Page:MaxwellEther1879.djvu/2

I have not thought it expedient to delay the publication of the letter on the chance that something bearing on the subject might be found among Maxwell's papers.

(Copy.)

Cavendish Laboratory,

Cambridge,

19th March, 1879

Sir, I have received with much pleasure the table of the satellites of Jupiter which you have been so kind as to send me, and I am encouraged by your interest in the Jovial system to ask you if you have made any special study of the apparent retardation of the eclipses as affected by the geocentric position of Jupiter.

I am told that observations of this kind have been somewhat put out of fashion by other methods of determining quantities related to the velocity of light, but they afford the only method, so far as I know, of getting any estimate of the direction and magnitude of the velocity of the sun with respect to the luminiferous medium. Even if we were sure of the theory of aberration, we can only get differences of position of stars, and in the terrestrial methods of determining the velocity of light, the light comes back along the same path again, so that the velocity of the earth with respect to the ether would alter the time of the double passage by a quantity depending on the square of the ratio of the earth's velocity to that of light, and this is quite too small to be observed.

But if $JE$ is the distance of Jupiter from the earth, and $l$ the geocentric longitude, and if $l'$ is the longitude and $\lambda$ the latitude of the direction in which the sun is moving through ether with velocity $v$, and if $V$ is the velocity of light and $t$ the time of transit from $J$ to $E$,

$JE=\left[V-v\cos \lambda \cos(l-l')\right]t\,$.

By a comparison of the values of $t$ when Jupiter is in different signs of the zodiac, it would be possible to determine $l'$ and $v\cos \lambda$.

I do not see how to determine $\lambda$, unless we had a planet with an orbit very much inclined to the ecliptic. It may be noticed that whereas the determination of $V$, the velocity of light, by this method depends on the differences of $JE$, that is, on the diameter of the earth's orbit, the determination of $v\cos \lambda$ depends on $JE$ itself, a much larger quantity.

But no method can be made available without good tables of the motion of the satellites, and as I am not an astronomer, I do not know whether, in comparing the observations with the tables of Damoiseau, any attempt has been made to consider the term in $v\cos \lambda$.

I have, therefore, taken the liberty of writing to you, as the matter is beyond the reach of any one who has not made a special study of the satellites.

In the article E (ether) in the ninth edition of the "Encyclopaedia Britannica," I have collected all the facts I know about the relative motion of the ether and the bodies which move in it, and have shown that nothing can be inferred about this relative motion from any phenomena hitherto observed, except the eclipses, &c., of the satellites of a planet, the more distant the better.

If you know of any work done in this direction, either by yourself or others, I should esteem it a favour to be told of it.

Believe me,

Yours faithfully,

(Signed) J. Clerk Maxwell

D. P. Todd, Esq.