Cannon : A method of measuring transpiration 523 



The polymeter gives ^o mm. as the maximum Vapor-pressure 

 at 84° F., and this number, 30, expresses also the weight in grams 

 of one cubic meter of water-vapor at that temperature, since for 

 the temperatures that usually occur the vapor-pressure in milli- 

 meters is expressed by almost the same number that gives the 

 weight of water-vapor per cubic meter in grams.* That is, a 

 cubic meter of saturated air, at sea-level, and at a temperature of 

 84° F., weighs (according to the polymeter) 30 grams.f At a 

 saturation of 23 per cent, there would be 6.9 grams in one cubic 

 meter (0.23 x 30). Multiplying this number by that fraction of 

 a cubic meter which the capacity of the bell-glass represents, 0.0175 

 (17^500/1,000,000), gives 120 milligrams. Therefore, at a tem- 

 perature of 84^ F. and at a saturation of 2;^ per cent, the meteoro- 

 logical conditions at the beginning of the experiment, there were 

 120 milligrams of moisture in suspension in Q.0175 of a cubic 

 meter, or in other words in the bell-glass. 



If the second reading of the experiment is reduced in like 

 manner it will give 262 milligrams as the amount of moisture 

 which was in the bell-glass at 10:00 o'clock. The difference be- 

 tween 262 and 120 milligrams is the increase in the absolute 

 humidity which is due to the transpiration of the plant. That is, 

 142 milligrams of watery vapor were given off by the plant in 2y 

 minutes, and from this result the rate fpr any period and for any 

 surface-area can be easily estimated. 



The preceding method of deducing the amount of moisture 

 from readings given by the polymeter is sufficiently accurate for 

 tests, for purposes of demonstration, or for comparison, but is 

 subject to certain very obvious errors. In the first place the 

 vapor-pressure given by the polymeter is not so accurate as that 

 provided by reliable hygrometric tables ; f it cannot be read so 

 closely. And, also, the maximum pressure of vapor in milHmeters 

 is not expressed at all temperatures that occur outdoors, at any 

 rate in the desert region, by the number that gives the weight of 

 vapor in grams per cubic meter. For example, the thermometer 



*Hann. Handbook of Climatology. English translation by R. DeWard, 1903. 



fThe number on the polymeter is really slightly less at the temperature, but as the 

 scale is so small it cannot well be read more closely. 



J Smithsonian Meteorological Tables, 1897, have been used in most of the reduc- 

 tions. 



