Suppose that humanity has spread across the universe, colonizing various habitable worlds, until their empire spans the entire galaxy (or near enough) - but there is no sign of alien life. There are worlds with water and an earth-like atmosphere, but no flora or fauna besides those the humans brought with them.

Then fast forward a few thousand years or so. Assuming inter-planetary migration is not too common, so that planets (or more likely systems) exist in relative isolation, how varied could these humans become? Living in survivable, but varied environments, is there a limit on how differently two groups of humans could evolve? Could they reach the point where they are no longer the same species?

  • 3
    $\begingroup$ Few thousands years is not that much. See, we don't look that different from ancient Egyptians, right? $\endgroup$
    – Mołot
    Jan 27, 2017 at 17:29
  • $\begingroup$ Would a timescale more along the lines of a few thousand generations be more useful? Or would we need more like millions of years to have significant change? $\endgroup$
    – Uzai
    Jan 27, 2017 at 17:33
  • 1
    $\begingroup$ Probably at least a quarter million to half a million years, and even that the differences wouldn't necessarily be that huge. $\endgroup$
    – ohwilleke
    Jan 28, 2017 at 2:55

2 Answers 2


Depending on the amount of Founder's Effect you employ, even a few thousand years could produce a pretty wide divergence in human appearance, but it is unlikely that they will become a new species (defined as not being able to reproduce with each other). Earth populations have been isolated from each other for far longer periods of time but are still one species, even though their external appearances are very different and there can be substantial differences in many genes.

One caveat would be a significant teratogenic effect driving an increased mutation rate; like increased radiation exposure, environmental contamination, or deliberate genetic alterations. This last one is what would be most likely to produce separate human species within a short time frame, depending on how liberal you are with it, virtually any level of genetic change is possible, leading to completely incompatible species, even on the same planet.

Millions of years is certainly enough time, we probably diverged from chimps 4-5 million years ago, but from Neanderthal just 600-800 thousand years ago. Here is a very technical paper about it.

Of course this is highly dependent on population growth (modern tech allows for a much greater population growth), frequency of large scale extinctions (the human population has probably dropped down to just a few thousand individuals several times in our past), and the amount of genetic variety in the initial colony (the Founder's Effect I mentioned earlier).


The only limit on evolution is time.

The idea that there's some hard limit to how different a species can become before it's non-viable is very popular in creationist circles, but it's totally unsupported biologically. Biologically and genetically, there is no hard limit to how much a given population can change.


A few thousand years really isn't long enough for major changes to take place. Europeans and east Asians diverged about 40,000 years ago - now, admittedly, there was some limited contact between the two groups, but still the populations remained 100% chemically interfertile with minimal phenotypic distinction.

Now, theoretically, it could be possible for a directed breeding scheme to produce some pretty massive changes much faster than pure natural selection. Speciation has been observed in fruit flies in around 35 generations - which, if every person bred at around the age of 20 would take 700 years in a human population.

So if you had a colony full of eugenicists actively breeding with the goal of speciation, you might be able to get a major change in a few thousand years. With pure natural selection on its own, though? You'd need to keep populations isolated for at least hundreds of thousands of years.

  • $\begingroup$ Agreed. The divergence time between the ancestors of S. African Bushmen and the most diverged humans outside of Africa is about 70,000 years. The divergence time by humans and Neanderthals, and humans and Denisovans which may have involved some hybrid gender preference for females was more than 400,000 years. On the other hand, the modern European phenotype is only about 5,000 years old and arose due to Indo-European and early farmer admixture, and the modern mestizo phenotype also due to admixture is only about 500 years old. Selective breeding, as noted, could produce more variation faster. $\endgroup$
    – ohwilleke
    Jan 28, 2017 at 1:52

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .