D O C U M E N T 7 7 M A R C H 1 9 2 2 9 5 The idea that conduction chains can be destroyed by magnetic fields is likely, indeed almost necessary, as is the fact that a sufficiently strong thermal motion would destroy conduction chains as would sufficiently large -energy quanta of which they are composed. Thus, the transformation of a superconductor into a nor- mal conductor by a temperature increase, and perhaps even the superconductor’s sharp temperature limit, may become comprehensible. Electrical conductivity at normal temperature may be based on the incessant thermal formation and destruc- tion of conduction chains. This conjecturing can only be excused by the momentary predicament of the the- ory. It is clear that new avenues have to be sought in order to be able to do justice to the facts of superconductivity. It appears probable, but not certain, that conduc- tivity at normal temperatures is based on superconductivity constantly being dis- turbed by thermal motion. This idea is backed by the consideration that the frequency of the electrons’ tran- sition to the neighboring atom may be closely related to the orbital frequency of electrons in the isolated atom. One thus arrives at the suspicion that the elementary currents of the individual conduction chains could be of substantial size. If this idea of the quantum dependence of elementary currents were proved valid, it would make sense that such chains could never contain atoms of various kinds. P.S. The suppositions suggested at the end, which incidentally do not raise any claim to novelty,(1) are partially disproved by an important experiment that Kamer- lingh Onnes performed over the last months ago. He showed that at the point of contact between two different superconductors (lead and tin) no measurable Ohmic resistance occurs.[22] (1) Comp., e.g., F. Haber, Sitz. ber. Ak. Berlin, 1919, pg. 506.[21] 77. From Paul Ehrenfest [Leyden,] 11 March 1922 Dear Einstein, Thanks very much for the card.[1] 1.° Please inform me when you want to be here.— As orientation: 27 April, first phys[ics] colloqu[ium] again after Easter. Circa 1 May, Onnes back in Leyden. When Lorentz will be back again I don’t know but probably unlikely before circa 5 May.[2] – hν [p. 435]