OB; and this time the screen CD, carried along with the frame in this movement, is no longer interposed in the path of the rays. The half-plate AO therefore receives the action of the spark while subjected to the

rays.

This being understood, the experiment is as follows: first the plate is kept in the first of the above-indicated positions during five seconds, then in the second position also for five seconds; it is then brought back to the first position, and the double operation just described is repeated several times.

After an interval of time equal to an even multiple of five seconds-for instance, one hundred seconds-each of the half-plates has been exposed to the spark for an equal period, only, while AO was exposed, "N" rays were in action, and while OB was exposed there

were none.

Thanks to an arrangement of guides and buffer-stops, the to-and-fro motion of the frame can be executed with perfect certainty and regularity, in spite of the darkness. A metronome is used to regulate the action.

The spark is produced by a small induction-coil, known as du Bois-Reymond's chariot apparatus; it strikes between two blunt points of platinum-iridium, carefully machined and polished. These are fixed to the two jaws of a pair of wooden pliers, which tend to close by elasticity, and are kept apart by a micrometer screw. At a distance of about 2 cms. from the spark, and facing the plate, a plate of ground glass is fixed. As I have previously mentioned, the light of the spark produces on this glass an extensive luminous patch, much easier to observe than the naked spark, and giving on the photographic plate impressions of much more regular form. The regulating of the spark is the delicate part of the experiment. The induced current must first be adjusted, by modifying the primary current on the one hand, and the position of the secondary coil on the other, till the spark becomes very small. The points are washed in alcohol, then a slip of dry paper is drawn between them, for the purpose of drying and repolishing their surface; then the micrometer screw is turned so as to make the spark as short as possible, yet without

F

incurring any risk of the points touching by any chance vibration, which would make it disappear intermittently. By a methodical process of trial and error, which sometimes demands much time and patience, one succeeds in getting a spark both regular and very feeble; it is then sensitive to the action of "N" rays. If one directs on it a pencil of these radiations, proceeding from any source, one sees the patch on the ground glass increase in size and glow; at the same time its central part becomes more luminous, appearing wrapped in a kind of nimbus. One can then proceed with the photographic experiment. I made about forty such experiments, employing in turn, as sources of "N" rays, a Nernst lamp, compressed wood, hardened steel, Rupert's drops, etc. I have varied the experiments in different ways— for example, by changing the side of the screen CD, by using a zinc screen transparent to "N" rays, etc. Several eminent physicists, who have been good enough to visit my laboratory, have witnessed them. Of these forty experiments, one was unsuccessful: the rays were produced by a Nernst lamp, and instead of

N.B.-The striæ and most of the spots in the figures do not exist in the original photograph, but are the result of the inability of the photogravure process to reproduce images of this kind.