ECOLOGY OF ISLE ROYALE. 123 



The quotations at the head of this article have an immediate bearing 

 upon the subject at this point. 



Not only is habitat preference, the association of avian species, their 

 succession, and the laws expressing these relations of much interest, but 

 they are of much importance scientifically as well as in a practical way. 

 It is therefore desirable that naturalists realize the necessity of under- 

 standing the "rules of the game" if the true relations at birds are to be 

 studied to the best advantage. No adequate substitute has yet been 

 devised to replace a grasp of general principles. 



Throughout this paper emphasis is placed on the definiteness of the 

 dominant major environmental influences and complexes because the 

 irregular features have apparently received undue emphasis and have 

 retarded the recognition of certain important definite relations. 



II. REPRESENTATIVE LITERATURE ON HABITATS AND SUCCESSION. 



1. Habitat Preference. The American literature on habitat prefer- 

 ence and succession, as a subject of special investigation, is very limited. 

 By succession is meant the change or replacement at a given place of 

 one or several species (an association) by others; as when a swamp is 

 invaded by a dune and the representative swamp birds are replaced by 

 those of the dune ; or even again when the dune becomes fixed by vegeta- 

 tion and is inhabited by still another association of bird life. This is 

 a much neglected subject; however, isolated observations on habitats 

 are abundant in the biographies of the various species. The fragment- 

 ary character of these biogi-aphies tends to make them composite and 

 they lose what pecularities they may have which are due to a response 

 on the part of the bird to its particular conditions of life. These un- 

 fortunate limitations clearly show that here is an extensive field worthy 

 of careful investigation. The work already done will be a useful guide 

 in many cases, but the student who wishes to develop this subject must 

 turn to the fields and forests rather than to the literature, both for 

 his inspiration and his data. 



Perhaps a further word should be added concerning the limitations of 

 the composite life-history method, as this will aid in making clear the 

 kind of work needed in the future. This composite or generalized method 

 of describing habitats and life histories and the response of birds to them, 

 tends to lay undue emphasis upon the average conditions of life and habits. 

 and tends to neglect those detailed responses to the environment which 

 reflect the laws of local influence. These results are similar to those pro- 

 duced by systematic students who are "lumpers" and who do not recog- 

 nize local races or varieties. Thus a nest may be built upon the ground 

 at the base of a shrub or bunch of grass, or in the brush, but what 

 conditions determine such sites ? In a dry meadow a Song Sparrow may 

 build directly upon the ground, but in a swamp, in order to have a dry 

 nesting site, it builds in a willow shrub. In many cases the causes of 

 these differences will be difficult to determine, but in others it is a 

 relatively simple question for any one familiar with the species to solve. 

 There are also geographic variations of habits as well as those of 

 habitats just cited, and for this reason it is necessary not to confuse such 

 variations with those confined to some restricted area. These local 

 and geographic relations are very intimately related, but they are sub- 



