202 
FORREST SHREVE 
winter temperature phases mentioned are of paramount importance 
as limiting factors. 
It is impossible to speak authoritatively of the importance of 
the temperature conditions of the growing season or those of the frost 
season in their relation to vegetation as a whole until we posess a 
large body of data regarding their influence on individual species. 
It is fundamentally important, furthermore, that our knowledge of 
the influence of these and other physical factors should be secured 
by observation of the distributional limits and behavior of species, 
by instrumentation, calculated to single out the critical factors and 
their critical intensities, and finally by investigation, in the laboratory, 
of the operation of these factors and of these intensities. Only in some 
such manner as this is it possible actually to interpret the underlying 
causes of the phenomena of distribution. Thus only can the observed 
or the instrumentally determined correlations of the field be given 
confirmation. Thus only can physiological plant geography place its 
generalizations on a secure logical basis. 
The Desert Laboratory, 
Tucson, Arizona. 
