[turnbull] a subjective PHENOMENON OF VISION 27 



reflected from pigments placed in a darkened room and illuminated by 

 ?.n "Auer " Inirner ; Fig. 2, the intensity of tlie light transmitted through 

 the coloured glasses placed in the Lambert apparatus as in Fig. 3, the 

 intensity of the light from the bright sky (sunshiny days) after trans- 

 mission through the same coloured glasses. Plate III, Figs. 1 and 2, 

 show cun^es for additional glasses (and for the white reflector without 

 any glass) used with the Lambert apparatus as in Fig. 2, Plate IL 



62. It will be noted that in all these curves the more refrangible 

 rays have the smaller, and the less refrangible, the larger initial period- 

 icities ; whereas white has a periodicity lying part way between that for 

 (t and that for Y. 



It will be also noted that with increasing intensity (within the 

 small limits of these experiments) the periodicity decreases. In all 

 of these curves there is the usual (initial) rapid slowing down of oscil- 

 lations observable. 



63. Eeadings in the different sets were, as much as possible, mixed, 

 in order to make all of these curves comparable with one another. 



64. Fig. 3, Plate III, represents the curves for red and indigo 

 and a mixture of the two (obtained with the Lambert colour mixer). 

 These curves are not comparable with the others of Plates III and IV 

 (as they were obtained a long time afterwards and the chronograph 

 had changed somewhat), but they are comparable with one another as 

 the sets were thoroughly intermixed. 



65. It will be noted that the curve for the mixture lies between 

 the curves for the simple colours, but (initially) somewhat nearer the 

 blue — this may be accounted for by the fact that the intensity of a 

 mixture equals the sum of the intensities of the simple lights making 

 up the mixture. The shade of this mixture was what is usually called 

 magenta, i.e., the red predominated a little. 



66. The curves, Plate IV, form an independent series of results 

 obtained with different coloured glasses and the Lambert apparatus. 

 They are, in general nature the same as curves Fig. 2, Plate II, and 

 Figs. 1 and 2, Plate III, and they were made, as nearly as possible, 

 under the same conditions. The only additional curve in the latter 

 series is N, Fig. 2, Plate IV, which represents the periodicities obtaining 

 when a dark object is "fixed" (the eyes having just emerged from 

 moderate daylight). The idea in obtaining this curve was to find the 

 initial "normal" periodicit}', i.e., the periodicity of the oscillations 

 which continually obtains during moderate daylight (it is probable that 

 black does not affect the periodicity initially, but allows the oscillations 

 to become gradually slower). It will be noted that the periodicity 

 for "normal" (453) is near that for white (-447). 



