Check out this posting over at Radiomuseum:
http://www.radiomuseum.org/forum/how_to ... kings.html .
I find this quite interesting, and will definitely give it a try next time I have a tube that I cannot read the number on or identify easily by other means.
I'm curious from a scientific point of view how this works. It must somehow relate to the wavelength of light produced by an LED flashlight (oops, I mean "torch"

), but it is not clear to me why there would be some residual markings more easily detectable with a particular wavelength when the main marking is rubbed off. Anyone know why this works?