Faith writes:
In fact, it seems to me that the controlled forced speeded-up conditions of domestic breeding could prove macroevolution if it really occurs, but in fact what is observed to happen is the reverse of anything in the direction of macroevolution. That is, the more you select, the less genetic potential you have for further breeding, as I've pointed out many times before
The problem with artificial selection is that it has a tendency to reduce genetic variation to the extreme. For example, there was a horse in the 18th Century, called
Eclipse. It is estimated that 80% of all thoroughbreeds have Eclipse in their pedigree. That seems indicative of a pretty small gene pool. I couldn't pretend to be an expert in animal husbandry, but my intuition tells me we wouldn't expect to see much in the way of large scale changes in that kind of lineage.
With natural selection, the selective pressures are wide and varied, with a large amount of possible solutions, both morphologically and genetically. With a large population and fecundity it could be expected that over time many different attempts to solve the problem of reproduction in a dangerous world will be tried. Some will fail, others won't. This is where macroevolutionary changes are going to occur - when the selection pressure acts much more generally all the way down to the genetic level.
Breeding doesn't do this so much. We can shape the animal (and as above, in a sense this is natural selection at work of a sort), but we select out far more aggressively than nature, and pick mates for the animal trying to steer evolution's course. This is more like taking a bush and chopping it so that it is very long and narrow. It might get taller, but it is not going to widen much.
Dawkins invites us to imagine a multidimensional genespace full of every animal that could ever be. Normal evolution takes small steps through the gene space and if it ever reaches a point where an animals immediate neighbours are not very fit it is likely that the tendency would be to 'retrace' a couple of steps through the gene space and try another direction.
With artificial selection we choose who mates with who and how often (sort of). This is kind of like taking slightly bigger steps through the gene space, without sending exploring fingers down other avenues (ie less genetic diversity). Is it any wonder that a breed might find themselves at a dead end quicker than a species might had it been able to wander down the paths of least resistance?
It is
possible that artificial selection based on the desires of men will arrive at a macroevolutionary solution, but given the restrictive nature of the selection process being used, I don't think it is particularly likely. It might be that in the background, the more sluggish natural selection might steer things generally towards a more macroevolutionary solution to keeping humans company/working with humans/feeding humans etc., and in some way that macroevolutionary change was influenced by artificial selection. Short term artificial selection is unlikely to arrive at such impressive solutions.
™ Version 4.2
