1897.] NEW YORK ACADEMY OF SCIENCES. 199 



II. 



If the rate of rotation is greater than twenty-five but less than 

 about forty revolutions per second, the changeable rotating sec- 

 tors are no longer distinguishable. In their place there is an 

 irregular changing or flashing of luminosity over the surface of 

 the disk. 



With two colored disks or a colored and a gray disk there is 

 an irregular changing or flashing of colors. This phenomenon 

 is observed whenever the di^ks differ considerablv in color, even 

 though there maj'^ be small or no difference in luminosities. 



III. 



AVith a rate of rotation between forty and sixty revolutions 

 per second the irregular flashing just described disappears, the 

 disk assumes a background of a nniform hue, and over this uni- 

 form background one notices a flickering very similar to that 

 which is noticed b}' a person looking at a picket fence while 

 running past it. 



If two disks of different colors are used, or a colored and a 

 gray disk, the phenomenon of flickering is the same in all cases 

 after the rotating compound disk has assumed a uniform back- 

 ground. 



If the two disks diff"ered in color but only slightly in lumi- 

 nosit}'^, it was found that there was veiy little true flicker after 

 the background had become uniform. 



IV. 



If the speed is increased be\'ond about sixty revolutions per 

 second the flickering disappears and we have only a uniformly 

 tinted disk. 



It should be remarked that for the phenomena of rotating 

 sectors described under I, and of flashing described under II, 

 the disk has no background of uniform hue or luminosity. As 

 soon, however, as the speed is sufficiently rapid to produce true 

 flickering the disk assumes a background of uniform hue and 

 luminosity. 



The transition from any one of the above phenomena to 

 another is not in general very siiarply defined. The transition 

 from the flickering described under III. to the uniform back- 

 ground without flicker described under IV. is, however, quite 

 abrupt. Next in definiteness is the transition from flashing to 

 flickering and this transition is much more sharply defined when 

 the difference in luminosity is more than ten per cent, than 

 when it is less. When the difference was ver}' small (less than 



