226 
REPORT—1896. 
8 cm. in diameter, and rotating about a horizontal axis. The part of the 
Fig. 16. 
NE- SW 
| 
SE 
paper on which the record is being made lies on a 
rectangular platform immediately above the driving 
cylinder. Two pens fixed to the platform record the 
time every half-hour on the edges of the strip of 
paper. As a rule the cylinder rotates once in an 
hour, so that the paper is driven at the rate of about 
30 cm. an hour. But when a shock occurs the 
velocity of the cylinder is immediately increased, so 
that for three revolutions it revolves once a minute, 
thus unrolling the paper at the rate of about 5 cm. a 
minute. 
The increased velocity is produced by means of a 
roller, started by an electrical seismoscope. This 
consists in the longer arms of the levers being con- 
tinued backwards to a length about fifteen times as 
great as that of the short arms. Beside, and very 
near the further ends, are two small vertical rods, 
which turn at their lower ends about a horizontal 
axis. A very slight movement of the levers closes 
an electric circuit, and at the same instant sets in 
motion the roller which gives the increased velocity, 
moves a collar which at once withdraws the vertical 
rods, so that they do not impede the oscillations of 
the multiplying levers, and also starts a clock pre- 
viously pointing to xii. The latter clock thus gives 
the time at which the increased velocity began. 
The increased velocity continues, as already men- 
tioned, for three minutes. At the end of this time 
the two vertical rods return to their original position. 
But if the pendulum is still in motion, electrical 
contact is immediately remade, the rods are again 
withdrawn, and the increased velocity re-established, 
so that with instantaneous interruptions this lasts 
until the movement is so slight that it ceases to start 
the seismoscope. 
Fig. 16 reproduces a diagram furnished by this 
seismometrograph on the occasion of the Caspian Sea 
earthquake of July 8, 1895. 
Dr. A. Cancani’s Seismometrograph.—The chief 
difference in principle between this instrument and 
the preceding consists in the omission of the arrange- 
ments for increasing the velocity at the time of a 
disturbance. Seismometrographs of this pattern have 
been in use for some time in the the geodynamic 
observatory of Rocca di Papa near Rome. Two 
apparatus of larger dimensions have recently been 
constructed, one for Rocca di Papa and the other 
for the observatory at Catania. These are described 
in a paper, ‘Nuovo modello di sismometrografo a 
registrazione continua’ (‘ Boll. Soc. Sismol. Ital.,’ 
vol. ii. 1896, pp. 62-65). 
In the Rocca:di Papa seismometrograph, the pendulum is 15 metres 
long and 200 kg. in mass. The weight is suspended by a steel wire 
