SPRING PRECIPITATION AND HEIGHT GROWTH 687 



with precipitation over a long period of years in many localities, how- 

 ever, because of a lack of reliable meteorological records. 



At Fort Valley the relation established for a level situation has been 

 found to hold on north and south exposures. The relation is also 

 strongly evident in planted trees below lo years of age. Near Flag- 

 staff, where meteorological records cover a long period of years, test 

 measurements have been made on level sites and on east, west, and 

 south exposures, and in every case the same relation was obtained as at 

 Fort Valley. What is most remarkable is that almost the same graph 

 is secured with a small number of trees as with a large number, when 

 taken on the same site. 



Instances have been observed where the relation here pointed out 

 apparently does not hold. This is entirely in accord with what should 

 be expected, because what is the critical factor on one site may not be 

 at all critical on another site. In the upper altitudinal limits of the 

 tree's range there appears to be no well-defined relation between height 

 growth and precipitation. Here, on account of the relatively high pre- 

 cipitation and low temperature, moisture apparently gives way to tem- 

 perature as the controlling factor. The same observation has been 

 made in canyons where moisture is abundant. 



Another factor which may assume a position of dominance is soil 

 composition. At Fort Valley and over almost the entire San Francisco 

 Mountain region the soil is a heavy adobe, which is penetrated with 

 difficulty by plant roots, with the result that, as has been previously 

 pointed out, the pine has a surprisingly shallow root system. On such 

 soils trees are more sensitive to temporary drought than on porous 

 soils, on which the roots penetrate deeper ; in fact, it would not be sur- 

 prising to find that on porous soils spring precipitation is subordinate 

 to winter precipitation. 



The age of the tree is another factor to be considered. Observations 

 on mature and nearly mature trees of the size usually felled in lum- 

 bering operations have yielded no definite conclusions one way or the 

 other, for the reason that they have passed the stage of vigorous height 

 growth. In such trees one would be more apt to find a relation between 

 moisture and diameter growth, as has indeed been found by Douglass. 

 In a number of instances saplings lo feet or more in height have been 

 observed to show a different sequence of long and short internodes 

 from that shown by smaller saplings on the same site. This condition 

 is strikingly illustrated near the Fort Valley Experiment Station and 

 at Prescott, Arizona. The probable explanation is that the older sap- 



