D O C U M E N T 4 5 5 M A R C H 1 9 2 9 4 0 5 455. From Chaim Herman Müntz [Berlin-Nikolassee,] 18 March 1929 Dear Professor, Here is a literal transcription of your manuscript and today’s extension[1] of it, for which I once again send many thanks It would make me very happy if I could keep the original: May I? The verification is in process [2] I am of course already convinced, but will have convinced myself soon on the basis of my own knowl- edge. Only the introduction appears to me to be unsuccessful in conveying the fac- tual situation.[3] Your anathema toward your own previous opinion would seem to me to be rather premature you most likely have nothing at all to regret in that re- spect even an error in your considerations would first have to be proved, I beg you! As you can see, I have been converted in this matter, not without reasons. Will you kindly examine the following countersuggestion “quite noncomittal- ly”:……… “The field equations provided in my earlier article*… are not yet convincing. Mr. L. and Mr. M.[4] have brought to my notice that the question of the admissibil- ity of the setting to zero of a tensor with N components with N unknowns cannot yet be seen as clarified in the literature but even in the case of its confirmation, the methods applied in loc. cit. according to Mr. M. leave open the possibility that every tensor divergence equation arbitrary fourfold equation could be taken as the electromagnetic equations, which would spoil the desired unambiguity even when conforming with the required adaptation to the Maxwell equations. Therefore, a method should be developed in the following, based on the same fun- damental ideas, but once again making use of Hamilton’s principle, that fulfills all the requirements of stringency and clarity that apply in the present case.” I will again gladly come to visit you personally in the near future! Yours truly, Müntz T / 0 =