152 BULLETIN 1059, U. S. DEPARTMENT OF AGRICULTURE. 
tions. For crude comparative purposes, however, the loss from a 
pan of water is probably as good a measure as any. of general, 
regional, atmospheric conditions, such as would affect lakes and res- 
ervoirs, provided only a standard exposure is employed. 
In plant ecology evaporation rates are measured for the purpose 
of determining how great or how little evaporation tendency, or 
stress, the plant has to withstand and can withstand without injury. 
As has been pointed out by Bates (151) under any circumstances 
the components of evaporation are (1) the heat necessary to trans- 
form water into vapor, and (2) the diffusion possibilities of the 
vapor. In ordinary evaporation there are two possible sources of 
heat; namely, sunlight, or radiation from any other source, and the 
heat of the air coming in contact with the evaporating body. By 
diffusion possibilities are meant the net opportunities for the vapor 
to move away from the evaporating body. These depend on me- 
chanical obstructions, on the number of vapor molecules already in 
the atmosphere, on the air movement near the evaporting body affect- 
ing this number, but most of all on the temperature, energy, and 
pressure of the molecules which are moving away. Hence sunlight 
and almost every atmospheric condition affect evaporation, and it is 
also plainly evident that each of these factors can not have an identi- 
cal effect on each of several different evaporating bodies. To use 
an extreme example, a highly polished metal vessel containing water 
might be set in full sunlight, and yet the evaporation of the water 
would not be appreciably influenced by the sunlight, because the heat 
of the sun would not reach it — the rays would be almost totally 
reflected. Similarly, the evaporating body may be very largely pro- 
tected from wind, so far as the wind might affect diffusion. This 
is to some extent the case in the leaf, where the vapor is largely 
formed internally. 
TTlien, therefore, it is said that in ecology an atmometer is desired 
which will react to the same external stimuli as affect the plant leaf 
most directly, it is not by any means implied that the instrument 
would parallel the plants' transpiration under all circumstances. 
There should be a certain similarity in their reactions when both are 
reacting freely. It is not desirable to compare an instrument which 
is mainly susceptible to wind with a plant which responds in the 
largest measure to sunlight. If there could be an atmometer which 
would receive its stimulus from sunlight, wind, air temperature, and 
humidity in about the same proportions as these four factors affect 
the plant, the evaporation factor of the habitat could be measured 
in an effective way. Otherwise, altogether too much emphasis may 
be placed on some one component, such as wind. After all is said 
this is almost exactly the same problem as attempting to integrate 
