I30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I43 



western South America. It will be noted that none of these species 

 shows a very high degree of intraspecific gregariousness. 



The sooty-capped bush-tanager, which is an important passive nu- 

 clear species but which does not show a high degree of intraspecific 

 gregariousness like the brown-capped bush-tanager and the golden- 

 crowned warbler, is also confined to the mountains of Panama and 

 Costa Rica. 



It is possible, therefore, that the brown-capped bush-tanager and 

 the golden-crowned warbler are more widely distributed than most 

 of their associates, not so much because they are passive nuclear 

 species in mixed flocks as because they are both highly gregarious 

 among themselves. All other factors being equal, a high degree of 

 intraspecific gregariousness probably is (or was until recently) rela- 

 tively much more advantageous to most montane species than to most 

 lowland species in most regions of tropical America. 



The forest and scrub areas inhabited by most lowland tanagers 

 and finches were probably nearly continuous, or interrupted by only 

 relatively small areas of other habitats, over the larger part of tropi- 

 cal America before human settlement became very dense (see also 

 comments below). Most of the lowland tanagers and finches of the 

 American Tropics probably occupied all or a large part of their ranges 

 more or less gradually. Individuals of many of these species must 

 have been able, in many cases, to move into new areas previously un- 

 inhabited by their species without going very far from the areas in 

 which they themselves were raised. Thus, even a single individual of 

 a species that did not show a high degree of intraspecific gregarious- 

 ness was probably often able to obtain a mate, when it moved into a 

 new area, by attracting one from an adjacent area previously occu- 

 pied by the species, or because other individuals of the same species 

 would tend to stray into the new area on their own initiative. In such 

 circumstances, many lowland species that did not show a high degree 

 of intraspecific gregariousness were probably able to spread as easily 

 as, or more easily than, many lowland species that did show a high 

 degree of intraspecific gregariousness. 



The original situation of the montane species in tropical America 

 must have been very different. Areas of montane forest and scrub 

 have probably always been rather scattered in much of Central and 

 South America, and many of these cannot have been occupied by a 

 process of gradual diffusion. In order to reach such areas, individuals 

 of montane species must have had to make long jumps over wide ex- 

 panses of unsuitable habitats. Individuals of montane species that 



