224 
W. A. CANNON 
been accomplished on experimental and observational lines, however, 
to suggest the possible importance of the root relation in the distri- 
bution of species. The basis of the significance of the root factor in 
this connection lies in two general features, namely, in the root char- 
acter itself and in the manner of response of roots to the soil environ- 
ment. In the former case, especially in obligate deeply penetrating 
roots, the limiting factor appears to be only the depth of the soil. In 
such species, on the other hand, as have generalized roots and roots 
which are essentially shallowly placed, the condition is different. 
The limiting factors here relate to root response to such environmental 
soil factors as moisture, aeration and temperature. 
From the position here taken, other distributional factors being 
equal, the species having the most plastic roots, and roots capable of 
the most catholic response to the soil environment, or both, should 
also be the most widely distributed, and conversely, species which are 
sharply limited in root growth by soil conditions, or whose roots are 
not plastic, should also have the most restricted distribution. So 
far as observations show the true condition, this conclusion is valid. 
From this it will appear at once that such factors as depth of soil, and 
soil moisture, aeration and temperature, are of the greatest importance. 
In addition there should be mentioned the exposure which, as already 
shown, greatly affects the temperature of the soil. It is of interest to 
note, also, that all of these, with the possible exception of the last 
named, are factors which are not commonly evaluated in a definition 
of a habitat and which do not immediately affect the shoots of plants. 
Desert Laboratory, Tucson, Arizona. 
