I've long wondered if it would be possible to hear the speech of ancient peoples by scanning sequential imperfections in crystal formations of fire-side stones as they cooled down and solidified (for the last time before the people moved on), assuming that ambient sound waves would be strong enough to create detectable shifts in the molecular positions and assuming that the formation would otherwise be precisely linear.
It probably sounds like a crazy idea on first inspection, but I think there's something in this.
So, are the crystals that are formed as stone cools down always in a precise formation regardless, or could their positions (on the picometre scale) be shifted slightly off linear by external vibrations and then stay that way?
As heat leaves the stone, I assume there's an imaginary line travelling inwards marking the point where the crystalline structure goes from malleable to fixed. If this line takes 10 minutes to traverse a 100mm stone (I think for the US that would be a rock), and assuming a crystal size of 500pm (0.5nm), you'd have an effective sampling frequency of over 300kHz, if the technology could be developed to measure the positions of crystals.
I hope there's an answer to this other than "you're mad!"