10 BURTON EDWARD LIVINGSTON 



40, respectively, referred to the air-paper surface of the water 

 test. They are directly proportional to the cobalt paper times. 



It is obvious that all of the leaves of a plant cannot usually 

 be subjected to the paper test, but the commonly used method of 

 random sampling ma.y here be resorted to. A sufficient number 

 of tests may be carried out to give to the resulting average 

 an adequate accuracy. Of course it is impracticable to obtain 

 indices by this proceedure, from surfaces which are covered by 

 a water film or which bear water droplets, as occurs with the fall 

 of rain or the formation of dew. Under such conditions, however, 

 absolute transpiration must approach zero or become actually 

 negative through slight foliar absorption. This is a consideration 

 which makes it seem unnecessary to attempt the measurement of 

 the internal resistance to transpiration at such times. Hairy 

 leaves and those which are much wrinkled or folded introduce 

 difficulties also, since with such surfaces the effect of the external 

 air upon the hygrometric paper is not readily excluded. 



When the transpiring power of a surface is very low, the time 

 required for complete disappearance of blue color in the test 

 paper may be shortened by using a thinner paper, or it is possible 

 to adopt a color standard from which all of the blue color has not 

 disappeared and to bring all test papers to this color rather than 

 to the complete response. Furthermore, the error in the observed 

 time of response, due to the momentary exposure of the dried 

 paper to the air, may assume considerable importance in a very 

 humid atmosphere. Nevertheless, the procedure above described 

 has been followed successfully at all hours of the day and night, 

 in a greenhouse, under the climatic conditions of a rainy summer 

 in JNIunich. 



The details of the application of the cobalt chloride method to 

 the more difficultly manipulated plant surfaces and of its employ- 

 ment under otherwise adverse conditions, remain mainly to be 

 elaborated. Under most circumstances, however, the method 

 above described is very simple of operation and requires no com- 

 plicated or expensive apparatus. It is well suited to operations 

 in the field, furnishing a ready means for the comparison of upper 

 and lower, or younger and older, leaves of the same plant, or of 



