322 ON LIGHT. 



central line devoid of colour. That these fringes origi- 

 nate in the mutual interference of rays which have passed 

 beside buth the edges of the object and entered the 

 shadow, is proved by intercepting the light on one side 

 only, leaving that on the other to pass freely. AH the 

 interior fringes and the central streak disappear, leaving 

 only the exterior ones on the illuininated side outstanding, 

 Tlie shadow (which is now formed under the same cir- 

 cumstances as in the former case) must, it is clear, be 

 still receiving one-half the total quantity of light which 

 it did before ; and if its edge be narrowly examined, it 

 will be seen not to terminate in any sharply-defined line 

 cutting it off from the fringes, but to graduate off insen- 

 sibly; and hence arises a very singular ph^enomenon. 

 If the light be readmitted on both sides, and the breadlh 

 of the shadow, or what under such circumstances must be 

 accepted as such, be measured, it is found to be nmch 

 broader than it ought to be were it limited by straight 

 lines drawn from the illuminating point through the edges 

 of the object. And, what is still more remarkable, if its 

 breadth be measured on a screen, successively placed at 

 different distances from the object, its increase of 

 breadth is found not to be in the simple proportion of 

 its increased distance ; as it would were the aerial shadow 

 (or the space shaded by the object) bounded by straight 

 lines ; but as if by curves starting from its edges, and 

 having their convexities towards the light. And, finally, 

 if the object and the SQx^<tYi^ preserving the same distance 

 between them, be moved gradually nearer and nearer to 

 the illuminating point, the fringes, both interior and ex- 



