Answering any questions on Evolution

[NailsII]

You must be a well paid researcher! Your examples all show the eardrum in front of the reflective surface.
The hearing horn being the least obvious. But in fact, I was wrong. The eardrums are not hidden behind the
owls feathers or below the curve of the reflective surface, as I stated knew they couldn't be.
And the owls face is not concave either as was originally argued.

Why can't you just accept that you are wrong?

 

[bobsterInJapan]

This thread is getting mighty long..

Is the OP sill open for questions?

I have heard of a type of owl, though I don't know the name of it, which has a face that is shaped like a dish, as in a statellite dish. It also has left and right ear holes that are vertically out-of-line with respect to each other. This owl has amazing hearing and can tell accurately where sounds are coming from, including vertical components.

How could such a bird evolve in such a way. How did his/her alleged ancestors evolve a dish face, especially, the beginning stages, i.e., the owl's face slightly becoming concave. Wouldn't an ever-so-slightly concave face have no real advantage over a non-concave face. Do you get what I mean?

Hi, I'm back, quoting my own question.

Firstly, I understand that this owl has a face that is shaped like a satellite dish, but doesn't operate exactly like a satellite dish. But I do understand that this owl has outstanding hearing capability due to the shape of its concave face and it's asymmetrically positioned ears.

I think I understand, in theory, how natural selection could work.

I assume we are talking about mutations that occur in the owl's DNA.

How does the DNA know to keep on mutating on the same direction? How does the DNA mutate so that the owl's face becomes concave and then more concave without going slightly less concave (or going slightly convex)?
I suppose that might still be due to whether the mutation exhibits an advantage or disadvantage.

What are the chances of the same type of advantageous mutation occurring time and time again? The DNA has no choice which DNA to mutate and has no say as to whether the mutation is going to be advantageous one or not. I also wonder how many advantageous mutations are needed in such a case. It seems like a very unlikely scenario to me.

 

[CabVet]

How does the DNA know to keep on mutating on the same direction? How does the DNA mutate so that the owl's face becomes concave and then more concave without going slightly less concave (or going slightly convex)?
I suppose that might still be due to whether the mutation exhibits an advantage or disadvantage.

What are the chances of the same type of advantageous mutation occurring time and time again? The DNA has no choice which DNA to mutate and has no say as to whether the mutation is going to be advantageous one or not. I also wonder how many advantageous mutations are needed in such a case. It seems like a very unlikely scenario to me.

DNA doesn't "know" to keep mutations in the same direction. What knows that is natural selection. Every population has natural variation, in this case, owls with more concave and owls with less concave faces. Because the owls rely on hearing to find their prey, the ones that find prey faster will do better (get more food, sustain more offspring), and therefore their genetic characteristics will pass on to a larger proportion of the next generation than that of owls with a less concave face. That is how things move into a certain direction, because of natural selection. Mutations are just the raw material that natural selection works on.

There is a very good analogy out there, think of evolution as water. Mutations would be the rain, and natural selection a river. Mutations drop everywhere but they always end up in (and are directed by) the river.

The fish (a species of butterflyfish) below is in big trouble because natural selection put it in a tricky positions. It feeds exclusively in one species of coral, but the coral is going away because of pollution and global warming, so the fish populations are declining.

 

[Mr Strawberry]

Why can't you just accept that you are wrong?

Yes, the trouble with the idea of being wrong is that it goes against the whole concept of faith, which is a deliberate attempt to believe something unbelievable. You can see the problems that introducing no-nonsense ideas like 'being wrong' could wreak with something like that.

 

[Wiccan_Child]

Hi, I'm back, quoting my own question.

Firstly, I understand that this owl has a face that is shaped like a satellite dish, but doesn't operate exactly like a satellite dish. But I do understand that this owl has outstanding hearing capability due to the shape of its concave face and it's asymmetrically positioned ears.

I think I understand, in theory, how natural selection could work.

I assume we are talking about mutations that occur in the owl's DNA.

How does the DNA know to keep on mutating on the same direction? How does the DNA mutate so that the owl's face becomes concave and then more concave without going slightly less concave (or going slightly convex)?
I suppose that might still be due to whether the mutation exhibits an advantage or disadvantage.

Exactly. The mutations cause change in both ways - but only one way causes an advantage. Mutation is random, but natural selection is not, and it's natural selection that picks which mutations get passed on and which don't.

Think about a bear in a normal environment. It's getting colder, so longer/thicker fur is advantageous. Mutations change fur length in both directions, but only longer fur is advantageous. You're right in that the mutations don't know which way to go, but natural selection does.

What are the chances of the same type of advantageous mutation occurring time and time again? The DNA has no choice which DNA to mutate and has no say as to whether the mutation is going to be advantageous one or not. I also wonder how many advantageous mutations are needed in such a case. It seems like a very unlikely scenario to me.

It's actually not. Mutations occur all the time, causing small variations in the average individual. It may seem counter-intuitive (and it is!), but the universe doesn't care about our intuition .

 

[Naraoia]

What are the chances of the same type of advantageous mutation occurring time and time again? The DNA has no choice which DNA to mutate and has no say as to whether the mutation is going to be advantageous one or not.

True. How often the same type of change is going to arise depends on the genetic underpinnings of a trait. For example, something like height is probably influenced by many, many genes. Mutations in any of them could potentially lead to the person being taller, so new mutations producing taller people would be relatively common. On the other hand, if a trait is determined by a few genes or even a single gene, advantageous mutations may be rarer, and it may be more difficult to collect more than one of them.

I also wonder how many advantageous mutations are needed in such a case. It seems like a very unlikely scenario to me.

This is a surprisingly difficult question to answer. Individual mutations can have very small or very large effects, and this is a bit unrelated to the type or size of the mutation. Some "small" mutations have large effects on the organism. For example, this terrible disease is often caused by the replacement of a single DNA base. So is the sickle-cell haemoglobin that can confer malaria resistance if you only inherit it from one parent and give you sickle cell disease if you inherit it from both. Flies with their antennae transformed into legs or their legs transformed into antennae can be made by changing the activity of just one gene. And then you can delete millions of base pairs from an animal's DNA and not see any obvious effect.

So... the number of mutations needed to "perfect" a new advantageous trait is very hard to predict. Perhaps only a few would be enough. For example, animals already have a genetic pathway in place that controls all kinds of asymmetries from which side your heart is to which way snail shells curl. Connecting up this asymmetry pathway with the genetic circuits that determine where the ear will develop may not be very difficult. (Disclaimer: I have no idea if this is indeed how owl ears became asymmetric.)

 

[Loudmouth]

What are the chances of the same type of advantageous mutation occurring time and time again?

I thought this was a really good question that needed more attention.

The answer is . . . extremely low. That is why any single advantageous mutation is much more likely the result of a single event that spreads through a population over several generations than several independent events.

This also has very important ramifications when we look at evolution overall. I think languages would serve as a good example analogy for this process. The Romance Languages are a lineage of languages that "evolved" from Vulgar Latin:

Changes to languages occur somewhat randomly and unpredictably. We would not expect two separate cultures with no meaningful contact to accumulate the same changes to their langauge. Instead, we would expect isolated cultures who share a common ancestral tongue to slowly change their common tongue in different ways thorugh time, and that is EXACTLY what we see. In fact, they change so much over time that when these cultures come into contact they are no longer able to understand one another even though they share a common ancestral tongue. This is equivalent to speciation in biology.

The fact that the same mutation does NOT occur because of the same exposure to the environment is a very important fact to understand. It has specific ramifications for what we should see in the genomes of living species. This is what allows us to test for common ancestry and evolutionary histories when comparing the genomes of modern species. It is very much like comparing modern languages to understand their common ancestral origin as well as the changes in each language that caused them to diverge over time.

It seems like a very unlikely scenario to me.

If science has taught us one thing it is that the universe could care less about what we find unlikely or likely.

 

[SkyWriting]

I thought this was a really good question that needed more attention.

The answer is . . . extremely low. That is why any single advantageous mutation is much more likely the result of a single event that spreads through a population over several generations than several independent events.

Which still points to it not happening. Most mutations are fixed by repair genes. Then there is the reproductive process that reduces the likelihood of expression by 50% in general. So vast numbers of the same mutation is required in order to show up in one offspring. Where it is again diluted by 50%

unless you mate with your sister who may happen to have the same mutation.

So DNA expression is not like language. Language sets can pick up a new word and promote it by a thousand fold or a million fold in one generation. Actually in about 2 hours most of the worlds population could be exposed to a new word. Intercourse is a slower process. At my house anyway.

 

[CabVet]

Then there is the reproductive process that reduces the likelihood of expression by 50% in general. So vast numbers of the same mutation is required in order to show up in one offspring. Where it is again diluted by 50%

Weren't you just criticizing people for twisting the facts to make points in another thread? This is completely wrong, learn some genetics before you post stuff like this.

We know, as a matter of fact, from sequencing entire human genomes, that every human being has on average 50 to 100 mutations on its genome. In other words, every child is 50 to 100 mutations away from the combination of their parents genes. And no, "vast numbers of the same mutation" are not required for it to show up. The only requirement is that the said mutation happens in an egg or sperm.

Now, if you multiply that 50 to 100 by the number of humans in this planet, times the number of generations that you want to go back you will get a grasp of how much genetic variation there is in humans.

If you still don't believe that genetics and mutation can produce enough variation for evolution to work on, just take a look at what happened to dogs in 10 thousand years:

 

[CabVet]

So DNA expression is not like language. Language sets can pick up a new word and promote it by a thousand fold or a million fold in one generation. Actually in about 2 hours most of the worlds population could be exposed to a new word. Intercourse is a slower process. At my house anyway.

Again twisting the facts, how about you take a look at the languages on that diagram again? Those are older languages, back when there were no internet forums around. So no, back then it did not take 2 hours for the world's population to be exposed to a new word.

 

[Loudmouth]

Which still points to it not happening. Most mutations are fixed by repair genes.

Some are not fixed. Each and every person is born with 50 to 200 mutations not found in either parent.

So vast numbers of the same mutation is required in order to show up in one offspring.

Nope. All it requires is a single mutation that is then spread through the population over several generations. It's called inheritance. You should look into it.

So DNA expression is not like language. Language sets can pick up a new word and promote it by a thousand fold or a million fold in one generation. Actually in about 2 hours most of the worlds population could be exposed to a new word. Intercourse is a slower process. At my house anyway.

It was only meant as an analogy showing how divergence can occur between two isolated populations. You are making the mistake of arguing against the analogy instead of the actual theory.

 

[Wiccan_Child]

If you still don't believe that genetics and mutation can produce enough variation for evolution to work on, just take a look at what happened to dogs in 10 thousand years:

BUT DEY'RE STILL DAAAAAGS

 

[bobsterInJapan]

Thanks everyone for your interesting replies. I feel as though I have learnt something.

I have a very different question this time. It is related to DNA but a somewhat different concept to natural selection, as far as I can understand it. Namely, taking the example of the dogs in that chart above. I think it would be fair to say that all dogs have one common first ancestor of dog. Suppose the first ancestor of dog was created with all the DNA of the dogs included in that chart. However the first generation puppies, though having approx. the same amount of variety of DNA, would have no choice but to mate amongst themselves, which in itself would pose no problems with birth defects because they still have a large variety of DNA. However, after a number of generations, these dogs divided up into different families, and each family moved further and further apart from each other. After more and more generations, these families became different 'breeds' of dogs. However, now, none of these breeds have anywhere near the amount of variety of DNA that the first ancestral dog had.

Now for my question, if someone were to study the DNA of these various modern dog breeds, would their DNA give the illusion of being evolved via natural selection? Richard Dawkins sometimes mentions the phrase "illusion of design", could an illusion of evolution be possible?

 

[Wiccan_Child]

Thanks everyone for your interesting replies. I feel as though I have learnt something.

I have a very different question this time. It is related to DNA but a somewhat different concept to natural selection, as far as I can understand it. Namely, taking the example of the dogs in that chart above. I think it would be fair to say that all dogs have one common first ancestor of dog. Suppose the first ancestor of dog was created with all the DNA of the dogs included in that chart. However the first generation puppies, though having approx. the same amount of variety of DNA, would have no choice but to mate amongst themselves, which in itself would pose no problems with birth defects because they still have a large variety of DNA. However, after a number of generations, these dogs divided up into different families, and each family moved further and further apart from each other. After more and more generations, these families became different 'breeds' of dogs. However, now, none of these breeds have anywhere near the amount of variety of DNA that the first ancestral dog had.

Now for my question, if someone were to study the DNA of these various modern dog breeds, would their DNA give the illusion of being evolved via natural selection? Richard Dawkins sometimes mentions the phrase "illusion of design", could an illusion of evolution be possible?

Only in the most contrived way possible. Such a scenario would only occur if the system were set up so that, at a specific point in time (say, the early 21[sup]st[/sup] century), dogs looked like they evolved from a common ancetsor. It wouldn't happen as a by-product of some other function.

Your analogy has some problems, though:

• "Suppose the first ancestor of dog was created with all the DNA of the dogs included in that chart" - what does that mean? What does it mean to say the ancestral dog was created 'with' the DNA of the dogs on the chart?
• "However the first generation puppies," - who did the ancestral dog mate with? Are there now two? This isn't a major issue, but it is unclear.
• "though having approx. the same amount of variety of DNA," - the same amount as what? Each puppy would have a distinct genetic code; are you saying the ancestral dog gave birth to, in the same litter, a genuine Chihuahua, Great Dane, Labrador, and all the other breeds? This doesn't match with what you later say about drifting families of dogs becoming the various breeds. So, are you saying that the puppies themselves have within them this variety of DNA - if so, what does "variety of DNA" mean? Is their DNA in some sort of quantum flux?
• "would have no choice but to mate amongst themselves, which in itself would pose no problems with birth defects because they still have a large variety of DNA" - assuming the puppies have DNA, incestuous breeding wouldn't be abated. They would have inherited their DNA from their parents, with a number of mutations thrown in.
• "However, after a number of generations, these dogs divided up into different families, and each family moved further and further apart from each other. After more and more generations, these families became different 'breeds' of dogs." - so, besides the wildcard of this "variety of DNA", it sounds like actual evolution, not an illusion. One ancestral species splits into several sub-species, the precursor to true speciation. If what you're describing isn't evolution, then what is it?
• "However, now, none of these breeds have anywhere near the amount of variety of DNA that the first ancestral dog had." - so, the ancestral puppies, who eventually sired the genetically varied dog breeds, were more genetically varied than their more genetically varied offspring? How does that work?

The key, I think, is in the phrase "variety of DNA". What does that mean?

 

[SkyWriting]

Weren't you just criticizing people for twisting the facts to make points in another thread? This is completely wrong, learn some genetics before you post stuff like this.

You are thinking of somebody else. Try citing your sources next time.

We know, as a matter of fact, from sequencing entire human genomes, that every human being has on average 50 to 100 mutations on its genome.

No we don't. Try citing your sources next time.

In other words, every child is 50 to 100 mutations away from the combination of their parents genes. And no, "vast numbers of the same mutation" are not required for it to show up. The only requirement is that the said mutation happens in an egg or sperm.

Try citing your sources next time.

Now, if you multiply that 50 to 100 by the number of humans in this planet, times the number of generations that you want to go back you will get a grasp of how much genetic variation there is in humans.

Try citing your sources next time.

If you still don't believe that genetics and mutation can produce enough variation for evolution to work on, just take a look at what happened to dogs in 10 thousand years.

Try citing your sources next time.

I consider natural background variation to be a separate thing from undesired, random mutations that are outside of design limits.

You can note from your dog example that 1000's of dog variations are possible to produce with a very tiny percentage of undesired mutations that die, are sterile, or even look or function odd. Your picture pretty much trashes the idea that most "mutations" are random. Most are planned. But a few slip through.

 

[bobsterInJapan]

I had a sneaky suspicion my analogy has some problems (sumimasen)

"Suppose the first ancestor of dog was created with all the DNA of the dogs included in that chart" - what does that mean? What does it mean to say the ancestral dog was created 'with' the DNA of the dogs on the chart?

I meant the ancestral dog was created having the entire gene pool for the dogs on the chart.

"However the first generation puppies," - who did the ancestral dog mate with? Are there now two? This isn't a major issue, but it is unclear.

Yes, let's say there are two, a male and a female.

"though having approx. the same amount of variety of DNA," - the same amount as what? Each puppy would have a distinct genetic code; are you saying the ancestral dog gave birth to, in the same litter, a genuine Chihuahua, Great Dane, Labrador, and all the other breeds? This doesn't match with what you later say about drifting families of dogs becoming the various breeds. So, are you saying that the puppies themselves have within them this variety of DNA - if so, what does "variety of DNA" mean? Is their DNA in some sort of quantum flux?

Good point.
I'm thinking of the ancestral female dog gave birth to, in the same litter, second generation dogs having DNA that included genuine Chihuahua, Great Dane, Labrador, and all the other breeds. For example, I have blond-hair genes from my father's side, but my hair is brown. My daughter, who is half Japanese, has very light-brown hair. So, I reasoning is that although I don't look as though I have blond hair genes, I do. In a similar manner, regardless of how the second generation dogs would look like, they (perhaps) could have DNA from which different breeds could stem from.

"would have no choice but to mate amongst themselves, which in itself would pose no problems with birth defects because they still have a large variety of DNA" - assuming the puppies have DNA, incestuous breeding wouldn't be abated. They would have inherited their DNA from their parents, with a number of mutations thrown in.

I was thinking that the ancestral male and female dogs would have a very large 'gene pool', so that the chance of defeats occurring from incestuous breeding would be greatly reduced. Is this fundamentally wrong?

"However, after a number of generations, these dogs divided up into different families, and each family moved further and further apart from each other. After more and more generations, these families became different 'breeds' of dogs." - so, besides the wildcard of this "variety of DNA", it sounds like actual evolution, not an illusion. One ancestral species splits into several sub-species, the precursor to true speciation. If what you're describing isn't evolution, then what is it?

Non-natural-selection evolution?

"However, now, none of these breeds have anywhere near the amount of variety of DNA that the first ancestral dog had." - so, the ancestral puppies, who eventually sired the genetically varied dog breeds, were more genetically varied than their more genetically varied offspring? How does that work?

I was thinking along the lines that each distinct breed would eventually have a smaller gene pool, a bit like the Amish people in America. Alternatively, if Europeans can be traced back to a black-African type of ancestor, why can't a European couple have a black baby? I think it's because they no longer have black-African genes.

 

[Wiccan_Child]

Yes, I meant the ancestral dog was created 'with' the DNA of the dogs on the chart.

OK, but what does that mean? It has the full genetic code of the Chihuaha alongside that of the Great Dane? How big would the dog be? Chihuahua-sized or Dane-sized?

Good point.
I'm thinking of the ancestral female dog gave birth to, in the same litter, second generation dogs having DNA that included genuine Chihuahua, Great Dane, Labrador, and all the other breeds. For example, I have blond-hair genes from my father's side, but my hair is brown. My daughter, who is half Japanese, has very light-brown hair. So, I reasoning is that although I don't look as though I have blond hair genes, I do. In a similar manner, regardless of how the second generation dogs would look like, they (perhaps) could have DNA from which different breeds could stem from.

These sorts of traits only code for a few alternatives - blonde or brown hair, for instance, or lobed or unlobed ears. Two white people won't be able to breed and produce someone who looks Asian.

So, the father and mother could both have two alleles for hair colour, the dominant brown and the recessive blonde, and thus both appear brown. 75% of their litter would be brown, and 25% would be blonde. But that's about as far as this 'hidden DNA' thing goes, and there's no guarantee it would weed out into pure-bred browns and pure-bred blondes - most likely these genetic alleles would continue to intermingle, absent some selection process.

I was thinking that the ancestral male and female dogs would have a very large 'gene pool', so that the chance of defeats occurring from incestuous breeding would be greatly reduced. Is this fundamentally wrong?

Fundamentally is the right word: the gene pool consists of all the breeding individuals of a population. Fewer breeders means a smaller gene pool. It doesn't mean that the individual has 'hidden DNA' that a given offspring will suddenly express.

Humans, for instance have a large chunk of DNA called junk DNA, and that basically comes from old bits of DNA we no longer use (and viral infections, defunct mutations, etc). It is tempting to think that this junk DNA can be switched on, and we get back our simian tails. What I think you're thinking of, is that someone could literally give birth to a monkey, if only the right genes were switched back on.

In the dog example, you could manufacture a highly contrived female dog whose eggs are sufficiently manipulated that each will yield a different breed. But, as I said, for this to appear like evolution, one would have to be specifically trying to create the illusion of evolution.

There are also problems with mainating a viable long-term population from just two individuals. Far as we can tell, outside of bacteria, it can't be done.

Non-natural-selection evolution?

Well, if there's some other selection process going on, it's not evident in the analogy - it sounds like this is a wholly natural process. Ironically, in reality, it wasn't natural selection, it was artificial selection.

I was thinking along the lines that each distinct breed would eventually have a smaller gene pool, a bit like the Amish people in America. Alternatively, if Europeans can be traced back to a black-African type of ancestor, why can't a European couple have a black baby? I think it's because they no longer have black-African genes.

Black skin is formed by how much melanin your skin makes, which goes back to genetics. Europeans don't have black babies because they don't have that gene. Not because the ancestral Africans had all human skin tones in their DNA, and modern 'races' inherited only one section - it's that the gene itself changed.

There is only a 'melanin' gene, which in Africans is highly expressed, and in Europeans is not so expressed. Africans don't have within their DNA the genetic code for white people, nor did our African ancestors.

And, even if they did, it would be a feat of genetic engineering designed solely that we 21[sup]st[/sup] century scientists would be fooled by the illusion of evolution. Quite the prank, eh?

 

[bobsterInJapan]

There are also problems with mainating a viable long-term population from just two individuals. Far as we can tell, outside of bacteria, it can't be done.

Why can bacteria maintain a viable long-term population from two 'individuals', whereas, e.g., two dogs cannot?

 

[sfs]

Black skin is formed by how much melanin your skin makes, which goes back to genetics. Europeans don't have black babies because they don't have that gene. Not because the ancestral Africans had all human skin tones in their DNA, and modern 'races' inherited only one section - it's that the gene itself changed.

There is only a 'melanin' gene, which in Africans is highly expressed, and in Europeans is not so expressed. Africans don't have within their DNA the genetic code for white people, nor did our African ancestors.

The gist here is correct, although the details are a bit more complicated. There are multiple genes that control the production of melanin, and many have them have had mutations under positive selection in human populations outside Africa. For example, a mutation in the gene SLC24A5 is present in all Europeans but not at all in sub-Saharan Africans. The European version clearly started as a single, fairly recent mutation, which spread rapidly because it was advantageous to have lighter skin at high latitudes. In Africa, however, the gene shows every sign of not having recently come from a single copy, but instead of being been around in a large population for a very long time. (Basically, the European version resides on a single DNA background that looks identical in everyone in the population, while the African versions reside on many different backgrounds, showing a long history of mutation and recombination.)

 

[bobsterInJapan]

..it was advantageous to have lighter skin at high latitudes

Why would it be advantageous to have lighter skin at high latitudes?