It's easy enough to figure out how a species might evolve a third biological gender in general (e.g. some sort of asexual drones/workers, ala insects).

But what I want to build is a system where the species evolved 3 "real" sexual genders, where all three take active part in sexual reproduction.

What would be a realistic evolutionary path for such a species (intelligent, but not necessarily humanoid) to develop?

I would accept plausible theoretical answers, but would prefer those based on actual evolutionary biology material.

Assume that the environment is somewhat Earth-like, but not necessarily exactly Earth; and no "unnatural" selection pressure occurs, such as alien experiments, magic, or panspermia introducing genes for 3 biological genders from another planet.

My own assumption is that this is NOT the only species with 3 biological genders on the planet, but that's from my understanding of biology and not some needs of the setting.

  • 1
    $\begingroup$ This is 100% wild speculation and opinions. We (as humanity) have no idea, never encountered anything like this in nature AFAIK, and have no guide/rules of thumb to distinguish implausible opinions from plausible. Some sci-fi tried to explore such settings but all are wild guess. $\endgroup$ Dec 17, 2014 at 21:57
  • 4
    $\begingroup$ Are you asking about a third gender or a third sex because those are different things. $\endgroup$
    – kuwaly
    Dec 17, 2014 at 22:13
  • 1
    $\begingroup$ A model with three sexes all which are needed to reproduce would be speculative AFAIK. Models with multiple sexes with a large number of viable pairs are pretty common and relatively well understood IIRC. (It is about increasing the chances that two individuals that meet are able to reproduce together.) Which of the two models are you interested in? $\endgroup$ Dec 17, 2014 at 22:22
  • 3
    $\begingroup$ The obvious answer is a situation where chromosomes group in threes rather than pairs, and meiosis splits a nucleus into a germ cell containing not half, but a third of the original. It seems a bit overcomplicated but once a system gets started it is hard to change. $\endgroup$
    – Oldcat
    Dec 17, 2014 at 22:55
  • 1
    $\begingroup$ Another sci-fi example - Flight of The Dragonfly (Rocheworld) aliens do not have sexes. Each parent splits off part of its body, the split off blobs merge and then form into a new individual. There can be a large range in possible numbers of parents involved in creating one child. $\endgroup$
    – Tim B
    Dec 18, 2014 at 9:43

7 Answers 7


No such reproductive systems have been recorded on Earth. It is postulated that a two-gendered reproductive system provides sufficient advantage in genetic variability of offspring while reducing the disadvantage of having to attract mates. In a two-sexed organism, each must attract one mate, but in a three-sexed organism, one must attract two separate mates - if that is part of their reproductive cycle.

However, evolution is not so much "survival of the fittest" as "survival of the adequate". If the tri-gendered organisms had sufficient advantage over asexual or bisexual organisms, then that would become the (or at least a) dominant system of sexuality.

A true 3-gendered reproductive system (as opposed to a reproductive system that required a male and a female, but had more than one would require a set of genetic material that could be split into three or more parts. Each of the three genders would be capable of transmitting one or more parts of this multi-part genome, but none would transmit more than N-2 of the parts, in order to keep the requirement for three genders.

In the following discussion, I will use the terms 'amale', 'bmale' and 'cmale' to refer to each of the three genders, to distinguish them from our two-gendered 'male' and 'female'

We could envision several different models for tri-gendered reproduction. There could be:

  • External fertilisation - the three genders get together (or just nearby) and release gametes, which fuse to form an embryo.

  • Single Accumulator Gender: an amale and a bmale are required to provide genetic material to the cmale.

  • Multiple Accumulator Genders: An amale produces a gamete, which is then fertilised by a bmale, and the resulting intermediate embryo is then passed to and fertilised by a cmale before the embryo matures.

All variants of this would require that in essence one gamete fuses with a second type of gamete, which then fuses with a third type - it would be much less practical to require fusion of all three gametes simultaneously. The relative sizes of each gamete could vary considerably, as could the quantity of genetic material in each, possibilities include 1:1:1 (ploidy 3), 2:1:1 (ploidy 4), 2:2:1 (ploidy 5), 3:2:1 or 4:1:1 (ploidy 6), and many others as ploidy (the number of similar copies of the genome the organism possesses) increases. The exact proportion of genetic material provided by each of the three parents would dictate how involved each is likely to be in raising any offspring - if any are involved at all - the more genetic material is contributed, the more involvement is likely.

Assessing the likelihood of each of the possible three-gendered reproductive systems, the most likely is probably an external-fertilisation water-based scenario - all genders accumulate in a particular area and release a cloud of single-celled gametes. Since no parental responsibility is required in similar reproductive strategies on earth, there are no mate-selection issues. However, this method of reproduction is typically used by r-strategists rather than K-strategists which means that a great many 'disposable' offspring are created, and parents play little or no role in offspring survival, and hence sentience is probably less likely.

A Single-accumulator strategy means that two genders, amale and bmale, must each attract only one mate, and the cmale must attract two. If the cmale contributes the majority of the genetic material, it is likely that this is a viable option for a K-strategist, with few offspring that are cared for to a great degree by one or more parents. This would be more likely to result in a sentient species, as transmission of ideas from parent to child is more likely.

Multiple Accumulator Genders requires that the amale mate with a bmale and then a cmale. The cmale would be most responsible for raising any offspring if that was part of the reproductive strategy, and the amale may be more selective than the bmale, and the cmale most selective of all. Again, the proportions of genetic material from each of the three parents would affect the degree of selectivity and involvement in any effort to nurture the offspring. Again, this is more likely to result in a sentient species.

A variant of this scenario is a system where the population alternates between a multicellular multiploid organism (having several copies of a set of genetic material) and multicellular organisms having lower ploidy. An amale may asexually produce (at the same time or temporally separated) bmales, cmales and dmales. A bmale, cmale and a dmale may then fuse, or mate in one of the patterns above, resulting in a new amale. This may be an option for either an r- or a K- strategist organism, and could result in one or more of the four different forms being sentient.


Monty Wild is incorrect in stating that "No such reproductive systems have been recorded on Earth." In fact, such reproductive systems have been recorded on Earth.

The most complex gender arrangements in real life involve fungi. Some fungi are tetrapolar (four gendered) and there are or have been tripolar fungi. It isn't clear that any tripolar fungi exist naturally today, but tripolar fungi can be and are bred as hybrids.

An evolutionary path to tripolar can involve going from bipolar to having the option to reproduce sexually or asexually, to having the ability to hybridize with another similar species with two somewhat different genders (often expressed A, a, B, b), to having to have all four genders to reproduce sexually, to having one parent that has two genders and one that self-reproduced and has one gender.

Another similar arrangement is called a patchwork virus:

Scientists found a virus that is made out of 4 to 5 separate components - it infects mosquitos, and they have to catch at least four of those components to get infected, the smallest, fifth component is optional. For plants and fungi, similar viruses were known before, but (at least according to the study) this is the first example in animals studied in detail.

The number of patches in a patchwork virus can vary greatly. You could have a female analog that would correspond to the "host" and might or might not be sentient (Ringworld by Larry Niven, for example, has a species with both a non-sentient gender - perhaps historically a reproductive host of a parasitic species - and a sentient gender), and then the elements of the patchwork virus function as gametes.

One can imagine a scenario in which sex involving just two of three genders is purely recreational, while involving all three genders is reproductive. One functional benefit of this would be that reproductive sex would require coordination, cooperation and deliberation. Something similar exists in modern hyenas. In hyenas, both the male and female must have the equivalent of an erection at the same time which makes a rape pregnancy almost impossible in hyenas, unlike almost all other mammals. This gives females much more reproductive choice and they have utilized it to produce a thriving more intelligent and socially coordinated species than any other megafauna carnivore. A tripolar gender system could work in a similar way - insuring reproductive choice. Also, simultaneous tripolar reproduction is much sexier than sequential tripolar reproduction.

Another way to get multiple genders is to have gender determined by environmental conditions at the time of conception or gestation which is quite common in many species (including vertebrates). Usually the environmental conditions involve temperature or pH (i.e. acidity), but one could imagine humidity or all manner of other conditions (e.g. pheremones from existing community members reflecting current gender ratios) playing a role in determining gender.

There are also three parent humans in real life, some via intentional genetic manipulation of multiple parent sperm and eggs, and others alluded to by HilWithSmallFields involving a sperm gamate donor and an egg gamate donor whose fertilized egg is implanted in a surrogate mother who contributed the mitochondrial DNA as I understand the process (apologies if I am incorrect, but the three part split is sensible fictionally, even if it doesn't exist in real life). Also, most women who have children in humans are at least partially in some parts of their bodies, chimeras, incorporating some of their children's DNA (and indirectly, their partner's DNA) into themselves.

Of course, humans have various non-binary gender options. Most notably, gay men, lesbian women, bisexual individuals, transgender individuals (both homosexual and hetrosexual), and even a very rare group of individuals who cycle through the course of a day or a month from one gender to another (a bit like transgender but unstable). In addition there are people are basically neuter (castrates and people born sexually ambiguous). There is strong anecdotal evidence for femme and butch sub-identities within a lesbian sexual orientation that have a biological basis associated with levels of testosterone exposure in utero, and for the equivalent in gay men, at least some of the time.

The Netflix series Hemlock Grove has an edgy scene in which a werewolf, a vampire and a human woman have a menage a tois that illustrates the dynamics of what a trigendered sexual relationship could look like (compare Twilight which has another love triangle of that type). Notably, many versions of the tales allow vampires and werewolves to reproduce both sexually and asexually - an asexual reproduction option seems to be a good foothold into a trigendered system. The Vampire Academy series by Richelle Mead also features a complex gender system involving living vampires, dead vampires, humans and dhampirs. And, the Vampire Diaries series also has some serious gender/species complexities.

"Being Human" (a TV series with a British and American version of exactly the same script) doesn't have gender complexity to it, it does have multi-species households that could provide further insight into the social side of the relationships.

I have read an interesting science fiction short story in which individuals had sex via a "bridge" which could be transferred from one partner to another during sex changing the gender of the people involved. One could imagine a "bridge" that only activates a seed when a sufficient number of people have contributed to it.

  • $\begingroup$ "non-binary gender options. Most notably, gay men.." I agree with most of this answer but this paragraph is incorrect. Sexual orientation is not the same as gender: Gay men are attracted to men but that does not make them "non-binary" or "neither male nor female", the "men" part makes them "male". You are correct about the transgender part however. $\endgroup$
    – Zxyrra
    Nov 20, 2016 at 0:20

There are multiple examples of third (pr even fourth or fifth) genders in nature. Mostly the follow about the same pattern.

Rather than requiring three or more individuals to reproduce, what's commonly found is the existence of multiple 'male' or 'female' genders within a population. These genders differ physiologically from one another, and often exhibit different breeding strategies. This is best indicated perhaps by the side blotched lizard.

Within an intelligent society, it may be that multiple genders persist due to an increased phenotypic range that is advantageous to society. For example, if there exists a 'small male' and a 'large male' gender, small males may represent an evolutionarily advantageous way of gathering resources, while the large males are better at defending a settlement. Small and large males would hold distinctly different Y-chromosomes, leading to a differentiation between these two male genders.

  • 1
    $\begingroup$ Very interesting, but those seems to be breeding strategies (3 for male, 2 for female) with phenotype (observable differences). All males/females have exactly same number of genes, right? Article does not say if breeding strategies are inheritable. Even if they are, it is little difference between breeding strategy and other genetically inheritable traits like height and hair color. $\endgroup$ Dec 17, 2014 at 22:43
  • 2
    $\begingroup$ I'm not sure why small male/big male would be a different gender any more than small men are sexually a different gender than tall ones. Or blondes versus redheads. That is a trait. To be a different gender it either has to have a unique role in sex, or be a mule of some kind. $\endgroup$
    – Oldcat
    Dec 17, 2014 at 22:57
  • $\begingroup$ @Oldcat I would argue that having a unique role in reproduction would qualify something as a third sex, not a third gender. $\endgroup$
    – ckersch
    Dec 17, 2014 at 23:00
  • $\begingroup$ @PeterMasiar I'm not sure about the lizards, but a similar pattern is definitely present (and heritable) in birds called ruffs. (en.wikipedia.org/wiki/Ruff) Interestingly, the gene locus that determines male mating strategy and physical appearance is located on an autosome. $\endgroup$
    – ckersch
    Dec 17, 2014 at 23:02
  • $\begingroup$ Fair enough - I assumed that's what the OP meant. Otherwise, its trivial, as something as simple as opting out of sex could be seen as a different gender. $\endgroup$
    – Oldcat
    Dec 17, 2014 at 23:03

Given the lack of evolutionary evidence to draw from, I would like to draw from a different source: mixology.

When making a Martini, there are three fundamental ingredients: Gin/Vodka, Vermouth, and a Garnish. The nature of each of these is essential to the taste of the Martini. Alton Brown has argued that these three ingredients act like a chord. You can have major cords, minor chords, diminished chords, but they always start from one-three-five.

I could see a system where three sexes come into play:

  • A "mother," the Gin/Vodka of the relationship. This is the workhorse who is going to have to carry the child through birth.
  • A "father," the Vermouth of the relationship. They add genetic material in a way which allows for a mixing of genes which is more rapid than asexual reproduction could have allowed.
  • A "muse," the garnish. They add genetic material which is selected for a higher rate of mutation. This gender could be responsible for more consciously selectable traits, such as "musclebound." Due to its high rate of variability, this could be an excellent candidate for storing "genetic memory" from generation to generation.

Socially, the mother and father could be tighter, with the muse being brought in to facilitate the creation of their "perfect child."

Evolutionarily it is reasonable. The purpose of sexual reproduction is believed to be the ability to increase adaptability by improving genetic mixing. It seems reasonable that a third gender could offer similar mixing on a much more rapid timescale.


Nice link about ants which @ckersch found. Different strains of ant males and queens exist, and different caste of ant will be produced after mating, depending on strain. But each individual still seems to have one or two parents - one (unfertilized egg) to grow males or two (fertilized egg) to grow queens (same strain male) or workers (other strain male). Good read.

But still, if it is evolutionary advantageous to have sexual reproduction (to get different set of genes from your own, which may give your offsprings advantage), it is not obvious why having more two sexual partners necessary to reproduce will be more beneficial than just one (or none).

So plausible solutions seems to be insectoid society with biologically separate castes, resulting from mating with different strains of males.

If you create more castes, you may have more strains and combinations for mating. To keep it plausible, each individual has max 2 parents.


One environmental setup that may have caused 3 genders would be an environment which:

  1. Has VERY little environmental mutagens in a specific location

    Low radiation, including solar; low rate of chemically-induced mutations as well

  2. High-mutagen environments on the borders between locales.

    #1+#2 means that individuals living far away from each other are very genetically diverse (because their ancestors crossed high-mutagen lines, AND their environments are somewhat different), while those living near each other are very genetically close.

    This introduces a strong pressure to introduce "far-away" genetic material into procreation if you want high genetic diversity

  3. Environment can change fast and unpredictably, meaning high genetic diversity is required to survive.

    Meaning that this "far-away" genetic material is a requirement need is a vital requirement, not a nice-to-have.

  4. High predation from tough predators, which means that at least one gender must evolve to be a strong fighter with good armor.

    This cannot be the child-bearing gender, because the biological cost of fighting/defending predators is too high if combined with biological cost of nurturing an offspring for the same individual (takes a LOT of biological energy to grow muscle and armor; and to train).

    Unfortunately, the same energy cost for a fighter means fighters aren't very mobile - that would require small/light body and preferably flight, which armor and fighting muscle heavily inhibits - see #5.

  5. High biological cost of geographical movement

    For example, small cells that can be used for foraging, separated by high walls (mountains/crater walls). As noted in #2, the highlands on tops of those walls are pretty much the only place where your genes are mutated.

    Let's make up some numbers:

    • An adult organism needs X energy to survive WITHOUT moving.

    • Birthing/raising childred needs X energy as well

    • Traveling far/well enough to gather 5*X energy requires additional 2*X energy for short term travel

      • Meaning, a "defender" sex energy budget is balanced (gather 5*X food, spend 2*X on travel, X on self-survival, 2*X on giving to birther-sex partner for survival/children).
    • However, traveling far/well enough to move from location cell to location cell requires additional 2*X energy.

      • Meaning, to travel very far, your energy budget is only balanced if you don't share energy with anyone - you gather 5*X, spend X on survival, 2*X on moving within a cell to feed, and 2*X on moving to the next cell by scaling the wall.

Based on #4 and #5, an average species's member can only forage enough energy to EITHER travel, OR reproduce/nurture offspring, OR gather food and defend the family from predation. But not 2 out of 3, never mind all 3.

In this situation, we have a conundrum:

  • To reproduce, you require at least 2 species, a birther/nurturer (equivalent of 2-sex female), and a provider/defender (equivalent of a 2-sex male).

  • BUT, with just those 2 sexes, the species stagnates and dies out, because NEITHER of those 2 sexes is adapted to traveling far, at all.

  • Ergo, the selection pressure favors an adaptation resulting in a Third sex, which is evolved specifically to travel far and wide (adapted to travel at this terrain, light, fast {possibly a flyer}, high endurance, able to withstand more variation in local conditions, and not good for anything except moving far and delivering genetic material there (because its unique adaptations make it poor defender, and poor birther). Basically, third sex acts as long-range sperm/seed, to put it crudely.


One possibility that isn't that different from what we now have would be to split the present female double role, of providing nuclear genetic material and providing the mitochondrial genetic material, into two, so the nucleus of the cell that becomes the new individual is produced from a large gamete (egg-like, but with negligible cytoplasm and no mitochondria) and a small gamete (sperm) that pushes into it. The resulting fused nucleus would then need to be implanted into a cell produced by a third sex, that provides the mitochondria and the bulk of the cytoplasm.

The nearest to this that I can think of in existing organisms is the reproduction of lichen (symbiotic systems of fungi and algae) where the fungi reproduce sexually and the newly produced fungus has to connect to some of the energy-providing algae.


You must log in to answer this question.

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