as I used in my researches on the velocity of propagation of "X" rays, might perhaps in this case play the part of analyzer, inasmuch as the properties of a spark may be different in the direction of its length, which is also that of the electric force producing it, and in directions normal to its length. Starting from this, I arranged an apparatus as shown in the accompanying diagram, so as to obtain a small spark during the emission of "X" rays.
A focus tube is connected to an induction-coil by wires BH, B'H', covered with gutta-percha (Fig. 1). Two other wires, also covered with gutta-percha, AIc and A'I'c', terminate at A and A' in two loops, which surround BH and B'H' respectively; a bit of glass tubing, not shown in the figure, keeps each loop separate from the wire which it surrounds. The wires AI, A'I are then twisted together, and their sharply pointed ends, c and c', are fixed opposite each other, at a very small distance, adjustable at will, so as to form a small spark-gap. By virtue of this disposition, the electrostatic influence exercised by the wires BH and B'H' on the loops A and A' produces
