“In 2003, for example, a team of scientists studied wings on stick insects. They found that the insects’ common ancestor had wings, but some of its descendants lost them. Later, some of those flightless insects evolved wings again.” He claims they can never get the same wings as before.
Basically, he says that 1 )random changes piled up to make something useful and 2) natural selection cannot be expected to undo random changes in reverse order because it’s not improving anything.
And my answer to that is simple. He admits that single changes can be lost (revert) randomly. And single random changes don’t have to be useful to continue, just not be fatal. So, you don’t need natural selection, but just random chance to reach the prior state. As long as the intermediate conditions are not fatal, natural selection does not act against them. And we know they are not fatal because that’s the path taken to arrive at the current configuration. As to repeating the same progression to wings, why not? We know it’s possible.
Anything that can happen once can possibly happen more than once, but the end result of any evolutionary process is such a vast and complex combination of interdependent variables that it is highly unlikely (although not actually impossible) for the exact same thing to happen more than once.
I’m no expert by any means, but don’t the recent chicken embryo experiments (changing one protein can cause chicken embryos to grow teeth, or long tails) show that some complex instructions are still packaged up and ready-to-go, but sort of ‘commented out,’ as it were?
Then couldn’t a reversal happen in just one generation, and become widespread if it were advantageous again?
I think the point he’s trying to show is evolution doesn’t do what’s best, it does what works.
And if evolution makes a mistake it doesn’t back up and start over, it rather does a work around.
The best example of this is the throat. Obviously having the windpipe cross the food pipe and allowing you to choke is a HUGE design flaw. But evolution evolved the “cough reflex” rather than backing up and having the windpipe not cross the food passage.
Yes, that can happen, and ‘commenting out’ is a remarkably apt analogy for it, but I get the impression that when people talk about things being lost, then evolving again, they mean something a bit more complex and than a few things getting switched off, but remaining otherwise intact.
It’s a bit of a semantic thing really, because technically, it may be all evolution either way, but the semantic meaning is the important one here, I think.
The insect wings weren’t the focus of the study. That was an example of a possible refutation of Dollo’s Law (which it may or may not be, depending on the particular evolutionary pathway for the second emergence of wings).
Rather, the study focused on a particular protein, and found that “restrictive mutations” essentially constrain the nature of future viable mutations. Simply undoing the steps of the useful mutations isn’t enough to get back where we started.
As mentioned in the article:
Which really isn’t anything particularly profound, and, indeed, seems rather like common sense: the more difficult something is to do, the more difficult it is to undo it. Because a good many traits are complex, “reverse evolution” isn’t really possible. Of course, that doesn’t mean that a new pathway can’t be found to produce the essentially the same effect.
To put it succinctly, it’s not “survival of the fittest”, but “survival of the adequate”. If you manage to reproduce, then no need to be fittest. The runts of the litter, male and female, can find each other and bring down the average but at the same time widen the gene pool.
It seems like the author is talking about evolution at the molecular level. A wingless stick insect can evolve wings and those wings can subsequently be evolved away. But the new wingless stick insect doesn’t have the same DNA as the old one and is not the same creature.
