller than between comparable human groups?

You evolved too. Have a say.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s