eS ee 
bi —_ 
ON THE EARTHQUAKE AND VOLCANIC PHENOMENA OF JAPAN. 169 
land and Sonth France. Professor M. S. di Rossi gives several interesting 
examples where magnets have dropped armatures or iron filings, or there 
have been sudden changes in magnetic elements at the time of earth- 
quakes. Amongst the observers of these phenomena we find Sarti, 
Count Malvasia, Palmieri, Secchi, Bertelli, Mascart, Lamont, and others. 
In Tokio I have often observed disturbances due to mechanical shaking, 
and one of the first seismoscopes I constructed about fourteen years ago 
consisted of a small magnetic needle held in a position of unstable 
equilibrium by the attraction of a piece of iron. On being shaken the 
needle flew to the iron, where it remained as evidence of a disturbance 
in every probability mechanical. The observations, however, of the 
greatest interest are those where the instruments which have been 
disturbed have been situated well outside any area of perceptible shaking, 
as, for instance, when magnetographs at Perpignan, Paris, Lyons, Kew, 
and other observatories were simultaneously disturbed at the time of the 
Riviera Earthquake on February 26, 1887 (see ‘ Nature,’ March 3, 1887). 
The magnetic disturbance following the eruption of Krakatoa in 1883 
progressed westwards aud northwards at rates of from 761 to 939 miles 
per hour, which is apparently a rate very quick even for a dust cloud to 
travel. 
At the Magnetical Observatory in Tokio, where magnetic elements 
have been recorded photographically for the last few years there do 
not appear to have been any disturbances at or about the time of earth- 
quakes excepting those which may be accounted for as being due to 
mechanically produced movements. 
The irregularities which exist are most noticeable in the lines indi- 
cating changes in deelination. They are occasionally visible in the 
record for horizontal force, but hardly ever in the record for dip. 
All the records respecting magnetic disturbances at or about the time 
ef earthquakes which I have been able to collect are being published in 
vol. xy. of the * Trans. Seis. Soc. of Japan.’ 
2. Electric Phenomena. 
At or about the time of earthquakes, electrical phenomena appear to 
be more frequent and more pronounced than magnetic phenomena, and 
the records of such phenomena are found in the description of many 
large earthquakes. The earthquakes in Catania 1693, at Lisbon 1755, 
in New England 1727, at Manchester 1777, in Ohio 1812, were all 
accompanied by electrical phenomena. Humboldt observed that during 
the earthquake of Cumana the electroscope quickly showed the presence 
of electricity in the atmosphere. 
Telegraphic land lines and submarine cables have often been disturbed 
by earth-currents at the time of earthquakes. In my second report to 
this Association, in 1881, I gave an account of earth-currents produced 
by the shaking of the ground at the time of an explosion of dynamite, 
and suggested that their origin might be due to the shaking, creating 
differences in contact between the earth and an earth plate resulting 
in varying degrees of chemical action. 
In Italy Professor Demenico Ragona observed that at the time of an 
_ earthquake there was a current passing through a galvanometer to a 
lightning rod-like conductor in the atmosphere. This observation led 
me to examine the photographic records of atmospheric electricity taken 
