act for a determined period on a sensitive plate, whilst this source is subjected to the action of "N" rays, and then repeating the experiment for the same interval of time and under the same conditions, save that the "N" rays are suppressed. The impression produced is notably more intense in the first case than in the second. As an example of the application of this method, I gave at the time two photo-engravings, whose comparison shows that water, even when used in very thin films, arrests "N" rays issuing from an Auer burner (see page 16). Since then I have extended the experiments to the registration of actions produced by "N" rays from various sources, and I have perfected the process, as will be shown.
A small, luminous spark is the most appropriate luminous source for this kind of investigation: for, on the one hand, it is very actinic, and, on the other, it can be maintained as long as necessary at the same intensity. Although it is impossible to obtain absolute steadiness of glow in the spark, since these variations are not produced systematically,
