704 Earthquakes and Electricity . [December, 
the child of the earth, and is supported by it ” (First Report, 
p.71). And again, in respeCt of the correspondence between 
earthquakes and terrestrial magnetism, Von Hoff loquitur : — 
“ In many instances in which an opportunity of observing 
the magnetic needle during an earthquake has presented 
itself an alteration in its direction for the time has been ob- 
served.” . . . “ More remarkable, however, are the changes 
in the direction of the dip and variation needles, which take 
place at a distance from the place where the earthquake was 
observed, and at a place where the shock itself is not per- 
ceptible ; as, for instance, in Paris on the 19th February and 
31st May, 1822, simultaneously with an earthquake which 
occurred in Savoy and some of the southern parts of France. 
If this observation should be established by others carefully 
made, the existence could not be denied of a connexion be- 
tween terrestrial vulcanism and terrestrial magnetism.” 
Mallet himself supplies the following remarkable concep- 
tion : — “ Thus, then, ignorant as we are of all within the 
outer surface or skin of our globe (and of how much of its 
exterior, for the ocean shrouds two-thirds of it from our 
eyes ?), we are compelled to see the close connexion of these 
mighty heating powers in which ignition is present on the 
vastest scale, yet without combustion, with the forces of 
terrestrial electricity and magnetism, forces which are those 
alone that within range of our observation are mutually 
convertible and both convertible into heat. Currents of both 
we know are ever passing, with variable activity, through 
enormous volumes of the earth’s crust, the different parts of 
which possess very different conducting powers. Can it be 
that these currents, constrained to pass through narrow and 
bad conductors at vast depths in some formations, ignite 
them in their progress ? Will it be found that the great 
lines of volcanic activity (as dreamed by Bylandt) are in 
some way connected with those of terrestrial magnetism ? — 
are possibly normals to the surface curves of equal magnetic 
intensity ? A glance at one of Gauss’s magnetic maps, and 
at another of the great bands of aCtive volcanoes on our 
planet, almost forces the mind into such conjectures ” (First 
Report, p. 77). Yet this brilliant speculation seems to have 
fallen entirely to the ground, for the author of it consistently 
attributed the cause of earthquakes either “to the sudden 
formation of steam from water previously in a state of 
repulsion from the heating surfaces ” (First Report, p. 80), 
or “ to the passage of a wave of elastic compression causing 
each particle of earth to perform a vibratory movement ” 
(Quart. Rev., July, 1881). 
