D O C U M E N T 3 1 5 S T E R N - G E R L A C H E X P E R I M E N T 2 4 5 Thus these are times of an order of magnitude that at least do not come into con- sideration at all for the experiment because there the adjustment must take place in a time of less than sec.[9] § 4. If one tries to find a way out of this dilemma, two alternative assumptions initially present themselves: A. The real mechanism is such that atoms never can get into a state in which they are not fully quantized. B. For rapid influx, states result that violate the quantum rule regarding orienta- tion the adjustments demanded by the quantum rule are produced by emission and absorption with a reaction speed that is extraordinarily much greater than for tran- sitions from quantum state to quantum state. A decision a priori between these two alternatives does not currently appear to be possible yet it is appropriate to take a clear look at their principal difference and the characteristic problems that each of them leads to. § 5. Discussion of Alternative A. 1. What it requires is particularly well exempli- fied by the Stern-Gerlach experiment: In the vaporizing chamber of the small melt- ing furnace, each silver atom is fully quantized after each collision, so its magnetic axis is oriented according to the magnetic field prevailing there, weak though it may be. After its last collision during its flight through the various parts of the field, its orientation constantly conforms to the field direction of the location con- cerned.(1) 2. In the process, one portion of the (one-quantum) moments will be set parallel, another antiparallel to the field, where the statistical distribution will be dominated by the temperature and the field strength in the small furnace’s vaporizing chamber and certainly not by the (radiation) temperature and field strength in the space that continues to be traversed! 3. One would accordingly have to decide to assume the following: Even weak fields must be decisive to the orientation immediately after the collision (i.e., the effect of very strong fields). In changes, e.g., of the direction of the magnetic field, which are arbitrarily quick compared to the Larmor rotation, the atom’s magnetic axis should therefore follow the direction of the field as fully as for arbitrarily slow changes. More generally: For an arbitrarily rapid change in the external conditions of a mechanical system, this system would have to adjust itself to the same final state as for an infinitely slow (adiabatic) execution of the change in the external (1) Dr. G. Breit already suggested such an assumption on the occasion of a discussion at the physics colloquium at Leyden.[11] 10 4– [p. 33] [p. 33]