Artificial Colour-blindness to Successive Contrast. 



218 



I next attempted to get the reversed image of the sun. This was 

 attended with difficulty, owing to the lack of a disc with a slit 

 sufficiently fine to reduce the sensation evoked by the direct light of 

 the sun ^vithin the limits required. I succeeded at last by increasing 

 the intensity of the illumination of the white card. This was effected 

 by holding in front of it a large lens by which the sun's rays could be 

 concentrated, the degree of concentration being regulated by adjusting 

 the distance of the lens from the card. In this way I was able to see 

 the sun's disc black upon a white ground. The experiment tended 

 to confirm the explanation already given of the appearance of the 

 incandescent filament under similar conditions. If the visual axis was 

 not fixed, three or four black discs would appear, and on looking 

 directly at one of the more central ones, the sun's disc seemed to be 

 partly visible behind it. A very curious effect was produced by 

 "sweeping" with the eye along a faint circle marked on the revolving 

 card. A whole series of black discs started into view one after the 

 other without a glimpse of the luminous disc that produced them. 



The principle underlying Sigmund Exner's method is illustrated in 

 an even more striking manner by the remarkable experiment of 

 Shelf ord Bidwell. In this a coloured object is placed behind a disc 

 half black and half white, with a sector 30° wide cut out of the white 

 portion. As the disc revolves, the eye is kept in darkness for a space, 

 then sees the coloured object for a short time, and immediately after- 

 wards a white surface for a considerably longer time. The retinal 

 fatigue induced by the colours of the object causes a negative after- 

 effect so strong that the object is seen in its complementary colours. 



From the point of view of my own investigations it was necessary to 

 repeat these experiments with the pure colours of the spectrum. 

 There are several positions in which a disc, such as Shelford Bidwell 

 employs, can be used in conjunction with a spectroscope. It may be 

 placed between the prism and the telescope, the latter being set back 

 an inch or two to make room for it, or it may work in a gap cut in the 

 body of the telescope, being illuminated by front light through a side 

 tube. But either arrangement involves some alteration of the spectro- 

 scope. The following method is free from this objection and has a 

 certain interest of its own : — 



The disc is placed in front of the eye-piece of the spectroscope, and 

 the spectrum viewed through a second telescope fixed in the optic axis 

 an inch or two from the eye-piece. But the second telescope magnifies 

 the spectrum and consequently renders it less bright. The definition 

 is, however, much better than would be expected, and is so little 

 affected by slight displacement of the second telescope from the optic 

 axis that it occurred to me to try the arrangement shown in figs. 5, 6. 



A telescope A, magnifying ten times, is placed with its eye-piece 

 close to the eye-piece E of the spectroscope. The disc B revolves 



VOL. LXVI. S 



