What exactly is natural selection and precisely where does it occur?

Unless there was some other way of differentiating the marbles (for instance, a sensor that selected for specific light wavelengths - color, or by weight, or something), then there is arguably no selection going on. Once again, we're pushing the illustration beyond what it is intended to convey, unfortunately. Living organisms and their environments are obviously more complex, so trying to stretch the analogy to encompass those complexities isn't going to work. As long as I was able to clarify the analogy (as far as it goes), then that works for me.

Hoot, forget Hardy-Weinburg. You'd have a much better case if you could show ONE SINGLE SOLITARY real-world natural sexually-reproducing population where HW is even relevant, let alone descriptive. Don't get wrapped up in theoretical mathematical constructs - it doesn't help the discussion.

I never said HW wasn't important in its particular context (all the nice equations generated by population geneticists describing nice neat theoretical - read non-existent - populations). Science needs theoretical constructs, and they truly have their place. I don't know where ecology would be today without the abstruse mathematics of Wilson/MacArthur's equations, for instance. Probably still back in the amateur stamp-collecting of natural history. However, they would not be my first choice of method for explaining why there are more endemics on oceanic vice continental islands, for instance - even though they were created specifically for this (among other things).On the other hand, what I was trying to point out is that HW has actually no relation to the real world except in the abstract. If you (that's a generic "you", not Hoot Man necessarily), want to try and explain natural selection to someone who doesn't know much about it, trying to simultaneously explain HW equilibrium and the difference between a mathematical model and reality AS WELL as trying to get them to understand NS would seem to be the most difficult way of going about it that I can imagine. So my advice, don't get wrapped up in the math until you are sure you grasp the fundamental concepts.Of course, all you have to do to prove me wrong is present one single study of a real-world, sexually reproducing population that HW actually describes.

Yes, I know you've said that repeatedly. I'm not sure you've actually gone so far as to demonstrate that contention, although you may believe you have.Since you prefer the reductionist viewpoint, riddle me this: how does one predict the fitness of an organism based solely on the activity of a gene or suite of genes? After all, the properties of an organism - its phenotype - are not simply the sum of the properties of its individual genes, correct? Is the fitness of an organism in its particular environment dependent solely on genetics, or is fitness a phenotypical characteristic? Do allele frequencies in a population change over time because of the activities of genes (IOW, is the gene an agent), or because of the phenotypical effects? Is it possible (outside of a purely theoretical mathematical construct) to calculate the fitness of an allele without merely abstracting from the measurement of fitness of an individual organism? Do genes play a causal role in fitness?ABE: OF you don't want to tackle the above in this thread, how about a simpler question. How does the genecentric view allow one to distinguish between selection and response to selection?Further to what I outlined for Hoot above with my little toy example, I readily agree there is selection of genes, suites of genes, alleles, etc. But what is being selected for is the phenotype - the expression of those genes, etc, in the individual. Edited by Quetzal, : added a bit easier question

Okay. Let me know when you're going to do that. Not that I necessarily support Gould, but why would the questions I asked be based on a misunderstanding of genecentrism? These are questions to which phenotpyical selectionists already have an answer, but couched in genecentric language. In other words, all I'm asking you to do is explain the implications of fitness calculations based on the genecentric viewpoint. You can call individual selection "inferior" all day long, but you still need to address the questions directly if you want this categorization to be taken seriously. Wow, this smacks of genetic determinism. Is that really the position you wish to advocate? I will absolutely guarantee you that Dawkins doesn't hold this view. Maybe you should expand your answer a bit? For clarity, I mean.This response especially breaks down in the face of behavior - which is Dawkins speciality, btw. Once you get beyond pure genetically programmed behaviors (such as many insects - like your ants - exhibit), the role of genes in behavior gets less and less. As Ehrlich puts it (paraphrasing), "genes don't shout commands, at most they whisper suggestions". To me, your response indicates you feel otherwise. Like I said, you should probably clarify. Do you consider behavior a phenotypical trait that has a fitness implication? If not, why not? If so, how does that square with genecentrism? Wait a second. This doesn't really make sense to me. Please explain what you think the relationship between selection and fitness might be? Perhaps that will clarify your position for me. 'Cause this appears at first glance to be moving WAY beyond any definition of fitness I've ever encountered. Is the "no" in response to "phenotypical effects" or is the "no" in response to the idea that a gene/allele is an active agent? 'Cause if the latter, you directly contradict yourself in the explanation. "Ability of the allele to...work with other genes" is pretty much the definition of "allele is an agent". Disentangle this for me, will you?Let's take a look at a real-world example to see how this works, shall we? As everyone knows, sickle-cell is a combination of two different alleles AA and SS. Homozygote AA's and SS's tend to die off before achieving optimal reproduction, although SS's produce far fewer offspring than AA's. AS heterozygotes, although mildly deleterious, do tend to produce many more offspring than either of the others. Thus the fitness of the three goes something like AS > AA > SS. However, these fitnesses result from the phenotypical effects of the combination of these alleles - effects that exist ONLY in organisms. If your interpretation of Dawkins' concept is correct, how, if the gene is not a causal agent in its own right, do these combinations "work together" - and why would they? Help me out, here. Unless you're trying to say (which I know you're not) that there's a bunch of loose genetic material floating around out in the wild replicating itself, you have not been able to divorce the measurement of a gene's fitness from the measurement of the organism's fitness. Even taking into consideration the contribution of the particular allele to the survival/reproduction of the organism, you still - except in the abstract, as an average over generations - haven't shown that it is possible to discuss gene fitness without recourse to the organism."I only touched on the issues..." is right.

No, I'm saying that no real world population has EVER exhibited anything resembling HW equilibrium in the wild. That HW isn't even intended to describe the real world. If you believe otherwise, then show me the study.Now back to your originally scheduled topic.

Right. I certainly haven't read anywhere that fitness is "just fecundity", so I'm not sure where you got that. Unfortunately, we're coming up to the end of this thread, so a long digression here may be counterproductive. Without getting into a very long discussion, suffice for the purposes of this thread I define fitness as the average lifetime contribution of individuals posessing a particular genotype to the population after one (or more) generations. In other words, not just the numbers of offspring that an individual can pump out, but the number of offspring carrying a particular genotype that themselves live to reproduce. So, simplistically, fitness of a genotype = (average fecundity) X (fraction surviving).Obviously, there is a lot of detail and nuance (relative vs absolute fitness, for instance) that I'm leaving out, but that's the gist, and probably sufficient for this particular discussion.

I wasn't aware that this was a challenge. I know that it was discussed earlier, but I didn't jump in there simply because that wasn't part of what I was arguing with you about. Unfortunately, as I mentioned to Justin (were you aware you had a gencentrist supporter on this thread?), we're coming to the end of the thread. Rather than move off into yet another side discussion, we should be summing up. I'd be happy to discuss this particular issue elsewhere, however. This doesn't answer the question, Mod. Come on, you can do better than that. Why are the questions bogus? How do the questions demonstrate a lack of understanding? You seem to be equating "disagreement" with "lack of knowledge". The two are not synonymous. I understand genecentrism. I simply don't agree with the approach. Merely denigrating the phenotype viewpoint doesn't serve to validate the "superiority" of genecentrism. Look, you're not Dawkins, and I'm not Gould. That being the case, I would have thought we could have a discussion of this topic without the personality aspects they bring (brought) to the table. It doesn't appear you agree. Pity.I understand why you're side-stepping the fitness questions. It is one of the key problems with genecentrism. The viewpoint provides a nice, very useful metaphor for looking at evolution. It absolutely bites rocks when you're trying to look at natural selection in isolation. As you've demonstrated in this thread, under genecentrism the two concepts simply cannot be divorced from each other. It makes absolutely NO distinction between selection (NS) and adaptive response (evolution). It can't. It isn't set up to be able to do this. It is an evolutionary metaphor. Justin mentioned Ancestor's Tale, and I would submit River Out of Eden as well, as examples of why genecentrism and fitness are incompatible concepts. This does NOT mean that genentrism is "wrong". It isn't. It does mean, however, that there are questions which it simply doesn't (or isn't designed to) address effectively. One of them being the topic of this thread. Where? Can you point to the post(s) where you discussed fitness in the context of genecentrism? I must have missed it/them. If I believed you had already "answered" the questions, I wouldn't have posted them - we'd already be discussing those answers. I'm not some ignorant creationist you can blow off with this type of one-liner. What? This makes no sense, and certainly doesn't clarify your previous response that I wanted you to expand upon. Here it is again, for reference: This is genetic determinism. I don't think you advocate this (or do you?). This is what I questioned you about. Your statement "fit individuals don't get naturally selected" doesn't even relate to the previous statement. I'm not trying to trap you here - I really am trying to understand what you're talking about. I'm not getting it, and so asked you to explain. Rather than explain, you pumped out an unrelated one-line soundbite. What's wrong, Mod? This isn't like you. This doesn't answer the question I asked. Or at best, only answers it partly. Focus on behavior, please.In addition, you've once again pointed up the main problem with genecentrism: the inability to separate natural selection from evolution. If you substitute the latter word in the last sentence you wrote, then I would agree with you. However, as written, the sentence is simply wrong. No, I'm not. I asked you to explain the relationship between selection and fitness in the context of genecentrism. Of course, you can also take the tack that "fitness" is meaningless in that context, but in which case you've got a pretty tough row to hoe. This points up another problem I have with the utility of the genecentric viewpoint. Proponents are apparently incapable of describing anything without recourse to (IMO) confusing, ambigious, and misleading anthropomorphic language. Can you possibly provide a more technically sound response to my question - leaving off the metaphors and anthropomorphisms? Thanks. I understand what you meant in the last sentence, even though you wrote it like you were agreeing with me at last - I know that's not the case. Be that as it may, this doesn't address my point. The claimed utility of genecentrism is that it provides a better explanation of what happens under natural selection than does the phenotypical viewpoint. However, if genecentrism is incapable of measuring or discussing fitness without recourse to measures of fitness of the individual organism, then I submit that all it's doing is adding an additional level of abstraction that doesn't advance our understanding. In fact, you've as much as said that fitness is an organismal (read, phenotypical) characteristic, but that anything that isn't genetic is ignored under genecentrism.Mod, I don't really understand what the problem is on this thread. You are normally one of my favorite posters: well read, well written, and incisive. You apparently don't consider me worth making that effort. If you are unable to provide more than off-the-cuff one-liners - which moreover are in essence no more than simple restatements of your position - in response to what I write here, I think I better simply bow out of this thread and let you persue your side discussion with Percy in the remaining few posts. Hopefully you're just having an off thread.Edited by Quetzal, : Fixed bad ubb coding

Although I'm not entirely sure of the purpose it would serve, since Mod took the time to summarize his position, I suppose (in the absence of anyone else interested in so doing) I better summarize the opposing viewpoint. Some of the points below have been discussed in this thread, others may not have been directly addressed.1. With the possible - albeit poorly understood and hence disputable - exception of the phenomenon known as segregation distortion (aka, meiotic drive), there is no known selection pressure that operates directly on a gene, suite of genes, or even genotype. Rather, selection affects the entire organism through either enhancing or restricting individual survival (the carrier of the gene), or over the longer term, the reproductive success or rate of the individual organism. Whereas a gene or polygene can be said to be "promoted" by selection if it increases in frequency in subsequent generations, it is actually the phenotypical expression of the gene that is affected. Since phenotype is ineluctably and directly tied to the interaction of the whole organism and its environment, attempting to understand this interaction in the context of genes (rather than individuals) tends to distort and/or add an unnecessary level of abstraction when discussing natural selection itself.2. Genecentrism is in essence an evolutionary metaphor. As such, the viewpoint is not designed to - and thus cannot - make a distinction between selection itself and the evolutionary response to selection. Thus genecentrism is unable to address a whole gamut of interesting questions concerning the ecological effects on individuals, populations, etc, of environmental factors in a single (or a limited number of) generations. These questions are, in fact, dismissed as "uninteresting" in an evolutionary context. I submit, however, that for numerous disciplines such questions are not only not "uninteresting", but actually key to our understanding of nature and natural history.3. Genecentrism dismisses most concepts of "fitness" as an organismal measure. At best, it allows only discussion of the relative fitness of particular alleles, and ignores the concept of absolute fitness. However, even here, there is really no effective way for genecentrists to measure such allelic fitness without implicitly or explicitly referring to the effect of those alleles on the individual which carries them - because, after all, it is the individual which passes those alleles on to subsequent generations. In other words, again we are adding an unnecessary level of abstraction to the discussion. Outside of the discipline of population genetics, this abstraction serves no function other than to obscure what occurs in nature.4. There are a number of biological sciences where it simply makes no sense to insist on a genes-eye-view, primarily because these disciplines are more holistic in nature. The one that immediately springs to my mind is (of course) ecology. With the exception of those genetically-related elements such as certain aspects of population dynamics (for instance, inbreeding depression) and extinction or extinction threat (for instance, the phenomenon of extinction vortex), the insistence of the proponents of genecentrism that theirs is the only "accurate" viewpoint would perforce render much of ecology unworkable - or perhaps only "uninteresting" - because ecology is concerned primarily with macroscale (vice gene-level microscale) interactions. Since ecology is arguably the science most concerned with natural selection (writ large) in the wild, the fact that genecentrism is not particularly useful - and in fact may be unusable in many contexts - for that science should give the genecentrists pause.5. And finally, on a very personal note, although I am aware there are many scientists - especially evolutionary biologists - who are very capable of discussing genecentrism in technical terms, there seems to be a nearly inescapable tendency even among themselves to use highly anthropomorphic language. Phrases like "selfish gene", "cooperative genes", and "A fit individual is one which works towards ensuring its alleles replicate." do nothing for understanding the actual biology of what is occurring. In fact, I submit that the terminology used in these cases is not only "inaccurate", but misleading, ambiguous, and obfuscatory.For these reasons, and others, I feel that the genecentric viewpoint - although having a place in evolution - is NOT the best and most accurate way of either looking at or describing natural selection.And may Darwin have mercy on us for wasting 288 posts on this subject.