METHOD FOR DETERMINING TRANSPIRING POWER 289 



THE USE OF PERMANENT COLOR-STANDARDS 



As this general method has been heretofore employed, the 

 cobalt-chloride paper slip is dried, usually by the apphcation of 

 heat, quickly placed upon the leaf to be tested and then covered 

 with a glass or mica plate. The time period required for the 

 color change is considered as beginning when the slip is placed 

 upon the leaf. There is considerable variation, however, in 

 the amount of time required to transfer the slip from the posi- 

 tion where it has been dried to the leaf, and the amount of 

 moisture taken up by the paper during this short interval must 

 vary with the length of the interval, as well as with the pres- 

 sure of water-vapor in the air at the time. It is certain that 

 every hygrometric slip takes up some moisture during this 

 transfer, and the amount thus taken up is not always the same, 

 thus introducing a variation in the moisture content of the slip 

 when the test is begun. To correct for such variations is prac- 

 tically impossible, and they frequently introduce a serious error, 

 especially when the air is moist, when the transfer is made slowly, 

 and when the leaf surface in question has a high transpiring 

 power. We may term this the initial error in the operation. 



When a test is finished its termination is shown by the attain- 

 ment of a certain color by the paper sUp. The end point of 

 the color change is taken as that point at which the test slip 

 has the same color (pale pink) as that of an originally similar 

 slip placed upon the same kind of leaf some time previously. 

 Or else, the test is considered as ended when blue color can no 

 longer be detected, this end point being less definite than the 

 one obtained with the control slip just referred to. Consider- 

 able variations in color judgments are frequently introduced 

 by both these methods, resulting in a corresponding error in 

 the determination of the end point of color change. This may 

 be termed the final error. 



To reduce both of these errors as much as possible a method 

 employing two permanent standard colors has been devised. 

 These colors are both blue, one of them being a little less in- 

 tense than is the color of the thoroughly dry cobalt-chloride 



