Cross-posted from substack.

"Everything in the world is about sex, except sex. Sex is about clonal interference."
– Oscar Wilde (kind of)

As we all know, sexual reproduction is not about reproduction. 

Reproduction is easy. If your goal is to fill the world with copies of your genes, all you need is a good DNA-polymerase to duplicate your genome, and then to divide into two copies of yourself. Asexual reproduction is just better in every way:

It's pretty clear that, on a direct one-v-one cage match, an asexual organism would have much better fitness than a similarly-shaped sexual organism. And yet, all the macroscopic species, including ourselves, do it. What gives?

Here is the secret: yes, sex is indeed bad for reproduction. It does not improve an individual's reproductive fitness. The reason it still took over the macroscopic world is that evolution does not simply select for reproductive fitness.^[1] 

Instead, the evolution of sexual dimorphism is a long sequence of strange traps, ratchets and outer-world eldritch cosmic forces that made it somehow inevitable. So let's talk about those things your parents never told you about.

The birds, the bees, and the fission yeast

What bugs me is that, not only most people have absolutely no idea why sexual dimorphism exists, but they seem entirely fine with that. Our lives are punctuated with all sorts of frankly weird practices related to it, but the reasons we ended up there remain obscure even to many biologists.

So I figured I would write up a summary of some popular theories. This way, when time comes, you can explain to your children the long evolutionary trajectory that culminated in VR ChatGPT cat-girlfriends.

(Note 1: As always with evolutionary biology, everything in this article is subject to uncertainty, controversy and mystery. Always keep in mind the Golden Rules of biology: all models are wrong; everything has exceptions; don't talk about fungi; mitochondria is the powerhouse of the cell.)

(Note 2: As this is a bottomless topic, I’ll have to make some cuts. I know you’re burning to learn about Pseudobiceros hancockanus’ penis fencing, but I can’t cover everything.)

First, let's get something out of the way.

Something something diversity-generation

I often hear vague explanations about sex being a way to generate genetic diversity. I don't find it compelling. If you want genetic diversity, you can do it in much easier ways than turning into a sexually-reproducing dimorphic species. One of them is, just increase the mutation rate, bro.

Bacteria are good at this. E. coli comes with a whole toolkit of DNA-polymerases with various degrees of accuracy. When everything is going fine, they use the most accurate one to faithfully replicate their genomes. But, in case of particularly bad stress, the bacteria start expressing error-prone polymerases, which increase their mutation rate. Who knows, if the mother cell is going to die anyway, some of the mutant offspring might stumble upon a solution to escape the bad situation.

All that is to say, raw genetic diversity cannot be the whole picture. It has to be a specific kind of genetic diversity.

Part 1: the evolution of sex

Most of the articles I read about the evolution of sex ask "what are the advantages of sexual reproduction?", then proceed to explain what are the advantages of sexual reproduction. The problem with this approach is that, if sexual reproduction really had such clear advantages, nobody would do asexual reproduction any more. But, to this day, asexual species are still very much around and successful. What we need to know is, "in what ways does sexual reproduction give access to new evolutionary niches?"

So, what do all sexually-reproducing organisms have in common? For context, sex evolved about 1-2 billions years ago, before multicellularity, but later than photosynthesis. It's very closely associated with the eukaryotes, a clade that appeared when archaea started eating bacteria, and the bacteria turned into organelles like mitochondria. In fact, mitochondria seem to be closely related to sex: to my knowledge, organisms with mitochondria always use sexual reproduction, as if mitochondria made sex necessary. Could there be a link between these two?

Before answering that, let's examine what sex does to your genome.

Genetic hitch-hikers and clonal interference

Let's start with an innocent, asexual bacterium. It reproduces by dividing itself into two daughter cells, who then proceed to do the same, and so on. Sometimes, a mutation occurs somewhere in the DNA. If it's a bad one, it will soon vanish from the population. But if it's a good one, the mutant will reproduce quicker than its siblings, and the descendants of this mutant will eventually take over the whole population. That's evolution 101.

What happens if new mutations occur at a faster rate than natural selection can sort them out? Two bad things can happen.

The first one is clonal interference. This is when a second beneficial mutation occurs in an unrelated cell before the first good mutant has time to take over. If the second mutation is even better than the first one, then it is that mutant that will take over, and the first mutation will be lost forever. That's too bad, because it would have benefitted the species.

The second problem is genetic hitch-hiking. This is when a beneficial mutation occurs in the same lineage where a detrimental mutation just occurred. If the good mutation has a larger effect than the bad one, the mutant will still grow in frequency, and the bad mutation will extend to the whole population.

That's quite a big problem. Since it's easier to break things than to improve them, the majority of possible mutations are bad ones. Thus, if a cell finds one beneficial mutation, it will often come with a bunch of detrimental hitch-hikers, and there's no way to get rid of them. This is called Muller's ratchet and in some conditions it can make fitness decrease as a result of natural selection.

The Fisher-Muller model^[2]

Enter sexual reproduction. Instead of getting all the mutations from the mother cell, the newborn cell receives an assemblage of random pieces of each parents' genome. As you repeat the process, you end up with many different possible configurations. Among them, hopefully, there will be newborns with all the good mutations, and none of the bad ones.

In effect, sexual reproduction parallelizes natural selection, as each variant gets tested separately in a different individual.

That’s the theory. Does it work in practice? McDonalds et al. (2016) had yeast evolve with and without sex for 1000 generations and sequenced them at regular points in time. In the following plot, the blue alleles are bad, the orange ones are good. It's pretty spectacular:

The asexual lines caught a lot of bad mutations by genetic hitch-hiking, while the sexual lines managed to purge all of them while retaining the good ones.

And, despite the cost of sexual reproduction, the sexual lineage (orange) was able to evolve more efficiently, and over the long run its fitness improved much faster:

This is a typical example of second-order selection.^[3]

I can hear you complaining, "this is entirely unrelated to mitochondria". Where do they fit in the picture? And why do organisms without mitochondria get away with the old asexual reproduction scheme?

Hot, hot DNA

Remember, we are ~1.5 billion years ago, and photosynthetic organisms have just released massive amounts of oxygen into the atmosphere, leading to the Great Oxidation Event. Some bacteria are starting to use this oxygen for breathing, turning them into little living powerhouses. Then an archaeon swallowed one such bacterium and made it into its very own intracellular powerhouse of the cell, starting the era of the eukaryotes. Unfortunately, heavy respiration releases a lot of oxidizing chemicals, and instead of going to the environment, these chemicals now accumulate in the eukaryote's cytoplasm, creating a lot of DNA damage. The mutation rate goes up.

Asexual reproduction works marvels in the low-mutation, high-selection regime: a mutation occurs, if it's good it takes over, if it's bad it disappears. Then the next mutation arises. Clonal interference is not a problem because mutations are rare enough they get selected one by one.

But if the mutation rate increases due to heavy respiration, both clonal interference and Muller's ratchet become much worse:

Thus, the increased mutation rate due to mitochondria is probably what kick-started the evolution of sex. 

This opened the door for entirely new opportunities: as genome size increases, Muller's ratchet gets worse and worse. Sex makes it possible to have larger genomes, packing more genes, and allowing for more complexity. This paved the way for exciting stuff like multicellularity.

Part 3: not my type

Now we have a cool mechanism for shuffling the genome, but this mechanism doesn't include any mating types yet. That is, we have only one type of individual, who can mate with any other individual. But this is very uncommon in nature. Basically every species we know differentiates into separate mating types, like "male" and "female", that cannot reproduce with themselves. Even baker yeast, who don't have any apparent male-female distinction, still have simple molecular components to switch between two mating types, so they can only mate with yeasts of a different type. (Yes, I know about snails. Hermaphrodites do have exclusive mating types, they just happen to be carried by the same individuals.)

Why aren't yeast pansexual?

There are many reasons why mating types could evolve, and it's not clear which one(s) really happened. Here's a good review^[4], I'll just go over the ones I find most interesting:

Selfing-prevention: exclusive mating types prevent an individual from mating with itself, also known as selfing. Selfing is easy, as an individual's own gamete are already around and immediately accessible. That would completely defeat the point of sex, not to mention inbreeding depression.

The molecular explanation: to have two gametes fuse together and combine their DNA, you probably need some kind of ligand-receptor pair. It could be two surface proteins that bind to each other and cause the membranes of the two cells to fuse. Obviously, this doesn't work well if both cells are identical, as their ligands/receptor pairs would constantly try to bind to each other within the cell's own membrane. Or gametes might release pheromones to attract each other – clearly this doesn't work if the gametes are attracted to their own pheromone. Hadjivasiliou and Pomiankowski, 2016 present a lot of evidence of this happening in various bizarre microscopic creatures. 

Organelle competition: eukaryotes have organelles like mitochondria or chloroplasts, who come with their own DNA and replicate independently. Now imagine a rogue, mutant mitochondrion that replicates much faster than the rest, instead of doing its powerhouse job properly. Nothing prevents this mutant from replacing all the good hard-working mitochondria until the cell is barely functional. But if you have exclusive mating types, you can have a mechanism such that only one parent will transmit its organelles to the progeny (e.g., in humans, the mitochondria come exclusively from the female). Now, there is no competition between individual organelles – the fitness of an organelle is locked to the fitness of the entire organism.

It takes 23,328 to tango

Ok, but why only two mating types? The male/female yin/yang mars/venus duality is something we take for granted, but in terms of evolutionary stability, it sounds like the worst possible arrangement. Say we start with two mating types. If a mutation creates a third type, it will be much easier for the new mutant to find a partner, as it can mate with everyone else in the population instead of only half of it. So it should rapidly invade until it reaches the 1/3-1/3-1/3 equilibrium. Then we can keep adding new mating types – here the optimum seems to be "as many as possible". The mushroom Schizophyllum commune gets it, with its 23,328 different mating types (which makes me wonder what discussions are like in their gender studies departments). But the two-sex binary is by far the most common arrangement in nature. Why don't we all have an interesting sex life like Schizophyllum?

The counter-balancing force here is genetic drift, the variation in a gene's frequency due to random sampling between generations. As you add mating types, the number of individuals of each type becomes increasingly small, and it's increasingly likely that all individuals of a type will be lost to random sampling.

In practice, the number of mating types depends on how many generations of asexual reproduction happen between two sex events, as this governs the relative importance of genetic drift compared to the benefits of an extra mating type. As it turns out, our unicellular ancestors probably did a lot of asexual generations between two mating events, so the number of mating types was pushed to the minimum of two. (In contrast, Schizophyllum, the sexy mushroom, uses sexual reproduction all the time, so it makes sense for it to be so non-binary.)

Note that we are not dimorphic yet. The "males" and "females" might express different receptors and secrete different pheromones, but they still have basically an identical body. The next transition, again, sounds absurdly complicated: you'd have to wire an entire gene regulation program so the population differentiates into two types, which means covering every cell in the organism with appropriate receptors so each tissue knows in what way to develop. Preposterous.

This is when the room temperature drops, rain starts pouring, the old wooden beams scream ominous screeches, and a band of contrabasses starts playing Arnold Schoenberg. Here enters our old friend, Moloch.

Part 3: symmetry-breaking

The next step in our journey takes us from two equivalent sex types producing symmetrical gametes, to males producing swarms of tiny motile minimalistic sperm cells and females producing huge oocytes packed with covid-survivalist levels of food.

How exactly the transition happens is hard to model, because it was certainly influenced by the spatial structure of the environment or the non-linearities in the function “material resources in a gamete → fitness”. 

But we can get a rough idea by considering a species with two types of gametes, whose size is controlled independently by different sets of genes. Each type of gamete can be either big and packed with resources, or small and massively-produced. We will see that, even if we start with everything symmetric, this configuration is unstable and the symmetry will inevitably break to create dimorphic organisms, where each sex produces either big or small gametes. 

What follows is very tedious, but I will do it so you don't have to. (Really, you can skip this and just trust me.)

Consider a diploid amoeba with two genetic loci, M and F. M controls the size of male gametes, and F the size of female gametes. Each locus has two possible alleles: B (big) and S (small). Therefore, a haploid gamete can have four possible genotypes: MB/FB, MS/FB, MB/FS and MS/FS).

What happens if an MB/FB population encounters a tiny MS/FB mutant population? In general, small gametes have a huge advantage. Say a big gamete contains 1000 units of resources, and a small one only 1 unit, but there are 1000 times as many of them. Two big gametes mating together make a 2000-resources zygote, while a big gamete mating with a small one makes a 1001-resources zygote. It means the small gametes can mate with 1000 times as many big gametes, but the zygotes’ resources are only reduced by ~50%, a pretty good deal. So we get a few MB-MB/FB-FB diploids (from MB-FB mating with itself), a lot of MB-MS/FB-FB from the wild-type mating with the swarm of MS/FB gametes, and a tiny amount of MS-MS/FB-FB from the mutant mating with itself. Therefore, the next generation of gametes will have a lot of MS/FB gametes, who will then mate together until they have completely taken over.

This works just as well for MB/FS. What happens if a population of MB/FS encounters one of MS/FB? We get the haploids MB-MB/FS-FS, MS-MS/FB-FB and MB-MS/FB-FS. This gives us a new type of gamete: MS/FS. This one is extremely good at fecundating other gametes, producing MS-MS/FS-FS diploids. However, these diploids cannot mate with each other, as a zygote made from two small gametes wouldn’t have enough resources to be viable. So the MS/FS gametes cannot possibly take over the population. We are left with MS/FB and FB/MS, who are tied, until one of them takes over for an unrelated reason (like a beneficial mutation somewhere).

You can do the same for other kinds of mating, like MS/FS vs MB/FB. You’ll see that, eventually, individuals with one big and one small gamete type always win. 

At this point, the advantage of having many small gametes isn’t so much to produce more viable offspring, but to keep all the eggs for oneself and prevent competitors from fertilizing them. On the collective level, this is far from optimal, since a lot of the small gametes are wasted. It harms the offspring, as they have to start with half as much resources than if the two parents' gametes were big. But the collective optimum turns out to be unstable.

Again, natural selection selected against reproductive fitness – an isogamous species would win over an anisogamous one. But here, we are not talking about the competition between two different species. We are talking about intraspecies competition: an organism versus its own mutants.

And that’s it, we have evolved anisogamy, sexual dimorphism for gametes. From now on, we define the "male" as the type who makes the numerous small gametes, and the "female" as the one who makes the scarce overpowered oocytes.^[5]

Can this dimorphism extend to macroscopic traits, like human breasts or peacock tails?

Part 4: the dimorphification

Here we enter the "look at this funny lizard I found" part of biology.

Before we get to sexual selection, the good old natural selection still plays a big role in generating dimorphism. Now the symmetry has been broken, the species as a whole can optimize the way it does sexual reproduction by specializing males and females' bodies in different ways. Various appendages appear to streamline the process. But I'm sure your parents already explained that part perfectly well.

[Small aside: the oldest known sexually-reproducing organism is the algae Bangiomorpha pubescens. If you think it's funny that the first sexually-active species is called this way, you will be delighted to learn that the first animal known to practice internal fertilization (that is, with a dick) is a fish called Microbrachius dicki. It is named after its discoverer, Robert Dick. Biologists have no sense of humour, this just keeps happening.]

As Darwin himself pointed out, once sexual dimorphism exists, a second kind of selection applies: sexual selection.

The Bateman

Now that female gametes are the limiting factor, the two sexes face new incentives:

• For males, what happened at the gamete-level roughly happens again to individuals. They produce gametes in excess: the best way to succeed is to claim as many partners for oneself as possible.
• For females, if you are already guaranteed to reproduce, mating with more people will not necessarily improve reproductive success. Devoting more resources to the survival of the zygote after fecundation is more important. This includes mating with the best males.
• As a result, what started with gamete size ends with society-level dynamics where males face intense competition, to the point of fighting each other, resulting in a wide variation in reproductive success among males.

This is called Bateman's principle, the Red Pill people love to bring it up, and it is kind of wrong.

First, Bateman's original research is pretty weak. Second, while it's true that the males are most often the ones facing intense competition, it's far from universal – in many species the males (despite producing the smaller gametes) are responsible for all parental care and in these case, it's big muscular females who fight for males. (I made up the "big muscular" part).

Overall, there is no clear pattern about how the two sexes diverge. Things can go all over the place. Perhaps you think humans' mating rigamarole is weird (a typical kenjataimu experience), but really, we are relatively tame. There are extremely unholy things like Sacculina carcini's parasitic castration cycle^[6], Bonellia viridis whose micro-males are so disposable they don't even have a mouth to eat with, or (god forbid) Pseudobiceros hancockanus's penis fencing.

Fisherian runaways

Imagine a species of birds who benefit from having elevated feathers on their foreheads for some reason. Maybe it's for dusting off cobwebs in the nest's ceiling, I don't know. This evolutionary niche impacts selection in two ways – direct selection, and sexual selection. Let's think in terms of genes:

• If a variant makes a bird's top feathers grow larger, it will be directly selected for,
• If another variant makes birds sexually attracted to birds with thick head-plumages, it is also selected for. This is because carriers of this variant will often mate with individuals who carry the head-plumage variant, producing offspring that carries the two variants together. While the sexual-attraction variant gives no advantage by itself, it will rise in the population by hitch-hiking with the adaptive plumage variant.

It sounds like all is good: females will instinctively be attracted to the males with the optimal amount of head-feathers and everybody will win. But we fall prey to another empyrean pagan god and our path runs into another ratchet.

Here's the problem: at the beginning, before selection for thick feathers happens, all the males in the population are below the optimum. There is some variation, but no one comes close to the optimal coiffure. So what instinct do you think will be evolved first?

• Being attracted to the optimal amount of feathers, which is unknown and doesn't occur anywhere in the population yet, or
• Being attracted to as much feathers as physically conceivable, the more the hotter?

You bet.

These two instincts are functionally equivalent – in both cases, the female will pick the male with the most feathers available among the current population. The hitch-hiking will work just as well, and the much simpler "moar feathers" instinct will be selected for.

And now we are in trouble. Let's say this goes on until the average male has just the right amount of feathers. At this point, all the females have acquired the instinct to find head-feathers super hot, and they still want moar. Having more feathers remains a fitness advantage, not because it helps clear up cobwebs, but because it attracts more sex partners. So the average head plumage will continue to expand, way beyond the optimal amount, until every male bird has turned into an 18th-century macaroni.

This whole thing is called the Fisherian Runaway, after our beloved eugenicist Ronald Fisher. And we cannot go back: if a new allele makes a female attracted to a less insane amount of feathers, she will mate with less conventionally-attractive males and will have less sexy sons, so the allele will encounter a barrier when it's time for her son to find a mate.

A possible example in humans is the boob. Other primates don't have boobs – they are flat most of the time and only swell for lactation. Maybe, at the beginning, swollen boobs was a sign of fitness, then human males got really into swollen boobs, then human females started padding them with fat to appeal to the males' instincts, leading to the persistent round boobs we witness today – even if the pad of fat isn't actually very useful for lactation.

(Note that this is one of many hypothesis about the evolution of breasts. It’s a highly controversial subject and an active research topic.)

Sexual selection can lead to all kinds of seemingly implausible phenotypes. If you want more, this review by Michael Ryan has a lot of funny shit ("sand pillars built by male crabs that approximate refugia to females, fins of male fish that mimic food, and male moths that mimic bat echolocation calls"). And let's not forget Basolo's classic Science paper about the sword-less ancestor to swordfish being attracted to human-made swords.

Epilogue: are aliens sexually dimorphic?

If there are other intelligent lifeforms in the galaxy, it sounds likely to me that they evolved through natural selection. It's also likely that they are relatively complex organisms, since they must have evolved some form of intelligence. They may be very different from us, but could they still be sexually dimorphic?

I would guess it's plausible. None of the evolutionary transitions that led to sexual dimorphism are obvious, but they seem almost impossible to escape. Based on what has been observed on Earth, sexual reproduction is virtually necessary to evolve into a complex fully-fledged multicellular organism.^[7][8] Even the very first step, about genetic hitch-hiking, could apply to biological systems completely different from DNA. It's about searching through the space of possible sequences for the fittest one, and how much information you get every generation.  So I would guess dimorphism is more frequent than, say, action-potential-based neurons or sound-wave communication. But, by the time we meet them, they might have engineered themselves into yet another stage of evolution, and none of this will be relevant any more.

Summary

• Unicellular organisms used to divide asexually, selecting one mutation after the other
• Photosynthesis opened the possibility of oxygen respiration, then an archaeon ate a respirating bacterium to make it its personal powerhouse. That generated a lot of oxidative damage on the DNA, leading to a lot of clonal interference between good mutations, and a lot of bad mutations hitch-hiked with the good ones
• This problem could be solved by mating with other individuals and shuffling homologous DNA to generate new individuals with random combinations of the existent mutations, allowing to purge the deleterious ones
• This improved evolvability and made it possible for macroscopic organisms to develop
• For various reasons this system worked better with exclusive mating types, and genetic drift pushed the number of types to two in most cases
• While initially symmetrical, intra-species competition made symmetry unstable and the two types diverged into big gametes and large gametes
• The asymmetry extended to the species' bodies in general, through natural and sexual selection
• Genes coding for sexual preferences started to hitch-hike with beneficial genes
• As the sexual preference instincts were misaligned with the optimal levels of traits, this led to Fisherian runaways, and organisms developed ridiculously exaggerated traits
• And this is how I met your mother.

1. ^{^}

   Here I define reproductive fitness as the average ability of your genes to reproduce. That's it. This may be different from the way you define fitness in general. For a deeper discussion, I recommend Hannah Kokko's great review on the stagnation paradox.

2. ^{^}

   This is also referred to as the "Vicar of Bray" parabola, but I never understood why.

3. ^{^}

   An alternative way to look at it is in terms of the distribution of fitness among a group. Sexual reproduction may decrease the immediate average fitness, but it also increases the variance of fitness, as it creates individuals with all the bad variants, and others with all the good variants (this is not the same thing as having more genetic diversity!). In conditions with only the few individuals with the highest fitness reproduce, then having a high variance in fitness means it's more likely that the fittest individual will be from your offspring.

4. ^{^}

   When the intro of a review ends with "we finally attempt to validate or refute these theories using data on fungi", you know things are about to get wild.

5. ^{^}

   Male and female are therefore not defined by the presence of an Y chromosome. Many species (like birds) use something completely different. Many other species don't use sex chromosomes at all, and differentiate in males/females based on environmental clues like temperature.

6. ^{^}

   Quoted for posterity: "The female Sacculina larva finds a crab and walks on it until she finds a joint. She then molts into a form called a kentrogon, which then injects her soft body into the crab while her shell falls off. The Sacculina grows in the crab, emerging as a sac [...] on the underside of the crab's rear thorax, where the crab's eggs would be incubated. Parasitic Sacculina destroy a crab's genitalia, rendering the crab permanently infertile. [...] The male Sacculina 'larva' looks for a female Sacculina on the underside of a crab. He then implants his cells into a pocket in the female's body called the "testis", where the male cells then produce spermatozoa to fertilize eggs. When a female Sacculina is implanted in a male crab, it interferes with the crab's hormonal balance. This sterilizes it and changes the bodily layout of the crab to resemble that of a female crab by widening and flattening its abdomen, among other things. The female Sacculina then forces the crab's body to release hormones, causing it to act like a female crab, even to the point of performing female mating dances. [...] When the hatching larvae of Sacculina are ready to emerge from the brood pouch of female Sacculina, the crab [...] shoots them out in pulses, creating a large cloud of Sacculina larvae. The crab uses the familiar technique of stirring the water to aid in flow."

7. ^{^}

   I don't count volvoxes or slime molds as "fully-fledged", and I don't want to hear about mycorrhizal fungi.

8. ^{^}

   A long time ago, scientists thought they had discovered an exception: a strange rotifer called Bdelloidea, who doesn't seem to reproduce sexually at all. All the individuals ever observed were female, so either the males are hiding very well, or they are reproducing asexually. But it turned out that they used to be sexual, then decided that the future is female and went for parthenogenesis. If the theory is right, this comes with a big cost in evolvability, so we'll see how they handle global warming. Likewise, there are many strictly-asexual plants, but all of them have made the switch recently. It doesn't look like asexuality in plants in very stable over long periods.