Mr. II. F. Talbot's Experiments on Light. 329 



portion between the whole time of observation, and the time 

 during which it disappears. This seems a very simple law, 

 and therefore likely to be true ; but such an experiment is 

 much too inaccurate to be relied upon to establish it. But as 

 it suggests a very important idea, namely, that time may be 

 employed to measure the intensity of light, it becomes desir- 

 able to establish it on solid grounds. And this is easily ac- 

 complished ; we have only to take a white circle, with one of 

 its sectors painted black, and make it revolve rapidly. It will 

 appear, as every one knows, of a uniform gray tint, without any 

 variation from the centre to the circumference ; and yet, if we 

 look first at a point near its centre, and afterwards at a point 

 near its circumference, a much greater portion of the white sur- 

 face will pass beneath the eye in the latter case than in the 

 former. But a greater portion of the black surface likewise 

 passes, and causes a compensation. In every point of the cir- 

 cle the white and black parts meet the eye during the same 

 proportion of time, and therefore the tint is uniform through- 

 out. 



This experiment, though very simple and well known, 

 seems yet well calculated to detect any error in the assumed 

 law, if it should deviate in the least from the truth ; for such 

 a deviation would cause a perceptible want of uniformity in 

 some parts of the circle which the eye would not fail to dis- 

 cover. I need hardly observe that it would be illogical to 

 assert a -priori the existence of this law of optics, however 

 simple and natural it may appear, unless we were perfectly 

 well acquainted with the circumstances which accompany the 

 action of light upon the retina, which is very far from being 

 the case. Its proof can rest upon experiment alone, and by 

 that it appears to be most satisfactorily established. 



Assuming white paper, at least for the present, as our 

 standard of whiteness, and applying to the revolving circle 

 black sectors of different angles, we obtain, when they are set 

 in motion, various gray tints, the obscuration in which is pro- 

 portional to the angle of the sector ; and therefore, if we make 

 an assortment of numerous shades of gray paper, and ascer- 

 tain by direct comparison with the revolving wheel that in 

 any particular experiment its tint matches one of the papers, 

 we have only to mark upon the paper the angle of the sector 

 employed, and we may ever after employ the paper with con- 

 fidence in questions of a similar kind. According to this no- 

 tation perfect whiteness would be represented by 0, and per- 

 fect blackness by 360. 



This method may be applied to determine experimentally 

 the intensity of a polarized ray. We may view through a 



Third Series. Vol. 5. No. 29. Nov. 1834. ' 2 U 



