584 Dr. G. Johnstone Stoney on 



any image that can be formed from it by any optical instru- 

 ment, more of the detail of the microscopical object than that 

 which is contained in image B. Tii fact no optical instrument 

 that can be made can form an image that will contain so 

 much of the detail as does image B, inasmuch as to do so it 

 should be able to receive and deal with all the light emitted 

 from the object which would enter the air space 8 if both the 

 cover-glass and the air space S had been extended laterally 

 without limit. Now in practice the more oblique rays of this 

 light fail to get through the objective, since the NA of a dry 

 objective cannot be quite so much as 1*00, nor can that of 

 an oil-immersion objective be quite so much as the index of 

 refraction in the oil — conditions which would need to be 

 fulfilled in order that the objective might be able to grasp 

 the whole of the light. 



35. This circumstance makes another standard image more 

 useful in studying the microscope than image B. This, which 

 we may call image C, is the image which would be produced 

 by the same supposed succession of events as image B, with 

 the sole difference that the light subjected to Reversal is not 

 to consist of all the u f w's into which the light emitted by the 

 microscopical objects may be resolved, but only of such of 

 them as advance in directions not so much inclined to the 

 optic axis as to make it impossible for the objective of the 

 microscope to transmit portions of them. The image formed 

 in this way would occupy the same position as image B in 

 space Q, which by one of the hypotheses made is to be 

 regarded as occupied by an extension downwards of the air 

 or oil which occupies stratum S. 



36. No image formed by light whose waves are not infinite- 

 simal in length can reproduce the whole of the detail present 

 in any real object, all beyond a certain limit of fineness being 

 necessarily absent from the image. The image which could 

 furnish all detail, would need to be formed by light of which 

 the wave-length is cipher. This impossible image is called 

 the Geometrical Image, and may be spoken of as image A. 

 Image B contains as much of the detail as light of the 

 wave-lengths that form that image can by any possibility 

 reproduce; while image C, which is formed from ufw's 

 whose inclination to the optic axis is limited by the NA of 

 the objective, becomes a different image when an objective 

 with a different NA is in use ; but is for each objective the 

 standard image for that objective, reproducing the most 

 minute detail of the object, which could by any possibility 

 exist in any image of the object formed by an objective of 

 that aperture. Accordingly, it is legitimate for us to treat 



