Besults of Some Experiments upon the Bate of Evaporation. 55 



With the exception of the last two rows all the observations were 

 taken when the sky was clear. It will be seen that the tendency 

 is for the tank to cool faster than the screened tub during the night, 

 and to become warm much more quickly by day. We should 

 expect then that the tank would lose more by day and less by night 

 than the tub, and should not be blessed. The following are some 

 comparative means and totals by night (XX. -VIII.) and by day 

 (VIII.-XX.), during the year 1900. 



Day. 



Air Temperature 



Wind-Velocity, m. per h. 

 Humidity % 



Tank Evaporation, inches 

 Screened Tub do do 

 Piche Tube do do 

 Tank Evaporation ratio . . 

 Screen Tub do do . . 

 Piche Tube do do . . 



72-0° 

 6-5 

 42-8 

 35-6 

 40-9 

 60-1 

 64-5 

 65-9 

 72-5 



Thus, while the tank did lose a little less than the tub by night, 

 it lost a great deal less by day, the ratios taking the relative 

 positions they would take if temperature instead of accelerating 

 actually retarded evaporation. The rapid evaporation from the 

 Piche Tube during the day is remarkable, considering that it is 

 suspended in the shade. No observations of its temperature are 

 available, but in virtue of its slender stem this is not likely to depart 

 greatly from that of the air. Various tests suggest, if they do not 

 prove, that this instrument is the most susceptible of all to the 

 strength of the wind. 



So far as the experiments made during 1900 go, they seem to 

 justify at the most the following conclusions : (1) The most potent 

 agency regulating the rate of evaporation from the different gauges 

 was the humidity of the air ; (2) A wind factor is suggested ; (3) The 

 great perturbing influence attributed to the temperature of the water 

 has not been exactly confirmed. The same experiments do not, 

 therefore, furnish sufficiently precise data out of which a general 

 evaporation formula, with accurate indices and coefficients, could be 

 constructed.'''' The main difficulty lies in the fact that it is in general 



* Even if it were sought, which is not my object. Professor Abbe quotes 

 approvingly the formula proposed by D. FitzGerald : — 



E = 0-0166 (V— r) (1 + iW), 

 in which V is the vapour pressure in inches at the temperature of the water, v the 

 vapour pressure at the temperature of the dew-point, and W the velocity of the 

 wind. (See Meteorohxjlcal Apparatus and MethoiU, 1888, p. 377.) 



