POSITION OF ECOLOGY IN MODERN SCIENCE 187 



causes determining the different areas of distribution, by an investi- 

 gation of the life-history of our present environment, or, when this 

 fails to give the required explanation, find out, by patient research 

 among the geological records of the past, the locations where ana- 

 logous conditions must have prevailed. 



The essence of geography is the endeavor to acquire knowledge 

 of the distinctive features of the earth, and for this purpose all the 

 data of related sciences must be collected. Since the organic sciences 

 are constructed from separate bits of knowledge, they must take each 

 fundamental element into account, that is, consider each individual 

 species according to its inherent qualities and external form. The 

 tendency of ecology to accumulate minor details obstructs the broad 

 view 'towards which it has painfully toiled, until the introduction 

 of the freer methods of geography aid it in uniting the results of 

 divided effort. We have here before us an excellent example of the 

 intimate connection between two branches of science, the one being 

 the complement of the other and acting as a stimulus to further 

 investigation. 



The attempt has been made to introduce into ecology various 

 phrases which shall have a definite, geographical application, but 

 in order to do this, due justice has not been done to existing facts. 

 The phrase recently used, "Climate creates a flora, soil determines 

 the formation/' seems to me to be unfavorable for the advancement 

 of future research along those lines where climate and soil furnish 

 a causal explanation. 



Climatology especially must be restricted within its proper limits, 

 where, however, it must be given all its privileges, as, for instance, in 

 the classification of continental zones. It seems to me entirely wrong 

 to allow the undue influence of a certain factor to establish unnatural 

 scientific divisions in a subject whose inherent worth lies in the cor- 

 relation of the most heterogeneous data. The defects which arise from 

 the use of a single morphological characteristic have been recognized 

 in systematic classification, so that here, where the investigator must 

 deal especially with the complexity shown by determining factors, 

 the giving prominence to one factor alone will enable him to illumine 

 only a small portion of the field and will give him but imperfect 

 means of taking his bearings. 



In the determination of both large and small divisions the natural- 

 ist is ever haunted by the same questions : for which forms of vegeta- 

 tion, as well as for their union into formations, is each district best 

 adapted, and by what vital conditions does it differ from neighboring 

 districts? 



It is strange that until now the ecological branch of animal geo- 

 graphy has almost neglected this, to me, most interesting side of the 

 question. Yet the periodic phenomena of animal life, so often con- 



