RESISTANCE BY LEAVES TO WATER LOSS 



9 



lie below, or the paper may be laid upon the leaf surface, after 

 which the clip may be placed. The clips are sufficiently light so 

 that they may be left hanging on the leaves of most plants during 

 the interval of the test, but it is a simple matter to arrange sup- 

 ports for them. 



A paper having been placed upon a leaf, the time required 

 for complete loss of blue color is carefully noted. It is well to 

 facihtate the deteriTiination of the endpoint by having previously 

 prepared a cUp with a piece of leaf of the kind to be tested, held 

 between two squares of hj^grometric paper. Thus are provided 

 color standards for both upper and lower foliar sui'faces, and the 

 bit of paper has, in each case, a margin or background of the same 

 nature as have those in the actual tests. Several parallel and 

 simultaneous tests are made and the results obtained are averaged 

 for each leaf surface. The results obtained from the two leaf 

 surfaces are averaged to give the index for the surface of the leaf 

 as a whole. 



The water test must be made under the same temperature 

 conditions as the leaf tests, it being assured that the water of 

 the standard surface has been allowed time to assume the general 

 ah* temperature. Between tests, the standard surface should be 

 kept covered, to prevent evaporation with its consequent cooling. 

 A stop watch is almost essential for the water test and is conven- 

 ient for the leaf tests. 



To illustrate the manner of manipulating the results an exam- 

 ple may be given. Let it be assumed that the water test gives an 

 average time period of 15 seconds and that the corresponding 

 average for the lower leaf surface of a certain plant is 300 seconds, 

 while that for a second plant is 600 seconds. It is clear that the 

 water-supplying power of the lower foliar surface of the first 

 plant is 15/300, or 0.05 of that of the air that lies against the paper 

 in the water test, while the lower sui'face of the leaves of the sec- 

 ond plant can give off water only one-half as rapidly, their tran- 

 spiring power being 0.025 of that of the standard surface. These 

 transpiring powers should be approximately the same as the rela- 

 tive transpiration ratios derived indirectly'. The indices of 

 transpirational retardation are, in the above example, 20 and 



