faith, or very important in themselves, could either have been
propounded by the priests or received by the audience. And it may be
further observed, in corroboration of so self-evident a truth, that it
was held an impiety to the popular faith to reject the initiation of
the mysteries--and that some of the very writers, most superstitious
with respect to the one, attach the most solemnity to the ceremonies
of the other.

XVI. Sanchoniathon wrote a work, now lost, on the worship of the
serpent. This most ancient superstition, found invariably in Egypt
and the East, is also to be traced through many of the legends and
many of the ceremonies of the Greeks. The serpent was a frequent
emblem of various gods--it was often kept about the temples--it was
introduced in the mysteries--it was everywhere considered sacred.
Singular enough, by the way, that while with us the symbol of the evil
spirit, the serpent was generally in the East considered a benefactor.
In India, the serpent with a thousand heads; in Egypt, the serpent
crowned with the lotos-leaf, is a benign and paternal deity. It was
not uncommon for fable to assert that the first civilizers of earth
were half man, half serpent. Thus was Fohi of China [49] represented,
and thus Cecrops of Athens.

XVII. But the most remarkable feature of the superstition of Greece
was her sacred oracles. And these again bring our inquiries back to
Egypt. Herodotus informs us that the oracle of Dodona was by far the
most ancient in Greece [50], and he then proceeds to inform us of its
origin, which he traces to Thebes in Egypt. But here we are beset by
contradictions: Herodotus, on the authority of the Egyptian priests,
ascribes the origin of the Dodona and Lybian oracles to two
priestesses of the Theban Jupiter--stolen by Phoenician pirates--one