126 MICHIGAN SURVEY, 1908. 



by a given species or some natural group is also an important and 

 neglected method of ecological study which possesses certain important 

 advantages. As an illustration of this method may be mentioned 

 Palmer's ('00) study of the Maryland Yellow-throat. He has shown 

 that different varieties haive different habitat preferences. Jacobs ('04) 

 has given us an interesting habitat study of a single species in Pennsyl- 

 vania, the Golden-winged Warbler. 



Let us now turn to another class of habitat studies, those which 

 through man's influence throw only a subordinate light upon "natural" 

 habitats and succession, and are primarily of economic importance but 

 contain valuable habitat data. 



An interesting and rather unique paper belonging to this class, based 

 on observations in Southeastern Michigan, is by Watkins ('00). It is 

 entitled 'Michigan Birds that Xest in Open Meadows.' A few of his 

 statements explain his point of view: "To make more plain the limit 

 and scope of this treatise, which, of necessity must be longer than I 

 hoped, I will include in my list only such species as I have found nesting 

 upon the ground in the open fields and meadows, excluding those found 

 nesting upon the boundary fences or ground; alsO' those nesting in the 

 open marsh land which are undrained and boggy to the extent of being 

 unfit for hay or pasture" (p. 67). The paper contains numerous notes 

 on the habitat preference and variations in these ti^aits. 



By far the most comprehensive and thorough study of any limited 

 farm area is that by Judd ('02) of a farm in Maryland. In this paper 

 habitat preferences are clearly recognized, and discussed rather fully 

 (pp. 12-20). The birds are associated thus: — birds that nest in the 

 open fieWs, birds that depend on covers, birds of less limited distribution 

 (consists largely of remarks on haunts), and birds of varied distribu- 

 tion. His last two sections are rather miscellaneous in character and 

 show that the principles of classification for habitats were not clearly 

 defined in his own mind. 



The only other paper discussing habitat preference in detail is also 

 the latest upon the subject, and is by Forbes ('07). This is a pi-e- 

 liminary report on a bird census across the corn belt of Central Illinois 

 in the early autumn; a study of the feeding grounds and preferences as 

 influeuiced by the dominant crops of the areai traversed, corn, pasture, 

 and stubble. By means of this census, the habitat preferences for differ- 

 ent crops and the association of species in them is statistically deter- 

 mined. The paper is particularly suggesti\e for its bearing on the sub- 

 ject of dominance; however, the suggested method of study has even 

 greater significance when applied during the breeding season. Doubt- 

 less opinions will vary as to the validity of the method as applied by 

 Foi'bes, even by those who would approve of it for the detailed study 

 of a limited area, or a breeding habitat. For large areas some co- 

 operative mlethod may be necessary. 



2. Succession. Turning now to the literature on succession, it is 

 found to be extremely limited in amount. So far as known toi the 

 writer, only two American authors seem to have realized the existence 

 of succession. In his discussion of the biotic succession in the Por- 

 cupine Mountains of Michigan, Ruthven ('00) clearly included the birds, 

 although they did not receive separate treatment, and might for this 



