For a species I am working on, there is a specific genetic trait that I want to introduce into the population. The exact nature of either is not important for the purposes of this question, but the trait is of such a nature that it is visible in the individual's phenotype to others of the species and it provides the individual with a pretty good advantage in certain situations without coming at too great a cost.

Normally, such a trait would spread readily throughout the population as it provides a net advantage and it is clear to potential mates whether a particular individual has it or not, leading to most individuals having it after some amount of time. However, in my world I want this trait to:

  • pop up (at least seemingly) randomly; even if both parents show this trait their offspring may not, and even if neither parent shows the trait their offspring might, so it cannot be a simple dominant or recessive inheritance model (maybe polygenic?)
  • occur in something like 0.01% to 0.1% of the population sustained over time (this is the hard part...)
  • while showing some variance in the degree through which it shows in the phenotype in the individuals that do have it, be a marked difference between those individuals that have it and those that do not
  • exist, and have similar effect, in both males and females

Without invoking magic or phlebotinum, can I make that work? How?

Assume Earth-like biology and evolutionary processes.

  • $\begingroup$ It could be recessive, or the population could simply have few evolutionary pressures that make it likely to change. For example, a recessive trait causes people to be able to make a coin hover, however, it's not like the ladies are lining up to breed with this guy. Probably. But in wolves, it's different. They work very hard to hunt prey, so an awesome hunting and fighting wolf can easily become he leader of his pack and have lots and lots of puppies. $\endgroup$ – Xandar The Zenon Feb 16 '16 at 2:42

10 Answers 10


We can look to sickle-cell anaemia for an example.

Three quarters of sickle-cell cases occur in Africa. A recent WHO report estimated that around 2% of newborns in Nigeria were affected by sickle cell anaemia, giving a total of 150,000 affected children born every year in Nigeria alone. The carrier frequency ranges between 10% and 40% across equatorial Africa, decreasing to 1–2% on the north African coast and <1% in South Africa. There have been studies in Africa that show a significant decrease in infant mortality rate, ages 2–16 months, because of the sickle-cell trait. This happened in areas that were known to be predominant areas of malarial cases

The heterozygous phenotype can be the "beneficial" phenotype, and the homozygous recessive phenotype can be the deleterious phenotype. As long as there is a fitness balance between the two, the selection pressure on the allele as a whole should be balanced, and allow there to be a constant amount of people with the beneficial phenotype. As long as the population is sufficiently large, the gene is unlikely to become extinct.

  • $\begingroup$ I think this is a good answer (the best one, potentially), but needs expanding with perhaps some popular-science style explanations. Not everyone understands "heterozygous" vs "homozygous", which is why I think you are not getting many up-votes. $\endgroup$ – Neil Slater Jan 18 '15 at 19:45
  • $\begingroup$ So basically, my designed trait would be the opposite of sickle-cell anaemia in terms of inheritance? If I'm reading Wikipedia correctly, SCA is caused at least in part by heterozygous inheritance of the haemoglobin gene. It seems to me like such an inheritance model, perhaps plus it being a recessive trait, could very well work. I like the idea; it is definitely plausible. $\endgroup$ – a CVn Jan 18 '15 at 21:41
  • 1
    $\begingroup$ And @NeilSlater, just look at how many upvotes people get for using $ \LaTeX{} $ in their answers. :P $\endgroup$ – a CVn Jan 18 '15 at 21:41
  • $\begingroup$ @MichaelKjörling Not the opposite of SCA, but the exact identical mechanism of inheritance (in the sense that the heterozygote has the beneficial trait but the homozygote does not). You can add some other genes for it to skip generations via polygenic inheritance, but the main gene at hand works with Mendelian inheritance. $\endgroup$ – March Ho Jan 18 '15 at 22:42
  • 2
    $\begingroup$ @AE Yes, this is correct. If the net effect is advantageous in terms of fitness, the allele will change in frequency by natural selection. The only way for allele frequency to stay fixed is for there to be zero net fitness for the allele. Under such a circumstance, the only effect leading to shift in allele frequency would be genetic drift, which can be countered by large population sizes. $\endgroup$ – March Ho Jan 19 '15 at 12:33

If it is caused by a combination of several otherwise unrelated dominant traits each of which in separation also causes a slight decrease in fertility or increase in mortality.

March Ho had an interesting idea about heterozygous phenotypes. This means that the gene needs two different alleles to have the trait. The benefit is that only half the children will have the phenotype, which prevents creation of pure-bred lineages, which using just dominant traits would allow. Downside is that half the children would have the trait, which would be noticeable by simple observation, after which new children with the trait could be produced by simply sacrificing the other half.

However, by combining this with my idea of requiring multiple genes in different chromosomes and making "incomplete versions" slightly negative... You'll end up with something that can't really be bred for.

Also your compound trait is rare enough that it is unlikely to make its component parts that much more common as long as there is enough of them and some of them are negative.

Something like that would probably be sensitive to mental state and health.

Human generations are fairly long, so as long as the balance of probabilities and selection pressures is roughly correct, nobody would notice slight variations.


What if it does permeate the society, but not everyone shows it?

Genes themselves don't do anything, its the proteins they describe that do interesting things. The production of these proteins gets upregulated and downregulated constantly. You could focus on them.

Consider crafting a beneficial gene for which upregulation is tricky. Perhaps that section of the genome is only active during the formation of the neural tube. Make the trait tricky to implement... messing up could be fatal for the baby. If the baby has the genes, it could be a roll of the dice whether the embryo implements it properly. On failure, it could simply downregulate the gene, saving the pregnancy.

This would allow the gene to spread through all of society (as we expect from evolution), but individuals only have a small chance of benefiting from it.

  1. Your right, it is probably a multi-gene combo

2-3 to keep it to a small % the visible variance could be a large piece of the puzzle. The visible items could be very unappealing for sexual partners, or even more simple that it also is paired with a reduced ability to procreate.

4b. I don't see why there would have to be any difference between in effect for male vs. female. The procreation can be affected on both sides. There could also be something that makes the pregnancy more likely to fail. Maybe they also always are born early, often too early.

4a. Or the pregnancy could cause some other issue that might often result in the mother dying. This would keep them with the visible characteristics more of a pariah, if the mothers often die then the women will be less likely to want children and the men will be less desirable as a partner if their kids will kill their mothers.


Don't forget that humans generally have two genetic lineages: the standard nuclear one we inherit from both parents and the mitochondrial one we get only from our mothers. This trait you talk about could be carried along in some analogue to the mitochondrial lineage. If you look at the mitochondrial diseases, you will find that some of them are exceedingly rare while other are more prevalent, depend on various factors, including interactions with nuclear genes and stubbornly persist across generations.

This mechanism could provide sufficient space to create a plausible scenario.

Or ... just look at the prevalence of some genetic conditions and focus on the ones with the prevalence you desire and note how they work.


Your trait, as you've described it, has the same patterns of inheritance as most cancers, mental disorders, and the majority of human traits. These phenotypes are the result of genes with incomplete penetrance. Penetrance is a term biologists use to describe genetic traits that do not always show their corresponding phenotype. For whatever reason, be it environmental or other genetic factors, only a percentage of the population that has the genotype for the trait show the trait phenotypically. The genotype does not completely determine the phenotype in an individual, but instead strongly influence the probability of the phenotype.

As for how to keep your trait from being positively selected for and becoming fixed in the population, you just need people with your trait to not have more children than people without the trait. There are many ways to do that both socially or biologically.


Random and rare: you want it to be an advantage to the species only at those low numbers.
Once you have that, evolution will have made the connection to the phenomena that trigger/regulate the process, no matter how ridiculous the connection may seem to us.

  1. The gene is present in most of the population, but manifests rarely, AND causes problems if it manifests too densely (preventing selection for increased manifestation chance):

    • social problems - possibly instinct-based, or just because of tradition. It doesn't do to have too many powerful leaders, and there might be a reason (like long-distance telepathy) not to let defeated rivals live.
    • renewable resource usage that is not a problem otherwise. Maybe even affects 'normal' beings - with harm/threat magnitude growing exponentially with the number of uplifted on the continent. No problem if regulated, though.
    • variant of resource usage: costly to attempt producing such an individual (stimuli are rare, but renewable, e.g. peculiar astronomical/atmospheric/geological phenomena, unpredictable and expensive to reproduce - sounds like magic, but happens in nature)
    • the presence of one uplifted individual suppresses the growth of any who are weaker (resource hogging, pheromones, state of mind).
    • manifesting the trait makes it less likely to be active in offspring
    • too great a density would bring malady, harming the targets and anyone nearby (descendants). Maybe affected individuals are more likely to attract the attention of something undesirable.
    • the trait lies dormant in many people, but most choose not to activate it (religious reasons? expectation of a different lifestyle? Suppression of mating instinct?), or don't get a chance to (traumatic experience + downtime needed?)
  2. Have the trait sustained in another species that have a symbiotic relationship with the Jed... ugh. The capability to uplift a host is useful to have, but the beneficent's evolutionary goals may prefer infrequent usage.

    • example: gut bacteria for whom hosts displaying the phenotype in even the least degree are less... hospitable. The bacteria can function, but outside of an already-changed host, the necessary adaptation is a burden, so they are inferior to the rest of their species. This keeps them defaulting to 'normal' mode, but the host species, overall, benefits from that occasional quirk, and indirectly so does the symbiote.
  3. "without coming at too great a cost" to the adult individual, I assume. The parents are a different matter, as is the infant.

    • sterility/sickness afterwards
    • high risk of failure (no effect, death, or opposite effect)
    • the child is indistinguishable from a sickly runt until the alternative growth starts

The key phrase is " beneficial in certain situations", as long as those situations are rare it will not spread quickly, if most people will go through their life without ever encountering those situation even better. if you can make it detrimental in other situation even better. if manifestation is not guaranteed again even better, the more muddy you can make the benefits of the GENE the slower the gene will spread. Another way is to make the benefits conditional on non-genetic factors, maybe you need to have built up sufficient body mass or bone density or live in a certain type of environment else the effect is detrimental not beneficial.

example super strong muscles are great unless you don't have bones or tissue that can take the strain, then its very detrimental.

Another example, The ability to read minds could be horribly detrimental if it manifests in the middle of a city or town but if it happens in the middle of wilderness you might be able to learn to control it before the constant noise drives you insane.


One of the major, if not the major, theories of how speciation occurs is allopatric speciation. You mention how a large population drowning out a new trait before it can spread is an issue with you, and so it is in conventional theory. This model posits that at the edge of a species' range, small groups will form a cluster breeding mostly internally, thus allowing a favorable change to take over the group. This group will also have additional pressure from the marginal fit it has to the environment, and thus a 'need' to evolve. Then if the environment in the whole changes to fit this group, it may spread out over the whole range.

  • $\begingroup$ Ehm... I would encourage you to re-read the question. My problem as stated in the question is that in most situations a net beneficial trait would spread throughout the population, leading to a large fraction of the population having that trait after some time; the goal of the question as asked is to maintain a given (net beneficial) trait in a small fraction of the population. Your answer seems to be the opposite of this. $\endgroup$ – a CVn Jul 16 '15 at 20:49

If the population is sentient, they can decide to restrict it. It doesn't matter how genetically beneficial it is if it is countered with, only the kings family can have it, if you aren't in the kings family we kill you, your parents, your cousins and just to be safe, your little doggie too.

Humans aren't immune to evolution, but we can uniquely influence it.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.