225 
Some of the foregoing results are fair approximations, and 
probably better results might be obtained with a more per- 
fect form of apparatus than I employed. The surfaces used 
were not perfectly flat. Moreover, there was in some of the 
experiments the indecision arising from the coexistence of 
several distinct tints on the surface of the disc due to the 
formation of caustic curves. This I afterwards remedied by 
working behind a screen of lawn. The effect of this is to 
render the illumination of the disc much more uniform. 
Instead of a screen we might also use glass cylinders ground 
internally. Another difficulty is the indecision respecting 
the depth at which equality of colour takes place. The de- 
gree of indecision will be different for different persons, and 
I suppose represents want of sensibility and want of skill in 
detecting shades of colour. So far as it arises from want of 
skill it is likely to be diminshed by more perfect experience 
with colours. It seems to me that it would be well to have 
some definite method for exercising the eye, or for testing 
one’s skill in this respect. This might be done by having a 
series of cylinders filled with coloured solutions, the quantity 
of colouring matter increasing in arithmetical progression and 
endeavouring to arrange these solutions in their order, the ar- 
rangement being made according to the judgment of the 
tints. The various success attending such exercises would 
afford the experimenter some method for testing his skill 
from time to time; it would also show him under what 
conditions he could use neither this nor any other method 
of colorimetry ; for when he has arrived at such a stage that 
he can perceive no difference of tint, notwithstanding an 
increase in the colouring matter, it seems as useless to try 
to do so as to use a balance of coarse sensibilitv to distin- 
*/ 
guish between two weights which are very nearly the same. 
By frequent exercise and concentrated attention I think the 
power of perception of difference of shade may be increased. 
If such a series of cylinders were taken containing, say 
