Consequences of the Electrical Theory of Matter. 51$ 



longitude for the drift which would make the components of 

 motion along the minor axes of the Earth and Venus orbits 

 very small, and therefore such as would have but little 

 effect on their perihelia, yet on page 166 Prof. Eddington 

 truly pointed out that, although the perihelia might thus 

 escape, the excentricities would be affected ; and, further, 

 that no contrivance of drift direction could avoid affecting 

 either the excentricity of the orbit or the longitude of its 

 perihelion, one or other, since these perturbations are 

 essentally at right angles. 



Well, I do not see my way at present to controvert this 

 conclusion, and yet I feel thai the last word has not been 

 said on the subject. The agreement for Mercury and Mars 

 is too marked to be readily abandoned. I claim that if once 

 the hypothesis be granted that the additional inertia due to 

 motion is not part of the body's true mass, and so is not 

 subject to gravity, the resulting perturbation comes so near 

 being able to explain the outstanding discrepancies of the 

 inner planets that the agreement can hardly be accidental. 



The whole thing turns upon whether the additional inertia 

 due to motion is or is not subject to gravity. If it is not, then 

 a vera causa is established which must be taken into account. 



In favour of the hypothesis of gravitative independence, I 

 adduce the analogy — admittedly not coercive — of a solid 

 moving through a fluid. The apparent inertia of such a 

 body is increased by an amount depending on the fluid 

 displaced, but its floating or sinking properties remain 

 unaffected : the extra inertia is no part of its mass, and is 

 not subject to gravity. If the extra electrical inertia of 

 moving matter is not part of the true mass, but represents 

 only aetherial reactions, which is what I expect, then an 

 astronomical perturbation is bound to be caused in rapidly 

 moving planets ; and whether this perturbation can be adjusted 

 to agree with observation, i. e. whether a solar drift can be 

 chosen which shall give a result neither in excess for one 

 planet nor in defect for another, becomes a matter for further 

 detailed calculation. 



Note 1. The treatment of orbits of any excentricity is remarkably 

 simplified by the singular little theorem, employed by Prof. 

 Eddington in October, p. 325, that the orbital velocity of an 

 inverse square orbit can be rigorously expressed as a constant 

 velocity v perpendicular to the radius vector, compounded with 

 another constant velocity v e perpendicular to the major axis. 

 The advantage of this theorem, when dealing with an orbit subject 

 to an unknown drift, is obvious ; and as it does not appear to be 



