THE PRESIDENT'S ADDRESS. 33 



tures, and on this account it, as well as the small cone by which 

 it is produced, ought to be got rid of as far as possible. Mr. 

 Wright has assigned it to a class of physical phenomena known 

 as Fresnel's interference bands. These images therefore will 

 in the future be only regarded as interesting examples of 

 experiments in physical optics. It is important to consider for 

 a moment the " true-false " image and its influence on that 

 obtained by the only correct method of microscopical illumina- 

 tion, viz., a 3'4 cone. It is, as we have seen, one of the com- 

 ponents of the resultant image, and it is formed by spectra 

 passing through the peripheral annulus. The new theory 

 shows that this image also partakes of the nature of Fresnel's 

 interference bands, but because it is a " true-false " image it 

 strengthens, by its superior resolution, Mr. Wright's new 

 " true" image, and as we must put up with it, full cones being 

 impracticable, it is consoling to know that it is a " true-false " 

 image that we have to deal with, which will assist, and not 

 injure, our new "true" image. 



Putting the case in another way we see that it is the image 

 formed by the central three-quarter portion of the whole objec- 

 tive that definitely fixes the focus, and consequently we are 

 unable to play upon focal adjustment for the formation by 

 means of the Fresnel bands of various pictures, which, however 

 beautiful they in themselves may be, yet have nothing what- 

 ever to do with the structure under the microscope. Some will 

 say that we have at last come back to Abbe's original theory, 

 which he has since abandoned, viz., that the microscopical 

 image is compounded of two superimposed images, one a 

 dioptric image and the other a diffraction image beginning at 

 23V0 inch- To this we reply that, although in words it may 

 be so, in meaning the case is far different. To mention two 

 differences : Abbe's double image was the essence of the micros- 

 copical image, but now the double image is an accident arising 

 from the impossibility of using full cones. Again the resolu- 

 tion in the centre conformed to the Abbe limit, whereas it now 

 possesses less resolving power owing to its dependence on the 

 Airy limit. 



With regard to the action of the four kinds of illumination 

 by means of axial cones, the following are from results obtained 



Journ. Q. M. C, Series II., No. 36. 3 



