DEPARTMENT OF BOTANICAL RESEARCH. 67 



In this laboratory experimentation is under way to compare the index 

 of transpiring power as determined by the cobalt tripartite sUps (/) 

 with the transpiring power as determined by the rate of the loss of 

 weight from the plant to the loss from an atmometer (T/E). Exten- 

 sive tests have thus far been made on five species of plants, both 

 methods being used simultaneously on every plant. In general, the 

 graph representing the daily march of I has the same shape as the curve 

 representing the dailj^ march of T/E. Shght variations occur fre- 

 quently, especially in the morning, when the tripartite slips often show 

 the beginning of the customary decrease one or two hours before it 

 appears in the other curve. In other words, the time of the beginning 

 of incipient drying can be detected sooner by means of the cobalt 

 sUps than by the weighing method. The results of the experiments 

 show that, for the species investigated at least, a comparison of the 

 transpiring power of different plants may be made as accurately with 

 the cobalt tripartite slips as by the use of the weighing method. The 

 fact that the cobalt shps may be used on plants naturally rooted in the 

 earth makes comparisons by this method more reliable thanby the other. 



As more data are being accumulated by the simultaneous use of the 

 two methods, it is becoming more and more probable that a numerical 

 constant will be found, the use of which will enable the experimenter to 

 compare the actual amounts of water lost per unit area from various 

 plants as well as their transpiring power. The search for this constant 

 is being made as follows: 



Transpu^ation, T, per unit area per unit time is obtained by weighing 

 plants in sealed pots at 2-hour intervals; evaporation, E, is obtained 

 by loss of weight from a white atmometer of the Livingston type; 

 the ratio T/E represents the transpiring power. Then for half an 

 hour in the middle of each transpiration period from 10 to 20 leaves 

 on the same plants are tested with the cobalt slips. The time required 

 for each slip to make the proper color change on the leaf is divided by 

 the time required for the same slip to make the color change over a 

 standard surface at the same temperature ; the reciprocal of this number 

 represents the index of transpiring power (/). When the values for 

 T/E are divided by the corresponding values for /, numbers are 



obtained which approach a constant. If — y- =c, then lXc=T/E; 



now, since E can be obtained from atmometer readings T, the actual 

 water-loss per unit area can be found. A much larger number of deter- 

 minations must be made before the validity of the constant can be 

 confirmed, but it now seems reasonably certain that by testing many 

 leaves in each experiment a constant can be found the use of which 

 will admit of at least an approximation of the actual amount of water 

 lost when the cobalt shps are used. This constant appears to be in 

 the neighborhood of 0.017. 



