CH. IV] LOSS OF WEIGHT. 89 



rubber which may be held in its place as shown in 

 fig. 15 ; the glass vessel c grips the rubber sheet, and 

 also serves to prevent evaporation from the bottom of the 

 pot. Since it may be necessary to water the plant during 

 the course of the experiment, a corked tube must be fitted 

 into a hole in the cork plate. 



(114) Transpiration compared with evaporation of a 

 surface of water. 



To estimate the transpiration from a given leaf 

 surface it will be necessary to take a plant small enough 

 to be placed on more delicate balance. Detmer recom- 

 mends a Phaseolus grown in a glass vessel having a ground 

 edge so that it can be covered by a divided glass plate. 



We have found it a simple plan to make use of Lambert 

 and Butler's | lb. tobacco tins. A small plant such as a 

 Pelargonium can be knocked out of its pot and trans- 

 planted to one of these tins. Owing to the stopper-like 

 arrangement by which the tins are closed, it is easy to 

 replace the tin-lid by a split cork through which the stem, 

 and a watering-tube pass. 



Ascertain the loss by transpiration in say 12 hours, 

 and at the same time ascertain the loss of weight from 

 a shallow dish of water of known area. 



Now calculate the transpiring area of the plant and 

 compare its loss of weight per unit of area with that of 

 the water. 



If a planimeter is not available the area may be 

 calculated by tracing the form of a leaf on stout paper, 

 cutting it out and comparing its weight with that of a 



