﻿Wide Angle Lens for Cloud Recording. 1001 



to show that a cloud at an angle 6 from the zenith will 

 appear on the plate PP at a distance a from the centre 

 of the image, given by 



x = Ji sin 0/ \/ jjb 2 — .sin 2 #. 



The lens is, of course, not corrected for chromatic aber- 

 ration, but this might be largely removed by employing a 

 monochromatic filter, which could be closen so as to facilitate 

 the photographing of clouds. 



The finite size of the image of a distant point source, 

 due to the plate PP not coinciding with the true image III, 

 can be shown to be very approximately d/7 , where d is the 

 diameter of the aperture in the screen S. If the hole in 

 the screen be of diameter d=r/20, the aperture of the lens 

 is roughly f. 50 ; and the finite size of the image mentioned 

 above results in a blurring at the zones which are most out 

 of focus, which is equivalent to an uncertainty in 6 of 

 about i°. 



The illumination of PP becomes less for large values of 0, 

 but the effect does not seem sufficiently pronounced to be 

 objectionable in cloud photography, and need not be 

 considered in detail. 



The angle of the field of view of the lens could be 

 increased yet further by replacing the outer plane plate Gr 

 by a plano-convex lens, having its curved face outwards. 

 This arrangement might be used if it were desired that the 

 photograph should show the horizon and slightly below it. 

 An advantage of this arrangement would be that the zone 

 for which the image is most cramped together would not 

 be at between 80° and 90° from the zenith, but say from 

 95° to 105°, thus enabling the record of clouds near the 

 horizon to approach more nearly the clearness of that of 

 clouds at the zenith than would be the case if a simple 

 hemispherical lens were used. 



It may be mentioned that the lens gives views similar to 

 those seen by a fish in water. The apparatus has, however, 

 probably been reduced to the simplest form advisable, though 

 a less perfect image could be obtained by placing the photo- 

 graphic plate in contact with one side of a thick parallel 

 plate of glass, the other side being covered by a thin screen 

 pierced by a small hole. 



Finally, it will be seen that when using the simple hemi- 

 spherical lens, or the thick plate of glass just described, only 

 two constants (viz. h and fi) need be known accurately, to 

 enable a complete network of degrees to be constructed for 

 use in interpreting the photographs. 



Phil. Mag. S. 6. Vol. 44. No. 263. Nov. 1922. 3 T 



