Chap. XII.] FISHES. 17 



where corals and other brightly-colored organisms abound, 

 are brightly colored in order to escape detection by their 

 enemies ; but according to my recollection they were thus 

 rendered highly conspicuous. In the fresh-waters of the 

 Tropics there are no brilliantly-colored corals or other or- 

 ganisms for the fishes to resemble; yet many species in 

 the Amazons are beautifully colored, and many of the car- 

 nivorous Cyprinidse in India are ornamented with " bright 

 longitudinal lines of various tints." 28 Mr. McClelland, in 

 describing these fishes goes so far as to suppose that the 

 peculiar brilliancy of their colors " serves as " a better 

 mark for king-fishers, terns, and other birds which are 

 destined to keep the number of these fishes in check ; " 

 but at the present day few naturalists will admit that any 

 animal has been made conspicuous as an aid to its own 

 destruction. It is possible that certain fishes may have 

 been rendered conspicuous in order to warn birds and 

 beasts of prey (as explained when treating of caterpillars) 

 that they' were unpalatable; but • it is not, I believe, 

 known that any fish, at least any fresh-water fish, is re- 

 jected from being distasteful to fish-devouring animals. 

 On the whole, the most probable view in regard to the 

 fishes, of which both, sexes are brilliantly colored, is that 

 their colors have been acquired by the males as a sexual 

 ornament, and have been transferred in an equal or nearly 

 equal degree to the other sex. 



We have now to consider whether, when the male difr 

 fers in a marked manner from the female in color or in 

 other ornaments, he alone has been modified, with the 

 variations inherited only by his male offspring ; or whether 

 the female has been specially modified and rendered incon- 

 spicuous for the sake of protection, such modifications be- 

 ing inherited only by the females. It is impossible to 



28 " Indian Cyprinidse," by Mr. J.. McClelland, ' Asiatic Researches,' 

 vol. xix. part ii. 1839, p. 230. 



