Most adaptations are what I'd call first-order. Cats evolve better eyesight; redwoods evolve to grow taller; male cardinals evolve attractive bright feathers. All of these changes were selected for because they directly make the organisms and their offspring more likely to reproduce. But it seems like there is a fundamentally different minority of adaptations, which I'll call second-order. Second-order adaptations do not make organisms and their offspring more fit directly, but rather they make an organisms' lineage more quickly responsive to selective pressures. Sexual reproduction is certainly the best example. The primary evolutionary benefit of sex doesn't come from increasing immediate fitness. In fact, I'd argue that many organisms' fitness is reduced by the fact that they cannot reproduce without a partner. Sex evolved not because it confers fitness to individuals, but because it confers adaptability to lineages: sexually reproducing lineages are quicker to evolve to fill new niches and avoid new threats. This is why I'd call sex a second-order adaptation, because it seems to be a rare adaptation which improves the process of evolution itself rather than the evolving organisms.

Is my understanding correct here, that sex is a fundamentally different kind of adaptation from, say, prehensile tails? If so, are there any other documented examples of second-order adaptations? Is there more accepted terminology than "first-order" vs "second-order" for this distinction?

Also, can anyone think of a third-order adaptation or pressure? Does that even make sense?

Possibly related: this post. But to be clear, I'm not asking about epigenetics.

  • $\begingroup$ When you say "fitness" do you mean individual or evolutionary? Because in the context of evolution it usually means the latter, which is defined by reproductive fitness. Evolutionarily, organismal aptitude is secondary to reproductive. That's why there's a often a distinction, or even discrepancy, between natural selection and sexual selection $\endgroup$ Commented Jun 18, 2021 at 18:57
  • $\begingroup$ @Punintended Even "reproductive fitness" defined as something like "number of offspring" or even "copies of gene in the next generation" is an operational definition of the overall theoretical fitness concept. It works as a good estimator in many common situations, but it's perfectly reasonable to think about it on a longer time scale over multiple generations. $\endgroup$
    – Bryan Krause
    Commented Jun 18, 2021 at 19:41
  • $\begingroup$ @BryanKrause Sure, but owls' better eyesight is an organismal fitness advantage that increases the probability / frequency with which that organism procreates. On longer timescales it absolutely makes sense to consider organismal fitness, but I'd call reproductive fitness the end goal because of creatures like salmon, which die shortly after procreating $\endgroup$ Commented Jun 18, 2021 at 19:48
  • $\begingroup$ @Punintended I mean reproductive fitness. Changing my list of first-order adaptations to include a clear product of sexual selection. $\endgroup$ Commented Jun 18, 2021 at 20:01

1 Answer 1


I think that you are talking about evolvability (as I see it).

Sex is indeed a sort of a meta-adaptation, see e.g. The Evolution of Sex by John Maynard Smith. The math for this is rather well developed at this point.

In terms of higher order effects like e.g. sex, I think that you are really talking about ideas like evolvability, and questions about constraints on paths over high-dimensional fitness landscapes. Examples of this might include the evolution of genome structure, and the idea that certain genome architectures are more conducive to generating novel variation. A specific example (though one could argue about it) would be V(D)J recombination's role in the adaptive immune system.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .