Here’s a simple experiment one can actually try. Take a bag of M&M’s, and without peeking reach in and grab one. Eat it. Then grab another and return it to the bag with another one, from a separate bag, of the same colour. Give it a shake. I guarantee (and if you tell me how big your bag is I’ll have a bet on how long it’ll take) that your bag will end up containing only one colour. Every time. I can’t tell you which colour it will be, but fixation will happen.
This models the simple population process of Neutral Drift. Eating is death, duplication is reproduction, and the result is invariably a change in frequencies, right through to extinction of all but one type. You don’t have to alternate death and birth; choose any scheme you like short of peeking in the bag and being influenced by residual frequencies (ie: frequency-dependent Selection), and you will end up with all one colour.
Is Chance a cause here? Well … yes, in a sense it is, in the form of sample error. Survival and reproduction are basically a matter of sampling the genes of the previous generation. More random samples are a distortion of the larger population than aren’t, so, inexorably, your future populations will move away from any prior makeup, increasing some at the expense of others till only one variant remains.
Selection is a consistent bias upon this basic process. If different colours also differed a little in weight, say, more of some would be at the bottom of the bag than others, so you’d be more likely to pick one type than another. In more trials, the type more likely to be picked would be picked more often, to express it somewhat tautologously. You’d get a sampling bias.
Both of these processes are random – or stochastic, to use the preferred term. In reality, they are variations of the same process, with continuously varying degrees of bias from zero upwards. It makes no sense to call selection nonrandom, unless by ‘random’ you mean unbiased. Where there is no bias, all is Drift. But turning up the selective heat does not eliminate drift – sample error – and so does not eliminate stochasticity.
With a source of new variation, these processes render evolution inevitable. Even with a brand new mutation, with no selective advantage whatsoever, 1/Nth of the time (where N is the population size) it will become the sole survivor. That’s the baseline. If there is a selective advantage, it will be more likely and quicker to fix, on the average. If at a selective disadvantage, it will be less likely and slower.
Conversely, without a source of new variation, all existing variation would be squeezed out of the population, and evolution would stop.
Actually all the heavy lifting seems to have happened all at once during the Cambrian explosion.
But yes, I agree with you, its almost impossible for small mutations to pinpoint exact morphological changes, we can almost no longer think of one gene relating to one feature. But that begs the question, how did this complicated interwoven system develop? Its pretty hard to say THAT happened by small step mutations, but now we need wholesale changes across many platforms to see any adaptations. How do we account for this completely new way of how organisms form?
This is why more and more biologists are beginning to question the role of Darwinism. If it is not really a Darwinian process now, how can we say it was before. That’s too much special pleading.
olegt,
Really, one can see that can they? How many offspring does each individual average? What is the gestation time, and what is the number of years to child bearing age? What is the average life expectancy of each individual? One can see this from your chart?
Because without this information, its completely fabricated nonsense.
phoodoo,
It has been known for many years that evolution consists of a complex interplay of NS, mutation, recombination, migration and drift. If a biologist ‘questions Darwinism’ (ie: downplays NS), this does not necessarily mean they favour ID, front loading, ‘natural genetic engineering’ or any of the other notions that occasionally resurface before withering from lack of evidential support (except, of course, in the world of intertube commentary).
phoodoo,
These are questions irrelevant to population genetics. The model Allan introduced has M&Ms, not mammals. It would also be irrelevant to population dynamics of bacteria, for which gestation and child bearing age are meaningless terms.
You are throwing up distractions at this point.
Why do we need tonnes of them cropping up all over them place if the complexity of life as we know it developed over a period of 3.8 billion years? It seems rather the other way around, that given such enormous timescales, the accumulation of novel changes are slow. In fact on human lifetime timescales, such changes should be pretty much undetectable with the naked eye.
If you think this is wrong, maybe you can find a way to quantify how many “tons” of changes are required to consistently emerge, and how notable they should be in your view. Perhaps we can try and link the frequency with which you imagine new features should be popping up with transitional periods inferred from the fossil record. Let’s take the proposed transition from fully terrestrial mammals to something resembling modern whales. The fossil record implies this transition took approximately 20 million years. Okay, can we then try to quantify how much “change” and “novelty” we should be expecting to see in a human lifetime, say 80 years? Let’s do a little, admittedly very simplistic thought-experiment here.
So if evolution is right and the nostrils at the snout of a terrestrial mammal slowly wandered from the front and coalesced into a blowhole at the top of the head on a fully aquatic whale over a period of 20 million years, we should expect to see something like the position of the nostrils of a mammal move a distance 1/250.000th the distance between the snout and the top of the head. (20.000.000/80=250.000)
Does this seem to you to be an infeasible level of change for mutations to accomodate? If not and you concede mutations can change the location of the snout by 1/250.000th the distance, why shouldn’t this be able to simply continue? What magic barrier sets in and prevents further change due to mutation? If mutation can move the nostrils 1/250.000th the distance, why not 2/250.000th? or 4/250.000th?
How do you know it doesn’t? How long have you been spending in the wild, measuring morphological change of entire populations of organisms? Would a change of 1/250.000th the distance from the snout to the top of the head in the position of the nostrils in your dog be readily apparent to you? Have you been around for a million years, or how about 20 million?
We certainly see small variations, which, given the timescales is actually what we expect. If we really did see the kinds of massive changes that you allude to, on tiny human timescales, it would make a mockery of the theory of evolution as we understand it, because the transitional times implied in the fossil record would be absurdly slow in comparison to the measured rates of change in living populations.
Sure it would, but if that’s not how evolution works, then it isn’t how evolution works. Nobody believes a completely new and additional brain can evolve on the timescale of a human lifetime.
What is this silly nonsense? Do you actually expect to see Crocoducks?
Actually, second brains have existed, and distributed nervous systems still exist.
A third arm implies radial symmetry, which does exist — in starfish, for example.
One of the evidences for common descent is that body plans tend to get frozen in lineages. But if you go back to the earliest animals, before complex adaptations evolved and before predation narrowed the scope of variation, you see that odd body plans can exist.
You’re kidding, right? Every human being on Earth has a hundred or more mutations i.e. “anomalies”. It used to happen and is happening and will happen.
One could argue that human civilization has altered the environment which selects those “anomalies”, but claiming that they don’t exist… well, any comment would break site rules.
This triggered me to spend a while (Mrs F is away looking after her sister’s house and kids so her sister and brother-in-law can take a second honeymoon) googling cephalopod nervous systems.. Fascinating!
It may not surprise you to hear that I disagree. Given deuterostomes and ectoderm, mesoderm and endoderm, the rest is origami.
@Allan Miller
Apologies for contributing to the derail. Though if the Lord wanted us to consume refined sugar in large quantities, he would have not created the lust for sugar, type 2 diabetes or sugar cane. Then the slave trade wouldn’t have got off the ground.
The lesson to take away is avoid refined sugar!
I’ll get me coat!
Precisely because of the limits of evolution. An intelligent designer, for sure, could give us a third arm or an extra brain, if she thought we could use one.
But evolution can only retrofit. It’s one of the reasons why evolution is the better model for what we observe. Or would be, if ID proponents would ever make an actual prediction (“We could use an extra arm, which is obvious to an Intelligent Designer, so ID predicts that soon our children will be born with one”)
I find that human designers generally design assembly and repair methods that require at least three hands.
If you google “third hand” you will find some interesting products, most of which are safe for work.
If I look at my computer, in terms of body plans, everything important seems to have arisen decades ago with the basic tower design.
When I look inside at the processor and memory, then I see that everything important has changed more recently, even though the basic body plan stays the same.
You are giving too much weight to body plan, and not nearly enough to small details.
JonF,
I think phoodoo was talking about what creationists tend to term ‘macroevolution’ (not quite the same as the technical use of the term) – big changes in ‘body plan’ (ditto): how come they‘ve stopped happening? Which they may or may not have, but how could we tell if one was in train?
Alan Fox,
A sugar coat?
Don’t worry, the only thread that fails to derail is one with 0 comments! 🙂
Isn’t the SRGAP2 gene that got duplicated 3 times in the human genome decent evidence that one duplication of a gene can make a big positive difference?
pastafarian,
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3365555/
Phoodoo likely doesn’t realize it, but he’s made use of an amusing motif that shows up in a fair few of Creationists’ anti-evolution arguments. Particular examples of this motif include “where’s the crocoducks?” and “if evolution is true, how come you never see any cats giving birth to dogs?” The general form of this motif is [Specific piece of evidence that would actually refute evolution if it were ever observed] hasn’t been found, therefore evolution is false, and the fact that Creationists are capable of using that motif indicates that at best, they really don’t know what the hell they’re talking about. Whenever you see this motif in action, you may be confident that it’s based on a weirdly distorted caricature of evolutionary theory
The particular form of the motif phoodoo is using here is based on the implicit presumption that evolution is capable of bestowing any ‘beneficial’ trait on any critter at any time. Needless to say (if you actually do understand evolution), this presumption is false. Fundamentally, evolution is about descent-with-modification; what that means is, a new Trait X can show up in a critter if and only if that critter’s ancestor(s) possessed a different Trait Y which could be modified, by mutation or whatever, to become that new Trait X.
In other words: Show me a species than doesn’t have toes, and I’ll show you a species that evolution can’t add toenails to.
Looking at phoodoo’s argumentation, okay, maybe a third arm would be useful to have. But a third arm will only show up in a species which possesses a physical structure that could be modified to produce a third arm! Tell me, phoodoo: What physical structure does the species Homo sapiens possess that could be modified to produce a third arm?
And let me echo Miller’s armor-piercing question: If some allegedly-spiffy Trait X would be so damn wonderful, why hasn’t the Designer given us that Trait X? The Designer isn’t subject to the limitations of evolution, after all—I mean, hell, that’s the whole friggin’ point of invoking the Designer, ennit?—so if that Trait X really is something that would be spiffy for us humans to have, the fact that us humans don’t have that Trait X is a bit of a conundrum for Creationists.
And yes, it’s not at all uncommon for Creationists to present arguments against evolution which would, if those arguments actually were valid, refute Creationism. Go figure, huh?
The more you try to justify your position, the less credible I find you as someone to waste time discussing something with. You were the one arguing that a generation couldn’t possibly be only 233 days or 23 days, and it was ME who had to remind you that not all organisms are humans.
Four new births, replacing four existing members of the population, in order to fixate a new mutation is NOT I repeat only one generation. You still can’t justify this dumb logic, and now you want to say well, the number is too small. If the logic doesn’t work for 4 it doesn’t work for 20. You start with a bad premise, and then expect it to make sense later-with no justification for such a claim.
Until any Darwinian can give a plausible explanation (using random mutations) of how a complex physical feature developed, your argument is bunk. It is your camp that claims things like an eye develop from a freak mutations which accidentally causes a light sensitive spot on an organisms, then many generations later another freak mutation causes that sunspot to get a depression right exactly at the spot of the light sensitive patch.
If this were true, number one, light sensitive patches couldn’t possibly have stopped happening randomly to some organisms to give them a chance to evolve, and likewise mutations that cause rods and cones in the backs of eyes couldn’t have stopped happening.
So do worms occasionally get random mutations which cause new light sensitive patches in other parts of their body, and rods and cones to develop anywhere on their body? Because why would we expect it to only happen once, a million years ago, in exactly the spot on the body that it is needed? The odds of that would be too preposterous.
So go ahead, come up with a feasible scenario of developing a brand new body plan, using Darwinian steps of random mutations. Humans have been studying millions of animals for long periods of time, any examples of random rods and cones mutations appearing anywhere on a body plan? How do you form hole sockets in a skull for the eyes through random mutations?
phoodoo,
If the 4 births were each to a different one of the starting 4, then of course it’s one generation. 4 parents, four children. A generation is a node in descent. If the 4 were successive descendants of one, it’s 4 generations: 4 original individuals, 3 sans issue and one leaving child, grandchild, great-grandchild and great-great grandchild. Those are the extremes (assuming death of all original members). Averaging out over larger and larger populations and longer runs leads to a distribution, centred on 1 generation per N replacements in this particular model. It’s hardly vital though, I don’t know why you are so hung up on this.
phoodoo,
You think skulls came first, then eyes?
Since all of your proposals involve a random mutation occurring exactly in a spot where it is fortuitous, it doesn’t really matter which you claim happens first, because any sequence becomes just as improbable. Which came first, a hole in the head for an ear, or a skull with holes where ears can form?
How does the sequence effect the problem of getting a skull to form in exactly the right shape by trial and error?
That is an amusing misunderstanding on your part, phoodoo. 🙂
I used as an example a human population in a town of N=100. With a lifespan of 70 years, one human dies every 70 years/100 = 255 days. Your proposal to equate one death with one generation means that a human generation in this town = 255 days. And in a town with N=1000 inhabitants, 26 days.
That is of course silly. It is silly because counting a generation as an interval between consecutive deaths is silly. Your solution? Declare that “not all organisms are humans.” Voila! A housefly lives just a day, so that solves the problem of a short generation! I am not kidding, this is exactly your logic!
It’s doubly silly of course to start with humans and then declare that we are talking about generations of houseflies. I am not making this up!
If you wish to work with houseflies, start with the lifespan of a housefly, 1 day. In a population of N=100 houseflies, deaths will occur every 15 seconds. Your count then yields 15 seconds as one generation of houseflies. And with N=1000 houseflies, one generation = 1.5 seconds.
This is sheer nonsense of course.
Again, you misunderstand the most basic things.
Take Joe’s recipe and paint numbers on the backs of M&Ms.
Start with three blue M&Ms and one black, all with number 0. Here is the sequence of events where only one new generation is produced:
Blue 0 dies.
Black 0 gives birth to Black 1.
Blue 0 dies.
Black 0 gives birth to another Black 1.
Blue 0 dies.
Black 0 gives birth to another Black 1.
We now have four black M&Ms, one with a 0 and three with a 1. There is only one new generation past the original one.
That wasn’t too hard, was it? 🙂
There is no “exactly right shape”. There are lots of possible shapes that would work well enough.
Right, and if the four births were each to different members of the same generation, then you obviously are talking about four births to four completely different lineages, that is the whole point. If its the same generation you are not going to get the novel mutation crossing into all four lines of heritage. That’s why you need successive generations, not concurrent ones. Calling it one generation is meaningless.
M&M’s are not people, I said this long ago. But Olegt thinks that just because he can fabricate any number story he wants and call it real, and then have the gall to go boasting about his tortured logic for a week, then I will call him on that. He even went so far as to say, “How can a generation be 23 days, and don’t you know how long it takes humans to have babies, and the average lifespan is 70 years, so how many deaths occur in 70 years….” Yea well, how long does it take M&M’s to have a baby, give me a freaking break. Four still doesn’t equal one.
So this time we just got lucky that the holes are right where the eyes and ears are? It would work just as well if the holes were in your shins, and the eyes were on your elbows?
Yea no kidding, you can make up any stupid story you want, that one progenitor outlives everyone else in the entire population and thus gives birth to everyone on the planet, but that doesn’t make it any more logical. And you also need one generation to first get the black mutation, so your four equals one argument already falls apart right there. Do you have OCD?
GUANO
Heh. 🙂
I have said, here and here, that fixation in one generation is possible, but not typical. Most of the time it will take more generations than one. Here is output from my code with N=4 M&Ms:
Run 145: {2, 0, 1, 1} at t=0; {0, 0, 0, 4} at t=11; 3 generations.
Run 151: {1, 2, 0, 1} at t=0; {0, 0, 0, 4} at t=3; 1 generations.
Run 153: {2, 0, 1, 1} at t=0; {0, 0, 0, 4} at t=8; 2 generations.
Run 161: {0, 0, 3, 1} at t=0; {0, 0, 0, 4} at t=4; 1 generations.
Run 163: {0, 1, 2, 1} at t=0; {0, 0, 0, 4} at t=5; 2.5 generations.
In run 151, fixation occurred in 3 steps as described above. It took 1 generation (as defined by Joe). In other runs, fixation took more steps; accordingly, more generations arose.
The average number of generations was 2.85. Not one.
My starting configurations had one black M&M. If you wish to start with no black M&M and count adding one as one extra generation, fine. The above result will change to 3.85 generations.
For a large population N of M&Ms, your +1 correction is a drop in the bucket. The average number of generations to fixation is N. Changing it to N+1 makes no difference if N is large.
You are talking to someone who is shocked by the idea that evolution could produce legs exactly the correct length to reach the ground.
The question about eyes and skulls is actually quite interesting. It’s a wonderful question. I wonder if biologists have ever thought about it, or whether phoodoo is the first.
What are the odds?
When your whole theory is based on story telling, just about anything can make sense.
olegt,
Run 145: {2, 0, 1, 1} at t=0; {0, 0, 0, 4} at t=11; 28 generations.
Run 151: {1, 2, 0, 1} at t=0; {0, 0, 0, 4} at t=3; 140 generations.
Run 153: {2, 0, 1, 1} at t=0; {0, 0, 0, 4} at t=8; 679 generations.
Run 161: {0, 0, 3, 1} at t=0; {0, 0, 0, 4} at t=4; 0 generations.
Run 163: {0, 1, 2, 1} at t=0; {0, 0, 0, 4} at t=5; 5487 generations.
Look, it took an average of 1,266.8 generations!
Aren’t numbers great when you can make up any formula you want!!
Seriously, you should check this out. It’s not story-telling and not even, strictly speaking, evolution.
phoodoo,
The point of Allan’s opening post was to encourage people to explore a very basic, and well understood, model of population genetics. He stated the rules of the model and made a prediction that sounds counterintuitive to the uninitiated:
You took the bait:
Allan pointed out that fixation will happen once in a while even in such a case:
To which you flippantly replied:
Well, guess what? You were wrong and Allan was right! Even with just one black M&M, the entire population can switch to black, with the probability of 1/N. Just like Allan said! You would learn that if you actually ran numerical experiments yourself.
For a smart person, this would be a good lesson. But then, again, a smart person would not make silly claims about a subject that he or she does not understand.
Well, learn the hard way.
And it took over 1,000,000 birth and deaths substitutions to do it! That is the point! In a population of only 1000 you need to pull out an M& M and replace it 1,000,000 times and kill off 1,000,000 existing M&M’s to get that result!
So yea, I agree, if you are willing to just throw out numbers until you have a big enough computer to get the results you want, it could happen in your theoretical, crazy math world. So like I said, aren’t numbers great if you want to make any formula you want! Congratulations. 4 still doesn’t equal 1.
A million deaths in a population of a thousand is not that long. A thousand generations.
A well-known result in population genetics. And even though I do not work in population genetics, I was able to learn that with a code that took 15 minutes to write.
You, too, could learn that. You chose not to. Everyone is entitled to their choice.
Heh. I used my old laptop.
olegt,
Its a thousand generations to you, because you seem to think that the only ones in the population who ever survive are the ones who have the “neutral mutation” and they give birth to the entire race on the planet while everyone else dies. I am not sure why you are bothering with the concept of neutral.
Now ramp it up to a population of 1,000,000 and you will need to pull out 10 zillion M&M’s before you get the result you want. But fortunately Intel chips have come a long way in the last decade.
Ah, you haven’t figured out the point of this exercise, have you? Here is a big secret, phoodoo. Don’t tell anyone.
Neutral drift is important in small populations. It overwhelms the effects of natural selection. In large populations, neutral drift is unimportant. As in your example. Generally, fixation through drift requires N generations in a population of N organisms.
OK, I was kidding. This isn’t a big secret. It’s a major, and well known, result in population genetics. See Wikipedia for starters.
Hee hee.
It’s fine to be rude about the argument or evidence given – or lack of it. It’s against the rules to act as if you assume the poster is not in good faith, eg. is lying, is bluffing, is deliberately telling you crazy things.
Which is what you just did:
I call them as I notice them, and of course I don’t notice them all — but I also don’t refrain from calling out my “friends” the same as my “opponents”.
Overall, I’d estimate I’ve been guano-producer #9 or 10 … worse than most of the long-time regulars who have been more polite and more restrained than I am … better than most of the UDers/IDers who it seems can’t restrain themselves from revealing their beliefs that “evolutionists” and “atheists” are just one big pack of liars.
I’m pretty sure I haven’t made a comment recently which implies that you are not posting in good faith, so I’m pretty sure that I’m keeping myself within the spirit and letter of the rules now. You’re welcome to give specific quotes where you believe I’ve flouted the rules; I’ll be happy to consider if I need to improve my behavior further.
I would say it’s a remarkably adaptable theory.
ToE can’t explain all possible observations and like all good scientific ideas is quite falsifiable. The fact that it does explain virtually all of the empirical observations we do make just highlights the strength of the theory.
Evolution is intended to explain the diversity of life on Earth. That diversity covers creatures that fly, and creatures that cannot fly; creatures that breathe water, and creatures that drown in water; single-celled creatures, and multi-cellular creatures; and so on and so forth. Any theory which can explain all those different creatures would have to be pretty damned adaptable, wouldn’t it?
One could argue that Creationism is even more adaptable than evolution, on the grounds that “the Creator did it” can explain absolutely any creature whatsoever. Evolution, contrariwise, cannot explain absolutely any creature whatsoever; rather, evolution can only explain those creatures we’ve actually seen up to now, and “those creatures we’ve actually seen up to now” is necessarily a limited subset of “absolutely any creature whatsoever”. Every time another formerly-unknown creature is discovered, that formerly-unknown creature could be a creature which evolution cannot explain… but so far, not one single one of them has been a creature which evolution cannot explain.
It is of course always possible that at some future date, a creature will be discovered that evolution can’t explain. Alas for Creationism, that day has not arrived yet.
Piltdown2,
Suppose you’re correct that macroevolution can’t be explained without invoking a guiding intelligence.
How do you explain the fact that out of zillions of alternative design patterns, the designer chose one of the very few that we would expect to see if macroevolution were in fact unguided?
Is the designer shy and attempting to hide? Is the designer a big fan of evolution who wants to emulate it?
Which of the following do you think is more likely?
a) Macroevolution appears to be unguided because it is unguided.
b) Macroevolution appears to be unguided because the designer is mimicking unguided evolution.
phoodoo,
If you have eyes first, and they are useful, all skull growth mutations that cover them or block their nervous connection to the brain would be detrimental and eliminated by natural selection.
If you have skulls first, eyes of the present type are very unlikely – that they develop first and then somehow hook up into the nervous system.
But there is no fossil evidence of the latter sequence. Bone evolved in fish with eyes and brains. Eyes came first, co-evolving with a ‘control centre’ that became the brain, which it proved useful to encase in hard material when that evolved.
phoodoo,
Ah, the zillion. Is that larger or smaller than the squillion? 😉
But (as I’ve said before) – remember mutation rate. If the neutral mutation rate is 1 in 1000 (one assumption for the presence of a black M&M in our starting population of 1000), then you have 1000 neutral mutations in a population of a million. Thus, you have 1000 ‘blacks’. The population of ‘non-black’ will be eliminated in 2N generations (ie, recent mutation will be fixed, and the prior eliminated), in 1 in 1000 replicates. Exactly the same.
Of course the mutation rate doesn’t have to be 1 in 1000. But whatever it is, a population of 1,000,000 will be producing mutations 1000 for 1 compared to the population of 1000.
You could argue (and be right) that the 1000 mutations are actually shades of black … er, ie grey. It takes much longer to fix a specific one. But the point is that the population has lost every representative of its prior allele when ‘shades-of-grey’ is fixed. And that happens at the same rate in a population of a million and one of a thousand. The upshot is that the fixation rate is the mutation rate.