In a pickle about Adam and Eve

_{vjtorley
March 5, 2014

Intelligent Design

4}_{Categories
Intelligent Design}

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

Professor Jerry Coyne can’t seem to leave the Adam and Eve question alone. In a recent post, Professor Coyne criticizes Bryan College in Dayton, Tennessee, for requiring its teaching professors to sign an updated “statement of belief” which, for the first time, explicitly affirms the existence of an historical Adam and Eve. Since Bryan College describes itself as “a nondenominational evangelical Christian college named after William Jennings Bryan: statesman, orator, and renowned prosecuting attorney in the famous Scopes Evolution Trial,” this requirement should hardly occasion surprise. What would be surprising is if the college didn’t require its professors to believe in a literal Adam and Eve.

In a related post published late last year, Coyne explains in detail why he is convinced that science has ruled out the existence of Adam and Eve:

The facts first. Sheehan et al., building on an earlier paper by Li and Durbin (references below), calculated that the minimum population size associated with the worldwide expansion of humans out of Africa about 60,000 years ago was 2,250 individuals, while the population that remained in Africa was no smaller than about 10,000 individuals. For population geneticists, this is the “effective population size,” invariably smaller than the census size, so these are minimum estimates, and ones derived from conservative assumptions. The population sizes are estimated by back-calculating (based on reasonable estimates of mutation rates and other parameters) how small an ancestral population could be and still give rise to the observed high level of genetic variation in our species.

Note: 2,500 is larger than two.

This means, of course, that Adam and Eve couldn’t have been the literal ancestors of all humanity.

Evidently math is not Professor Coyne’s forte.

Note: 2,500 isn’t the same as 2,250.

Note: 2,250 + 10,000 = 12,250.

The math lesson is over.

Coyne goes on to say that even these figures are under-estimates: they represent “the ‘effective population size,’ invariably smaller than the census size.”

I invite readers to have a look at the following article by Luke J. Harmon and Stanton Braude, of Princeton University:

Conservation of Small Populations: Effective Population Sizes, Inbreeding,
and the 50/500 Rule

I shall quote a brief extract:

There is no such thing as “the effective size” of a population. Different effective population sizes help us to estimate the impact of different forces. The effective size you estimate will depend on the scientific question you are trying to address (Box 12.1). Estimating the appropriate effective population size is crucial in biology; in most (but not all) cases, effective population size will be smaller than the actual number of organisms in the population. Think for a moment about why
this is so. A conservative rule of thumb used by some biologists is
that N_e [the effective population size – VJT] is usually about one-fifth of the total population size (Mace and Lande, 1991). Using such a rough estimate is risky because N_e can be larger than the census size of the population, depending on the history of the population and the particular N_e under consideration.

It’s rather embarrassing when a biology professor makes mistakes in his own field, isn’t it?

UPDATE: A final suggestion for Professor Coyne. Coyne claims that the effective population sizes he cites are “based on reasonable estimates of mutation rates.” Coyne is assuming here that the mutations are natural and undirected. If Coyne wants to refute the Adam and Eve hypothesis as entertained by believers in intelligently guided evolution, then the question he really should be asking himself is: what would the effective population size need to be, if the mutations that gave rise to the human line were artificial and directed?

Comments

I wouldn't listen to Joe - he's an evolutionary biologist and and the son of communists! You know who those people re-write history :)wd400_{March 15, 2014
March
03
Mar
15
15
2014
03:08 PM
3
03
08
PM
PDT}

Would you like to rephrase your remarks concerning this, wd400? No. It's not possible for there to be more than one mtEve and multi-regionalists did not expect there to be more than one (rather, they thought mtEve would be much older than the OOA school did) You comments to goodusername suggest a lack of understanding about who mtEve is. The title can obviously change hands, and you don't wars or famine or floods for other lineages to die out. In fact, there must have been an molder tEve for the population in which "our" mtEve lived!wd400_{March 15, 2014
March
03
Mar
15
15
2014
03:00 PM
3
03
00
PM
PDT}

Hi PaV and wd400, Professor Felsenstein was kind enough to respond to a query of mine regarding Adam and Eve, over at http://theskepticalzone.com/wp/?p=4200 . See especially http://theskepticalzone.com/wp/?p=4200&cpage=1#comment-43066 , http://theskepticalzone.com/wp/?p=4200&cpage=1#comment-43193 and http://theskepticalzone.com/wp/?p=4200&cpage=2#comment-43375 . You might find his remarks helpful. Cheers.vjtorley_{March 15, 2014
March
03
Mar
15
15
2014
02:59 PM
2
02
59
PM
PDT}

PaV,
Would you like to rephrase your remarks concerning this, wd400? You see, Communism prevails. And goodusername, what I was stating was right there in the article. So I won’t have to point anything out to you.
It appears that WD400 was right in #89, this is confusion regarding the multi regional theory vs the out-of-africa theory. No one - including the proponents of the multiregional theory - were saying that there wouldn’t be a single mito Eve at some point going back. As explained, there HAS to be one due to the facts of how mtdna is passed on and that we’re a sexually reproducing species! Many of the proponents of the out-of-Africa theory did make much of the results of the study. Not because there IS a mito Eve (everyone knew that there would be one), it’s that she lived perhaps more recently than some predicted, and came from the very region where the out-of-Africa proponents were suggesting that modern humans originated from. It’s perhaps a point in favor of the out-of-Africa theory, but I’d say a small one, at best. Had the study said that mito Eve from from eastern Asia 400k years ago, it would perhaps be a point in favor of the multiregional theory, but again, a small one. It’s pretty random who mito Eve is going to be, and as explained, it’s a moving target. A thousand years from now, mito Eve may indeed be someone from Asia. Even as a proponent of the out-of-africa theory myself, I think much too much was made of this study. This is why I said it’s irrelevant to the question of human origins. As explained, someone today will someday be a “mito eve”, but it doesn’t mean that this is the start of a new species or anything.
Sorry to ‘rain on your parade,’ but will it turn out that this eventual “one woman” came out of Africa? You know what I mean.
It can be any woman today that has, or will have, daughters. (If you meant something else by the question I’m afraid I missed it.)goodusername_{March 15, 2014
March
03
Mar
15
15
2014
02:57 PM
2
02
57
PM
PDT}

goodusername: Thanks for the link. Here is what is in the third column of the 1987 Science paper:
Because the origins of modern humans was thought to have occurred at some time in the past half million years, mtDNA seemed to offer a genetic route to answering a question that had eluded anthropologists for decades. The question was not just when the first modern humans evolved, but also how. Did they evolve simutaneously throughout the Old World, deriving from populations of archaic sapiens already established there from Homo erectus forerunners? Or did they arise in one location and then migrate throughout the rest of the world, replacing poulations of archaic sapens as they went? ……….. One this was clear,” says Brown (a student of Wilson, who led the investigation--pav), “the degree of variation between individuals was much less than might have been expected, given the variation known fro the great apes for instance.” This low level of mtDNA variation was a surprise, not least because it appeared to preclude deep genetic roots for Homo sapiens.
Would you like to rephrase your remarks concerning this, wd400? You see, Communism prevails. And goodusername, what I was stating was right there in the article. So I won't have to point anything out to you.
(Note that the answer to who’s Mito Eve is irrelevant to the question of human origins. It’s also not a fixed individual. The mito eve of humans a hundred years from now might be someone different.) Expecting a war are we, goodusername? Does this mean you believe in the flood? Same phenomena, isn't it?
So, you were wrong again.
No, you were wrong again. Wilson had his own ideas, as the article points out; but there was plenty of uncertainty in the methods and other competing points of view at the time. And those views were being bandied around in the media all the time. So, start listening to your elders, please.
(Oh, and it hardly matters, but the best evidence still points to most of the genome being junk, and this is indeed evidence for that fact out genomeis the product of evolution.)
You will one day have to 'eat' those words, I'm afraid.
A woman alive today will be a future mitochondrial eve. Every woman alive today is a mitochondrial lineage. Each generation from now, there will be women who either don’t have children, or have sons but no daughters. Thus, each generation from now, there will be fewer and fewer women of today that can claim any mtdna descendants. Eventually it will be just one woman.
Sorry to 'rain on your parade,' but will it turn out that this eventual "one woman" came out of Africa? You know what I mean.
PaV_{March 15, 2014
March
03
Mar
15
15
2014
02:15 PM
2
02
15
PM
PDT}

And I'm not going to. My own experience is irrelevant to the fact that speaking of mulitiple mitchondrial eves makes not sense. To suggest (without evidence, I might add) that genetcists expected to find multiple eves is very strange indeed.wd400_{March 15, 2014
March
03
Mar
15
15
2014
01:38 PM
1
01
38
PM
PDT}

wd400: You still have not answered how old you are.PaV_{March 15, 2014
March
03
Mar
15
15
2014
01:26 PM
1
01
26
PM
PDT}

PaV: A woman alive today will be a future mitochondrial eve. Every woman alive today is a mitochondrial lineage. Each generation from now, there will be women who either don’t have children, or have sons but no daughters. Thus, each generation from now, there will be fewer and fewer women of today that can claim any mtdna descendants. Eventually it will be just one woman. So the question simply was, how far back do we have to go before there is a single woman who is the mdna ancestor of all women today? (Note that the answer to who’s Mito Eve is irrelevant to the question of human origins. It’s also not a fixed individual. The mito eve of humans a hundred years from now might be someone different.) If you go back to a point where there are still several mito lineages from which mdna today originates, than, by definition, you haven’t gone back far enough. So not only is it not true that several mito eves were expected, it doesn’t even make any sense. It’s nonsensical. What would it even mean? Here’s an article from Science mag 1987: “The Unmasking of Mitochondrial Eve”: “Because mitochondria pass from generation to generation only through the female line… the phylogenies inferred from mtDNA data essentially trace maternal inheritance: ultimately, a single female is reached at the root of the tree, hence the reference to Eve.” http://www.sciencemag.org/content/238/4823/24.extract It’s the same for Y-Adam. At some point in the past there is a man who’s the Y-chromosome ancestor of all men today. It’s not much of a prediction that such a man exists, it’s what HAS to happen, for the same reason a mito Eve must exist. In Dawkins’ “River Out of Even” (1995) he predicts that the Y-Adam will be a man who probably lived long after mito Eve. This was (IIRC) about 5 years before the results of the first Y-Adam study were announced. http://news.sciencemag.org/2000/11/eden-chronicles-tracking-y-chromosome-adam If you can find any articles that predicted otherwise, I’d really love to see them - it would be fascinating to see what on earth they were talking about.goodusername_{March 15, 2014
March
03
Mar
15
15
2014
12:04 PM
12
12
04
PM
PDT}

It's not possible for their to be more than one mitochondrial eve. How could that possibly work? Since Cann and Wilson and other geneticists in the 1980s were not idiots I think we can safely rule out the claim that they were expected to find multiple eves, and put all this "Darwinists are like communists,re-writing history" stuff aside. I guess you are confused by the difference btween multi-regionalism and Out of Africa. But neither of those schools thought there would be mulitple eves (and indeed, even in out of africa there are regionally unique mitochondrial lineages) So, you were wrong again. I won't hold my breath waiting for an apology about all this childishness comparing evolutionary biology to communnism and Nixon... (Oh, and it hardly matters, but the best evidence still points to most of the genome being junk, and this is indeed evidence for that fact out genomeis the product of evolution.)wd400_{March 15, 2014
March
03
Mar
15
15
2014
11:25 AM
11
11
25
AM
PDT}

wd400: We've lived through Darwinists pounding away at IDists, night and day, for years and years, that the presence of so much "junk-DNA" is proof that Darwinism, and not ID, was correct. Then, when they could do this no longer in the face of so much conflicting evidence, they now, quite conveniently, just say: "Oh, when did we ever say that?" This sounds very much like your: "Oh, they never expected there to be just ONE "mitochondrial Eve." When they asked Nixon how history would view him, he replied: "It depends on who the historians are."PaV_{March 15, 2014
March
03
Mar
15
15
2014
08:42 AM
8
08
42
AM
PDT}

Thanks, Querius, for your comments. I remember all of this happening in the early 90's, but it could have easily been the late 80's too. Of course, then it was off to "Y-Adam." And another surprise.PaV_{March 15, 2014
March
03
Mar
15
15
2014
08:38 AM
8
08
38
AM
PDT}

wd400: From what I know, aren't you about 30? That means you were what, ten years old when all this was going on, while I was in my forties. Do you want me to deny reality? Is that what you're asking me? It was in the papers. It was a topic of discussion. I remember it vividly. I waited expectantly for their results exactly because if there were multiple origins of "Eve" this would prove troubling. And then they were surprised when it turned out that there was only 'one' Eve. I remember all of this very well. We're not in a communist state, yet; so I'm in no way going to deny reality. Unless you're older than forty, I'm not going to accept a word you say. Darwinists, like Communists, rewrite history to please themselves. Don't drink the Kool-Aid, wd400. Now, shifting back to our discussion: I hammered away at making the point that science cannot see clearly enough beyond these "bottlenecks." However, there's another way of looking at all of this. When we look at Genesis, Cain, after killing Abel, is sent off. He's afraid that he will be killed; so he is given a special 'mark.' It sure sounds like we're dealing here with other human-like primates. Let's remember that what gives us freedom, and the use of reason, is our consciousness. Without this, we are mere animals. Think of someone who sleepwalks: they move around, talk with people, and do other things; but they're not at all self-aware of what they are doing. It is consciousness that elevates us above all the other animals. It is entirely possible that the account of man's creation in the second chapter of Genesis has to do with God infusing consciousness into human-like primates. If this is true, then genetically, these "persons"---human-like primates now having "consciousness" (likely changed in some ways)---would be linked to the genetic history of the group from which they were formed. The Bible, in my estimation as a Catholic, is not a history book at all times (it is quite a good history book at other times), and this applies to the Genesis accounts of Creation. It says there that God formed man out of the clay. But, of course, in Chapter 1 it just finished saying that God made men and women and told them to fruitful and multiply. So there is really no clear and unequivocal understanding here, and so caution as to how we understand all of this is in order. But, I invite your comments about the genetic history perspective I've presented. BTW, this is why I said earlier that I don't see linkages of humankind (as we now know it) to chimps and such as a problem. Now, do I consider this to be "common descent"? I don't think so. I think it is very likely that God intervened in some manner. For example, in order to accomodate "consciousness" was a larger cerebral cortex needed? Etc. I don't know that we will ever have a full story of this. As I see it, neither religion nor science, neither Darwinism nor Intelligent Design, will ever know enough as to make a definitive conclusion about all of this. We're all called to understand this as best we can, and in the best way possible. Anyway, your comments.PaV_{March 15, 2014
March
03
Mar
15
15
2014
08:35 AM
8
08
35
AM
PDT}

PaV might not be wrong at all---but you might be, wd400. From what I've read, there seems to have been a reaction by paleoanthropologists in the 1990s against the 1987 Mitochondrial Eve theory. While Darwinists claimed that ME was represented by many women, Paleoanthropologists observed genetic traits in Homo erectus fossils simultaneously in various parts of the world---arguably MEs in Asia, Africa, and Europe (Neanderthals). Interestingly, there actually *three* primary mitochondrial lineages that are observed in the world today (M, N, R) that could nicely correspond with the three fertile mothers on Noah's arc, and also explain the genetic bottleneck as due to a global flood. For anyone's interest, Dr. Robert Carter speaks on mitochondrial DNA and other genetic topics starting at 30:30) at http://www.youtube.com/watch?v=CDuLEVu1C4A&feature=youtu.be&t=9m40s. -QQuerius_{March 15, 2014
March
03
Mar
15
15
2014
01:26 AM
1
01
26
AM
PDT}

You said this.
And, now, let’s remember a little bit of history: in the 90?s, they went looking for “mitochondrial Eve”, and were certain that they would find more than ONE such “Eve.”
Can you admit you got this wrong?wd400_{March 14, 2014
March
03
Mar
14
14
2014
07:46 PM
7
07
46
PM
PDT}

wd400: This is from Caballero's 1994 article: The simplest possible conditions under which the dispersive process can be studied are met in the Wright—Fisher idealized population (Fisher, 1930; Wright, 1931). This consists of an infinite, randomly mated base population subdivided into infinitely many subpopulations, each with a constant number, N, of breeding individuals per generation Yes, the "subpopulation" is not infinite (I'm sure the infinite population is being used to justify some kind of mathematical trick). And, yes, it is the Wright-Fisher model. I stated that somewhere along the line. And, yes, that's not the coalescent theory, but coalescent theory is built upon the foundation of the neutral theory. And, yes, as I stated twice already, each generation produces an infinite number of alleles. The thing I think you haven’t quite grasped is the difference between an instantaneous value for Ne (which could be arrived at from many histories) and the histroical reconstruction of Ne over many time periods that these coalescent methods produce. That’s why these graphs extend out past the OOA bottlenecks in non-African populations. It seems to me that "coalescence" takes place when you run out of 'heterozyogisity.' You end up with one allele, or one form, etc. That's where you're going to determine where the MRCA is. This number obviously depends on the methodology employed, and it is not always going to be the same estimated time. And, of course, these methods suffer--- as all methods involving molecular biology do, like molecular clocks and phylogenies---with inconsistencies. Nevertheless, some helpful information is gained, and it is certainly gained in as scientific a manner as is available. Again, I just notice the limitations of such methods, and we're wrong to not notice them. (do you at least admit no one though there would be multiple mtEves?) Can you rephrase this? There may be a word missing. As to the discussion, I would agree that it has been more than thoroughly hashed out.PaV_{March 14, 2014
March
03
Mar
14
14
2014
06:23 PM
6
06
23
PM
PDT}

The infinite allele model does not include an infiniate population size. In fact, the most famout eqn in the paper is F = 1 /[ 4Ne.u + 1] I think(?) you are talking about the idealised Wright-Fisher model. The infinite thing in an infinite sites model is the number of possible alleles (i.e. each mutation is making a new allele). You yourself described the derrivation of why both populatins have genetic reduced diversity, so I don't why you keep on going on about the semantics of 'bottle necks'. The thing I think you haven't quite grasped is the difference between an instantaneous value for Ne (which could be arrived at from many histories) and the histroical reconstruction of Ne over many time periods that these coalescent methods produce. That's why these graphs extend out past the OOA bottlenecks in non-African populations. Anyway. It's become increasingly clear you are so wedded to your ideas that you want give them up (do you at least admit no one though there would be multiple mtEves?). So I think I'm going to find something better to do than this.wd400_{March 14, 2014
March
03
Mar
14
14
2014
04:54 PM
4
04
54
PM
PDT}

wd400:
They aren’t caveats, they are consequences of the fact the harmonic mean gives us the value of Ne under these circumstances. Ad you can’t ‘choose to focus’ or one thing or the other, they’re both true. With the harmonic mean long-term stable populations have higher Ne, and populations recovering from a period of very low populatin retain low Ne for a long time.
This is just "preachy". It's like you're in a classroom teaching some young 19-year old. I'm sure they'll accept your answers. I don't. You'll have to do better than you've done so far.
Because there is much less diversity in such a population than one that has been at 2,250 for ever. As the text you quoted says, it takes a long time for the population to get back towards the equilibrium
I'm sure you really think this. And I'm sure you think this because it's the prevailing dogma. But I'm looking for answers, not just dogma. Let me illustrate what I mean. You have made no attempt whatsoever to try and understand the objections I have made in terms of the model being used. You're content to give me bromides. The model that is in play is the "infinite allele" model, is it not? This model says that you begin with an "infinite population" with EACH generation itself producing an INFINITE number of "alleles." The next generation is randomly selected from this host of "infinite alleles." Now, where on earth will you find an "infinite sized population" that each generation produces an "infinite" number of alleles? Nowhere. (And, yes, I know about permutations of a genome that is billions of nucleotides long is quasi "infinite", etc. I understand all that. So no lectures, please) So, when dealing with actual populations, but, in particular, populations that fluctuate over time, an "effective population" calculation is used to correct for the idealized assumptions. However, as far as I can see, you fail to engage the implications of the model. That is, the ENTIRE CONCERN is that when the population size is reduced, the notion of an infinite number of individuals producing an infinite number of alleles is thrown way off. Another way of looking at it is to simply say that when there is a "bottleneck", a lot of diversity is lost because the diversity 'exists' over the entire population. Once it is lost, it will take a long time to recover it. In fact, depending on the lineage, and how far back in time it goes, you may never fully recover it. Now that is all you've been saying. But, that's not the end of the story. For you, however, that is the end of the story. And when I get you to think it through more carefully, I get one lecture after another. Yes, indeed, the Ivory Tower is quite high. I will now point out how you are completely unresponsive to what I'm trying to point out. Answer this simple question: which of the two populations below has suffered from a "bottleneck"? (1) Population A: here is part of its history (population sizes of each generation separated by slashes): 100/200/600/2,000/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250 and so on for another 5,000 generations. (2) Population B: here is part of its history (slashes again separate generation size): 9,450/9,200/9,760/8,760/3,700/2,700/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250/2,250 and so on for 5,000 generations. I won't wait for your answer, because I don't think you'll want to answer. So I'll just avoid all the gamesmanship. According to you, both of these should be considered "bottlenecks." Right? And what does this mean? It means what I've been saying now for over a week. It's saying that if there is a "bottleneck" then you have no idea whether the population size of the population before the "bottleneck" was ten times greater or ten times smaller than the calculated N_e. If you a before and after picture, and you can't tell whether something gets ten times bigger or ten times smaller, then I suggest you must be 'blind'. Are you willing to admit that my interpretation is the correct one?
PaV: So, you use one population, the African one, as the ancestral population, and then you compare the European populations with that of the African one. When you compare Europeans, you go back, what, was it 67,000 years. That’s where the “bottleneck” supposedly occurred, based on heterozygosity and assuming NGD; and, meantime you’ve backed into a N_e. Isn’t that how the computer simulation is done? This is extradondairy. No, this is not how the study was done. Inf fact, the Li and Durban paper uses a single genome to infer past population sizes. These methods use the distrbutions of times-to-coalescnce of blocks of chromosomes to to inder past populatin sizes. When effective population size is small you get many such coalesences quickyl, when it’s large that are more spaced out.
I don't see what you think is so extraordinary. Let me tell you what I see that is extraordinary: I am not talking about Li and Durban. I'm talking about Sheehan's paper and they use more than one genome. But this is not big thing, since both are basically using a newest variation of the "coalescent" method. And what are they doing with the method? They're going back in time using some rule for RGD, instead of going forward in time. And what is it they're looking at which is going to "coalesce"? They're simply taking some consensus genome for the European population and another consensus one from the African and using some form of simulated Markov method until the present day chromosome length "coalesces" to the "original" allele. Here's what they say in their Discussion section: Applying our method to a 2 Mb intergenic region of chromosome 1 from ?ve Europeans and ?ve Africans, sequenced as part of the 1000 Genomes Project, and using a per-generation mutation rate of µ = 1.25×10?8 per site, we have inferred a severe (out-of-Africa) bottleneck in Europeans that began around 117 kya, with a drop in the e?ective population size by a factor of 12. In contrast, we have observed a much more mild population size decrease in the African population. We remark that our estimate of the timing of the bottleneck may not be very accurate, since we used only 16 discretization intervals and 7 free population size parameters. Furthermore, all of our inferred times and population sizes would be smaller by a factor of two if we had used µ = 2.5 × 10?8. What is the difference between "inferring" these numbers and "backing into them"? So, it's a bit extraordinary that you get so worked up by what I stated. Did you notice it wasn't "one" genome, but five; and it wasn't just one population, but both African and Central European. And the factor of 12 they're talking about is a drop from their calculated N_e for Africans at the time of split of 28,000, and the 2,250 they calculated as the N_e for the Europeans. Isn't it quite obvious that based on Sheehan's paper, from which Coyne (remember him? That's how this all got started) then said that 2,250 is greater than 2, implying that Europeans are directly descended from Adam and Eve, rather than what is quite obvious from Sheehan's paper, that the Europeans are descended from the African lineage. So, we go back to Africa, and then we go back in time, and then we estimate an effective population size of 10,000, and then we say, "Obviously we're not descended from Adam and Eve, but some chimp lineage that numbered around 10,000." This isn't science. This is assuming evolutionary theory to be true, and then interpreting everything accordingly, and then saying that this is all demonstrable. The fact of the matter is we don't know enough, and probably will never know enough. But when scientists want to trumpet "effective population sizes" as evidence that there was never an Adam and Eve, well, this sounds to me much like what it must sound like to you when a true Creationist says that the world was created in six days. It has been against this improper extension of science that I've been railing against.PaV_{March 14, 2014
March
03
Mar
14
14
2014
04:05 PM
4
04
05
PM
PDT}

Believe it or not, I can actually read. You’ll notice I included the quote you’ve posted. But I didn’t underline it. There are two caveats: one about ‘bottlenecks’ and one about ‘constant population size.’ You choose to focus on the ‘bottleneck’ and I choose to focus on the “constant population size.’ They aren't caveats, they are consequences of the fact the harmonic mean gives us the value of Ne under these circumstances. Ad you can't 'choose to focus' or one thing or the other, they're both true. With the harmonic mean long-term stable populations have higher Ne, and populations recovering from a period of very low populatin retain low Ne for a long time.
Please explain to me how going from 2 to 50 to 500 to 2,000 census population, and then remaining at N = 2,250 for the next hundred generations, represents a ‘bottleneck’. In what way?
Because there is much less diversity in such a population than one that has been at 2,250 for ever. As the text you quoted says, it takes a long time for the population to get back towards the equilibrium
So, you use one population, the African one, as the ancestral population, and then you compare the European populations with that of the African one. When you compare Europeans, you go back, what, was it 67,000 years. That’s where the “bottleneck” supposedly occurred, based on heterozygosity and assuming NGD; and, meantime you’ve backed into a N_e. Isn’t that how the computer simulation is done?
This is extradondairy. No, this is not how the study was done. Inf fact, the Li and Durban paper uses a single genome to infer past population sizes. These methods use the distrbutions of times-to-coalescnce of blocks of chromosomes to to inder past populatin sizes. When effective population size is small you get many such coalesences quickyl, when it's large that are more spaced out. It should be obvious from this that bottlenecks are not 'blindspots' or barriers through which such methods can't see. For wthat it's worth an Ne of 10,000 is most commonly used for humanity as it is calculated from observed genetic diversity in modern populations.
I don’t see any other alternative understanding.
And that, ultimately, is the problem. You've laboured in this entire thread, attacking your own misunderstanding of how these method work,without, it seems, considering the idea that specialists in population genetics know what they are talking about and don't extend their estimates back beyond 'blindspots'.wd400_{March 14, 2014
March
03
Mar
14
14
2014
09:53 AM
9
09
53
AM
PDT}

wd400:
(2) Nope. Your own source, which you pasted here, says “Secondly, N_e is most strongly affected by periods of reduced population size. In other words, if a bottleneck occurs, causing an increase in inbreeding, this is not restored by a later expansion of the population size.” (3) Nope. See above.
Believe it or not, I can actually read. You'll notice I included the quote you've posted. But I didn't underline it. There are two caveats: one about 'bottlenecks' and one about 'constant population size.' You choose to focus on the 'bottleneck' and I choose to focus on the "constant population size.' Please explain to me how going from 2 to 50 to 500 to 2,000 census population, and then remaining at N = 2,250 for the next hundred generations, represents a 'bottleneck'. In what way?
(4) Sure. Why is this important?
One reason is that Jerry Coyne said that 2,250 is greater than 2. Does this mean he thinks that effective population sizes of 2 can actually exist? Second reason: it is silly to think that one can look at what came before a "bottleneck". I've already shown this a number of times. You seem not to want to accept this reality.
You might be able to create some post hoc reason that saves a literal Adam and Eve, just like it’s possible to say the world was invented yesterday complete with memories, but it’s clear genetic evidence doesn't support such a couple.
Clear genetic evidence? Really? What's so 'clear' about it? As I've stated over, and over, and over again, you can't see what's 'in front' of a "bottleneck". The only reason that they talk about a population size of 10,000 going back in time, is because of the Out of Africa theory. So, you use one population, the African one, as the ancestral population, and then you compare the European populations with that of the African one. When you compare Europeans, you go back, what, was it 67,000 years. That's where the "bottleneck" supposedly occurred, based on heterozygosity and assuming NGD; and, meantime you've backed into a N_e. Isn't that how the computer simulation is done? But the point is is that this doesn't allow you to look back past the 67,000 year mark using the European population data alone. If, in the original paper, they said something like the effective population of the African population is 10,000 going all the way back in time, why, then wasn't it 9,350? Probably because that is their best guess for N_e. Then you use that to figure out some time to a "bottleneck" based on the heterozygosity of the African populations. And what comes before the 10,000 at it's earliest? I'm sure they have no idea. It's guesswork. And, so, the same is true of the European data set. Bottlenecks end in blindspots, not actual population sizes. Unless, of course, you've actually done some kind of field study. Using an infinite allele model--as does NGD---I don't see any other alternative understanding.PaV_{March 13, 2014
March
03
Mar
13
13
2014
09:51 PM
9
09
51
PM
PDT}

wd400:
Progress, you’ve admited to getting something wrong :)
The ONLY mistake I made was in saying "a particular place along the genome" instead of "a specified place along the genome." For some reason you EBs like "pre-specified." I think it only confuses things. Why not just use "specified"?
Let’s look at it per gene, as the numbers are easier to follow, and you have the exonic portion of DNA wrong in your ealier comments. The average polypeptide is 400 amino acids long = 1200 nt.
Yes, the "exonic portion" is probably about 3%; but, as usual, I give the EBs a break and used a higher percentage. And for this, you tell me I got it wrong.
Using back-of-an-envelope numbers, the per nucleotide mutation rate is ~2.2e-8 and ~25% of mutations are synonymous. Over 6 million years, each lineage would have 240 000 25yr generations so we’d expect 1200 nt * mu = 2.2e-8 * 240000 gen * 2 lineages ~ 13 substitutions per protein. 13 per protein rather than 13 per genome.
My calculations used 4,500 generations. Sticking with only one lineage, since the only reason to include 2 lineages is if, as you stay further on, if you want to study when they separated, then we get: 1200 nt * 2.2 * 10^-8 *4,500 = .122 which, though different, means that during those 4,500 generations it is likely that no mutation occurred in any of the protein coding area. Using your same numbers, but looking at just ONE lineage, there are about 6 mutations per exon. That's 6 nucleotides within an exon of 1200 exons. Now what percentage of the average exon is conserved? We're talking here about NGD, so are we not talking about "neutral" substitutions? But you go on to note that non-synonymous mutations "fix" at 1/5 the rate of synonymous mutations. So that means, on average, 6/5, or, ONE nucleotide per exon; or, at most, ONE a.a. difference per exon. Now this is with purifying selection turned on. I guess this means that if you compare two exons, one from a chimp and a corresponding one from a human, out of the 1400 nucleotides only ONE nucleotide is different. And this is what converts a chimp into a human. Is this what you want me to believe?
and 13/5 is 2.6 which is about the average number of amino acid differnce between human and chimp proteins)
But, just a little bit more. You've made your calculation using 2 lineages. The reason for that, I presume, is that you're considering the NGD that takes place within the chimp population. So this means that the chimp lineage changed from what it was like 6 million years prior by about ONE a.a., while the human lineage changed from what the chimp lineage was like 6 million years prior. So if you want to say that NGD is responsible for changing chimps into humans, then, in the same amount of time (you've assumed this) the chimps, having undergone an equal amount of change, nevertheless, still remain chimps! Why haven't they changed into something different? If the numbers impress Darwinists/EBs to have confidence in macro-evolution, wherein a chimp, little bit by little bit, changes into a human, then why didn't chimps change, little bit by little bit, into something else more similar to themselves, not as profoundly different as mankind? IOW, is there a chimp species that shows itself diverging from another chimp species less than 6 million years ago? I don't know. I'm asking the question. But if there isn't, then this absence appears to be rather peculiar. Another curiousity. Is the effect of purifying selection that of reducing the mutation rate of non-synonomous mutations to one-fifth that of synonomous mutations determined mathematically, stemming from some kind of first principles; or do you just simply back yourself into it by noticing the a.a. differences between humans and chimps and then calculating the numbers? Again, what mechanism, exactly, are you proposing for macro-evolution to take place. And, again, if you leave Darwin behind, saying that the EBs have moved on from his original theory, then surely science requires you to propose some new mechanism. And, of course, science would further require you to propose a sensible, logical mechanism. Is there one you would now propose?PaV_{March 13, 2014
March
03
Mar
13
13
2014
09:05 PM
9
09
05
PM
PDT}

(2) Nope. Your own source, which you pasted here, says "Secondly, N_e is most strongly affected by periods of reduced population size. In other words, if a bottleneck occurs, causing an increase in inbreeding, this is not restored by a later expansion of the population size." (3) Nope. See above. (4) Sure. Why is this important? Actually, can you succinty point out what you are trying to do with any of this? I'll remind you that orignal paper finds now evidence of a severly small population in more than a millon years, the newer one extend back less far but still no sign of a severely small population. We know that Ne is "strongly affected by periods of reduced population size", so if there was a bottle beck prior to these evens it would still show up these studies. It hasn't. You might be able to create some post hoc reason that saves a literal Adam and Eve, just like it's possible to say the world was invented yesterday complete with memories, but it's clear genetic evidence doesn't support such a couple.wd400_{March 13, 2014
March
03
Mar
13
13
2014
07:56 PM
7
07
56
PM
PDT}

wd400: I've asked you why the 'harmonic mean' was used in calculating Ne. You told me that I should look at how it is derived. I've done that. Now I'm reporting back. First of all, we're dealing with the theory of random genetic drift that is attributed to Sewall Wright and R.A. Fisher. It's been around since the 30's and 40's. It is based on the "infinite allele" model. This model assumes an infinite population that produces an infinite number of alleles, that is then subdivided into N sub-populations, each of which itself produces an 'infinite' number of alleles. So this is quite an 'idealized' model. Therefore, when implementing it, one has to move from the 'ideal' population model and actual populations. Corrections, therefore, must be made in applying the model; one of these has to do with population size, and Ne, effective population arises within the context of making this correction. Any quotes I include in the following come from a Nature's Review article by Armando Caballero, published in Heredity in 1994. "Under the simple conditions of the idealized population, sampling of gametes is binomial and the variance of the change in gene frequency is: ?²(&#916 q) =q(1-q)/2N, where q is the allele frequency of a gene in the infinite base population. The coefficient of inbreeding at generation t, the probability that two gametes which unite to produce a zygote in generation t carry identical by descent copies of a gene (Wright, 1922; Malecot, 1948), is: F_t = 1/2N + (1-1/2N)F_t-1, where the first term denotes identity by descent from copies of a gene of an individual in generation t-1 and the second, that from copies of a gene of an individual in previous genrations. The rateof incresae in inbreeding per generation is thus: &#916 F = 1/2N where &#916 F =F_t-F_t-1/1-F_t-1. The observable consequence of this increase in inbreeding is a reduction in the expected heterozygosity (H) each generation, &#955=H_t/H_t-1 = 1-&#916F ......... (3), or, relativ to that in the base population, H_t/H_t-1 = 1 - F_t = (1-&#916F)^t. ..............(4) [(N.B. the 1-F_t is from a straight substitution of the equation above (3) directly into (3). And, since, this equality is true for EACH generation, gen 0 to gen t, then the power of "t" is arrived at) PaV] . . ." Caballero gives a few more equations that won't be interesting us, and then says: "It is obvious that real populations are very unlikely to meet the conditions of the idealized population defined above and, therefore, the number of breeding individuals does not describe appropriately the effects of inbreeding and gene frequency drift in most practical situations. The concept of effective population size (Ne) was introduced by Sewall Wright (1931, 1938, 1939) to overcome this problem and has been developed subsequently by others, mainly James F. Crow and coworkers (Crow & Kimura, 1970, pp. 345-364; Crow & Denniston, 1988)." How do we get effective population size, Ne? "If the variance of change in gene frequency or the rate of increase in inbreeding are known, because the genotypes can be distinguished and hence the genotypic fequencies estimated or because pedigrees are available, the effective population size can be estimate or computed directly from the expressions above. For example, if we can trace an observable quantity such as the heterozygosity so that we know its rate of decay (H_t/H_t-1), we can use eqns (2) and (3) to estimate the asymptotic N_e. . . . When information on genotypic frequencies or pedigrees is not available, effective size can still be predicted under certain circumstances (in which one or more assumptions of the idealized population are removed) when demographic data such as census numbers and variances and covariances of the number of progeny per parent are available. Effective size can be derived following a sampling drift approach or an inbreeding approach when the matter of interest is in the gene frequency drift or the increase in homozygosity, respectively (Crow 1954)." So, just as I suspected, and as I stated above, one either had the numbers and information, or one "backs into" N_e. When you have to estimate it without available information, then you have to remove "one of more assumptions of the idealized population). This is exactly where the HARMONIC MEAN comes in. We move down in the paper: "4. Variable population size over generations In the idealized population the number of breeding individuals (N) is constant over generations. Let us consider a situation where the population size varies over generations, with size N_i in generation i. From eqns (3) and (4), the expected heterozygosity in generations t relative to that in the base population is H_t/H_t-1=&#928i=1-t (1-1/2N_i). This can be equated to the relative heterozygosity in the idealized population (eqns (2) and (4), replacing N by N_e, i.e., (1-1/2N_e)^t. When population sizes are large and t much smaller than any of these, this latter equality can be approximated by 1-&#931_i (1/2N_i)&#8776 1-t/2N_e, from where: 1/N_e&#8776 1/t &#931_i [1/N_i] (PaV: summations are from i=1 to i= t) (Wright, 1938; Crow & Kimura, 1970pp. 109-110), i.e., using the harmonic rather than the geometric mean." So, there you have it: the derivation and the reason for using the harmonic mean. But, we're not through yet. I've argued here, over and over again, that it is plain silly to say that a growing population represents a "bottleneck." Well, here's what Caballero has to say to us: "Because N_e is a harmonic mean, two important points similar to those explained for eqn (7) appear. [N.B., eqn (7) deals with a different assumption of the idealized population, and its only relevance here is that as in the case of using the harmonic mean, caution must be exercised when applying this deviation from the idealized model] Firstly, the maximal N_e, given a total &#931_i N_i, is achieved with constant population size over generations. Secondly, N_e is most strongly affected by periods of reduced population size. In other words, if a bottleneck occurs, causing an increase in inbreeding, this is not restored by a later expansion of the population size. . . . As was mentioned before, genetic drift begins one generation (if selfing is allowed) or two (if not) earlier than inbreeding. If population size varies over time, for a given generation, genetic drift depends on the number of individuals in that generation whereas inbreeding depends on the number of their parents (if selfing is allowed) or their grandparents (if not)." So, when I describe a population that grows from 2 individuals to 28,000 individuals over a 7 or 8 generation time frame, this means that only (depending on the mutation rate) 1,400 to 1,600 mutations will have been fixed. Then after that, it's "harmonic mean" will be that of its maximum value since it will have a "constant population size" for the next 9,900 generations. To summarize: (1) I was right about scientists having to "back into" the N_e when actual information is lacking. (2) I was right about a 'constant' population size over many generations. (3) This means I was also correct in saying that an "effective population size" of 2,250, or even, 28,000 can be reached in a very few number of generations, and is, therefore, blind to the method being employed. (4) Just to underline all of this, from an earlier post, an "effective population size" of 2 looks like this (generation numbers as well as generations, separated by slash marks: 2/2/2/2/2/2/2/2/2/2/ .....................2/2/2/2/2/2/2/ .......................2/2/2/2/2/2/ ..............PaV_{March 13, 2014
March
03
Mar
13
13
2014
07:18 PM
7
07
18
PM
PDT}

That’s right. It isn’t
Progress, you've admited to getting something wrong :)
Why am I mistaken? Why am I mistaken? Because what I point to is something that Darwinists/evolutionary biologists refuse to look at?
Because the equations following your sentences don't appear to be related in any obvious way. You get yourself in another mess in this post which I don't have time to umtangle. If you wan to know what's the proability that pre-sepcified mutation will fix in a population by dirft in less that a million years then the answer is close enough to zero to make no difference. After all, chimps and humans have been evolving apart for 6 million years (=12 million years of evolution) and we have very few fixed differences. But why is that important? No one things drift is responisble for adaptation. Using neutral models to explain demography of genomic patterns doens't preclude us from using selective models to explain other parts of biology. As I've said from the start, there is a lot of evidence for the fact natural selection has been at work in our genome.
So, tell me, the time it took to go from a ‘chimp’ to a ‘human’, what is it? 6 million years? Per my calculation, via random genetic drift, this “buys” you 13 a.a.s in places where the ‘chimp’ to ‘human’ transition happens. Just out of curiosity: are there more than 13 a.a. differences between chimps and humans?
Well, your calculations are wrong. Let's look at it per gene, as the numbers are easier to follow, and you have the exonic portion of DNA wrong in your ealier comments. The average polypeptide is 400 amino acids long = 1200 nt. Using back-of-an-envelope numbers, the per nucleotide mutation rate is ~2.2e-8 and ~25% of mutations are synonymous. Over 6 million years, each lineage would have 240 000 25yr generations so we'd expect 1200 nt * mu = 2.2e-8 * 240000 gen * 2 lineages ~ 13 substitutions per protein. 13 per protein rather than 13 per genome. In fact, the observed number is smaller that the one we just calculated, due to the effect of purifying selection. The rate at which non-sysnomous mutation fix is ~1/5 the rate at which synomous ones fix (and 13/5 is 2.6 which is about the average number of amino acid differnce between human and chimp proteins). So the neutral rate would generate more differences than we observe, the fact we don't see so many is evidence for the operation of selction.
he answer is: Of course! Don’t you think, then, that there’s something wrong in presenting this supposed method as a way of bringing about “macro-evolution”?
When did I do that? I said we can use a neutral methods to infer demographic histories. That doesn't mean neutral theory explains all of "macro-evolution", what ever that term means to you.wd400_{March 13, 2014
March
03
Mar
13
13
2014
03:01 PM
3
03
01
PM
PDT}

wd400:
Which is clearly not the same as “the time required for a neutral mutation to be fixed at a particular location along the length of the genome”.
That's right. It isn't. But the "time required for a neutral mutation to be fixed at a particular location along the length of the genome" is the only way you can BEGIN to have "evolutionary change." So what if one million "neutral" mutations are fixed all over the length of the genome. As I've already asked you---and you have yet to respond---what, exactly, does this buy you? What value is there?
You now seem to be mistaken in a new way. though it’s hard to follow precisely what youa are trying to calculate. What exactly are you trying to calculate, and who is specifying these “needed” changes?
You admit that you don't know what I'm trying to calculate, and you admit you don't know why the changes are "needed", yet you're convince, it would appear, that "I'm mistaken in a new way." Why am I mistaken? Because what I point to is something that Darwinists/evolutionary biologists refuse to look at? Here's an example. I have a genetic defect within my hemoglobin protein coding area that results in defective hemoglobin molecules, which makes me chronically anemic. Now, let's say you're a geneticist, and it's your goal in life to 'cure' this disease. So you look along the length of the HG coding area and see the actual loci (there are two bad a.a.s involved) where these wrongly-coded nucleotides are found. You then devise an entirely new technique which allows you to replace these two or three (possibly four) "bad" nucleotides with the "correct" ones. The person is cured. Now, using random processes to 'cure' the same person requires that, at a minimum, two EXACT locations along the genome must be replace with the "CORRECT" nucleotide. The probability of replacing the nucleotide with the "correct" nucleotide via random genetic drift is 1/3 x 2,250/3,000 as calculated above. But this is ONLY HALF the problem. You need TWO mutations. So, for the probability of ONE of getting the "correct" replacement at the first location along the length of the genome, 3,000/2,250 x 3 x 112,500 years will be needed = 450,000 years. Now the FIRST mutation is in place. If you're lucky, and the FIRST mutation doesn't change itself again in the intervening 450,000 years (although it is likely that this will happen based on neutral drift ---REMEMBER: 100 mutations are fixed EACH generation, and all over the genome), it will take an additional 450,000 years to get the SECOND mutation. And, of course, much more than that if, which is likely, the FIRST mutation disappears. So,almost ONE MILLION YEARS to change TWO mutations. And, in my case, these two mutations are NEEDED. You don't mind if I do the specifying, do you? So, tell me, the time it took to go from a 'chimp' to a 'human', what is it? 6 million years? Per my calculation, via random genetic drift, this "buys" you 13 a.a.s in places where the 'chimp' to 'human' transition happens. Just out of curiosity: are there more than 13 a.a. differences between chimps and humans? The answer is: Of course! Don't you think, then, that there's something wrong in presenting this supposed method as a way of bringing about "macro-evolution"?PaV_{March 13, 2014
March
03
Mar
13
13
2014
01:06 PM
1
01
06
PM
PDT}

You said this So, just think what the brilliant population geneticists are telling us. That from the time than humans lived in Africa, only ONE!!! allele has become fixed, Which is clearly not the same as "the time required for a neutral mutation to be fixed at a particular location along the length of the genome". You now seem to be mistaken in a new way. though it's hard to follow precisely what youa are trying to calculate. What exactly are you trying to calculate, and who is specifying these "needed" changes?wd400_{March 12, 2014
March
03
Mar
12
12
2014
02:29 PM
2
02
29
PM
PDT}

I am on a tour in Africa. Our tour guide is from South Africa and like all good tour guides provides interesting tidbits about the local culture etc. He said there is a theory proposed by some South Africans researchers that there was a mass killing off of the human species several thousand years ago and the species was reduced to a small group along the eastern coast of South Africa. My comment was that he was saying South Africa was/is paradise. I have no ideas if this is bunkum or not but will ask him if he knows of the source of the data. The tour guide is well educated and was a computer programmer before heading off to the bush to live with the animals. My reaction to this is that science is a Whack a Mole process. Just when one thinks they have killed off one theory another comes out of nowhere that may be the answer.jerry_{March 12, 2014
March
03
Mar
12
12
2014
01:31 PM
1
01
31
PM
PDT}

wd400:
This is a common error. It would take ~112,500 years for any given mutation to fix. But there many mutations in each generation. In fact, the fixation rate is equal to the per-individual mutation rate. The brilliant population geneticists are telling us ~100 mutations fix in each generation.
This is NOT an error. It is the time required for a neutral mutation to be fixed at a particular location along the length of the genome. 4,500 generations x 100 mutations per generation = Total Mutations over 112,500 years = 4.5 x 10^5. These are "randomly" distributed along the length of the genome (if we assume NS is turned off, it can even happen in the just 5% of the genome that codes for protein). So, to change a protein, you would need a nucleotide change at a particular location = .05 X 3.0 x 10^9 nucleotides within protein coding regions=1.5 x 10^8 nucleotides. The probability of such a genome occuring during 112,500 years of "neutral evolution" is .......drum rolls please: 4.5 x 10^5/1.5 x 10^8 = 0.003. Now the Ne is 2,250. So the odds of such a mutation actually occurring in the population is 2,250/3,000. So there is less than ONE chance of ONE needed mutation to occur in this population over 112,500 years. But, wait a second. It's supposed to be "neutral", so it's likely not a single amino acid will be changed by this mutation---should it even occur. So, how, exactly, is "evolution" supposed to happen? That's starting at 112,500 years ago. So, starting 112,500 years ago, if you push back another 112,500 years, again, less than one chance of a needed mutation occurring. How do you propose overcoming these dismal statistics? Michael Behe and Snoke rans some of these kinds of numbers, looking just for a two amino acid substitution, assuming immediate fixation, but also elimination of deleterious mutations, and they came up with staggeringly huge amounts of time required for simple a.a. substitutions to occur; and this only occurs if your population size is enormous. That study led to "The Edge of Evolution." Have you read that book? I'll have more to say about harmonic means and effective population sizes tomorrow.PaV_{March 12, 2014
March
03
Mar
12
12
2014
12:23 PM
12
12
23
PM
PDT}

wd400:
The brilliant population geneticists are telling us ~100 mutations fix in each generation.
Except that "brilliant" shpuld never be used to describe population geneticists as they are far from it. And their methods do not apply to a Special Creation.Joe_{March 11, 2014
March
03
Mar
11
11
2014
12:37 PM
12
12
37
PM
PDT}

I mean the derrivation of the forumula for this effective population size. As I say, it's not straightforward but you can look it up in most popgen texts. It’s not clear to me what you’re getting at here. What are you saying exactly? it's (a) not true that 'in the 90?s, they went looking for “mitochondrial Eve”, and were certain that they would find more than ONE such “Eve.”' (b) not even possible that there could be more than one "eve"wd400_{March 11, 2014
March
03
Mar
11
11
2014
12:28 PM
12
12
28
PM
PDT}

wd400:
The derrivation of the forumlae for Ne are not straightforward, but are covered in many textbooks.
I'm not asking you to tell me how the formula is derived. It appears to simply be the reciprocal of the reciprocal of the arithmetic mean. So, kind of N=N/N(1/N). Why is it used here? Can you answer that question. I think I know, so I will be interested in your answer.
It’s not true, or even possible so far as I can tell, that Wilson and Cann though there would be more that one MRCA mitochondrial DNA (or Y-chom). How would that even work?
It's not clear to me what you're getting at here. What are you saying exactly?PaV_{March 11, 2014
March
03
Mar
11
11
2014
12:15 PM
12
12
15
PM
PDT}

Prev 1 2 3 4 5 Next

You must be logged in to post a comment.

Leave a Reply