" Experiments and Observations on the Properties of Light." 3 



the length of route of the rays, bvit it has on the angle at which 

 they intersect ; so that (3) in the case represented in the author's 

 fig. (20), supposmg abstractedly two pairs of interfering rays 

 (such as BC, AC, and BD, AD), it is evident that the fringes at 

 D ought to be broader than those at C, not owing to any differ- 

 ence in the routes, but because the angle BDA is less than BCA ; 

 while (-4) interference perfectly explains fringes, even when the 

 action is wholly on one side of the ray or edge. 



But passing from these points of confessedly less importance, 

 we will proceed to the most material and fundamental experiment 

 (Prop. II. Exp. 1), in which, when fringes are formed by the edge 

 of an opake bcdy, if a second edge be placed at a greater distance 

 along the ray from the origin on the same side as the first edge, 

 it produces no change in the fringes, but on the opposite side it 

 does, the fringes being shifted in position towards the first side ; 

 or in other words, in the one case it has no power of producing 

 further difi"raction, in the other it has : and this is viewed by 

 the author as supporting his theory of a peculiar action exerted 

 by the edge upon the ray passing near it, by which it is disposed 

 or indisposed for further flexure according to the conditions above 

 expressed. 



The experimental fact in general is easily verified. There is, 

 however, one material condition necessary to be attended to for 

 reproducing the result exactly as described by the author. 



"N^Tien two edges are at the same distance from the origin and 

 from a narrow aperture, they give, as is well known, fringes on 

 each side extending into the shadow, with a white centre (fig. 1). 

 As one edge is removed successively further from the origin and 

 nearer to the screen, the fringes on that side dilate (fig. 3), 

 Fig. I. Fig. 2. Fig. 3. 



mi 



HE 



become faint, and at length disappear (fig. 3) ; so that beyond a 



B2 



