This has already been boingboing'd but I liked it so much I joined the sheepish bb-followers: Dylan Stiles at Stanford was bored and had spare runtime on the lab's NMR machine. What to do? Obvious, really: run your earwax through it and check that Wikipedia is correct about its composition. It was. Then you could confirm it with TLC. He did. One commenter rebuked him for not co-spotting, which is entirely correct and not at all in the spirit of the thing.