at each break of the current in the coil a small spark at the gap cc', at the same time as "X" rays are being emitted by the tube. Owing to the flexibility of wires, AIc, A'I'c', the straight line cc', along which the spark occurs, can be set in any direction we please. sheet of aluminium foil, 40 cms. square, is interposed between the tube and the spark, so as to prevent any direct influence of the electrodes of the tube on cc'.

A

In order to define easily the relative positions of the tube and the spark cc', take three rectangular axes, of which one, Oz, is vertical.

Fix the focus tube so that its length, and, consequently, the pencil of cathode rays, coincides with OY, the anticathode being placed near the origin, and sending "X" rays in the positive direction of OX.

Place the gap cc' at a point on the positive side of OX, so that its direction is parallel to OY. The spark being properly regulated one observes that the "X" rays act upon it in such a way as to increase its luminosity, for the interposition of a sheet of lead or glass manifestly diminishes the brightness.

Now, without altering the position of the gap, turn it so that it comes parallel to OZ, i.e. normal to the cathode rays. The influence of the "X" rays on the spark is then seen to disappear, and the interposition of a lead or glass plate causes no change in its brightness.

66

"X" rays have therefore a plane of action, which is the one passing through each "X" ray and the cathode ray which gives rise to it. If the direction given to the spark-gap is intermediate between the two above mentioned, the action is seen to diminish from the horizontal position to the vertical.

The following is another experiment, still more striking: if the spark is made to turn about OX, parallel to plane YOZ, the spark is seen to pass from a maximum brightness when horizontal to a minimum when vertical. These variations of brightness are similar to those observed when a pencil of polarized rays traverses a rotating Nicol's prism, the small spark playing the part of analyzer. The pencil of "X" rays presents the same asymmetry as a pencil of polarized light. According to Newton's definition, it has sides differing from

each other; in other words, it is polarized in the complete sense of the term.

The phenomenon is easy to observe when the spark is well regulated; this means that the spark must be very small and faint.

If the focus tube is made to turn about its axis, which is parallel to the cathode rays, the observed phenomena do not change, so long as "X" reach the gap. rays The plane of action

is thus independent of the orientation of the anticathode, being always the plane passing through the "X" rays and the generating cathodic rays.

The spark being kept in this plane, and turned round from the position in which it is at right angles to the "X" rays to that in which it is parallel to them, we observe that the effect of the "X" rays on the brightness of the spark is a maximum in the first position, and diminishes to nothing in the second.

Now, an "X" ray and its generating cathodic ray only determine a plane when their directions are different. Again, amongst the emitted 'X" rays, some are in a direction very nearly the same as that of the cathode rays, being

those which graze the cathode.

One should expect these to be very incompletely polarized; and, indeed the small spark enabled me to confirm this.

I noted several important facts, which, however, I will merely allude to in the present paper. Quartz and lump-sugar rotate the plane of polarization of "X" rays in the same sense as that of light. I obtained rotations of 40°.

Secondary rays, styled "S" rays, are also polarized. Active substances rotate the plane of polarization of these rays in a sense contrary to that of light. I observed rotations of 18° (note 2).

It is extremely likely that magnetic rotation also exists for "X" rays as well as for "S" rays. One can also surmise that the properties of these rays, with reference to polarization, extend to tertiary rays, etc. I intend shortly to publish the results at which I have arrived concerning these different points.

On a New Species of Light (March 23, 1903).

The radiations emitted by a focus tube are filtered through a sheet of aluminium foil or a screen of black paper, in order to eliminate the luminous rays which might accompany them. While studying these radiations by means of their action on a small spark, I discovered that they are plane-polarized as soon as emitted. I further proved that when these radiations traverse a plate of quartz in a direction at right angles to its axis, or a lump of sugar, their plane of action undergoes a rotation just like the plane of polarization of a pencil of light (see pp. 5 and 6).

I then asked myself if a rotation could also be obtained by passing the radiations of the focus tube through a pile of Reusch mica sheets. I observed, in fact, a rotation of from 25° to 30° in the same direction as that of polarized light. This action of a pile of micas made me at once infer that a single sheet of mica must act, and that this action must be depolarization, or, rather, the production of elliptic polarization; this is