TRANSPIRATION IN CACTI. 113 



means of the porous cup evaporimeter devised for this work and 

 described in the publication referred to above. Each increment 

 of Avater loss was divided by the number of hours in tlie time 

 period, to obtain the hourly rates of evaporation and of absolute 

 transpiration. The rates of relative transpiration were found 

 by dividing the rates of absolute transpiration by the correspond- 

 ing rates of evaporation. The rate of relative transpiration as 

 thus derived may be simply defined as the quantity which ex- 

 presses the number of evaporimeters of the form used, which 

 would be required to evaporate as much water per hour during 

 any time interval as is transpired by the plant during the same 

 interval and in the same position. According to the size of the 

 plant in question the average magnitude of this ratio may be 

 less than, equal to, or greater than unity. 



In plotting tlic three curves it is convenient to have them 

 reduced to the same scale. This was accomplished by taking as 

 unity the lowest evaporation rate and dividing each evapora- 

 tion rate thereby, thus obtaining a new series of rates, one of 

 which is unity and the others greater than unity; and by taking 

 as unity the rates of absolute and of relative transpiration for 

 the time period of least evaporation, and dividing each series by 

 the appropriate unity rate, thus obtaining reduced curves for 

 these two functions. When the three corrected curves are 

 plotted on the same co-ordinates they will all intersect at the 

 lowest point in the curve of evaporation, and can tlius be readily 

 compared. 



In the figures, the time intervals are plotted as abscissas 

 and the hourly rates as ordinates. The points are placed in the 

 middle of the time periods which they represent. Heavy hori- 

 zontal lines denote the extent of the night period, from 6 a. m. 

 to <*) p. m. The curve marked E represents evaporation, that 

 marked A. T. represents absolute transpiration, and that marked 

 E. T. represents relative transpiration. 



In Figure 1 are presented the curves for three cacti, all 

 derived at the same time, beginning May 2, 1906. The first 

 plant was a small specimen of Opunfia Engelmannii, the second 



