More Recent Comments

Wednesday, March 12, 2014

Everything you thought you knew about sex is probably wrong

The evolution of sexual reproduction is one of the great mysteries of biology. I've been teaching this to undergraduates for several decades but it seems that most undergraduates don't get the message. Most of them think that sex has been explained and the answer is that sexual reproduction generates diversity.

I usually give them some reviews to read and, invariably, the experts who write the reviews will say that sex is a great unsolved mystery. If the experts think that this is a mystery then why do so many people think they have the answer?

Here's what Douglas Futuyma says in the 2nd edition of Evolution (p. 340).
The recency of most parthenogenic lineages suggests that sex reduces the risk of extinction. If this were the reason for its prevalence, sex might be one of the few characteristics of organisms that has evolved by group selection. But recombination and sex also have serious disadvantages. One is that recombination destroys adaptive combinations of genes.... In general, asexual reproduction preserves adaptive combinations of genes, whereas sexual reproduction breaks them down and reduces linkage disequilibrium between them. An allele that promotes recombination, perhaps by increasing the rate of crossing over, may decline in frequency if it is associated with gene combinations that reduce fitness.

Sexual reproduction has a second disadvantage that is great enough to make its existence one of the most difficult puzzles in biology. This disadvantage is the cost of sex. Imagine two genotypes of females with equal fecundity, one sexual and one asexual. In many sexual species, only half of all offspring are female. However, all the offspring of an asexual female are female (because they inherit their mother’s sex-determining genes). If sexual and asexual females have the same fecundity, then a sexual female will have only half as many grandchildren as an asexual female. Therefore, the rate of increase of an asexual genotype is approximately twice as great as that every sexual genotype (all else being equal), so an asexual mutant allele would very rapidly be fixed if it occurred in a sexual population.
Note the phrase, "... one of the most difficult puzzles in biology," How many readers think that sex and/or recombination is a great mystery? Not many, I'll bet. I'm guessing that most of you are quite satisfied with the answer that sex confers evolutionary advantage because it increases diversity (combinations of alleles).

You all should read what Joe Felsenstein is saying on What did Joe Felsenstein say about sex?. You don't have to agree with him, or with Futuyma, but if you are a teacher you should make sure that students understand the controversy. The mystery of sex has not been solved.

I like the review by Sarah Otto in The American naturalist (Otto, 2009). Here's a little excerpt to make you think ...
... sexual reproduction breaks apart favorable combinations of genes built by past selection. To hammer this point home, consider an analogy. Imagine entering a poker hall after a game has been played. If you were to offer the winners (holding, say, a 3♣, 4♣, 5♣, 6♣, 7♣ straight at one table, a three‐queen hand Q♦, Q♥, Q♠, 2♥, 8♦ at another, etc.) the opportunity to keep their hands or to shuffle their cards with those of another, everybody would hold his or her cards. Winning hands—those that have “survived” previous rounds—have cards that work well with one another. Shuffling these cards together produces descendant hands with no guarantee of success (creating, e.g., a lousy hand of 3♣, 4♣, Q♠, 2♥, 8♦). In all card games of interest, it is not enough to know the suit and number of each card in isolation; rather, the interactions among cards are what determine whether the card is in a winning hand or a losing hand. Similarly, genes do not work in isolation; the interactions among an individual’s genes in the context of its environment are what determines whether that individual will successfully survive to reproduce or fail. Sexually mixing one's genes with those of another destroys the network of alleles that worked well in the parent, creating a new network that may or may not function....

Given the costs of sex and the widespread potential for asexual reproduction, why do so many species reproduce sexually? This question has been called the paradox of sex. Most biologists would answer that sex and recombination have evolved because they generate variation needed by selection. Indeed, this is one of the oldest explanations for sex, attributed to August Weismann (1889, p. 279):
Sexual reproduction can also increase the differences between individuals …. Such differences afford the material by means of which natural selection is able to increase or weaken each character according to the needs of the species.
That sex evolved to generate variation may very well be correct, but there are two holes in the argument that make it a much less obvious answer than it would at first seem. Because these holes are not widely appreciated, even among evolutionary biologists, they deserve some attention.
She goes on to discuss "I. Sex Need Not Increase Variation" and "II. Generating Variation through Sex Often Reduces Fitness."

Isn't it interesting that an expert on the subject would use a title like "The evolutionary enigma of sex"?

Otto, S.P. (2009) The evolutionary enigma of sex. The American Naturalist 174, S1-S14. [doi: 10.1086/599084[


heleen said...

Another take altogether might be found here:
Glansdorff, N . Xu, Y, ; Labedan, B. 2009. The Conflict Between Horizontal Gene Transfer and the Safeguard of Identity: Origin of Meiotic Sexuality. JOURNAL OF MOLECULAR EVOLUTION 69: 470-480
From abstract:
“ Among higher eukaryotes HGT appears of limited scope except in asexual organisms. We suggest that meiotic sexuality (a hallmark of eukaryotes) emerged in the genetically redundant and protoeukaryotic LUCA as a molecular identity check providing a defence mechanism against the deleterious effects of HGT.”

NickM said...

The points in this post assume that there are one or a few optimal genotypes. If there is diversifying selection, for example, on proteins that either avoid or recognize proteins used on the surfaces of pathogens, the whole argument changes. Since there have probably been pathogens basically forever...

Joe Felsenstein said...

The Otto review and the Futuyma comments are very good. There is still a major enigma. But what there is not is a Giant Crisis. People have often claimed that, but usually when they have done a piece of work that they think explains the evolution of "sex" -- declaring a Giant Crisis is very self-serving in that case. The uncertainties of how to explain "sex" actually does not impede other work on evolution.

I put quotes around "sex" because it is important to realize that we aren't discussing why male and female gametes are produced by different individuals (hint: they aren't always), or why those individuals look different. We are actually trying to explain the evolution of recombination in outcrossing organisms, who might be hermaphrodites.

AllanMiller said...

I'm afraid I'm one of those you criticise, though I do see why it is perceived as problematic. Futuyma's treatment exposes a fundamental issue: sex is seen as most problematic in dioecious species. Yet (as is well known) that cost is absent in isogamous species. It's not a cost of meiosis, nor of sex per se. Isogamy-to-anisogamy is not a binary flip, but almost certainly an incremental transition. At what point does the cost go from non-existent to twofold? After a 1% differential? 10%? 20%? Meantime, the genetic transaction remains symmetrical, and genes invest in both strategies alternately. Anisogamy is also associated with multicellularity (being difficult to achieve in single cells). This affects the dynamic, since the 'interests' of somatic cells are suspended in favour of a gametic future.

There are, I am well aware, many issues, but it seems that, implicitly at least, it is proposed that (on present knowledge) the eukaryotic biota should consist of two classes of organism:

1) Happily isogamous sexuals.
2) The asexual descendants of multicellular anisogamous females

and we don't know why it doesn't!

whimple said...

I used to teach graduate students about the evolutionary implications of sex (although I phrased it as an unsolved challenge rather than as a "controversy"). We discussed the pros and cons of the Red Queen hypothesis, and molecular genetic analysis of rotifers as the exception to the generalized observation of the relative recency of asexual lineages. The topic was also a great segue into what it means to apply the term 'species' to organisms that reproduce asexually.

whimple said...

"We are actually trying to explain the evolution of recombination in outcrossing organisms, who might be hermaphrodites."

In my field this is explained by recombination being the molecular mechanism by which parental chromosome homologs are identified prior to segregation in meiosis. Specifically, recombination precedes and is required for the pairing of chromosomal homologs, rather than the more commonly held view that homologs pair up (by some unspecified means) and only then undergo recombination.

AllanMiller said...

Either way the retention of at least one crossover seems necessary for correct disjunction. This pushes the problem back slightly - why a 2 step meiosis with bivalents separated in the 1st - but it does seem plausible that recombination's wider effects could be an incidental consequence of the cytological constraints.

Georgi Marinov said...

I am absolutely not an expert in the field, and I have not read the literature on the subject that has been built up over the years (I was one of those who has been taught the standard explanation too). I do plan to correct that some day, whenever that is, but for now, I speak as layman.

Anyway, to me the key observation has to be the phylogenetic distribution of sex on the grandest scale - what we usually refer to as sex is only found in eukaryotes. Prokaryotes do exchange genetic material, and they seem to do so quite freely as evident by the fact that you can't even talk about "the genome of species X" in the traditional sense of the term. But it's a very different thing from what goes on In eukaryotes. So it would seem to me that it has something to do with the barriers to that free exchange of genetic information that exist in eukaryotes due to the organization of the genome and the cell. The observation that the asexual lineages are all young seems to support that (though how one explains the Bdelloids is another mystery). So we're back to the "you need recombination or you go extinct"/group selection explanation but there isn't anything else I can think of that make sense

AllanMiller said...

[...] we're back to the "you need recombination or you go extinct [...]

I think reduction is at least as promising an avenue as recombination for 'refreshing' a lineage. Gene conversion (recombinational, ironically) tends to increase homozygosity over time, favouring periodic haploidy (provided the recessive is not expressed in the haploid state).

Joe Felsenstein said...

A naïve calculation suggests that if male gametes are x times as big as female gametes, Maynard Smith's "cost of meiosis" should mean that fitness of the species is reduced from 1 to (1/2)(1+x). Thus a 2/(1+x)-fold reduction. (Note that a 1-fold reduction is, in that terminology, no reduction at all).

Robert Byers said...

Is it possible that the origin of sex is a mystery because it confounds normal speculation of evolutionary actions. I mean both are untestable but the regular stuff adds up in a species of reasoning. The sex stuff doesn't. Thats the only difference.
In short there is no biological scientific evidence for evolution. its just lines of reasoning from raw data.

John Vreeland said...

Sarah P. Otto's analogy is useful for illustrating her point but as an argument against sexual reproduction it is a straw man. A straight flush will win perhaps 99.99% of all games in poker but in real organisms such combinations are unlikely to exist. At best
3♣, 4♣, 5♣, 6♣, 7♣ might have only a slight advantage over 3♣, 4♣, 5♣, 6♣, Q♠, winning say 51% in the competition for reproduction. This would leave plenty of room for other forces to come into play.

Joe Felsenstein said...

People have suggested recombination is necessary for chromosome segregation, certainly. But why bother to segregate at all? Why not reproduce clonally and just do mitosis. American dandelions (Taraxacum officinale) do just fine without meiosis at all.

AllanMiller said...

Same with British dandelions (Taraxacum officinale)! Although I believe they do have meiosis, but have abandoned the second step.

But no, I understand the point you are making.

As I suggested in the above post, the 'why not do mitosis' question seems to be asked in relation to the costs, primarily of males. If isogamy avoids the major cost (males), then any and all of the proposed counterbalances would suffice to maintain it (granted the lesser costs of recombination). People are trying to repay a twofold cost, but even if they don't get all the way there, isogamous organisms (lacking that cost) reap that benefit. So the fundamental existence of sex seems less mysterious. The mystery is one for zoologists and botanists: it's so overwhelmingly beneficial, sexual types can even afford males!

Of course the 'why not do mitosis' question remains for isogamy.

But I'm not clear on why we should expect them to. If the ancestral state was haploid, and syngamy occurred, the haploids did not abandon their former mode of life, but underwent a temporary union. They may derive benefit from deferring the severance of that union for a few mitoses, or even (eg aphids) a few somatic generations, but eventually return. Certain types have 'deferred' segregation eternally, and we seem to be asking why everything hasn't.

AllanMiller said...

Given that anisogamy seems primarily restricted to multicellular organisms, the proportion of overall costs provided by gametes is unlikely to rise above the background cost of a body!

Corneel said...

what we usually refer to as sex is only found in eukaryotes

That is because sexual reproduction is most likely the ancestral state of all extant eukaryotes. That is, all modern asexual eukaryotes have reverted back to asexuality at some point in their evolution.
I believe it is also largely unproven that asexual lineages have higher extinction rates. There are some alternative explanations to explain why most asexual lineages are young.

Alex SL said...

It depends on what you mean when you say "sex is a mystery". How exactly it evolved? Maybe. But what it is good for? Yes, I think that is clear. It lets you get good combinations of genes faster.

The two arguments above work somewhat similar to Joe Felsenstein's argument from the previous thread: They assume that we already have the winning hand and argue from there.

But that's unrealistic. Consider this scenario, which is perhaps a better approximation of what likely happens in reality:

1. Assume two teams of, say, ten players each.

2. Everybody starts with random cards.

3. In team A, people are allowed to discard one card every fifth turn and pick up another one (mutation). Then each of their hands is "copied" ten times (offspring), and of all the 100 copies in the team the strongest ten are picked up by the players for the next round (selection).

4. In population B, people are also allowed to discard one card every fifth turn and pick up another one. But they are also required to form groups of two, and pool and shuffle their cards (sex). Then each of their hands is "copied" ten times, and of all the 100 copies in the population the strongest ten are picked up by the players for the next round.

After, say, twenty turns, which population of players is likely to contain the strongest hand? Which of them is likely to have the stronger hands on average?

The trick is to get the strong hands quickly, before somebody else does. Again, assuming that you start with the winning hand is cheating.

Yes, once you have the winning hand, some of your children will lose it through recombination (at least as long as the population isn't overwhelmingly there yet), but, as cynical as that sounds, throw out enough children and it doesn't matter. Say only a quarter of a shrub's five hundred thousand seeds will carry the kick-ass metabolic pathway, fine, but the asexual sister species is still thousands of years away from even accidentally getting that pathway in the first place. Before it gets it, it is toast, as empirically demonstrated by the young age of most asexual lineages.

Joe Felsenstein said...

To see which theory of the evolution of recombination is at work here, he fitness scheme here needs to be made clear. I am assuming it is the usual order of hands in, say poker. It is extremely strong selection, favoring the highest hand. That hand, once it appears, gets given to all players on the team.

It might be interesting to make this game have only two cards per player, with only two kinds of cards, 3s and 4s. So there are only two kinds of hands: 33, 34, and 44. Also, have there be only two players per team. Draws of new cards draw either 3 or 4 with equal probabllity.

That would start to approach mathematical analyzability. Now there are only 6 states of a team: (33,33), (34, 33), (34, 34), (33,44), (34, 44), and (44, 44). How quickly does each strategy get a team to hold all (44,44)? In the long run how much time is spent in each state if we continue the process/

It is not entirely obvious even with this simplification of the game.

I'll stop here, I am busy and have other work to do. Maybe Alex SL can take it the next step.

Joe Felsenstein said...

Winning 51% is a fairly strong selective advantage in nature, and the genotype will take over in a few hundred generations.

Donald Forsdyke said...

At its most fundamental level, sex is recombination. The conjugation required for the meeting of paternal and maternal human gametes is an elaborate preliminary to the final meiotic meeting of parental genomes in the gonads of their offspring, where recombination occurs. The early microscopists who witnessed this final meeting described it as a “conjugation of the chromosomes” that was necessary for a rejuvenating “interchange of substances.” Thus in 1901 T. J. Montgomery wrote:

“The conjugation of the chromosomes in the synapsis stage may be considered the final step in the process of conjugation of the [parental] germ cells. It is a process that effects the rejuvenation of the chromosomes; such rejuvenation could not be produced unless chromosomes of different parentage joined together, and there would be no apparent reason for chromosomes of like parentage to unite.”

In essence, this means that sex is about error correction. A big topic, but Montgomery is a good starting point (see "A study of the chromosomes of the germ cells of metazoa," Trans. Am. Phil. Soc. 20 (1901) 154–236.)

DAK said...

Is it possible that the conditions which drove the evolution of sex are no longer present? If so, it'd mean that looking for present day benefits might be a case of barking up the wrong tree.

AllanMiller said...

With respect, defining sex thus seems to be begging the question. The 0-45 (female) or 0-70+ (male) years of diploid co-existence that intervene between syngamy and the occurrence of the recombinational event, and reduction itself, can hardly be just incidental to the transaction? Things must have moved on a bit since 1901!

Michael Le Page said...

Bill Martin and Nick Lane have proposed that the origin of sex is inextricably linked with the acquisition of mitochondria that (in their view) led to the evolution of eukaryotes:

Tom Mueller said...

@ Joe re: "why bother to segregate at all"?

To repeat some comments in an earlier post and extend it to chromosome segregation:

I can imagine zygospores as some evolutionary quiescent exaptation that eventually gave rise to sexual reproduction… the exaptation possibly being DNA repair.

In other words, recombination was an evolutionary add-on… in teleological terms, some fortuitous after-thought as it were.

To continue some earlier musings: Correct me if I am wrong – but reproduction exclusively by sex is by far the exception to the rule meaning in my mind most organisms manage to enjoy the both benefits of asexual reproduction together with the benefits of sexual reproduction.

Most organisms employ asexual reproduction in constant environments as their default setting and sexual reproduction kicks in under scenarios of environmental stress.

In other words; if environment changes, not only may offspring be poorly adapted but ALL offspring will be poorly adapted because they are identical.

Zygospores are quiescent resting bodies awaiting environmental conditions to return to "normal" ...

or alternatively...

...allow recombination resulting in novel variable offspring, some of which can to adapt to the “new normal”.

So why is sex beneficial? Parent’s chances of producing offspring that can survive competition are greatly enhanced, remembering that each generation produces far more progeny than can possibly survive.

Would the above summary pass muster for presentation to a High School class? I do not want to number among those Biology teachers who "get sex wrong".

One nagging question remains? Why are dioecious male to female ratios ubiquitously and persistently remain 50/50 when in sexual reproductive mode and skewed to favor females in asexual reproduction mode except in exceptional circumstances?

I would appreciate a ready quick answer to provide to my own students.

Thanx in advance…

Tom Mueller said...

"sex is only found in ekaryotes"?

Oh dear - I hope I do not need to undo inadvertent damage commited in my classroom.

In my classroom I teach that sex is by definition some reshuffling of genetic material to generate new allelic combinations.

In humans sex cannot be separated from reproduction.

In bacteria sex is separated from reproduction.

In other words, sex is ubiquitous!

AllanMiller said...

Tom - you'll find a lot of support for this view, including our illustrious host! Me, I think it has the potential to mislead.

Tom Mueller said...

@ Allan Miller

"sex is ubiquitous" has "the potential to mislead"?

how so?

Tom Mueller said...

Georgi Marinov re: "you need recombination or you go extinct"

I agree with your synopsis

The accumulation of deleterious alleles places SMALL asexually reproducing populations at a disadvantage compared to sexually reproducing populations.'s_ratchet

Sex must be beneficial for small populations subject to Genetic Drift allowing them to escape Muller’s ratchet?

i.e. In a small population subject to genetic drift - the burden of accumulating deleterious mutations may eventually cause some populations to go extinct (bad luck of the draw due to drift, as it were).

Sexual reproduction short-cuts Muller’s ratchet. E.g. two individuals each with one copy of some deleterious mutation will produce offspring that are free of either deleterious mutation 25% of the time.

Georgi Marinov said...

Tom MuellerThursday, March 13, 2014 10:36:00 AM
"sex is only found in ekaryotes"?

Oh dear - I hope I do not need to undo inadvertent damage commited in my classroom.

To me, "sex", "recombination", and "exchange of genetic material" are three different things. I maybe wrong. But within my understanding of the terms, what I said is correct.

AllanMiller said...


I think people can be over-committed to seeing what is perhaps eukaryotic sex's most significant feature - recombination - as its raison d'etre. If you define sex as recombination, that naturally follows!

Lumping superficially similar bacterial mechanisms in has some justification, because there is some continuity of genetic underlay, but they are more different than similar. Sex in the eukaryotic sense involves syngamy, diploidy and reduction; recombination is a cytological necessity at the end of that process, invoked by generating DSBs and allowing ancient DSB repair pathways to link the chromosomes. Similar genes are used to integrate 'foreign' DNA in prokaryotes, but the rationale may well be very different.

It is possible but not necessarily true that the reason eukaryotes perform syngamy and reduction is to get close enough to perform recombination. Or that they perform reduction and syngamy in order to propagate recombinant DNA. Either way, it's a lot to evolve all at once!

Bacterial mechanisms, meanwhile, seem more akin to infection. Conjugation is closest to eukaryotic sex, but mainly due to a superficial resemblance to multicellular conjugation. Transduction and transformation? Come on!

Eukaryotic sex is just 'different'. No prokaryote swaps half its genome with another. Nor do they spend a significant amount of time in the diploid state before separating and serially swapping partners in haploid/diploid alternation.

Tom Mueller said...

@ Georgi Marinov

Now I am really confused... how can recombination occur without exchange of genetic material? ... what do you call "sex"?

Georgi Marinov said...

Does recombination in chloroplasts occur because of sex?

El PaleoFreak said...

"which population of players is likely to contain the strongest hand?"

OK, there is a winner population. But why that population started to play that complicated strategy in the first place? How could it "know" about the future? ;o)

Tom Mueller said...


I take issue with your primary premise: “eukaryotic sex's most significant feature - recombination - as its "raison d'etre”

I always understood eukaryotic sex's raison d'être to be DNA repair with needed back-up from another partner's copy of genetic information.… recombination was a fortuitous add-on.

Let’s say the evolutionary benefits of recombination are questionable (at best neutral if not deleterious) the overwhelming added benefits of DNA repair would win the day.

@Georgi re: chloroplasts... ditto. it is not about recombination, it is about DNA repair.

ITMT - am I correct to understand the genomes of chlorplasts and mitochondria are not static?

Forgive my naiveté

Georgi Marinov said...

There is some controversy, but mammalian mitochondria are generally thought to not recombine, at least that's my understanding.

But in other groups they do. The diversity of mitochondrial genome organization is extreme across the eukaryotes as a whole.

Tom Mueller said...

@ Larry & @ Joe

Gentlemen - I think I may finally have discerned your intent.

Correct me if I am wrong - but are misconceptions and fuzzy thinking regarding sex due to the teleological premises presumed when discussing the "reason for sex"?

Are you both attempting to purge us inadvertent teleological fuzzy thinking?

If so – I for one, thank you for focusing my thoughts on this vexing question.

best regards

AllanMiller said...


I take issue with your primary premise: “eukaryotic sex's most significant feature - recombination - as its "raison d'etre”

You may not be among their number, but most theories of sex are theories of recombination. Even the repair hypothesis can be classed as a recombinational theory (Donald Forsdyke, elsewhere in the thread, appears to take that stance).

I doubt the fundamentality of the repair hypothesis for similar reasons to recombination, though the objection depends on how the diploid arose - syngamy or endomitosis.

Somehow, organisms thought to be in 'need' of repair can't maintain an accurate haploid chromosome across the G1 phase of the cell cycle (the only part where it doesn't have an available sister), so they keep an extra copy. Hmmm ...

Endomitosis is the obvious route - if you want an accurate backup, use a recently replicated copy, not some distant relative! But it rather assumes some incompetence on the part of the organism, has nothing to do with sex, and is individually costly. That is, the best route to repair is the worst route to cyclic sex.

I'd agree that the many consequences of sex reinforce, though. When you have a homologue in G1, use it - why wouldn't you? But it would be a weak driver to cyclic syngamy, and susceptible to asexual invasion by a permanent diploid if that was all it was 'for'.

S Johnson said...

Most discussions of sex in popular literature, as near as I can tell, seem to focus on animals, particularly on sex ratios as demonstration of gene selection. But the contrasts between prokaryotes and unicellular eukaryotes, and between fungi, plants and animals are suggestive.

The greater biomass of bacteria as opposed to unicellular eukaryotes does not suggest that the eukaryotic chromosome is overwhelmingly advantageous on a microscopic scale. Further, the relative paucity of sexual reproduction in unicellular eukaryotes other than yests suggest that sexual reproduction on a microscopic scale is relatively disadvantageous. This of course is another way of saying that the evolution of sex is not a straightforward process of natural selection for advantage.

Continuing with the yeasts, I gather the haploid varieties reproduce asexually and the diploid varieties reproduce sexually. Or I suppose you could rephrase the problem of sex as the problem of diploidy? It seems many varieties of yeasts (the unicellular fungi)
switch rather easily between haploid and diploid forms.

Again this confirms that sexual reproduction does not have an overwhelming reproductive advantage. I noted that one form of asexual reproduction in yeast, budding produces pseudohyphae, which can be construed as an intermediate between unicellularity and multicellularity. In the multicellular fungi, some species have no observable organs dedicated to sexual reproduction (the imperfecti.) Sexual reproduction when it occurs has the cytoplasm of two haploid cells fusing, then at some point (which I gather can be well separated in time?) the nuclei fuse, then meiosis occurs. Notably, not all fungi has just two sexes, plus they are commonly isogamous, which is why mycologists talk about mating types instead. Some have four, others more. The interesting thing about multicellular fungi to me is that they don't really have a body plan, existing in bulk as the opportunistic expansion of hyphae. In the life cycle of a fungus, part of the time it is haploid, another diploid. (I have no idea how the classical arguments in favor of equal sex ratios apply to tetrapolar fungi.)

This is called alternation of generations, which is particularly associated with plants. (Fungi get no respect.) Plants do not have the same kind of tightly defined body size and shape, not as a group, that animals do? What is the shape of a grass plant? What is the size of an adult oak tree? The plant genome does not specify the adult form in the same way an animal genome does. The reason I think is that mobility, the basic animal survival strategy, puts severe constraints on animal morphology. (Sessile sponges depend on invisible to us movement to circulate water.) Still, although a sea weed is pretty amorphous when compared to an ant, it is far more defined than a fungus. In a sense, all plants are separate from their environment in a way which fungi, which riddle theirs to consume their food with extracellular enzymes, are not.

At any rate, in the plants, the unicellular green algae exhibit alternation of generations, that is, haploid and diploid phases. Notably, unicellular green algae can form intermediates with multicellularity, such as colonial or filamentous forms. (Red and brown algae are multicellular.) It is also notable that the basal, earlier evolved land plants show a marked equality in size of diploid gametophyte and haploid sporophyte.
Over time, as plants evolved, the diploid gametophyte phase became a dependent organ on the sporophyte, much, much smaller. To me, this suggests that for the everyday business of growing, haploidy is favorable to mitosis.

Animals, with the most tightly constrained body forms, do not alternate generations.

S Johnson said...

Sorry to be so long, but I can't see how to shorten it without getting even more cryptic. Here's the rest:

Comparing organisms like this may ignore gene selection but I still can't help but feel that if you are so bold, then we can conclude that sexual reproduction is pretty much the same as diploidy. And that diploidy is not particularly advantageous for reproduction of cell. The green algae have not outcompeted the cyanobacteria. Even in plants, if anything the diploid phase does not appear to have any advantages, rather the contrary. But diploidy always associated with multicellularity. I would suggest that diploidy/sexual reproduction is an inherited feature that cannot be easily dispensed with. The repeated re-evolution of asexual reproduction in multicellular organisms, particularly under harsh natural selection, confirms that sexual reproduction has some strong disadvantages. But evolution is not an engineer. Doing away with diploidy but retaining multicellularity is a radical redesign that is so far apparently beyond nature's capacity. (Dollo's law comes to mind.)

Most species go extinct, so it is not entirely certain what it means to say that asexual species "fail." Many species that do appear to survive are certainly not the same genetically. Is it really so certain that a modern horseshoe crab could successfully breed with the ancestral form? And in what sense is the horseshoe crab a "surviving" species?

In diploid cells, the phenomenon of dominance can mask the expression of deleterious genes, or modify the expression of alleles by their combined effect. Diploidy provides a way of increasing information storage. Simultaneously, for essential genetic information, homozygy provides error correction, which is ever more important the more tightly constrained the phenotype is by mechanical or physicological necessitites. The strategy of multicellularity is fairly successful, achieving perhaps half the biomass.

As for the question of recombination, the object seems to be that it doesn't necessarily increase genetic diversity. Well if it increases genetic uniformity, perhaps it is helpful in forming intermediate stages of multicellularity. It seems likely to me that the eukaryotic chromosome emerged as a compromise between the costs of producing the novel organelle and the benefit of suppressing deleterious information. It appears to be an adaptation for gene regulation. As such it was probably orginally haploid. As these unicellular organism reproduced, similar cells were able to reap the advantages (when there were any) of intermediate stages to multicellularity, such as colonies and filaments.

But, all flesh is grass, and haploid eukaryotic chromosomes will still mutate. Thus inevitably in the course of time, the colonial or filamentous cells begin to diverge. The mechanisms that keep them cooperating begin to fail. (To be sure, the interesting question is how much advantage these give in a unicellular ecology?) Stressed as the benefits of cooperation are lost (whether it's a "simple" biofilm or something like Volvox?) the organisms' old machinery for transformation or conjugation is coopted for crossing over. The crossing over process increases the uniformity of the daughter haploid cells. However slowly the advantage of increased uniformity exerts selection pressure to optimize the crossing over behavior. Then at some point, the cells fail to separate. The ensuing diploid cell of course is even more uniform, plus it can store more information correctly, an exaptation for full multicellularity.

Alternatively, crossing over may just be an unavoidable event in the replication of diploid dells, one that isn't damaging enough to be excised by natural selection.

S Johnson said...

PS The first case seems to be less likely. If it were so, you should expect to see crossing over between the male and female chromosomes in fertilization. But the second case, even though it isn't very compatible with gene selection (wait, genes are hopping around on chromosomes and it doesn't make enough difference to be selected out? That's not very deterministic!) does have the merit of biochemical/genetic simplicity. That is, if you look at things like alternation of generation and the dominance of haploid sporophytes in plants, etc.

Larry Moran said...

Just for the record. Recombination is prevalent in all eukaryotic cells, even those that are dividing by mitosis. It's a common myth that recombination is restricted to meiosis but it's a myth.

Larry Moran said...

Joe Felsenstein says,

Winning 51% is a fairly strong selective advantage in nature, and the genotype will take over in a few hundred generations.

I think that what you mean to say is that if a certain new genotype has a 1% fitness advantage then it has about a 2% chance of taking over the entire population. Right? It also has a 98% chance of being lost before it becomes fixed.

Larry Moran said...

Joe Felsenstein says,

I'll stop here, I am busy and have other work to do. Maybe Alex SL can take it the next step.

Thank-you very much for spending as much time as you did. This is complicated stuff and I don't understand the math behind a lot of population genetics. That's why I nudged you to jump in. I apologize for the cheap trick of naming you in the title to one of my posts but I hope the result was informative for my students and for other Sandwalk readers.

Larry Moran said...

You start with both teams playing by the same rules. What is the immediate selective advantage for individuals that "evolve" the team B strategy?

I can see why the B strategy might pay off in the long run under some circumstances (the payoff is much exaggerated) but as far as I know natural selection or random genetic drift doesn't see into the future.

You're postulating that something like this occurred in population B: two cells fuse and then produce 20 offspring that each contain random mixtures of chromosomes from each of the parental cells. (Let's assume that there are ten or so chromosomes per cell.) This results in different combinations of alleles in each of the twenty offspring. (You don't need recombination to get this kind of mixing of alleles.)

What is the evolutionary explanation for why this arose in population B? Are you saying that it's because at some time in the distant future the B population will hit upon a combination if successful alleles faster than if it had just stuck with simple cell division?

That seems like a lot of effort for a distant and uncertain payoff.

Rosie Redfield said...

Sorry to jump in late - I was travelling and missed this post. Interested readers might want to look at my 2001 review Do Bacteria have Sex ( It lays out the evidence that recombination of chromosomal genes in bacteria occurs largely as unselected accidents of processes evolved for infectious transfer (conjugation and phage infection), nucleotide acquisition (natural competence) and DNA replication and repair (strand invasion and physical recombination).

DAK said...

I like the DNA replication and repair stuff you mention. It reminds me a little of the psuedo-diploid phenomenon observed in retroviruses that helps them deal with the heavy damage their genomes receive.

Sex could have evolved as part of a repair mechanism before efficient DNA repair mechanisms or high fidelity replication was achieved.

Joe Felsenstein said...

That's OK, you thereby gave me license to blather on endlessly.

AllanMiller said...

Sex could have evolved as part of a repair mechanism before efficient DNA repair mechanisms or high fidelity replication was achieved.

I think it's probably the other way round. The recombinational part of sex (which isn't the whole of sex btw) invokes repair pathways that are probably quite ancient.

Joe Felsenstein said...

Absolutely, but it has a far greater chance of fixing than a neutral mutant would. In a population of (say) 1,000,000 individuals, a mutant with a selective advantage of 0.01 will fix only 0.0198 of the time. But a neutral mutant would fix only 0.0000005 of the time, so the advantageous mutant is 40,000 times more likely to fix.

In the card hand example above, if both hands were near 50% frequency, fixation of the favored one would be nigh-unto-certain.

whimple said...

Sort of... the recombination that happens in mitotic cells is predominantly between sister chromatids. Recombination between parental homologs does happen in mitotic cells, but with much lower frequency.

whimple said...

Recombination, in the sense of exchanging genetic material between parental homologs, seems more of a mechanistic necessity for sexual reproduction than a reason for sexual reproduction. Through crossover interference most organisms actively reduce the amount of crossing over, though not to less than the one crossover per bivalent mechanistically necessary for segregation. If the point of sex was to do recombination, why then actively suppress recombination through crossover interference? In other words, most organisms tolerate some minimal amount of recombination in order to faithfully segregate homologs during sexual reproduction, which must have (as yet unknown) long-term advantages.

Arguing that organisms reproduce sexually in order to do recombination is roughly equivalent to arguing that organisms reproduce sexually in order to have testicles on males.

Alex SL said...

As I wrote in the first paragraph, I do not have any great idea how precisely it evolved.

As for the rest, as we have seen previously we have very different views of evolution. As far as I understand, you look at nature and see lineages randomly ambling around without a lot of adaptation or selection happening, relatively speaking; at the same time, the argument you just made hinges on the assumption that there is strong selection to reduce waste (effort without certain payoff).

I, on the other hand, look at nature and see organisms in a constant fight for survival against viruses, bacteria, herbivores or predators, and myriads of abiotic stresses such as heat, cold, salinity, radiation, etc, and all of that happening red queen style because the first of these also evolve and the last of these change all the time, in the case of plants simply because you ended up in a random place when you were a seed. Everything is highly adapted to not being killed; get one detail wrong in that regard and you are gone.

At the same time, at least eukaryotes do not actually appear to be under strong selection for or well adapted to efficiency. Nearly everything that happens inside the cells is fantastically wasteful compared to how bacteria do it, and that is before we come to what would appear to be highly wasteful macroscopic processes such as casting of all the leaves come winter or moulting every time you grow a bit.

So from my point of view, Eukaryotes appear to tolerate a lot of inefficiency and slack, which might give them the room to accumulate currently useless pre-adaptations and thus also provide a plausible route to evolving something as odd as sex. It seems to be more important to survive at all (where sex makes a lot of difference because you get good combinations of genes faster) than to be as efficient at growing as E. coli.

That's my best shot at this.

Tim Tyler said...

The "red queen" explanation of Hamilton and Van Valen proposes sex is advantageous largely because it generates diversity that hinders parasites. This is one of the main modern explanations of the existence and maintenance of sexual recombination (the other most notable one being the "gene repair" theory). The naive "generates diversity" idea thus went on to become the most important theory - a status it has held since the mid 1970s.

un said...

Hi Larry,

I thought that you might like this remark by John Thomson and Patrick Geddes who wrote about the problem of sex more than a century ago (in 1889, to be exact), before genetic recombination was even discovered:

"It has been raised as a reproach against the now fortunately dominant school of evolutionist naturalists that they could give no account of the origin of sex. Some people, like children, wish everything at once. Yet it must be admitted that there has been a lack of any sure and certain voice on this question. Apart from the simple fact that evolutionist biology is still young, there are three reasons for the comparative silence in regard to the origin of sex.

(I.) The first of these is still the curiously prevalent opinion, that when you have explained the utility or advantage of a fact, you have accounted for the fact, - an opinion which the theory of natural selection has done more to foster than to rebuff. Darwin was, indeed, himself characteristically silent in this regard to the origin of sex, as well as of many other "big lifts" in the organic series. Many, however, have from time to time pointed that the existence of male and female was a good thing. Thus Weismann finds sexual reproduction the chief, if not the sole source of progressive change. Be that as it may at present, it is evident that a certain pre-occupation with the ulterior benefits of the existence of male and female, may somewhat obscure the question of how male and female have in reality come to be.
" [The Evolution of Sex, P. 126]

And then they go on to list 2 other reasons why the biologists of their time were unable to resolve the problem of the origin of sexual reproduction. It's actually interesting that someone was criticizing adaptationism way back in the 19th century.

AllanMiller said...

I offer the opinion that we know just about all we're ever going to know about the dynamics of sex. The lack of a consensus stems in part from the fact that sex is so multi-faceted, and cuts across numerous areas of specialism. Evolutionary biologists look at it one way, cytologists another, geneticists a third and so on.

The certainty in the 1970's that fame and fortune awaited he or she who came up with 'the' killer reason seems to have given way to batting about the same fundamentals. "It must be adaptive. It just must because ... well, twofold cost. Some types turn it on and off.". But there probably won't be a killer reason. Take your pick of your favourite (other than those that are wrong!). It's all of 'em - Red Queen (not just restricted to parasites), Frozen Niche, Complementation, Fisher-Muller, Variation, Flexibility ....

To convince, one has to offer a model system. But by its very nature, sex does rather poorly in model systems. Homogenise the background, as you must for analysis, and sex is unconvincing. But it is necessary to enfold the whole of messy nature. Sex doesn't exist in a homogeneous background. Not least, once the clade's up and running, there are other sexuals to contend with.

Larry Moran said...

Maybe, just maybe, we're making an unnecessary assumption. Maybe sex isn't an adaptation. Maybe it's just an accident or maybe the first sexual organisms were maladaptive?

That could be one of the lessons that the facts are telling us.

Corneel said...

You seem to have missed the first part of the discussion , where Joe Felsenstein showed that sex does not generate extra diversity if there is linkage equilibrium. Sex does seem to help in fixing novel beneficial mutations more rapidly, which could be beneficial in a "red queen" like arms race like you suggest.

Corneel said...

But then again, the force of purifying selection is greatly diminished in SMALL sexually reproducing populations, so this only buys a little more time. Moreover, your argument relies on group selection over long evolutionary time scales (thousands of generations). That does not give a big selection advantage, compared to the obvious costs of sexual reproduction.
I don't think there is a lot of empirical support either for the idea that muller's ratchett increases the extinction rate of asexual populations.

AllanMiller said...

Yes, that was part of my point. Adaptive considerations apply across a sharp boundary in this instance, and importing within-population thinking can be misleading. I think that at the initiation of sex, when 'the competition' consisted only of asexual haploids, temporary diploidy could be seen as a sufficient benefit, through complementation or increase in size. It was a one-off. From then on, a sexual clade grew, and with it the population-level effects. Perennially asexual offshoots leave the arena in which 'adaptation' (in the selfish-gene, 'cost-based' sense) has traction. They become a different species. They suffer genetic degradation, by increasing homozygosity from gene conversion, and lack a mechanism to do much about it.

AllanMiller said...

The basic version, perhaps not, but gene conversion is a bit of a ticking time bomb for uncovering mutations, many of which had already happened when the asexual 'froze':

Corneel said...

That's a cool story allright. Still, I have to see whether this finding will generalise. Several ancient asexual lineages (such as bdelloid rotifers) don't seem to be bothered by it.

AllanMiller said...

Oh, absolutely; every rule has its exception in biology. Best be careful that one's model system is not actually the exception!

But gene conversion does provide a potentially general contributory explanation as to why diploid lineages are not as immortal as they 'ought' to be. One might expect a stochastic distribution of 'genomes with the potential to be troubled by unmasking recessives'. Some will slip through unharmed, or possess mechanistic countermeasures.

Piotr Gąsiorowski said...

Several ancient asexual lineages (such as bdelloid rotifers) don't seem to be bothered by it.

They have discovered alternative strategies, such as living on stolen DNA.

Tom Mueller said...


How the issue of sex and evolution is taught in some high schools, a downloadable case study based on a 2005 article in Nature written by three scientists from the Imperial College London that deals with the issue of sexual vs. asexual reproduction and their relative merits.

I had completely forgotten I had filed this away.

Anonymous said...


Pardon me for the naive question (my background is in the computer industry) but I was wondering if, since mitochondrial DNA is preserved through the female line - somewhat similar to the preservation of adaptive combinations through asexual reproduction - might this not go some way to counter-balance the "cost" to males?

The prevalence of transferring genes from mtDNA to nucei in clonal and self-fertilizing plants over out-crossing plants ( seems to suggest that transferrence of genes is important.

Which leads me to another naive question: does mtDNA to some extent effectively preserve adaptive combinations of genes, thus compensating for the breaking-up of such combinations in asexual reproduction?

Kindest regards,


Joachim Dagg said...

"It's actually interesting that someone was criticizing adaptationism way back in the 19th century."

That period has also been called the eclipse of Darwinism. Scholars discussed evolutionary theories like Spencer style Lamarckism, Orthogenesis, Saltationism and soon Mutationism. The question seems to have been whether natural selection was a necessary part of evolutionary theory at all. Nowadays, the criticism of adaptationism seems to be about the relative importance of natural selection, instead, while everybody agrees that natural selection is a necessary part of evolutionary theory in the first place.