TJieories of Microscopical Vision. By A. E. Conrady. 629 



But we cannot strictly realise illumination from a point-source 

 of light and as soon as we employ an extended source of light 

 we shall secure a still more definite focus of the image. For 

 different points in the source of light will illuminate the object 

 from different directions ; there will be correspondingly displaced 

 pairs of points P and F ; and the amount of confusion by over- 

 lapping of images, corresponding to different points of the source 

 of light and to light of different colours, will become so pronounced 

 as to destroy all semblance of a recognisable image beyond the 

 immediate neighbourhood of the plane 1 1. Hence the latter 

 will be instinctively found by focussing, because it is the only 

 position yielding a sharp image. 



When other spectra in addition to the direct light and one first 

 diffracted wave are admitted, the same effect results from the use 



Fig. loo. 



of an extended source of light ; but as there is a difference in this 

 case when a point-source of light is used, I will briefly treat an 

 example — viz. the case when the direct light and the two first 

 diffracted waves are admitted. 



Starting in this case from a bright point in the true image, 

 where the two diffracted waves meet the direct one with a whole 

 number of wave-lengths difference of phase, we can apply the 

 same reasoning which we used before, to show that the two 

 diffracted waves will meet in phase anywhere along the line con- 

 necting the bright point in the image with the focus P of the 

 direct light ; for that point P is midway between the foci P' and 

 P" (fig. 100) of the diffracted waves. But, as the path from P to a 

 point on this connecting line shortens more quickly than the 

 paths from P' and P" (owing to the inclination of the latter paths), 

 the combined diffracted waves will meet the direct wave succes- 

 sively in all possible phase-relations, with the result that we get 



Dec. 21st, 1904. 2 x 



