find
in limited areas, serve to convince us how imperfect and fragmentary
must be our knowledge of the earth's fauna at any one past epoch; since
we cannot believe that all, or nearly all, of the animals which
inhabited any district were entombed in a single lake, or overwhelmed by
the floods of a single river.

But the spots where such rich deposits occur are exceedingly few and far
between when compared with the vast areas of continental land, and we
have every reason to believe that in past ages, as now, numbers of
curious species were rare or local, the commoner and more abundant
species giving a very imperfect idea of the existing series of animal
forms. Yet more important, as showing the imperfection of our knowledge,
is the enormous lapse of time between the several formations in which we
find organic remains in any abundance, so vast that in many cases we
find ourselves almost in a new world, all the species and most of the
genera of the higher animals having undergone a complete change.

_Causes of the Imperfection of the Geological Record._

These facts are quite in accordance with the conclusions of geologists
as to the necessary imperfection of the geological record, since it
requires the concurrence of a number of favourable conditions to
preserve any adequate representation of the life of a given epoch. In
the first place, the animals to be preserved must not die a natural
death by disease, or old age, or by being the prey of other animals, but
must be destroyed by some accident which shall lead to their being
embedded in the soil. They must be either carried away by floods, sink
into bogs or quicksands, or be enveloped in the mud or ashes of a
volcanic eruption; and when thus embedded they must remain undisturbed
amid all the future changes of the earth's surface.

But the chances against this are enormous, because denudation is always
going on, and the rocks we now find at the earth's surface are only a
small fragment of those which were originally laid down. The
alternations of marine and freshwater deposits, and the frequent
unconformability of strata with those which overlie them, tell us
plainly of repeated elevations and depressions of the surface, and of
denudation on an enormous scale. Almost every mountain range, with its
peaks, ridges, and valleys, is but the remnant of some vast plateau
eaten away by sub-aerial agencies; every range of sea-cliffs tell us of
long slopes of land destroyed