122 BULLETIN 1233, U. S. DEPARTMENT OF AGRICULTURE. 



There are, then, several ways in which the availability of soil 

 moisture may be more or less precisely determined. The most 

 precise method is undoubtedly to determine the actual osmotic 

 pressure shown by the soil water when it is present in various amounts 

 or percentages. This of course implies somewhat exhaustive study 

 of each soil whose moisture condition in the field is a matter of 

 interest. In the present study a single approximate determination 

 of osmotic pressure is available for each 01 only about half the indi- 

 vidual soils w T hich are of interest. Tins determination was made 

 by. the vapor-transfer method under conditions far from ideal. 



Another method is that of comparing the current moisture per- 

 centage for each soil with that percentage which represents the 

 completely nonavailable moisture. This, of course, must have been 

 determined by a wilting test in connection with each soil, or approxi- 

 mately through some such physical measurement as the moisture- 

 equivalent or capillarity determination. As fully explained in 

 "Research Methods in the Study of Forest Environment," the 

 "availability" is expressed by the ratio of the available moisture 

 to the whole. This gives a quantity always less than unity, which 

 has a certain logical value, as — 



, M- WC 



in which M represents current moisture and WC the wilting coefficient. 



Or, finally, granting that within a single soil type there may be a 

 fairly constant relation between the water-holding properties as 

 determined mechanically and the water-wdthholding properties in the 

 osmotic sense, the current moisture may be related to the moisture 

 equivalent of each soil, and a purely arbitrary ratio may be obtained 

 for purposes of comparison. This method is almost certain to be 

 misleading if radically different types of soil are involved in the 

 comparison. 



Method of securing soil-moisture data. — During 1910 and 1911, when 

 only three forest sites were being studied at the Fremont Experiment 

 Station, the attempt was made to secure samples of the native soil 

 at frequent intervals in the usual manner, namely, by means of a 

 tubular soil borer. These samples were usually taken in triplicate at 

 points near the centers of the respective stations, and, like all other 

 samples considered in this study, were dried in a water-bath oven at 

 the temperature of boiling water, which at the Fremont Forest Experi- 

 ment Station is approximately 92° C. Until 1915, drying for 8 hours 

 was considered adequate, in view of the small size, coarseness, and 

 mineral character of the samples; but later this period was extended 

 to 24 hours, and the water in the bath was kept boiling for at least 

 the last 5 or 6 hours of the period. No allowance was made for 

 varying vapor pressures, which, being generally Low, probably made 

 the drying as complete as though brought about by a higher tempera- 

 ture in a more humid locality. 



This plan for sampling the native soil at the several stations was 

 qoI satisfactory, because, as has been said, the soil of the Pikes 

 Teak region in most situations and except for a few inches at the 

 surface, is disintegrated granite in situ, and is not only very unyield- 

 ing to penetration but almost entirely lacking in cohesive properties 

 when once loosened. Hence it required, in the first place, great 

 labor to sink the soil borer, and. in the second place, great ingenuity 





