No. 496] ASPECTS OF THE SPECIES QUESTION 239 



of plants which differ from those of other areas, either 

 contiguous or widely separated, and, in cases where 

 types inhabiting different areas so characterized are ap- 

 parently similar, though different, their separation or 

 isolation has been given weight in regarding them as 

 specifically distinct. No doubt this is a rational course 

 to pursue if it is not carried to extremes. The question 

 whether the environments to which the ancestors of such 

 types have been exposed have been the cause of their 

 differentiation, or whether the elementary species have 

 been perpetuated which were best adapted to the soil, 

 climate or other features of the environments, is one of 

 the most interesting of unsolved problems. That similar 

 types have for the most part come from common ances- 

 tors we must regard as most probable, even if now in- 

 habiting widely separated regions, segregated by the 

 disappearance of related types in intervening space, be- 

 ing thus remnants of the more general distribution of 

 the ancestral forms in earlier geologic eras. 



Geographic distribution must, however, in cases of 

 contiguous land districts, be cautiously used as a deter- 

 mining factor. There are many instances in which a 

 species with certain well-marked characters in one region 

 is, apparently, at least, completely connected through 

 intermediate characters with what is readily regarded as 

 a perfectly distinct species in another region. Instances 

 of this kind are within the experience of every one who 

 has given attention to geographic distribution of plants. 

 I say this is apparently the case ; the conclusion is based 

 on long series of herbarium specimens and on field obser- 

 vations made over large areas of country. Neither of 

 these methods of information is wholly satisfactory, 

 because the herbarium series must necessarily be limited 



observations have to be taken at different times and usu- 

 ally at widely separated intervals. Still, the consensus 

 of opinion of plant geographers leans strongly to the 



