660 Mr. McClung on the Relation between Bate of 



The measurements of the different values of ?i were made 

 in exactly the same way, except that the levers were separated 

 at known intervals from one another along the line xy. By 

 separating these levers at definite distances from each other, 

 the electromotive force could be applied to E at definitely 

 known times after the rays were stopped. By this means 

 different values of ?2 were obtained corresponding to different 

 known intervals of time t. Yalues corresponding to N and 

 n were thus obtained. The values of t corresponding to 

 the different values of n were determined by calibrating 

 the pendulum for the different distances betvveen the two 

 levers. 



In making these measurements, as is almost always the 

 case in working with Rontgen rays, some difficulty was 

 experienced on account of the tendency of the rays to vary 

 slightly in intensity. Even with an automatic regulating 

 bulb, it is almost impossible to keep the intensity of the rays 

 quite constant during a whole series of observations. In 

 order to obviate this difficulty as much as possible the bulb was 

 run at regular intervals, as far as could be done throughout 

 the series ; and in addition, instead of depending upon a single 

 reading for any one quantity, several readings were taken in 

 each case and the mean of these taken. This system of 

 means was followed throughout the whole investigation. 



To test the validity of the law of recombination at the 

 various temperatures, experimental curves were obtained 

 showing the relation between the number of ions n and the 

 time t. These curves were plotted, and then taking two of 

 the experimental points on the curve with which to calculate 

 a from the formula, several arbitrary values for t were chosen, 

 and the corresponding theoretical values for n were calculated 

 from the equation (2). These theoretical points were then 

 plotted and compared with the experimental curve. Experi- 

 mental curves were obtained for several different temperatures 

 ranging between 15° C. and 300° C. Some of these curves 

 are shown in figures 3, 4, and 5. In these curves the 

 ordinates represent numbers corresponding to the different 

 values of n, while the abscissse represent the corresponding 

 values of t in seconds. The experimental points in these 

 diagrams are marked with crosses, while the theoretical 

 points are marked with circles. To avoid confusion the 

 experimental curve alone is drawn in each case. 



The agreement between the theoretical and the experimental 

 points, as shown by these diagrams, is quite close. Any 

 slight variations which may occur can easily be accounted 

 for by variations in the intensity of the rays, experimental 



