Hacker News Reader

Comment by forgotpwtomain | Hacker News Reader

Interesting comment in the article's native discussion:

> Chris Jankowski • 2 hours ago

> The author says that sex "imposes a huge cost on a species, and that cost is called “males.”" Well, if the males are such a cost to a species, then why nature does not respond by heavily skewing the sex ratio of the offspring towards females. Males produce sperm which is cheap compared with the cost of producing eggs. A male could fertilize eggs of many - tens or hundreds of females. So the ratio could be 1 to 10 or even 1 to 100. But I do not know of a species that would do this.

> Some marsupials - possums and kangaroos create some skew, but is is slight and in response to environmental conditions or is more sequencing oriented depending on the age of the breeding female.

View on HN

The answer to the first question is Fisher's Principle [0]:

Suppose male births are less common than female.

A newborn male then has better mating prospects than a newborn female, and therefore can expect to have more offspring.

Therefore parents genetically disposed to produce males tend to have more than average numbers of grandchildren born to them.

Therefore the genes for male-producing tendencies spread, and male births become more common.

As the 1:1 sex ratio is approached, the advantage associated with producing males dies away.

The same reasoning holds if females are substituted for males throughout. Therefore 1:1 is the equilibrium ratio.

[0] https://en.wikipedia.org/wiki/Fisher%27s_principle

Thanks, that's very interesting.

The only solution would be to ignore the 1:1 ratio and randomly kill some infant every once in a while to bring the ratio close to 2:1.

Sorry? Can you go first? Please, show us the way!

That's not really an answer though. Females could choose to reproduce asexually. Even some large animals can still do it, like sharks. The mystery is why they don't do it all the time. What's the genetic benefit of making daughters that need males to reproduce, instead of making daughters that can clone themselves? Handwaving about "introducing variation via sex" doesn't help, because bacteria are the fastest evolving critters around and they reproduce asexually.

Bacteria reproduce asexually, yes, but they have developed an even more effective means of introducing variation than sex: horizontal gene transfer.

That sex introduces lots of variation is undisputed, however, and the mechanism of sexual selection accelerates the process over that of the random drift in asexually reproducing species.

horizontal gene transfer is basically what sex is, unless you do it standing up, then I guess it'd be vertical gene transfer.

You're taking it back a step to asking "why sexual reproduction"?

The original question is... assuming that we have mandatory sexual reproduction, why do we have a 1 to 1 ratio?

Because upper vertebrates normally can't exert influence on the sex of the offspring before birth. Your DNA will have an X chromosom at least. In principle [1], you can have either a second X or an Y. 50% probabilities. 1:1.

Only reptile, amphibious, fishes and it seems that very primitive mammals also can do this. Our system to get rid of the excedent is to make males sillier, much more prone to take life-threatening risks than females.

1 [Those aren't our only options, we have the so called supermales (one X, two Y), and superfemales (XXX) but they are not common and some combinations ("1.5 X") are sterile]

> Because upper vertebrates normally can't exert influence on the sex of the offspring before birth. Your DNA will have an X chromosom at least. In principle [1], you can have either a second X or an Y. 50% probabilities. 1:1.

That's not a sufficient explanation, as there could be a different ratio during the sperm cell generation process or selection during insemination or from the egg cells themselves. Y and X sperm cells could easily be handled differently at any point of the whole process.

I could have sworn i have read about a species of mice that do not have Y at all, but still produce males.

Humans have a ~1:1.05 F:M ratio so evolution really can alter the ratio. (Though this ratio varies by population.) https://en.wikipedia.org/wiki/Human_sex_ratio

Bacterial genomes are much, much simpler than animal genomes. They reproduce very quickly, and even if many bacteria suffer from a problematic (or fatal) mutation, others will be OK, and some may even have adaptations to changing conditions.

Animals have much more complex genomes. They are more at risk from mutations, and they cannot reproduce as quickly. There is more incentive to "get it right", if you will. With asexual reproduction, selection occurs at the level of the genome, meaning you are as good as your "worst" gene. You can't get rid of harmful mutations, and it seems like (complex) species that get stuck in asexual reproduction go extinct.

With sexual reproduction, selection can occur at the level of the gene, allowing beneficial mutations or recombinations to spread through a population AND correcting (many) harmful mutations.

The dropout regularization method for neural networks was sort of inspired by this observation, i.e. that recombination (combining genes of more than one individual) makes the resulting structures more robust [1]. The idea is basically that it avoids overfitting to the evolutionary context. This has two effects: the fitness cannot change very quickly in response to single mutations, and, conversely, it avoids that genes co-adapt such that the entire genome becomes inflexible at exploring alternatives.

To achieve the same without sex, mutations would would have both break up the co-adapted genes without breaking their functions and also find a good alternative mutation. With sex, it does not matter if co-adapted genes are broken up, because they are forced to evolve to deal with it. (This is kind of similar to the concept of learning to learn in neural network research.) Also, if a parent gene is unfit, it is possible that the child can escape this mutation by receiving that part of the genome from the other parent instead by a 50% chance.

[1] https://arxiv.org/abs/1207.0580

Because it leads to overfitting. Sexual reproduction is just more robust to the environment changes.

Snails are all hermaphrodites. No complicated ratios, sexual reproduction, equality and interesting sex.

Bacterial have something far better than sex - horizontal gene transfer [1]. They can share genes with any organism.

1. https://en.m.wikipedia.org/wiki/Horizontal_gene_transfer

Corals are an example of animals that can do both. They reproduce asexually through fission, which keeps clones in local similar conditions where they can thrive.

While sexual reproduction occurs through spawning, where genetically diverse offspring travel far and wide on currents to possibly different habitats and niches.

The idea here is that asexual reproduction is a benefit to the species during relatively stable conditions, while sexual reproduction works best to populate new areas or recover from population disasters.

It'd be really interesting to get to see this happen on a wide scale sometime soon. It'd be better not to have to, as well...

Clones don't work well with the evolution. All of the individuals are exactly the same, so they are very prone to a whole population being wiped out at once. The key of evolution are mutations that promote survival of the best adapted. And the bacterias are different from what I understand because they reproduce at extremely fast rates. The more complex creatures could do the same but it would take centuries to do the same number of cycles and they (as a population) just wouldn't survive that long. Mixing genes is thus a better way for bigger creatures.

So how does the model account for species like ants and bees?

It works just fine in solitary bees. Eusocial insects like the best-known species of ants and bees have a totally different system [0]. Their colonies operate as "super-organisms" that tend to be selected for/against as a whole, rather than individually. The vast majority of them are sterile workers anyway.

[0] https://en.wikipedia.org/wiki/Haplodiploidy

Thanks! Next questoin: do we have any idea whether haplodiploidy is a result or driving force for eusocial behaviour?

Probably a driving force, as Wikipedia lists quite a few haplodiploid Eusocial species that independently evolved Eusociality.

(The interplay of Eusociality and genetics is one of the most interesting wiki dives I ever did, highly recommended it)

Humans have a ~1:1.05 F:M ratio at birth. So evolution really can and has altered the ratio. https://en.wikipedia.org/wiki/Human_sex_ratio

1:1 is really only a theoretical balance point. 1:1.05 is better balanced because boys and men have a higher rates of death. (every year except right before puberty ~10 and 11 oddly enough.) https://www.ssa.gov/oact/STATS/table4c6.html

PS: Men are around 3x as likely to die at 20 than woman. And the oldest woman lived 6 years longer than the oldest man. The 10 oldest living people are all women. https://en.wikipedia.org/wiki/Oldest_people

So the ratio is evaluated at time of reproductive maturity, rather than time of birth. The selective pressure to keep the ratio at 1:1 still exists.

To me, this means Fisher's principle belies the notion that males are a cost borne by a species.

I wonder how this might bear out for crocodile and turtle species, whose sex if determined in the egg due to nest temperature?

Changes to average beach temperatures can skew sex ratios in hatchlings. (Will find for refs but mobile right now)

Could this mechanism of rebalance work for turtles, even though the mortality of hatchlings and juveniles might be most important to the survival to breeding age, and subsequent numbers of offspring?

In many species, even though the ratio is mostly equal, most males don't reproduce. A single male becomes dominant and fathers most/all of the offspring. So it achieves the same effect as he's suggesting, a 1/n ratio of reproducing members of the species, but also adds test(s) to select out the weaker males.

I believe this would be advantageous in species with longer gestational periods where you'd want to put as much of the fitness test up front and not wait until the babies are born to eliminate the weaker offspring.

Based on genome analysis, that's been true about human males also. http://www.denisdutton.com/baumeister.htm

What an incredibly interesting analysis.

I reread that essay every year or two. It explains so much.

> Of the small minority of people who do like math, there are probably more men than women. Research by Jacquelynne Eccles has repeatedly concluded that the shortage of females in math and science reflects motivation more than ability. And by the same logic, I suspect most men could learn to change diapers and vacuum under the sofa perfectly well too, and if men don’t do those things, it’s because they don’t want to or don’t like to, not because they are constitutionally unable (much as they may occasionally pretend otherwise!).

Its these kinds of lines that make it very hard to take this analysis seriously.

How so? It's more tongue in cheek than the other parts of the analysis, but it's pretty straightforward in any case.

> If a man said “not today, I have a headache,” he might miss his only chance.

Ouch

For one thing, sexuality enables sexual dimorphism. We can compete and evolve as a symbiotic pair of subspecies with the advantages and disadvantages of both.

> imposes a huge cost on a species, and that cost is called “males.”

Incidentally, the cost of males is only a 50% overhead when the males contribute nothing to raising the offspring. In some species, the energetic/resource cost is dominated by caring/feeding/raising long after birth, which males may contribute to as much as females.

When male/female determination depends on which chromosome you inherit from a parent, it is hard for a mechanism to arise for the chance to be anything other than 50/50.

A lot of mammal species eliminate excess males using other methods - mainly bloody fights to the death, which also tends to result in the remaining males being the strongest ones.

Except that it's not 50/50, at least in the case of humans. With humans, it's about 1.07/1, in favor of males.

The mechanism is much more complicated than a 50/50 chance of inheriting some chromosome.

The first answer is that it appears there is no way for the embryo to have its sex selected non-randomly. It's like throwing a coin, you end up with 50% chances for each sex.

The second answer is that there are indeed cases the environment can influence the chances for either male or female in several species. For example, mostly for reptiles, https://en.wikipedia.org/wiki/Temperature-dependent_sex_dete...

Reply on news.ycombinator.com