ATMOMETRY AND THE ATMOMETER 99 



periods utterly destroy the value of the instrument as one for 

 measuring the evaporating power of the air. As has been 

 stated, the absorbing mounting above described is suitable only 

 where rain and heavy dew do not occur; for of course dew formed 

 on the cup is absorbed just as is rain. In moist time's, of low 

 evaporation rates, a shower lasting less than an hour may inject 

 into the reservoir more water than is removed by evaporation 

 during the entire remainder of the day. Hence the evaporation 

 rate for that day, although evaporation actually occurred for 

 almost the entire time, cannot be determined from the instrument. 



To avoid this difficulty, and give to the porous cup atmome- 

 ter its greatest advantage over the open pan of water, the non- 

 absorbing mounting was devised 21 (see fig. 7) . In its essentials 

 the non-absorbing porous cup atmometer differs from the ab- 

 sorbing form only in the introduction into the tube, between 

 reservoir and cup, of a simple mercury valve which allows water 

 readily to pass from reservoir to cup but practically prevents 

 any movement in the opposite direction. Figure 7 (reprinted 

 here from the author's paper just cited) shows the essential 

 construction of the non-absorbing mounting as usually arranged. 

 The following description is largely drawn from the earlier 

 publication. 



Two glass tubes (A and B) extend downward from the rubber 

 stopper, one of them reaching upward to the tip of the cup, 

 the other only through the stopper. These tubes are slightly 

 offset laterally, so as to allow the attachment of the remaining 

 members. They are of small bore — 1 mm. is sufficient — as are 

 also the other two tubes mentioned below. Each of these tubes 

 passes a 2-perf orated stopper into a small shell vial (with round 

 bottom, though this detail is not shown in the diagram). Tube 

 B nearly reaches the bottom of the vial, tube A merely pene- 

 trates the stopper. Through the remaining perforation in each 

 stopper extends a second tube (E, F). Tube E is bent, first 

 laterally about 5 cm. and then downward. It reaches nearly 

 to the bottom of both vial C and reservoir G. Tube F reaches 



21 Livingston, B. E., A rain-correcting atmometer for ecological instrumenta- 

 tion. Plant World 13:79-82, 1910. 



