422 
Transactions of the Boyal Society of South Africa. 
different gauges at their different respective distances, raises three points 
for consideration in the framing of a correct evaporation theory. First, 
that however favourable we may try to make the conditions, the rate 
of evaporation from a given surface area cannot be made indefinitely 
rapid.'^ If this be so it follows that when the maximum rate of evapora- 
tion from the given surface is reached, any further increase of wind 
velocity is of no effect, because there is no additional quantity of water 
vapour to be carried away. The wind factor in Fitzgerald's formula can 
therefore at the best only apply at velocities below a certain fixed upper 
limit. Second, and arising directly out of the first, if the rate of evapo- 
ration can be accelerated from a given mass of water by every increase of 
wind velocity, it can only be because the increased agitation of the water 
increases the area of the evaporating surface, and not that the evaporation 
from a given surface is affected at all. Now the wet bulb assumes its 
lowest temperature in a comparatively weak draught,! so that it seems 
a fair inference that the evaporation from an undisturbed water surface 
reaches its maximum rate on account of the wind alone, also in a com- 
paratively weak draught. For that reason it is not very likely that much 
success will attend the proposal to determine the moisture in the air from 
the measured rate of evaporation from a water surface.]: Third, that the 
agitation of the water mixes the cool evaporating surface with the warmer 
lower layers more effectually than would be the case if the cool surface 
particles had to sink by their own gravity. This factor reduces the 
effective temperature of the surface of a small mass of water much more 
quickly than it does a large mass. 
Leaving out of account the small differences in the final temperature 
of the water in the different gauges, the loss by evaporation in the gauges 
at 15, 24, and 33 inches is closely represented by the formula — 
where E is the loss of water in grains, and W the velocity of the air 
current in feet per minute. 
In the large Table at the end will be found details of the last series of 
experiments arranged in order of relative humidity, to which is added, for 
* It has been stated that the evaporation from water in vacuo is instantaneous. But 
Hertz, in describing one of his experimental researches, has pointed out that the existence 
of a limited rate of evaporation for every fluid is demanded by the kinetic theory of gases. 
In air, of course, the rate will be much slower than in a vacuum. 
f Unless the depression of the wet bulb be very great. See Sutton, " A Comparison 
between Glaisher's Factors and Ferrel's Psychrometric Formula," Quarterly Journal, 
li. Met. S., January, 1906. 
I On this point, however, see Cleveland Abbe's " Treatise on Meteorological Apparatus 
and Methods," 1888. 
