Evolution Requires Reduction in Genetic Diversity

Absolutely. I was isolating the effect of inbreeding from other factors to show that inbreeding alone did not change allele frequency. Maybe I did get a little technical for this discussion, but I thought it was important to make the point that inbreeding is not what changes the proportion of alleles; it is the other evolutionary factors - such as drift. See Message 605 and Message 615.Another point to make is that in plant breeding, when a particular trait of interest is found, breeders will produce a highly inbred line (7+ generations) starting from a single individual or a small group of individuals. The point is to produce a population that is highly homozygous (enough inbreeding and they can be homozygous at virtually every loci). They then use this population to do QTL studies and identify what marker the particular trait is associated with. Interestingly, no evolution occurs in these inbred lines. Why? because they are shielded from drift, selection and migration (you can't prevent mutation)(some animal lines such a lab mice as well are highly inbred, but I am not as familiar with animal breeding programs)HBD

Congratulations, Faith, you have just discovered the founder effect! This topic got started because you want to take known evolutionary processes and ignore certain parts of them and redefine others so you don't have to acknowledge that what you are talking about IS evolution. You wanted the founder effect to be based on inbreeding alone. I made a simple, straight-forward comment about inbreeding not changing allele frequency but simply shuffling them into homozygotes and that in order for allelic frequency to change there had to be other evolutionary forces at work - like drift. But that sounds too much like evolution to you. It IS important to your argument. You don't want to acknowledge that mutations can add genetic diversity to a population so you require genetic incompatibility to come about by shuffling alleles around into different genetic combinations. Or do you have another mode of incompatibility in mind?HBD

Maybe not, but there are places where using the term "gene frequencies" has been terribly confusing. Allele frequency is much clearer. Genotypic frequencies should be used when talking about combinations of alleles. Oh, this is definitely true. Some genes, such as ribosomal genes, are particularly prone to variable copy numbers. Sub-species level organization is often identified by these types of copy number variations or extra genes, for example a pathovar may have a gene that allows it to disable the defenses of a particular host that no other members of the species has.I would just point out that to say "individuals in a population can differ as to which genes they have and how many" could give the wrong impression. For the most part genomes are very stable and we would not expect this kind of variation at the individual level. Where we usually see it is at the sub-species level. Of course, this sub-species variation undoubtedly arose from an individual, so it is not an incorrect statement.HBDEdited by herebedragons, : spelling

There are three basic categories of reproductive isolation,1) Ecological or geographical isolation
2) Prezygotic isolation - ie. different pollinators, different breeding patterns/timing, different mating habits (bird songs/displays)
3) Postzygotic isolation or hybrid incompatibilityWhat we seem to be focused on here is postzygotic isolation, which seems like it would be the most important barrier because of our general concept of species, however, it turns out that postzygotic factors usually play little role in speciation. Ecological isolation is of course the big player in allowing two subpopulations to diverge, but prezygotic isolation is really the major factor in the reproductive isolation of two subspecies - at least initially. In the case of the greenish warblers, IIRC, where the two ends of the ring meet the two species have different songs and so do not recognize each other as potential mates. Could they produce viable hybrids if forced to mate? Maybe, but does it matter?This is really a huge subject and maybe we could spin this off as a separate thread? Genetic incompatibility can come from a number of things. It could be from incompatible genes or from failure of the chromosomes to form viable pairings during recombination or from changes in regulatory networks. (I suspect that the latter to be the most common).So in theory, one mutation could result in prezygotic or postzygotic incompatibility but it is highly unlikely. It more probable that it is due to a series of incompatibilities that accumulate.ABE: I should make it clear though that I agree that shuffling alleles around is not sufficient to cause genetic incompatibility and that is the point I have been making all along. Introducing a new gene could theoretically do it, but that would require mutations not just working with the same alleles or a subset of them. /ABEHere is a good paper by Allen Orr that discusses the Dobzhansky-Muller Incompatibility. Look especially at figure 1 which shows how incompatibilities increase rapidly as the number of mutational differences increase. The population genetics of speciation: The evolution of hybrid incompatibilitiesAlso check out this paper Hybrid incompatibility and speiationWe can talk about them later.HBDEdited by herebedragons, : clarification

Don't tell me you accept evolution now... But the problem with creationist's view of "microevolution" is that pretty much everything is microevolution. So even if, for example, the order Diptera diverged from a single fly pair on the ark into the more than 240,000 species we have today, that would be "microevolution." - "they are still flies!" So saying you are talking about "microevolution" simply obfuscates what we are really talking about, and that is the formation of distinct, differentiated populations - ie. speciation. But that doesn't mean that they ARE wrong. That is where you need to provide evidence that they are; what you keep doing is repeating your assertion and accusing me of dishonesty. I try REALLY hard to understand what point you are trying to make and address that - if there is misunderstanding it is an honest mistake. Let me sum up what I see as your explanation as to how new species form: A small subset of the population breaks off and moves into a new geographical location. This sub-set of the population has only a portion of the alleles in the original population and because the population size is so small they undergo a period of inbreeding that brings about new genetic combinations. Over time, this sub-population becomes quite distinctive from the original population to the point that it no longer interbreeds with the original population.Is that about right? Show me the evidence of that. Not just re-stating your hypothesis, but provide some data. You missed an important aspect of outbreeding depression. It is more than just "genetic problems caused by two different genotypes coming together." There is an important aspect of those two different genotypes that you appear to have overlooked. Faith, you always seem to be right even when you are wrong. So what does this sentence actually mean? This would be a great time for data. So there is no way a subspecies can give rise to another subspecies? All dog breeds were derived directly from the wolf? Or do you allow a certain number of branching events? This seems to be an extremely bold claim with even very little logical support. So you claim. But it should be clear that I don't uncritically accept the ToE. Just as one example, do you not see Taq and I arguing over the term "junk DNA?" I have been convinced of the ToE by critically studying it and I still think many aspects are controversial - such as gradualism verses punctuated equilibrium, is sympatric speciation possible in natural populations, is "junk DNA" an appropriate term to use. So the claim that I am "so busy swallowing the ToE whole" is a "big fat lie" to use your own words. The funny thing is that your ideas only work in your head; it was the same situation in the geology threads. If you want to show me how things "really" work... show me some data; show me the research that demonstrates how it really works. And you can't claim that the lack of data is due to the fact that no one is looking for the things you claim because you also claim that we are all working with the same data just interpreting it differently. So find some data that has been interpreted "wrong" and show me how it should be interpreted. Part of the problem is that you don't address so many of the points we bring up. I know you are only one person and so it is understandable that you don't have time to address every argument, but it may be helpful to getting this thread back "on track" to go back and address some of them.Your claim is that population splits cause an irreversible depletion in genetic diversity. However, I think it is still unclear what you mean by "genetic diversity." I would ask that you go back and address my Message 459 and carefully consider the questions about genetic diversity that I posed. And don't just tell me what you "think" give some justification (preferably with examples) for you position. Maybe then I won't have to debate with a phantom Faith.HBD

I think I better start by clarifying a few things. I don't think we are in disagreement on any of this or that you don't understand it, but I may (once again) be approaching this in a highly technical way that may not be readily apparent. Reproductive isolation is any barrier that prevents or reduces gene flow between two populations. Reproductive isolation is the first step in speciation but is not synonymous with it, nor is it synonymous with genetic incompatibility. Speciation cannot be thought of as an event, but as a process. It is essentially impossible or impractical to identify the point or event at which speciation occurred. We can however, identify the key steps in the process and even determine their relative importance and the likely order in which they occurred. Genetic incompatibility is when there is a genetic basis for the inability of the hybrid offspring to contribute to the recombining of the parent populations. Or another way to put it: when there is a genetic basis for the reduction in gene flow (as opposed to a physical barrier). ABE: Genetic incompatibility could be synonymous for postzygotic isolation, although technically "isolation" is what it *does* and "incompatibility" is what it *is* (if that makes sense). /ABENow, I will admit that this is my own definition so I would be happy to debate it but I'll give some justification here for this definition.I think it is pretty obvious that if the hybrids are not viable, (ie. they don't develop into adults) that they will not themselves breed and produce offspring and so cannot contribute to recombining the parent populations. It should also be obvious that if the hybrids spread through the population and completely homogenize it, they are NOT genetically incompatible. However, in-between these two extremes is a whole gradient of possibilities.Here is an example. Let's say we have a population of plants that is adapted to higher elevations and one that is adapted to lower elevations. Where these two population meet there is a hybrid zone. However, the hybrids are not well suited to the higher elevations; but neither are they well suited for the lower elevations. So the hybrids remain confined to this narrow zone and the two populations retain their distinct character, thus gene flow is restricted to a small section of the two populations. This would be considered partial incompatibility, but it would be enough to allow each population to remain differentiated. We could even measure the amount of partial incompatibility by creating a cross and transplanting the hybrids to both habitats and measuring fitness. A reduction of fitness by say 90% in each habitat would indicate a 90% incompatibility. No, I wouldn't combine them. They are two different ways of restricting gene flow. Prezygotic isolation could occur without geological isolation and geological isolation could occur without producing prezygotic barriers. The factors that prevent most of the two northern varieties from interbreeding are probably genetic (although bird songs can be learned from the parents, so if that were the only difference then it may not be genetically based). So perhaps it could be partial genetic incompatibility but unless we know something about the hybrids, more likely simply prezygotic isolation. Yea, of course defining the line between species is always a controversial subject in taxonomy. Perhaps you are a "lumper"? I don't think that the biological species concept requires 100% incompatibility, so perhaps I'm a "splitter?" Without having done much study on this, I would guess those populations are distinct enough to consider separate species even though they hybridize at very low rates. But maybe they could all be considered sub-species of one species, but really, what's the difference? Just a longer name to write Phylloscopus trochiloides subsp. trochiloides . But I do believe that they are now considered to be subspecies by most authorities, not completely sure though. At first I was a bit uncomfortable about requiring the "function" of the gene to change, but the more I think about it, that is pretty accurate. However, it wouldn't necessarily be "protein production." I think I would leave that out and just say the function of the gene would need to be different. It seems a bit difficult to make a generalized statement regarding this. I think we would need to talk about some specific examples. However, I think by definition that if an allele is recessive then its expression is masked by the dominate allele. But could a mutation change a normally recessive allele into a co-dominate allele? Yes, certainly. Two things; First, I was unclear in my statement. I should have said that "in theory one mutation could result in prezygotic isolation." I don't expect postzygotic incompatibility to arise from a single mutation (although I would still say it's not completely inconceivable, but I am just not sure how it would). And as I explained above, I would also consider there to be incompatibilities even if they were not 100%.Second, even if one mutation resulted in pre or post-zygotic isolation, speciation would still be a process. Even in the case of polyploidy, that mutation would need to establish as a breeding population before speciation would be considered to have occurred.Here is a study on Monkeyflower where a substitution at one locus (YUP) caused a dramatic shift in pollinator preference and sympatric populations are now 99% isolated - because of this one mutation resulting in prezygotic isolation. They will hybridize in the lab, which is how they made their NILs (near isogenic lines - which is single gene from one species introgressed into the background of the other). But they are reproductively isolated in nature.Genetic incompatibility can be very simple as in the case of Mimulus

quote:

---

Much evidence has shown that postzygotic reproductive isolation (hybrid inviability or sterility) evolves by
the accumulation of interlocus incompatibilities between diverging populations. Although in theory only a
single pair of incompatible loci is needed to isolate species, empirical work in Drosophila has revealed that
hybrid fertility problems often are highly polygenic and complex. In this article we investigate the genetic
basis of hybrid sterility between two closely related species of monkeyflower, Mimulus guttatus and M. nasutus.
In striking contrast to Drosophila systems, we demonstrate that nearly complete hybrid male sterility in
Mimulus results from a simple genetic incompatibility between a single pair of heterospecific loci. W

---

The following chart is adapted from a lecture on DMI (Dobzhansky—Muller Incompatibilities) in my Evolutionary Biology course.

# subtitutions	# incompatibilities
1	0
2	1
3	2
4	3
5	6
6	9

It shows how one substitution causes no incompatibility but the number quickly rises as the number of substitutions increase.So in the case of this monkey flower (Mimulus) there are substitutions at 2 loci, hms1 and hms2 resulting in 1 incompatibility which causes almost complete male sterility in the hybrids.HBDEdited by herebedragons, : No reason given.

Just a couple quick notes here as I have just a little time right now. Just a point of clarification in case it is not clear to everyone reading this. In this case, (a) and (A) do not refer to dominant and recessive genes but to derived and ancestral character states. So (a) is ancestral and (A) is derived. So, the point is you only get the potential for incompatibility when there are at least two derived character states. Yes, hmmm. I went back and looked at the lecture and my professor drew his "V" diagram a bit different; he only considered derived - derived relationships - he did not include derived - ancestral, but I am not sure why. Maybe it just simplifies the situation since derived - derived incompatibilities are 3 times more likely to occur than derived - ancestral incompatibilities when rates of substitution are equal in each lineage (see bottom of page 1807). So, I think derived - ancestral makes more sense so, not sure why he only included derived - derived. I don't think we disagree, maybe just using different terminology?? I initially list 3 types of isolation:1. geological/ecological (or spacial) isolation
2. prezygotic isolation (mating choice would be a prezygotic isolating mechanism)
3. postzygotic isolation (genetic incompatibility)And yes, isolation is what is critical for further divergence, that is why geological/ecological isolation is the big player in speciation.HBD

Not necessarily. The point is that the founding population is small enough that the original distribution of alleles will be significantly under-represented. It is not really about the number of individuals, but it does usually involve a small number of individuals. For example, would 1,000 individuals be considered a small, founding population? Probably not. But what if the parent population had a billion individuals? In that case 1,000 would be a very small number and would definitely under-represent the allelic distribution of the parent population.It seems to me this is the situation you are describing. The allelic diversity of the parental population is significantly under-represented in the daughter population. That is the whole point of the founder effect, not how many individuals established a new population. Migration has a specific meaning in population genetics - it means gene flow. You are simply describing formation of a daughter population separated from the parent population by a physical/geographical barrier. Allopatric speciation. It is fine to focus on only allopatric speciation, since not only is it the easiest to discuss, but also is the most common situation. In fact, the other modes of speciation are quite controversial, so better to stay away from them for now. I don't deny that. The daughter population would probably have a different allele frequency that the parent population, but it is not inevitable. Yes, a small founding population would significantly change allele frequency; no one denys that. I am not sure what this means, but I think you are saying that you would need to count all alleles at all loci. ??msg 670 This... is the same as "change the genotype frequency"msg 673 Inbreeding means something specific in population genetics, which is what we are discussing. If you are using it to mean a population that breeds with itself, that is just silly. All populations breed with themselves. Inbreeding is a form of non-random mating and it means that individuals are more likely to mate with a close relative than would be expected by chance. Yes, you would expect inbreeding with a small population (again this is part of the consequences of the founder effect). However, once the population reaches a certain size, you would expect random mating to resume and inbreeding to diminish. Well, if that's your position then there is no inbreeding at all. There is just breeding. Ok, fine. I don't want to go back and find where you said what. If you never said it and you agree that inbreeding does not change allele frequency, then there was no point in you arguing about it.msg 674 Where did I say anything about loci being fixed? Loci being homozygous is not the same as being fixed. Homozygous is on an individual level, fixed is on a population level. How you create a recombinant inbreed line (RIL) is to mate two parents that have complimentary character states of a trait of interest. For example, if you are interested in what genes control flower color you might cross a white flowered individual with a pink flowered individual. You would then grow an F₁ progeny and would cross it with itself to produce the F₂ generation. This F₂ generation will have a random distribution of genes from both parental lines. Next you will grow out the F₂ and self-fertilize again. Then you will take one seed from each F₂ plant and grow it out and repeat this cycle for 6 - 10 generations.The size of the RIL population varies, but usually between 100 and 5000 individuals. The point of these lines is to produce as many combinations of homozygous genotypes as possible - each individual will have its own unique combination of alleles. The phenotype of each individual can then be mapped onto genetic markers so that an association can be made such as when a trait is present so is a particular molecular marker. This is known as QTL mappingBut this actually goes against what you are claiming, which is... Why do we not have 500 new species in a RIL? Because new allele combinations are just not enough. In the case of an RIL, you haven't decreased the genetic diversity of the metapopulation, just of individuals. If you allowed the RILs to begin randomly mating again, the population would return to equilibrium genotype frequencies.HBD

I was thinking about this issue this morning and was already going to comment on it. It is unfortunate that science often uses terms that already have common meanings but are used in a specific way - like the word theory. Clarity depends on how the word is used. For example, if I were to say "My theory is..." it would indicate that I am going to speculate about a specific topic and make my best guess about it. However, if I say "Well, that's only a theory..." that instantly gives the impression that I don't understand what the word theory means as used in a scientific context.Migration IS appropriate to use when referring to populations moving from one area to another. But when it is used to refer to an evolutionary process, it refers specifically to gene flow - not movement of individuals. If an individual moves from one population to another but does not breed in the new population, migration has not occurred. Without going back and quoting Faith exactly, I got the impression she was using the word migration in the context of an evolutionary process which is why I asked questions such as "Why would gene flow only occur in one direction?"In the context of population genetics, movement of individuals is sort of irrelevant. I mean, of course individuals need to move in order for genes to move, but we are not really concerned about the individuals per say. Furthermore, if a population can move into an area there is no reason it cannot move back. What would prevent gene flow back and forth is the generation of a barrier - which is really the concern - isolation. And what if there is no movement of populations but a barrier develops that simply separate them - such as a road, that would also produce the same effect without "migration" at all.So my advice is rather than talking about a "subpopulation migrating to a new area" just say "when a daughter population is isolated from the parent population." Yes, no need to invent new terminology.HBD

Is this not a discussion of population genetics? It is not a matter of being nitpicky, it is a matter of understanding. If basic terminology and concepts are not clear, how are we going to be able to look at research on these subjects? For example, when I see you have written something like this: it just gives me this nagging feeling that you really don't understand the concepts you are pontificating about. In this case, inbreeding and gene flow are two separate issues; because a population is not inbreeding does not give me a clue about what kind of gene flow is going on. So I think, "Wait, she is confused about something here" and I try to clear up the confusion. Thank You. And yes, you should have anticipated this. Population genetics is my sub-specialty after all. What you are describing is a daughter population that is breeding in isolation from the parent population. It is that simple. Inbreeding may or may not be a part of the situation, but unless an increase in the proportion of homozygotes is important to your case, maybe there is no need to bring it up. However, inbreeding can affect genetic drift and so alleles can become lost or fixed more readily. Great. I think more discussion about what is meant by "genetic diversity" is seriously needed.I would like to know what you would consider to be a gain in genetic diversity. Would changing the proportion of heterozygotes be changing the genetic diversity?How would you measure genetic diversity? What criteria would you use to compare genetic diversity in two populations?What would you expect to see in a population that has lost genetic diversity? (should be something measurable)HBD

I am a graduate student studying Plant Pathology (plant diseases) and will also have a degree in EEBB (ecology, evolutionary biology and behavior). The emphasis of my research is the diversity of a pathogenic fungi, Rhizoctonia solani but I am not trying to argue from authority. Not at all, I am simply convinced that the things I have studied and learned are true. I would change anything I accept as true in regards to science should it be shown to be false. The math can get rather complicated, but it is not all that necessary to understanding the principles. Most of the basic ideas can be explained with simple mathematical representation. The difficult part comes with the statistics, but you can get by without knowing how to do the statistics as long as you can understand what they mean. Very likely so. However, your reaction is usually defensive and accusatory rather than explanatory. That certainly doesn't help the situation at all. I think there is something about genetic diversity that needs to be clarified. You probably do understand this, but there are two broad ways in which genetic diversity is important. One is diversity within a population and the other is diversity between populations. Diversity within a population provides the raw material for populations to diverge and differentiate and so increase diversity between populations. This is what you seem to be disputing actually happens.But, I think you have worded your argument all backwards. I think your argument should be "Evolution requires an increase in genetic diversity but what happens in natural populations is that diversity is actually reduced."When a subpopulation splits off of the parent population it must then become different from the parent population. The diversity between populations must increase. You can think of diversity as variation or differences so that the question we are asking is how much diversity or how much difference is there between a parent population and a daughter population. This diversity or difference must increase before these two populations are considered differentiated or considered to be different species or subspecies.Take a look at the table below and see what you think. We have already talked about migration and you have already recognized that migration would slow evolution or in other words would decrease differentiation between populations. See if your idea of drift fits with what the table shows.

	Variation within subpopulation	Variation between subpopulations	Affect all loci
Mutation	increase	increase	No
Migration	increase	decrease	Yes
Drift	decrease	increase	Yes
Selection	increase/decrease	increase/decrease	No

(Adapted from: Conner and Hartl (2004). A primer of ecological genetics. Sinauer Associates)If you have any questions or are in disagreement with how these processes affect diversity I think it important to discuss them.Also pay attention to whether these processes affect all loci or not. This will be important to our discussion on how to measure and interpret genetic diversity.HBD

I understand that, but we need to establish what we are talking about when we say genetic diversity. Let's use the terminology "genetic differences" for the time being. Genetic difference, genetic variation and genetic diversity mean essentially the same thing, just have subtle differences in their usage. I think that using genetic differences will much clearer for the purposes of this discussion. Nope. I am referring to genetic diversity (that is genetic differences) throughout. Maybe because the title of the thread is "Evolution Requires Reduction in Genetic Diversity."I would like to focus on some of these basic principles before we discuss your specific scenario. Not only will it ensure that we are talking about the same thing, but it should also help you understand your own argument better so that you can better express it. (although I will compliment you that you have done a much better job of clarifying what you are trying to get across in the last couple messages, at least with me). I also suggest that there is no need to specify "micro" or "macro" evolution. We will be talking about only one issue... how populations diverge and become distinct enough that we recognize them as different species or subspecies. The diagram below should help clarify what the table is referring to At this point, both subpopulation 'A' and subpopulation 'B' are the same species but live in distinct geographical regions. They are both part of the same metapopulation. Now look at the table again. I made a couple changes so that it relates to the diagram above.

	differences within 'A'	differences between 'A' and 'B'	Affect all loci
Mutation	increase	increase	No
Migration	increase	decrease	Yes
Drift	decrease	increase	Yes
Selection	increase/decrease	increase/decrease	No

Now, what we are trying to understand is how subpopulation 'B' becomes different enough from population 'A' so that we can recognize it as a distinct species or subspecies.If your argument is truly that there is more genetic difference between individuals within 'species A' than there is between individuals in 'species A' compared to individuals in 'species B' (assuming that they have somehow become distinct enough that we recognize them as separate species or subspecies); then that not only seems to go against basic logic but against every observation ever made about different species. Perhaps your argument is that the genetic differences that increase between populations are irrelevant and when only the relevant genetic differences are considered, the diversity decreases.But we can address all that later. First let's come to an understanding as to how population geneticists view this issue and then we can talk about what is wrong about that view and how your idea better fits the observations.ABE:Nope. It's about genetic differences. That would be fine. We can talk about each process separately and I can provided experimental observations to explain.
/ABEHBDEdited by herebedragons, : No reason given.

This would be where you would quote reputable sources that define genetic diversity, genetic variation and genetic differences and explain why you think they are too different to use genetic differences at this time for the purposes of this discussion. I could define them for you and explain why we don't need to distinguish between them at this time. But it would be better if you provide your definition. Yes, I would expect the disconnect IS huge, that is why we are having a debate.There are differences between those terms, but for the purpose of this discussion we don't need to be concerned with them at this point. First you tell me what you think the differences are and then I will respond in kind. A Great Debate would be fine with me, it would allow us to have a tighter focus of issues.Part of the issue here is we need to put this in terms that we can actually measure. If we are talking about some idea of genetic diversity that cannot be measured or demonstrated, then we will get no where but speculation. But if we look at the parts we can measure, we can then apply those principles to the bigger issue.HBD

which is:

quote:

---

Genetic diversity refers to the total number of genetic characteristics in the genetic makeup of a species.

---

It would be quite a feat to count the TOTAL number of genetic characteristics in an individual let alone an entire species. I also disagree that genetic diversity should refer to total characteristics of a species, it should be of a population. Not really. Genetic diversity is more of an abstract concept. We don't report that "genetic diversity of population A is "X value." It's not hard to define the concept of genetic diversity, but to actually report the "genetic diversity" of a species is not realistic. Well, you are wrong. Heterozygosity is a measure we use to estimate diversity, to have an idea how much variability there is within a population, but it is not synonymous with genetic diversity. Neither is alleles per locus. Think about it, we could have a population with 2 alleles at a given locus with each allele at a frequency of 0.50 which would mean the heterozygosity is 50%. Then we could have another population with 3 alleles that is highly inbreed where heterozygosity is <10%. Which one has higher genetic diversity? What is the value of genetic diversity (and units) for each population? What numerical value for genetic diversity would be considered low or high? Genetic variability is the tendency of the individuals in a population to vary or to be different from one another. The measure of variability is how much difference there is between individuals within a population. A highly homogeneous population will have little difference between individuals while a highly variable population will have a large number of differences between individuals.Whenever we are making comparisons of populations what we are actually analyzing is the amount of variability. Variability is ubiquitous in natural populations - it is how much variation that we are concerned with, thus the types of questions we usually ask are "How different are these two populations?" (genetically or phenotypically)Genetic diversity depends on genetic variability. Populations with high variability will have high diversity. Populations with low variability will have low diversity. Variability is what we measure (heterozygosity is a measure of variability) to make inferences about diversity; diversity is not measured directly - it is too abstract, too large of an umbrella term. Think about it, if we measure the heterozygosity and the number of alleles for 10 loci of 100 individuals randomly selected from a population of 10,000, what would that tell us about the overall TOTAL diversity in the entire population? What about diversity at the other 19,990 loci? Do we need to include them in the study as well? I only proposed that in order to discuss the chart and the processes involved we use the easier to understand term of genetic variation or genetic differences. It IS relevant because that is what happens, populations in isolation tend to become genetically different from each other. It is these differences in genetic makeup that drives phenotypic divergence. See, I don't want to just say this is wrong. I want to explain WHY it is wrong and WHY it doesn't work and that it simply doesn't match what we observe in nature. But it is not just a simple matter to explain, it takes working through a process. I even proposed that in working through that process you could better understand your own argument and possibly make some adjustments so it might be more convincing. It is doing no good to just keep restating your position. We get it and it's wrong. You keep thinking that if we understood your argument we would see that it is right, but it's not that we don't understand your argument as you keep claiming, it's that your argument doesn't match what we observe in nature. (Or at least that is our counter-argument). I could just point you to some studies that show you what we actually observe and that show your model to be wrong, but... I am only trying to explain what we actually examine and observe in studies on natural populations, which flies in the face of what you are proposing. If you only want to discuss hypothetical experiments and tell us what the hypothetical results should be, then this will never move forward. Unless, of course, you have some real data to support your position. It has not been forthcoming as of yet.HBD

That doesn't match any understanding of genetic drift I have ever heard of. It doesn't even make sense.You mentioned once that you thought of drift as random selection, and that is a better description of drift than the above. Drift removes individuals from a population by random events regardless of the individual's fitness. Drift happens within populations - all populations, not just daughter populations.The parent /daughter split would NOT be considered genetic drift. After separation, the two populations could experience genetic drift at different rates and to different effects, but the split itself is not drift.HBD