It also probably works both ways - the termites evolve to better detect parasites, so simple mimicry that worked thousands of years ago wouldn't work today. Similar to the relationship of AV and malware.
This might or might not be true. Birds are a good example - they are subject to parasitic mimicry.
One problem that a bird might have is that another bird lays one or more eggs in the first bird's nest. Theoretically, the victim bird will then incubate the egg and potentially feed and rear the nestling as if that nestling were the bird's own child.
There are several points in this process where the victim has the opportunity to defend itself:
- Ducks (at least some kinds) recognize the threat that another duck will lay an egg in their nest, and will fight approaching ducks to prevent them from doing this. But once the parasitic duck manages to get an egg into a victim's nest, the victim is not able to detect the foreign egg, and is stuck incubating and raising the adoptive duckling.
- Some birds learn to recognize their own eggs. If they come home and find a foreign egg in their nest, they will dump that egg onto the ground.
- Birds could theoretically be able to recognize nestlings of their own species. This would not help the ducks from the first example, who are victimized by their own kind. But it would help birds to defend themselves against cuckoos (and similar parasites), who lay their eggs in the nests of other species of bird. I am not aware that any species of bird has actually developed this defense, though I wouldn't be surprised if some have.
Anyway, what's relevant here is that all of these defenses have failed to develop in large numbers of birds that are subject to the relevant pressures. It's definitely possible that the same simple mimicry that worked 100,000 years ago works just as well today. But it's also possible that there's been an arms race.
Only about 40% of fairy wrens abandon cuckoo chicks, which suggests that this may be a relatively new development in these birds' evolutionary arms race.
My understanding of the usual expression of cuckoos is that they hatch alongside the genuine nestlings and are then given all of the food, leaving the genuine ones to starve.
If that's right, abandoning the whole nest seems like a waste - it should still have viable wren chicks in it?
Particularly since the mechanism you describe involves recognizing the cuckoo as an individual chick (assuming there are also wren chicks in the same nest, they will repeat the song, so it's necessary to keep track of which chicks have done so and which haven't), it seems like a better approach would be to just kill it. Do you know if any wrens do that or might be moving in that direction?
(On second thought, I see that the National Geographic piece appears to say that the cuckoo chick will kill the wren chicks? That's not quite clear, since it also says the cuckoo will hatch first, so I'd expect it to be shoving "eggs" out of the nest rather than "siblings".)
In any case, fairy wrens are very fast reproducers (raising 2-4 clutches per year), so I imagine they just start over with a new nest when this happens.
It can also works the other way round. Remove an albatross chick from its nest and the parent doesn't recognise them and the chick will starve https://www.dailymail.co.uk/news/article-7619547/Fans-Sir-Da...
Nature eh?
(Apologies for the Daily Mail link)
> Grown Albatrosses are unable to recognise their babies by sight and only know its [sic] them if they are settled in the nest.