382 
Transactions of the Society. 
element of green in it in place of pure red, as before, and in con- 
sequence the green images A will not be contrasted so well with it. 
The dark images B show a still more marked diminution of contrast, 
because these being almost entirely dioptric images, scarcely any of 
the additional light which we obtained by enlarging the iris opening 
is utilised in their formation. 
In the next photograph (plate X. fig. 3), taken with a 7J mm. 
iris diameter, the background becomes still more greenish in colour, 
so that the green images A begin to change from the appearance of 
light on a dark ground to dark on a light ground. The images B 
become still fainter than before. 
The last of the series (plate X. fig. 4) is taken with a cone of 
light (12 mm. iris diameter) equal to the full aperture of the objective, 
so that the whole of the colour-disc receives direct light. The back- 
ground appears completely green, and is therefore light coloured on 
the photograph, and now the images A appear quite dark in strong 
contrast to it. On the other hand, the images B have practically 
disappeared ; the excess of direct light passing through the green part 
of the disc has flooded them out, and the photograph is to all intents 
and purposes the same as if taken without the red central stop at all. 
1 may here mention that these photographs were taken on the 
most colour-sensitive plates obtainable, viz. Cadett’s spectrum plates. 
Had less colour-sensitive plates been used, the transition of the back- 
ground from dark to light would have been better shown at the 
expense of not rendering the images B so well. 
Under one of the Microscopes exhibited this evening, you will 
find objects arranged with the double image colour-discs, and you can 
there see the results much more clearly than in the imperfect photo- 
graphs which I have been able to take. A great peculiarity about the 
visual perception is that as the iris is gradually opened, the images 
A from the green rim appear to change from green to red, whilst 
the image B from the red central part appears to become greenish. 
This is purely an effect of subjective colouring. 
It seems to me that an important fact to be learned from these 
photographs with the double image colour-discs, is that in using 
increasingly wide cones of illumination we are working more and more 
with the marginal zones of the objective, and less and less with the 
central portion. The reason is apparent. Dividing an objective having 
a back lens of, say, 8 mm. into four zones by constructing imaginary 
circles of 1, 2, 3 and 4 mm. radius respectively, the illuminating 
powers of the zones are as — 
Now it is evident that in focusing an image we adjust it to work 
more particularly with that zone of the objective which has the 
greatest illuminating power, i.e. through which most rays of light 
2 2 , 4 2 - 2 2 , 
and 8 2 — 6 2 respectively, 
or as 1 3 
7 
