ro8 THE PLANT WORLD 



the ratios of rainfall to evaporation furnish a compound 

 climatic factor which is more satisfactory in explaining the 

 distribution of the forests of eastern North America, than 

 is either rainfall or temperature taken alone. The paper 

 is accompanied by several very instructive charts, showing an 

 excellent agreement between the precipitation-evaporation 

 ratios, on the one hand and the limits of the three great 

 forest centers of the region, on the other. 



The evaporation data on which Transeau s work is 

 based are somewhat unsatisfactory, having been calculated 

 by RusselP, from the readings of whirled, wet and dry 

 thermometers, taken in the instrument shelters of the U. S. 

 Weather Bureau. It was therefore deemed worth while to 

 attempt a series of direct observations of evaporation, with 

 an instrument which would simulate, to as great an extent 

 as possible, the position and condition of living plants, and 

 to test the value of evaporation alone as a criterion for re- 

 lating plant distribution to climatology. The porous cup 

 atmometer, described in the first paper cited above, was used 

 for this purpose. Instruments were sent to twenty-four 

 stations in the United States, and were read weekly by ob- 

 servers at the several stations, for a period extending from 

 the middle of May till the end of September, 1907. There 

 was some variation in the time at which the instruments were 

 installed, but nearly all were in operation by June i, and 

 continued to the end of the period. The ev^aporating cup 

 was placed in the open, with its center fifteen centimeters 

 above the ground, so that the records may be taken to rep- 

 resent the evaporation conditions which are met by a plant 

 of about the above named height, growing in the open. 



Before the instruments were sent out to the stations 

 they were assembled and carefully compared, and coefficients 

 of correction were determmed for all of them, thus making 

 it possible to reduce the readings to standard. Unfortunately, 



3 Russell, T., Depth of Bvapi->r;itinii in the T'nitecl Stutes Month 

 Weath. Rev. 16; 2.^5-239 1SS8. 



