500 CORRELATION OF DISTRIBUTIONAL FEATURES. 



We may not, therefore, look for any of the factors controUing the dis- 

 tribution of the Desert region in these features of the temperature. 



The amphtude of the number of days in the longest normal dry 

 period of the frostless season is also great for the Desert (from 127 to 

 283). This is a case, however, in which even the minimum value for 

 the Desert indicates dry conditions, and the maximum value signifies 

 conditions of extreme aridity. The lower values of this condition may 

 be of importance elsewhere in differentiating vegetation, and the 

 various values within the Desert area may be of importance in connec- 

 tion with the minor vegetationa,l differences of that diversified region. 

 The position of the isoclimatic lines of plate 51 does not indicate that 

 any particular intensity of this condition is critical in limiting the 

 Desert region, which is in accord with the evidence of figure 27. The 

 case of the daily mean evaporation is a very similar one. 



The number of days in the longest normal rainy period in the frost- 

 less season ranges, in the Desert region, from its minimum value of no 

 days at several stations to a maximum of 10 days. The isoclimatic 

 line of 25 days embraces the entire Desert region and portions of the 

 Semidesert, the Grassland, and the western section of the Northern 

 Mesophytic Evergreen Forest. This factor appears to be an impor- 

 tant one and it is possible that numerous and adequately distributed 

 stations would show that the isoclimatic line of 10 days lies near the 

 boundary of the Desert area. 



The three derivations of the moisture ratio (plates 58, 59, 60) are 

 very similar here, as they are in all of the vegetations and other botani- 

 cal areas. The narrow amplitude of this condition for the Desert area 

 suggests an importance which is borne out by an examination of the 

 isoclimatic lines. The line representing values of 0.20 for the moisture 

 ratio is roughly similar to the outline of the Desert, and suggests that 

 the isoclimatic line of the ratio 0.25 might be a still closer approxima- 

 tion to the limits of the Desert. The area with values below 0.25 

 would then include more of western Texas, would separate the MogoUon 

 Plateau of Arizona from the Rocky Mountains, and would extend 

 somewhat further north, thereby coming into still closer agreement 

 with the outlines of the map of generalized vegetation. The bend of 

 the isoclimatic line for 0.20 which extends through the Tehachapi Pass 

 into the San Joaquin Valley of California is in accord with the pro- 

 nouncedly desert character of this valley, which has been included in 

 the composite Semidesert Area. 



All the evidence which we have been able to bring out in this case 

 points to the moisture ratio as being the climatic condition of most 

 importance in determining the boundaries of the Desert region. 



Semidesert (fig. 28). — The Semidesert embraces two areas — one in 

 California and one in Texas. These areas are not only diversified 

 within themselves, but are somewhat dissimilar in their vegetation and 



