Talk:Evolution/Archive 46

Third Sentence
The third sentence in the intro states,
 * Mutations in genes can produce new or altered traits, resulting in the appearance of heritable differences between organisms.

The implication is that the only, or at least the primary, source of heritable differences between organisms is mutation. But the source of the obvious heritable differences between any two (non-identical-twin) sibling organism is not, so far as I know, mutation. Yet natural selection may favor one sibling over the other. Yes, mutation introduces more radical differences in one generation, but are these radical changes required for evolution to take place? --Unflappable (talk) 16:49, 12 March 2008 (UTC)


 * No, mutation is not required for evolution to occur. Variation is required, but mutation is only one of several sources of variation. This is an illustrative example that is expanded upon in the following sentence, where the other sources of variation are discussed. This is moving from what will probably be familiar to the reader (mutations) to what will probably be the new concepts of the alternative mechanisms. Tim Vickers (talk) 16:53, 12 March 2008 (UTC)


 * Yes, and this is important to understand, because it is rarely one single mutation of one single gene that gives rise to an "advantageous trait" -- it is much more often a new combination of genes that cause a differentially selectable trait to occur. (See eg. Recombination frequency for more on this issue).  —  Wi ki sc ie nt  — 17:02, 12 March 2008 (UTC)
 * Great! That's what I suspected -- that it is much more often a new "combination" of genes that cause a differentially selectable trait to occur -- but could not find anything that made that clear, and certainly not this article.  Can this point be cited and made more clear in the introduction, which currently implies something else?  Here are the two sentences:
 * Mutations in genes can produce new or altered traits, resulting in the appearance of heritable differences between organisms. Such new traits also come from the transfer of genes between populations, as in migration, or between species, in horizontal gene transfer.
 * There is no mention of what is (apparently?) the most common source of new advantageous traits: new "combinations" of genes that result from ordinary sexual reproduction. What is mentioned - mutation, migration and horizontal gene transfer - are all influential, but not the primary sources of new traits.  Right?  --Unflappable (talk) 17:14, 12 March 2008 (UTC)


 * Well, yeah. At least: that's my story, and I'm stickin' with it!
 * Anyone else want to voice a POV before I try to make the recommended changes to the article itself...?
 * — Wi ki sc ie nt  — 17:22, 12 March 2008 (UTC)
 * It would be nice to cite a book or website that makes this clear. --Unflappable (talk) 17:28, 12 March 2008 (UTC)


 * Google.
 * ;) — Wi ki sc ie nt  — 17:48, 12 March 2008 (UTC)


 * In a large population, re-assortment cannot change allele frequencies. From the 'gene-level" view of evolution, which is my favored approach, other genes are part of the environment. Novel traits cannot be produced by reassorting the alleles in a population, all you are doing is expressing a phenotype from alleles that already exist. You need a novel allele for a novel trait. Tim Vickers (talk) 17:43, 12 March 2008 (UTC)


 * I disagree, but have to go at the moment and will not be able to respond further until "later"... — Wi ki sc ie nt  — 17:48, 12 March 2008 (UTC)


 * I disagree too. If combinations of genes define traits, then a new combination can create a new trait.  Consider the apparently unique combinations of genes that transpired to create the geniuses of Mozart and Einstein.  If such combinations, or similar combinations, produce differentially selectable traits that are advantageous, then it follows that natural (or artificial) selection will favor them.
 * On a related note, it's also possible that such combinations are not unique. Consider the "intellectual genius" trait, or the "musical genius" trait, or the "exceptional athlete" trait.  Individuals with those traits (and thus having the combination of genes that create those traits) can be found in the human population who are not necessarily even closely related.  If the environment changes to suddenly favor such a trait, then it follows that natural selection will favor combinations of genes that manifest that trait, and that trait will become more and more prevalent in the population (with successive generations).   Isn't that how giraffe necks keep getting longer?  That is, if the longest neck is 1 meter at generation N, then at generation N+100 the longest neck might be 1.25 meters, without the influence of novel alleles.  Given enough generations, the shorter neck today could be longer than the longest neck in an earlier generation.  Hence the development of the novel "long neck" trait without a mutation-created "long neck" allele. --Unflappable (talk) 19:03, 12 March 2008 (UTC)


 * I've tweaked this a little to merge the sentences, so mutation is not singled out inappropriately. Tim Vickers (talk) 17:55, 12 March 2008 (UTC)


 * That still implies the only or primary way to produce new or altered traits is through mutation, migration or horizontal gene transfer. Again, what about novel traits introduced by new combinations of genes that result from ordinary sexual reproduction?  Isn't that not only another way, but the most commmon way to produce new or altered traits?  --Unflappable (talk) 19:03, 12 March 2008 (UTC)


 * If you have a reproductively-isolated population at Hardy-Weinberg equilibrium how can a new trait appear through reproduction alone? By definition such a population is not evolving. Tim Vickers (talk) 19:10, 12 March 2008 (UTC)


 * A new trait can appear through reproduction alone for the same reason that nobody on earth looks exactly like you (unless you have an identical twin), or has your DNA, or has your fingerprints - which are all unique and novel traits that result from reproduction alone, are they not? In reality, is an environment really ever stable enough to manifest a Hardy-Weinberg equilibrium?  And note that when it's close to that, species do basically stop evolving.  Isn't that why some sharks have not evolved for millions of years?  Whether variation in traits is introduced by mutation or genetic drift through reproduction alone (see my question below), if those novel traits are not advantageous natural selection will not favor them and the population will remain relatively stable genetically. --Unflappable (talk) 19:33, 12 March 2008 (UTC)


 * This article might make the definition of Hardy-Weinberg equilibrium, and why it answers your questions, a bit clearer for you. Tim Vickers (talk) 19:44, 12 March 2008 (UTC)

Well, right off the top that article states:

What I'm saying is that if the dominant phenotype is advantageous, then the BB and Bb offspring are preferred by natural selection, while if the recessive phenotype is advantageous, then the bb offspring are preferred. In the latter case, we would expect to find more and more bb members in the population, and eventually there may be no BB or Bb members left at all. That's evolution by natural selection, is it not? And not a single mutation required. Just normal sexual reproduction and natural selection. Note that this article only applies to advantageously neutral phenotypes. It states that H-W fails when natural selection applies. That is, when either the dominant or recessive trait is preferred by natural selection. From the article you suggested I read (thank you):


 * To see what forces lead to evolutionary change, we must examine the circumstances in which the Hardy-Weinberg law may fail to apply. There are five:


 * Natural Selection


 * If individuals having certain genes are better able to produce mature offspring than those without them, the frequency of those genes will increase. This is simple expressing Darwin's natural selection in terms of alterations in the gene pool. (Darwin knew nothing of genes.) Natural selection results from


 * differential mortality and/or
 * differential fecundity.


 * Mortality Selection


 * Certain genotypes are less successful than others in surviving through to the end of their reproductive period.


 * The evolutionary impact of mortality selection can be felt anytime from the formation of a new zygote to the end (if there is one) of the organism's period of fertility. Mortality selection is simply another way of describing Darwin's criteria of fitness: survival. Link to an example of powerful mortality selection in a human population causing a marked deviation from Hardy-Weinberg equilibrium.


 * Fecundity Selection


 * Certain phenotypes (thus genotypes) may make a disproportionate contribution to the gene pool of the next generation by producing a disproportionate number of young. Such fecundity selection is another way of describing another criterion of fitness described by Darwin: family size.


 * In each of these examples of natural selection certain phenotypes are better able than others to contribute their genes to the next generation. Thus, by Darwin's standards, they are more fit. The outcome is a gradual change in the gene frequencies in that population.

"The outcome is a gradual change in the gene frequencies in that population." That's evolution, is it not? None of this is explained in the article, so far as I can tell. --Unflappable (talk) 23:07, 12 March 2008 (UTC)

Yes, if an allele exists that produces an advantageous or disadvantageous trait in a population, then natural selection can change the frequency of that allele. However, if no mutation occurs and no forces act on allele frequencies, the organisms will continue to breed and produce offspring with new combinations of the existing alleles, but over time allele frequencies remain constant - no new alleles appear and the old alleles remain in a constant proportion to each other. This is the definition of H-W equilibrium. If you think this isn't explained in the article than you simply can't have read it. Tim Vickers (talk) 23:13, 12 March 2008 (UTC)
 * But H-W equilibrium, as I understand it, assumes an environment so stable that natural selection for genetic variation due to any cause is not happening. That's not evolution; that's non-evolution.  There have to be environmental forces of natural selection acting on the phenotype driven traits in order for evolution to occur. And, to be clear, my beef (at least so far) is with the introductory paragraphs, not with the whole article.  In this case it's the third sentence of the intro.   --Unflappable (talk) 00:47, 13 March 2008 (UTC)

If producing new combinations of genes through sexual reproduction did introduce genetic variation and new traits in a population, then a sexual species could never be at HW equilibrium - since new variation would be constantly produced they would be constantly "evolving" under that definition. However, all sexual reproduction does is shuffle the alleles around. Think about sexual reproduction like shuffling and dealing a pack of cards, with the population the entire pack and each organism a hand of cards. If all you do is shuffle, deal and replace the cards in the pack, the pack stays the same. No new cards appear and the frequency of one particular card does not change. This is a breeding population in HW equilibrium that isn't evolving. However, if you start to discard cards out of the pack - a form of selection - then the pack will change over time. Tim Vickers (talk) 01:04, 13 March 2008 (UTC)
 * Per your logic, if producing new combinations of genes through mutation did introduce genetic variation and new traits in a population, then a species that experienced mutation could never be at HW equilibrium - since new variation would be constantly produced they would be constantly "evolving" under that definition. That's absurd, of course, for the same reason the same assertion using sexual reproduction is absurd: one of the necessary requirements of the HW equilibrium is the neutralization of natural selection - and that's the reason it stays at HW equilibrium despite the production of new combinations of genes and introduction of genetic variation (regardless of the source).
 * You see, even though sexual production and mutation introduce genetic variation, a species that reproduces sexually and experiences mutation can be at HW equilibrium, as long as the new variation -- whether introduced through mutation or sexual reproduction -- is not advantageous through natural selection.
 * While "all" (as if that's no big deal!) sexual reproduction does is shuffle the alleles around (no argument there), shuffled alleles arranged in new combinations create differentially selectable traits that may have not been expressed before. Again, that's why your look is unique. Your looks are not the only novel heritable traits that you possess, all through sexual reproduction. Same with all other member of our species, and all species. No?   I guess what I'm saying is that mutation plays a role, of course, it adds a whole new dimension to evolution, but since sexual reproduction alone creates differentially selectable traits (expressed genotypes), that is enough to have evolution (assuming an environment in which some traits are advantageous in order to avoid HW equilibrium, but this is necessary to have evolution caused by mutation as well).  --Unflappable (talk) 01:26, 13 March 2008 (UTC)
 * Here is another point that supports what I'm trying to say: Because of sexual reproduction, "beneficial mutations from separate ancestries can be combined" link. If sexual reproduction alone (without mutation) can combine beneficial mutations from separate ancestries, and presuming new traits can be expressed as a result of these new combinations, then it follows that sexual reproduction alone produces new differentially selectable traits.  This method of introducing new and altered traits into a population is arguably the primary driver for evolutionary change, and probably why sexual reproduction is so successful relative to asexual reproduction, and yet is unmentioned in the introduction of this article, despite the inclusion of arguably less significant sources of trait introduction and alteration.  --Unflappable (talk) 01:48, 13 March 2008 (UTC)

Proposal: mods to first and third paragraphs
Okay, I'm proposing changing the first three paragraphs per the discussions above. Here is the current state:

"In biology, evolution is the changes seen in the inherited traits of a population from one generation to the next. These changes are relatively minor from one generation to the next, but accumulate with each subsequent generation and can eventually cause substantial changes in the organisms. Inherited traits come from the genes that are passed on to offspring during reproduction. Mutations in genes can produce new or altered traits, resulting in the appearance of heritable differences between organisms, but new traits also come from the transfer of genes between populations, as in migration, or between species, in horizontal gene transfer. Evolution occurs when these heritable differences become more common or rare in a population, either non-randomly through natural selection or randomly through genetic drift.

Natural selection is a process by which heritable traits that are helpful for survival and reproduction become more common in a population, while harmful traits become more rare. This occurs because individuals with advantageous traits reproduce more successfully, so that more in the next generation inherit these traits. Over many generations, adaptations occur through a combination of successive, small, random changes in traits, and natural selection of those variants best-suited for their environment. In contrast, genetic drift produces random changes in the frequency of traits in a population. Genetic drift arises from the role chance plays in whether a given individual will survive and reproduce.

One definition of a species is a group of organisms that can reproduce with one another and produce fertile offspring. When a species is separated into populations that are prevented from interbreeding, mutations, genetic drift, and the selection of novel traits cause the accumulation of differences over generations and the emergence of new species. The similarities between organisms suggest that all known species are descended from a common ancestor (or ancestral gene pool) through this process of gradual divergence."

And here is what I propose to have it say instead. Most of the changes are in the first paragraph, none in the 2nd, and just a minor one in the third.

"In biology, evolution is the changes seen in the inherited traits of a population from one generation to the next. These changes are relatively minor from one generation to the next, but accumulate with each subsequent generation and can eventually cause substantial changes in the organisms. Inherited traits come from the genes that are passed on to offspring during reproduction. In species that produce sexually, new and altered traits are produced by combining genes inherited from separate ancestries in new combinations, resulting in the appearance of heritable differences between organisms. Other ways that new or altered traits are produced include mutations in genes, the transfer of genes between populations, as in migration, or between species, in horizontal gene transfer. Evolution occurs when these heritable differences become more common or rare in a population, either non-randomly through natural selection or randomly through genetic drift.

Natural selection is a process by which heritable traits that are helpful for survival and reproduction become more common in a population, while harmful traits become more rare. This occurs because individuals with advantageous traits reproduce more successfully, so that more in the next generation inherit these traits. Over many generations, adaptations occur through a combination of successive, small, random changes in traits, and natural selection of those variants best-suited for their environment. In contrast, genetic drift produces random changes in the frequency of traits in a population. Genetic drift arises from the role chance plays in whether a given individual will survive and reproduce.

One definition of a species is a group of organisms that can reproduce with one another and produce fertile offspring. When a species is separated into populations that are prevented from interbreeding, sexual reproduction, mutations, genetic drift, and the selection of novel traits cause the accumulation of differences over generations and the emergence of new species. The similarities between organisms suggest that all known species are descended from a common ancestor (or ancestral gene pool) through this process of gradual divergence."

--Unflappable (talk) 02:08, 13 March 2008 (UTC)

We could mention the role of recombination in shuffling alleles, but sexual reproduction doesn't produce new traits, and mating (as long as it is random) has no effect on speciation. Tim Vickers (talk) 03:50, 13 March 2008 (UTC)


 * Tim, I don't know how you can say that sexual reproduction does not produce new traits. The ever lengthening neck of giraffes was produced by natural selection and sexual reproduction, was it not?   Isn't the long neck of the giraffe a trait?
 * With regard to sexual reproduction not causing speciation, what about ring species? If a connecting population in a ring species dies out, then you're left with two non-breeding populations - two separate species.  And when populations are separated, say on the Galapagos, are you saying that it is only mutation that causes them to no longer be able to mate with each other after sufficient numbers of generations?  All that has to happen is for the populations to be isolated and for sufficient genetic change to occur for them to no longer be able to breed.  Note that in humans, there are certain males and certain females who already cannot successfully breed with each other due to genetic incompatibilities.  Each can successfully breed with the majority of the opposite sex, but not with a certain minority of the species, due to some genetic incompatibility (e.g., blood type).  The longer two populations are isolated, the more members in each population become reproductively incompatible with more members in the other population, until some undefined threshold is crossed where we might call them separate species. And, so far as I know, mutation is not required for this divergence and reproductive incompatibility to take place.  In fact, genetic drift alone might be sufficient to cause it, according to that article.  Finally, speciation has been produced artificially, and repeatedly, in as few as eight generations of isolated sexual reproduction.  --Unflappable (talk) 17:34, 13 March 2008 (UTC)
 * Sexual reproduction does not per se produce new genoytpes. What it does is it redistributes genotypes.  During sexual reproducton, new mutations can occur through transcription errors, but this is viewed as a kind of mutation and not a principle effect of sexual reproduction.  Again, you seem to be confusing changes in gene frequences with the actual creation of new traits. Slrubenstein   |  Talk 17:43, 13 March 2008 (UTC)


 * Sexual reproduction does not alter allele frequencies. As I explained earlier it is like shuffling a pack of cards - the numbers of cards and the chance of being dealt an ace doesn't change no matter how often you shuffle the pack. If two populations can interbreed, they are the same species. If you have a population of sexual organisms that is divided for a while into two identical isolated populations and they continue to breed within these new populations, no change in allele frequencies will occur due to this breeding. Since there is no change in the alleles of the population, evolution has not occurred, and when the populations come back into contact with each other they will still be able to interbreed. For speciation to occur, evolution must occur, and that requires a force that can change allele frequencies - such as drift or selection. Tim Vickers (talk) 17:46, 13 March 2008 (UTC)


 * As a note, I've added some material to the section on recombination that explains the difference between sexual reproduction's effects on individuals, and its effects on populations. Tim Vickers (talk) 17:56, 13 March 2008 (UTC)


 * You are right Tim, what I wrote was sloppy, Slrubenstein  |  Talk 19:04, 13 March 2008 (UTC)

Recombination section

 * No probs. Could you have a look over the recombination section, Slrubenstein, to see if I managed to explain the ideas at all clearly? Tim Vickers (talk) 19:10, 13 March 2008 (UTC)
 * I'm a little uncertain about your line about positive selection - variation in LD is due mostly to variation in recombination rate, but the relevant quantity when discussing positive selection is the short coalescent time (that is, the long haplotype block results from a very short history, which precludes the occurrence of recombination). So the inclusion of positive selection in the brief mention of LD is a bit of a non-sequitur here. Also, the quoted figure for rate is wrong - the rule of thumb is "one recombination event per chromosome arm per generation", which in the human genome works out to about 10^-8 per base per generation. See the hapmap paper for figures. Graft | talk 20:21, 13 March 2008 (UTC)
 * Good catch, thank you. I think the mistake was confusing a 1 cM/Mb with one crossover per Mb. Yes, we could write a bit more about why the haplotype signature occurs, but I'm tempted to simply remove it as confusing. What do you think? Tim Vickers (talk) 20:45, 13 March 2008 (UTC)
 * I feel like LD is an obscure enough topic that it should go on positive selection rather than in here. The reader of this article doesn't need to know how to detect positive selection. Graft | talk 16:04, 14 March 2008 (UTC)

Slrubenstien, I don't think I've said anything about sexual reproduction producing new genotypes. Yes, sexual reproduction redistributes genetic material which creates and alters differentially selectable phenotypes (expressed traits), thus leading to evolution and potentially speciation without necessarily changing any genotypes through mutation. No?

Tim, I have not stated that sexual reproduction alone changes allele frequencies. I am saying that it is my understanding that sexual reproduction combined with selection and/or drift can, even without any assistance from mutation, change allele frequencies, and can do so enough to cause speciation, and I gave specific examples of how this happens (giraffe necks getting longer, Galapagos, ring species, and artificial speciation). --Unflappable (talk) 19:33, 13 March 2008 (UTC)

Thanks for doing the update. That helps. Now, since you concede that sexual reproduction affects populations, and these effects are presumed to be differentially selectable, why do you insist that evolution cannot occur from sexual reproduction and selection alone, that mutation is also required? In short: isn't each individual an example of evolution, as compared to his parents and siblings? I assume we agree the only difference between "microevolution" and "macroevolution" is a matter of degree (number of generations). Similarly, isn't the only difference between sexual reproduction and "micro evolution" also a matter of degree (number of generations; namely one vs. enough to notice discernible differences in the population)? Isn't that what is meant when we say we are constantly evolving? That we evolve with each generation? --Unflappable (talk) 19:33, 13 March 2008 (UTC)


 * The process of evolution by natural selection involves two classes of mechanisms - those that introduce new alleles into a population and those that remove them. Evolution is simply a change in allele frequencies, so all you need in the short-term is a force that can alter the frequency of an allele in a population - usually either drift or selection. Both of these processes act to eliminate alleles from a population and decrease variation. Over time, a population experiencing drift or selection (an evolving population) will become more homogeneous. Sexual reproduction will not offset this process, since all it does is mix the existing pool of alleles. You can't get new versions of genes through sex - only new combinations of the existing alleles. For evolution to continue in the longer term, you need processes that introduce new alleles into the population. These mechanisms are mutation, horizontal gene transfer and gene flow from other populations. Tim Vickers (talk) 19:47, 13 March 2008 (UTC)


 * It's easy to imagine a scenario in which sexual reproduction (i.e., recombination) produces new phenotypes - though it won't produce new variation in the strictest sense, it will produce new alleles in the classical sense (where we're considering, say, genes), and these can of course have a HUGE phenotypic effect. And in fact this is extremely important for evolution, since it allows you to dump a lot of your genetic load (by combining multiple deleterious variants into a single set and losing them at once), which lets you get by with much smaller population sizes. Humans and other sexually-reproducing organisms would not be viable as an asexual population - the mutation load would be simply too high. We probably have an article on this subject... ah, here. Graft | talk 20:37, 13 March 2008 (UTC)


 * Yes, epistasis means that new combinations of genes can have new phenotypes, but as you say, since this doesn't involve the introduction of new alleles of genes, this isn't introducing variation at a population level, only at the level of an individual's offspring. Tim Vickers (talk) 20:43, 13 March 2008 (UTC)


 * No, I'm saying this CAN produce new alleles of genes. I.e., if I have mutation A at position 1 and mutation B at position 10, each of which individually do nothing, but when recombined into the same gene produce a dramatic effect on fitness, this is a new allele, and this IS new variation at a population level. It's just not variation at the level of SNPs, but why should we pretend that all genetic variation is functionally independent? It's obviously not, and epistasis is an important consideration in evolution - in fact epistatic interactions might be driving a huge part of evolutionary changes. And recombination has implications for the total deleterious mutation load in an individual genome, which is obviously important to evolution. I'm not sure I agree with the suggested phrasing, but sex definitely contributes to variation in fitness, and that absolutely should be stated. Graft | talk 20:52, 13 March 2008 (UTC)

Yes, I suppose in the rare cases where a recombination event hits a gene that contains SNPs and produces a recombinant then you have a novel allele without introducing a new mutation, but as genes are well-spaced by flanking sequences that will be an extremely rare event and will occur at a very low frequency indeed. In most crosses, all that happens is existing alleles being made into new combinations. Such a recombination event also does not change the information content of the genome, so can't be described as introducing increased genetic diversity at the genomic level, all it is doing is moving existing variation around. At the moment we have this discussion of the effect of recombination on fitness. Tim Vickers (talk) 21:24, 13 March 2008 (UTC)


 * I'm not sure I agree with the suggested phrasing either, and it's my phrasing! But I think it's important to be clear in the introduction of the evolution article that evolution ultimately means huge changes resulting from many generations of tiny changes, including just the relatively small but differentially selectable changes that result in one generation from normal sexual reproduction, and that the ultimate result of these changes is non-random due to natural selection. I do not think this point is clearly made in the current wording at all. --Unflappable (talk) 21:07, 13 March 2008 (UTC)


 * You're still missing the point that the variation produced from a sexual cross in a single pair of organism's offspring does not change allele frequencies or introduce novel sequences. Imagine a population composed entirely of clones. If they breed sexually for a couple of generations they will remain genetically identical. Sex doesn't introduce variation, it just mixes what you have. Tim Vickers (talk) 21:24, 13 March 2008 (UTC)


 * You're considering a definition of "variation" that is unnecessarily strict. You're wrong when you suggest (above) that recombination in genes is so rare as to be unimportant - genes typically span at least tens of kilobases - remember introns - and recombination events between a pair of SNPs in a single gene could easily occur in the typical lifetime of most slightly-deleterious or neutral variation. But when we say "allele", in the modern context we frequently refer to a single base-pair. Back in the day, an allele was a variant of an entire gene - thus, a particular variant could definitely be introduced by the act of recombination. Also, sex DOES introduce variation in fitness, regardless of whether it introduces "genetic variation", however we define it, via "new and advantageous gene combinations", and also, let us not ignore, new and deleterious gene combinations. Graft | talk 21:44, 13 March 2008 (UTC)


 * Yes, a sexual cross does produce phenotypic variation, and certainly new allele combinations, but all these effects depend on there being variation in the first place for reassortment and recombination to mix around - sex isn't the ultimate cause of variation but it does influence the selective process on the variation that is there. What I don't agree with in the proposal above is the idea that sex is itself a source of variation and that sexual reproduction could cause speciation. I changed the recombination section a bit more and retitled it to "Sex and recombination", since it discusses reassortment just as much as recombination. Tim Vickers (talk) 22:00, 13 March 2008 (UTC)

Tim, well, if your point is that sexual reproduction alone, and even with selection, cannot cause variation and thus evolution starting from scratch, yes, of course. But, this article is not about the origin of life. It's about the evolution of existing life. We're long past the point of starting from scratch. What has occurred since variation was introduced is what is usually thought of in terms of "evolution". Once variation is introduced into a population (by whatever means), sexual reproduction in and of itself, along with selection of course, causes more variation and evolution, including speciation. That's the point, which I think is critical to understanding evolution, that is missing from the introduction of this article. Mutation just accelerates the rate of evolution now and then, by bringing about more fundamental underlying changes. But evolution including speciation can and does happen without mutation. Evolution caused by selection on variation resulting from sexual reproduction is arguably the driving force in evolution. The underlying technical details of how all this happens covered later in the article are interesting, but I think it's more important to get the basic high level concepts right up front in the introductory paragraphs. --Unflappable (talk) 23:29, 13 March 2008 (UTC)


 * Do you understand the point made in the second section quoted above that sex can under some circumstances reduce the genetic variation within a population? Tim Vickers (talk) 23:47, 13 March 2008 (UTC)


 * Yes. But just because the underlying variation in available alleles is reduced doesn't prevent the combinations of alleles that produce differentially selectable expressed traits to increase, which is ultimately what natural selection acts upon. Reduction in genetic variation is like going from base 16 to base 10 - what matters are variations in the expressed traits that are actually manifested, not the underlying genetic variation.  The software on your computer and on my computer and on all the computers between us, including those hosting this discussion, are all systems whose fundamental building block can only be expressed in one of two ways: a 0 or 1.  You basically can't have any less underlying variability than that.  Yet the number of variations that can be expressed in combinations of various lengths of strings of 8, 16, 32 or 64 of these bits are infinite.  So pardon me for being underwhelmed by the fact that under some circumstances sex can reduce underlying genetic variation within a population.  As long as there is some underlying variance genetically, that's enough to ultimately produce infinite variance phenotypically.
 * To underscore the irrelevance of underlying genetic variation, consider that humans have 23 pairs of chromosomes, barley has 14 and algae (!) has 148. The fruitfly has only 8 but has been used to demonstrate speciation (without mutation, I believe - would like to confirm that) with artificial selection in as few as 8 generations.  I think you're putting far too much emphasis on the importance of underlying genetic variation in evolution, or at least not enough emphasis on the importance of sexual reproduction and its role in creating and altering the differentially selectable expressed traits on which natural selection operates  --Unflappable (talk) 00:27, 14 March 2008 (UTC)


 * I would disagree with this. Without mutation, reproduction would quickly eliminate variation. Variation has a finite lifetime in ANY population, due to drift (or selection). This means evolution would stop after a number of generations.
 * Also, sexual reproduction cannot lead to speciation, since by the "biological" definition an interbreeding population of individuals IS a species. How can sexual reproduction lead to barriers to itself? Graft | talk 00:10, 14 March 2008 (UTC)


 * That's why isolation is always required for speciation. If you isolate one population from another from the same species, with enough generations and the influences of selection and drift, you will have speciation.  Mutation is not required for speciation to occur, only isolation and evolution by natural selection operating on the variation in expressed traits introduced by sexual reproduction at each generation is all that is required, as I understand it. --Unflappable (talk) 00:27, 14 March 2008 (UTC)


 * Also, I don't understand why you say reproduction would quickly eliminate variation without mutation. Perhaps so in a stable environment, but that's true with mutation in the picture too (like sharks), but as long as the environment keeps changing, natural selection will continue doing its thing, and various genetic combinations (including newly expressed ones, for which there are infinite possibilities) and their corresponding expressed traits will become more or less prevalent in the population.  --Unflappable (talk) 00:42, 14 March 2008 (UTC)
 * See genetic drift. Graft | talk 00:47, 14 March 2008 (UTC)


 * Per genetic drift: "genetic drift ... is the evolutionary process of change in the allele frequencies (or gene frequencies) of a population from one generation to the next due to the phenomena of probability ... the biological traits that it confers, to become more common or rare over successive generations, and result in evolutionary change over time. ... now held to be one of the primary mechanisms of biological evolution". And what causes genetic drift?  Sexual reproduction.  So, we have
 * sexual reproduction → genetic drift → biological evolution
 * We also have:
 * natural selection (variation in phenotypes from sexual reproduction) → biological evolution
 * In both cases evolution, along with population isolation and sufficient numbers of generations along with environmental forces, may lead to ring species, and speciation, without mutation. No?  --Unflappable (talk) 04:38, 14 March 2008 (UTC)

You clearly do not understand drift - it is not "caused" by sexual reproduction - indeed, the colloquial definition of the word "cause" does not neatly apply to many scientific phenomena ... it sound slike you do not understand evolution either. Drift is an effect of sampling error. It is greatest in small populations, and least in large populations, and you could just as well say population size is a "cause" of drift. In any event, drift is a statistical phenomenon and we are better off not talking about causes at all. If you ant to talk about events that are relevant to drift, yes, sexual reproduction can be one. So can being killed by a large anvil falling out of the sky. So can a whooping cough epidemic. Now, do you have an improvement to suggest? So far all of your suggestions would move the article from being accurate to being inaccurate. If you think the article is unclear or misleading, tell us exactly what sentence is unclear or misleading and why, please. Slrubenstein  |  Talk 12:26, 14 March 2008 (UTC)
 * If each member of a species is represented by a deck of cards, and the "signature" of a population is determined by the sorted order of the top card of each member's deck, then, if the cards are randomly shuffled periodically, you will have varying signatures if the populations are small, and little variance if the populations are sufficiently large.  The stochastic  variance that results from random shuffling is drift.  In that sense the random shuffling is what causes the drift, and the random shuffling in biological evolution is sexual reproduction.  That's what I meant when I said sexual reproduction "causes" drift.  Do you think someone who "clearly does not understand drift" could explain by analogy like this?

Quite obviously, yes. Slrubenstein  |  Talk 16:59, 14 March 2008 (UTC)
 * Perhaps I've been reading too much Dawkins, but he explains things by analogy - and it works for me. This article does not work for me.  It seems to get too bogged down in the technical details at the cost of confusing and misrepresenting the fundamental high-level concepts.
 * I have listed specific statements that I find to be misleading or wrong, explained why, and have made alternative suggestions (see the previous 3 or 4 sections).
 * Maybe we should take a step back and think about what evolution means in general - not specifically in biology. Evolution occurs in any system in which there is change over time.
 * All that is needed for evolution by selection and/or drift to occur is for there to be change in differentially selectable traits at each generation. What brings about those changes is irrelevant to that point.  But, the causes of these changes include the random shuffling of sexual reproduction, as well as mutation and introduction of new genes due to migration, etc.  Since change (in differentially selectable expressed traits) from mutation is relatively rare, and change from the shuffling of sexual reproduction is constantly present, I find it inaccurate for the latter to not be mentioned in the introduction, while many (all?) of the other mechanisms of change are mentioned.  Yes, sexual reproduction is mentioned, but not as a source of change upon which evolution acts.  --Unflappable (talk) 16:03, 14 March 2008 (UTC)
 * No, that isn't anywhere close to explaining genetic drift. Since you seem happiest dealing with analogies, imagine you have a purse containing 50 marbles, half red and half blue. In each round of the game you throw away (at random) half of the marbles and you are given replacement marbles of the same color as the ones that remain. Since the numbers of red and blue marbles you throw away will fluctuate, so will the composition of the population slowly change over time, sometimes more red, sometimes more blue. It is even possible that you may, by chance lose all the red marbles and be left with a bag of blue marbles. This is genetic drift - chance fluctuations in which genes survive, leading to chance fluctuations in the allele frequencies of the population. This mechanism applies to both sexual and asexual reproduction. Tim Vickers (talk) 16:12, 14 March 2008 (UTC)
 * Does the purse represent an individual member, a population or an entire species? --Unflappable (talk) 16:37, 14 March 2008 (UTC)
 * The purse is the population and the marbles a gene locus, with the different colors different alleles of the gene. Tim Vickers (talk) 16:54, 14 March 2008 (UTC)
 * Ah. So in a small population the purse is small and holds a relatively small number of marbles, thus the likelihood of exchanging all of the reds for blues is relatively high, while the larger the purse (population), this becomes less and less likely.  But where are the individuals?  Where are the expressed traits upon which evolution acts?  I see that you are viewing variation within the population as the alleles themselves, independent of arrangement (or particular combinations, if you will).  But it's their arrangement that determines the practically infinite variety of expressed traits upon which evolution acts.  I don't see that represented in your analogy.  --Unflappable (talk) 18:55, 14 March 2008 (UTC)
 * Phenotypes are irrelevant to the process of genetic drift. An allele could be neutral and still be eliminated or fixed by this entirely random process. Equally, in small populations where drift may dominate, disadvantageous alleles can be fixed by chance. Tim Vickers (talk) 19:03, 14 March 2008 (UTC)
 * Of course drift also affects neutral alleles, but its the effect it ultimately has indirectly on phenotypes that makes it significant to evolution (and thus to this article). Similarly, it's the effect that sexual reproduction shuffling has on altering and producing new phenotypes that makes that process significant to evolution (and thus to this article). Am I missing something? --Unflappable (talk) 19:12, 14 March 2008 (UTC)
 * So do you now see why people objected to the idea that sexual reproduction causes genetic drift? Tim Vickers (talk) 19:24, 14 March 2008 (UTC)
 * Not completely. I understand that the variation in allele frequencies is inversely related to the population size, and why, but the same random shuffling and recombining of alleles (due to sexual reproduction) that produces new allele frequencies in small populations and averages out to the same stable ratios in larger populations must also produce new and altered phenotypes, some of which are not differentially selectable.  So it seems to me that drift would have an effect in even large populations (in terms of phenotype expression, not change in allele frequencies).  Anyway, I still think it's not inaccurate to say that it's the shuffling from sexual reproduction, along with other factors, that causes genetic drift.  --Unflappable (talk) 20:00, 14 March 2008 (UTC)
 * Drift is not the shuffling of the deck or any effect of shuffling the deck. Drift is the effect of some cards - at random - disappearing from the deck entirely. If you remove 25 cards from a 52 card deck, the effect on the hand dealt will be noticible very quickly and will radically alter our ability to play most games. If you remove 25 cards from a 5000 card deck, the effect on the hand dealt will be slight and will not seriously affect the ability to play a game. The key thing is that the cards that disappear do so randomly. Sexual reproduction is one of many possible factors, but even here numbers count. If a couple has four children, it is conceivable that there will be no drift for a given trait. If they have ten or twelve kids, the drift may well be minimal. If they have one kid, the drift will be higher - that is, if they are in a small population. In any event, sexual reproduction is but one factor, falling anvils, reckless drivers, lightening also "cause" drift. Slrubenstein  |  Talk 20:12, 14 March 2008 (UTC)

Summary of my objection to the intro to this article
My objection to the introduction of this article, and particularly the first paragraph, is not only that it discounts the role of sexual reproduction in evolution, but it totally ignores it (as a source of change upon which evolution acts). Sexual reproduction is what shuffles and thus recombines the genes that ultimately determine the differentially selectable traits upon which selection acts, and by which drift is defined. Sexual reproduction shuffling alters existing differentially selectable traits, and creates new ones. In other words, if mutation suddenly stop happening sexually reproducing populations would never-the-less continue to evolve by selection (as long as the environment changed) and drift (if the population is sufficiently small), simply because of the variance in differentially selectable traits produced by sexual reproduction. This is fundamental to evolution and the introduction, particularly the first paragraph, as currently written, implies otherwise. --Unflappable (talk) 16:25, 14 March 2008 (UTC)


 * This is wrong in almost every respect. But I do agree with the general point that a mention of sex and its effect is appropriate in the intro. I'll give it a shot... Graft | talk 16:54, 14 March 2008 (UTC)
 * I changed my mind. The Intro is far too general for this to be appropriate. Graft | talk 16:59, 14 March 2008 (UTC)
 * Would you mind identifing a statement, assertion or even implication in what I wrote that you think is wrong? These are the fundamental points:
 * Selection acts on differentially selectable expressed traits.
 * Gene combinations are what determine the differentially selectable traits upon which selection acts.
 * Sexual reproduction shuffles the genes from both parents and creates combinations of genes that alter existing and produce new differentially selectable traits.
 * Do you disagree with any of these statements? If so, which one(s) and why? Thanks. --Unflappable (talk) 18:46, 14 March 2008 (UTC)

I added a sentence to the first paragraph, but we can't say simply "increase genetic diversity" since under some circumstances sex will decrease diversity, that's why I used the slightly more general "can further increase the variation between organisms." What do you think Graft? Tim Vickers (talk) 17:07, 14 March 2008 (UTC)


 * I just noticed this Thanks. It goes a long way towards what I'm looking for. Here is the sentence as it currently stands:


 * In species that produce sexually, new combinations of genes are produced by genetic recombination, which can further increase the variation between organisms.


 * The "can further increase" seems to downplay the profound effect it actually has. Can we say:


 * In species that produce sexually, new combinations of genes are produced by genetic recombination, which increases the genetic variation within a species, and may account for most of the genetic variation in sexually reproducing species.


 * Note that the "increases the genetic variation with a species" assertion, which you seem reluctant to state, is supported by the PBS FAQ quote below. --Unflappable (talk) 19:42, 14 March 2008 (UTC)

Here, from a PBS FAQ on evolution, is another expression of what I think the introduction does not say, and actually implies is not the case:

In short, this concept, which is fundamental to evolution, is not conveyed in the introduction:


 * sexual reproduction → shuffling/recombining → new combinations of genes → increase in genetic variation → evolution

--Unflappable (talk) 19:24, 14 March 2008 (UTC)


 * We can't say that since it is wrong. Sex increases diversity in some instances, but decreases it in others. See and this paper. Tim Vickers (talk) 19:44, 14 March 2008 (UTC)


 * I haven't tuned-in to this discussion in a while, and I haven't read too much of what's been discussed since then, but I think I see the point you're trying to make here, Unflappable.
 * And I think I agree with it, too. But the wording of this issue is very sensitive.  (Eg.: maybe something more like "→change in relative distribution of genes→" above instead of "...→increase in genetic variation→...", etc?)
 * I think the problem is more in how to phrase the point you're trying to make, though, than with the point itself. I wonder if it might help if some of the others (who have apparently been involved in the article for a while), could try to look at the "conflict" here in those terms?
 * — Wi ki sc ie nt  — 19:55, 14 March 2008 (UTC)

The idea that sex increases genetic variation does have some recent experimental support in some model systems, eg, but this isn't enough data to say it will do this in all cases - this is why I phrased that sentence so carefully. Tim Vickers (talk) 19:57, 14 March 2008 (UTC)


 * What I think he's saying, though, is that the same random-walk in gene assortment that accounts for genetic drift can also give rise, randomly, to certain gene-combinations that are the selectively advantageous -- ie. on which natural selection can then operate the same way it would on gene-differences that are due to eg. mutation or cross-over events -- right...?
 * — Wi ki sc ie nt  — 20:04, 14 March 2008 (UTC)
 * That's exactly how I see drift connected to this. --Unflappable (talk) 20:10, 14 March 2008 (UTC)


 * Tim, again, I think you're continuing to discount the profound and fundamental role that sex has in increasing genetic variation (in terms of phenotypes). Mark my words (and those of PBS).  ;-).  Also, isn't your objection to saying that sex increases diversity since it sometimes can decrease diversity a bit like saying we can't say "food is nutritious" because sometimes food can be bad for you?  --Unflappable (talk) 20:10, 14 March 2008 (UTC)


 * No, see, now, Unflappable, there really is a phrasing issue here. I think it can be worked out, but I don't think your comment immediately preceding is helping to move this in that direction....
 * — Wi ki sc ie nt  — 20:15, 14 March 2008 (UTC)
 * Also, the problem is Unflappable that your changes would imply that any change can result from recombination caused by fusion of gametes. Humans aren't going to suddenly appear with blue hair due to crossing over, independent assortment, and gamete fusion. Mutation must occur for new alleles to be introduced. Recombination ≠ mutation. Der Wohltemperierte Fuchs ( talk  ) 20:19, 14 March 2008 (UTC)


 * Ahem... " suddenly appear "...?!— Wi ki sc ie nt  — 20:22, 14 March 2008 (UTC)


 * No, David, I'm not suggesting that any change can result from recombination (though I won't rule it out). I am suggesting that most changes that do occur in evolution may result from this. But, in your example, you're assuming that there is one allele that controls hair color.  I bet that there is some combination of alleles currently existing in the human gene pool that could cause humans to have blue hair.  But, admittedly, that would be hard to prove.  This week.    --Unflappable (talk) 20:39, 14 March 2008 (UTC)


 * Wikiscient, sorry. I didn't mean to offend anyone or impede progress in anyway!  Perhaps I didn't word it clearly.  Let's try another take.  My understanding is that the way drift can "reduce diversity" is by randomly shuffling out of the population some particular allele.  But so what?  What really matters is phenotype, and we know that it is combinations of alleles that produce phenotypes.  There is no "wing" allele, for example.  The combinations of alleles that "build" the wings of eagles, bats, and bees are probably totally different, and there are probably countless other combinations of alleles that could do it.  Humans, for example, probably have the genetic capacity to build wings, in countless ways even.  So losing some allele does not necessarily mean losing any variation, or potential variation, that is significant to evolution.
 * I just don't see how the loss of variation at the allele level necessarily translates to loss of variation at the level that matters to evolution: phenotypes. To me, it's like the fact that we can represent any numerical value using a base 10 system does not preclude us from representing any numerical value using a base 8 or even a base 2 system.  Losing one allele is like going from base N to base N-1.  As long as N-1 remains greater than 2, no ability to express variety is lost.  --Unflappable (talk) 20:39, 14 March 2008 (UTC)


 * Hmmm... you're sort of losing me here... that seems less clear than your statment of the problem a few posts above...— Wi ki sc ie nt  — 20:47, 14 March 2008 (UTC)

Yes, this is just a phrasing issue. I don't object to saying that "sex can increase diversity", what is wrong is just saying that "sex increases diversity", since this isn't always true. This is why I wrote In species that reproduce sexually, new combinations of genes are produced by genetic recombination, which can increase the variation between organisms. The "can" is vital here, as the literature I've linked to above shows. Tim Vickers (talk) 20:20, 14 March 2008 (UTC)


 * "which can increase the variation between organisms", leading to evolution by natural selection, is the point being made here, right? — Wi ki sc ie nt  — 20:25, 14 March 2008 (UTC)

Is what is currently says. Tim Vickers (talk) 20:28, 14 March 2008 (UTC)


 * Ah. I see. Okay.  Well, what's the problem then on your end, Unflappable...?—  Wi ki sc ie nt  — 20:33, 14 March 2008 (UTC)


 * The problem is that there are two different usages of "diversity" at play here - phenotypic diversity and genotypic diversity. Then when we speak of genotypic diversity, we may speak of some really precise thing, like 'average heterozygosity', which recombination (or sex in general) won't change, or we can speak of allelic diversity or novel haplotypes, which sex CAN change. The valuable contribution of sex is that it increases variation in fitness - recombining chromosomes means you can create "more fit" organisms and "less fit" organisms by recombining genomes (cf. the movie "Twins" with Schwarzenegger and DeVito), which lets you increase the total genetic load in your population. The line Tim added that I amended said exactly that - sex increases variation in fitness. I think that's an acceptable single-line phrasing of the issue. Graft | talk 20:31, 14 March 2008 (UTC)

Also, there seems to be some confusion - drift has nothing to do with sex. You can still have drift in asexual populations. Graft | talk 20:31, 14 March 2008 (UTC)


 * Yes re. asexual drift, good point. Maybe that is part of the confusion --  Unflappable wants to stress the importance of the random-walk-of-gene-assortment, and is simply using the phrase "sexual" too broadly...? —  Wi ki sc ie nt  — 20:41, 14 March 2008 (UTC)


 * Yes, fitness is the key issue, but the problem with adding that word is that we haven't introduced selection at this point in the introduction, so while that would certainly be more precise, it would also be a bit confusing if you were reading it for the first time. That's why I think simply saying "variation" is still right, but a bit clearer. Tim Vickers (talk) 20:35, 14 March 2008 (UTC)


 * "Variation in traits" is great, precise and not introducing anything new. Thanks Graft! Tim Vickers (talk) 20:38, 14 March 2008 (UTC)


 * So where do you stand at this point, Unflappable?— Wi ki sc ie nt  — 20:41, 14 March 2008 (UTC)
 * I still think the role that sexual reshuffling/recombining of genetic material plays in introducing the kind of variation that matters to evolution (in phenotypes) is introduced later than it should be and is generally underplayed in the first paragraph, but I think it's much better now with respect to this point than it was a few days ago. Thanks to Tim, Graft, Wikiscient and everyone else who indulged me with my objections/suggestions.  FWIW, I sure learned a lot in the process. Hopefully we all agree the article has improved as a result of this!  --Unflappable (talk) 20:51, 14 March 2008 (UTC)


 * [[Image:Face-grin.svg]]— Wi ki sc ie nt  — 20:53, 14 March 2008 (UTC)


 * Yes, its certainly refreshing to discuss biology, rather than religion on this talkpage! ;) I also agree the article has improved, thank you everybody for your comments and edits. Tim Vickers (talk) 20:55, 14 March 2008 (UTC)

Graft, please correct me if I'm wrong. The reason you can have drift in asexual populations is because you can still have genetic variation in asexual populations. It's just that in asexual populations it all comes from mutation, right? Regardless of any other sources of asexual genetic variation that I may be currently ignorant of, the point is that genetic drifts happens because of random changes in genetic material that does not manifest itself in phenotypes in a way that is differentially selectable. And in sexual populations, I believe the most significant contributor to genetic variation is the shuffling/recombining of genes caused by sexual reproduction. So just because you can have drift in asexual populations does not mean that drift has nothing to do with sex. In sexual populations, I suspect most drift can be attributed to sexual shuffling/recombining. So unless mutation (and other sources of asexual genetic variation) is proportionally more prevalent in asexual populations than in sexual populations, I would suspect that, in general, there is less drift in asexual populations. --Unflappable (talk) 21:02, 14 March 2008 (UTC)
 * No - drift is due entirely to sampling, which occurs in any reproducing population. There is no difference whether the population is sexual or asexual. In fact, common models of drift don't even bother to specify which they're modeling. In *all* populations, the most significant contributor to *genetic* variation is mutation. Without mutation, new genetic variation does not arise. Sex might be the biggest contributor to *phenotypic* variation, but that's a separate issue. Graft | talk 21:10, 14 March 2008 (UTC)

As I have said several times already, drift is about stuff like anvils falling, not sexual reproduction - and not mutation. Slrubenstein  |  Talk 23:11, 14 March 2008 (UTC)


 * Hmmm. Here's what I was thinking you were talking about, Graft:
 * (From This site on asexual reproduction)
 * — Wi ki sc ie nt  — 21:22, 14 March 2008 (UTC)
 * — Wi ki sc ie nt  — 21:22, 14 March 2008 (UTC)

Graft, okay, drift is due entirely to random sampling from a pool of variation. But what is that pool comprised of? You act like it's the gene pool, and indirectly it is, but when you step on a beetle, you're acting on the beetle. Yes, that affects the allele distribution/variation in that beetle's population (particularly if it's a small population of beetles), but it also affects the phenotype distribution/variation in that beetle's population. And it's the latter variation that matters to evolution. From the opening sentence in the article: "evolution is the changes seen in the inherited traits of a population" -- evolution, including evolution from drift, is changes in the inherited trait (phenotype) frequency, not changes in allele frequency. In other words, if a population loses a neutral/unexpressed allele due to drift, that's arguably not even evolution. It's just ditching apparently unused genetic baggage. And it's the phenotype variation for which the biggest contributor in sexually reproducing populations might be sexual reproduction, not mutation. --Unflappable (talk) 23:21, 14 March 2008 (UTC)
 * The opening sentence is, in my opinion, bad, since it excludes all of neutral evolution. No one in the field would agree with the statement that losing neutral alleles is not evolution. . Graft | talk 23:42, 14 March 2008 (UTC)
 * Fair enough. Then we must define evolution as changes in genotype or phenotype (recognizing that change in phenotype can be the result of change or just reshuffling in genotype).  So, then, drift would be due to random sampling from the pool of phenotypes as well as genotypes, and you've already agreed that sexual reproduction might even be the biggest contributor to phenotype variation in sexually reproducing populations.  So, then, drift does have something to do with sex, and arguably quite a bit in populations that reproduce sexually.  --Unflappable (talk) 00:00, 15 March 2008 (UTC)
 * As I wrote above, drift is not the shuffling of the deck or any effect of shuffling the deck. Drift is the effect of some cards - at random - disappearing from the deck entirely. If you remove 25 cards from a 52 card deck, the effect on the hand dealt will be noticible very quickly and will radically alter our ability to play most games. If you remove 25 cards from a 5000 card deck, the effect on the hand dealt will be slight and will not seriously affect the ability to play a game. The key thing is that the cards that disappear do so randomly. Sexual reproduction is one of many possible factors, but even here numbers count. If a couple has four children, it is conceivable that there will be no drift for a given trait. If they have ten or twelve kids, the drift may well be minimal. If they have one kid, the drift will be higher - that is, if they are in a small population. In any event, sexual reproduction is but one factor, falling anvils, reckless drivers, lightening also "cause" drift.


 * The real issue here is not drift but the relationship between sexual reproduction and evolution. We have discussed this before, at times, in discussions about lateral genetransfer.  In a recent essay in The New York Review of Books Freeman Dyson argued that evolution as modelled by the Modern Synthesis really did not begin until the evolution of sexual reproduction.  So I see one of two ways to handle this: either there is a debate as to whether the views expressed by Freeman are right or not - this would be a big theoretcial debate - or Freeman was expressing an ideosyncratic or personal view, and while sexual reproduction does play a role in evolution it is only on piece of a much bigger model covering all known forms of life past and present.  I am not enough of an expert to know which one of these two possibilities is right.  But as a Wikipedia editor, I think whichever one is right it is best handled by a section n the body of the article, rather than in the introduction. Slrubenstein   |  Talk 10:54, 15 March 2008 (UTC)
 * Agree. Dyson's arguments aren't exactly accepted widely, and after reading it I was under the impression is was a pet theory, not necessarily something he was setting all stake on. Unflap, remember H-W equilibrium is thrown out the window when at least one condition is invalidated- sexual reproduction is not actually one (random mating is). Natural selection selects individual's genes, in a manner, through their children- those more successful will have more children, and more of their genes amongst the population or what have you. Sexual reproduction is more the means to an end, not the impetus. Der Wohltemperierte Fuchs ( talk  ) 12:45, 15 March 2008 (UTC)
 * Are you talking about "Our Biotech Future"? That's a stupid, stupid essay. Designing genomes will NEVER become the domain of the little guy, just like designing semiconductors is still something mom and pop will never do. Dyson should stick to physics, I think. Graft | talk 17:17, 15 March 2008 (UTC)
 * Yeah, that is the one. My own opinion aside, I simply did not know if he was drawing on views held by biologists or just promoting his own.  In any event, I was using it to make a point which believe still stands: whether the impact of the development of sexual reproduction on our understanding of evolution, and thus evolutionary theory, is a matter of notable debate, or whether sexual reproduction is but one among many factors that enters into accounts of evolution, it is best handled in the body of the article, and not the introduction. Slrubenstein   |  Talk 18:55, 15 March 2008 (UTC)

The significance of the role of sexual reproduction in evolution
In reference to the above comments, I haven't read Dyson. Biology never interested me much until recently I read Dawkins' The Selfish Gene. Now I think I'm some kind of expert. Just kidding. But I do find the mechanisms of evolution to be a fascinating topic, and when I came to this article what I thought I understood seemed to be contradicted in the introduction. Hence all the back and forth since my first post here.

Speaking of Dawkins, one of the concepts that struck me was the imprecise definition of gene: "...any portion of chromosomal material that potentially lasts for enough generations to serve as a unit of natural selection... a replicator with high-copying fidelity" (p. 28). He also defines a genetic unit as "any length of chromosome [no matter how long or short it is], not physically differentiated from the rest of the chromosome in any way." (p. 29)  He also points out on the same page that the  length of a genetic unit is inversely proportional to its expected life span (in generations). That is, the longer it is, the more likely it is to be split by any crossing-over during sexual reproduction. Then, on p. 31, Dawkins writes,

Again, I'm not suggesting that evolution is sexual reproduction (as Dyson apparently is arguing). Clearly mutation is "of great evolutionary importance", and it occurs in asexual reproduction too. However, according to Dawkins, "the usual way for a new genetic unit to be formed" is through the "chance coming together, through crossing-over, of previously existing sub-units". That, to my understanding, is what sexual reproduction achieves, and the significance of this role it plays in evolution still seems to be underplayed in the current version of the first paragraph of the introduction. This is the part that I think still needs to change:

In particular, in the second sentence above, by mentioning mutation first in terms of producing new or altered traits, and then some other ways new traits are introduced in the same sentence, but not mentioning the role of sexual reproduction until the next sentence, we are not at all conveying that the usual way for a new genetic unit to be formed is through sexual reproduction, and that these other methods are very important to evolution, but much more rare. As I understand them, the meaning of these words is very different from the meaning of Dawkins' words that I quoted above. The following suggested rewording would convey what I understand from reading Dawkins better than the current wording:

This is the same point I've been trying to make all along, but perhaps it is more understandable now that I've taken the time to express it in Dawkins' words. Is it? --Unflappable (talk) 15:35, 15 March 2008 (UTC)


 * Dawkins is one of the most effective popularizers of science of our time. Stephen Jay Gould - someone he often disagreed with - is another, although Gould's dissertation research was on evolutionary biology, and he has made original contributions to evolutionary theory (which is not to say they are universally accepted) whereas Dawkins has not (the selfish gene idea derives from important work by Hamilton).  In any event, no one should rely on Dawkins as an authority on evolutionary theory.  You might do better reading the sections on evolutionary theory in a good physical anthropology (e.g. Nelson and jeurmaine) or evolutionary biology textbook.  Personally, I love G.G. Simpson's The Meaning of Evolution if you want to rely on a popular account.  Though dated, Simpson (even more than Gould, and in contrast to Dawkins) was a towering figure in the real science of evolutionary biology.  And the book is very accessible. Slrubenstein   |  Talk 19:04, 15 March 2008 (UTC)
 * As much as Dawkin's popularized science, he's not that great a writer (or at least, someone else coulda made The Selfish Gene for straight-forward for biology newbies.) Der Wohltemperierte Fuchs ( talk  ) 22:06, 15 March 2008 (UTC)

As a note, recombination is not restricted to sexual organisms, see this paper and. Also, the usual way for new genes to evolve seems to be duplication of an existing gene, or domain, and then divergence of this sequence through further mutation. See and this review for good overviews of this. A critical point here is that gene duplication is a form of mutation, so all these steps in the process are mutational events. Tim Vickers (talk) 21:50, 15 March 2008 (UTC)

Slrubenstein, I appreciate the references, but are you suggesting that because Dawkins has not made original contributions to evolutionary theory, what he writes about it should not be considered to be accurate? In particular, is anything I quoted above inaccurate? If so, what? If not, I don't understand your point. Anyway, while I understand that much of the Selfish Gene is derived from Hamilton and others, I think his contribution of memetics is original, particularly as a way to explain evolutionary theory more generally, independent from biological evolution, to help distinguish between general principles of evolution from those aspects that are peculiar to biological evolution.

Tim, yes, we can define gene duplication as a form of mutation, but that seems to be a much broader usage of the term mutation than what you (in your comments above), the evolution article, and the mutation article, seem to mean by it (which came across to me to mean what Dawkins refers to as point mutation). But, if we do define it that way, that's yet another way to make my point, since every instance of sexual reproduction leads to the mutation (in this broad sense) of countless genes, and causes the genetic recombination that accounts for most of the genetic variation that occurs in biological evolution. --Unflappable (talk) 05:36, 16 March 2008 (UTC)


 * Unflappable, you do not understand my point? I guess I have to bullet-point it for you:

I won't go into memetics, which I think it total pseudoscience on rank with astrology, but it is not important here as it is unrelated to the article.
 * The only point you have been making, which you now repeat with lengthy quotation from Dawkins, is that in the course of sexual reproduction genes recombine. Before you initiated this lengthy discussion, the article had an entire section dedicated to recombination, and Tim or Graft have now added a clause or sentence to the intro mentioning recombination.  So the point you keep breinging up is made in the article.
 * You clearly do not understand drift
 * Just because you one reads one book by anyone about anything does not mean you one understands the subject matter
 * thus, just because you one have has read the one book by Dawkins does not mean that you one understands evolution and evolutionary theory.
 * I have suggested some other authors whose works you could profitably read, and I think you could benefit from reading some textbooks as well


 * I'll further add that most contributors to this article have read Dawkins and many othe books as well. The way to improve this article is not to make the contents fall more in line with the views of one popularizer.  Certainly, your views as a non-expert on evolution are valuable, as a reader.  What sentences are unclear?  What paragraphs do not flow?  Articles can alwsya be improved!  Slrubenstein   |  Talk 15:31, 16 March 2008 (UTC)
 * I would just like to point out that most of your points are almost purely ad hominem. Asserting that I do not understand drift does not make it so.  Pointing out what is stated in other parts of the article does not address my points about the impression created in the introduction.  Nothing you say even states whether Dawkins is correct when he says the usual way new genes are created is through recombination, much less explain why it is incorrect if it is.  --Unflappable (talk) 16:05, 16 March 2008 (UTC)

The classes of mutation, as this article explains, are point mutations, transversions, translocations, insertions and deletions. See Mutation. Usually sexual recombination does not cause such effects, these come from unequal crossovers, which are thankfully quite rare! Tim Vickers (talk) 16:07, 16 March 2008 (UTC)


 * Hi Tim. Please correct me if I'm wrong, but I'm still sensing a resistance, which I don't understand, to accept Dawkins' assertion that the usual way in which new genes are formed is by crossing over (not unequal crossover, but the normal crossover of meiotic recombination during sexual reproduction), especially a reluctance to reflect that notion in the introduction of this article.  --Unflappable (talk) 22:20, 17 March 2008 (UTC)


 * Hey there, yes, you'd be right in thinking that! :) Did you have a look at the reviews I posted above about how new genes tend to appear? It's also reviewed in and an interesting but rarer alternative mentioned in this paper. All these routes to evolutionary innovation require a mutation of some kind - either simply a point mutation, or more commonly a segment of the genome being duplicated and then a series of point mutations that cause the duplicated sequence to acquire new functions. All of these mechanisms operate just as well in prokaryotes as eukaryotes - indeed if they didn't we wouldn't have eukaryotes! Tim Vickers (talk) 22:41, 17 March 2008 (UTC)


 * also pushes horizontal gene transfer as a major source of new genes, and focuses on bacteria/bacteriophages and eukaryotes/mobile elements. Although it also mentions that "The most broadly held view is that regions of the genome are duplicated and that subsequent functional diversification can produce genes conferring novel properties" Again, nothing to do with sex. Tim Vickers (talk) 22:48, 17 March 2008 (UTC)


 * So, Dawkins is wrong when he says the usual way new genes are formed is through meiotic crossing over? Or do we have disagreement of semantics - based on strict and more broad definitions of "gene"?  --Unflappable (talk) 23:03, 17 March 2008 (UTC)


 * Gene duplications can occur during meiosis, as the result of unequal crossover events, so I wouldn't say he was entirely wrong, but these are mutations that occur during mistakes in the process of homologous recombination, rather than a normal part of the process. Such duplications can also occur in non-sexual organisms, and in germ-line cells as part of DNA repair, so I'd guess he was simplifying the topic a little. There are several good reviews that deal with this area and are cited in the article in the section on mutation, if you can't access the full text just e-mail me with requests and I'll send you the Pdfs. Tim Vickers (talk) 23:09, 17 March 2008 (UTC)

Russian interwiki link
Need to add the following:  MarsRover (talk) 02:25, 17 March 2008 (UTC)
 * Done. Tim Vickers (talk) 15:31, 17 March 2008 (UTC)

First sentence
In biology, evolution is the changes seen in the inherited traits of a population from one generation to the next.

The first sentence is inaccurate and grammatically incorrect.

Might I propose that the first sentence be changed to reflect the fact that evolution is a process and that the process can lead to, not only changes, but also to conservation of inherited traits.


 * 1. Evolution is a process not the result of a process. One of the results of the process of evolution is call speciation, the arising of a new species.


 * 2. The word "seen" implies the strictly visual appearance of an organism. Inherited traits can also be unseen, i.e. resistance to disease or nutritional requirements.


 * 3. The modern term for inherited traits is "genes".


 * 4. Inherited traits (genes) can also be conserved. If the trait is successful it is unlikely to be changed or go away.


 * 5. Inherited traits (genes) are not uniformly found through a population. Some traits (genes) are more common, or found with more frequency, than others.


 * 6. Although it is strictly true that the frequency of an inherited trait (gene) can change from ONE generation to the NEXT it would be more accurate to imply a multi-generational process.

Thus:

"In biology, evolution is the name given to the process which can lead to changes or conservation in the frequency of inherited traits (genes) found in a population of organisms from generation to generation."

GeeOh (talk) 04:36, 18 March 2008 (UTC) —Preceding unsigned comment added by GeeOh (talk • contribs) 04:32, 18 March 2008 (UTC)


 * Elegant like a sink full of dishes. The first sentence lacks a lot, but simplicity is worth it. We have a whole article to make caveats. Graft | talk 05:00, 18 March 2008 (UTC)
 * Could we at least get rid of the unnecessary phrase, "seen as?" That is one especially slimy old dish. Slrubenstein   |  Talk 13:01, 18 March 2008 (UTC)
 * GeeOh, I agree with most of your points. I definitely agree it's a process and should say so.  However, it's the changes that make it evolution.  After all, if there was only conservation of traits and no changes, it wouldn't be evolution.  How about this?
 * "In biology, evolution is the process of change in the frequency of inherited traits (genes) in a population of organisms from generation to generation."
 * --Unflappable (talk) 17:30, 18 March 2008 (UTC)


 * In biology, evolution is the process that changes the inherited traits of a population from one generation to the next. Tim Vickers (talk) 17:37, 18 March 2008 (UTC)


 * I agree that's even better. But is it the process that changes the traits, or do the changes comprise the process?  I think it's the latter, hence:


 * In biology, evolution is the process of changes in the inherited traits of a population from one generation to the next.
 * --Unflappable (talk) 17:49, 18 March 2008 (UTC)


 * I like it. Let's put it in! Graft | talk 17:53, 18 March 2008 (UTC)

Uh, I think here we run up against the old "evolution as theory" and "evolution as fact." According to many, the fact of evolution is simply the change in gene frequencies in a population over time (not any particular process that cuases changes in frequencies or genes); the theory of evolution is a model of those processes that account for changes in genes and gene frequencies. Slrubenstein  |  Talk 17:54, 18 March 2008 (UTC)

That last version still has the "from one generation to the next" problem, which excludes the multi-generational changes. How about:


 * In biology, evolution is the process of changes in the inherited traits of a population that accumulate with each generation.

--Unflappable (talk) 18:02, 18 March 2008 (UTC)


 * Or, to avoid "process":
 * In biology, evolution is the accumulation of changes over time in the inherited traits of a population that occur at each generation.
 * --Unflappable (talk) 18:06, 18 March 2008 (UTC)

The first paragraph currently is:


 * In biology, evolution is the process of change in the inherited traits of a population of organisms from one generation to the next. Though changes between generations are relatively minor, differences accumulate with each subsequent generation and can, over time, cause substantial changes in the organisms.

How about:

--Unflappable (talk) 18:23, 18 March 2008 (UTC)
 * In biology, evolution is the accumulation of changes over time in the inherited traits of a population of organisms that occur at each generation. Though changes between generations are relatively minor, differences accumulate with each subsequent generation and, over time, cause substantial and even fundamental changes in the organisms, including the creation of new species.


 * There has never been a fundamental change in the makeup of the organisms on Earth, they all use DNA, proteins and lipids, and usually share much of their metabolic wiring - the changes between E. coli and elephants are just matters of detail, at the fundamental level. Tim Vickers (talk) 19:03, 18 March 2008 (UTC)
 * Indeed, one of the most revolutionary and profound elements of Darwinian thought is to conceive of species as quantitatively rather than qualitatively different - this is one of the issues where Darwinian thought at creationaism are clearly and irreconcilably at odds. Slrubenstein  |  Talk 19:10, 18 March 2008 (UTC)


 * DNA, protein and lipids are all products of biological evolution, and their evolutionary development represents fundamental differences between the life forms that existed before and after this development.  I believe that the only characteristic that is fundamental to evolution is replication. --Unflappable (talk) 19:25, 18 March 2008 (UTC)
 * You just switched the terms of the discussion. Tim and I were responding to a sentence you proposed in which the object was the evolution of new species.  Now you are changing it to the evolution of life itself.  It is my impression that this article is basically about the former, not the latter.  Are you proposing to change that? Slrubenstein   |  Talk 19:35, 18 March 2008 (UTC)
 * The proposed wording does not say or imply that the object is evolution of new species; it says that new species creation is included in what results from evolution. I see it all as the process of biological evolution starting with the primordial soup.  It is conceivable that in a billion years or so some new fundamental mechanisms will evolve and DNA, protein and lipids will no longer be present.  The opening sentence/paragraph of an article should define the scope, and I don't see why the evolutionary beginnings of life on earth, or the conceivable future, should be excluded from the scope of a general article explaining biological evolution.  To the contrary, covering that is probably essential to understanding evolution  --Unflappable (talk) 19:49, 18 March 2008 (UTC)
 * Note that the UCB on-line information about evolutionary biology starts with the origin of life here. While the origin of life is an important subtopic and deserves its own article, my point is that it is in scope for the topic of evolution, and the definition of evolution implied in the intro wording to this article should not exclude it. --Unflappable (talk) 20:03, 18 March 2008 (UTC)
 * Chemical evolution isn't the topic of this article, which deals only with how evolution occurs in biological systems. The origin of life isn't an area we should, or need, to cover in any more detail than we have. Tim Vickers (talk) 20:57, 18 March 2008 (UTC)
 * Mechanisms and processes of evolution is also not the topic of this article, though that title currently redirects here and perhaps it should be a separate article. Also, the evolution of DNA, proteins and lipids is not chemical evolution.  We don't know how exactly these substances evolved, but I think it's pretty clear that they evolved (and continue to evolve) integrally with biological life, and are hence part of biological evolution.
 * I think it's interesting that you say this article deals with how evolution occurs (present tense) in biological systems. Essentially, the implication is that the title of this article should be something like Specific modern mechanisms and processes of biological evolution.  But that's not the title.  I think that explains some of the discord we're having.  It seems to me that the article with a title as general as Evolution should not be limited to such a specific subject, as worthy as it is.  This article should be about the general mechanism of biological evolution, explaining how and why it's both theory and fact, etc.  The details of the underlying mechanisms seems to me to be out of scope; it's hiding the forest with the trees.
 * Finally, one more plug for saying in the opening sentence that evolution is the accumulation of changes over time rather than the process of change... By calling it a process we imply that there is intentional purpose, begging the question of whose intent it is.  Noting that evolution is simply the accumulation of changes over time is more accurate.   --Unflappable (talk) 22:33, 18 March 2008 (UTC)

You are skirting an NOR and NPOV violation. I understand your view, but has it been published in a reliable source, and is it notable? NPOV insists we include all notable views found in reliable sources, and NOR forbids us from inserting our own views. This is not what you think evolution is, this is how mainstream evolutionary scientists view evolution. And they view it as an empirical fact - changes in gene frequencies over time - and as a theory - processes like natural selection and drift which, acting on mutations, reshape gene frequencies in a population. This is very different from those forces that led to the creation of amino acids for the first time. You seem to be using a broad, colloquial perhaps dictionary definition of "evolution." If you chose to follow that path you could say the introduction should discuss how ideas evolve as well as the evolution of stars and punk rock. No, this isn't about "all" evolution, it is about evolution as understood by evolutionary scientists e.g. evolutionary biologists, physical anthropologists, population geneticists, molecular geneticists, zoologists, etc. There is a large set of phenomena they are concerned with, and a particular model, the modern synthesis, they use to understand this phenomenon and to generate new questions. The bottom line: to say "DNA evolved" and to say "human beings evolved" is to use the same word - evolved - in two very different ways. Scientists do not confuse these two very different ways and we would be betraying our educational mission if we mixed them up. Slrubenstein  |  Talk 23:12, 18 March 2008 (UTC)


 * I agree completely Slrubenstein, well put. Tim Vickers (talk) 23:14, 18 March 2008 (UTC)


 * I also have to agree with Slrubenstein. I am afraid the comments from Unflappable appear to be further and further into the weeds and more and more confused. We have other articles on the meaning of the word evolution like Evolution (term). It does not matter what I believe the word "evolution" means, or what you want it to mean, or Tim, or Slrubenstein; we are just reporting on the mainstream accepted definition of the word in a biological context. No more, no less. --Filll (talk) 23:39, 18 March 2008 (UTC)


 * I understand and agree that this article should be limited to the mainstream accepted meaning of evolution in the biological context, and everything I'm saying is consistent with that. If it seems otherwise, then I've not explained myself clearly.   I've even cited a source (below).  --Unflappable (talk) 00:05, 19 March 2008 (UTC)


 * Yes, my view - that the origin of life and how everything evolved since is part of biological evolution - has been cited in a reliable source, which I linked above. Here is the actual quote:
 * Good enough? Again, I'm not saying that we need to say anything more about the origin of life.  I'm just saying that biological evolution, as understood by evolutionary scientists, started right after life originated, probably billions of years before DNA, proteins and lipids evolved, and the opening paragraph should not imply otherwise, nor be based on any assumptions to the contrary.  --Unflappable (talk) 23:34, 18 March 2008 (UTC)
 * Good enough? Again, I'm not saying that we need to say anything more about the origin of life.  I'm just saying that biological evolution, as understood by evolutionary scientists, started right after life originated, probably billions of years before DNA, proteins and lipids evolved, and the opening paragraph should not imply otherwise, nor be based on any assumptions to the contrary.  --Unflappable (talk) 23:34, 18 March 2008 (UTC)


 * You can tell sentences that make assumptions, because they contain words like "probably". This article makes no such assumptions. Tim Vickers (talk) 23:42, 18 March 2008 (UTC)
 * I'm not sure what you mean Tim, but all sentences make assumptions, if nothing else that the meanings of the words used in the sentence have certain meanings. Words like "probably" are not required to make assumptions.  In this case, there is an assumed scope to the article, and depending on how the introductory sentences are worded, different assumptions about what that scope is are made.  And what those words should assume the scope should be is part of what we're talking about here.  The current wording of the first paragraph allows for a broad scope, and that's fine.  The reason we're talking about this is your objection to my proposal of changing the wording to say that evolution has caused fundamental changes in life.  Perhaps I misunderstood, but you seem to think that this article should only cover evolution since the point fundamental changes in life have occurred, and you object to the rewording because you think it extends the scope to prior to that point (but I think that scope is already implied with the current wording).  --Unflappable (talk) 00:01, 19 March 2008 (UTC)

Looking through the Berkeley website, I can see that you tried hard to find a place where maybe these concepts could be confused. However, I believe you are misreading the text there. It does not say that evolution has anything to do with abiogenesis. And if you look in the glossary for evolution, you see it says it is common descent. However, that Berkeley website is a bit more elementary than what we are aiming for with this article, I think.--Filll (talk) 00:13, 19 March 2008 (UTC)


 * It's just what I found with a quick google search. I'm not saying evolution has anything to do with abiogenesis, except that after abiogenesis is when evolution of common descent started.  Anyway, the only point in issue relevant to this article is whether the differences accumulated through evolution ultimately amount to fundamental changes, or are merely substantial changes.  And I'm not even insisting that it should say "fundamental"; my main point is that substantial seems to be an understatement.  For example, the differences between robins and eagles are substantial, but aren't the changes that occurred between dinosaurs and modern birds more than merely substantial?  Going from cold to warm blooded might not require fundamental changes at the genetic level, but the differences in terms of inherited traits, which is the context in question, are, I think, fundamental.  --Unflappable (talk) 00:26, 19 March 2008 (UTC)

Now I see what you are doing. Look, I see here a fundamental confusion about what science is, and what a scientific theory is and what constitutes data etc. An accepted scientific theory is just a temporary explanation for data that makes successful predictions. That is it. It is not truth; in fact it is provisional so probably will be replaced soon. The Theory of Gravity will probably be replaced before the Theory of Evolution, but both are provisional and temporary etc. Both make predictions which we can check. As for all the nonsense where you are dancing around about macro and microevolution, there is so much evidence for speciation it is not even funny. I would point out the fusing of our 2nd chromosome and the past records of endogenous retroviruses in our genetic code, for example. But there is just oodles of evidence, including different species of plants on either side of the Great Wall of China. There are literally tens of thousands of examples of speciation, established in the laboratory, in the field, in fossil records, in genetic evidence etc. And no matter how much you try to torture the language here, you will not get around this.--Filll (talk) 00:51, 19 March 2008 (UTC)
 * I don't understand how this is a response to anything I've written. You didn't answer any of my questions.  Why are you trying to convince me there is incontrovertible evidence for speciation?  Of course there is.  I'm the one trying to add wording to the intro paragraph that says as much.  More to the point, do you have any objection to my proposed rewording of the first paragraph (above - posted at 18:23, 18 March 2008 )?  If so, what is it?  Anyway, I'm out of here for a few days.  Have fun, folks.  --Unflappable (talk) 01:20, 19 March 2008 (UTC)

"... evolution has caused fundamental changes in life" is ambiguous: it's true that the to-day's most complex organisms are more complex than those of 500M years ago; but the dominant life-form now and probably 500M years ago is bacteria, and it's hard to say how much they've changed in that time; and it would be best to avoid any wording that suggests saltationism. Philcha (talk) 09:15, 19 March 2008 (UTC)


 * I must repeat: one of Darwin's greatest and most profound insights was to view differences between species in quantitative rather than qualitative terms. Changes may appear to be fundamental or substantial but in fact they are due to the accrual of many finite changes in, we now know, the sequencing of a finite and universally shared set of nucleotides.  Viewing species as fundamentally different is creationism, not evolutionary science. Slrubenstein   |  Talk 09:59, 19 March 2008 (UTC)


 * I don't think fundamental in this context suggests saltationism any more than suggesting that the Colorado River created fundamental changes on the Colorado Plateau suggests anything other than slow and steady erosion by the river carving into the uplifting plate created the Grand Canyon. The fundamental changes are at the phenotype context, not the genotype (note that the context is changes in the organisms).  It's a question of context, not creationism vs. evolutionary science.
 * Anyway, maybe fundamental is problematic, but I still think the current wording, "can, over time, cause substantial changes in the organisms", is understated. The changes that do (not just can) occur over time in the organisms are much more than merely substantial. --Unflappable (talk) 14:20, 22 March 2008 (UTC)

Second sentence
Looking at the discussion above about "fundamental" changes, it's a good point, and perhaps the current wording wrongly implies that the organisms change, rather than the populations. Current sentence "Though changes between generations are relatively minor, differences accumulate with each subsequent generation and can, over time, cause substantial changes in the organisms." Proposed "Though changes between generations are relatively minor, differences accumulate with each subsequent generation and can, over time, result in divergence in the characteristics of the organisms, often to the extent that new species arise. Just my tuppenceworth, .. dave souza, talk 20:17, 18 March 2008 (UTC)

Frequency of harmful mutations
According to the mutation section:
 * Studies in the fly Drosophila melanogaster suggest that about 70 percent of mutations are harmful, and the remainder are either neutral or have weakly beneficial effects.

According to the mutation article:
 * It is believed that the overwhelming majority of mutations have no significant effect on an organism's fitness.

These seem to be in contradiction. --Tgr (talk) 17:36, 19 March 2008 (UTC)


 * Good catch, the mistake is mine, this figure refers to nonsynonymous mutations within genes, rather than mutations in general. I've corrected the article. Tim Vickers (talk) 18:18, 19 March 2008 (UTC)


 * I'd like to point out that the subsequent sentence (that there is an 'optimal' mutation rate set by a tradeoff between beneficial and deleterious mutations) is contradicted by the cited reference, which says: Intuition might suggest that the mutation rate is set by a tradeoff between the effects of deleterious and beneficial mutations, since mutation is required for long-term adaptation. Recombination, however, weakens the effect of beneficial mutations on mutation rate modifiers and thus the mutation rate in many populations may instead be set by a tradeoff between deleterious mutations and constraints on the fidelity of replication. That latter - essentially the notion that perfect fidelity in replication or repair places an inordinate burden on the cell - is itself somewhat hypothetical, though a much better hypothesis than the idea that the mutation rate is set to encourage adaptation. If no one has objections I will amend the sentence. Graft | talk 20:04, 19 March 2008 (UTC)
 * Actually, reading further, the citation suggests that the sentence we have in the article is flatly wrong - that is, there is no situation in which a mutation leading to a higher mutation rate could be beneficial, and it will only be fixed in the situation that it is linked to some other variant of high benefit. Graft | talk 20:07, 19 March 2008 (UTC)
 * This can be true in asexual organisms such as bacteria, where mutator alleles are found in wild populations and can be observed in the laboratory eg . However, this is a bit of a special case, and the sentence would probably be more general if it said "Therefore, the optimal mutation rate for a species is a trade-off between costs of a high mutation rate, such as deleterious mutations, and the metabolic costs of maintaining systems to reduce the mutation rate, such as DNA repair enzymes." Tim Vickers (talk) 20:29, 19 March 2008 (UTC)
 * Ah that's true (not the way I'm used to thinking about linkage). But those are exceptional circumstances, and anyway as far as I can tell they are transient (i.e. unstable because of the high mutation rate), and they don't suggest that the *base* mutation rate, fairly uniform across taxa, is set or maintained by this kind of selection. For that, actually we'd have to show that mutations that improve fidelity were disadvantageous and selected against because they prevented future adaptation (which seems impossible to me from a theoretical perspective). Graft | talk 20:55, 19 March 2008 (UTC)

lead length
It may just be me, but the lead to this article seems to be very long. Would anyone object to me shortening it? J.d ela noy gabs adds 17:14, 20 March 2008 (UTC)


 * I feel like we've had this conversation before, and the consensus was that since this is a keystone article describing a very vast and complicated field of biology, then it's okay if it gets a little bit long. What would you propose to trim? Graft | talk 18:49, 20 March 2008 (UTC)


 * The lead is about the average length for a Featured article, I don't think it needs to be shortened at all. Tim Vickers (talk) 20:35, 20 March 2008 (UTC)

No editing?
Why can no one edit the article? Does this not destroy the essence of wikipedia, that anyone can make reasonable changes to the articles? --Albert Einstien&#39;s ghost (talk) 12:57, 25 March 2008 (UTC)


 * This is a temporary situation while we deal with a long-term vandal. Tim Vickers (talk) 13:55, 25 March 2008 (UTC)

This looks like it will continue for a while. Please edit Talk:Evolution/draft article, and constructive edits will be transferred to the main article. Tim Vickers (talk) 02:40, 31 March 2008 (UTC)

Reduction back to semi-protection
I've reduced editing back to semi-protection and disbanded the /draft_article idea. (1) It completely ruins the history of the page and appropriate attribution. (2) Fully-protecting high-profile pages like this is simply absurd. If there is a persistent vandal, let's deal with him / her. We've got 1500 admins, 1000 of which are active. Let's deal with the problem user without punishing everyone else. --MZMcBride (talk) 18:27, 4 April 2008 (UTC)


 * Attribution is maintained by linking to the user who edited the draft page, as in diff, this is not a concern. I'd advise you to look at the article history and the protection history before describing this action as "absurd", it was not taken lightly and should not be dismissed in those terms. Tim Vickers (talk) 20:43, 4 April 2008 (UTC)


 * I replied on AN, but I will say that this has been thoroughly discussed in the past for articles like George W. Bush. As far as I know, the consensus has always been to semi-protect and revert any vandals on high-profile pages. --MZMcBride (talk) 21:07, 4 April 2008 (UTC)


 * If you look at the history for even just today, you can see why this page needs serious protection. I am not sure if you have been watching this page for a while or not, as I can find no edits that you have made in the past four months (until these recent changes).  Note that I am not saying you shouldn't be able to edit, nor am I saying you haven't been watching the article, as I am sure there are many individuals that watch the article without editing (I only have 1 edit on this page myself).  When I look at my watchlist, and 70% of edits in a day are major vandalisms (like what constantly happens here), it is easy to miss smaller edits on other pages that warrant my attention.  I don't think fully protecting this page is too much to ask for, nor is it a detriment to users. IanCheesman (talk) 00:31, 6 April 2008 (UTC)


 * You don't think blocking almost every user (with the exception of about 1000) from editing the article is a detriment to users? Wow. Twenty minutes of vandalism about 12 hours ago, and none since. You've left me nearly speechless with your comments. --MZMcBride (talk) 02:05, 6 April 2008 (UTC)


 * You don't appear to be familiar with the history of this article, why not go back through the last few months and get a better idea of the problem? Tim Vickers (talk) 03:40, 6 April 2008 (UTC)


 * I don't understand why we are discussing this any further. There appears to be a consensus to roll out this little experiment to help insure that this article is no longer victimized by one of the WP's most prolific vandals.  If MZMcBride chooses to flaunt this consensus, tough cookies.  Even admins can't stand in the way of consensus, especially in dealing with this problem.  Roll it out and if he wants to continue to fight it, report it like we would with any other editor who intends to be disruptive.  Baegis (talk) 05:37, 6 April 2008 (UTC)


 * Just to note the results of the experiment, protection was reduced to semi-protected and within 24 hours four socks of the same vandal had made the same tedious edit that has been reverted many times before, each sock being blocked after making the same old edit. A better method of preventing vandalism by socks of that user is required. .. dave souza, talk 17:32, 6 April 2008 (UTC)


 * If the vandal keeps making "the same tedious edit" why couldn't a Bot handle the situation? --MarsRover (talk) 18:28, 6 April 2008 (UTC)


 * I seem to remember some discussion which said that User:ClueBot was reprogrammed to deal with it. Hut 8.5 18:30, 6 April 2008 (UTC)

Bleh. I'm not going to waste any more time in a discussion with people who simply aren't rational. I re-read the protection policy, and couldn't find anything that stated that preemptive protection was allowed (i.e., fully-protecting when there is no activity on the page), and I couldn't find anything justifying full protection for cases like this. The section on semi-protection clearly covered pages that are "subject to heavy and persistent vandalism." Oh well. I'm right, and I don't say that to sound like an ass, I'm right because I've been doing this long enough to know that I'm right, and I've spoken with other admins (who I respect and trust) who agree with me. I've also looked at other cases of indefinite full protection, and everything suggests that this approach is the wrong one. Oh well. Let a vandal win, I'll live. --MZMcBride (talk) 20:09, 6 April 2008 (UTC)


 * I tend to think TimVickers is on the right track. It is ridiculous to assume a constant vigil will protect this article. Why would anyone want to use a source for evolution starting with Genesis 1:1. Besides vandalism, no one who has ever read this article (who knows anything about evolution) is ever really happy with it. The article can become like a sand dune changing with the winds, but I wouldn't want to limit useful growth either. I like Tim's method that prevents vandalism of the main article, prevents re-inventing the entire article, but allows useful growth with peer review of commenting editors. Seems a good compromise with this "hot topic". I would think the "end product" would be of greater concern than any Wikipedia policy. GetAgrippa (talk) 20:27, 6 April 2008 (UTC)


 * If we get a workable bot as a solution, then that would be the optimal result. However, we've had no success with that so far. Tim Vickers (talk) 21:20, 6 April 2008 (UTC)


 * MZM - I don't think anyone, Tim included, is suggesting this is the correct approach. It is merely the *best* approach under the circumstances, as I said to you the other day in chat. I've seen no indication that admins are capable of successfully fending off this vandal, based on the past few attempted unprotections - once the vandal realizes the page is unprotected, a significant fraction of the page's life is spent as a copy of Genesis. While we all dearly want this page to be editable, leaving a high-profile page vandalized a significant portion of the time is NOT a better option.
 * What we want is A BETTER SOLUTION. We do not want vandalism to continue, and we do NOT want the page to be fully protected. But Christ, PLEASE, we need to bring some other resources to bear here. I don't give a shit what they are - bots, hard-coded avoidance, more protection levels, whatever. Just, something. Graft | talk 23:12, 6 April 2008 (UTC)

Looking through the edit history, for every day the article was not hard-protected it sat in a vandalized state for probably 10 minutes, at a very high estimate. This is fairly mild and not bad for such a high-profile page. I don't believe that any more than semi-protection is needed here. Christopher Parham (talk) 03:11, 7 April 2008 (UTC)
 * At our current rate of page views for this article, 10 minutes = 60 readers presented with vandalism rather than the article text. I don't see that as even remotely acceptable. Tim Vickers (talk) 15:49, 7 April 2008 (UTC)
 * "The section on semi-protection clearly covered pages that are "subject to heavy and persistent vandalism."". Surely the whole point of applying semi-protection is that it's supposed to stop the "heavy and persistent vandalism"?  If the vandal cannot be stopped this way, then why only semi-protect it?  What would be the point?  As for the "legality" of applying full protection: according to WP:PROTECT, full protection can be applied for "content disputes": and repeatedly replacing the entire page with Genesis sure looks like a "content dispute" to me! --Robert Stevens (talk) 20:53, 7 April 2008 (UTC)
 * This is a novel interpretation, but if we accept it then administrators should refrain from making any disputed changes while the page is protected (presumably, the dispute in this case would extend to any change beyond minor spelling corrections). Christopher Parham (talk) 02:04, 8 April 2008 (UTC)

MZM, you might have missed WP:IAR in your searching of policy. In fact, this is exactly the type of situation we have that for - we have a vandal who's acted beyond the ability of our normal procedures to handle. We have a solution which is better than the status quo. The fact that there's no policy in place to handle these extreme circumstances is no reason not to do what's best for the encyclopedia. --Infophile (Talk) (Contribs) 22:05, 7 April 2008 (UTC)


 * Very good argument Infophile!!I agree the "end product" gets the highest priority-after all it is an encyclopedia and not a bible in this instance. I agree with Tim that one reader no less sixty is too many readers to be ignored. I guess so much for "reliable" information, huh. We've entered a new age the "age of misinformation". Evolution is the process whereby God created all life. Sounds good! Print it! This guy is a sneaky little $h!t and he will return. We need to nip it in the bud. GetAgrippa (talk) 00:21, 8 April 2008 (UTC)

Are you really targeting this text at readers too dimwitted to realize that the vandalized version is not, in fact, the intended content? If so, you have written the article very poorly: such a reader wouldn't understand half the words in the lead. In any case, readers who aren't prepared to deal with inaccurate information, or the occasional vandalized version, simply shouldn't be using Wikipedia. The vast majority of readers are sensible enough to use Wikipedia appropriately, within the constraints necessitated by the limitations of the wiki process. Christopher Parham (talk) 02:04, 8 April 2008 (UTC)


 * We need to strike a balance here between inconveniencing of our readers and inconveniencing of our editors. For the last several years IPs have been blocked from editing the article, now these people are free to edit the draft article and have their edits incorporated into the main article. I see that as a large step forwards, done in a way that increases the reliability of this article for readers. This arrangement therefore provides benefits to both of the sets of people who access this article. Tim Vickers (talk) 03:02, 8 April 2008 (UTC)

I agree that the use of a draft article is the best option out of the current set available to us. Would it be possible to put a note on the main article advising readers that anyone can edit the draft (even IPs) and that useful content will be incorporated. At the moment this is only on the draft page and this talk. Cheers Suicidalhamster (talk) 18:48, 8 April 2008 (UTC)


 * The note is at the top of the article, but in hidden text, so you only see it when you try to edit. Tim Vickers (talk) 18:51, 8 April 2008 (UTC)


 * Only problem I see is that the edit tab instead reads "view source" here (not sure if admins still see it as "edit this page"), so a prospective editor may be turned off by that and never see the hidden text or the note on the talkpage. I don't think we'll be able to change that, though. However, one solution might be to modify the protection template specifically for this article so that on mouseover it mentions the existence of a draft article. How does this sound? --Infophile (Talk) (Contribs) 20:31, 8 April 2008 (UTC)