Intersting. I would have guessed that in the tropics natural selection would have FAVORED darkly pigmented individuals. I had NO idea that a theory could be imbued with the power of creation.
It's an example of an amusing argument, though .... the explanation supplied for some particular evolutionary mechanism -- in this case, for the evolution of dark (or light) skin -- almost invariably relies on some ex post facto application of a human design. It's that old fallacy, post hoc, ergo propter hoc, dressed up in a fancy lab coat.
Even if we grant that evolution somehow explains it, the truth is almost certain to be much more complicated.
A different example that shows the danger: it's been observed that domesticated animals often have coats with large spots. And it turns out that there is some evidence that these spots are in some way correlated to selective breeding. But the correlation turns out to be for selective breeding for docility, not coat patterns.
So, sure, one could come up with a dandy explanation for why skin is dark.... but the spotted cow suggests that there's probably a different explanation than the easy, human-designed one.