Human evolution: Ardipithecus, humans, and chimps

_{Denyse O'Leary
December 21, 2009

Darwinism, Human evolution

1}_{Categories
Darwinism
Human evolution}

Share: Facebook; Twitter/X; LinkedIn; Flipboard; Print; Email

Someone wrote to me recently, asking

Ever since the reporting of Ardi, I expected a commentary of it on your blog but so far I have found none unless I missed it. I’m curious to what you have to say about it since the researchers of Ardi claim chimps may have descended from us. That being the case then evolution’s tree of life would have to be reimagined. Thoughts?

I think Jean Auel did the best job in Clan of the Cave Bear, and she even admitted that she was writing fiction. Which, in my view, puts her way ahead of dozens of profs who can tell me exactly how long-dead people – who never left any writings behind – thought about stuff like religion and family life.

Anyway, I am putting this to the commenters. What do you think about Ardipithecus? An ancestor of us? Of chimps? Both?

By the way, this blog is a volunteer enterprise. Unlike the Darwinists, we are not part of your tax burden. If you feel like contributing financially, don’t let me deter you. We could expand our coverage and services if we had more resources. Otherwise, you see only what a volunteer found time to contribute. That’s hardly a quarter of what we could say.

Comments

Haldane's dilemma - the cost of selection - is not alive and well. This idea is based on the follwoing simplifying premises: 1) Species are optimised for their environments, and therefore any mutation is deleterious until the (biotic or abiotic) environment changes. In this sense, Haldane excludes the possibility of positive selection. Selection is only purifying and the cost of selection is the cost of being out of step with the environment. It is a measure of how much change can occur in an environment before a species becomes extinct. However, as we now understand that environments are dynamic, there is substantial scope for positive mutations in natural populations. Soft selection and hard selection both occur, a point Haldane was unaware of. 2) Species use fixed, delineated resources. However, species can encroach on the resources of other species (i.e. the basis for Van Valen's Red Queen hypothesis). Because of this, differential intrapopulation selection cannot be invoked to set an upper limit on the rate of molecular evolution. 3) Mutations fix individually and independently. This is the most important assumption in my opinion. The main stochastic force in a sufficiently large population is genetic draft (hitchhiking). This is the primary assumption which discounts the generality of the neutral theory. An interesting discussion of the genome-wide reach of this can be found here (Hahn, 2008). Even though a large proportion of mutations being fixed in populations are neutral as Kimura correctly predicted, they are not being fixed by drift. Further a series of positive mutations at linked loci are able to fixed concurrently, provided meoitic recombination breaks the linkage. This is the most important point because it is the same mistake that PaV @ 28 makes when stating:
it takes 14,000 years for mutation A to arise, and then it takes mutation A 4 x 10^5 years to become fixed. Total number of years is 414,000 years for the first beneficial mutation to occur and become fixed. Now the genome must search for mutation B. Well, of course, this will take another 414,000 years.
A small amount of knowledge is a dangerous thing.paulmc_{December 23, 2009
December
12
Dec
23
23
2009
01:49 PM
1
01
49
PM
PDT}

Joseph:
That is false- the dilemma stands firm.
No, it doesn't.
Also there isn’t any evidence that demonstrates any amount of mutational accumulation can account for the transformations required. IOW you don’t have anything beyond imagination. Is that how your “science” operates- via imagination?
I don't like your tone very much, Joseph. Unless you make an actual argument, I will refrain from responding to your insulting remarks.IrynaB_{December 23, 2009
December
12
Dec
23
23
2009
11:40 AM
11
11
40
AM
PDT}

PaV: The mutations were speaking about have specifically to do with the differences between chimps and humans.
As long as you don't try to argue that an ordinary hand of bridge is so unlikely as to be implausible (1 in 600 billion). Any single genome is incredibly unlikely. The number of mutations are more than sufficient to account for the changes. Even in a moderate size population, every mutation is tried every few thousand years. 10^9 genome / (10 mutations per year * 10^5 population)
PaV: Zachriel, your understanding of neutral theory is wrong.
Meanwhile, neutral mutations fix at approximately the rate they occur.Zachriel_{December 23, 2009
December
12
Dec
23
23
2009
10:00 AM
10
10
00
AM
PDT}

Collin and Zachriel: The mutations were speaking about have specifically to do with the differences between chimps and humans. The genetic difference between humans and other species becomes far greater. Zachriel, your understanding of neutral theory is wrong. I'm going to get Kimura's "The Neutral Theory of Molecular Evolution" and quote some of his exact numbers. I'm sure you're not going to trust my opinion; so I'll give you the opinion of the grandaddy of neutral theory.PaV_{December 23, 2009
December
12
Dec
23
23
2009
09:30 AM
9
09
30
AM
PDT}

Collin: I generally agree with you, but isn’t their argument that if the mutations that separate us from apes were not there that there would be different mutations that separate us from… whatever other species would exist had evolution worked differently. In other words, it is highly unlikely that a certain mutation would occur, but it is not quite as unlikely that any beneficial or neutral mutation would occur, be whatever mutation it might.
Quite so. What we expect is a radiating pattern and diversity. A population may diverge to become humans or chimps. A population can be highly diverse. Indeed, nearly every human being has a unique genome. Leaving aside selection, if we divide a population, drift alone will assure divergence.
Joseph: Being “fixed” means that everyone in that population has it.
That's correct. And the rate of fixation for neutral mutations is approximately equal to the neutral mutation rate, regardless of population size. (2Nm) · (1/2N) = m. Thus far we have only considered point mutations, but there are many other sources of genetic novelty.
Joseph: And in a sexually reproducing population even the most beneficial mutation has a better chance of getting lost than it does at becoming fixed.
That depends on the selection coefficient. The probability for a specific mutation being fixed is about 2s / ( 1 - e^( -4 Ns) ). For small s and large N, that approximates to 2s, and is therefore much more likely to fix than a neutral mutation.Zachriel_{December 23, 2009
December
12
Dec
23
23
2009
04:40 AM
4
04
40
AM
PDT}

IrynaB:
Haldane’s dilemma has been refuted a long time ago, but that’s for another post.
That is false- the dilemma stands firm. Also there isn't any evidence that demonstrates any amount of mutational accumulation can account for the transformations required. IOW you don't have anything beyond imagination. Is that how your "science" operates- via imagination? Strange...Joseph_{December 23, 2009
December
12
Dec
23
23
2009
04:12 AM
4
04
12
AM
PDT}

Zachriel:
The rate is 175 mutations per genome per generation or roughly 10 mutations fixed per year.
Being "fixed" means that everyone in that population has it. And in a sexually reproducing population even the most beneficial mutation has a better chance of getting lost than it does at becoming fixed.Joseph_{December 23, 2009
December
12
Dec
23
23
2009
04:10 AM
4
04
10
AM
PDT}

PaV, I generally agree with you, but isn't their argument that if the mutations that separate us from apes were not there that there would be different mutations that separate us from... whatever other species would exist had evolution worked differently. In other words, it is highly unlikely that a certain mutation would occur, but it is not quite as unlikely that any beneficial or neutral mutation would occur, be whatever mutation it might. I think that ID's best arguments concerning DNA include those that show that mutations are almost always deleterious and are inadequate in explaining complex structures, whether they be human or ape.Collin_{December 22, 2009
December
12
Dec
22
22
2009
05:35 PM
5
05
35
PM
PDT}

IrynaB [30], If I compare species A to species B, then I notice that specific mutations at specific locations along the genome are part of the causes bringing about the difference between species A and species B. So to talk about it not mattering what kind of mutation we’re considering, or where a mutation occurs, is to ignore the actual fact of genomic differences. You are thinking as though any mutation, anywhere along the length of the genome, can be tolerated. Well, we know that just one mutation in the wrong spot can cause the death of an individual organism. So it is wrong to think so many mutations can occur without any damage to the organism. When we invoke the neutral theory, we’re simply assuming randomness in how the mutations come about; we’re not assuming mutation is itself neutral, though a great number of them are. Getting back to the problem at hand, again, specific nucleotide substitutions at specific locations is what separate species. What is needed, therefore, is a sequence of specific mutations. If you take species A and species B, and gene A1 differs from gene B1 at three specific locations, then you need three specific mutations. There are 70,000 genes in humans, and my calculation shows that only three of these 70,000 genes can be helped along by accepted Darwinian mechanisms. Obviously, this isn’t sufficient. Zachriel [31], You’re making the same mistake that IrynaB is making. If gene A is 1500 bp long, and three nucleotide changes are needed, then this 1500 bp section, surrounded as it is by 3.5 x 10^9 bp, has the probability of all three mutations occurring of (1500/3.5 x 10^9)^3, which is a trillionth of a trillionth of a chance of happening. That’s where fixation helps you. But if a mutation that has no effect on a protein fixes (a neutral mutation), then this is no help whatsoever. Only when one of the three critically needed nucleotide mutations occurs and fixes is any progress made. But the probability of this---not just any old mutation---is quite low as the above calculation demonstrates. When you invoke positive selection, you are simply invoking the mechanism that involves the very paradox that Nachman and Crowell present: positive selection is the flip side of genetic load---and they see a great big problem there. As should you.PaV_{December 22, 2009
December
12
Dec
22
22
2009
05:12 PM
5
05
12
PM
PDT}

PaV: The point that you’re missing is that not just any old mutation will do. You need a specific mutation at a specific location along the length of the genome.
That's a profound misunderstanding of evolution. There isn't a specific adaptation that evolution has "in mind." Rather, it's an opportunistic process. If a particular mutation is beneficial, it will tend to propagate through the population.
PaV: Total number of years is 414,000 years for the first beneficial mutation to occur and become fixed.
Your calculation of fixation is incorrect. The presumption is that critical mutations are beneficial which fixes faster than neutral mutations. If the population has structure (subdivisions, as are commonly found in natural populations), then it can fix even faster.
PaV: Your calculation of 10 million mutations becoming fixed only points out the true function of natural selection:
You keep conflating neutral and non-neutral mutations. Natural selection doesn't influence neutral mutations. If a mutation is beneficial, it will be more likely to fix. If it is detrimental, it will be more likely to go extinct. Under selection, it's P ? 2s / ( 1? e^(?4 Ns) ). For small s and large N, that approximates to 2s. P, probability N, effective population s, selection coeffecient If there are large numbers of neutral mutations per individual in each generation, then we expect large numbers to accumulate. A mutation will be effectively neutral if s is much less than 1/2N.Zachriel_{December 22, 2009
December
12
Dec
22
22
2009
03:58 PM
3
03
58
PM
PDT}

Pav:
The point that you’re missing is that not just any old mutation will do. You need a specific mutation at a specific location along the length of the genome.
No, we're not talking about specific mutations -- we're talking about the total number of mutations that have accumulated in the genome for the last few million years. Focusing on a specific mutation is like asking what the odds are that a specific person wins the lottery, while here we are talking about the odds that someone -- anyone -- will win the lottery. The latter odds are of course much higher. It just so happens that my estimate coincides nicely with independently estimated divergence of humans and chimps, whereas your estimate is way off.
Your figure of 10 million mutations fixed per million years amounts then to a heterozyosity of around 50 mutations per 1500 basepairs, which is roughly 3%. This is perfectly in line with heterozygosities for mammalian species.
I agree. This supports my calculation.
But we’ve calculated the heterozygosity per one million years. The highest heterozygosities occur in insect species. For mammals H is in the range of 6 to 13%. So, any species that diverged more than 8 million years ago (I’m doubling here because of the split lineages) couldn’t tolerate this amount of heterozygosity, and that would mean that natural selection would have to kill off members of the species as a way of eliminating the accumulated number of mutations.
No, because our calculations so far assume neutral mutations.
And, again, Nachman and Crowell end their paper with the problem of genetic load. Translated, this means that to keep harmful mutations away, assuming Darwinian mechanisms, “intolerable” amounts of offspring would have to die. So, what does this say about Darwinian mechanisms? First we had Haldane’s Dilemna—which basically led Kimura to his Neutral Theory, and, now, we have Nachman and Crowell’s Paradox. You see, RM + NS just doesn’t add up. So, what’s going on? Or, better yet, what went on?
It seems to add up just fine if you do the calculations correctly. Haldane's dilemma has been refuted a long time ago, but that's for another post.IrynaB_{December 22, 2009
December
12
Dec
22
22
2009
03:49 PM
3
03
49
PM
PDT}

IrynaB: If 1/Ne is the probability of fixation (which assumes neutrality), then out of 10^6 mutations per year, 10 will become fixed.
Which is consistent with an even simpler calculation. Assuming neutrality, the rate of fixation is approximately equal to the rate of mutation, independent of population size. The rate is 175 mutations per genome per generation or roughly 10 mutations fixed per year.
PaV: Typical rates, as I stated in the post, is 1 x 10^9 mutations/generation= 1 x 10^9 mutations/year, which equals 20 x 10^9/20 years = 2.0 x 10^8, which is the exact figure that Nachman and Crowell give.
Are those supposed to be negative exponents? Nachman's estimated rate of mutations per nucleotide is 2.5*10^-8 per generation.
PaV: This means, then, that for an effective population size of 10,000 to 100, 000, mutations will occur, on a per year basis, at the rate of 10^9/10^5,
It would be the yearly rate of mutations per diploid genome (10) * population (10^5).
PaV: This means, then, that for an effective population size of 10,000 to 100, 000, mutations will occur, on a per year basis, at the rate of 10^9/10^5, and will become fixed at a rate of 1/4Ne, or 1 fixation for every 400,000 years.
Doesn't matter because new mutations are constantly being added and constantly being fixed. In any case, beneficial mutations fix much more rapidly.Zachriel_{December 22, 2009
December
12
Dec
22
22
2009
03:25 PM
3
03
25
PM
PDT}

IrynaB [26]:
Here’s my calculation: 175 new mutations per individual, for a population size of Ne=10^5, means 2×10^7 new mutations per generation of 20 years, equals 10^6 new mutations per year in the population. If 1/Ne is the probability of fixation (which assumes neutrality), then out of 10^6 mutations per year, 10 will become fixed. That means 10 million mutations become fixed every million of years. A factor 10^7 higher than your estimate of 2.5. That’s better than your previous error of 10^11. Four more posts, and we might agree!
The point that you’re missing is that not just any old mutation will do. You need a specific mutation at a specific location along the length of the genome. So, at some specific spot, lineage A diverges from lineage B. Any mutation somewhere else won’t do because the likelihood of the mutation being deleterious is much greater than it’s being beneficial. So, the question then becomes this: to get a particular mutation at a particular point (thus rendering it ‘beneficial’), how many mutations do we need? Well, the probability of getting a particular mutation at a particular site along the length of the genome is 1 divided by the total number of nucleotides. Now the 175 mutations is for a ‘diploid’ genome. So the odds of a particular mutation at a particular site is 1/7 x 10^9 = 1.4 x 10^10. Thus, 1.4 x 10^10 mutations have to occur before we can be sure that a particular mutation---let’s call it mutation A---occurs. Now, if 10^6 mutations are produced each year by the population, then 1.4 x 10^4 years are needed for mutation A to occur. Now simply occurring is not sufficient; it has to become fixed. If we need mutation B elsewhere on the genome if some new species is to arise, then mutation B has show up on a genome that already has mutation A on it. This means that, roughly, mutations A,B,C,D ….. and so forth, need to be added sequentially (note that I said roughly, for on average this is true, but not strictly true). The fixation time for a particular mutation is 1/4Ne, with Ne considered by Nachman and Crowell to be 10^4 or 10^5. You’ve used 10^5. So, it takes 14,000 years for mutation A to arise, and then it takes mutation A 4 x 10^5 years to become fixed. Total number of years is 414,000 years for the first beneficial mutation to occur and become fixed. Now the genome must search for mutation B. Well, of course, this will take another 414,000 years. So, take the 2.75 million years since lineage A and lineage B diverge, and this means, on average, we would expect 2.75 x 10^6 years/one fixed mutation/4.1 x 10^5 years = 6.6 fixed, beneficial mutations. I guess my 8 fixed mutations figure was overstating it! Your calculation of 10 million mutations becoming fixed only points out the true function of natural selection: the elimination of deleterious mutations. If, indeed, all 10 million mutations are truly “neutral”, then let’s look at another number Nachman and Crowell give us: 70,000 genes. Now, on average, these 37.5 million mutations, all fixed per your calculation, would amount to 37.5 x 10^6 mutations/7 x 10^4 genes, equals, roughly, 5 x 10^2 mutations per gene. N&C use 1,500 basepairs as the average size of a gene, and they calculate that the total number of coding basepairs is about 10% of the genome. Your figure of 10 million mutations fixed per million years amounts then to a heterozyosity of around 50 mutations per 1500 basepairs, which is roughly 3%. This is perfectly in line with heterozygosities for mammalian species. But we’ve calculated the heterozygosity per one million years. The highest heterozygosities occur in insect species. For mammals H is in the range of 6 to 13%. So, any species that diverged more than 8 million years ago (I’m doubling here because of the split lineages) couldn’t tolerate this amount of heterozygosity, and that would mean that natural selection would have to kill off members of the species as a way of eliminating the accumulated number of mutations. And, again, Nachman and Crowell end their paper with the problem of genetic load. Translated, this means that to keep harmful mutations away, assuming Darwinian mechanisms, “intolerable” amounts of offspring would have to die. So, what does this say about Darwinian mechanisms? First we had Haldane’s Dilemna---which basically led Kimura to his Neutral Theory, and, now, we have Nachman and Crowell’s Paradox. You see, RM + NS just doesn’t add up. So, what’s going on? Or, better yet, what went on?PaV_{December 22, 2009
December
12
Dec
22
22
2009
02:55 PM
2
02
55
PM
PDT}

PaV, really good stuff. Haldane's Dilemma is alive and well, and evolutionists continue to pretend it doesn't exist.Mung_{December 22, 2009
December
12
Dec
22
22
2009
12:29 PM
12
12
29
PM
PDT}

PaV, Thank you very much for your detailed reply. It's good to see a pro ID commenter here take the time to seriously work the numbers. But (you knew there was a but coming), your calculations are still way off because you calculate on a per nucleotide basis -- not a per genome basis.
Typical rates, as I stated in the post, is 1 x 10^9 mutations/generation= 1 x 10^9 mutations/year, which equals 20 x 10^9/20 years = 2.0 x 10^8, which is the exact figure that Nachman and Crowell give. This means, then, that for an effective population size of 10,000 to 100, 000, mutations will occur, on a per year basis, at the rate of 10^9/10^5, and will become fixed at a rate of 1/4Ne, or 1 fixation for every 400,000 years. This means that it will take, on average, 10,000 years to get a particular needed mutation, and 400,000 years to fix it. IOW, we can expect roughly 2.5 mutations to become fixed every million years of evolution.
Here's my calculation: 175 new mutations per individual, for a population size of Ne=10^5, means 2x10^7 new mutations per generation of 20 years, equals 10^6 new mutations per year in the population. If 1/Ne is the probability of fixation (which assumes neutrality), then out of 10^6 mutations per year, 10 will become fixed. That means 10 million mutations become fixed every million of years. A factor 10^7 higher than your estimate of 2.5. That's better than your previous error of 10^11. Four more posts, and we might agree! I will address the rest of your post later.IrynaB_{December 22, 2009
December
12
Dec
22
22
2009
12:22 PM
12
12
22
PM
PDT}

The Riddled Chain Pure. Dumb. Luck.Mung_{December 22, 2009
December
12
Dec
22
22
2009
12:19 PM
12
12
19
PM
PDT}

StateMachine [4]:
Worth noting, I believe, is that the mutation rate quoted deals with that per *nucleotide*, not per genome. With a genome of several million nucleotides, a mutation will most definitely appear in a shorter amount of time. If you are referring to a mutation fixating, that is a different story.
Actually, the mutation rate, per year, is rather constant, and is 1 x 10^9. In my response below I’ll spell that out. As to genomes of “several million nucleotides”, this does not apply to human evolution, which has a genome of about 3 x 10^9 nucleotides. And, since we’re dealing with small effective population sizes, the time needed to arrive at the needed mutation is the bottleneck, not the fixation time. IrynaB [5]:
Actually, the number of new mutations per individual offspring is estimated to be about 150 in humans, off the top of my head. You’re off by 11 orders of magnitude. Also off the top of my head, humans produce about 400 novel gene products compared to our nearest relatives. Factor in the fact that regulatory genes can have large downstream effects, do you still feel that the “huge anatomical differences” (which I don’t find so huge at all) cannot be accounted for?
I’ve just finished reading the article. The actual number is 175 per “diploid” genome. [I suppose what is implicit in the “per diploid genome” is our understanding that “anit-sense” transcription takes place, which means we should divide this number by two. But let’s just leave the number alone for right now.] This number is calculated using a generation time of 20 years, and rather low effective population sizes. The actual equation is: mu=k/(2t + 4Net)=mutation rate. You can see that by increasing the effective population size, Ne, the result is a lower rate of mutation. But, we’ll leave that to the one side as well. Now, typically, mutation rates are given as mutations/generation while assuming one year as the average generation time. Typical rates, as I stated in the post, is 1 x 10^9 mutations/generation= 1 x 10^9 mutations/year, which equals 20 x 10^9/20 years = 2.0 x 10^8, which is the exact figure that Nachman and Crowell give. This means, then, that for an effective population size of 10,000 to 100, 000, mutations will occur, on a per year basis, at the rate of 10^9/10^5, and will become fixed at a rate of 1/4Ne, or 1 fixation for every 400,000 years. This means that it will take, on average, 10,000 years to get a particular needed mutation, and 400,000 years to fix it. IOW, we can expect roughly 2.5 mutations to become fixed every million years of evolution. Nachman and Crowell use a figure of 2.75 million years based on the last common ancestor having been 5.5 mya (we divide the time by two since both lines are mutating [N.B. N&C were comparing humans to chimps] simultaneously, thus giving twice the mutational difference). So, how many mutations of all sorts can we expect? It’s 2.5 x 2.75 = 7.925. Let’s round that off to 8 mutations that are fixed. Now there have been a huge number of mutations that will have occurred, but only 8 will have found their way through the entire population. One would think this would have the effect of humbling Darwinists, but we know that it doesn’t. There is the dogma to preserve, you know! So, my basic argument still stands. The whole ‘excitement’ about Ardi is because it pushed the 5.5 mya back a bit. Yet, again, we’re dealing with orders of magnitude of difference between what is needed to be explained and what Darwinism, specifically neo-Darwinian population genetics, can explain. Let me add these two further points: (1) I believe it is due to the small effective population size and the very long generation time used for humans that gives these larger than average numbers for mutation rates. With this in mind, I’ll make the second point. (2) As I’ve just demonstrated, correcting for different generation times makes the “rate” of human evolution the same as that of other organisms. Yet we hear that human evolution has “speeded up”. I believe this is entirely due to using numbers without making needed corrections. My second point, then, is that, quite typically, we who challenge the Darwinian orthodoxy are forced to wade through the addled thinking of Darwinists. This muddled thought happens because words are simply invented by the Darwinists to force-fit the data into Darwinian thought. We, then, are forced to unravel their muddled thinking so as to get to the bottom of things. paulmc [8]
Nachman and Crowell (2000) produced this well-known estimate. It means that across the global human metapopulation there is enough mutation to replace every single nucleotide in a period of about 20 years (i.e. each generation). So a lack of mutation to generate variation is a poor argument indeed.
Well, first of all, your claim depends on what number you use for the effective population size of humans. Your claim, that in one generation every nucleotide can be replaced means that 7.0 x 10^9 nucleotides (diploid genome) must be replaced in one generation with a mutation rate of 175 mutations per diploid genome. Let’s assume 200 mutations per diploid genome just to make the calculation straightforward. This means the effective population size of the ancestral originator has to be 3.5 x 10^7, that is, 35 million strong. Given that Nachman and Crowell use population sizes of 10^ and 10^5, your statement is off by at least a factor of 350. Second, as I’ve pointed out, it’s not just a matter of getting the mutation, but of it fixing. With a population size of 35 million, it would take 140 million years for a single mutation to become fixed. Isn’t it obvious that the numbers just don’t add up? paulmc [8]
A series of terrible analogies about whisky and garage sales from PaV does not change this basic fact.
Do my analogies seem so terrible now? Having answered the challenges made, let me note how NACHMAN AND CROWELL’S PAPER ends:
The high deleterious mutation rate in humans presents a paradox. If mutations interact multiplicatively, the genetic load associated with such a high U would be intolerable in species with a low rate of reproduction (Muller 1950; Wallace 1981; Crow 1993; Kondrashov 1995; Eyre-Walker and Keightley 1999). The reduction in fitness (i.e., the genetic load) due to deleterious mutations with multiplicative effects is given by 1-e^2U (Kimura and Moruyama 1966). For U = 3, the average fitness is reduced to 0.05, or put differently, each female would need to produce 40 offspring to survive and maintain the population at constant size. This assumes that all mortality is due to selection and so the actual number of offspring required to maintain a constant population size is probably higher. . . . The problem can be mitigated somewhat by soft selection (Wallace 1991) or by selection early in development (e.g., in utero). However, many mutations are unconditionally deleterious and it is improbable that the reproductive potential on average for human females can approach 40 zygotes.
So, there you have it. A paper that ends up presenting Darwinism with a “paradox”---which is no more than Haldane’s Dilemna revisited---is quoted as a source telling us just how many mutations are possible. So a paper that should have Darwinists shaking their heads is used as a debating device in their favor, and, it appears, because they are misinterpreting it under the guise of "sped up" human evolution. Finally, let me note that Kimura, in his book The Neutral Theory of Molecular Evolution says that when he first proposed the “neutral theory,” he did so based on this very problem: that is, the problem of genetic load. But he said there was another reason why he was proposing it which he left unstated. It was the problem of the huge number of beneficial mutations that are needed to account for the polymorphism found in human DNA. Because most mutations are deleterious, a huge number of mutations is needed so as to winnow out the few good mutations that go into any kind of progressive evolution of the genome.PaV_{December 22, 2009
December
12
Dec
22
22
2009
11:23 AM
11
11
23
AM
PDT}

Why do some people want to be related to chimps?Joseph_{December 22, 2009
December
12
Dec
22
22
2009
04:52 AM
4
04
52
AM
PDT}

IrynaB, What is the scientific data that demonstrates the transformations required are even possible? Also it is most likely that even the most beneficial mutation will get lost, as opposed to becoming fixed. BTW "beneficial" is relative- what is beneficial for one generation in one environment may not be so in the next generation or in another environment. Also there isn't just one beneficial mutation- meaning there could very well end up being competing beneficial mutations.Joseph_{December 22, 2009
December
12
Dec
22
22
2009
04:52 AM
4
04
52
AM
PDT}

Mung: And yet every single nucleotide is not replaced every 20 years, or every generation, so apparently there are in actuality not enough mutations across the human metapopulation to replace every single nucleotide in a period of about 20 years (i.e. each generation). Quite the conundrum.
No conundrum. The mechanism is called selection (Darwin, Wallace 1858).
Mung: Can’t have too much mutational noise or the signal would get wiped out.
Quite so, but not an issue. No single individual has every mutation, rather every single-point mutation is represented in the population over time. With a population of a few million, every mutation would be tried every few thousand years. There is sufficient variation to explain extensive changes in populations, which was the question raised. - Plasma storms are apparently still causing a delay in Zachriel's comments appearing on this blog. Our teams are working to resolve the problem.Zachriel_{December 22, 2009
December
12
Dec
22
22
2009
04:23 AM
4
04
23
AM
PDT}

vjtorley: Here are some other articles by John Hawks that may be germane to the debate on whether Homo could have evolved gradually: John Hawks, Keith Hunley, Sang-Hee Lee and Milford Wolpoff. Population Bottlenecks and Pleistocene Human Evolution at http://mbe.oxfordjournals.org/.....ull/17/1/2 . In Molecular Biology and Evolution 17:2-22 (2000).
Everything in the paper you cite is based in and lends support to the Theory of Evolution, that is, evolution by natural variation and selection. From the abstract:
Although significant population size fluctuations and contractions occurred, none has left a singular mark on our genetic heritage. Instead, while isolation by distance across the network of population interactions allowed differences to persist, and with selection, local adaptations were able to develop, evolution through selection, along with gene flow, has promoted the spread of morphological and behavioral changes across the human range. It is this pattern of shared ancestry that has left its signature in the variation that we observe today.
The study indicates a bottleneck two million years ago, no subsequent bottleneck, and insufficient resolution in genetic data to provide a more detailed look at fluctuations in population.Zachriel_{December 22, 2009
December
12
Dec
22
22
2009
04:12 AM
4
04
12
AM
PDT}

vjtorley: Here’s an illuminating paper on Ardpithecus from a paleoanthropologist named John Hawks.
Nice blog. But what is your interest in the difficulty of reconstructing a relatively minor detail of common descent? Hawks certainly agrees that the overall pattern of evidence strongly supports Common Descent, the question being where to fit this particular organism. His request for better access to the original evidence is reasonable. Regardless, it doesn't call into question evolutionary theory, just some details of a lineage that you seem particularly interested in (perhaps because you share a possible family relationship). The paleoanthropologist named John Hawks also thinks ID goes well beyond fantasy into the realm of delusion.Zachriel_{December 22, 2009
December
12
Dec
22
22
2009
03:56 AM
3
03
56
AM
PDT}

Mung:
And yet every single nucleotide is not replaced every 20 years, or every generation, so apparently there are in actuality not enough mutations across the human metapopulation to replace every single nucleotide in a period of about 20 years (i.e. each generation). Quite the conundrum.
The word replace was a bit unfortunate. Most conceivable single nucleotide mutations occur within a relatively short time. Since the probability of fixation of a beneficial mutation is roughly 2s I think, where s is the selective advantage, only a small fraction of beneficial mutations will become fixed. For neutral mutations that's 1/N and for deleterious mutations even smaller. Assuming a distribution of fitness effects of mutations (based on estimates, if available), isn't it possible to work out how many beneficial mutations have accumulated since the split-off some 6 or whatever millions of years ago? Doesn't sound like too much work, so I guess someone already did that.IrynaB_{December 21, 2009
December
12
Dec
21
21
2009
10:37 PM
10
10
37
PM
PDT}

paulmc:
It means that across the global human metapopulation there is enough mutation to replace every single nucleotide in a period of about 20 years (i.e. each generation).
And yet every single nucleotide is not replaced every 20 years, or every generation, so apparently there are in actuality not enough mutations across the human metapopulation to replace every single nucleotide in a period of about 20 years (i.e. each generation). Quite the conundrum. And if there were, it would play heck with trying to reconstruct phylogenetic trees, wouldn't it? Can't have too much mutational noise or the signal would get wiped out.
So a lack of mutation to generate variation is a poor argument indeed.
He was assuming a much smaller population size. Of course, I think the trade off there is with a smaller population size a mutation can be fixed much more quickly.Mung_{December 21, 2009
December
12
Dec
21
21
2009
09:59 PM
9
09
59
PM
PDT}

More About ArdiDavem_{December 21, 2009
December
12
Dec
21
21
2009
09:57 PM
9
09
57
PM
PDT}

If we descended from apes, why did we lose our hair? What advantage does hairlessness convey? Is it because we discovered fire and how to make clothing to keep warm? Did fur have a tendency to catch fire so our ancestors with less hair survived more than the ones who had fur? Why did we retain eyebrows and hair on our heads? Didn't we know how to make hats? Eyebrows can be helpful for keeping debris out of our eyes, so did our ancestors who lost their eyebrows get killed more by predators and other enemies because they got stuff in their eyes at crucial moments, therefore losing eyebrows was an evolutionary disadvantage?Davem_{December 21, 2009
December
12
Dec
21
21
2009
09:47 PM
9
09
47
PM
PDT}

Paul, when you say, “Skepticism is one thing, stating “That’s because the oldest ancestor of man was a man” is rather another. That is a statement that requires some evidence” you are right.
Well it's nice to know we agree on that much! Who knows about the rest...paulmc_{December 21, 2009
December
12
Dec
21
21
2009
09:32 PM
9
09
32
PM
PDT}

Oops, that last post had the most incredible argument for intelligent design ever conceived, but I accidentally deleted it and now I forgot it. Oh well. :) Paul, when you say, "Skepticism is one thing, stating “That’s because the oldest ancestor of man was a man” is rather another. That is a statement that requires some evidence" you are right.Collin_{December 21, 2009
December
12
Dec
21
21
2009
08:33 PM
8
08
33
PM
PDT}

Collin_{December 21, 2009
December
12
Dec
21
21
2009
08:31 PM
8
08
31
PM
PDT}

Hi everyone, Here's an illuminating paper on Ardpithecus from a paleoanthropologist named John Hawks. It makes for VERY interesting reading. The trouble about Kenyanthropus and Ardi .
There are three skulls from putative "hominins" that date to 3.5 million years or earlier. Every one of these skulls is known now from extensive reconstruction or correction for distortion in the original. By itself, the extensive reconstruction might not be a problem. But as Tim White has repeatedly shown, the specialists on these crania actively and vociferously disagree about the basic anatomy due to problems reconstructing them. (Emphases mine - VJT.)
It gets better:
We can't see the scans, no independent reconstructions are possible, and the people who can see the scans refuse to present comparisons of these three skulls that together represent the supposed origin of the hominin lineage.
By the way, the emphasis and italics in the above quote are John Hawks', not mine. Watch out for the AAAARRRRGGHHHH! near the end of the post. All of us have fresh memories of the HarryReadMe file in the Climategate scandal. One lesson we all learned from that is that when a scientist says Aaarrgghh!! , you know there's a major problem. Here are some other articles by John Hawks that may be germane to the debate on whether Homo could have evolved gradually: John Hawks, Keith Hunley, Sang-Hee Lee and Milford Wolpoff. Population Bottlenecks and Pleistocene Human Evolution at http://mbe.oxfordjournals.org/cgi/content/full/17/1/2 . In Molecular Biology and Evolution 17:2-22 (2000). A revised chronology for early Homo . Is a lack of fossils the problem with early Homo? News flash: Dmanisi hominids were not short . Enjoy!vjtorley_{December 21, 2009
December
12
Dec
21
21
2009
07:14 PM
7
07
14
PM
PDT}

Prev 1 2 3 Next

You must be logged in to post a comment.

Leave a Reply