POE Sa ee ee hg Pe SERED ee eee MBE OL ig gee Se BETES Si NR eae Tat SOI De NP RRR pCR ae Decree AEM See ease ash ages fr ULES LAT Ne ea a eee Tea Neo Se ROE EE ES Ee Eig Seen Oat ie Pp eee ee 
Igto] LIVINGSTON—SOIL MOISTURE 251 
line of these graphs. They express percentage of dry weight. 
At the base of each figure, the dates of sampling are indicated. 
The major portion of the calculations and of the drawing of the 
graphs are the work of Grace J. Livincsron. 
DISCUSSION OF GRAPHS.—It is to be remembered that several 
simultaneous samples at the same station would be expected to 
vary to some extent, especially in the stony soil of the hill andof the 
deeper layer of the slope. Therefore it is not surprising to note 
minor fluctuations in the graphs, which do not appear traceable to 
conditions of precipitation. 
The graphs bring out clearly the lagging of soil moisture behind 
precipitation; it requires considerable time for rain water to reach 
the upper level of sampling and a still longer time for it to reach 
the lower level. A slight precipitation may not alter the condi- 
tion of the soil at either level, the water being lost by evaporation 
before it can penetrate even to a depth of 10 or 15 cm. It also 
occurs that a rain which is only sufficient to moisten the surface 
layers, thus placing them in capillary connection with the deeper 
ones which are already moist, may actually accelerate the drying 
out of the deeper layers. Thus a slight rain is sometimes directly 
deleterious to the vegetation; it removes the dry mulch from the sur- 
face and fails to add markedly, if at all, to the deeper moisture con- 
tent, so that before the dry mulch can regenerate, the deeper soil 
has lost considerable water by upward movement and evaporation. 
From data given in Publication 50 of the Carnegie Institution 
(Pp. 66, 67) it appears that a number of plants of the hill, when 
grown in pots (in hill soil), wilted with a soil moisture content of 
6-12 per cent by wet volume. On the basis of dry weight these 
numbers become about 7 per cent and 14 per cent. Boerhavia 
(one of the summer annuals of the hill) wilted in the open soil with 
4 moisture content of 6-7 per cent by wet volume, or about 7-8 
per cent on the basis of dry weight. These observations were 
taken in July and August, the period of the summer rains, when the 
€vaporating power of the air was far below its magnitude of May 
and June or of October and November. With greater evaporation 
Tate, the plants should of course wilt with a correspondingly 
€r soil moisture content. It is probably safe to assume, there- 
