Make them live longer by artificial selection? You might need something with shorter lifespans…

So think about what you’re doing.

Is it comparable to training someone to swim or dive for longer periods or are you breeding for better swimmers/divers?

One of them is genetic, the other one is physiological.

Study design is fun. As you said at the beginning of your post, this will probably take much longer than the timeframe you’re willing to devote. And… there’s a chance it may never happen even if you spend a thousand years on it.

The issue is really quite complex. There are an intense amount of variables to be considered. I would suggest you do some intensive research into pillbug morphology, genetics, life-history, and evolution.

It looks like you have some knowledge of evolutionary processes but you might want to make sure you concretely understand them before undertaking this, or even writing up a procedure. As mith mentioned, you need to distinguish between behavioral changes and genetic changes. Even if you can breed a pillbug that LOVES living in the water 24/7, its biology might not allow that. I have my doubts as to whether the pillbugs can really retro-adapt back to aquatic environments.

So, my advice is research. research research research.

Aside from that… just my personal brainstorm on the matter (without research) would be:
You would need a pretty large population with plenty of genetic variability to start with. So if what you have right now is one hatch from a single set of parents, you might not even have enough diversity to accomplish anything.
Get a huge population, and make the conditions such that slightly (or more than slightly) more individuals die than are born to begin with. If the population gets below a certain level, and you don’t notice any individuals or families competing better than others, you might need to add more pillbugs. keep food levels constant at their designated locations though. To make such high mortality rates, you might introduce a predator on the land portion..? The idea behind this is that you’re trying to select for an attribute that doesn’t necessarily exist, genetically, in these bugs. So, by increasing your population, you’re increasing the chance that (if it does exist) one of your individuals will posess a desired genetic trait, or that a favorable mutation or allele expression will occur. Still, it’s an evolutionary crap-shoot imho ;D.

As far as distributing the food, your idea of having little poor-quality food on land, and more high-quality food in the water isn’t bad. Perhaps make gradations within the water, having better food further into the water so that individuals who can forage further into the water eat better. But, if you’re just using a small aquarium for this, the distance might be negligible anyway.

Again, I’d suggest considering adding a predator on the land, or at least a competitor (but it would have to be something that can’t live or function in the water).

You’re going to need to consider the ecology of your tank as well. How are nutrients going to be redistributed, or are they? Are you going to clean the tank occasionally? Are there other neutral species of bug, plant, or plankton you could use as a buffer to toxins etc?

ENDING THOUGHT: Look up the difference between acclimation and acclimatization. I think what you want from your study is acclimatization, but you’re trying to set up a system that would acclimate.

I dunno.. anyway, have fun.

Of course bugs can learn. They aren’t just hard-wired to do one thing, or they’d never survive. Now they might not be doing calculus in their head… 🙂

If your pillbugs could not modify their behavior, then you’d have to just dump a bunch of them in the water until a few survived. That would be the only way. Otherwise, they would never learn to venture into the water, or stay in the water for any length of time.

The point where the behavioral changes and genetic changes seperates is here:
The pillbugs that behaviorally adapt to spend more time in the water might not be physiologically (genetically) adapted to spend as much time in the water as they like. The ones that are better adapted to water (better genes) will compete better and survive longer, having more children.

so the behavioral modifications will lend themselves to the eventual (if it works) genetic shift. OR they’ll all just die 😀

Think about it this way
A group of humans one day decided they were going to become aquatic mammals and developed a colony at the waters edge. Every day they would spend 99.9% of their time in the water. They learned how to hold their breath underwater for 3 minutes! They could swim almost as fast as any fish. They could even catch food with their bare hands, enough food to survive solely in the water.

Just because they learned to succeed in the water, doesn’t mean they’ve adapted genetically. Even if they have children and their children learn to live in the water, like them, that still isn’t an indication of genetic change. Get my drift?

Of course this is a somewhat unlikely scenario because, as far as I know, they’d die off pretty quickly if they spent 99% of their time in the water, what with temperature, exposure, osmotic imablances, etc.

One fast-and-sloppy approach is to take all of your specimens, submerge them, and when some fraction has drowned, take the remainder as your breeding pool. Breed and repeat. You may not wind up with a population that WANTS to be underwater, but you will be selecting for greater and greater tolerance of it.

Your water resistant strain would have to exist in those 400 to begin with for selection to work. For the bugs to gain any new genetic material it would a long long long time. Maybe it’s time you think about actually injecting foreign DNA not the larvae of the bugs and see if something or anything happens.

A true demonstration of their underwater abilities due to evolution would be if they somehow decide to live under water for generations and can no longer walk on land. Again, this would be a demonstration to convince others not necessarily has to be this way.