RESEARCH METHODS IN STUDY OF FOREST ENVIRONMENT. 157 



somewhat according to atmospheric conditions by regulating the 

 hydrostatic pressure. With the outside walls of the cup always 

 moist and yet not dripping, the rate of evaporation will of course 

 be governed by atmospheric conditions. It must not be expected, 

 however, that the evaporation from this instrument may be com- 

 pared under a variety of conditions with that from the Piche in- 

 strument, or with that from a free-water surface. While the ab- 

 sorption of heat from radiant sources and conduction from the air 

 will be practically the same for the water surfaces in the three 

 cases mentioned, yet the further absorption beyond the first water 

 surface will depend on the nature of the substance behind that 

 water surface— in the one case, water; in the second, paper and glass; 

 in the third, clay or some similar earthy substance. Therefore,, 

 the three instruments will respond quite differently to the stimuli 

 of warm air and sunshine. 



For these reasons, comparative data will be of value only when 

 the same instrument is used in all measurements of the comparison. 



Skive's nonabsorbent porous-cup atmometer. 



It has been the experience of various investigators that the Liv- 

 ingston porous cup atmometer measures the evaporating pow T er of the 

 air with a very considerable degree of accuracy during periods 

 when the temperature is not recorded at or below zero centigrade. 



In 1910, Livingston (158) described a rain-correcting atmometer. 

 This atmometer, while giving great satisfaction in the hands of 

 many inexperienced workers, was difficult to operate in some locali- 

 ties. Thus it was found impossible to obtain continuous records in 

 the dry climate of the Wasatch Mountains of the Manti National 

 Forest in central Utah, principally on account of the connections 

 and joints of the equipment, all of which occurred outside of the 

 water reservoir. Hail storms and objects carried by strong wind 

 were often so severe as to disjoint or break the more delicate equip- 

 ment. In connection with this instrument, it is understood that the 

 automatic mercury valves which operate to prevent the water ab- 

 sorbed by the porous cup in times of rain from entering the reservoir 

 are externally situated. This makes it essential to have all valves 

 very tightly connected in order to prevent leakage. Much to the 

 satisfaction of those who have used this instrument, Shive (165) 

 has described one so modified as to be self-contained, and at the 

 same time to reduce to the minimum the liability of breakage and 

 the difficulty of adjustment. This was accomplished by eliminating 

 jointed mercury valves and decreasing the leakability and breakage 

 to a nominal degree. The arrangement of the different parts of the 



