Here is a passage from The Selfish Gene

If selection goes on between groups within species, and between species, why should it not also go on between larger groupings? Species are grouped into genera, genera into orders and orders into classes. Lions and antelopes are both members of class Mammalia, as are we. Should we then not expect lions to refrain from killing antelopes, 'for the good of the mammals'? Surely they should hunt birds or reptiles instead, in order to prevent the extinction of the class. But then, what of the need to perpetuate the whole phylum of vertebrates?

So shouldn't it eat the other animals? Also in the next line, it says about vertebrates that since reptiles and birds are vertebrates, the lion should go for invertebrates.

It sounds illogical, but what stops lion from eating his "relatives", or from sparing a deer? And till what extent is the term "relatives" valid?

Is the question is too broad, please give a source where I can get an answer.

  • $\begingroup$ It would be helpful if you could cite the actual passage. It is a bit unclear when reading your post whether the subject is lineage selection or group selection (aka kin selection). $\endgroup$ – Remi.b Mar 28 '17 at 18:43
  • 1
    $\begingroup$ It feels to me that the title of the post ask another question from the content. "What level applies" is different from "How does selection at this level make sense". Generally speaking evolutionary biologists assume that selection at level higher than variance in individual's fitness within population is often negligible but I don't think there is much data and much of a good argumentation for the extent of the validity of such assumption. $\endgroup$ – Remi.b Mar 28 '17 at 18:46
  • $\begingroup$ @Remi.b so individual fitness and that of close ones determined by Hamilton's equation. And that would also explain the lion example as they are too distantly related to outweigh the advantage. $\endgroup$ – YAHB Mar 28 '17 at 18:54
  • $\begingroup$ I did not say that. I said I would need a clear quotation from the book to be able comment on it. $\endgroup$ – Remi.b Mar 28 '17 at 19:01
  • 2
    $\begingroup$ I haven't read the book, but these questions seem rhetorical. It seems like Dawkins is setting the stage for an answer, which I suspect he gives later. $\endgroup$ – canadianer Mar 28 '17 at 20:25

Natural selection favors genotypes that produce phenotypes that survive to reproduce. To the extent that natural selection drives the instincts that drive predation, the (essentially tautological) answer is that, "It has proven selectively beneficial for that particular genotype to eat whatever it's eating."

Indeed, nature has shown us a variety of solutions to these selective pressures. Cannibalism is not hard to find. In the extreme case, mothers from diverse species are known to kill and eat their own offspring, which at first glance may seem like the height of failure when it comes to propagating one's genes. And yet we can hypothesize a number of selective advantages of that behavior. (E.g., the mother can produce large litters when food is plentiful, and then adjust for shortages to shore up a smaller litter when food is scarce.)

We also know that there are selective pressures against eating similar animals. For example, pathogens that kill an animal are likely to afflict its genetic relatives, and so that could drive the aversion of some carnivores (including humans) to eating close relatives.

There are plenty of examples of evolved symbiotic organisms. Alligators don't eat plovers, and sharks don't eat remoras.

But it is hard to imagine selective pressure for altruistic behaviors that don't benefit a particular genotype. E.g., if there are mammals to be eaten, the lions that eat them will survive and the ones that don't will die. If the entire class Mammalia is under pressure, population dynamics trump evolved behavior: If prey become scarce, predators will become scarce. But if any predators survive, it will be the greedy ones.


Not the answer you're looking for? Browse other questions tagged or ask your own question.