omposed and made to enter into new combinations,--the quantity of
heat which they evolve during these processes; when we recollect the
expansive power of steam, and that water itself is composed of two gases
which, by their union, produce intense heat; when we call to mind the
number of explosive and detonating compounds which have been already
discovered, we may be allowed to share the astonishment of Pliny, that a
single day should pass without a general conflagration:--"Excedit
profecto omnia miracula, ullum diem fuisse quo non cuncta
conflagrarent."[759]

The signs of internal heat observable on the surface of the earth do not
necessarily indicate the permanent existence of subterranean heated
masses, whether fluid or solid, by any means so vast as our continents
and seas; yet how insignificant would these appear if distributed
through an external shell of the globe one or two hundred miles in
depth! The principal facts in proof of the accumulation of heat below
the surface may be summed up in a few words. Several volcanoes are
constantly in eruption, as Stromboli and Nicaragua; others are known to
have been active for periods of 60, or even 150 years, as those of
Sangay in Quito, Popocatepetl in Mexico, and the volcano of the Isle of
Bourbon. Many craters emit hot vapors in the intervals between
eruptions, and solfataras evolve incessantly the same gases as
volcanoes. Steam of high temperature has continued for more than twenty
centuries to issue from the "stufas," as the Italians call them; thermal
springs abound not only in regions of earthquakes, but are found in
almost all countries, however distant from active vents; and, lastly,
the temperature in the mines of various parts of the world is found to
increase in proportion as we descend.

The diagram (fig. 93) in the next page, may convey some idea of the
proportion which our continents and the ocean bear to the radius of the
earth.[760] If all the land were about as high as the Himalaya
mountains, and the ocean everywhere as deep as the Pacific, the whole of
both might be contained within a space expressed by the thickness of the
line _a b_; and masses of nearly equal volume might be placed in the
space marked by the line _c d_, in the interior. Seas of lava,
therefore, of the size of the Mediterranean, or even of the Atlantic,
would be as nothing if distributed through such an outer shell of the
globe as is represented by the shaded portion of the figure _a