40 BRITISH BIRDS. [vol. viii. 



upon a preliminary reconstruction of the country as it 

 existed prior to man's interference ; together \A'ith analysis 

 of the fragments of primitive groimd that remain, and their 

 associated bird-life. 



From their environmental relationships, the bird- 

 associations of this country from one point of view fall 

 into two groups — natural and '' artificial " — the former 

 relatively clear-cut and faimistically definite, its faunal 

 stability due to stable environmental conditions ; the latter 

 relatively nebulous and fluctuating, its characteristic insta- 

 bility referable in part to the inconstancy of the environment, 

 in part to imperfect faimal adjustment. The generic 

 association -groups, such as the "natural" ones of alpine, 

 moor, forest, grassland, aquatic, coastal belt ; or the 

 " artificial " ones of arable land, pasture, plantation, fall 

 respectively into sub-divisions or specific associations, the 

 faunal relationships of which nm jDarallel to the environ- 

 mental. How far the faunal correspondence between 

 environments of similar type is affected by dissimilarities 

 in other conditions : that is, how far environmental influence 

 in the naiTO\\' sense outweighs that of all other factors 

 combined, remains to be shown ; but it is clear that a birch- 

 wood and a cottongrass moor in southern England will show- 

 greater avifaimal resemblance to their respective Scottish 

 prototjqoes than to each other ; which implies that environ- 

 mental resemblance is more potent than climatic divergence ; 

 or, conversely, that climatic similarity is less influential 

 than environmental contrast. Or, to take a less extreme 

 comparison : given two similar environments with differing 

 climatic conditions, the respective avifaunas will correspond 

 more closely than in the comparative instance of two related 

 but specifically distinct environments with common climatic 

 conditions. A south of England birch-wood will show 

 greater, if incomplete, approach in avifauna to a Scottish 

 wood of similar type, than to a pine-wood in its own neigh- 

 bourhood. Here, so far as other conditions are equal, the 

 resemblances are due to environment, the differences to 

 climate. Bat it is evident that it is soon enough to look 

 to clunatic factors for an exj)lanation of faimal divergence 

 when it is shown that the envkonments are truly homologous. 

 In degree, as two envkonmental tyjDes are contrasted in 

 character, their dependent bird-associations are distinct ; in 

 degree as they approach each other in form, so do their 

 respective avifaimas ; and this remains broadly true whether 

 the climatic features are similar or contrasted. On the other 

 hand, divergent climatic features do not of themselves entail 



