[neelin] the sensitiveness OF THE EYE TO LIGHT 237 



observation is represented by -1170 of the total intensity of the spec- 

 trum in portion observed. But this fraction of total intensity gives 

 a luminosity in the eye-piece equal to the standard of brightness 

 which in this paper is represented by unity, and all portions of the 

 spectrum, as observed, were of equal brightness. Therefore, we may 

 represent the luminosity of each of those portions by unity, and hence 

 also their intensities as observed in the eye-piece by unity. Now 

 Konig's Law stated mathematically is of the form Di/ (I +Di) =k where 

 Di represents the least perceptible increment and I a constant in- 

 tensity in the source of light as first observed. Hence making the 

 proper substitutions in the above formula from data in Table XV and 

 plotting. Curve 2, Fig. 4 was obtained. 



Discussion of Results. 



In general, we may conclude that the experiments described in 

 part I of this paper strongly support the view that, except in the case 

 of spectra from light at very low intensity, the general character of 

 the sensibility curve does not vary with varying intensity of the source. 

 If, however, the spectrum is one of very low intensity the two pro- 

 nounced maxima, one in the yellow and one in the green tend to 

 diminish while the two slight maxima in the blue and red apparently 

 maintain their prominence. When the spectrum is of uniform intensity 

 in all parts these maxima appear either to disappear entirely, leaving 

 a uniform curve parallel to the horizontal axis, or to become scarcely 

 more than noticeable. Referring to Table XV it will be noticed that 

 in the regions of jM=-564 and //=-648 slight maxima still persist. 

 I do not know whether these indications have a real physical signifi- 

 cance or are due to inaccurate observations. I am inclined to adopt 

 the latter reason for their appearance. This could only be settled 

 absolutely by the work of several observers upon uniform spectra at 

 different intensities. If, indeed, these maxima really exist in how- 

 ever slight a degree, we are forced to conclude that Konig's Law as 

 interpreted in this paper is not absolutely correct, for these maxima 

 still persist as is shown under Di/(I+Di), Table XV. Conclusions 

 reached regarding the steady fall in sensibility with increasing inten- 

 sity, and the application of Fechner's Law, are quite in accord with 

 accepted results. 



I desire to acknowledge my indebtedness to the kindness and 

 valuable advice of Professor Frank Allen, director of the Department 

 of Physics, University of Manitoba, at whose suggestion the investi- 

 gations described in this paper were undertaken; also to Dr. R. K. 



