ON SEISMOLOGICAL INVESTIGATION. 227 
45 mm. in diameter. Near its lower end, the wire passes through slits 
in the short arms of two horizontal levers. The long arms of the levers 
are made of two very light brass tubes which, soldered to a small metal 
plate, form a very elongated isosceles triangle in a horizontal plane. The 
short arms are inclined at 45° to the long arms in opposite directions, so 
that, while the former are at right angles to one another, the latter are 
parallel. The weight of each lever is 25 grammes ; the length of the long 
arm is 40 cm., and its ratio to that of the short arm is at present 10 to 
1; but this ratio can, if desired, be increased to 20 tol. At the free 
end of the long arm is a small ‘V-shaped frame, which carries a light pen 
furnished with a counterpoise, similar to those used in the meteorological 
instruments constructed by MM. Richard of Paris. The levers are 
arranged so that they are perfectly free throughout their whole range, 
passing one over the other without striking. 
The instrument at Catania differs only in details. The pendulum is 
26 metres long and 300 kg. in mass, and the horizontal levers are made 
of thin aluminium plate. 
In both apparatus, the strip of paper on which the registration is 
made is 14 cm. in breath, and is driven by a brass cylinder 60 cm. in 
circumference, which rotates once an hour. <A strip of paper, which costs 
one franc, lasts for about seven days, and can be used at least four times, 
twice on each side, so that the daily cost is less than four centimes. 
Paper is also wrapped round the driving cylinder to prevent the loss of 
any part of the diagram, in case the moving strip should come to an end 
unexpectedly. 
For ten seconds at the beginning of every hour, the traces are inter- 
rupted, the levers being raised from the paper by means of a system of 
levers connected electrically with a chronometer. The experience of five 
years with another instrument shows that this is the best of the methods 
which have been devised for marking the time. The subsequent move- 
ment of the levers does not seem to be in the least affected by their 
removal, and the missing part of the diagram is so small that it can be 
reconstructed with ease. 
The diagrams corresponding to distant earthquakes which are obtained 
with this seismometrograph are too large to be conveniently reproduced. 
The velocity of the paper being so great (60 cm. an hour),! the diagrams 
are exceedingly clear, showing the individual undulations so distinctly 
that all the elements of the motion and the epochs of the different phases 
can be determined with great precision. 
Professor G. Grablovitz’s Geodynamic Levels.—For many of the details 
given below I am indebted to the kindness of Professor Grablovitz. The 
levels are installed in the R. Osservatorio Geodinamico of Casamicciola, in 
the island of Ischia, one being directed north and south, and the other 
east and west. 
The account of these levels is contained in the following papers :— 
1. ‘Livelli geodinamici a registrazione continua:’ ‘Boll. Soc. Sismol. Ital.,’ 
vol. i. 1895, pp. 39-43. 
2. ‘Nuovi metodi per indagini geodinamiche:’ Ibid. vol. ii. 1896, pp. 41-61. 
‘ The reasons which have led Dr. Cancani to regard this as the most suitable 
velocity for the study of pulsations from a distant earthquake are given in a valuable 
paper, ‘Sugli strumenti pid adatti allo studio delle grandi ondulazioni provenienti 
da centri sismici lontani’ (Rend. delle R. Accad. dei Lincei, vol. iii. 1894, 
pp. 551-555). 
Q2 
