Still, toys remain on my list of science class essentials--I'd toss the textbook before I give up marbles, Jacob's ladders, drinky birds and the like.
I have a magic stone. Oblong and deep gray, smoothed by years of massage by a local stream, the stone nestles well in a worried palm. That by itself makes it special, but not magical.
One day, while idly spinning it, it awoke. It spun a few circles, shuddered to a stop, then reversed itself and spun in the opposite direction.
I spun it again. Again, it rattled to a stop, and spun the other way. Spin. Wobble and rattle. Counterspin. I found myself a rattleback.
Rattlebacks have a long history. Stones of similar shape have been found buried in Egyptian tombs (hey, a good toy helps pass away eternity), and people have played with them long before anyone understood Newtonian physics.
The rattleback is also known as a "wobblestone" (for obvious reasons), and as a "celt." As per the OED Celt comes from the Latin word celtes, meaning stone chisel, and is used to describe certain chopping tools used by prehistoric peoples.
I imagine a bored archaeologist spinning a celt as she sat in her tent during bad weather at a dig site. "Why, look, Professor James, this bloody implement is possessed!" If anyone has an idea of the real etymology, I would be much obliged.
You can find rattlebacks sold as scientific toys, sometimes at outrageous prices. Some rattleback merchants wrongly tout the toy as a model demonstrating the coriolis force. Even physicists have a time wrapping their minds around the toy's mechanics. Sir Hermann Bondi wrote: "Many people, even trained scientists, find it hard to understand that the behaviour of the toy doesn't violate the principle of conservation of angular momentum."
When I lend them to students, I try not to tell them what to do. Just let them play.