122 EVOLUTION AND ANIMAL LIFE 



Curagao, and still darker bay is the crown of Dendroica ruficapilla 

 capitalis, the golden warbler of the Barbadoes. 



Still other goldeii warblers exist, with the chin and throat 

 chestnut as well as the crown. One of these, ohve green on the 

 back, is Dendroica rufigida, of Martinique. The others are 

 more yellow. One of these, with the sides heavily streaked, 

 inhabits the isthmus region, Dendroica erythacoides , callad a 

 distinct species, because no intergradations have been made 

 out. Another, more faintly streaked, replaces it on the Atlantic 

 coast from Yucatan to Costa Rica, Dendroica bryanti, while 

 the Pacific coast, f]'om Sinaloa to Costa Rica, has another form, 

 with still fainter markings, Dendioica bryanti castaniceps. An 

 extreme of this iorvn with the tliroat and breast tawny, but not 

 the crown, is found in Jamaica again and is known as Dendioica 

 eoa. In this case, which is one typical of most groups of small 

 birds, the relation of the species to the barriers of geography 

 is so plain as to admit of no doubt or question. 



Given the facts of individual fluctuation and of heredity, 

 it is manifest that while nn.tural selection may produce and 

 enforce adaptation to conditions of life, the traits which dis- 

 tinguish these species bear little relation to utility. The 

 individuals which, separated from the main flock, people an 

 island, give their actu<al traits to their actual descendants, and 

 -the traits enforced by natural selection differ from island to 

 island. If external conditions were alike in all the islands the 

 progress of evolution would perhaps run parallel in all of them, 

 and the only differences which would persist would be derived 

 from differences in the parent stock. As some difference in 

 environment exists, there is {i corresponding difference in the 

 species as a result of adaptation. If great differences in con- 

 ditions exist, the change in the species may be greater, mors 

 rapidly accomplished, and the characters observed will bear 

 a closer relation to the principle of utility. 



Doubtless, wide fluctuations or mutations in every species 

 are more common than we suppose. With free access to the 

 mass of the species, these are lost through interbreeding. 

 Isolate them, as in a garden, or an enclosure or on an island, and 

 these may be continued and intensified to form new species or 

 races. Any breeder or any horticulturist will illustrate this. 



It is not claimed that species are occasionally associated 

 with physical barriers, which determine tlioir range, and which 



