[thaumasiotes]

What happens by chance in the first generation will happen by design in later generations. Unless the altered mosquitoes differ from normal ones in a new and unpredictable way every generation from now to the end of time, they will eventually be detected and avoided.

[delecti]

They didn't do anything to the existing mosquitos. They released sterile specimens. All of the fertile wild specimens were still going about their business making more Mosquitos, but the matings between the normal wild ones and the sterile lab ones didn't result in offspring.

If you roll a collection of dice, and remove all of them when they roll something other than 6, the other dice haven't "evolved" to only roll 6s.

[kd5bjo]

No, but dice that roll sixes more often are contributing more offspring to the next generation. If the selection pressure keeps up, you will have most dice consistently rolling 6s.

[maxerickson]

The technique has been used against other species of insects since ~1950.

There should be a horror story by now.

[thaumasiotes]

Horror story? The "horror story" would be "the insects are still around".

[maxerickson]

Indeed, I was being flip. It's actually been successful at eradicating populations of insects, rather rapidly.

https://en.wikipedia.org/wiki/List_of_sterile_insect_techniq...

[thaumasiotes]

I see that "eradication" has been declared in most of those trials, but I also see that formal "eradication" is often followed by renewal of the trial, recovery of the population, or "complete eradication". I'm forced to conclude that "eradication" in this context does not have its usual meaning.

That said, it does look like this is usually a pretty successful method of ongoing population suppression.

[delecti]

There's no reason to think there's any difference between the wild mosquitos who bred with fertile wild specimens and the wild mosquitos that bred with sterile lab specimens. If there's no difference, then there's no "selection" going on.

If you want to argue that there is a difference, then you'll have to provide some evidence of that. Currently you're asserting it as a premise.

[JoshCole]

A list of differences:

- Birth environment.

- Coordinates of release into the wild environment.

These two differences alone are already enough to represent a selection pressure. A mosquito pair which is fortunate enough to have a statistical deviance in its position relative to a released population of sterile mosquitos will tend to reproduce more often than others. We absolutely expect that a pressure will exist. We absolutely expect that a selection will take place.

Assuming that the selected feature will be one which corresponds with detecting a difference between sterile and non-sterile mosquitos might be a stretch though. There is an enormous space of potential features which might be selected for. For example, mating multiple times to increase the chance of being with a fertile mate is another possible selection. Another is flying away from regions with large numbers of mosquitos which are appear to be newly introduced to the environment. Another is that mosquitos which exist in hazardous environments which usually kill mosquito possess unique adaptions which may prevent the sterile mosquitos from hindering their reproduction. The space of possibility is too large to enumerate.