6 5 0 D O C . 4 5 2 H A M I L T O N ’ S P R I N C I P L E ADfX. [18 322]. The document consists of seven pages. Page numbers presented here in the margin in square brackets depart from those in the original, where they appear in the top right corner of the page. [1] This is a draft of a paper written as a follow up or correction of Einstein 1929n (Doc. 365). An abortive draft with a similar aim but different approach is Doc. 368 see also the fragment Doc. 367. Therefore the date of this document is no earlier than the date of presentation of Doc. 365, which is 10 January 1929. It is likely that Doc. 455 from Herman Müntz suggests changes to the first paragraph of this draft, which would place the date of this document before the date of that letter: 18 March 1929. [2] At this point in the text, Einstein indicates a footnote saying: “Diese Berichte 1929 I.” The ref- erence is to Einstein 1929n (Doc. 365). [3] Cornel Lanczos and Herman Müntz had already been acknowledged for their discovery of an error in Doc. 365 in Einstein’s first effort to modify that paper see Doc. 368. [4] Doc. 365, p. 6, equations (10a), (11). [5] Doc. 365 had given field equations without deriving them from variational principles, as Einstein had done in Einstein 1928o (Doc. 219). See the discussion in the Introduction, sec. V. [6] The equation number given here is a peculiar mixture of a 6 and a 7, probably indicating uncer- tainty of the equation numbering during the writing process. It probably should be read as a 6 that has been turned into a 7 with a curly tail. Note that the preceding equation is numbered 6 and the succeed- ing equation is numbered 7. Note that the next reference to equation (6) in the text seems to refer to this equation. [7] The method of Doc. 365. [8] The reference here seems to be the equation number discussed in note 6 above. [9] From this point on, the manuscript underwent multiple revisions. The remainder of p. 3 was first struck out and replaced with two pages labeled p. 4 and p. 5 (as suggested by the numbering of the equations). Then these were in turn struck out and replaced with two pages also labeled pp. 4 and 5 and in this text identified as p. 4a and p. 5a. The order of presentation is, first the struck-out text at the bottom of manuscript p. 3, then the final text given in p. 4a, then the struck-out p. 4, then the final text given in p. 5a, then the struck-out p. 5. [10] This quantity and notation was first introduced in equation (1) of Doc. 368.