Evolution and randomness

[tdn]

Several times, on this board and elsewhere, I've heard that evolution is not a random process. When creationists say "Evolution is just random chance", evolution accepters say "It's chance, but it's not random."

Isn't this wrong? I was of the understanding that randomness is the very cornerstone of evolution. Within finite limits, of course.

Am I wrong about this?

 

[Donks]

The operators (mutation and crossover) that modify the solutions (the genes) are random. But the evaluation of those solutions is anything but random. A random evaluation would imply that poor solutions have the same chance of survival as good solutions. That's definately not the case, poor solutions tend to get eaten.

 

[tdn]

That's pretty much what I thought. My simplest evaluation of evolution is this:

1) Random mutation
2) Natural selection

So the random part is half of the equation. In fact, one could even argue that the natural selection bit is merely a byproduct of the mutation. In which case the random mutation would be the one and only aspect of evolution.

 

[Ashles]

So the random part is half of the equation. In fact, one could even argue that the natural selection bit is merely a byproduct of the mutation. In which case the random mutation would be the one and only aspect of evolution.

I don't quite understand your logic there.

Random mutation certainly isn't the only aspect of evolution otherwise, as Donks ponts out, every mutation would have an equal chance of survival. But we know this is not the case.
Adaption suitability, thus survival and reproduction are what selects the most successful genes.

To say that "random mutation would be the one and only aspect of evolution" misses the whole point of Natural Selection.

 

[roger]

Evolution would work even if the variation in the genes was not random. Genetic variation, combined with selective survival, yields evolution. It just so happens that this comes about via random variation, but it's not a cornerstone of the theory.

 

[tdn]

Random mutation certainly isn't the only aspect of evolution otherwise, as Donks ponts out, every mutation would have an equal chance of survival.

True, but the same can be said if there were no mutations (random or otherwise). Natural selection would have no basis on which to select.

The fact of natural selection is just the way the world works. That which survives, survives. Random mutation is the primary mechanism behind change in a species. Which in itself is no big deal, until put into the arena of natural selection.

I'm probably not explaining this very well.

 

[jan]

Evolution would work even if the variation in the genes was not random. Genetic variation, combined with selective survival, yields evolution. It just so happens that this comes about via random variation, but it's not a cornerstone of the theory.

If the number of parameters you can change is small enough, it might be feasible to test all possible slight changes of those parameters. For example, Dawkins in his "Blind Watchmaker" has some model organisms, his "biomorphs", that have only 9 "genes", each "gene" being an integer number. If you only increment or decrement one single gene at the time, there are only 18 possible mutant children, and you may check them all to see which one of them has the most of a certain quality (beauty in this case of artificial selection, simply fitness in the case of nature).

So one could imagine a fictional universe in which the whole process of evolution is perfectly deterministic and free from chance.

But of course, with so many parameters as the genome of a typical living being exhibit, the number of possibilities for "slight changes" is far too high to have them all realized.

 

[Ashles]

True, but the same can be said if there were no mutations (random or otherwise). Natural selection would have no basis on which to select.

Not true - it would still apply in exactly the same way, but in a population which is all equally suited to an environment there is nothing to be selected for or against. They are all identical.
This doesn't mean that they aren't still constantly subject to Selection, or the mechanisms by which it works.

For example if an creature injures itself it is immediately less suited to the environment and is being 'selected against'.

Or if there was no mutation and the environment changed in a way that was no longer suitable for the population they would likely all die. Without mutation they would not survive.
So Natural Selection would apply even without mutation.

It is about suitability to change as much as suitability to an environment.

The fact of natural selection is just the way the world works. That which survives, survives. Random mutation is the primary mechanism behind change in a species. Which in itself is no big deal, until put into the arena of natural selection.

I'm probably not explaining this very well.

The theory revolves around a combination of the two.

The mutations are random, but the suitability to an environment means that only certain mutations will be beneficial. So the environment is the non-random influencing factor.
It is non-random in that it is equally applied to all creatures - a constant by which to judge their random mutation.

 

[CBL4]

Evolution can easily occur without mutations.

For example, a climate in one area might favor bigger, furrier bears. The existing genes for bigness, sexual preferance for bigness, furriness and sexual preference for furriness would be favored by natural selection. Meanwhile somewhere else, the climate might favor smaller, hairless bears. These two groups of bears may become separate species without any mutation.

It is less spectacular evolution but it may be the kind that could be seen in Galapogos finches. Remember Darwin knew nothing about genes or mutation but he still got the big idea right.

CBL

 

[daenku32]

10 sided dice can RANDOMLY provide any number between 1 and 10. But it cannot give 11 or 0. Randomness within limits.
Am I right?

 

[tdn]

10 sided dice can RANDOMLY provide any number between 1 and 10. But it cannot give 11 or 0. Randomness within limits.
Am I right?

That's how I meant it. I think it's funny (and sad) how creationists insist that any mutation is harmful, as if a fish is suddenlygoing to grow a trunk or something. By within limits I mean slightly darker skin or braoder leaves or something like that.

 

[El_Spectre]

That's how I meant it. I think it's funny (and sad) how creationists insist that any mutation is harmful, as if a fish is suddenlygoing to grow a trunk or something. By within limits I mean slightly darker skin or braoder leaves or something like that.

Well, most mutations are have very little/no effect, or are harmful.

Only 1 in a million (or billion, I dunno) are helpful, but that's enough!