Hacxett—The Photometry of N-Rays. 137 
certain there was a difference. Finally, the screen, under the influence of N-rays, 
could be unhesitatingly pronounced the brighter. 
Some idea of the difference in the screens was got by asking the observer to 
re-adjust the balance between the screens, The change of distance required was 
relatively large. The experiment can be subjected to the most rigid tests. The 
observer may be ignorant of what screen is being influenced by the rays; but he 
will have no doubt which screen is the brighter. The obsérver was usually tested 
in this way. 
But unless some precision is attained in comparing indistinctness, it would be 
futile to proceed with the latter part of the experiment. No difficulty has been 
found in getting this precision with any person of normal sight, and several 
experiments made with such persons have been uniformly successful. After some 
practice, a person of defective vision—short-sighted, for example—can acquire the 
necessary precision ; and when he has done so, there is no difficulty in repeating 
the experiment. 
I shall now give some estimations of the N-ray effect under these conditions 
made by a trained observer. When the screens are equally indistinct, their 
intensities are assumed to follow the law of inverse squares, which I have already 
used in the course of this paper. After some training, the two screens can be 
compared with great exactness. One screen will be moved back into almost the 
same position time after time. From many observations of this class, it seemed 
that two screens could be compared within 3 per cent. 
The N-ray effect was deduced from the change of distance required to restore 
the balance after it had been disturbed, combined with the distance of the screen. 
) . 
The same formula BULD was also used to reduce these observations. The 
R 
numbers obtained varied from 20 per cent. to 40 per cent., according to the 
source used and the conditions of the screen. 
Another way of looking at the effect is to consider the change of distance 
required to restore the balance of the screens. This varied from one-sixth to one- 
tenth of the total distance. For example, if the movable screen is at a distance of 
40 cm., and N-rays influence the second screen, the first screen will be moved 
forward from 4 to 7 cm. to restore the balance of brightness. 
I have tried to describe how these effects may be seen, but there are many 
things which cannot be communicated in a paper. I am sure that any person 
who takes a little patience in working with the screens in the manner mentioned 
will have no difficulty in repeating the experiments. 
So far I have only dealt with the simplest property of N-rays—that of 
increasing the brightness of a phosphorescent screen. I am satisfied that this effect 
