[p.viii]It would seem, from the evidence regarding Petra which may be
collected in ancient history, that neither in the ages prior to the
[p.ix]commercial opulence of the Nabataei, nor after they were deprived
of it, was Wady Mousa the position of their principal town.

When the Macedonian Greeks first became acquainted with this part of
Syria by means of the expedition which Antigonus sent against the
Nabataei, under the command of his son Demetrius, we are informed by
Diodorus that these Arabs placed their old men, women, and children upon
a certain rock [Greek text], steep, unfortified by walls, admitting only
of one access to the summit, and situated 300 stades beyond the lake
Asphaltitis. [Diod. Sic. l.19.c.95, 98.] As this interval agrees with
that of Kerek from the southern extremity of the Dead Sea, and is not
above half the distance of Wady Mousa from the same point; and as the
other parts of the description are well adapted to Kerek, while they are
inapplicable to Wady Mousa, we can hardly doubt that Kerek was at that
time the fortress of the Nabataei; and that during the first ages of the
intercourse of that people with the Greeks, it was known to the latter
by the name Petra, so often applied by them to barbarian hill-posts.

When the effects of commerce required a situation better suited than
Kerek to the collected population and increased opulence of the
Nabataei, the appellation of Petra was transferred to the new city at
Wady Mousa, which place had before been known to the [p.x]Greeks by the
name of Arce [Greek text], a corruption perhaps of the Hebrew
Rekem.[Joseph. Antiq. Jud. l.4,c.4.] To Wady Mousa, although of a very
different aspect from Kerek, the name Petra was equally well adapted;
and Kerek then became distinguished among the Greeks by its indigenous
name, in the Greek form of Charax, to which the Romans added that of