Page:PoincareDynamiqueJuillet.djvu/47

 so that the total force can be divided into three components corresponding to the three brackets of expression (12); the first component has a vague analogy with the mechanical force due to the electric field, the other two with mechanical forces due to a magnetic field; to complete the analogy I can, under the first point, replace $$\tfrac{1}{B^{3}}$$ by $$\tfrac{C}{B^{3}}$$ in equations (11), so that X1, Y1, Z1 only depend linearly on the velocity ξ, η, ζ of the attracted body, since C has disappeared from the denominator of (11bis).

We pose then:

it follows that C had disappeared from the denominator of (11a):

and there will also:

Then λ, μ, ν or $$\tfrac{\lambda}{B^{3}},\ \tfrac{\mu}{B^{3}},\ \tfrac{\nu}{B^{3}}$$ is a kind of electric field, while λ', μ', ν' or rather $$\tfrac{\lambda^{\prime}}{B^{3}},\ \tfrac{\mu^{\prime}}{B^{3}},\ \tfrac{\nu^{\prime}}{B^{3}}$$ is a kind of magnetic field.

3° The postulate of relativity would require us to adopt solution (11) or solution (14) or any solution that would inferred by using the first remark; but the first question that arises is whether they are compatible with astronomical observations; the discrepancy with 's law is of the order ξ², that is to say, 10000 times smaller when it were of order ξ, that is to say, if the propagation happens with the speed of light, ceteris non mutatis; it is permissible to hope that it will not be too great. But only a thorough discussion will be able to teach it to us.


 * Paris, July 1905.

H.