368 FISHES 



taneous variations, as would be required to bring about the 

 evolution of flat-fishes on the view that acquired characters 

 cannot be inherited ; the mutations are all in the direction of 

 greater symmetry, or a mere reversal of asymmetry, whereas in 

 the evolution of a flat-fish from symmetrical ancestors, mutations 

 in the direction of asymmetry would be required, and it has 

 never yet been shown that mutations which could lead to the 

 condition of flat-fishes occur in ordinary symmetrical fishes, 

 which swim in a vertical position. A still stronger argument 

 may be founded on the gradual development of the peculiarities 

 of the flat-fish, the process of metamorphosis. The mutations 

 to which we have referred in these fishes, and many other con- 

 genital variations in other animals, develop directly, not by a 

 gradual change from a preceding condition. 



We have every reason to believe that the abnormalities above 

 described are present from the beginning of the metamorphosis. 

 There is no reason to suppose that the flat-fish was first normal 

 and gradually became abnormal. On the other hand, the 

 mutations which would be required to produce the evolution of 

 the normal Flat-fish from a symmetrical ancestor would consist 

 in gradual changes such as we now see in the metamorphosis. 

 We are not, therefore, justified in assuming, as most evolutionists 

 at the present day would assume, that the abnormalities in 

 question are examples of the same kind of variation as that 

 which gave rise to the Flat-fish originally. On the contrary, it 

 is more reasonable to regard the two things as quite distinct 

 and different: the abnormalities as due to mutations arising 

 in the ovum, the original evolution of the flat-fish as due to 

 modification caused by the peculiar habits. 



Various well-known variations have arisen in certain species 

 of fish which have been domesticated for long periods. The 

 origin of these variations has not been scientifically studied by 

 modern methods but it seems most probable that they arose as 

 mutations. The most familiar of them is the colour of the 

 gold-fish, which is a domesticated variety of Carassius auratus, 

 a native of China ; it is doubtful if this species itself is really 

 more than a variety of Carassius vulgaris, which occurs naturally 

 in Europe and Siberia. The colour of the wild Carassius auratus 

 is greenish or olivaceous, and that of the gold-fish is due to the 

 reduction or absence of the black pigment which leaves the 



