128 Hacxertr—The Photometry of N-Rays. 
N-ray effect to look for a small change in such a surface. We have then to 
remember brightness—an exceedingly difficult thing to do. If this were our sole 
guide in these observations, Blondlot’s experiments would deserve the severest 
criticism. 
But the eye, consciously or unconsciously, guides itself by another property of 
such a surface. Looking at a dimly-lighted surface, we find it difficult to 
distinguish its outline unless we are near. The further off we go, the more 
blurred and indistinct does it become; an increase of brightness diminishes the 
indistinctness until the whole outline comes out clearly again. By paying special 
attention to this aspect of the matter, I hope to be able to show why so many 
have found such observations difficult, and how, by taking advantage of this 
property, it is possible with practice to make such observations easy. 
The object observed in many experiments was simply a diaphragm, with two 
vertical slits fixed in front of a phosphorescent screen. ‘The slits were separated 
by a distance equal to their own width, which varied from 2 to 8 mm. Ata 
short distance, the two slits came out very distinctly, separated by a black line. 
If now the screen is gradually moved away from the eye, the black interval 
between the slits gradually gets more and more indistinct, and finally disappears. 
The two slits appear to be a uniform spot of light. In this condition, or rather 
slightly before this condition is reached, the screen is extremely sensitive to 
changes of brightness. A slight increase of brightness causes the re-appearance 
of the line between the slits. 
It is difficult to say what are the conditions of intensity and distance when 
. this indistinctness takes place. If we look upon the outline as a series of points, 
the illumination of the outline will then follow the same law as the illumination of 
a point—that is, the illumination or indistinctness of the outline will be dependent 
on the intrinsic brightness, and the inverse square of the distance. Indicating 
the intrinsic brightness by J, and the distance of the screen from the observer by 
. 2 MH 
R, we might say, then, the indistinctness would be some function of ie So 
that under the same conditions of indistinctness = would be constant. Such a 
law would be in accordance with the fact that the greater the brightness of the 
screen, the greater its distance from the eye in order that it should be equally 
indistinct. 
A simple experiment was made to sce if there was any foundation for assuming 
such a law. ‘I'wo slits, 2 mm. apart, were eut in a sheet of cardboard, and 
covered with white paper. This was illuminated from behind by a very minute 
gas flame. The arrangement gives two bright slits separated by a dark line, 
