230 Bulletin 37. 



Changes from April 12 to June iS. 



The set of samples taken June 18, 1899, showed that marked 

 changes had taken place in the moisture content of the respective 

 feet of the different soil strata. The greatest change was in the 

 upper five feet, this part of the soil having lost over half of the 

 water it contained April 12. This great change was due, evi- 

 dently, to several factors — evaporation, percolation, and the use 

 of water by the trees. The stratum above the gravel is the only 

 one that could lose water by capillary action and evaporation. 

 There was probably no great amount of percolation from this 

 stratum after taking the first set of samples, April 12, as two 

 weeks had already intervened since irrigation ceased. The great- 

 est loss of w r ater was undoubtedly from its use by the trees, as 

 the surface mulch was intended to, and probably did, prevent the 

 rapid loss of water from below the first foot. The four feet ly- 

 ing between the first and the sixth foot contain a much larger 

 number of roots and rootlets than any stratum below, and the 

 withdrawal of water by the trees would be much more rapid from 

 this stratum than from any other. Furthermore, the trees made 

 nearly all their growth between April 12th and June 18th. 

 Hence it is not surprising that during this period the upper five 

 feet lost over half the water contained upon the former date. 



The next greatest change occurred in the upper two feet of 

 the clay below the gravel, the water that had been received from 

 the gravel above more rapidly than it could percolate downward, 

 having had time to settle. Where part of it went will be indi- 

 cated by a reference to the tables and to Fig. 7, which show that 

 between the 20th and the 26th foot the moisture content actually 

 increased during the summer. By Table II it will be seen that 

 the net loss of water from each column of 25 feet, from April 1 2 

 to June 18, was approximately 82 pounds — equal to a layer of 

 water a little over 15 inches deep. Of this amount the upper five 

 feet lost over 50 pounds, or nearly two-thirds of the total. 



Changes from June to September. 



During the three and one-third months that intervened be- 

 tween the taking of the samples June 18 and September 30, the 



