D O C . 1 2 7 O C T O B E R 1 9 1 9 1 0 9 supply us with details on it.[1] This has not happened until now and so I do not want to wait any longer. I heard about the result obtained by Eddington from a report by Mr. van der Pol, curator at the laboratory here.[2] He attended the meeting of the British Association in Bournemouth and told me upon his return what Eddington had presented. As the plates have been only preliminarily measured, a final value could not yet be given however, in Mr. Eddington’s opinion, the reality of the effect is established and one can say with certainty that the deflection (at the solar limb) lies between 0.87″ and 1.74″.[3] The value found is thus a little too small, since your theory requires 1.7″. Van der Pol told me, furthermore, that a discussion took place (I would have liked to have been present) in which Sir Oliver Lodge expressed his congratulations to you and to Eddington for the obtained result.[4] I made a little calculation that probably contains nothing that’s new to you and relates to refraction in a gas enveloping the Sun. For lack of anything better, I there- by assume the gas is homogeneous and has the same temperature everywhere. The rapidity with which the density diminishes in the vertical direction, that is, e.g., the density gradient at the solar surface, then varies, depending on the nature and tem- perature of the gas. If this gradient is known, deflection of a ray passing through the gaseous mass depends on the refractive index n of the gas at point P, where the ray is at the shortest distance from the solar center O. One can calculate how large n has to be so that the deflection comes to 1″. If I now take for the mentioned den- sity gradient the value that would be valid for hydrogen at 0° C, I find n = 1 + with a gradient 20 times smaller (hydrogen at 5000° C helium at 10000° C), the result is n = 1 + . It is thus evident that an extremely low gas density could produce a deflection of 1″.[5] Luckily, this deflection would then diminish very rapidly if the ray and hence point P were situated farther away from the Sun. The increase in distance OP would have to amount to just a 6 millionth part, or a ten-thousandth part, resp., of OP in order to allow the deflection to drop to an eth part. Consequently, it will be easy, provided many stars appear on the plates, to distinguish your effect from this refraction in a solar atmosphere. We surely may believe (in view of the magnitude of the detected deflection) that, in re- ality, this refraction is not involved at all and your effect alone has been observed. This is certainly one of the finest results that science has ever accomplished, and we may be very pleased about it. ————— I heard from Ehrenfest that you are thinking of departing shortly on a trip to Hol- land and I hardly need tell you that it will be a very special pleasure for me to see you here once again. Still, I must not conceal that, after what you unfortunately have had to report to me about your health condition, your plan fills me with concern.[6] Can you really dare to subject yourself now to the burdens and fatigue associated with the journey? It is obviously of greatest importance that you stay in 47 10 10– ⋅ 210 10 10– ⋅