CORRELATION OF DISTRIBUTIONAL FEATURES. 571 



conditions is relatively not nearly as great. If the humidity of our 

 most humid areas were to be increased (within the limits set by world 

 climate), little or no alteration in the vegetation would be expected. 

 Nor would any great alteration in the vegetation of our most arid 

 areas be expected from increased aridity, excepting that vegetation 

 would finally be prohibited altogether. On the other hand, if the 

 intensity or duration factor of the temperature conditions of our 

 warmest provinces were increased, or if that of our coldest provinces 

 were decreased, great changes in vegetation would be expected. Both 

 north and south of the area of the United States the same humidity 

 conditions are concomitant with very different vegetation characters, 

 and this difference is to be related mainly or entirely to temperature 

 differences. 



We present below some of the most definite cases of concomitancy, 

 considering first the temperature provinces, then the moisture prov- 

 inces, then the provinces based on the temperature-moisture product, 

 and finally the two-dimensional provinces based on temperature and 



moisture. 



2. TEMPERATURE PROVINCES. 



Two charts of temperature provinces have been employed for these 

 comparisons, the one based on the average frostless season (plate 34) 

 and the one based on physiological summation indices (plate 40). It 

 will be convenient to consider the comparisons in the order of the 

 vegetation features as these have been presented in plates 1 to 33. 



As has been mentioned several times, there is no primary correlation 

 between temperature conditions and the general types of vegetation 

 as shown by plates 1 and 2, and a comparison of these plates with those 

 of the temperature provinces emphasizes this statement once more. 



In the case of plate 3, if the evergreen broad-leaved trees and the 

 microphyllous trees are taken together as a single group (characterized 

 by relatively low transpiring power), it is found that the geographic 

 area occupied by this group very nearly corresponds with the area of 

 the very warm, warm, and medium temperature provinces as brought 

 out by the chart of the average frostless season. Near the Pacific and 

 Atlantic coasts the correspondence is not good when the chart of 

 physiological summations is employed. 



The two eastern palms, Sahal palmetto and Serenoa serrulata, occupy 

 nearly the same area as does the very warm temperature province on 

 either of the two temperature charts here employed. 



On plate 29, Phorodendron flavescens shows an area of distribution 

 that closely agrees with the form and extent of the combined very 

 warm, warm, and medium temperature provinces, as shown on the 

 frostless-season chart. As in the case of the broad-leaved and micro- 

 phyllous group of trees, the correlation is not good with the physiologi- 

 cal summation chart. 



