430 Prof. Powell's Remarks on Mr. Barton's Paper 



periment of Newton's was confessedly of a rougher description 

 than most of his others ; and the passage in which it is described 

 appears to me a little obscure throughout. (See Optics, Book iii. 

 p. 300 et seq. Ed. 1721.) But there is one circumstance at- 

 tending the experiment, which alone suffices to place it out of 

 all fair comparison with Fresnel's formulas; viz. that the origin 

 of rays, instead of being a single point (the essential assumption 

 in the theory), was a hole a quarter of an inch in diameter. 



But we may view it in some of its other relations : Newton 

 shortly after describes very precisely his well-known experi- 

 ment with two straight edges meeting at a small angle, and 

 gives a representation of the fringes. In this case it is evident 

 that the part of the image corresponding to the very narrowest 

 part of the opening is by no means dark at the centre ; but 

 just beyond, between the diverging hyperbolic branches, there 

 is the commencement of a dark space represented. Now if 

 we suppose two such constructions placed with the points 

 towards each other, there would result an appearance on the 

 whole similar to that in Mr. Barton's diagram ; and the ar- 

 rangement of the edges would resemble his two curved edges, 

 except that it is implied they were not actually in contact at 

 their point of nearest approach ; but if they were so near as 

 to allow no sensible portion of light to pass, even at one point, 

 the case would be precisely that just described. 



But it is not necessary to suppose the edges actually in con- 

 tact to produce this effect; for even when they are at a sensi- 

 ble distance, it results from the well-known fact of the enlarge- 

 ment of the shadows of the two edges beyond their geometrical 

 boundaries, that these shadows (projecting as it were before 

 the edges to which they belong,) will coalesce, before the edges 

 meet ; and this is a result of the undulatory theory. 



The same thing will be true with parallel rectilinear edges, 

 and may be seen even without the necessity of a single lumi- 

 nous point as the origin ; and this seems very likely to have 

 been the real result observed by Newton, in the experiment at 

 first referred to. 



I have repeated the experiment with slightly curved, as well as 

 with rectilinear edges, in several forms and at various distances; 

 and I have invariably observed, when the curved edges ap- 

 proached very near, an appearance similar to that represented 

 in Mr. Barton's diagram, but with this important difference, viz. 

 that the central dark portion, which (if 1 understand his dia- 

 gram rightly,) is represented as absolutely dark and isolated, 

 and having a continuation of the bright fringes on each side of 

 it, in my experiments always appears to join continuously with 

 the dark shadow on each side ; the bright fringes always bend- 



