Cannon : A method of measuring transpiration 527 



comparative studies are carried on, and it should not be over i 

 per cent, a minute. This condition is fairly well attained by vary- 

 ing the size of the bell-glass in accordance with the rate of tran- 

 spiration. It can be told only b}^ trial. 



Although the relative humidity of the atmosphere of the bell- 

 glass cannot be lowered to suit the convenience of the operator, 

 without subjecting the plant to grossly unnatural conditions even 

 on the desert, it is altogether best if at the beginning it does not 

 greatly exceed 25 per cent. The experiment can then be so con- 

 ducted that a considerable range in the relative humidity is pos- 

 sible without greatly modifying the moisture conditions; the range 

 need not be greater, if so ample, as that to which the plant is daily 

 subjected in the open. If this precaution is taken the decrease in 

 the rate of transpiration which accompanies an increase in the 

 humidity can largely be avoided. 



The foliage should be uniformly distributed throughout the 

 bell-glass so that the diffusion of the watery vapor may be as 

 quickly attained as possible. This may be facilitated by the use 

 of some device by which the circulation of the air may be main- 

 tained. . ^ 



The temperature can be kept fairly uniform if the apparatus is 

 allowed to become of the same temperature as the plant and the 

 base, provided a permanent base is used, but in case the day is 

 unusually hot and a large increase in heat is observed, notwith- 

 standing the shade, a dampened blotter or cloth tied to the out- 

 side of the bell-glass and large enough to cover one-third of its 

 surface, will in a very satisfactory manner keep the temperature 

 down. 



Summary and conclusions 



The polymeter method has been in use one year, during which 

 time about 100 experiments were performed with it All of these, 

 with the exception of a single series, were conducted upon plants 

 in place, and the experiments, an account of which will be given at 

 another time, showed very conclusively that the new method is a 

 practicable one for outdoor use. The following summary is given, 

 with sufficient illustrations drawn from the work of the year to 

 give some conception of the scope of the method and more espe- 

 cially of its efficiency. 



