rocks do not break and give vent to the imprisoned gas.
There is some connection between volcanoes and earthquakes; the former
are, to a certain extent, "safety-valves." The column of smoke from the
volcano of Pasto suddenly disappeared just before the great earthquake
at Riobamba. In the spring of 1868 Pichincha and Cotopaxi showed signs
of increasing activity, but in the summer became quiet again. Cotocachi
and Sangai, 200 miles apart, were awaked simultaneously; the former,
silent for centuries, sent forth dense masses of earth and volcanic
matter to a distance of many miles, covering thousands of acres; the
latter thundered every half hour instead of hourly, as before. Still,
the greatest earthquakes do not occur in the vicinity of active
volcanoes. Lisbon and Lima (where, on an average, forty-five shocks
occur annually, and two fearful ones in a century) are far distant from
any volcanic vent; likewise Northern India, South Africa, Scotland, and
the United States.

An earthquake is beyond the reach of calculation. Professor Perrey, of
Dijon, France, is endeavoring to prove that there is a periodicity in
earthquakes, synchronous with that in the tides of the ocean, the
greatest number occurring at the time of new and full moon.[96] If this
theory be sustained, we must admit the existence of a vast subterranean
sea of lava. But all this is problematical. Earthquakes appear
independently of the geology of a country, though the rate of undulation
is modified by the mineral structure. Earthquake waves seem to move more
rapidly through the comparatively undisturbed beds of the Mississippi
Valley than through the contorted strata of Europe. Meteorology is
unable to indicate a coming earthquake, for there is no sure prophecy in
sultry weather, sirocco wind, and leaden sky. The Lisbon shock came
without a warning. Sudden changes of the weather, however, often occur
after an earthquake. Since the great convulsion of 1797 the climate of
the Valley of Quito is said to be much colder. A heavy rain often
follows a violent earthquake in Peru.

[Footnote 96: Professor Quinby, of the University of Rochester, has, at
our request, calculated the position of the moon at the late earthquake:
"August 16th, 1868, 1 A.M., the moon was on meridian 137 deg. 21' east of
that of Quito, or 42 deg. 30' past the lower meridian of Quito, assuming the
longitude of Quito west of Greenwich to be 79 deg., which it is very nearly.
This is but little