1 7 8 D O C . 9 1 G E N E R A L R E L A T I V I T Y A N D M O T I O N similar to eq. (15), except that the second equations is missing a factor of 2. In the manuscript, the expressions for and are indicated by horizontal dashes. [28] The way Einstein here turns the operator into the wave operator is analogous to the way he had obtained the wave operator from the Einstein tensor in Einstein 1916g (Vol. 6, Doc. 32). The covariant divergence of the Einstein tensor vanishes as a consequence of the Bianchi identities for the Riemann tensor from which and R are obtained by contraction. [29] There are more equations given by (18) and (19) than components to be determined by these equations. If (19) were a generally covariant equation, (18) could be regarded as a coordinate condition. In the approximation scheme at hand, (18) represents an additional constraint on the admis- sible solutions to (19). For a similar sense of “overdetermination,” see Einstein 1924d (Vol. 14, Doc. 170), where he already linked the question of finding particle solutions to field equations and the question of whether the field equations determine the equations of motion of the particles. [30] At the end of this paragraph, the manuscript contains the following deleted footnote: “Die gleichzeitige Erfüllung von (18) und (19) hat ferner das Verschwinden von zur Voraussetzung, welche Bedingung gemäss (15) tatsächlich erfüllt ist.” [31] Obviously, Einstein implicitly defines . [32] After “genügt,” the manuscript contains the deleted insertion “und ausserhalb des Nullpunktes regulär ist.” [33] After “wären,” the manuscript contains the deleted insertion “in sowie in.” [34] In Ritz and Einstein 1909 (Vol. 2, Doc. 57), Ritz and Einstein agreed to disagree with respect to the restriction of electrodynamics to retarded potentials. Ritz believed that such a restriction was justified and that the resulting introduction of an arrow of time in the foundations of electrodynamics (and from which he expected the thermodynamic arrow of time to be derived) was an advantage. For a detailed assessment of the debate between Einstein and Ritz, see Frisch and Pietsch 2016. Below the footnote, the manuscript contains in pencil the following question, possibly in a different hand: “Ist die Lösung durch ret[ardierte] V stets nicht zu speziell?” [35] should be . [36] In the manuscript, directly after equation (26b), Einstein wrote and later deleted: “Damit ist aber noch nicht gesagt, dass die Feldtheorie mit der Quantentheorie unvereinbar ist. Denn die Gleichungen (7) und (10) erlauben noch andere Ansätze für die erste Näherung für eine punktartige Singularität, deren Bewegungsgesetz.” [37] Instead of the first sentence of this final paragraph, the manuscript contains the following lon- ger statement: “Was die Beziehung zu der Quantentheorie anlangt, so bleiben verschiedene Möglichkeiten: 1.) Die Fortsetzung zur Betrachtung zu höheren Approximationen liefert von selbst Quanten- Bedingungen. 2.) Quanten-Bedingungen ergeben sich bei abgeänderten Hypothesen über die Natur der Singula- ritäten. 3.) Die zugrunde gelegten Feld-Gleichungen sind mit den Quanten-Thatsachen unvereinbar. Zwischen diesen Möglichkeiten lässt sich einstweilen nicht unterscheiden.” Most likely, Einstein intends the “nature of the singularities” to refer to the same that, earlier in the text, he referred to as the “character of the singularities.” For the meaning of the latter, see notes 5 and 8 above. Q 24 Q 34 L ik R ik 1 2 --g ik R – R ik g ik  k x Qik 1 2  ik Q –    L L  = x 1 x 2 x 3 0 = = =  ik ik