Data or Dust Speck?

Shortly after the invention of the laser, a torrent of discoveries began pouring in thanks to the previously unreachable intensities that became available. Many of these discoveries fall under the general category of "nonlinear optics", which you could more or less say is the study of the behavior of light in a medium whose optical properties are themselves a function of the light intensity. Pretty much all material exhibit nonlinear optical effects if the light is intense enough, but "intense enough" is frequently in the neighborhood of 10^20 watts per square meter.

Once lasers were invented however, high intensities became available. One of the more important discoveries was second harmonic generation, which happens when light of frequency f is sent into a medium with nonlinear properties and light of frequency 2f is generated. Most green laser pointers work this way, frequency-doubling infrared light of 1064 nm wavelength into visible green light of 532 nm wavelength.

This second harmonic generation was first reported in Phys. Rev. Lett. 7, 118-119 (1961) by researchers at the University of Michigan with pulsed ruby laser light of around 3 kilowatts instantaneous power focused into a very tiny area. The paper they published is now unfortunately not just a famous paper in laser physics but a famous publishing screw-up in laser physics:

The big glob is the main beam on the photographic plate. The arrow points to a much smaller spot on the photographic plate from the part of the beam that's been converted to its second harmonic. Well, it did until someone in the printing process (this was before the online PDF era, after all) mistook the signal for a speck of dirt and erased it from the image. So this paper went to press with the actual experimental data completely erased!

Well, the paper was right anyway and today second harmonic generation isn't merely of pure science interest but is pretty much ubiquitous in commercial laser applications. It's a cute story anyway.