I was thinking of a sci-fi story where a large population of human beings (let's say anywhere between hundreds and millions) live on a planet with robot guardians. But after a certain amount of time living in isolation from the rest of humanity, their genetic code changes enough that their DNA is no longer marked as homo sapiens by the computers. The machines turn on them, having been programmed to protect their human wards from potential "alien species."

How many years would the human colony have to be on the planet for that twist to be plausible?

Side question: If I made the planet more radioactive, could I plausibly speed up the evolutionary process in a way that wouldn't horribly scar the actual people? If not, is there another way to speed up the process while keeping the community as close to human as possible?

  • $\begingroup$ This really depends on the criteria for "different species". In biology, two groups of organisms are usually determined to be different species if members of group A can no longer produce viable offspring with members of group B, with "viable" meaning the offspring can grow to adulthood and produce offspring themselves. Note that this problem needs to apply to the whole group: individuals can have fertitlity problems. I can't see how this criteria could apply here since there's no old humans for the neo-humans to mate with. So, how do the robots decide whether an individual is human? $\endgroup$ Commented Sep 28, 2015 at 11:05
  • 1
    $\begingroup$ I would say this is a duplicate of How long would a modern human population..., but you've also asked about accelerating the process. $\endgroup$
    – Frostfyre
    Commented Sep 28, 2015 at 12:10
  • $\begingroup$ I'm going to say what I've said in other similar questions, that the best way to make two groups of humans sexually incompatible is to turn one of them into robots. Or cyborgs, whatever you prefer. If they upload their minds into robot bodies, they may still be considered human, but can no longer reproduce with natural humans. $\endgroup$ Commented Sep 28, 2015 at 13:25
  • $\begingroup$ A problem here is that the changes you describe happen on an individual level, not on a population level. If the robot guardians aren't going to wait for genetic divergence to happen in the whole population, they'll pick out the small handful of people every generation who have insufficiently human DNA. This machine would be selectively breeding its human population, essentially. $\endgroup$ Commented Aug 20, 2016 at 12:55

5 Answers 5


This depends on the population of the planet. There are several different views about how origin of new species occurs.

Origin Of New Species: Method 1

Some scientists believe that:

  • The process of evolution begins with a need for adaptation. There must either be some big incentive (a new food source which is not utilized by any other species etc) or a big threat (adapt to the new conditions ... or go extinct).

  • The species goes through a general evolutionary phase. As in, several individuals go through different types of genetic changes, which continue to be mixed and matched as the individuals breed, forming more genetic variation in the next generation.

  • Soon (in a few hundred thousand years), the environmental conditions decide which type of individuals are best suited in the changing world around them. These individuals proliferate while the others wean out and go extinct. In all of this process, a group of creatures evolve into another species.

Under this scheme of evolution, your humans would take at least several hundred thousand years to have a 0.1% DNA change that could mark them as a different species (depending on how the robots determine it). You would also need to provide the reason for change. Is the atmosphere slightly different than earth's and the respiratory system has to change a bit to adapt? Is the gravity of the planet slightly more or less than earth's and the bones of the people undergo a slight change? Are there new food sources (plants/animals) on the planet that require modified teeth/stomachs/intestines/kidneys?

Origin Of New Species: Method 2

As much as I like the above mentioned theory of origin of new species, most scientists believe that origin of new species follows these steps instead:

  • One individual in the population undergoes a random mutation. This mutation helps its offspring to be slightly superior to its competitors. It runs faster/its hearing is keener/its jaws are stronger etc.

  • Due to more success than the other individuals of its kind, this individual has more chances of procreating, so with time the individuals of this type increase.

  • With time, more random mutations occur and the same process goes on rinse-and-repeat lines.

In this scenario, you would have more trouble explaining why all of the population underwent the evolutionary change, as reproduction in humans is no longer a game of physical characteristics anymore.

My Recommendation

If I were the writer of this fiction, my story would follow these lines:

  • The atmosphere of this planet has slightly lesser oxygen (~19.5% as compared to 21% on earth).

  • People have slight difficulty in breathing, but it is mostly subconscious, they do not consciously notice that they are breathing slightly faster than humans on earth.

  • Scientists invent a respiratory aid for people. People undergo a minor surgery which dilates their trachea and helps them inhale more air with each breath.

  • A group of genetic scientists figure out a means to improve the tracheal and lung structure of the next generation so that they would not require any surgery at all. This would involve changing a set of 8 genes in the human DNA.

  • The genetic change is applied at some test mice (imported from earth) and the results are satisfactory. A 3-day schedule is announced for this genetic change to be applied to all the people on the planet.

  • The gene-changing machinery is very large and runs on a lot of power. Since the population of the planet is high (several billions) and the robot guards take a lot of energy to keep running, the defense mechanism is shut down for these 3 days so that there is no electric overload.

  • All the people in the population (you might save some, the lazy ones like me who missed the schedule due to one reason or the other) go and get this change of genes.

  • After the process is finished, the robotic defense mechanism is turned on, and voila ...!

  • $\begingroup$ Small populations can more rapidly speciate than that; ie: Darwin's finches. But they're also, often, following other rules (get big) which helped. $\endgroup$
    – user3082
    Commented Sep 28, 2015 at 11:41
  • $\begingroup$ The reason larger populations speciate slower is not that the rate of change is slower. Rather the gene pool is larger and there is more genetic diversity within a larger population. Which means that a larger population changes at the same rate as a small one, but the changes are very very less spread out evenly in the population. $\endgroup$ Commented Sep 28, 2015 at 11:50
  • $\begingroup$ Not according to what I've read, but it was re: bacteria populations :) Lots of land means no competition. And you only get diversity if travel is slowed/stopped otherwise the frequency of the alleles is about the same across the board, versus drifting because the frequencies change within the population pool. Larger populations have less drift. $\endgroup$
    – user3082
    Commented Sep 28, 2015 at 13:03
  • 1
    $\begingroup$ The adaptation of the Sherpa population to high altitudes (similar to your low oxygen planet) appears to have developed within the last 3000 years. $\endgroup$ Commented Sep 28, 2015 at 13:47
  • $\begingroup$ And then they say evolution occurs randomly without any direction ... $\endgroup$ Commented Sep 28, 2015 at 14:03

Much faster than you might think. One generation!

For this story, you'll have to dump those silly robots. Instead, imagine this. Some years in the future, doctors become aware that a rapidly increasing number of young couples are attending fertility clinics. Plenty of young couples are still giving birth to healthy babies, but far more than in previous generations, can't.

The penny drops. It is discovered that these couples are infertile because one of them has an odd number of chromosomes. He, or she, is sterile for the same reason that a mule is sterile. And of course, this means that part of humanity has evolved, un-noticed, very suddenly, into a new species with two more, or two less, chromosomes. You can't tell them apart, other than using a microscope to count chromosomes. Or more likely, you can hardly tell them apart. Once you know what you are looking for, you can tell, though with rather less than 100% reliability absent scientific confirmation.

Over to you, the author. I've no doubt that the new homo species will have inherited the same amount of bigotry and boneheaded stupidity as the original. It might be interesting to make the narrator or hero one of those sterile hybrid children.

OK, to the hard science. Consider the horse and the donkey. Two separate species sharing a common ancestor. Yet a donkey has 62 chromosomes, and a horse, 64. How can they share a common ancestor? How can any creature have a different number of chromosome pairs to its parent, and yet manage to breed non-sterile offspring?

If a single proto-donkey had once given birth to a single mutated foal with an extra chromosome pair, that would be a genetic dead end. The only way that foal could reproduce is if, in the same generation, a significant fraction of other proto-donkeys were also giving birth to proto-horses. To repeat: unless something triggers the same genetic repackaging in a significant percentage of a species within the same generation, a new species with one more or one fewer chromosome pairs cannot arise.

It is believed that a widespread virus or retrovirus infection is the probable triggering mechanism.

I've talked of horses and donkeys because mules are well-known. In many ways they are superior to both their parents. But they are sterile. Something similar must have happened in the very recent evolutionary history of homo sapiens. We have a different number of chromosome pairs to our nearest extant related species, the chimpanzee. The causative virus may still be out there. Or if it's a retrovirus, it may still be latent within us.

  • $\begingroup$ You've forgotten, or are neglecting to mention, the other way to get a change in the number of chromosomes. $\endgroup$
    – user3082
    Commented Sep 28, 2015 at 11:38

Big initial population means low genetic drift. Go a lot smaller to start, or bottleneck the population at some point (massive planetary disaster, or biological problem).

Selection effects or environmental pressures could conceivably move things along (ie: omega-3 fatty acids in Inuit populations, lactose-tolerance in Europeans, etc).

And I'd suggest that robot guardians are the way forward. Have them select against the violent, the criminal and the insane... And all of a sudden pleasing robot guardians is a selective effect/environmental pressure on the population.... Irony express, here we come.

Radioactivity will increase cancers to start, and will probably just make more broken people. If the overlords didn't cull the humans when the first million were brought, then we've already got a ton of broken genes - a few more won't register with their filters (as they're already loose enough that all of those people qualified as 'human' to start).

Might be better if the robot's programming / reference sample were corrupted, hit by cosmic rays after the shielding got destroyed in the disaster (strike by extinction-level sized meteor - backups already lost due to criminally insane taking out a couple facilities)

The protection from alien species would probably result in the elimination of any mutants who didn't fit whatever the profile is, instead of the robots turning on the population as a whole - if you want to have the humans attempt to speciate, you're going to have to let mutants run rampant in the population, and out-breed the ones who do fit the human profile. Which means the robots have to do a pretty bad job for quite a long time before getting back to work.

400 is pretty much as low as you can go, and be viable, and will definitely have potential drift. However, that population needs to be kept small (think Darwinian small islands with open niches vs. continents). If you have explosive growth (16+ kids per family, say) because there are no natural limits; ie: it's a paradise, and everyone is living on the beach and fish are jumping into your hands.... then the drift doesn't happen, except if there's a 1/400 rare gene and that guy dies initially. It's when the population gets crunched that lines die out, and drift occurs.

Differences in these variables result in very different time-frames (if anything happens at all). We're also tool and culture using, which almost completely eliminates most selection from happening. How many people do you know who died in childhood, or were unable to breed?


What is the definition of "homo sapiens"?

Would a chimp at a local zoo be targeted for elimination by the guardians? Any alien is going to be far more different than a chimp (or a cow, or a pig, or a chicken).

Babylon 5 had an episode that investigated a similar set up for a destroyed civilization. Soldiers had been created (using biotech on suitable individuals) and were programed to purge the impure from the population.

The definitions were more ideological rather than physiological, so no one was actually pure enough, and the entire population was wiped out.



A retro-virus inserts its DNA into the DNA of the host cell. Usually that is a regular somatic cell. However, it is theorized that retro-viruses can also insert their DNA into a gamete (sperm/egg) cell, and when that gamete creates an offspring, the virus' DNA is permanently incorporated into the somatic cells of the offspring. Then that offspring will create gametes with the virus' DNA and pass it on to the next generation!

Let's say a retro-virus infects a single sperm. That lucky sperm joins an egg and contributes its half of the DNA+virus DNA to the new baby. That half could have genetic dominant traits that cause enough changes to mark the baby as a new species. Now multiply that out:

For the population at large: you could have a highly contagious and infectious but harmless to the host retro-virus that attacks gametes both sperm and egg, thus making every baby diploid with the new DNA. (You might have to explain what effect the retro-virus has on such babies.) The new DNA should not be fatal, obviously, since you want the babies to be killed by the robots. How cruel you are!

But you're storyline has to wait a while. "Species" is often defined by being un/able to produce viable offspring. You'd probably have to wait till these babies reached sexual maturity to find out that they cannot reproduce with non-infected but can reproduce with other infected. (Unless you have a way to determine fertility or reproductive compatibility strictly through blood or DNA or other tests. I mean obviously after you discover the incompatibility issue then you can just screen for the new DNA. I'm just talking about the initial realization that the infected constitute a new species.)

  • $\begingroup$ Let me just add that while random mutation does happen, you are basically changing a single gene in a single person. With retro-virus, you suddenly insert an entire new chain of foreign DNA inside the host DNA. Also, retro-viruses can infect entire populations quickly. $\endgroup$
    – Xplodotron
    Commented Aug 19, 2016 at 21:52

You must log in to answer this question.

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