More Recent Comments

Sunday, January 15, 2017

Why are most biologists adaptationists?

I enjoyed listening to Michael Lynch's talk on Friday. Much of what he said has been covered in Sandwalk over the past few years. His main point was that nothing in biology makes sense except in the light of population genetics. He laments the fact that most biologists, and even most evolutionary biologists, don't have a firm grasp of population genetics and the importance of random genetic drift.

I asked him why he thought this was true. He said he didn't know why. I think he was being polite. If you read his book, "The Origins of Genome Architecture," you'll see that he attributes this phenomeon to ignorance of modern evolutionary theory.
Contrary to popular belief, evolution is not driven by natural selection alone. Many aspects of evolutionary change are indeed facilitated by natural selection, but all populations are influenced by nonadaptive forces of mutation, recombination, and random genetic drift. These additional forces are not simple embellishments around a primary axis of selection, but are quite the opposite—they dictate what natural selection can and cannot do. Although this basic principle has been known for a long time, it is quite remarkable that most biologists continue to interpret nearly aspect of biodiversity as an outcome of adaptive processes. This blind acceptance of natural selection as the only force relevant to evolution has led to a lot of sloppy thinking, and is probably the primary reason why evolution is viewed as a soft science by much of society.

A central point to be explained in this book is that most aspects of evolution at the genome level cannot be fully explained in adaptive terms, and moreover, that many features could not have emerged without a near-complete disengagement of the power of natural selection. This contention is supported by a wide array of comparative data, as well as by well-established principles of population genetics. However, even if such support did not exist, there is an important reason for pursuing nonadaptive (neutral) models of evolution. If one wants to confidently invoke a specific adaptive scenario to explain an observed pattern of comparative data, then an ability to reject a hypothesis based entirely on the nonadaptive forces of evolution is critical. (pp. xiii-xiv)
Not only do biologists not appreciate random genetic drift but neutral evolution should be the null hypothesis of all investigations.

This theme was reiterated by Eugene Koonin in a recent paper [You MUST read this paper if you are interested in evolution]. That paper was summarized by Forbes writer John Farrell in "Why Biologists Waste Time Looking For Adaptations That Don't Exist. It's worth reading his article if you don't like the more difficult scientific papers and books.


70 comments :

Jonathan Badger said...

I think there are (at least) kind of two different sorts of evolution going on and people can be adaptationists at one level and not another. While I think nearly everybody who works on genome evolution and molecular evolution in general recognizes that most variation seen at those levels is not adaptive (as it generally has no phenotypic value as far as we can see), it is *much* harder to argue that major phenotypic traits of organisms did not arise through selection (whether or not we can actually understand what was being selected for is another matter, of course).

Petrushka said...

The deep problem here is that ignoring drift makes adaptive mutations look like miracles.

Robert Byers said...

Amen. The genetic separation between biologists and evolutionary biologists. is rightly announced here. A creationist complaint of old. We are fighting the few and not the many.
It seems there must be something bigger then mere ignorance for why so much adaptionsist biologists.

It must be the new ideas are not welcome or seen as important to change general ideas.
Its like saying, like Gould, that the old ideas were wrong or/and just don't work to explain mechanism for biological change.
I think it means a great wHOPPS WE WERE WRONG and now correct it and shhh.
Thats why they are slow i think.


Corneel said...

Robert said
It must be the new ideas are not welcome or seen as important to change general ideas.
Its like saying, like Gould, that the old ideas were wrong or/and just don't work to explain mechanism for biological change.
I think it means a great wHOPPS WE WERE WRONG and now correct it and shhh.
Thats why they are slow i think.


Do you know the historical roots of ID? Have you ever heard of Michael Paley and the watchmaker argument?

S Johnson said...

From Coyne, publicizing Steve Jones' new book: "Part of the new Ladybird Expert series, Evolution is a clear, simple and entertaining introduction to Charles Darwin’s pioneering and revolutionary theory of how all life changes through natural selection."

Steve Jones is an adaptationist because he's an eminent geneticists and expert on evolution, who unlike Gould et al. isn't out to make a name for himself. If I recall correctly Steve Jones is also worried about the slow disappearance of the Y chromosome in human beings.

More generally, adaptationism via natural selection, is Nature's (read, God's) seal of approval on what is. And it is foundational to evolutionary psychology and the reality of biological races, both things that triumphantly affirm the sanity (read, decency) of orthodoxy.

Petrushka said...

You appear to be attributing an endorsement by Coyne, and you imply things that Coyne didn't say. Or perhaps you can supply a link to back up your statements.

Jass said...

His main point was that nothing in biology makes sense except in the light of population genetics. He laments the fact that most biologists, and even most evolutionary biologists, don't have a firm grasp of population genetics and the importance of random genetic drift.

I asked him why he thought this was true. He said he didn't know why. I think he was being polite. If you read his book, "The Origins of Genome Architecture," you'll see that he attributes this phenomeon to ignorance of modern evolutionary theory.


Hmmm...I'm just wondering why would such an important mechanism of evolution (if not the saving one for you) as you emphasize random genetic drift, woulds be overlooked by soooo many experimental evolutionary biologist?

Unless it is an act of desperation on your and Lynch's part, to save your faces, I find it really bizarre.

How do you miss such an important opportunity to stick it to creationists?

Did you happen to mention "your baby" (random genetic drift) to the gods at the Royal Society Meeting? If this mechanism is such a great evolutionary creative power, why none of them never, even mention it?

I would hope you did and at least tried to convince some of the evolutionary gods at the RSM to see your and now Lynch's experimental evidence...

So, how did you do at the royals Larry? I'm sure that you and all your fans of the random genetic drift would love to hear how you overwhelmed all those disillusioned evolutionist with your simple but experimentally proven solution of random genetic drift...

Larry, please don't keeps us and the world in the dark! Let the world see the evidence you and now Lynch have been preserving until today as a punchline...

Joe Felsenstein said...

Excellent point. I have tried to point this out to Larry, with not much success. In a population of size N (say 1,000,000) a mutation that will be neutral only if it has the absolute value of its selection coefficient less than 1 part in 1,000,000. Can we imagine that any noticeable difference in morphology has that small an effect on fitness? I can't. But many people simply assume that if we can't see any noticeable difference in fitness then the change must be neutral. Taking neutrality as a natural null hypothesis for changes of morphology is not a wise practice.

Neil Taylor said...

Isn't the relevant population the effective population size?

Does any invertebrate have an effective population size much over 10,000?

Neil Taylor said...

Opps that should be vertebrate!

William Spearshake said...

This may be a little off topic, but I have been confused by the latest argument by IDists that adaptations are not evolution. Can anyone here explain their (supposed) rationale. The IDists certainly can't, but they are adamant about it.

Joe Felsenstein said...

Of course. White-footed deermice are found all over North America. Not to mention deer themselves. Song sparrows? Sardines? I think the idea that effective population sizes are typically as low as 10,000 is wrong. By the way, it is *not* common for effective population size to be even 10x lower than census population size.

Larry Moran said...

Joe Felsenstein asks,

Can we imagine that any noticeable difference in morphology has that small an effect on fitness?

I can imagine that the ability to role your tongue, the ability to wiggle your ears, whether your second toe is longer than your big toe, the size of your lips, male pattern baldness, and the shape of your ear lobe have a negligible effect on fitness. I can also imagine the the effective population size of the humans who exhibited those traits in the ancient past was a lot less than one million.

Do you think that any of these variable traits, which are clearly present at reasonable frequencies in today's population, have an effect on fitness? If so, can you identify which allele is beneficial and what kind of selective coefficient you expect?

I'd love to know if my grandchildren, who can all role their tongues, are going to have many more descendants than their friends who can't. I'd love to know how losing my hair when I was in my late 50s affected my ability to have children. I'd love to know whether having "free" earlobes (a polygenic trait) makes me more fit or less fit.

I'd love to believe that having a longer second toe, as I do, is correlated with high intelligence, greater success at mating, and enhanced fitness. Is this true? Should we just automatically assume that there's an adaptation in there somewhere or is it more scientific to withhold judgement and adopt the null hypothesis until there's evidence of a fitness benefit?

Neil Taylor said...

Thanks for that - but isn't the estimate for humans in the 10s of thousands? I was seduced by the assumption we would be typical and going from 7 billion to 10K seemed quite a reduction!

I found this: Effective population size/adult population size ratios in
wildlife: a review. Though it's very old, 1995.

It says:

(iv) Comprehensive estimates of Ne/N

The mean of NJN estimates that included all relevant
variables was 0.11 in both the full- and pooled
comprehensive data sets (Table 1). The estimate of
Ne/N from the multiple regression equation utilizing
fluctuation in population size, variance in family size
and unequal sex-ratios was 01.0 for both the full data
set and the pooled data set. Explicitly correcting the
above estimates so that they had NA as a divisor
resulted in essentially no change. The mean of the
estimates given by Nei & Graur (1984) was 0.06, not
significantly different from the means above (95%
confidence interval of 0-02-0-11)

The estimate of 0.06 is 16 times smaller than the census population.

Has it stood up to the test of time - Prof Felsenstein you imply not.

Joe Felsenstein said...

Yes, human effective population sizes in the past few hundred thousand years were quite low, near 10,000. Chimpanzee effective population sizes were larger, maybe 100,000. Human numbers were unusually low among mammals.

It is likely that a lot of traits (such as tongue-rolling) were not neutral, but were correlated with other traits involving, say, tongue musculature, that were selected, and the tongue-rolling got dragged along for the ride. Of course it is naïve to imagine that every trait is optimized separately by natural selection. But it is equally naïve to imagine that these traits are neutral.

txpiper said...

The most strident young earth creationists have to accept radical adaptation and rapid speciation. If those things are evolution, there is no argument.

What is actually in dispute is whether or not there are any conceivable limits as to what the mechanisms of evolution can accomplish.

Joe Felsenstein said...

I am happy to accept the figure that effective population sizes are 16 times lower than census sizes, and be corrected on that. But what were the census sizes. For mice? For sparrows? For sardines? Not just for humans. There are estimated to be 5,000,000 ring seals in the arctic, for example. How many mule deer are there in North America? How many otters? How many great blue heron?

Anonymous said...

Velhovsky,

You sound very sick. What makes you so mad about this?

Also, what makes you think that making stupidly ignorant comments works for your "cause"? Maybe you don't really care about your beliefs, if you're so willing to "defend" them by making such displays of stupidity and ignorance.

You could have saved some embarrassment if you just checked a bit around and read for understanding. For example, neutral theory has been around for quite a while. It's not something that Larry and Lynch invented out of the blue. So, read around. I mean, if you care. Just a suggestion.

You might also want to get some help about your mental health. (I'm not joking.)

Larry Moran said...

Joe,

I am not naive. I am not arguing that all those traits HAVE to be neutral. I am trying to make two points.

1. It is wrong to assume that all those traits have to affect fitness. That's the adaptationist fallacy. Enjoy the pleasures of uralism.

2. It is preferable to adopt neutrality as the null hypothesis and demand evidence for adaptation rather than just assuming it as the default hypothesis.

txpiper said...

Joe,

You will not find a more tolerant host than Professor Moran. You are a guest here. Please be courteous.

Joe Felsenstein said...

Larry: "uralism"? I am not sure what that is. I did pass through the Urals once, but on a train, at night, so I don't know much about them.

Robert Byers said...

ID is only recent unless your saying all ideas about God and creation are ID.
I know about the watch thing.
ID is really getting into the modern nuts and bolts of biology etc and disputing claims that chance could organize great organiztions in biology. then proving it.
Complexity is the great old idea about god being the creator.
iD is not just about complexity but the proposed rejection of it by chance evolutionism etc just is impossible. then they show why and how. Thus modern ID.

txpiper said...

Joe, this is an open forum.

Diogenes said...

Joe G is the infamous Security Clearance Joe Gallien, who lied about being a scientist, doing "national security" research he can't tell you about until you get a security clearance, getting injured outrunning RPGs in Iraq, and most relevantly, his claim that he has a method to determine whether any sequence in intelligently designed or not. *This guy* is calling Larry a "proven liar."

Corneel said...

Just pointing out that as a self-proclaimed YEC, maybe you should be a bit more self aware when you are accusing other people of being reluctant to accept new ideas.

Larry Moran said...

"uralism" = pluralism

S Johnson said...

https://whyevolutionistrue.wordpress.com/2017/01/15/a-new-general-book-on-evolution-by-steve-jones-and-one-on-climate-change-by-prince-charles/ I don't recall the words "I endorse this book" but if Steve Jones felt slighted by this, I would be concerned for his emotional well-being.

As to Coyne's opinions on Gould, I quote myself: "Coyne wrote: "Steve Gould and Niles Eldredge, with their theory of punctuated equilibrium proposed in the Seventies, essentially made the extreme non-Darwinian claim that big evolutionary changes happens when small populations somehow lose their 'genetic equilibrium' (how wasn’t specified), and, further, that species selection was responsible for trends in the fossil record as well as adaptations themselves." (Original at https://whyevolutionistrue.wordpress.com/page/6/)"

The link for the quote: http://sandwalk.blogspot.com/2017/01/you-must-read-this-paper-if-you-are.html#comment-form

S Johnson said...

Again Coyne makes a post relevant to the adaptationist stance of evolutionary biologists holding sound views as to the folly of critiques of the Modern Synthesis: https://whyevolutionistrue.wordpress.com/2017/01/16/a-superbug-resistant-to-all-antibiotics-kills-nevada-women/#comments

Coyne knows that natural selection is so effective that bacteria should be expected to lose antibiotic resistance to old, long-unused antibiotics. His post is about prompting discussion of this evolutionary puzzle. It is to be expected that a lay crank like me should see other reasons for the mainstream evolutionary biologists to hold equate natural selection and evolution, and conclude adaptation is pervasive. But this is perhaps the best place to remember they tell us they believe the critics of this to be charlatans who have never succeeded in supporting their claims.

Petrushka said...

"I don't recall the words 'I endorse this book' but if Steve Jones felt slighted by this, I would be concerned for his emotional well-being."

He hadn't even read the book. He was just passing on a recommendation. Why not go to the original source? Your other link doesn't work

Unknown said...

His other link doesn't work, because he didn't link to a specific post, but to a page on Coynes history (which since it moves as now post appear, doesn't help).
https://whyevolutionistrue.wordpress.com/2016/12/26/why-are-scientists-always-saying-that-evolutionary-biology-needs-urgent-and-serious-reform/
is the post S Johnson refers to.
Now, there are some issues with Coynes post as far as it pertains to Gould. For starters species selection wasn't (and still isn't) intimately linked to PE - it was proposed by Steven Stanley and not by G&E. The idea that PE was how Gould made his name is also somewhat flawed. When G&E first presented it, Gould already had tenure at Harvard. He didn't get it because Harvard thought he'd do something spectacular sometime in the future, but on the strength of his research, say his 1966 article "Allometry and size in ontogeny and phylogeny", which currently has 1811 citations or his various other articles on allometric growth patterns. You could also easily deduct PE from Goulds CV and he'd still be one of the most influential paleontologists of the 20th century just on the basis of his work on allometry, the MBL papers and the spandrels paper. PE seems to overshaddow his other accomplishments - similarly to how people think of Einstein mainly as the father of relativity, but he would have been a big name in Physics on the strengths of his other 1905 papers (on brownian motion and the photoelectric effect) alone.

John Harshman said...

It seems to me that if you are trying to test for selection, neutrality is a necessary null hypothesis. But failure to reject the null hypothesis is not the same as accepting the null hypothesis. Perhaps a confidence interval on the selection coefficient would be a better statistic? If the confidence interval fails to include 1/N, we might suppose that at least the practical neutrality hypothesis is supported.

Unknown said...

I am not naive. I am not arguing that all those traits HAVE to be neutral. I am trying to make two points.

1. It is wrong to assume that all those traits have to affect fitness. That's the adaptationist fallacy. Enjoy the pleasures of uralism.


Why ignore the core message of the spandrels paper? Which is: There are more "traits" than parameters. You should aim to isolate parameters before you errect just-so stories about traits. You've made the same mistake here, making an ad-hoc assertion about traits, without trying to figure out the parameters. Neutrality is different from non-adaptivity of a trait.

2. It is preferable to adopt neutrality as the null hypothesis and demand evidence for adaptation rather than just assuming it as the default hypothesis.

A null hypothesis is not the same as a default hypothesis. And there is a difference between failing to reject a null and affirming it. Arguably no mutation ever is neutral (argument: s is an absolutely continuous random variable. QED.), but we don't generally care about that. We can not measure s with perfect accuracy and all that matters is that s is so close to 0 that we couldn't tell the difference. It's worth noting that in terms of morphology a more reasonable null is passivity, which is a weaker claim than neutrality (for a quantitive trait it assumes that selection has a centered effect. Neutrality of all alleles affecting a trait implies passivity, but passivity does not imply neutrality).

txpiper said...

Gould and Eldredge marked a new direction in theoretical science with PE. The idea was to reconcile supposed gradualism with an uncooperative fossil record. PE wasn't proposed on the basis of evidence. It was an attempt to explain the lack of it.

Anonymous said...

txpiper,

It would be much better if you investigated before posting those ignorant comments. PE was proposed because of evidence. Gould worked with some pretty rich fossil records, if memory serves, it was trilobites. He noticed that the species of these trilobites remained anatomically "static" for a while, and that those periods of stasis (equilibrium) were "punctuated" by periods or relatively (relatively being a very important word here) fast change.

Unfortunately, to sell this observation as if it was some amazing departure in evolutionary understanding, rather than a well observed phenomenon and an improvement in our understanding of evolutionary tempo, Gould discussed it as if "gradualism" was meant to mean "constant, even, and uninterrupted change," when all it meant was that changes were not expected to happen magically in one step (which is why we don't expect a duck to be born from a crocodile). Anyway, PE is not "a fix," but the description of very well documented observations of evolutionary equilibria punctuated by evolutionary "revolutions." The fossils "cooperated" all along.

Unknown said...

If one assumes the neutral hypothesis for a gene sequence or a trait, then one can falsify this hypothesis by showing the advantage to the sequence or trait. One would never ‘prove’ neutrality, just as nobody has ‘proved’ general relativity.
If one assumes the sequence or trait has advantage, then one can never falsify this as it is impossible to show neutrality. We end up with ‘just so stories’ that are unscientific. This leads to the impression that there isn’t any real science involved in the subject of evolution.
But there is a science of evolution- it involves population genetics and the neutral hypothesis. I’m so glad I see that now.

John Harshman said...

Gould worked with some pretty rich fossil records, if memory serves, it was trilobites.

That was Eldredge. Gould's thing was Cerion. Neither had much to do with his claims for stasis. Whether the fossils document punctuated stasis is an interesting question which I think is so far unresolved.

Unknown said...


Adaptationist thinking appeals to the intuition and creative thinking.
This is why it competes with creationism- they both appeal to the intuition and allow for the creation of ‘just so stories’.

The actual science (population genetics) is rather dry by comparison.

My two cents.

Joe Felsenstein said...

There have been many substantive remarks on this subthread (plus the usual pure noise remarks by trolls). Let me just repeat a little in response to Larry:

1. I agree with him that we should not assume that natural selection optimizes each morphological (or behavioral or physiological) character that we look at. But this is not just because it might be neutral -- another possibility is that its value may be a byproduct of selection on a correlated trait.

2. Most of these characters must show fitness differences, if you change them enough. It may not be obvious what is the fitness effect of increasing the length of your thumb by 1%. But increase it by 100% and it is a reasonable supposition that this has an effect on fitness.

3. For molecular changes such as single-base substitutions in intergenic sequences, particularly far from the nearest gene, neutrality is a highly plausible null hypothesis.

4. For changes in visible morphological characters, it is highly implausible. You might use it as a null hypothesis, but on finding no evidence for selection, you ought to keep in mind that selection coefficients too small to be detected by us are quite possibly big enough to affect fixation probabilities.

'Nuf said.

S Johnson said...

Sorry about the bad link.

txpiper said...

photosynthesis,

“Gould discussed it as if "gradualism" was meant to mean "constant, even, and uninterrupted change," when all it meant was that changes were not expected to happen magically in one step”

No, that is not what he was doing at all. He was noticing that stasis in the fossil record does not match the expectations of the theory. Decades later, in 2004, Eldredge was still wrestling with this frustrating issue:

”Studies of extant taxa with rich fossil records provide mounting evidence that morphologically defined species-level lineages recognized in fossil sequences often correspond to genetically defined species in the modern biota (Jablonski 2000). Such studies are crucial to the demonstration that patterns of stasis in the fossil record constitute a genuine problem for evolutionary theory.”
http://www.tiem.utk.edu/~gavrila/papers/stasis.pdf

Larry Moran said...

Joe Felsenstein says,

4. For changes in visible morphological characters, it is highly implausible. You might use it as a null hypothesis, but on finding no evidence for selection, you ought to keep in mind that selection coefficients too small to be detected by us are quite possibly big enough to affect fixation probabilities.

Gould & Lewontin listed several kinds of arguments commonly used by adaptationists.

1. If one adaptive argument fails, try another.

2. If your best adaptive arguments fail then just assume that anther one must exist even if you don't know it right now.

3. If you don't have a good just-so story then claim it's because you don't understand the exact environmental conditions.

4. Concentrate on possible adaptive features in today's species and ignore the fact that the feature may have arisen for other reasons in the past.

5. Ignore the fact that not only neutral alleles but also deleterious alleles can become fixed. A particular feature might actually be maladaptive.

6. If no adaptive explanation makes sense then attribute the trait to adaptation occurring somewhere else (hitchhiker effect).

7. If all else fails, invoke sexual selection.

And now we have ....

8. It is highly plausible that some version of variable visible morphological traits in a population is adaptive (and, consequently, the other allele(s) are maladaptive). 'Nuff said. (A version of #2.)

Anonymous said...

You should have read the paper txpiper. It's ironic that you quoted from that article and you didn't notice that its content demonstrates my points. Starting with this one: it would be much better if you investigated before posting those ignorant comments.

txpiper said...

I've read it more than once. Stasis is an ongoing problem because it means that things don’t evolve.

The closing statement says:

"The solution to these and related problems will demand further integration of the fields of evolutionary ecology and evolutionary developmental biology into evolutionary genetic and paleontological approaches”. In other words, “We have no idea”.

Misconceptions in evolution theories said...

I do not know about how IDists define adaptation compared to evolution, but I base my different meaning of the words on Darwin. I have written about it in my new blog, in this blog post. I have also made an introduction to the theme here. In my view adaptation is about short term changes of allele frequencies in a population, while evolution is about novelties like new features and speciation.

Corneel said...

@Jarle Kotsbak

Isn't that distinction already covered by microevolution and macroevolution respectively?

Misconceptions in evolution theories said...

@Corneel

You are right. There are different definitions and different terms in use. I know Stephen J Gould used those terms. But I am not sure he used these terms in the same way as I use adaptation and evolution. As I have understood his definition, macroevolution is changes that can be detected by fossil records, while he saw e.g. the evolution of an eye as microevolution. But maybe you have another definition?

Corneel said...

@Jarle Kotsbak

In my understanding, macro evolution refers to evolutionary processes at or above the species level, such as mass extinctions, adaptive radiation and such.

I note that our dependable host has already written an essay on it.

Under this definition, adaptation should simply be subsumed under the heading of evolution.

Misconceptions in evolution theories said...

@Corneel

Thank you. I will study it, although I think I have seen it before. I have also looked at Wikipedia, and written a comment to by own blog, that I referred to above. Do you have specific comments to my blog, so give them there. I am now writing a new blog post that includes randomness, so stay tuned.

S Johnson said...

Adaptationism never rests. Since natural selection is all powerful, the zebra's stripes must be adaptive. https://whyevolutionistrue.wordpress.com/2017/01/18/the-secret-of-zebra-stripes-solved-or-so-scientists-say/

It is unclear to me why the possibility that horses outside Africa aren't striped because stripes didn't happen to be a trait of ancestral populations. Nor is it clear how it's been demonstrated that striped horses were selected over unstriped horses. That seems to suggest that the differential reproduction from being bitten less (not at all?) is so great that stripe were fixed in African populations. Nor is it all clear that biting flies in other continents were so much less of a detriment to reproduction. All I'm really getting from this is that evolutionary biologists have a rock hard conviction that all traits are adaptive because of natural selection.

S Johnson said...

From what I understand, the fossil record for many species tends to show a geologically slow change in morphology. And, as a result, it is technically correct to speak of "chronospecies," since species in paleontology are morphological defined, rather than by observation of interbreeding as per the biological species concept held by Mayr et al. Apparently that's why it's so common for discussions of ancient organisms to speak most often of genera, rather than species? The thing is, it is not at all clear to me that this kind of evolution (and surely it is evolution too) doesn't demonstrate random genetic drift on a species wide scale (rather than in isolated populations.) At any rate, it seems to me the adaptationists should be able to see natural selection at work here too. As far as I know that's not the case at all.

rich lawler said...

It's been awhile since I (re)read Gould/Lewontin, but did they really mention sexual selection? (your point 7).

I do remember that the Clutton-Brock/Harvey paper that preceded Gould/Lewontin's paper at the same conference made more use of sexual selection as an explanatory concept for putative adaptations.

Larry Moran said...

@rich lawler,

No, they didn't mention sexual selection. I made that up.

It is, however, the fall-back position for adaptationists when all other explanations don't make sense. Problem is, they can't decided whether men prefer women who can role their tongues or reject them! In other words, |s| must be greater than zero but the sign is tricky. :-)

rich lawler said...

Thanks for the response, makes sense. And I agree, sexual selection gets soundly abused, particularly when applied to human mate choice (i.e., the Ev-Psych nonsense).

rich lawler said...

Stasis has never been a problem for evolutionary theory and doesn't imply that things don't evolve (since a population held at an optimum by stabilizing selection or some other mechanism will still accumulate neutral mutations). An excellent analysis of stasis in the fossil record is by Estes/Arnold in Am. Nat. 2007. They show how a moving optimum can explain many different patterns of morphological evolution in the fossil record, particularly stasis.

The fossil record is rife with examples of gradualism (e.g., the slow dental evolution of early Adapids), and punctuation (e.g., Paleozoic brachiopods). There is no "rule" about fossils and how they should evolve, so there is no need to invoke generalizations as if the past was some mysterious magical thing to which well-documented present-day evolutionary mechanisms don't pertain. But certainly one contributing factor to debates about fossil evolution is simply the incompleteness of their record. Several people have written about this.

Donald said...

I am not seeing my post appear, so I will repeat it. Morton's toe, whichnis the condition of having the second toe the longest, is thought to cause various running injuries. For instance, it can supposedly cause too much pressure on the second toe when running.

I am not a doctor so I qualified all this, but you can find this referenced online and I learned about it in an old book I have on common running injuries.

Here are some lay people talking about it on a running site.

http://community.runnersworld.com/topic/morton-s-toe-morton-s-foot

The point being that sometimes small morphological differences can matter. Actually, as a layperson I wonder why Morton's toe wasn't selected out back when being able to run long distances might have mattered.

Larry Moran said...

Do runners have more babies?

Unknown said...

The Estes & Arnold paper does not address the central issue, which is that the distribution of rates of change is bimodal for a wide range of taxa and quantitative traits. It is worth noting that referring to the incompleteness of the fossil record often amounts to nothing more than handwaving, unless there is a concrete discussion on how it affects observations. In this particular case it is worth noting that if the underlying rate distribution is bimodal (i.e. PE holds) then the observed distribution will tend to become unimodal as sampling gets worse. The reverse isn't true - an underlying unimodal distribution stays unimodal. The idea that the PE pattern observed for many taxa is an artifact of incomplete preservation is nonsense for this reason - but you could have the appearance of gradualism due to a lack of fossils. Adapidae have just 60 occurences (PaleoDB querry, just now), while paleozoic brachiopods have 133,527. That's a tad more. PE is generally found for taxa that are sexually reproducing and have an extensive and comparably excellent fossil record. It's not found for taxa that have shoddy records. There's no equivalence here, just like you can't say "well, there are papers with large sample sizes and long durations demonstrating a link between smoking and lung cancer, but I compared 10 smokers to 10 non-smokers over a 2 year period and found no significant health effects, so I guess the issue is wide open. And anyway there's never enough data, so maybe the studies that found a link between smoking and cancer have too little statistical power".

AllanMiller said...

ISTM that the determination that the selection coefficient definitively takes any value - which would include zero - is on rather shaky ground. It's a reasonable null hypothesis against which to check departures. But it is damned hard to measure with the accuracy required to distinguish zero, or even the effectively neutral zone - to confirm the null, rather than to fail to reject it - within the confidence limits chosen.

rich lawler said...

I'm not handwaving about how sample sizes in the fossil record influences our interpretations of rates of change...plenty has been written about it--a simple google scholar search will demonstrate that (e.g., work by Pete Wagner).

I also wasn't talking about PE, and nor were Estes/Arnold (they bring it up only to say that their moving optimum model accounts for both bursts of change and stasis) but apparently you were.

S Johnson said...

Rich Lawler "The fossil record is rife with examples of gradualism (e.g., the slow dental evolution of early Adapids)..."

And there are the classic charts of the progression of man from ape to modern human. Or the equivalent for the evolution of the modern horse. Given the reserve with which we should treat those poularizations, I think lay people should be very cautious in claiming continuous evolution of phenotypes leading to multiplication of species.

But more to the point, the kind of adaptationism on hand as the topic of discussion is one which sees natural selection as the pervasive and powerful, so much so that all traits are conceived to be adaptive. Despite the occasional concession that mutation, random genetic drift etc. do exist, they are in practice never to be considered.

It seems to me this kind of adaptationism implies a nearly universal sequence of gradual changes in morphology as part of the multiplication of species. It also seems to me that adaptationism of a less imperial kind says that natural selection is the primary force conserving the phenotype despite mutation, random genetic drift etc. This kind natural selection, as the agent of stasis and protector of adaptations, instead of the agent of change, the creator of new species, does indeed seem to be a "well-documented present-day evolutionary mechanism...) I don't think the charge against PE that it posits some magical past processes is true at all.

rich lawler said...

I should clarify to Simon that I wasn't talking about PE when I was bringing up the Estes/Arnold paper. Obviously, my second paragraph I do bring up PE.

My apologies.

Anonymous said...

txpiper,

"I've read it more than once."

No, you didn't read it. Had you read it, you would not have said that "fossils didn't cooperate," since the article mentions many examples of fossils "cooperating." It also mentions that the "problem" they're talking about is understanding the tempo of phenotypic changes in evolution, as I explained to you. Not only that, they explain potential solutions to this "problem."

"Stasis is an ongoing problem because it means that things don’t evolve."

Stasis means that some organisms keep a phenotype for quite a while. Nothing in evolutionary theory demands that species can never show stasis. Only your idiocy and ignorance demands such a thing. Why did you link to that paper if you didn't read it? Did you think I would not read it either? Are you really that stupid?

"The closing statement says:"

I read the whole thing. You just looked for the word "problem" and didn't care to check what they might be talking about. Sad news to you: they don't mean what you thought they meant.

Learn to read. That would save you some embarrassment. Also read the other answers offered to you above.

Why do you insist on looking like an illiterate buffoon? Do you really think that you, looking like an idiot, will convince anybody here that you might as well be right?

rich lawler said...

S. Johnson: I have no idea what you're on about. I can't really understand the main points you are trying to make in almost every post you write. Sorry.

txpiper said...

photosynthesis,

“Stasis means that some organisms keep a phenotype for quite a while. Nothing in evolutionary theory demands that species can never show stasis.”

That isn’t exactly accurate. It is a problem. The paper opens and closes noticing that it is a problem. Some kind of theoretical accommodation has to be made to explain why some things stay stable for tens of millions of years, while others, like whales, are supposed to have raced from Pakicetus to a fully marine animal in just four million*.

I can sympathize with your embarrassment and anger. (It’s a religious deal, like someone pointing out to a muslim that the prophet Muhammed didn’t record any prophecies.) But I wouldn’t worry about it if I were you. Like rich lawler says, “Stasis has never been a problem for evolutionary theory”. That isn’t because it isn’t. It is because evolutionary theory is fiercely protected, and problems are not allowed. But, they can be easily solved by dropping a suitable adjective in front of ‘selection’, or by just using your imagination.

I know the term is somewhat taboo, but how do ‘living fossils’ fit in your worldview?

*http://news.nationalgeographic.com/news/2011/11/111116-antarctica-whales-oldest-evolution-animals-science/

S Johnson said...

rich lawler: since I don't post a lot here, it will be easy to skip my posts.

photosynthesis: "Nothing in evolutionary theory demands that species can never show stasis." Strictly speaking, beneficial mutations, neutral and nearly-neutral mutations, random genetic drift, recombination, etc. do strongly imply constant evolution in the sense of change in gene frequencies, so these normal processes of evolution do argue against stasis. And the presumed power of natural selection to optimize the body plan suggests that changes in environment (which includes other organisms also changine) will modify morphology (or phenotype or whatever you wish to call it.) I have often thought a popularization of evolutionary theory should be entitled "All Flesh Is Grass."

It's hard to tell though whether the statement was about evolution as multiplication of species. Or whether the tacit assumption is that drift, neutral etc. aren't really parts of modern evolutionary theory.

Anonymous said...

I would expect stasis in many organisms for long periods of time. Mutations change, but if the environment stays similar, most variations will die out and the population won't change. When the environment changes cyclically (ice ages vs. not, El Nino vs. not years etc.), selection pushes organisms a little this way and a little back; from any moderately long view, that will look like stasis. And when the environment changes in a consistent but slow way, populations may well migrate, staying in conditions to which they are well adapted, rather than change.

Anonymous said...

The claim by Gould and Eldredge that big evolutionary changes happen in small populations makes sense to me.

In little populations, genetic changes can be pretty random. Neutral or somewhat harmful traits can become fixed. Odd combinations of alleles can come about. If the small population survives (it probably won't) it may become quite different from its parent population, in unpredictable ways. Its environment is likely to be different from that occupied by the main population from which it arose, too, so selection pressures, to the extent they can operate, will be different.

Big population have a certain inertia. Yes, any slightly beneficial allele is likely to become common, but also it will shift range with the conditions it's adapted to, so neither selection nor drift will easily or quickly move it from its current adaptive peak.

Consider a cold-adapted plant as the ice-age ends. Main population shifts range north. Small populations are left on north-facing slopes, in coastal bogs, and in other cold pockets. There's a lot of genetic diversity in the newly isolated populations, though that will fall with time. New mutations, of course, aren't shared with the parent population.

What happens to the little populations? Mostly they go extinct, either because of bad luck or because inbreeding causes too many harmful alleles to come together. In some, harmful alleles are lost and the plants hang on for centuries if the habitat remains stable, in (apparent) stasis. Sometimes the parent population migrates back while interbreeding is still possible, and the potentially distinctive population is lost. And sometimes, sometimes, the genetic changes in this small population cause it to change in ways that leave it more or less well adapted to a different set of conditions than its parent population. It may hang on and stay small, or spread widely.

How small are the isolated populations? I've worked with ones of 20 or less (census, not effective populations) and many of a few hundred to a thousand. Some isolated in the last 200 years, some for thousands of years.

I think small populations are key to rapid or big evolutionary changes.

Larry Moran said...

@bwilson295

The important thing about PE is that change is associated with speciation by splitting (cladogenesis). In most cases the daughter species come to occupy the same environment as the parent species. This suggests that the environment is not playing a role in the speciation event. Furthermore, speciation in one lineage is not correlated with speciation in other lineages found in the same location suggesting that environmental change is not driving speciation.

The number of species in the same environment is always increasing by splitting but after over time some species go extinct. This led to development of species sorting as a higher level analogy to population genetics.

The idea that evolution must be linked to environmental change is undoubtedly true in many instances but the link is often exaggerated. Species still evolve in an unchanging environment. In order to explain stasis this way you have to make the assumption that all species are perfectly adapted to their current environment so they don't change unless the environment changes. That's an unreasonable assumption.

It's better to explain relative stasis as just the buffer effect seen in large populations.

Anonymous said...

S Johnson,
Yes. That's why I wrote "phenotype" and "phenotypic" as necessary. I did not mean genetic stasis.

Anonymous said...

I agree with you that relatives stasis is just the buffer effect seen in large populations. I agree that species can evolve to be different in an unchanging environment (new mutations, result of hybridization, etc.) and also that environments may not be as unchanging to the organisms as they appear to be to us (especially if we're looking at the fossil record). See oscillating variation in Galapagos finches, for example.

I do think that speciation is often associated with small populations either at the margins of a range or left behind as the large population moves. Uneven cladogenesis, one might say. That speciation may be associated with PE at the time of species or afterwards if the new species succeeds and spreads, when no only does it have different mutations than in the parental species, but it changes the environment (competative interactions, etc.).

Normally when you say something I disagree with, I figure I have to learn a lot quickly because I'm wrong. However, the importance of small populations seems to fit with what I see in the world. Therefore, in this case I figure that either you're wrong or (more likely) I've communicated badly so we wouldn't actually disagree if I could make clear what I think with all the appropriate qualifications.