348 TROPICAL NATURE. v 
weed of these two colours. Experiment shows, however, 
that when blinded the change does not occur; so that 
here too we probably have a voluntary or reflex sense- 
action. 
These peculiar powers of change of colour and adaptation 
are, however, rare and quite exceptional. As a rule there is 
no direct connection between the colours of organisms and 
the kind of light to which they are usually exposed. This is 
well seen in most fishes, and in such marine animals as por- 
poises, whose backs are always dark, although this part is 
exposed to the blue and white light of the sky and clouds, 
while their bellies are very generally white, although these 
are constantly subjected to the deep blue or dusky green 
light from the bottom. It is evident, however, that these 
two tints have been acquired for concealment and protection. 
Looking down on the dark back of a fish it is almost invisible, 
while, to an enemy looking up from below, the light under- 
surface would be equally invisible against the light of the 
clouds and sky. Again, the gorgeous colours of the butter- 
flies which inhabit the depths of tropical forests bear no 
relation to the kind of light that falls upon them, coming as 
it does almost wholly from green foliage, dark brown soil, 
or blue sky; and the bright underwings of many moths, 
which are only exposed at night, contrast remarkably 
with the sombre tints of the upper wings, which are 
more or less exposed to the various colours of surrounding 
nature. 
Classification of Organic Colours 
We find, then, that neither the general influence of solar 
light and heat, nor the special action of variously tinted rays, 
are adequate causes for the wonderful variety, intensity, and 
complexity of the colours that everywhere meet us in the 
animal and vegetable worlds. Let us therefore take a wider 
view of these colours, grouping them into classes determined 
by what we know of their actual uses or special relations to 
the habits of their possessors. This, which may be termed 
the functional and biological classification of the colours of 
living organisms, seems to be best expressed by a division 
into five groups, as follows :— 
