422 PROCEEDINGS OF THE ACADEMY OF [1896. 



ment if not too intimately complicated with others, will produce in 

 organisms of the same general nature similar results wherever situa- 

 ted ; and, lastly, that the resulting features strikingly similar 

 though they may be, are, conversely, no evidence of ontogenetic 

 relationship. In any census undertaken with a view to determin- 

 ing systematic relationship, such characters must be eliminated in 

 order to avoid an erroneous conclusion. 



It is only by close and minute study of the details of the situs of 

 species and of their minor, though by no meaus unimportant, char- 

 acters of form and surface, that we shall be able to recognize those 

 features which may be classed as dynamic as opposed to those which 

 even if dynamic also in their ultimate origin, have become geneti- 

 cally constant. The noxious and stupefying multiplication of spe- 

 cific names, which has been characteristic of a certain school of 

 workers during the last twenty years, could never have gained scien- 

 tific recognition had there been any general appreciation of the 

 extent to which dynamic modifications affect all organisms. It is 

 much easier to describe and name a character than it is to search 

 out its reason for existence. It is even easier, with proper appara- 

 tus, to count the cells in an organism of moderate size than it is to 

 recognize and discriminate the influence of the environment upon 

 the organic total of those cells. By inspecting the fragments of a 

 building one may learn something of construction, but it is only by 

 contemplating it as a whole that the higher elements of architecture 

 can be recognized. 



Recognizing the imperfection and inadequacy of our knowledge, 

 even of the limited groups discussed in this paper, the writer thinks 

 that some glimmerings of light may be had on the subject of 

 dynamic characters from the accompanying study of insular land 

 shells. 



The following summary will express, tentatively, such of the con- 

 clusions as appear justified from the study of the specimens: 



A. Given a region of volcanic origin and mountainous character, 

 with local or seasonal aridity, more or less arboreal vegetation as 

 well as herbage and a tropical or nearly tropical climate, moderate 

 isolation and safety to propagate and increase. 



B. Into this region let land shells of the principal continental 

 types be introduced, and allowed the necessary time to become dis- 

 persed over the region, multiply abundantly and respond to the 

 ■environment. 



