TRA NSP1RA TION. 3 5 



exposed in a normal way in the air of the room or in the open. 

 To do this we should select two plants of the same kind growing 

 in pots, and of approximately the same leaf surface. The potted 

 plants are placed one each on the arms of a scale. One of the 

 plants is covered in this position with a bell jar.^ With weights 

 placed on the pan of the other arm the two sides are balanced. 

 In the course of an hour, if the.air of the room is dry, moisture 

 has probably accumulated on the inner surface of the glass jar 

 which is used to cover one of the plants. This indicates that 

 there has here been a loss of water. But there is no escape of 

 water vapor into the surrounding air so that the weight on this 

 arm is practically the same as at the beginning of the experiment. 

 We see, however, that the other arm of the balance has risen. 

 We infer that this is the result of the loss of water vapor from the 

 plant on that arm. Now let us remove the bell jar from the other 

 plant, and with a cloth wipe off all the moisture from the inner 

 surface, and replace the jar over the plant. We note that the 

 end of the scale which holds this plant is still lower than the 

 other end. 



75. The loss of water is greater in a dry than in a humid 

 atmosphere. This teaches us that while water vapor escaped 

 from the plant under the bell jar, the air in this receiver soon 

 became saturated with the moisture, and thus the farther escape 

 of moisture from the leaves was checked. It also teaches us an- 

 other very important fact, viz. , that plants lose water more rapidly 

 through their leaves in a dry air than in a humid or moist atmos- 

 phere, We can now understand why it is that during the very 

 hot and dry part of certain days plants often wilt, while at night- 

 fall, when the atmosphere is more humid, they revive. They lose 

 more water through their leaves during the dry part of the day, 

 other things being equal, than at other times. 



76. How transpiration takes place. Since the water of 

 transpiration passes off in the form of water vapor we are led to 

 inquire if this process is simply evaporation of water through the 

 surface of the leaves, or whether it is controlled to any appreci- 

 able extent by any condition of the living plant. An experiment 



