e difficulty of mounting them and keeping them in
order, tend to prevent them from being generally used.

Seismology will not be able to make any serious progress until it has
at its disposal very certain and very numerous data as to telluric
movements registered at a large number of points at once by accurate
instruments. I have endeavored to construct a simple apparatus capable
of automatically registering such facts as it is most necessary to
know in scientific researches on the movements of the earth. After
numerous experiments I believe that I have succeeded in solving this
delicate problem, since my apparatus, put to the test of experience,
has given me satisfactory results. I have consequently decided to
submit it to the approval of men of science.

My seismograph is capable of registering (1) vertical shocks, (2)
horizontal ones, (3) the order in which all the shocks manifest
themselves, (4) their direction, and (5) the hour of the first
movement.

[Illustration: CORDENONS' SEISMOGRAPH.]

The apparatus is represented in the accompanying cut. The horizontal
shocks are indicated by the front portion of the system, and the
vertical ones by the back portion. The hour of the first shock is
indicated as follows: The elastic strip of steel, C, is fixed by
one of its extremities to a stationary support, d. When, as a
consequence of a vertical motion, the free extremity of this strip
oscillates, the leaden ball, x, drops into the tube, c, and, on
reaching the bottom of this, acts by its shock upon a cord, i, which
actuates the pendulum of a clock that has previously been stopped at
12. The other strip, B, is very similar to the one just described,
but, instead of carrying a ball, it holds a small metallic cylinder,
u, so balanced that a vertical shock in an upward direction causes
it to drop forward into the anterior half of the tube to the left. A
second vertical shock in a downward direction causes it to drop into
the other half. The cylinder, u, and the ball, x, are regulated in
their positions by means of screws affixed to a stationary support.

The portion of the apparatus designed to register horizontal
(undulatory) motions consists of four vertical pendulums, z z z z,
each of which is capable of moving in but one direction, since, in the
other, it rests against a fixed column.

Telluric waves, according to modern observations, almost invariably
in every region follow two directions that cross each other at