l idea
regarding the Picts. Here the identification was closer still. Then
came the consideration: The fairies lived in hollow hillocks and under
the ground: what kind of dwellings are the Picts supposed to have
occupied? The answer to this question still further strengthened Mr.
Campbell's conjecture. There yet exist numerous underground structures
and artificial mounds whose interior shows them to have been
dwelling-places; and these are in some places known as "fairy halls" and
in others as "Picts' houses." (Illustrations of these are shown in the
present volume, and are specially referred to in the annexed paper.)

The examination, therefore, of this interesting theory not only helped
greatly to bear out its probable correctness, but it further began to
appear that by following this method of inquiry new lights might be
thrown upon history--perhaps upon very remote history. It was clear that
the question was not a simple one. All tradition is obscured by the
darkness of time, and genuine fact is mixed up with ideas which belong
to the world of religion and of myth. Even in Mr. Campbell's own
statements there were seeming contradictions. These, however, it is not
my present purpose to discuss; since they do not vitally affect his main
contention.

The Lapp-Dwarf parallel was gone into very fully by Professor Nilsson in
his _Primitive Inhabitants of Scandinavia_, written twenty years before
the "West Highland Tales." Not that he, either, was the originator of
that theory, for it is frequently referred to by Sir Walter Scott, who
accepted it himself.[3] "In fact," he says, "there seems reason to
conclude that these _duergar_ [in English, _dwarfs_] were originally
nothing else than the diminutive natives of the Lappish, Lettish and
Finnish nations, who, flying before the conquering weapons of the Asae,
sought the most retired regions of the north, and there endeavoured to
hide themselves from their eastern invaders." Scott, again, refers us
back to Einar Gudmund, an Icelandic writer of the second half of the
sixteenth century, whom I would cite as the earliest "Euhemerus" of
northern lands, were it not for the fact that he is obviously much more
than a theorist, and is beyond all doubt speaking of an actual race, as
may be seen from an incident which he relates.

But, although the popular memory may retain for many centuries the
impress of historical facts, these become inevitably blurred and
modified by the lapse of