122 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I43 



(Hardin, i960), has been questioned or denied (see, for instance, 

 Cole, i960) ; but it does seem to be a valid generalization in the case 

 of many species of birds (see, for instance, Lack, 1944 and 1947, and 

 Moreau, 1948). All other factors being equal, species that are closely 

 related to one another phylogenetically probably tend to compete with 

 one another more strongly, in most cases, than species that are dis- 

 tantly related to one another, simply because the habits of closely re- 

 lated species (including their feeding habits) are usually more sim- 

 ilar than the habits of distantly related species. 



Although closely related species that occur in the same areas and 

 habitats must almost certainly differ from one another in some aspects 

 of their ecology, the ecological isolation between them can only be 

 partial. For this reason, and because such species are usually or always 

 similar to one another in many other characters, the ecological differ- 

 ences between such species probably tend to disappear in certain cir- 

 cumstances. Unless strongly reinforced by other factors, the partial 

 ecological isolation between such species is almost certain to break 

 down or become increasingly ineffective if the species should begin to 

 associate with one another increasingly frequently and closely — 

 as they are bound to do if they become incorporated in the same 

 highly integrated type of mixed flock. There will thus be strong selec- 

 tion pressure, in many or most cases, to prevent closely related species 

 from associating with one another very frequently and closely and/or 

 to increase and strengthen the ecological differences between closely 

 related species when they do associate with one another. Thus, highly 

 integrated flocks that regularly include two or more closely related 

 species are relatively rare (among passerines) ; and when they do 

 occur, the closely related species tend to have very distinctly different 

 feeding habits (e.g., the Parus species in mixed flocks in European 

 woodlands — see Hartley, 1953; Gibb, 1954; and Snow, 1954). 



Among the tanagers, honey creepers, and finches of mixed flocks in 

 Panama, the most conspicuous examples of two closely related species 

 frequently occurring in the same mixed flocks is provided by the palm 

 tanager and the blue tanager. These two species coexist over a large 

 part of tropical America; but they have slightly different habitat 

 preferences, and the blue tanager seems to be slightly more insectivo- 

 rous than the palm tanager (at least around the clearing on Barro 

 Colorado Island). It is probably also significant that the palm tanager 

 and the blue tanager are more often separated by their different habi- 

 tat preferences during the breeding season, when competition be- 



