456 ON METHODS FOE DETECTING SMALL OPTICAL EETARDATIONS, [415 is required. This was afforded by a small cylinder lens, acting as sole eyepiece, whose axis is best adjusted by trial to the required parallelism with the slits. Fairly good results were obtained with a glass tube of external diameter equal to about 3 mm., charged with water or preferably nitro-benzol. Latterly, I have used with advantage a solid cylinder lens of about the same diameter kindly placed at my disposal by Messrs Hilger. With this arrangement a wire stretched horizontally across the object-glass in front of the slits is seen in fair focus. When the adjustment is good, the bands are wide and the blacknesses well developed, so that a local retardation of -£Q\ or less is evident if suitably presented. The bands are much disturbed by heated air rising from the hand held below the path of the light. The necessity for a high magnifying-power is connected with the rather wide separation of the interfering pencils as they fall upon the object-glass. The conditions are most favourable for the observation of very small retardations when the interfering pencils travel along precisely the same path, as may happen in the interference of polarized light, whether the polarization be rectilinear, as in ordinary double refraction, or circular, as along the axis of quartz. In some experiments directed to test whether " motion through the sether causes double refraction*," it appeared that a relative retardation of the two polarized components could be detected when it amounted to only X/12000, and, if I remember rightly, Brace was able to achieve a still higher sensibility. The sensibility would increase with the intensity of the light employed and with the transparency of the optical parts (nicols, &c.), and it can scarcely be said that there is any theoretical limit. Another method by which moderately small retardations can be made evident is that introduced by Foucaultf for the figuring of optical surfaces. According to geometrical optics rays issuing from a point can be focussed at another point, if the optical appliances are perfect. An eye situated just behind the focus observes an even field of illumination; but it' a screen with a sharp edge is gradually advanced in the focal plane, all light is gradually cut off, and the entire field becomes dark simultaneously. At this moment any irregularity in the optical surfaces, by which rays are diverted from their proper course so as to escape the screening, becomes luminous; and Foucault explained how the appearances are to be interpreted and information gained as to the kind of correction necessary. He does not appear to have employed the method to observe irregularities arising otherwise than in optical surfaces, but H. Draper, in his memoir of 1864 on the Construction of a Spherical Glass Telescope]:, gives a picture of the disturbances due to the heating action of the hand held near the telescope mirror. Topler's work dates from * Phil. Mag. Vol. iv. p. 678 (1902); Scientific Papers, Vol. v. p. 66. t Ann. de VObserv. de Paris, t. v.; Collected Memoirs, Paris, 1878. t Smithsonian Contribution to Knowledge, Jan. 1864.tact with the closed end, at the height of 70 inches or more " (Young's Lectures, p. 626,1807). If the mercury be wet, boiling may be dispensed with and negative pressures of two atmospheres are easily demonstrated.