NO. I DISTRIBUTION OF THE ONYCHOPHORA CLARK 7 



which the type as a Avhole can maintain itself, a total range which is 

 always duplicated within the tropics. 



Bird types which exist only in a great number of local forms cannot 

 be assumed to be living under optimum conditions for the type as a 

 whole. Such bird types, living always within tropical conditions, are 

 probably all of ultimately tropical origin, their progenitors having 

 gradually extended their range outward from the tropics with the 

 annual outward extension of the tropical conditions, and eventually 

 having colonized, though in the summer only, the temperate regions. 



To a type with a highly developed power of migration, such as 

 many birds, the temperate regions in the summer represent the 

 border of a tropical habitat, and thus we should expect to find such a 

 type occurring in the temperate regions in summer obeying the laws 

 of peripheral distribution of animal types in general. 



In the winter these migratory birds, in order to remain within the 

 economic range necessary for their existence, of necessity withdraw 

 within the tropics (where, as non-breeders, they are perfectly well 

 able to exist) there to remain until, with the advent of summer, the 

 tropical conditions are again extended. 



But in the tropics the sum total of the range of each type is dupli- 

 cated, and conditions are such that there is no closely circumscribed 

 local and ecological differentiation comparable to that which occurs in 

 the temperate regions. 



Therefore there is no compelling reason for the various races to 

 maintain their summer segregation, and a number of these races may 

 be found living together. 



Many of these bird types have breeding representatives in the 

 tropics, especially on isolated islands where the factors which, after 

 the summer colonization of the temperate regions, caused their extir- 

 pation as breeding residents from their original tropical home, have 

 not operated ; the non-breeding individuals of many others appear to 

 prefer always to remain within the tropics. 



These bird types within the tropics are secondary colonists, re- 

 turned to their original area of optimum conditions, where they are 

 able to exist as adults in a great number of closely related forms, but 

 where nidification, unless of a newly acquired highly specialized type, 

 or in especially protected localities, has now, thanks to the develop- 

 ment of certain enemies, become impossible. 



In any area in which the optimum conditions for a given animal 

 type are represented by the mean of the conditions under which that 

 type is able to maintain itself, the progressive development of that 



