, such as the apparently sudden
introduction of new forms throughout all past time, and the common
occurrence of wholly isolated types, which cannot be explained
in this way. Whilst it seems certain, therefore, that many of
the phenomena of the succession of animal life in past periods
can only be explained by some law of evolution, it seems at the
same time certain that there has always been some other deeper
and higher law at work, on the nature of which it would be futile
to speculate at present.

Not only do we find that the animals of each successive formation
become gradually more and more like those now existing upon the
globe, as we pass from the older rocks into the newer, but we also
find that there has been a gradual progression and development
in the _types_ of animal life which characterise the geological
ages. If we take the earliest-known and oldest examples of any
given group of animals, it can sometimes be shown that these
primitive forms, though in themselves highly organised, possessed
certain characters such as are now only seen in the _young_ of
their existing representatives. In technical language, the early
forms of life in some instances possess "_embryonic_" characters,
though this does not prevent them often attaining a size much
more gigantic than their nearest living relatives. Moreover, the
ancient forms of life are often what is called "comprehensive
types"--that is to say, they possess characters in combination
such as we nowadays only find separately developed in different,
groups of animals. Now, this permanent retention of embryonic
characters and this "comprehensiveness" of structural type are
signs of what a zoologist considers to be a comparatively low
grade of organisation; and the prevalence of these features in
the earlier forms of animals is a very striking phenomenon, though
they are none the less perfectly organised so far as their own
type is concerned. As we pass upwards in the geological scale,
we find that these features gradually disappear, higher and ever
higher forms are introduced, and "specialisation" of type takes
the place of the former comprehensiveness. We shall have occasion
to notice many of the facts on which these views are based at
a later period, and in connection with actual examples. In the
meanwhile, it is sufficient to state, as a widely-accepted
generalisation of palaeontology, that there has been in the past
a general progression of organic types, an