L E C T U R E O N L I G H T E M I S S I O N 8 5 1 TRANSLATION On Light Emission Professor Einstein started with a brief reminder of the classical view on light emission by oscillating electrons. An oscillating electron undergoes radiation damping, as a result of which the process of light emission practically ends in seconds. Because of this, waves can be made to interfere even with large path differences, and in this way classical theory explains these phenomena satisfactorily. Quantum theory, however, supposes that radiation only takes place when an electron in the atom jumps from a level with energy to another stable orbit with energy , where- by the frequency ν of the emitted light is given by where h is Planck’s constant. Now what does quantum theory say about the duration of the emission? The lecturer makes a distinction between the duration of the state Z and the duration of the transition U. The duration of the state is the time during which the electron remains moving in a Bohr orbit with higher energy, whereas the duration of the transition is the time in which the tran- sition takes place from a stable orbit into another one with smaller energy. Experiments by W. Wien on the attenuation of light emission by canal rays moving on their trajectories in vacuum show that Z + U is of the order of magnitude predicted by classical theory. It cannot be decided from these experiments, however, how large the ratio Z/U is. For various reasons, such as the experiments by Franck and Hertz, it seems likely that U is many times smaller than Z. It would moreover be very desirable to find out some- thing about the magnitude of the absolute duration U. The lecturer described an experiment that at first sight appears to be able to clarify this question. One makes a bundle of canal rays move with 1/300 of the speed of light past a slit that is placed in the focal plane of a lens. If the emission time is not extraordinarily short, behind the lens one obtains a fanlike bundle of planes of equal phase. If the waves are also made to go through a dispersive medium, one may expect that the light ray does not follow a straight path in it, and that the deviation of this straight path becomes larger as the light ray has to cover a longer distance in the dispersive medium. This would, in the end, lead to a displacement of an image of the slit created by a second lens. But if one looks more close- ly at the theory of this phenomenon, it becomes clear that the alleged displacement will not take place. In fact, the displacement could not be observed experimentally. The lecturer, however, now feels that he can outline an experiment that might provide definitive information on the duration of the light emission. One once again makes a bundle of canal rays move in the focal plane of a lens. Behind the lens there is a pair of semi- silvercoated plates of glass, separated by a certain distance. In the case of a stationary light source one would now observe interference rings in the focal plane of a second lens. If, on 10 8– E2 E1 E2 E1 – hν, =