158 



TEXTBOOK OF PLANT PHYSIOLOGY 



tion. By this simple method, it can readily be demonstrated that 

 there is a considerable difference in the rate of water loss from 

 the upper and lower surfaces of leaves which have their stomata 

 mainly or wholly on one surface. The cobalt-chloride method is 

 a qualitative one. But, during recent years, Livingston and his 

 coworkers have made an endeavor to raise it to the level of a 

 quantitative method by measuring with a stop watch the time in 

 which the blue paper turns pink. 



The most reliable methods of studying transpiration are those 

 based on changes in weight of the plant resulting from loss of 



Fig. 65. — Cobalt chloride method of measuring transpiration (after Nathansohn). 



water. It may be supposed that such methods might introduce a 

 considerable error, for changes of weight depend not only on the 

 loss or gain of water, but also on the gain or loss of dry substance, 

 due to assimilation and respiration. Numerous experiments have 

 shown, however, that the weight of water lost by a plant per unit 

 of time is several hundred times as great as the increase in dry 

 weight. The error resulting from not taking into consideration 

 assimilation and respiration is therefore negligible. 



For the study of transpiration, whole plants rooted in soil, or 

 cut branches, or leaves may be used. Rooted plants permit the 

 study of transpiration under more natural conditions, but the 

 necessity of weighing such plants together with the soil in which 



