598 MR WILLIAM SWAN ON THE GRADUAL PRODUCTION OF 



on the eye in equal times ; from which it obviously follows, that the brightness of 

 an impression on the eye increases with a rapidity exactly proportional to the bright- 

 ness of the light which produces it. 



This conclusion seemed so remarkable, that I determined to try whether the 

 direct light of the sun produced a given portion of its impression on the eye with 

 no greater rapidity than ordinary artificial light. For this purpose I made use of 

 a selaometer, represented in fig. 5, where K L represents a plate of brass with 

 two apertures A B, ,lth of an inch in diameter, and half an inch distant. A plate 

 of ground glass is placed before the apertures, and behind the aperture B, a 

 tube B C is fixed, in which is placed a Nicol's polarizing prism. A longer tube 

 B D, is fitted so as to turn freely upon the outside of the tube C D, and another 

 Nicol's prism is placed in its further extremity, so that, by turning round the 

 tube B D, the illumination of the aperture B can be varied at pleasure. A disc 

 EF, with a sector of 7 30 revolves rapidly in front of the plate, by means of the 

 band H I passing over the pulley G, so as to project beyond the aperture A, which 

 is only visible when the sector passes before it at each revolution of the disc* The 

 apertures were first illuminated by gas-light, and the disc being made to revolve 

 so rapidly as to produce a continuous impression, the apparent brightness of the 

 apertures was made equal by turning one of the prisms. When the apparatus 

 was next illuminated by the direct light of the sun at noon, and the disc made 

 to revolve so as to produce a uniform impression, the apertures were still equally 

 bright, although the position of the prisms remained unaltered. This experiment 

 was repeated several times with the same result, and a similar result was ob- 

 tained when moon-light was compared with gas-light. Now the effect of turning 

 round the prism is to diminish the brightness both of the sun-light and gas-light 

 in the same proportion. Since, therefore, the two apertures were always equally 

 bright, it follows, that the apparent brightness of the aperture behind the revolv- 

 ing disc, had also, in both cases, the same ratio to that of the light seen by unin- 

 terrupted vision. But the ratio of the apparent brightness of the aperture behind the 

 revolving disc to that of the direct light, evidently depends on the rapidity with 

 which the light acts on the eye at each passage of the sector before the luminous 

 aperture. Hence it is obvious, that if the sun-light and gas-light required dif- 

 ferent times to produce like portions of their total effect on the eye, the apparent 

 brightness of the flashes produced by the revolving disc would have different ratios 



* By means of tliis arrangement, the brightness of the impressions produced during the revolu- 

 tion of the disc, can be compared with the light transmitted through the aperture B. Since the in- 

 tensity of a ray of polarized light when transmitted through a doubly-refracting crystal, varies as the 

 square of the cosine of the inclination of the ])rincipal section of the crystal to the plane of polariza- 

 tion of the ray ; by attaching an index to tlie tube B D, so as to measure the angle through which it 

 has been turned, tlie intensity of the transmitted light might be estimated, and thus the brightness 

 of the impressions produced by the revolving disc might be determined. (See SuppUment au Traite 

 de ta Lumiere de Sir J. F. W. Herschel. Pur A. Quetelet, p. 595.) 



