136 Hacxett—The Photometry of N-Rays. 
JOO, 
I shall conclude my paper with a description of such an experiment. The 
observations mentioned above were taken with screens which had been exposed 
to daylight. It is probable that most observations have been hitherto made 
under these conditions. Such screens are comparatively bright, and N-rays cause, 
as we have seen, only small changes in them. 
But if we take very dim screens, the N-ray effect is relatively much larger. 
When the phosphorescent screens are kept,for a long time in total darkness, their 
phosphorescence disappears. If they are now exposed from one to two minutes 
to gaslight, at a distance of about 6 feet, they acquire a slight phosphorescence, 
which decreases quickly for a short time. After the initial rapid decrease is over 
they are suitable for observation. In such screens the outline alone suffices for the 
guiding indistinctness. In the experiment, it is well to have another screen which 
has gone through the same treatment. These two screens are placed so that the 
outline is slightly indistinct. This occurs at a distance of 30 to 40cm. In this 
state the two screens can be adjusted so that the indistinctness of each screen is 
the same. It is usually stated that objects appear of equal brightness at all 
distances. If the screens were initially of different brightness, it would then be 
impossible to make them appear of equal brightness. But in the case above we 
are solely paying attention to indistinctness—in fact, it 1s well to neglect the 
intrinsic brightness of the two screens altogether, as it is lable to confuse one, 
However, for convenience, the screens will be called equally bright when they 
appear equally indistinct. 
The experiment with N-rays was carried out in this way. The observer 
placed the screens at such a distance that the indistinctness of outline could be 
easily observed. One screen was fixed in position, and the observer was asked to 
move the second one until it seemed as bright as the first. When this had been 
done, the position of the screen was noted; several comparisons were made in 
this way. After about half a dozen attempts the observer could generally com- 
pare the indistinctness with a certain amount of precision. 
If & were the distance of the movable screen from the observer when its 
indistinctness is equal to that of the fixed screen, the precision generally attained 
was such that the various values of & did not differ by more than 1 or 2 cm. 
When such precision was arrived at, the observer could be quite sure the screens 
were equally indistinct. If some source of N-rays was then brought up behind 
either screen, the balance of brightness was disturbed after a short time. The 
observer could no longer say one screen was as bright as the other. He would be 
