EESEAECH METHODS IN STUDY OF FOREST ENVIRONMENT. 145 



are the " Psychrometric Tables " of the United States Weather 

 Bureau, contained in its Bulletin 235. These have been worked out 

 for barometric pressures of 30, 29, 27, and 23 inches. In accordance 

 with American custom, vapor pressures are given in inches of 

 mercury. Through the courtesy of the Weather Bureau, it is possi- 

 ble to produce in the Appendix an additional Table of vapor pres- 

 sures for a mean barometer of 21.42 inches, prepared by B. C. Kaclel 

 for the special use of the Wagon Wheel Gap Experiment Station, 

 at an elevation of 9,300 feet. This table will doubtless be of con- 

 siderable assistance in ecological studies in the western mountains. 



The vapor pressure may also be determined very quickly and 

 precisely by means of dew-point apparatus and a table of satura- 

 tion j^ressures corresponding to various temperatures. This appa- 

 ratus is, however, far less convenient for field use than the 

 psychrometer. 



The ideal record of humidity is, of course, one which shows the 

 atmposheric condition for every hour of the day. Theoretically, this 

 is obtained by the use of the hair hygrograph; but, actually, the 

 instrument is of very little use. 



The atmospheric conditions are measured in terms of relative 

 humidity, which fluctuates rapidly with every change in air tem- 

 perature. The record must then be transposed, in conjunction with 

 the continuous temperature record, into terms of absolute humidity 

 and saturation deficit, before it can have much value. Furthermore, 

 the hygrograph is probably the least reliable and accurate of the 

 automatic instruments commonly used. 



Since the absolute humidity or vapor pressure usually does not 

 change through a wide range in a short time, but shows a general 

 tendency to increase as the air warms and to decrease with the 

 cooling at night, it is possible to determine a fairly satisfactory mean 

 humidity for any day (except of course during general disturbances) 

 by means of two or three observations with the psychrometer. For 

 example, the hours of 7 a. m., 1 and 7 p. m., have been used, or 7 a. m., 

 2 and 9 p. m. After hourly observations for a few days at any 

 season a in I point, it should be possible to select one or more con- 

 venient hours when, in the ordinary sequence of events, the mean 

 humidity of the day may he approximately measured, cither at each 

 observation, or through averaging unlike valuations. As has been 

 suggested, the absolute humidity varies less than the relative humid- 

 ity or saturation deficit. Therefore, lor calculating the mean sat- 

 uration deficif for the day it is logical to arrive first at the mean 

 vapor pressure, and then, niter calculating the mean temperature 

 for the whole period, to ohtain the saturation deficit by deduction. 



ree 22 to 



