3 2 0 D O C U M E N T 3 3 5 D E C E M B E R 1 9 2 8 This is in any case extremely simple. Try it out on a centrally symmetric field! Forgive me for my restlessness. But why should one make things complicated when there is such a simple possibility? Best regards, your A. Einstein P.S. Please send both of today’s letters back to me, since everything is so nicely written down in them.[3] 335. To Chaim Herman Müntz [Berlin/Gatow?,] 15 December [1928] Dear Mr. Müntz, Setting up the field equations starting from the identity is a more subtle exercise than I had originally thought.[1] I will convey my considerations of this problem to you here they would seem to have solved the task, even though I am not quite sure that my solution is the only one possible. The identity that forms our starting point[2] is the following: (1) In order to illustrate my method, I first want to show how, in a quite simple man- ner, one can arrive at some physically not useful equations, which however fulfill all of the formal conditions. Let me first note that is none other than that is, what one usually denotes as the electric field. We will call it for short (without a comma). We write in abbreviated form: … (1a) We now take its divergence with regard to l, whereby here and in what follows I take the liberty of leaving off the writing out of the raising of the index, so as to increase clarity.[3] We find for which, owing to the commutatability of differentiation, we can also write … (1b) For the field equations, we take … (2) These are 16 equations. From them, according to (1b), follow the electromagnetic … (3)  l   l    l –     l  0  – +  l    l –     l  –  l,  ,l –  l x  --------   x l --------- –       l  l    l 0  –  l  l   l l – 0  l   l  l l 0  –     l 0 =  l l 0 =