So, after weighing the pros and cons, I decided to accept certain risks and take a hybrid approach. That's a test tube inside of another test tube with a scrumdiddlyumptious fly in the one. So far, the eggs/larvae/pupae remain.
So I thought I'd give an update and report that this "hybrid" approach seems to be working quite well for S. molesta. It seems to result in 100% humidity, but the ants are actually preferring this setup over a vacant neighboring test tube setup. The colony is probably around 70 workers now, and the pile of brood is massive.
