352 VICTOR E. SHELFORD. 



the area of optimum. A study of the ecological distribution of 

 the animals thus located gives no correct idea of the ecological 

 optimum, but makes the relations of the species to conditions 

 seem particularly variable. From the point of view of ecological 

 generalization, data on distribution under agricultural condi- 

 tions are of questionable value. 



6. Development of Ecological Classification. 



The separation of animals into marine, fresh-water, and 

 terrestrial has long been practiced; such a classification has 

 about the same significance as the division of animals into 

 vertebrates and invertebrates. Some other divisions have been 

 recognized but usually rather loosely and little could probably 

 be added by the study of literature w r hich exists. A few of the 

 recent attempts at ecological classification deserve mention. 

 Three principal classifications by zoologists apparently not in 

 close touch with progress on the plant side must be noted. 

 Morse ('08) divided the Orthoptera into geophiles and phytophyles 

 representing certain "structural adaptations" (to strata). Both 

 of these main groups he further divided into xerophiles and Jiygro- 

 philes, and each of these in turn into campestrian and sylvan, etc. 

 The classification is dependent primarily upon strata. The 

 factors involved are unanalyzed and the scheme fails to distin- 

 guish the differences in physical conditions, which will probably 

 come to be the basis for all ecological classification. For ex- 

 ample, shrub-inhabiting species are not divided into those in- 

 habiting thickets exposed to sun and atmosphere in the open and 

 those inhabiting thickets or shrubbery in the shaded forest. Two 

 groups wholly unlike in their relations to physical factors are 

 thus put together. Shull ('n) points out the failure of the 

 classification in practice. 



Hancock ('n) follows a plan similar to that of Morse but 

 makes it much more complete, on the whole, contributing much 

 to the knowledge of ecological distribution of the Orthoptera. 

 He fails, however, to separate primary (primeval) conditions from 

 secondary (agricultural and human) conditions, and like Morse, 

 has made divisions primarily upon the basis of strata or 

 levels, with only partial consideration of physical factors or 



