236 SMITH : 



versa has no effect whatever. An important conclusion from 

 this is that the rise of temperature of a junction registered 

 when it is placed in a leaf exposed to direct sunlight must all 

 be due to the absorption of radiation by the leaf, and no part 

 of it to the absorption of radiation by the junction itself. 



The method of ascertaining the temperature by reading the 

 deflections of the galvanometer was checked on several occa- 

 sions, e.g., January 22, 27, and 29, and November 16, by using 

 hot water in the funnel containing the control junction. In 

 this way the temperature of the junction in water was made 

 equal, or nearly equal, to that of the junction in the exposed 

 leaf. The deflection obtained under these circumstances was 

 either zero or was quite small. When the deflection is zero, 

 the accuracy of the method is a maximum, and in all cases 

 the readings thus obtained were confirmatory of the results 

 obtained by the more usual method. The following figures 

 are an example of this : — 



Table IV.— January 25, 1906-. 

 Junction in leaf of Tradescantia discolor. 



Hot water was then added until the junction in water was at a 



temperature of 42*8, and the following readings were obtained: — 



Time : 11.24 a.m. 



Conditions : Exposed to svm. 



Zero: -1. 



Deflection : • 2. 



Temperature of junction in water : 42-8. 



Shade temperatmre of air : 27 ■ 5. 



Calculated temperature of junction in water : 42-3. 



Thus the temperature registered when the control junction 

 was at 27*4 and the galvanometer showed a large deflection 

 agreed very closely with the temperature registered immedi- 

 ately afterwards when the control junction was 42 '8 and the 

 deflection only "l. 



