D O C U M E N T 2 5 1 A U G U S T 1 9 2 8 2 4 5 This is highly implausible. It would therefore seem preferable to employ the invariant[5] . This is to be sure only a feeling, which I cannot justify logically. It would be good in any case if you could carry out your method for the equations belonging to as well, where the legs are correctly coupled to each other. 4) When the singularity problem has been solved, then the problem of equilibrium is also solved for those [equations], assuming that the strict solutions are free of singularities.[6] For the field equations lead to “conservation equations”: (field equations) . If the , or the , are independent of , then this equation gives upon in- tegration a surface integral relation (vectorial), which has the sense of an equilib- rium condition. Then it will be seen as to whether this condition can be understood/ interpreted in terms of experience! Best regards, your A. Einstein 251. From Roland Weitzenböck Laren (N. H.), 8 August 1928 Dear Colleague, Thank you kindly for your letter of the 3rd of this month.[1] I have now finished writing my note and have included more in it than I had originally intended.[2] I derived the field equations that result from the four simplest action functions A, B, and in a general way a calculation that is not quite effortless.[3] I did not dis- cuss any physical interpretation. In particular, I take the liberty of calling your at- tention to my remark on p. 8 of my manuscript: From ,[4] one obtains the equations ,[5] which you had already mentioned. I 1 g = I 1 h a H h a ------------ x -------- H h (a) ---------------- – 0 = x -------- H H h a ---------------- -h a – 0 = I I h a x 4 = rot 0 =