192 



BOTANICAL GAZETTE 



[SEPTEMBER 



The dewpoint of the incoming and outgoing air was determined 

 by the interjection of the base of a polished nickel test-tube into 

 the air stream. Ether within the tube was cooled by blowing a 

 current of air into it with an atomizer bulb. When the ether and 

 the tube were cooled to the dewpoint of the air stream, the for- 

 mation of a film of dew on the outside of the test-tube could be 



observed. 



thermometer 



degr 



to be sufficiently accurate for all practical purposes. In the 

 following discussion only the dewpoints of the incoming and out- 

 going air are given. Any- 

 one desiring to know the 

 absolute evaporation can 

 quickly calculate it from 

 these data; but for the pur- 

 poses of this discussion 

 this is not necessary. Let 

 y = rise in dewpoint of out- 

 going air over that of the 

 incoming air; t = tempera- 

 ture of air; / x = dewpoint of 

 incoming air; 2 = a con- 

 stant, the value of which 

 will depend upon the rate 

 of air movement (w) and 

 the size and nature of the 

 evaporating surface. The value of y may then be expressed by 

 the formula y = z(t—t I ). The agreement of this formula with the 

 experimental results is shown in fig. 1. This gives the results 



» 



of an experiment in which the temperature was artificially raised 



2 H » 10 |1 If /6 J8 20 5 It « £ 



Fig. i 



and controlled, 



giving a range 



from 16.4 to 25. 2 C. Since 



the dewpoint of the air used remained constant at 3.0 , there 

 was a range in the value of t—t z of 8.8°. The lines ot and ok 

 give the value of t and f, respectively. The line oy gives the 

 calculated value of y when z = $t> P er cent - The dots show the 

 agreement of the experimental with the calculated value of y. So 

 long as other factors remain the same, the value of y appears to 



