It puzzles me. In Einstein's paper on the photoelectric effect he proposed that photons with [tex] E = nhf [/tex] were the explanation. Wouldn't a more elegant explanation be that the tangent of the electromagentic wave must take on discrete values because of the boundary conditions between the emitter and the metal plate? This seems to give the same results, the acceleration and therefor the energy of the wave would be in quanta [tex] E = nhf [/tex]. I don't see a need to use the particle theory of light to describe this experiment in it's original form.