xlvi 



One set of these facts is exemplified by the well-known case of dis- 

 tribution of insects between the east and west in the southern part of 

 our own island. I am not aware that comparative lists have yet been 

 published, but it will not be disputed that many hundreds of species 

 of Coleoptera, for instance, are known in the east, many of them 

 abundant, which are totally unknown in the west, and a smaller 

 number are known in the west which are not found in the east. In 

 cases like these a difference of climate may be the cause of the 

 limitation. But there is another set of facts requiring quite a 

 different explanation : this is the limited ranges of closely- allied 

 species in the plains of Tropical America. I have already elsewhere 

 recorded the fact that, in the forest plains of the Amazons, where 

 there is no difference of level worth mentioning, and no physical 

 barriers, the species of a large number of genera are changed from 

 one locality to another, not more than 200 or 800 miles apart. This 

 is most distinctly marked on the Upper Amazons, where the country 

 may be mapped out into areas of a few hundred square miles each, 

 every one containing numerous species of such genera as Ithomia, 

 Meliniea, Eubagis, Doryphora, Erotylus, &c., &c., allied to but quite 

 distinct from their representatives in the others. From what I have 

 seen of Mr. Buckley's collections on the eastern side of the Andes 

 I think the same limitation of areas must occur there also ; and 

 judging from the few species I know as coming undoubtedly from the 

 Guayaquil side of the Cordillera, the butterfly faunas of these areas 

 in the uniform country of the east are pretty nearly as distinct from 

 each other as the species east of the Andes are distinct from those 

 west of the mountains. We here again feel the want of facts, such 

 as Mr. Buckley collected, but which have not yet been published, 

 to teach us exactly what species are found east and what west of 

 the mountains, and how the great multitude of closely- allied spe- 

 cies are distributed in the narrow tract explored on the east. My 

 own observations in the level plain a few hundred miles further east 

 show distinctly, however, that the most effective possible barriers 

 are there opposed to the spread of hosts of species without any 

 physical barrier existing which is perceptible by our senses. The 

 explanation of the fact, T believe, is this, that there really are 

 subtle differences of physical conditions from place to place, even 

 in a uniform region ; slight differences in soil, humidity, succulence 

 of foliage, and so forth, which require in each area a re-adjustment 

 of the constitution of any new immigrants from adjoining areas ; 



