Weirdly enough, when it comes to selenium meters I've noticed trends that go more by specific meter than how the meter was stored. I don't think I've seen a Weston III or earlier that didn't work and wasn't accurate. At the same time, GE made a reasonably common handheld meter that always, to my eye, has always looked a bit like an electric razor. I have a few of those lying around-one of them will twitch a bit if I take it out in the sun and open the light shield, but the others are completely dead. I've handled one that worked and was accurate-it was one that someone brought into a local camera store along with a bunch of other stuff to sell when I happened to be there killing time(or at least it gave the expected light reading when pointed at one particular fluorescent light fixture in the store that is used so much for testing cameras/meters that anyone who works or hangs out in the shop can tell you what it should read). Selenium meters built into cameras are almost always suspect in my experience, but I got a pleasant surprise when I bought a Canonflex RM and found that it was good. Not much later, I bought a Canon 7, which seemingly has the same meter mechanism, and it's also good. My general experience with selenium has also been that if they give what I'd call a reasonable response(i.e. take them into full sun with the baffle-if present-closed and they deflect a good bit, or deflect in a reasonably bright room with the baffle open) they are PROBABLY reasonably accurate.