36 BRITISH BIRDS. [vol. viii. 



the more uniform and enduring the habitat conditions, 

 the more speciahzed and non-adaptable the dependent 

 fauna. Under relatively constant and undifferentiated 

 forest conditions evolve specialized forest-d\^'elIers, pro- 

 portionately dependent upon the existence of forests. 

 Species with a mixed inheritance, whose racial environ- 

 ment has through relative fluctuation or complexity 

 prevented specialization, thereby retain readier power of 

 adjustment to an altered or new environment. 



However indirect in control, the environmental influence 

 in bird-distribution is so far predominant as to suggest 

 the need for a systematic presentation of the facts from 

 this side. That classification which most successfully 

 arranges species in characteristic or mutually exclusive 

 groups, is based upon the most fimda mental or differen- 

 tiating factors. It \\ould be hopeless on this test to 

 attempt a grouping of British bird-distribution purely 

 according to climatic, or topographical, or geological 

 aspects — or even to food-supply. The efficient factor- 

 combinations in distribution, and the related species- 

 groups, are broadly recognized in the popular divisions 

 of Avoodland birds, moorland birds, aquatic, and so on. 

 The nesting-distribution of birds is closely correlated with 

 that of vegetation, and in minor degree, with surface 

 geological features ; and with the partial exception of the 

 food-supply, it seems probable that the influence of the 

 other factors is mainly indirect, effective through its 

 action on the vegetation. 



It is natural to relate the localization of many birds 

 to the question of food-supply. Nevertheless, there are 

 some instances of range-restriction which can hardly be 

 assigned to this cause ; and there appear to be sufficiently 

 good reasons for believing that \\hile the nesting-range 

 must necessarily be contained within environmental limits 

 which hold the necessary food-supply, it is in many 

 instances not co -extensive with the latter ; that is to say, 

 in such cases the food-supply is not the limiting factor in 

 distribution. Even in cases where it could be shown that 

 certain species whose distribution is markedly '' local," 

 derive their food from sources equally local, the question 

 would still remain : are these species restricted to such 

 areas because they depend largely or mainly upon certain 

 food -supplies thereto confined ; or do they utilize the 

 latter because they are characteristic productions of the 

 areas preferred by the birds for other reasons ? It is 

 necessary to assume that the differentiated environmental 



