v COLOURS OF ANIMALS 393 
abnormal excrescence and gain additional vitality by the 
adoption of the views here imperfectly set forth. 
Although we have arrived at the conclusion that tropical 
light and heat can in no sense be considered as the cause of 
colour, there remains to be explained the undoubted fact that 
all the more intense and gorgeous tints are manifested by the 
animal life of the tropics; while in some groups, such as 
butterflies and birds, there is a marked preponderance of 
highly-coloured species. This is probably due to a variety of 
causes, some of which we can indicate, while others remain 
to be discovered. The luxuriant vegetation of the tropics 
throughout the entire year affords so much concealment that 
colour may there be safely developed to a much greater 
extent than in climates where the trees are bare in winter, 
during which season the struggle for existence is most severe, 
and even the slightest disadvantage may prove fatal. Equally 
important, probably, has been the permanence of favourable 
conditions in the tropics, allowing certain groups to continue 
dominant for long periods, and thus to carry out in one 
unbroken line whatever developments of plumage or colour 
may once have acquired an ascendency. Changes of climatal 
conditions, and pre-eminently the glacial epoch, probably led 
to the extinction of a host of highly-developed and finely- 
coloured insects and birds in temperate zones, just as we 
know that it led to the extinction of the larger and more 
powerful mammalia which formerly characterised the tem- 
perate zone in both hemispheres; and this view is supported 
by the fact that it is amongst those groups only which are 
now exclusively tropical that all the more extraordinary 
developments of ornament and colour are found. The obscure 
local causes of colour to which we have referred will also 
have acted most efficiently in regions where the climatal 
condition remained constant, and where migration was unneces- 
sary ; while whatever direct effect may be produced by light 
or heat will necessarily have acted more powerfully within 
the tropics. And lastly, all these causes have been in action 
over an actually greater area in tropical than in temperate 
1 These views have been restated and enforced by much fresh illustration 
and argument in Darwinism, chap. x. 
