CHKOMATIC FUNCTION. 91 



darkness, and in fact is so developed normally in many 

 animals.^® 



In absolute antagonism to the old hypothesis which ascribes 

 the origin of the pigment in the skin of animals to the direct 

 influence of light, there is another which, under the almost 

 supreme influence of Darwin's theories, is now as generally 

 accepted as the other was formerly. It is now almost uni- 

 versally asserted that the colours of animals have arisen from 

 either natural or sexual selection. We will postpone the dis- 

 cussion of this view to a future chapter, in which the uses 

 accruing to animals from their colours will be considered ; but, 

 since it is proved by abundant evidence that at least one par- 

 ticular kind of protecting resemblance — i.e. the adaptation of 

 the colour of the skin of certain animals to the colours of the 

 objects that surround them — depends on the influence of light 

 through the medium of the eyes, it will be convenient to treat 

 of it here. Pouchet applied the term ' chromatic function ' to 

 that adaptation of colour to the surroundings of the creature 

 which is indirectly the result of sight, in order to distinguish it 

 clearly from other cases in which — so far as we can at present 

 tell — the distribution of colour is not influenced by light at all. 



The tei-m ' chromatic function ' refers neither to constant 

 colouring, even when this causes a protective resemblance, nor 

 yet to such variations in colour as are occasioned in Ohamse- 

 leons and Cuttle-fish by physical irritation without any protec- 

 tive resemblance being the result. The expression, which is not 

 altogether a happy one, is new ; but the fact it designates, that 

 such protective changes do occur in many animals, has long been 

 known. In the year 1830Starkmadeanumberofobservationson 

 the subject, on species of the genera Leuciscus, Gasterosttus 

 (the fresh- water Stickleback), Gobitis 6ar6a<MZa,andthe common 

 Perch, Percajluviatilis. All these fishes change colour with 

 some rfipidity, some in a, few hours, others in from two to three 

 minutes ; and we know now that many splendidly coloured 

 sea-fish have the same power, often in a quite extra ordinaiy 

 degree, as, for instance, species of Serrarms. Shaw seems to 

 have been the first to observe, in 1838, that such fish as are 

 capable of changing their colour, apparently at will, must be 



