780 
DR. OLIVER LODOE OR ABERRATION PROBLEMS. 
In practice, when the telescope is being carried along by the earth at angle 6 to 
the light, it is customary to assume the medium stationary, in which case a = 0 and 
the first term alone exists; or we may assume that the ether is more or less carried 
along by the earth, in whicli case 9 = (f), and the second term attains its maximum 
value, ga/3, when the angle is 45°. But even if a is as great as /3, the most favourable 
case, unless the ether has a proper motion of its own, this only means a discrepancv 
in a star’s absolute position of ‘OOl'' ; so it is hopeless to look for a motion of the 
medium this Avay. 
50. The experiment of filling a telescope tube with water as suggested by 
Boscovich, and tried by Airy, and more exactly by Hoek,'“ aims at a motion quite 
different from that above contemplated. It aims at moAung a portion only of the 
ether in a partitioned-off region of space. It is easy to show that on Fresnel’s 
tlieory no different aberrational effect can be thus observed than is observed with an 
air telescope ; in fact, Fresnel himself, in his original letter to Arago (§ 4) contem¬ 
plated this experiment, and predicted its negative result, but it^may be instructAe to 
enter on its consideration in a geometrical fashion. 
Kjfect of Motion combined with Change of a Portion of the Medium. 
So long as the Avhole medium is changed or moved, we have seen Avhat effect there is 
on aberration. Motion has no effect on it, it drifts the Avave-normals but cannot affect 
the rays ; increased g Avithout motion increases it. But if the ray has to pass through 
a bounding surface, and if the change or motion occurs only on one side of that 
smface, tlien circumstances are different. 
If, for instance, the source S sends a ray SA, Avhich AAmuld have gone to B, a 
change of medium may carry it to C, Avhile a drift of the medium, carrying A to A' 
and B to E, Avill slant the ray along AE (fig. 13). 
It is, therefore, just possible for a drift to neutralize a refraction, and to let a ray 
enter a dense medium Avithout bending. To this end the drift must equal CB or 
A A', and the ray Avill then be straight. 
* ‘ Arcliives Neerlandaises ’ (1868), vol. 3 p. 180. 
