226 PROCEEDINGS OF THE ACADEMY OF [Mcll., 



The range of individual variation in Ashmtinella among specimens 

 from one place is (with the exception of A. I. heterodonta) not greater 

 than in Polygyra. Among specimens I have measured or examined, 

 I have seen no lot which would yield a markedly bimodal curve were 

 the variations plotted. The variations between different colonies or 

 ■gens are often appreciable, sometimes conspicuous; but here also the 

 €ase may readily be paralleled in Polygyra, although usually not in 

 such restricted areas, for the reason that in the Polygyra country the 

 topographic and climatic features are less emphatic, and the life-zones 

 are not crowded upon one another as in the land of the Ashmunellas, 

 but are spread over larger areas. 



The conception of species in such sedentary animals as snails is far 

 from simple. A "species" comprises a multitude of colonies or com- 

 munities w^hich at any one time are isolated one from the other by the 

 existing topographic and other surface features of the country. This 

 is and always has been the case, even with the common, widespread 

 forms of the more level parts of the country; but the colonies there have 

 always been subject to frequent mixture with their neighboring colo- 

 nies, by the ever slightly fluctuating conditions of woodland and local 

 moisture, so that their network over the country has been here and 

 there made practically complete within comparatively short periods. 

 As a consequence, we have in many cases no tangible difference 

 between individuals from colonies hinidreds of miles apart. 



In regions where the local physical features are more accentuated, 

 the colonies or communities are often less subject to mixtiu-e. More- 

 over, the range of conditions within a limited area is far greater. Thus 

 snails of the same original stock living in the rocky talus on opposite 

 sides of a canyon are often subject to very diverse conditions of heat, 

 moisture and consequently cryptogamic food. They are often wholly 

 unable to cross from one side to the other by reason of a wide, freshet- 

 swept or arid space. Moreover, subsequent changes, such as the for- 

 mation of lateral canyons and the localization of suitable stations in 

 the talus, tend to further isolate the several colonies, and to preserve 

 their individuality for long periods. 



Thus each colony follows its own bent; and differentiation ensues, 

 cither by the cmnulation of organic changes induced by varying 

 conditions of growth and nutrition, determined by the local 

 enviromncnt as mentioned above, or by the occurrence of diverse 

 "mutations " in the several colonies, or by both causes. My idea of the 

 practical isolation of snail colonies is based upon the experience of 

 many years. Similar views have been expressed by Hemphill, in the 



