Perhaps this is nothing new, but it just struck me that there are two very different ways one might try to ground ethics upon biological evolution. They differ in whether evolution is considered to have substantive or methodological significance. That is, evolution might come in at the level of normative ethics, or else meta-ethics.
The 'substantive' approach takes the evolutionary telos to underpin ethical normativity. It makes the normative-ethical claim that we ought to share and promote Mother Nature's "purposes". We might further distinguish 'forward-looking' and 'backward-looking' versions of this thesis. On the former, we ought to do whatever will promote the survival of our species, or our own genes, or something like that. On the latter, we ought to behave in ways that have previously been selected for by natural selection. Either variant seems terribly misguided, since there's no reason to think that evolutionary functions have any moral significance, or that humans are subject to such arbitrary 'externally imposed purposes'. (Though of course these goals might happily correlate with genuine values; survival of the species clearly bears some connection to human wellbeing, for example. They just aren't foundational to ethics, is all.)
Further, note that there's nothing particularly scientific about this view. There's nothing in evolutionary theory which says what humans morally ought to do. Rather, this view is a substantive philosophical thesis which (seemingly arbitrarily) claims that the moral good is tied to evolution in the ways described above. It's not good science, and it certainly isn't good ethics.
How about the 'methodological' approach? Rather than trying to build evolutionary considerations into the content of our moral theories, this meta-ethical approach remains neutral on the content and instead makes higher-order claims about the nature of ethical discourse itself. In particular, it suggests that, to gain insights into our ethical practices, we should look to their evolutionary origin. Suppose our moral behaviour evolved for the (biological) purpose of creating more stable societies and thus boosting our biological fitness. This explanation makes no mention of moral facts, and so might be used as the basis for a nihilistic 'error theory'. (I suppose this is not so much grounding ethics upon evolution, as using evolution to argue that ethics is groundless.) [Update: see comments for another, non-nihilistic, example.]
While attempting to build evolution into normative ethics is transparently stupid, its methodological use is much more interesting. I'm not quite sure what to make of it, actually. I'll explore the underlying ideas some more in my next post.