How Predictable Is Evolution? 209
sciencehabit (1205606) writes "If the clock rewound, would organisms evolve the same way they did before? Humble stick insects may hold the answer to that long-running question in biology. Through studies of these bugs, whose bodies match the leaves the insects live on, researchers have found that although groups of the bug have evolved similar appearances, they achieved that mostly via different changes in their DNA. 'I think it says that repeatability of evolution is very low,' says Andrew Hendry, an evolutionary biologist at McGill University in Montreal, Canada, who was not involved with the work."
Seems somewhat predictable ... (Score:5, Insightful)
The fact that a trait may be expressed by different DNA sequences doesn't really seem to undermine this. The DNA sequences are implementation details. Evolution is about solutions and environments not DNA sequences.
Re:Repeatable as Fuck (Score:4, Insightful)
The key question is whether the same results would come from different ends, again.
And the key evidence is that parallel evolution uses different changes from different genes to achieve the same end.
The question that I have to ask is, if different changes result in the same end, can the follow-on changes result? Or are they stopped?
Flippers turn into hands, but using different gene combinations - does that stop the thumb from differentiating? Or would evolutionary pressure still reward the mutant with the thumb?
I haven't read the whole thing, but I'm not swayed on any part of the question other than someone is now thinking about this. It is far from the foregone conclusion you think it is. In fact, in your statements, it stops at the interesting point. Will eyeballs that evolved differently be able to further evolve in similar ways? Or are they forever doomed, due to their makeup of different proteins, to be different? Or is it somewhere in the middle, which sounds plausible pending further research?
Environment shapes evolution (Score:5, Insightful)
Looking at cows, dolphins and horses genetic proximity [ucsc.edu] shows unexpected results, as cows and horses are not the closer in the trio, despite their similar features.
That suggests environment drives evolution in a predictable way, while the genetic evolution is not. This is the really amazing point: evolution find similar solutions to similar problems, but it does so through different ways.
Re:Repeatable as Fuck (Score:4, Insightful)
I think the answer is self-evident: alternate reality results would be just as diverse as species are today, and while they would bear superficially similar results, they would be "different animals." Commenters above have noted that vastly diverse organisms in a common environment still successfully evolve common features: they may have similar means of locomotion, means of food detection, means of sexual partner selection, and on and on, yet the specifics for any given species will be completely different from the other species.
Would the appearance of an opposable thumb on a flipper cause the lengthening of the appendage into something more useful, like an arm? Maybe, because arms are a useful advantage for food gathering; or maybe not because arms aren't as hydrodynamic as flippers. Or maybe there'd be a fork with two successful species resulting. I don't think the follow on changes would stop, they just would be different changes.
But as to the original article, why would anyone think that if we rewound the clock that a chaotic process would repeat? It's not like the universe called rand() with a common seed when it started mutating DNA.
Re:Seems somewhat predictable ... (Score:3, Insightful)
So both bloods were evolved to perform the same task of moving oxygen, however they use two different mechanisms to perform the task.
Re:What? (Score:4, Insightful)
no, a better example is - we handed the spec to a bunch of different developers...
They each gave us wildly different code to achieve the same goal.
If we give the same specs to other developers, we expect the same result. That the code will be wildly different each time for the same goal - so we can't predict what it will look like.
Re:Bah (Score:4, Insightful)
Re:Repeatable as Fuck (Score:5, Insightful)
It's not that simple. Something like a blind spot can't just be evolved away. There needs to be a pathway from "has blind spot" to "doesn't have blind spot" that doesn't go through "vastly decreased eyesight" along the way. Otherwise evolution will stick with what it has, and no amount of selection pressure can cause it to change.
We're vastly suboptimal in many ways. We're not perfectly tuned machines, we're cobbled-together from evolutionary scraps, and you can see it by looking at any part of our physiology. That's precisely the thing that makes intelligent design a stupid idea. Yet, we "work", and are capable of survival, and that's enough.