1 0 6 D O C U M E N T 9 0 O C T O B E R 1 9 2 5 It follows from the first in relation with (13)[6] and out of (21) in connection with (14)[6] The last two equations are in good agreement with the second group of Maxwell’s formulas All of this agrees with what you found and I also notice that if one defines , it follows from (7) that and from (6) and (5). This means that if , equation (3) holds for all the values of the indices α, μ, ν. It now suggests itself that if vector is not set equal to zero one could perhaps obtain a link to the second group of Maxwell equations as they read in the presence of charges, namely: , It goes without saying that , will be related to . Nev- ertheless, it is not as simple as that on the contrary, it can be shown that the men- tioned link is scarcely possible. In order to see this, it suffices to look at the elec- trostatic case (e.g., of an electron at rest). Then (17) changes into . Because , it follows that . If we hold that then follows, which is not valid in general, of course. grad div E ∂ ∂t ---- rot H E) · – ( + 0. = rot (rot H E) · 0 =– rot H E · = div E 0. = ϕα 0= Γαα α 1 2 --gαα---------αxα∂ - ∂g = Γμα μ 1 2 -- - gμμ© ∂gμα ∂xμ ----------- - ∂gμμ ∂xα ----------- ∂ -----------¹·αμxμg∂ -–+ § = ϕα 0= ϕα rot H E · ρv += div E ρ = a1, a2, a3, a4 ρvx, ρvy, ρvz, ρ ΔE rot a –= div rot a 0= Δ div E 0= div E ρ ,= Δρ 0=