232 THE ROYAL SOCIETY OF CANADA 



thus showing sensibility to sligiit differences in intensity using the 

 equivalent of a spectrum of uniform hnninosity. 



Tdhlc of (Idtd. 



In Talkie X\', \ indicates the portion of the spectrum upon 

 wliicli observations was made. "jS" denotes the inclination of the 

 plane of the polariser to the plane of the analyser. In the column 

 under " a " will be found the inclination of the plane of the polariser 

 to the plane of the analyser after rotation has taken place. Cos-/3 will 

 represent the amount of light going through the niçois before the 

 rotation of the polariser, and Cos-a the amount of light going through 

 when this rotation was completed. In the column under "Least 

 Perceptible Increments" will be found the value of (Cos-a — Cos-/?) 

 which Avill represent the fraction of total intensity which had to be added 

 to the initial intensity (Cos-^) to effect a change of sensation that was 

 just perceptible. The reciprocals of those values plotted as a function 

 of the corresponding wave length gave the luminosity curve found in 

 Fig. 3. As explained above those increments were reduced to light 

 of uniform intensity by dividing by Cos"/9 in each case. The recipro- 

 cals of these latter results were taken to represent sensibilit}' and will 

 be found under "Sensibility." Under Di/ (I +Di) will be found the 

 value of the least perceptible increment divided by the total intensity 

 (after rotation) which was used to get curve 2, Fig. 4 to demonstrate 

 Konig's Law. 



IL 



Sensibility to Radiation of Varying Intensity. 



As a demonstration of the steady falling off of sensibility with 

 increasing intensity and to show that it is approximately inversely 

 proportional to the intensity over a wide range, observations were 

 made as follows: 



The same apparatus as used in the preceding was used in this; 

 and the same an'angement. First, the telescope of the spectro- 

 meter was adjusted for light of a given wave length, for example -420 // 

 (see Table VII), and rigidly fixed. The niçois wci-e then crossed so as 

 to exclude all light from the eye-piece. The angle indicated on the 

 attached scale was read and this reading was taken to indicate an 

 inclination of 90° between the principal sections of the niçois. The 

 polariser was then rotated and at the first sensation of light ex- 

 perienced through the eye-piece, rotation was stopped. The angle 

 indicated on the attached scale was again read. Let a,, denote this 



