Tuesday, January 27, 2009

The Power of Darwin

 
Richard Dawkins has published an article in Free Enquiry titled The Power of Darwin. Here are the opening paragraphs.
Charles Darwin had a big idea, arguably the most powerful idea ever. A powerful idea assumes little to explain much. It does a lot of explanatory “heavy lifting” while expending little in the way of assumption or postulation. It gives you plenty of bang for your explanatory buck. Its Explanation Ratio—what it explains divided by what it needs to assume in order to do the explaining—is large.

Power of a theory = That which it explains/That which it needs to assume in order to do the explaining

If any reader knows of an idea that has a larger explanation ratio than Darwin’s, let’s hear it. Darwin’s big idea explains all of life and its consequences, and that means everything that possesses more than minimal complexity. That’s the numerator of the Explanation Ratio, and it is huge. Yet the denominator is spectacularly small and simple: natural selection, the non-random survival of genes in gene pools (to put it in neo-Darwinian terms rather than Darwin’s own).

Power of Darwin’s theory = The diverse complexity of life/Non-random survival

Natural selection is an improbability pump—a process that generates statistical improbability. It systematically seizes the minority of random changes that have what it takes to survive and accumulates them, step by tiny step over unimaginable timescales, until evolution eventually scales mountains of improbability and diversity whose height and range seem to know no limit.
Natural selection does NOT explain all of life and its consequences. A great deal of what we see in modern species is a consequence of accident and happenstance where contingency rules over natural selection. Natural selection does NOT explain diversity.

Darwin's big idea was to convince us that life has evolved rather than being created. That's a simple concept that explains a lot.

Natural selection explains adaptation. That's extremely important and extremely interesting but it's only a small part of evolution. Random genetic drift, which Darwin does not get credit for, explains much more because more of evolution is due to drift than to adaptation.

The contributions of Charles Darwin are enormous. That's why he gets credit for being the greatest scientist who ever lived. It does a disservice to his achievements to exaggerate them in celebration of the 150th anniversary of the publication of Origin of Species.


35 comments :

  1. Funnily enough, just before I came to look at Sandwalk, I was reading this paper on drift and speciation:

    Uyeda, J.C. et al. (2009) Drift promotes speciation by sexual selection. Evolution, advance online

    Quantitative genetic models of sexual selection have generally failed to provide a direct connection to speciation and to explore the consequences of finite population size. The connection to speciation has been indirect because the models have treated only the evolution of male and female traits and have stopped short of modeling the evolution of sexual isolation. In this article we extend Lande's (1981) model of sexual selection to quantify predictions about the evolution of sexual isolation and speciation. Our results, based on computer simulations, support and extend Lande's claim that drift along a line of equilibria can rapidly lead to sexual isolation and speciation. Furthermore, we show that rapid speciation can occur by drift in populations of appreciable size (Ne≥ 1000). These results are in sharp contrast to the opinion of many researchers and textbook writers who have argued that drift does not play an important role in speciation. We argue that drift may be a powerful amplifier of speciation under a wide variety of modeling assumptions, even when selection acts directly on female mating preferences.

    ReplyDelete
  2. How are you weighing selection and drift? What's the criterion to decide one process more important than the other?

    ReplyDelete
  3. Dominant relative frequency

    ReplyDelete
  4. I've read Sandwalk long enough to know that you think drift gets short shrift and I'm starting to agree but this confuses me again. When you say "Natural selection does NOT explain diversity" do you mean that it doesn't explain all of observed diversity or that it doesn't explain any diversity?

    ReplyDelete
  5. I mean that it doesn't explain a great deal of diversity, especially within species diversity.

    Natural selection eliminates diversity within a species. That's why the discovery of abundant diversity in the '60s was such a blow to the adaptationist position.

    ReplyDelete
  6. The most important scientist who ever lived was Issac Newton. Period, end of discussion.

    ReplyDelete
  7. The most important scientist who ever lived was Issac Newton. Period, end of discussion.

    Oh, what cobblers. Wrong about gravity, wrong about light, less good at calculus than Liebnitz.

    Still, quite useful at the Royal Mint...

    ReplyDelete
  8. I still put Einstein ahead of Darwin. Someone was inevitably going to come up with special relativity at some point (e.g. Poincare)...but no one but Einstein I think would've come up with general relativity.

    ReplyDelete
  9. Just ran through the uncorrected version of the Uyeda, et al article…

    I wouldn’t go so far as to say that drift explains “much more” of evolution than does NS; however I would agree that drift has a role, and on more-rare occasions - under certain conditions - a large one. A larger role under conditions such as those modeled in the Uyeda article; when low to moderately sized populations are allopatrically diverging “under favorable circumstances.” (Those favorable conditions listed on Page 7 of my copy)

    Although Natural Selection, when functioning within a species, pushes towards genetic uniformity, saying that it “eliminates diversity” requires the assumption that NS operates independently of sexual selection (both epigamic and intrasexual) and other evolutionary components. Indeed, one of Darwin’s “mistakes” may have been focusing too narrowly on the removal of variation – Natural Selection, red in tooth and claw – and ignoring the recombination power of sexual selection. To the best of my understanding, neither Dawkins nor “adaptationists” interpret NS has the sole and only driver of evolution – they just establish a different ranking of significance than do “pluralists.”

    The models depicted in the cited article are based on the assumption of selectively neutral preferences in the female line being drawn towards a terminal point for diversification. This assumption, though appropriate for the conducted research, is very much a reduction and simplification of the processes as they exist naturally. Through this single trait model, clarity and mathematical precision are obtained, but I’m certain that we all understand that selective pressures (from any mode of speciation) in the natural world are highly dynamic and exist along a continuum of multiple traits. Blue tail feathers are important, but you have to know how to sing and dance too! Nature utilizes a model based on the whole organism.

    As I think Carl may have been alluding to, placing values on the various modes and components of evolution and then establishing a relative ranking system is a difficult task. There’s tremendous disparity between mating and reproductive systems; species with more than two sexes, various methods of dispersal, environmental sex determinism, asexual reproduction, hermaphroditism, etc… I have doubts that any all encompassing rule like “Drift is more important” could be formulated as to be accurate with all taxa, and at all times.

    Although I’d be surprised if drift wasn’t a primary driver in at least some systems, I’m inclined to think that it spends the majority of its time piggybacking on NS and SS.

    ReplyDelete
  10. "How are you weighing selection and drift? What's the criterion to decide one process is more important than the other?"

    Like a number, huh

    "Dominant relative frequency"

    A dominant frequency can result from drift. To tell away from selection you need to make some assumptions about the structure of the population and know its history (which is not always possible)

    Measuring natural selection is a tough one, but it doesn't stop many from trying.

    A problem is that there is NO PRECISE LIMIT as to what kind of "trace" can natural selection leave on organismal form. To Dawkins, the answer is close to everything (some brits tend to get a little bit overdarwinian).

    Others think when comparing sequences betwen species, a rate of non-synonymous substitutions that is higher or lower than synonymous in some stretch of DNA is the mark of positive or negative selection. I'm not sure biochemical explanations have been explored enough. Plus, a great deal of studies finding selection this way use the wrong tests. There's lots of controversy with these studies, so watch out.

    Some paleontologists attempt to measure natural selection for instance in the decrease of the number of spines in sticklebacks in a well-documented population.

    To others, any adaptation is evidence for natural selection according to the mantra "Natural selection explains adaptation"
    Often, the origin of an adaptation is sudden, involving a single mutation, phenotypic plasticity and exaptation. In all, "positive selection" may result as consequence but it is not causal in the origin of the adaptation.
    These are well-documented empirically. However, the origin of a new adaptation by selective, directional accumulation of many mutations has resulted much more troublesome.

    To others, it is sufficient to point to any evolutionary transformation and coflate it with selection. We can for instance point to the sequence of changes in the foot in the lineage leading to the modern horse and pronounce that an example of "selection for running". You can point to the evolution of the vertebrate eye and pronounce it to be "selection for seeing". And so forth

    Some let Paley-phobia get the best of them and are ready to pronounce any biological complexity in itself as sufficient evidence for the action of selection.

    In summary, there are so many ways of "seeing" selection, it seems that we can always have people saying selection is the most important thing.

    ReplyDelete
  11. Dawkins seems to think SELECTION, the "picking out" from a given variety, is a process that in itself generates complexity. We are left to wonder, then, why natural selection has not made it into the books of physichal chemistry and thermodynamics. Can Dawkins, give us a single example of natural selection generating any complexity in a non-biological system (computer simulations don't count as they only exist in the context of the human organism)

    It seems there is SOMETHING about selecting ORGANISMS, isn't there.

    Has Dawkins given DEVELOPMENT two neurons of decent thought? There is quite a bit of biological complexity arising there, and it's no natural selection.

    ReplyDelete
  12. I am not sure what you mean Vargas. Natural Selection acts if you have survival and replication value. Computer simulations, viruses, memetics, and life all have this, so it is present. No one, including Dawkins, thinks Natural Selection would give birth to complexity from a system that does not have fitness via S&R value.

    And I also am confused at your second statement. Yes there is complexity in development, but how a species develops is largely based on it's genes and how they interact with the environment. This is largely determined by the species evolutionary history (selection + drift + flow , etc.). An organisms development doesn't just happen, it is based largely on it's evolutionary past.

    This said, Dawkins never said all complexity came from selection. Just that it is a source for diversity and complexity, which it is.

    I do wish he'd stop acting like selection is all that matters though. If you really ask him, he acknowledged full well that drift and the like play a definite role, but when he is talking to the general public he always leaves it out of the picture. I understand the use of simplification but this is far too much, it is just misleading.

    ReplyDelete
  13. Selection, as Larry Moran said, is not a source of diversity.

    ReplyDelete
  14. "No one, including Dawkin, thinks Natural Selection would give birth to complexity from a system that does not have fitness via S&R value"

    It also has to be a system whose variation includes increases of complexity. Selection only selects, increases of complexity require another explanation (In evolution, the greatest source of complexity is the combination of several organisms)

    "how a species develops is largely based on it's genes and how they interact with the environment"

    As simple as that? Don't you think you should mention everything else other than genes that's in the cell? It so happens this is not putty. All cells in your body have the same genes, despite having widely different cell types. How do you think that comes to be without cytoplasmic assymetries? Without considering cytoplasmic environments, your explanation of development is going to be pretty worthless.

    "This is largely determined by the species evolutionary history (selection + drift + flow , etc.). An organisms development doesn't just happen, it is based largely on it's evolutionary past"

    Jeez. Look, even if the past evolutionary history were explained by natural selection (which is not the case, of course), this does not mean that the increase of complexity as observed in development is occurring by selection. You've fallen in the ideologicla fallacy of "evolutionism". If you want to make sure, try convincing developmental biologists. See how useful they think your idea is.

    ReplyDelete
  15. These discussions tend to cause polarization and fundamental stances on both sides of the argument. Let’s bear in mind that evolution is a dynamic process, and that there isn’t a dichotomy in which we have to choose it’s only “this,” or only “that.”

    For clarity, “adaptations” are viewed from the rear view mirror only; they are the byproducts of selection. In fact, every generation of an organism that has survived –avoided elimination – can be said to have “adapted” or been selected. There is no “picking out,” teleological goal or forward looking purpose in selection.

    To list barriers, contingencies or other considerations, such as embryological development, cytoplasmic environments or pure chance (catastrophe), as somehow an argument against natural selection is meaningless and only confuses the issue.

    Genetic drift is equitable to sampling error. If a population finds itself isolated, bottlenecked or decimated by a catastrophic event, its allele frequencies may be of different ratios than those that existed in the pre-catastrophe population; because the survivors are a random sample from that original. Such a change in allele frequencies (in part, or their entirety) may ultimately, if passed on to the next generation (SELECTED), result in speciation. Here, drift generated novel conditions; however that novelty can only be said to have contributed to diversity if it becomes fixed – fixed via selection. Diversity arises from evolutionary processes – not from sampling error.

    Asteroids and orogenies may cause selection pressures to shift in one direction or the other, but their very presence says nothing about fundamental evolutionary processes – they might even start the game anew, but it’s the same game!

    ReplyDelete
  16. Natural selection eliminates diversity within a species.

    Granted, but can't natural selection create new species and increase inter-species diversity? It's a forest-vs-trees view, isn't it?

    Dawkins is talking about explanatory power and one of the key features of life which needs explanation is the appearance of design and the diversity of species rather than the diversity within a species. To that extent, focusing on natural selection sounds valid. Perhaps it is somewhat overstated but I'm not sure it's as wrong as people are making it out to be.

    Of course maybe I don't understand just how significant genetic drift is.

    ReplyDelete
  17. I'm not an evolutionary biologist and have only progressed beyond Popular Science accounts relatively recently. Here's some research that I've found interesting that relate to the matters being discussed here. Hopefully they'll be of use (particularly to laymen like myself).

    Andreas Wagner recently published a review trying to reconcile neutralism and selection:

    Wagner, A. (2008) Neutralism and selectionism: A network-based reconciliation. Nature Reviews Genetics 9, 965-974

    http://www.bioc.uzh.ch/wagner/papers/NRG08.pdf

    I've also enjoyed Michael Lynch's recent work:

    Lynch, M. (2007) The evolution of genetic networks by nonadaptive processes. Nature Reviews Genetics, 8, 803-813.

    Lynch, M. (2007) The frailty of adaptive hypotheses for the origins of organismal complexity. PNAS, 104, Suppl: 8597-8604

    Lynch, M., and J. S. Conery. (2003) The origins of genome complexity. Science, 302, 1401-1404.

    These can all be found on his website:

    http://www.indiana.edu/~lynchlab/LynchPublications.htm

    Finally, Austin Hughes has published some important papers showing that some of the evidence for positive selection may be seriously flawed. His papers don't appear to be widely accessible for free, but I do recommend the following if people can get hold of them:

    Hughes, A.L. (2008) Near Neutrality: Leading Edge of the Neutral Theory of Molecular Evolution. Annals of the New York Academy of Sciences, 1133, 162-79.

    Hughes, A.L. (2008) The origin of adaptive phenotypes. PNAS, 105, 13193-13194.

    Hughes, A.L. (2007) Looking for Darwin in all the wrong places: the misguided quest for positive selection at the nucleotide sequence level. Heredity, 99, 364–373.

    a couple of quotes from the latter paper:

    For the past 20 years, there has been a tendency on the part of journal editors and reviewers to assume that every case of alleged statistical evidence for positive selection is worthy of publication, even in the absence of a plausible biological mechanism underlying the alleged selection.

    and

    Reasoning that evidence for positive selection can be obtained from the relative numbers of synonymous and nonsynonymous substitutions, biologists developed a number of new tests for positive selection. Unfortunately, the most widely used such approaches are fundamentally flawed as tests for positive selection, because they do not effectively rule out alternative interpretations (Hughes et al., 2006).

    ReplyDelete
  18. I take Darwin over Einstein any day of the week. The many waited decades to release his theory. Thats a patience and understanding that only Darwin could have.

    ReplyDelete
  19. The many waited decades to release his theory. Thats a patience and understanding that only Darwin could have.

    Well, no. He only rushed to publish when he realized Wallace was going to get the jump on him. In that sense, he was a little too patient.

    ReplyDelete
  20. Johnny says,

    These discussions tend to cause polarization and fundamental stances on both sides of the argument. Let’s bear in mind that evolution is a dynamic process, and that there isn’t a dichotomy in which we have to choose it’s only “this,” or only “that.”

    Spoken like a true pluralist!

    Please join me in opposing all those who would reduce evolution to just a single process; namely, adaptation.

    ReplyDelete
  21. Tyro says,

    Granted, but can't natural selection create new species and increase inter-species diversity? It's a forest-vs-trees view, isn't it?

    I suppose that natural selection can "create" new species. I'm not sure if it's the dominant mechanism of speciation.

    Speaking of forests. I'm told there are thousand of different species of beetles in the Amazon. Many are from the same genus and the different species can be found on adjacent trees. Are all of these species adapted to a particular small part of the tropical rain forest, or is this an example of drift?

    Dawkins is talking about explanatory power and one of the key features of life which needs explanation is the appearance of design ...

    When you are talking about things that look designed then natural selection is the answer.

    However, the bigger question is whether everything looks designed. When I look at living things I'm impressed by how poorly designed many of them are. In most cases, I could do a much better job than evolution (or God). Natural selection isn't very good at explaining poor design.

    So the fundamental question is how much of life looks designed. To Dawkins it's almost everything and that's why he advocates a dominant role for selection and adaptation.

    ... and the diversity of species rather than the diversity within a species. To that extent, focusing on natural selection sounds valid.

    Maybe I'm just stupid but to me natural selection doesn't explain the diversity of species and focusing on that single mechanism isn't very good science.

    ReplyDelete
  22. Carl Zimmer asks,

    How are you weighing selection and drift? What's the criterion to decide one process more important than the other?

    I've answered that question many times. I count up the number of fixed nucleotide substitutions that have occurred in a particular gene lineage then I try and decide which ones are adaptive and which ones aren't.

    In all cases the non-adaptive mutations seem to vastly outweigh the adaptive ones. Thus, random genetic drift is much more common than natural selection as a mechanism of evolution.

    It's shame that Richard Dawkins has given much the same answer in some of his writings but he then goes on to claim that the only interesting part of evolution is adaptation so random genetic drift doesn't even count as evolution.

    What's interesting to Richard Dawkins is not the only way to define evolution. I happen to be very interested in the kind of evolution where selection is irrelevant. This is the domain of molecular phylogeny with its approximate molecular clock that cannot be explained by natural selection.

    I wish Dawkins would add a note to every article he publishes making it clear that all of his statements are based on the premise that evolution is defined as that which is interesting to Richard Dawkins.

    It would make things much clearer.

    By the way, Carl, have you ever asked Richard Dawkins the question you asked me? :-)

    Care to share his answer?

    ReplyDelete
  23. I'm told there are thousand of different species of beetles in the Amazon. Many are from the same genus and the different species can be found on adjacent trees. Are all of these species adapted to a particular small part of the tropical rain forest, or is this an example of drift?

    Very good point. You've convinced me I was wrong, and that I was inadvertently focusing on a few specific examples rather than on evolution in general. Thank you.

    ReplyDelete
  24. Larry,

    You’re correct; I would be best classified as a pluralist. Multi-mechanistic is the way to go – all of the cool kids are into it, and NS doesn’t operate in a vacuum. I was under the impression that we were discussing relative contributions to diversity, and evolution as a whole. I’d never intentionally imply that drift doesn’t exist – I’d be surprised if Dawkins did either. If it’s a matter of scale however, natural selection wins hands down in my assessment! Does that make me an adaptationist too?

    I’m a little confused by the beetles in the rain forest example… If there are separate species on each tree how is it that a “founding few” became sexually isolated without passing-on a phenotypic sexual incompatibility to subsequent offspring? Sexual incompatibility (rather it is morphologically or ethologically based) rising from some prezygotic or postzygotic barrier preventing cross breeding between groups sounds a lot like a phenotype to me.

    If this phenotype isn’t eliminated, but rather passed on – it has been Selected. Even if drift seeds the novelty, it can only be fixed IF it’s selected.

    ReplyDelete
  25. How do you pluralists feel about self-organization?

    ReplyDelete
  26. Maybe I'm just stupid but to me natural selection doesn't explain the diversity of species

    If there are lots of environments to select for, why can't natural selection (operating on the genetic diversity within species created by mutations) be an engine of speciation, thus diversity of/among species? If you mean it doesn't seem to explain *all* the diversity, well:

    and focusing on that single mechanism isn't very good science.

    Hard to argue with. ;-)

    Even if drift seeds the novelty, it can only be fixed IF it’s selected.

    Are there any forms of contingency you *wouldn't* consider selection?

    ReplyDelete
  27. "How do you pluralists feel about self-organization?"

    Larry probably thinks it's wahooey because he thinks he already understands evolution with no need for that notion. He thinks all he neeeds to do is think about gene frequencies and poulation (He thus ignores development, too)

    Self-organization as an organismal property is pretty much a factI think it is much more on the right track than natural selection, as far as understanding the origin of complexity. Think about homeostasis, regeneration, development.

    ReplyDelete
  28. One thing I think we might agree upon: natural selection does not explain everything it is claimed to explain.

    "When you are talking about things that look designed then natural selection is the answer"

    "looks designed" is a pretty wishy-washy notion to adscribe to a specific process like natural selection. What is "looking designed"? Snow crystals seem perfectly designed. And guess what? natural selection does not shape Snowflakes. Nor are they being put together by some intelligent being. This is a topic for physical chemistry, thermodynamics.

    These thing also occur in organims and may end up "looking designed". Emerging order, dissipatory structures, laws of physisc and geometry can all add to what someone may consider "looks designed"

    The problem rappears when we want there to be a single mechanism that trivially explains complexity or an 'appereance of design", such that we can simply blame it. Many processes other than selection are there, yet selection gets credit for all. This is very poor science. Ultimately, it is looming creationism that is getting all the attention.

    ReplyDelete
  29. Larry--Thanks for your clarification. I'm a bit slow keeping up with this comment thread (I blame Twitter).

    In answer to your question, I have not asked Dawkins about this (or anything--never spoken to him). If I had to guess, I'd speculate that he and a number of other evolutionary biologists would say, "Sure, a lot of the nucleotides you see fixed in genomes today are the result of drift, but when we say 'evolution,' we really care about is the origin of adaptations and speciation." And in those cases, a number of researchers (but not all evolutionary biologists) have made arguments that natural selection is key.

    As a writer, I find that I often have to get biologists to define their terms before I can compare their ideas. Otherwise, I end up with apples and oranges.

    ReplyDelete
  30. Johnny says,

    If it’s a matter of scale however, natural selection wins hands down in my assessment! Does that make me an adaptationist too?

    Yes, probably.

    I’m a little confused by the beetles in the rain forest example… If there are separate species on each tree how is it that a “founding few” became sexually isolated without passing-on a phenotypic sexual incompatibility to subsequent offspring?

    If they are true biological species—which is doubtful—then what's the problem? We've already talked about this in previous postings. There has to be some phenotypic change that makes the two varieties incapable of interbreeding and it's far more likely that this became fixed by random genetic drift than by natural selection.

    Sexual incompatibility (rather it is morphologically or ethologically based) rising from some prezygotic or postzygotic barrier preventing cross breeding between groups sounds a lot like a phenotype to me.

    I agree. What's your point? Are you claiming that all phenotypic differences have to be subject to selection?

    If this phenotype isn’t eliminated, but rather passed on – it has been Selected.

    Wrong. You seem to lack understanding of how alleles can be fixed by random genetic drift.

    Even if drift seeds the novelty, it can only be fixed IF it’s selected.

    That's so wrong that I hardly know where to begin. Please read Random Genetic Drift.

    ReplyDelete
  31. Carl Zimmer says,

    In answer to your question, I have not asked Dawkins about this (or anything--never spoken to him). If I had to guess, I'd speculate that he and a number of other evolutionary biologists would say, "Sure, a lot of the nucleotides you see fixed in genomes today are the result of drift, but when we say 'evolution,' we really care about is the origin of adaptations and speciation." And in those cases, a number of researchers (but not all evolutionary biologists) have made arguments that natural selection is key.

    I agree with you that this is what Dawkins would say.

    Here's the problem. Most evolutionary biologists have a common definition in mind when they talk about evolution. It's the one that's in all the textbooks and it's the one I decribe in my essay What Is Evolution?.

    Evolution is a process that results in heritable changes in a population spread over many generations.

    This is a minimal definition of evolution that covers any type of genetic change whether it is due to random genetic drift or natural selection, or anything else, for that matter. When a scientist writes about evolution, this is what most other scientists will assume to be the definition.

    Dawkins writes about a subset of evolution; namely, the parts that he is interested in—design and adaptation. There's nothing wrong with that as long as he makes his bias clear to all readers. But when he says, "Darwin’s big idea explains all of life and its consequences, and that means everything that possesses more than minimal complexity," he implies that natural selection is the sole explanation for all of evolution.

    That's very different than saying that Darwin's big idea explains everything that Richrad Dawkins cares about.

    I find it annoying that when a pluralist argues that there's more to evolution than adaptation they are asked for supporting evidence but when an adaptationist claims that natural selection explains all of life nobody asks them to support their claim with evidence.

    ReplyDelete
  32. I completely overlooked your excellent "What is Evolution?" piece--thanks for linking to it!

    ReplyDelete
  33. Larry,

    Thanks for pointing me to the Random Drift article; I’m relatively new here so I may need to work through the archives to truly understand your perspective.

    I don’t think that I have any serious disputes with Random Drift as detailed at the provided link. The article gives several examples of bottlenecks; elephant seals, cheetahs, etc… That’s all well and good, but what wasn’t detailed by the author is the idea that drift leads to SPECIATION more frequently than does selection or other evolutionary mechanisms, which seems to be what you’re implying here.

    So, when you comment that, “There has to be some phenotypic change that makes the two varieties incapable of interbreeding and it's far more likely that this became fixed by random genetic drift than by natural selection,” are you saying that the “phenotypic change” may have been created by a novel mutation, for example, and that it spreads through the population by way of drift – OR – are you saying that both the “phenotypic change” and its spread arise via drift?


    Larry asks, “Are you claiming that all phenotypic differences have to be subject to selection?

    No, I’m not making that claim. In regards to phenotypes, I see three possibilities:
    1. The phenotype is beneficial
    2. The phenotype is neutral
    3. The phenotype is detrimental

    1. If the phenotype is beneficial, the organisms possessing it will experience increased fecundity and fitness and pass the trait on to offspring in greater numbers than those not possessing the trait. Adaptation is the word given to describe improved efficiency in an ecosystem – a better fit. The non-random accumulation of beneficial phenotypes is called selection. Both adaptation and selection are present in this beneficial phenotype scenario.

    2. If the phenotype is neutral, the organisms possessing it will not experience any change in fecundity or fitness as a result of the trait, it’s neutral – not subject to selection. The neutral phenotype may still increase its concentration in the population but only randomly or if it happens to be present in organisms that survive due to other traits that are not neutral. This is called genetic drift; it’s a random and neutral process that is in no way linked to fitness.

    3. If the phenotype is detrimental, the organisms possessing it will have decreased fecundity and fitness and will therefore have less offspring than those not possessing the trait. The non-random removal of detrimental phenotypes from a population is called selection.

    So when I said, “If this phenotype isn’t eliminated, but rather passed on – it has been selected.” I was equating “phenotype” to scenario 1 and 3 above.

    Contrastingly, my read of you is that you are asserting that scenario 2 leads to sexual isolation without piggybacking on 1 or 3 to get there. My argument is therefore, that if this is true, it must be rare. Rare because it’s a reproductive barrier created without having a link to sexual fitness. In short, this seems a dramatic result from simply increasing concentrations of already existing alleles.

    Of course all this is pointless if we’re only talking about nucleotide bases… You can’t discuss natural selection without talking about phenotypes. If you’re argument is that genetic drift is a greater contributor to evolutionary processes because you’ve counted the fixed nucleotide substitutions, that seems reasonable enough and I’ll take your word for it. But, if you are saying that those substitutions you counted are somehow equivalent to greater frequencies of speciation, sexual isolation or phenotypic change exhibited in nature, I’ll wait for the published research.

    ReplyDelete
  34. SteveF, thanks for posting those references. Otherwise: gah!!! So many straw-figures, so little time.
    Imperfect design is hardly an argument against the importance of selection; evolution mostly tinkers and everybody acknowledges the importance of developmental constraints and historical contingency. Moran is talking about counting nucleotide substitutions and Dawkins is talking about phenotypic evolution at the population level, and they're both talking past each other. That Moran's favorite population-genetics-based minimum definition of evolution is in the textbooks does not make it the only interesting level at which evolution can be studied, any more than Dawkins' writings argue against random nucleotide substitutions. Dawkins' (and, I confess, mine) level of interest is organisms in populations. Adaptation is an emergent property of populations; to trace the "origin of an adaptation" to an individual mutation does not address the population-level cause of fixation. And yes! Every phenotypic variant is exposed to selection. Either it is statistically correlated to relative reproductive success (positively, negatively, or quadratically)--in which case its change in frequency is due to selection, or not...drift. You can then ask whether any correlation is causal (direct selection) or correlational (indirect), etc. There is then the separate question of the phenotypic trait's heritability, and the resultant genetic response to selection. Its a case-by-case thing, and your Amazonian beetle question makes no sense. (As a pluralist, you ought to be open to the idea of "sympatric" speciation through host-plant switching; selection is hardly irrelevant to that process.)
    gah! No time for this!

    ReplyDelete
  35. "Moran is talking about counting nucleotide substitutions and Dawkins is talking about phenotypic evolution at the population level, and they're both talking past each other"

    Not quite. Larry does argue phenotypic traits are neutral too (tongue-rolling). His example of beetles reflects his belief that much non-adapative evolution occurs.
    I think you should take a closer look at the peculiarities that separate closely related species. Many of them make very little adaptive sense.

    Still, I can understand some frustration at the beetle example. The relationship of insects to different plant-hosts (true "microenvironments") is crucial to their speciation. However, it is unrealistic to describe this evolution as directed by darwinian gene selection. The forefront of this speciation process is behavioral and developmental plasticity (e.g switching hosts).
    The kind of thing Larry and many others are willing to leave out of evolution as "non-inheritable", or "non-selectable"

    ReplyDelete