This sort of thing makes me think that if I were engineering a biome, this sort of specialized spider would be exactly how I would solve the problem of a species that I accidentally made too prolific. The spider is a targeted fix: only preys on the ants, and if the ant problem ever resolves itself, the spider goes away.

I think this is a case where the theory of evolution feels pretty handy-wavy and doesn't answer the mechanism of how this happens. The article even talks about "bioengineered" silk! There are certainly lots of cases of evolution, so it is clear that things do evolve, but the problems of a theory are revealed in the edge cases, and I think this spider suggests that the theory really needs a lot of work to be a robust theory. I don't think a simple gradient descent (evolving toward more fitness) has explanatory power in this case. A theory that hand-waved around edge cases like this would never fly in Physics; at best it would be considered a holding theory until a better one could be found, like dark matter.

(Plus, I think it's fun to think about bioengineering in a sci-fi, or even Babylon 5 way)

Amazing specialization. I was wondering the same thing. Cave glow worms cast a "fishing line," and this is similar-ish. I wonder if N million years ago, a couple of fishing-line-like spiders started anchoring their lines, and the ones with a more conic shape anchor may led to more success over time. And the anchor may have only worked on territorial prey. Fun stuff to imagine.

Injured Himenoptera are known to send pheromones that trigger a vicious defensive response from other members of the colony. On a typical web the companion ants would do what the ants do. Go to war and flood the place surrounding the danger until eventually killing it. The spider does not have neither the stamina, nor the venom amount to deal with that. This web is designed to extract just one ant, while cutting the path that the ant rescuers could follow.

This is the first spider web known designed to catch only one species of prey. That alone would make the finding extraordinary. The trap can lure only green ants and serve the food exactly were the spider wants it; granting access to a common source of food that is everywhere, but also that is very dangerous to hunt (as much big as the spider, with powerful jaws, and much stronger).

The video shows one most interesting thing: Notice that the spider is carefully moving out of the way, just a second before the ant is launched. The spider knows in advance that its current location is about to be hit by a bungee jumping ant, and acts accordingly just in time to avoid the "bullet". We can easily imagine the spider thinking 5,4,3... This means that spider brains can predict the future outcome of a complex movement of objects in the physical system of its trap, and also calculate how much time the fibers will resist the jaw of the ant.

There are many such mysterious mysteries in evolution. Some wasps paralyse a caterpillar of a particular species by stinging it very precisely in its nervous centres, then carries the caterpillar into the nest it previously dug out, lay its egg on the caterpillar then close the hole to never come back.

This is a completely automatic, unintelligent behaviour; if you remove the paralysed caterpillar at any point in the process the wasp simply goes on with its business (it will close its nest without any caterpillar inside, where the larva will die out from lack of food).

In the late 19th century, Jean-Henri Fabre studied these wasps (and many other strange insects) and had a copious correspondence with Charles Darwin on this very matter. His books are absolutely fascinating (ditto the letters Darwin and him exchanged).