THE OEIGIN AND THE EXTINCTION OF SPECIES 347 



That a horse should grow wings is beyond the limits of the 

 possibilities of equine variation — there are no determinants which 

 could present variations directed towards this goal; but that any 

 multicellular animal which lives in the light should develop eyes lies 

 within the variational possibilities of its ectoderm determinants, and 

 in point of fact almost all such animals do possess eyes, and eyes, too, 

 whose functional capacity may be increased in any direction, and 

 which are adaptable and modifiable in smy manner in accordance with 

 the requirements of the case. As soon as the determinants of the 

 most primitive eye came into existence, they formed the fundamental 

 material by whose plus- or minus-variations all the marvellous eye 

 structures might be brought about, which we find in the different 

 groups of the Metazoa, from a mere spot sensitive to light to a 

 shadowy perception of a moving body, and from that again to the 

 distinct recognition of a clear image, which we are aware of in our 

 own eyes. And what wonderful special adaptations of the eye to 

 near and to distant vision, to vision in the dusk and at night, or in 

 the great ocean-depths, to recognition of mere movement or the 

 focussing of a clear image, have been interpolated in the course of 

 this evolution ! 



All such adaptations are possible, because they can proceed 

 from variations of determinants which are in existence ; and in the 

 same way it is possible, at every stage of the evolution of organisms, 

 for eyes to degenerate again, whether they have been high up or low 

 down in the scale of gradations of this perhaps the most delicate 

 of all our sense-organs. As soon as a species migrated permanently 

 from the light into perfect darkness its eyes began to degenerate. 

 We know blind flat worms, blind water-fleas and Isopods, also Ijlind 

 insects and higher Crustaceans, and even blind fishes and amphibians, 

 the eyes of which are now to be found at very different levels of 

 degeneration, as Eigenmann has recently shown in regard to several 

 species of cave-dwelling salamanders of the State of Ohio. In all 

 these cases it is only necessary for the determinants of the eye to 

 continue to vary in the minus direction, and the disappearance of the 

 eye must be gradually brought about. 



We must picture upward development in quite a similar way. 

 The forest butterflies of the Tropics could not possiblj^ all have their 

 under surfaces coloured like a leaf if the protective pattern depended " 

 solely upon the chance of a useful variation presenting itself. It 

 always presented itself through the fluctuations of the determinants, 

 and thus the appropriate colourings were not merely able to develop, 

 but of necessity did so in gradually increasing perfection. If chance 



