D O C U M E N T 2 3 2 J U N E 1 9 2 2 1 9 3 important confirmation of quantum theory was caused, on one hand, by the great experimental difficulties (measurement of the absorbed ultraviolet radiation energy and the minuscule converted amounts attaining the required gas purity), on the other hand, by the theoretical interpretation of the experimental finding. These lines can only give a faint notion of the life work of such a multifaceted researcher. But perhaps it will inspire some fellow members of the profession to delve into one or the other of his original papers, the following itemization of which will perhaps be more welcome than the scant allusions given thus far about the con- tent of a little part of it. Scientific Articles by Emil Warburg[5] [. . . See the documentary edition] 232. To Aurel Stodola Berlin, 13 June 1922 Esteemed Professor Stodola, It pleased me exceedingly that you have taken an interest in my paper of 1917,[1] as it relates to my special hobby that is currently finding little favor among physi- cists. They tend determinedly to uphold electrodynamics in a vacuum, which in my opinion leads to contradictions.[2] I do have to admit, though, that the more precise mechanism behind the elementary processes has remained completely obscure to me. I imagine the production of heat by the absorption of visible or ultraviolet radi- ation in two ways: 1) An isolated electric elementary particle is unable to absorb because it has as its precondition an infraction of the law of conservation of momentum. Two collid- ing electric elementary particles certainly emit radiation at a loss of mechanical energy. So they also have to be able to absorb radiation at a gain of mechanical energy, i.e., the production of heat. 2) A molecule or atom can absorb in the Bohr manner (through electrons). If such a molecule collides with another, this energy can transform during the colli- sion into kinetic energy (heat). This is related, e.g., to the fact that at rising temper- atures fluorescence tends to diminish at the cost of heat production. A quantitative execution of these things is possible but presently not attractive insofar as it requires quite a few detailed hypotheses. [p. 826]