Mr. J. Barton on the Inflexion of Light. 265 



the crescent form of the sun during an eclipse, received through 

 a pinhole in a darkened room; and the still more familiar fact 

 of the distinct vision of a multitude of distant objects through 

 a refracting telescope; since the rays proceeding from all those 

 objects repeatedly cross one another in passing through the in- 

 strument. 



Secondly. Though the theory of Fresnel agrees pretty well 

 with the results of his own experiments, it is far from agreeing 

 with the results of other experiments made by observers of 

 acknowledged accuracy. Sir Isaac Newton and M. Biot have 

 each of them recorded a series of observations on the inflexion 

 of light, which may be employed as tests of the accuracy of 

 FresnePs theory. Newton, having admitted a beam of light 

 through a hole the 42nd part of an inch in width, let it fall 

 on a slit between the edges of two knives, at the distance of 

 8 feet 5 inches. Placing a sheet of white paper behind the 

 slit, he observed that the shadows formed by the edges of the 

 knives were bordered with a succession of coloured fringes. 

 Varying the width of the slit, he observed at what distances 

 the paper must be placed, so that the first of the dark inter- 

 vals between these fringes, coming from either side, might 

 cross one another in the centre of the spectrum *. Now, by 

 the theory of Fresnel, any one of these observations should 

 give us the length of the undulations supposed to constitute 

 light; and of course that length will be » the same when de- 

 d need from any other observation. The following Table will 

 show how far this is from being the case. 



Distances of the paper 



from the knives 



in inches. 



4 



3* 



81- 

 32 

 96 

 131 





Length of an undula. 

 tion by Fresnel*s 

 theory. 

 •00001385 

 •00001763 

 •00002075 

 •00001901 

 •00001896 

 •00001888 



Distance between 

 the edge of the 

 knives. 

 •012 

 •020 

 •034 

 •057 

 •081 

 •087 



In all these observations the length of an undulation turns 

 out smaller than it should be. According to Fresnel, the 



length of an undulation in red light is '00002512f inch. 



According to Young, it is '0000266 



In yellow light, the most luminous part of 

 the prismatic spectrum, the length of an 



undulation, according to Young, is '0000235 



Whereas the highest value deduced from 

 Newton's observations above, is no more 

 than -00002075 



• Optics, Book III.Obs.ix. p. 105. f -000368 millimetres 



Third Series. Vol. 2. No. 10. April 1833. 2 M 



