iqiq] 



WEAVER &• MOGENSEN—TILiNSPIRATIOX 



399 



Continuous records of the humidity of the air and the tempera- 

 ture of both air and soil were obtained by means of Friez's hygro- 

 thermographs. The last was for a soil depth of 6 inches among 

 the containers. A thermometer placed at a similar depth in one 

 of the medium sized containers, and protected from external atmos- 

 pheric conditions by means of a felt-lined brass case fitted with 

 a cap, gave readings very similar to those of the soil, as may be 



seen in table I. 



TABLE T 



Soil temperature at a depth of 6 ksxhes inside a 



container, and at a similar depth 



outside in the soil 



While the factor data will be discussed in connection with the 

 transpiration graphs, it may be said here that at no time was the 

 soil in the containers frozen solidly to a depth greater than 2 . 5- 

 3 .0 inches, a point extending not far into the root zone. 



Transpiration losses were determined by weighing the con- 

 tainers. A large long-armed Troemner balance was used which 

 was sensitive to o . 5 gm. under the maximum load of about 8-9 kg. 

 imposed upon it. In making the weighings the containers were 

 transferred from their place out of doors into a weighing room cooled 

 to nearly a similar temperature, for, as shown by Winkler (23), 

 leaves of evergreens and twigs of other trees can endure from four 

 to six times as much cold if the changes are gradual as if they 

 are sudden. Unless the water loss was rather insignificant as 

 compared with the amount of soil concerned, the practice followed 

 was to replace the loss at each weighing by adding the proper 

 amount of water from a burette through the temporarily uncorked 

 glass tube in the side of the container. Thus the soil moisture 

 was kept at an almost imiform condition throughout. 



