The three forms of natural selection

[lucaspa]

I think it's past time that we discussed that natural selection comes in 3 forms. Too often speculations in this forum (and others) are based upon the premise that natural selection only directionally changes a population.

 

Purifying or stabilizing selection. Also called normalizing selection, this type of selection acts to preserve a certain array of phenotypes because of their selective advantage. Due to mutation and recombination, any Mendelian population can generate an enormous array of phenotypes, but selection limits the phenotypes to those of selective advantage. This is the type of selection seen when a species is well-adapted to a constant environment and is the type of selection responsible for stability. When a population is well-adapted to an environment (on a fitness peak), then ANY change will likely move the individual off the peak, making it less adapted. Thus natural selection in purifying selection will eliminate variation and cause a population to become genetic homogenous.

 

Directional selection. This is the form of natural selection people associate with the words "natural selection". This happens when either 1) a population moves into a new environment or 2) when the environment is changing in a single direction. Directional selection takes the form of shifting the norm of variations. Now the population is formed from phenotypes that were presently at low frequencies and the bell-shaped curve of the parameters of a trait shift either right or left.

 

Disruptive selection. This happens when a population encounters 2 (or more) separate environments in different areas of the region it occupies. Here the bell-shaped curve develops a valley in the middle between the two peaks representing the different regions. If there is little gene flow between populations, the phenotypes of the species will split into 2 separate bell-shaped curves. This would be allopatric speciation. An example of a species undergoing disruptive selection is the pocket gopher Thomomys bottae. We see different coat color, body size, and skeletal proportions. The herring gull is another example of disruptive selection. Disruptive selection has also been used by Thoday and others in the lab to produce reproductive isolation (species).

[CDarwin]

Are these different "kinds" of selection or different results of selection? It seems like the mechanisms would be the same, the differences come from the environment and the like. I'm not really making a point, just observing that.

[Dak]

but, as CDarwin said, aren't they mechanically identicall? and aren't they just slightly different instances of a phenomena that acts to 'idealise a species to its environment'?

 

iow, isn't there just one 'natural selection'?

[PhDP]

iow, isn't there just one 'natural selection'?

 

It's probably why he said "natural selection comes in 3 forms".

 

In truth, there's many different ways to classify the different forms of natural selection.

[Dak]

doh! i missed that bit

[SkepticLance]

There is also a common misconception that mutations appear to guide directional natural selection just in time to do the job. In fact, any population contains within its gene pool numerous mutations which do not really perform any useful function. However, with a change in environment, previously useless mutant genes may become useful. In that case, natural selection acts to increase their frequency in the gene pool.

 

For example: the sickle cell anemia mutant gene - normally harmful. If malaria is present in the population, it may become useful and be slected for.

[Sisyphus]

I'm wondering if there ever really is such a thing as a "fitness peak." Surely even in a stable environment there is almost always some change, right? Or am I just being too pedantic in demanding an absolute interpretation, and these are just abstract forms that represent always competing tendencies?

 

Also, I've been told that natural selection is "more like a sieve than a selector," and any mutation that isn't harmful above a certain threshold gets absorbed into the population. That would mean there would always be change, right?

[PhDP]

I'm wondering if there ever really is such a thing as a "fitness peak." Surely even in a stable environment there is almost always some change, right? Or am I just being too pedantic in demanding an absolute interpretation, and these are just abstract forms that represent always competing tendencies?

 

The fitness landscape is never totally stable. For example, the concept was often used to predict the optimal age at maturity, which depends on many environmental factors, including mortality. When mortality is high, it's less advantageous for an animal to invest in growth so they tend to start reproducing earlier. Obviously, mortality is not stable so the optimal age at maturity is never exactly the same. It might, in part, explain some variation in the observed traits.

 

Also, I've been told that natural selection is "more like a sieve than a selector," and any mutation that isn't harmful above a certain threshold gets absorbed into the population. That would mean there would always be change, right?

 

There's always change. A mutation with little effect on fitness will randomly reach fixation or extinction by drift. Unless the effective population size is very large (which is rarely the case for animals), random drift will be stronger than selection even for slightly deleterious allele and some will reach fixation.

[lucaspa]

It's probably why he said "natural selection comes in 3 forms".

 

In truth, there's many different ways to classify the different forms of natural selection.

 

Thanks for the closer reading.

 

Phil, I haven't found any other classification of the forms of natural selection. These are the forms discussed in all the textbooks of evolutionary biology I've seen, i.e. http://www.evotutor.org/Selection/Sl5A.html

 

If you've seen a different way to classify the forms of natural selection, I'd appreciate if you would post the source, BUT

 

I see Wikipedia lists a number of "forms" of selection. This is another reason why Wikipedia cannot be used as a source in a serious academic or science discussion.

 

I'm wondering if there ever really is such a thing as a "fitness peak." Surely even in a stable environment there is almost always some change, right? Or am I just being too pedantic in demanding an absolute interpretation, and these are just abstract forms that represent always competing tendencies?

 

You are being a bit pedantic. "Some change" may not be enough to change the fitness peak. Look at sharks or horseshoe crabs. Yes, there is enough change to give slightly different species, but overall their environment is constant such that there has been stabilizing selection for tens of millions of years.

 

Also, I've been told that natural selection is "more like a sieve than a selector," and any mutation that isn't harmful above a certain threshold gets absorbed into the population. That would mean there would always be change, right?

 

There are 2 ways to look at natural selection: elimination of traits/designs that aren't quite as good or preservation of traits/designs that are good. Darwin looked on NS as a preserver or selector of good traits/designs:

 

"But if variations useful to any organic being do occur, assuredly individuals thus characterized will have the best chance of being preserved in the struggle for life; and from the strong principle of inheritance they will will tend to produce offspring similarly characterized. This principle of preservation, I have called, for the sake of brevity, Natural Selection." [Origin, p 127 6th ed.]

 

"Change" depends on the environment. And while yes, if looked at in total the environment might change, that does not mean the environment for each and every locus will change. One locus can be under directional selection and another under stabilizing selection.

 

Futuyma cites Travis J, The role of optimizing selection in natural populations. Ann. Rev. Ecol. Syst. 20: 279-296, 1989 as a source documenting stabilizing selection.

 

Much of the data on the existence of stabilizing selection was gathered before 1980. Some of the recent papers documenting stabilizing selection are listed first, followed by some of the older papers:

 

1: Trends Genet. 2006 Aug;22(8):456-61. Epub 2006 Jun 27.

 

Natural selection on gene expression.

 

Gilad Y, Oshlack A, Rifkin SA.

 

Department of Human Genetics, University of Chicago, Chicago, Il 60637, USA.

gilad@uchicago.edu

 

Changes in genetic regulation contribute to adaptations in natural populations

and influence susceptibility to human diseases. Despite their potential

phenotypic importance, the selective pressures acting on regulatory processes in general and gene expression levels in particular are largely unknown. Studies in model organisms suggest that the expression levels of most genes evolve under stabilizing selection, although a few are consistent with adaptive evolution. However, it has been proposed that gene expression levels in primates evolve largely in the absence of selective constraints. In this article, we discuss the microarray-based observations that led to these disparate interpretations. We conclude that in both primates and model organisms, stabilizing selection is likely to be the dominant mode of gene expression evolution. An important implication is that mutations affecting gene expression will often be deleterious

and might underlie many human diseases."

 

1: Hunt J, Blows MW, Zajitschek F, Jennions MD, Brooks R.

Reconciling Strong Stabilizing Selection with the Maintenance of Genetic

Variation in a Natural Population of black field crickets (Teleogryllus

commodus).

Genetics. 2007 Jul 29; [Epub ahead of print]

PMID: 17660544 [PubMed - as supplied by publisher]

 

2: Bottin L, Isnard C, Lagrange A, Bouvet JM.

Comparative molecular and phytochemical study of the tree species Santalum

austrocaledonicum (Santalaceae) distributed in the New-Caledonian archipelago. Chem Biodivers. 2007 Jul;4(7):1541-56.

PMID: 17638336 [PubMed - indexed for MEDLINE]

 

3: Zhang XS, Hill WG.

Multivariate stabilizing selection and pleiotropy in the maintenance of

quantitative genetic variation. Evolution Int J Org Evolution. 2003 Aug;57(8):1761-75. PMID: 14503618 [PubMed - indexed for MEDLINE]

 

4: Burger R, Gimelfarb A.

Genetic variation maintained in multilocus models of additive quantitative traits under stabilizing selection.

Genetics. 1999 Jun;152(2):807-20.

PMID: 10353920 [PubMed - indexed for MEDLINE]

 

5: Brooks R, Hunt J, Blows MW, Smith MJ, Bussiere LF, Jennions MD.

Experimental evidence for multivariate stabilizing sexual selection.

Evolution Int J Org Evolution. 2005 Apr;59(4):871-80.

PMID: 15926696 [PubMed - indexed for MEDLINE]

 

6: Lemos B, Meiklejohn CD, Caceres M, Hartl DL.

Rates of divergence in gene expression profiles of primates, mice, and flies:

stabilizing selection and variability among functional categories.

Evolution Int J Org Evolution. 2005 Jan;59(1):126-37.

 

Here are some of the older papers:

 

1: Kaufman PK, Enfield FD, Comstock RE.

Stabilizing Selection for Pupa Weight in TRIBOLIUM CASTANEUM.

Genetics. 1977 Oct;87(2):327-341.

PMID: 17248766 [PubMed - as supplied by publisher]

 

2: Aslam M, Browder LE.

Stabilizing selection for pathogenicity in cereal rust fungi.

Basic Life Sci. 1976 Mar 1-7;8:205-12. No abstract available.

PMID: 1032100 [PubMed - indexed for MEDLINE]

 

3: Curtsinger JW.

Stabilizing selection in Drosophila melanogaster.

J Hered. 1976 Jan-Feb;67(1):59-60.

PMID: 816847 [PubMed - indexed for MEDLINE]

 

4: Dawson PS.

Directional versus Stabilizing Selection for Developmental Time in Natural and

Laboratory Populations of Flour Beetles.

Genetics. 1975 Aug;80(4):773-783.

PMID: 17248688 [PubMed - as supplied by publisher]

 

5: Ellis WS.

Mortality and birth weight in Philadelphia Blacks: an example of stabilizing

selection.

Am J Phys Anthropol. 1973 Jan;38(1):145-9.

[PhDP]

Lucaspa,

 

I'm happy to see we're going to have a civil discussion

 

Phil, I haven't found any other classification of the forms of natural selection. These are the forms discussed in all the textbooks of evolutionary biology I've seen.

 

It depends of what is discussed and the area of study. In truth, there's just so many ways to classify natural selection;

 

Ecological selection vs. Sexual selection. I don't have the reference with me right now, but this classification must be included in Futuyma's "Evolutionary Biology".

 

Hard vs Solf Selection ("Life History Evolution", Roff, 2002, Rice's excellent "Evolutionary Theory" also covers this);

 

Hard selection = Density- and frequency-independent

Soft selection - Density- and/or frequency-dependent

 

Of course, we often see the oppositions; frequency-dependent vs frequency-independent and density-dependent vs density-independent selection.

 

In molecular evolution, very often, natural selection is simply divided in two forms; positive/advantageous selection (pushing allele frequencies toward fixation) and negative/purifying selection (pushing allele frequencies toward extinction).

 

Also, selection can be classified by units of selection; species selection, group selection, kin selection. It has been often said that group selection did not exist, it's not really true. T. Ryan Gregory wrote an interesting article about this (Macroevolution, hierarchy theory, and the C-value enigma, Paleobiology, 2004). He makes a clear distinction between the old "naive" theory of group selection and a modern perspective.

 

There's just so many other names; genic selection, dominant selection, codominant selection. I'm not saying the classification you used was wrong, it's the standard definition used in most basic textbook on evolution and population genetics. But natural selection can be divided differently.

[lucaspa]

Ecological selection vs. Sexual selection. I don't have the reference with me right now, but this classification must be included in Futuyma's "Evolutionary Biology".

 

Sexual selection isn't natural selection. Two different things. Darwin separated the two. Sexual selection falls under "nonrandom mating" in ways to disrupt a Hardy-Weinberg equilibrium. Natural selection is a different way to disrupt the equilibrium.

 

As it turns out, in many cases natural selection underlies sexual selection. By that I mean that the traits selected for by females are linked genetically to other traits for overall fitness. But the female is not selecting for those overall traits. An example is:

3. E Pennisi, Females pick good genes in frogs, flies. Science 280:1837-1838, (19 June) 1998. Discusses recent studies that show how "bad" genes associated with male display are actually connected to survival genes in males, so that females actually pick survival traits. In frogs the descendents of "long callers" did better in every fitness test.

 

Hard vs Solf Selection] ("Life History Evolution", Roff, 2002, Rice's excellent "Evolutionary Theory" also covers this);

 

Hard selection = Density- and frequency-independent

Soft selection - Density- and/or frequency-dependent

 

Of course, we often see the oppositions; frequency-dependent vs frequency-independent and density-dependent vs density-independent selection.

 

These are not subtypes of natural selection. Instead, frequency dependent and frequency independent selection describe the fitness of the allele in terms of the frequency of the allele in the population

 

"the fitness of a genotype depends on the genotype frequencies in the population" Futuyma, pg 389. Futuyma then goes on to discuss inverse frequency dependent selection where the fitness is greater when the allele is rare. The frequency will then go to an equilibrium. In positive frequency dependent selection, "whichever allele is inititally more frequent will be fixed".

 

Notice that these are processes within the 3 categories of natural selection. Positive frequency dependent selection can move a population to a new trait in a changing environment or, if a population is already established at a fitness peak, keep it there.

 

So, they are not really different forms of natural selection but ways in which natural selection works at the level of alleles.

 

In molecular evolution, very often, natural selection is simply divided in two forms; positive/advantageous selection (pushing allele frequencies toward fixation) and negative/purifying selection (pushing allele frequencies toward extinction).

 

You just changed the definition of "purifying selection". This is too simplistic (why am I not surprised at that from molecular biologists) because it ignores things like heterozygote fitness and equilibria.

 

Also, selection can be classified by units of selection; species selection, group selection, kin selection.

 

That is still not "forms" of natural selection. It is deciding what the units of selection will be for directional, purifying, or disruptive.

 

It has been often said that group selection did not exist, it's not really true. T. Ryan Gregory wrote an interesting article about this (Macroevolution, hierarchy theory, and the C-value enigma, Paleobiology, 2004). He makes a clear distinction between the old "naive" theory of group selection and a modern perspective.

 

Gregory may make a distinction, but that doesn't make group selection exist. It doesn't say anything about the new perspective being any more valid than the old. I have just finished reading that paper and, in fact, Gregory makes many more valid arguments against group selection (particularly species selection) than he does for it. Mayr makes an overwhelming counterargument in What Evolution IS. Species are groups of individuals, and the unit of selection remains the individual. Gregory basically says the same thing in his paper.

 

Gregory tries to make an argument that the genome itself is a unit of selection. I find his argument correlating genome size to body size particularly unconvincing, since one of the the largest genomes (the amoeba) has a very small body size. And no, Gregory is not talking size of cell, but of body size of multicellular organisms. After admitting that both the smallest and largest genomes are represented in protists, he then seems to forget that and look only at multicellular organisms!

 

There's just so many other names; genic selection, dominant selection, codominant selection.

 

Yes, there are different names. But they are not necessarily different "forms" of selection. They are ways of describing different details of how selection is working -- in all 3 forms.

 

By the way, Futuyma does not view frequency independent (what we normally talk about) and frequency dependent selection as different "forms" of natural selection:

 

"Thus a population is not necessarily driven by natural selection to the most adaptive possible genetic constitution." pg 392

[PhDP]

Sexual selection isn't natural selection. Two different things. Darwin separated the two. Sexual selection falls under "nonrandom mating" in ways to disrupt a Hardy-Weinberg equilibrium. Natural selection is a different way to disrupt the equilibrium.

 

I don't really care what Darwin did or said, sexual selection is definitely a form of natural selection. I know some people think otherwise, but my guess is; most evolutionary biologists would agree with me on this. Endler covered this question quite well in the first part of "Natural Selection in the Wild".

 

These are not subtypes of natural selection. Instead, frequency dependent and frequency independent selection describe the fitness of the allele in terms of the frequency of the allele in the population

 

Notice that these are processes within the 3 categories of natural selection. Positive frequency dependent selection can move a population to a new trait in a changing environment or, if a population is already established at a fitness peak, keep it there.

 

So, they are not really different forms of natural selection but ways in which natural selection works at the level of alleles.

 

Yes, there are different names. But they are not necessarily different "forms" of selection. They are ways of describing different details of how selection is working -- in all 3 forms.

 

It's a different way to classify, that's all. Different need = different classification, of course frequency-dependence could be included in the "3 forms" you named. So what ?

 

Much ado about nothing, I don't understand why you get so upset at the notion that natural selection is divided differently according to the discipline. 'Your' 3 forms, they're the standard "types/forms" of population genetics (arguably the core of evolution), and I never disputed that. But I stand on my claim that natural selection can be classified differently, depending on the need.

 

You just changed the definition of "purifying selection". This is too simplistic (why am I not surprised at that from molecular biologists) because it ignores things like heterozygote fitness and equilibria.

 

It's what I'm trying to explain to you, different area of study = different classification. You might think what you want of "positive" selection, but it's a very useful expression in molecular evolutionary biology. When studying long-term patterns of evolution, molecular evolutionary biologists are mostly interested in 'positive'/'negative' selection (and of course, sexual selection is included).

 

Gregory may make a distinction, but that doesn't make group selection exist.

 

It's pretty clear to me that he claims goup selection is a fact;

 

Like organism-level (microevolutionary) natural selection, group selection has been shown to be remarkably effective under experimental conditions (for reviews, see Goodnight and Stevens 1997; Sober and Wilson 1998). However, although these models are theoretically sound and supported by experimental evidence, the overall role of group selection under natural evolutionary conditions has yet to be determined. As outlined below, one of the most significant inputs of group selection to macroevolution may actually have come from its early operation at the subgenomic level.

[lucaspa]

I don't really care what Darwin did or said, sexual selection is definitely a form of natural selection. I know some people think otherwise, but my guess is; most evolutionary biologists would agree with me on this.

 

Actually, most evolutionary biologists disagree. The major textbooks on evolutionary biology -- the ones evolutionary biologists teach from -- view them as separate. Things that mates select can irrelevant to competition for scarce resources.

 

In recent years, some studies have determined that, in some cases, genes for sexually selected traits are linked to genes for adaptive traits:

3. E Pennisi, Females pick good genes in frogs, flies. Science 280:1837-1838, (19 June) 1998.

 

However, in other cases, sexual selection gives traits irrelevent for adaptation:

2. G Arnqvist, Comparative evidence for the evolution of genitalia by sexual selection, Nature 393, 784-786: 1998 (June 25).

 

A good review: 1. LA Dugatkin and J-GJ Godin, How females choose their mates. ScientificAmerican, 278: 56-61, April 1998

 

 

It's a different way to classify, that's all. Different need = different classification, of course frequency-dependence could be included in the "3 forms" you named. So what ?

 

Then it's not an independent classification of how natural selection works.

 

'Your' 3 forms, they're the standard "types/forms" of population genetics (arguably the core of evolution), and I never disputed that. But I stand on my claim that natural selection can be classified differently, depending on the need.

 

You can always "stand by your claim". But you do need some data for your "stand" to be considered valid.

 

 

It's what I'm trying to explain to you, different area of study = different classification.

 

I was talking about your attempt to change the definition of "purifying selection". When you change a definition so that you leave out critical components, then you have made an invalid definition.

 

When studying long-term patterns of evolution, molecular evolutionary biologists are mostly interested in 'positive'/'negative' selection (and of course, sexual selection is included).

 

"Positive/negative selection" tells you the effect of natural selection: some alleles are increasing in frequency (= positive selection) and some are decreasing in frequency (= negative selection). Basically, what molecular biologists have is a shorthand label for the effect of the selection coefficient. I'm very surprised that you would consider this a form of NS.

 

It's pretty clear to me that he claims goup selection is a fact;

 

"Claim" does not necessarily = true. You should know that. The models and data he refers to are still under dispute.

[bombus]

Sexual selection isn't natural selection. Two different things. Darwin separated the two. Sexual selection falls under "nonrandom mating" in ways to disrupt a Hardy-Weinberg equilibrium. Natural selection is a different way to disrupt the equilibrium.

 

I would tend to disagree with that most strongly. Sexual selection is definitiely natural selection. It's not man-made selection so it is natural selection!

 

Natural selection does not just include competiton for scarce resources. Sexual selection produced the peacocks tail, and the plumage of many birds, and badgers stripes, and probably human's big brains.

[SkepticLance]

Sexual selection vs natural selection

 

We could get bogged down here in semantics. I could see valid arguments for separating the two, and other valid arguments for combining the two. Why don't we simply accept the scientific convention, and just regard them as two kinds, in spite of the valid arguments to the contrary?

[ecoli]

I would tend to disagree with that most strongly. Sexual selection is definitiely natural selection. It's not man-made selection so it is natural selection!

I would disagree that not "man made" = natural.

 

Natural selection does not just include competiton for scarce resources. Sexual selection produced the peacocks tail, and the plumage of many birds, and badgers stripes,

Which has little other function than to attract more mates. Interestingly, some sexually-selected traits are negative frequency dependent, so the more common they are, the less they are selected for. (there's also a tradeoff between attracting mates and attracting predators to consider)

 

and probably human's big brains.

I would tend to disagree with that, but won't get into it here. (see "ape evolution" thread)

[CDarwin]

Why don't we simply accept the scientific convention, and just regard them as two kinds, in spite of the valid arguments to the contrary?

 

That's not the scientific convention, as far as I know. Sexual selection is natural selection directly for reproductive fitness. Natural selection can also select indirectly for reproductive fitness by selecting traits that increase the ability to commit resources to baby-making. At the end of the day, though, if you're not reproducing, evolution couldn't care less about you.

 

Perhaps the terminology doesn't work as I thought it did.

[ecoli]

That's not the scientific convention, as far as I know. Sexual selection is natural selection directly for reproductive fitness. Natural selection can also select indirectly for reproductive fitness by selecting traits that increase the ability to commit resources to baby-making. At the end of the day, though, if you're not reproducing, evolution couldn't care less about you.

*or helping others reproduce*

[CDarwin]

*or helping others reproduce*

 

True. If you're not affecting the distribution of genes in a population in any significant way then evolution couldn't care less about you.

[Mr Skeptic]

The trick of course is that "negative" traits show that you are fit enough to survive despite them. Being easy to spot means you have to outrun more predators, so if you're still alive you're probably faster/more agile/better endurance than others. This is similar to how guys tend to do risky/"stupid" things to impress the girls.

 

In the case of birds, doesn't the male plumage also serve to distract a predator from the female and the nest?

[ecoli]

The trick of course is that "negative" traits show that you are fit enough to survive despite them. Being easy to spot means you have to outrun more predators, so if you're still alive you're probably faster/more agile/better endurance than others. This is similar to how guys tend to do risky/"stupid" things to impress the girls.

 

In the case of birds, doesn't the male plumage also serve to distract a predator from the female and the nest?

Yes... it's called a tradeoff.

[PhDP]

lucaspa,

 

Ok, from your comment about me changing the definition of purifying selection, I think the problem is that you fail to make the distinction between quantitative characters and other kinds of characters/genes.

 

In truth, when talking about purifying selection I used a different definition, yes, but it was based on the same idea, applied in a different context (it's why, in the case of molecular evolution we often call purifying selection "positive selection").

 

You use Mather's classifications, this classification is very useful for quantitative characters. But what is purifying / disruptive / directional selection when we're following a single gene ?

 

Most basic textbook on evolution are mostly concerned with quantitative characters, so they make this the standard forms, but it's only for practical reasons. This classification is used to follow the distribution of a trait and how selection acts on this distribution. Mather's definitions might be the most commonly seen in basic textbooks, but it's certainly not the most general classification. While you might think the classification of natural selection based on the selection coefficient might be simplistic, it underlies all of Mather's forms.

 

If you claim there is only ONE way to classify NS, it means you either don't acknowledge basic mendlelian genetics, or you think there's no classification for NS except for quantitative characters.

 

I think you would get into a lot of trouble if you were reading articles with this overtly rigid view of natural selection. First because it simply doesn't work, when following a single mutations you're not going to see Mather's classification. Also, in most textbooks, directional selection and stabilizing selections are defined to be mutually exclusive. In the scientific litterature, however, it's often not the case. In fact, when adopting the most correct definition of stabilizing selection, it become completely independent of directional selection.

 

"Positive/negative selection" tells you the effect[/b'] of natural selection: some alleles are increasing in frequency (= positive selection) and some are decreasing in frequency (= negative selection). Basically, what molecular biologists have is a shorthand label for the effect of the selection coefficient.

 

Yes, sometime it's a very handy definition, and yes, it's based on the selection coefficient. In fact, we could say the same thing about the classifiction you used; it tells you the effect of natural selection; some alleles related to a particular region of the phenotype distribution are pushed toward fixation (= positive selection), and some are pushed toward extinction (= negative selection).

 

So... it's a different classification... It's a classification of NS based on a criterion; selection coefficient. If it's not a classification, what is this ? We take natural selection, make distinctions between subtypes based on a criterion. How do you call this ? You must really use a strange definition of "classification" to refuse to call this a classification.

 

Actually, most evolutionary biologists disagree. The major textbooks on evolutionary biology -- the ones evolutionary biologists teach from -- view them as separate. Things that mates select can irrelevant to competition for scarce resources.

 

I wouldn't quote basic textbooks on evolution as proof of anything, and you certainly don't work in the field, I really wonder how you can say that most evolutionary biologists disagree.

 

In recent years, some studies have determined that, in some cases, genes for sexually selected traits are linked to genes for adaptive traits:

3. E Pennisi, Females pick good genes in frogs, flies. Science 280:1837-1838, (19 June) 1998.

 

However, in other cases, sexual selection gives traits irrelevent for adaptation:

2. G Arnqvist, Comparative evidence for the evolution of genitalia by sexual selection, Nature 393, 784-786: 1998 (June 25).

 

A good review: 1. LA Dugatkin and J-GJ Godin, How females choose their mates. ScientificAmerican, 278: 56-61, April 1998

 

It's completely non sequitur. IMO, you're making a typical mistake, you look at the result, not the mechanism. If we look at the dynamic of a mutant and the forces acting on it, we could makes a pretty clear distinction between drift, mutational pressure and selection. But sexual & ecological selection are basically the same thing, and will have the same kind of effect on the allele.

 

Seen as a diffusion process, they are statistical forces causing drift in the probability distribution of an allele. Can sexual selection give totally different results than ecological selection ? Yes. But it's still the same basic mechanism.

 

What I find completely unnaceptable is that you don't even admit that some scientists believe SS is a different manifestation of NS. I feel strongly about this issue, but I acknowledge that it's mostly semantic. Many scientists think sexual selection is a form of natural selection. I'm going to quote a classic;

 

Darwin made a careful distinction between natural selection and sexual selection: sexual selection is a result of different mating success, including fertilization and pairing. The distinction was made because because traits favored by sexual selection may sometime be disadvantageous, or opposed by other components of natural selection. Thus the outcome, as well as the dynamics, can be quite different from what Darwin and many biologists would regard as "natural selection". Explicit as well as implicit differences of opinion abound on whether or not sexual selection is a subset of natural selection.

 

And he goes to explain why he consider sexual selection a subset of natural selection. So, there's 2 things;

 

#1. Some scientists think SS is a subset of NS, some dont.

 

#2. AFAIK, there's no open debate about this, because it's trivial.

 

Even if I think it makes no sense to consider them separate mechanisms, I can't see a situation where it would really matter.

 

"Claim" does not necessarily = true. You should know that. The models and data he refers to are still under dispute.

 

Well yes, there's a debate about group selection (and sometime it's group selection v. kin selection). I think the theoretical argument is quite valid in favour of some form of group selection, however that was not my point.

[bombus]

Sexual selection vs natural selection

 

We could get bogged down here in semantics. I could see valid arguments for separating the two, and other valid arguments for combining the two. Why don't we simply accept the scientific convention, and just regard them as two kinds, in spite of the valid arguments to the contrary?

 

It's not semantics, it's just that sexual selection IS one facet of natural selection. Others are competion (intra and inter species), environmental factors, behaviour, the list goes on. There is no need to separate out sexual selection, and it's often not easy to that anyway. Natural selection doesn't care about the reason - only if the individuals genes are successfully passed on.