ENVIRONMENTAL CONDITIONS. 385 



another index based on atmometric measurements carried out from 

 the Desert Laboratory in the summer of 1908. From our work with 

 these and other atmometric indices, as well as on a priori grounds, 

 it appears that evaporation is one of the most important climatic 

 features, as far as the climatic control of vegetation is concerned. It 

 deserves more attention than does precipitation, and as much as does 

 temperature, but observational data on the evaporating power of the 

 air are as yet exceedingly meager, and serious interest has only recently 

 been directed toward atmometry. 



Transeau's ratio, the normal annual precipitation divided by the 

 annual evaporation for the single year tested (Russell) , is a very valu- 

 able index of climatic moisture conditions, and we have used this ratio 

 in our series. The method employed by Transeau has been modified 

 in several ways, to give other moisture indices. Of these modifi- 

 cations two appear to be particularly worthy of special mention here: 

 (1) the ratio calculated for the period of the average frostless season, 

 and (2) a ratio derived by dividing the total evaporation for the period 

 of the average frostless season (Russell) by the total normal precipita- 

 tion for the period of the average frostless season -plus the preceding 

 30 days. The use of the frostless season as duration factor needs no 

 comment in this place, but it may be emphasized that the second index 

 of the two just mentioned (or some similar one) should prove worthy 

 of much more study than is now possible. It is based on the idea that 

 the precipitation of the early spring, before the beginning of the frost- 

 less season, is effective to offset a portion of the evaporation of the 

 earlier part of the frostless season itself. 



Aqueous-vapor pressure and relative air humidity furnish several 

 moisture indices. Special attention should be directed to the normal 

 mean percentage of relative humidity for the period of the average 

 frostless season, an index that is probably the most valuable of all the 

 desiccation indices we have used. The reason why this index is here 

 accounted as more valuable than those of evaporation or aqueous- 

 vapor pressure apparently lies in the fact that relative humidity data 

 have been accumulated for a long period of time, while the other data 

 are unsatisfactory in this respect. It should be emphasized that the 

 saturation deficit is what ought to be recorded, instead of relative 

 humidity, however, and that the simultaneous observation of moisture 

 conditions throughout an area embracing as many degrees of longitude 

 as does the United States is at least misleading and apt to introduce 

 artifacts in the charts. For use in climatological ecology a much better 

 method of making the records needs to be devised. 



From the available data on wind velocity and sunlight intensity we 

 have derived climatic indices that may be regarded as pertaining to 

 the moisture aspect of the climatic complex, but the information at 

 hand, and especially the very unsatisfactory method now used for sun- 

 shine records by the United States Weather Bureau, are not adequate 

 for a general consideration of these conditions for the area studied. 



