486 Prof. A. 0. Rankine : Relation between Illumination 



the best straight line through the points. It is nearly equal 



to 22*0 microamperes, the observed value of the " dark " 



current. 



In the numbers in the last column the greatest variation 



from the mean is rather more than 2 per cent., and this 



occurs at low values of C, where an error o£ observation 



would be more effective than elsewhere. It has already 



been mentioned that the observations extend over a range 



corresponding to a ratio of light intensities of 500 to 1 ; 



j n ^ • £ i- l a a candles . „ AA candles 



actually the ranee is rrom about 0'4 ; — „ to 200 ; — =- 



J metre * metre** • 



The fourth-root law holds with considerable accuracy 



throughout. 



Fig. 2. 



Several series of observations were also taken in which the 

 voltage was applied continuously to the selenium. Before 

 starting any series, however, the current was allowed to flow 

 through the selenium (in darkness) until it had become 

 practically constant — i. e., it was not increasing more rapidly 

 than 0*1 microampere per hour. This state of affairs took 

 several hours to be reached. The procedure was otherwise 

 the same as that already described. Curve II (fig. 2) shows 



