ch the Medes seem to have carried with them from their early
eastern abodes, and to have applied to some high upland plains west
of the main chain of Zagros, which were peculiarly favorable to the
breeding of horses. As Alexander visited these pastures on his way from
Susa to Ecbatana, they must necessarily have lain to the south of the
latter city. Most probably they are to be identified with the modern
plains of Kbawah and Alishtar, between Behistun and Khorramabad, which
are even now considered to afford the best summer pasturage in Persia.

It is uncertain whether any of these divisions were known in the time of
the great Median Empire. They are not constituted in any case by marked
natural lines or features. On the whole it is perhaps most probable
that the main division--that into Media Magna and Media Atropatene--was
ancient, Astro-patene being the old home of the Medes, and Media Magna a
later conquest; but the early political geography of the country is too
obscure to justify us in laying down even this as certain. The minor
political divisions are still less distinguishable in the darkness of
those ancient times.

From the consideration of the districts which composed the Median
territory, we may pass to that of their principal cities, some of which
deservedly obtained a very great celebrity. Tho most important of all
were the two Ecbatanas--the northern and the southern--which seem to
have stood respectively in the position of metropolis to the northern
and the southern province. Next to these may be named Rhages, which was
probably from early times a very considerable place; while in the
third rank may be mentioned Bagistan--rather perhaps a palace than
a town--Concobar, Adrapan, Aspadan, Charax, Kudrus, Hyspaostes,
Urakagabarna, etc.

The southern Ecbatana or Agbatana--which the Medes and Persians
themselves knew as Hagmatan--was situated, as we learn from Polybius and
Diodorus, on a plan at the foot of Mont Orontes, a little to the east of
the Zagros range. The notices of these authors, combined with those of
Eratosthenes, Isidore, Pliny, Arrian, and others, render it as nearly
certain as possible that the site was that of the modern town of
Hamadan, the name of which is clearly but a slight corruption of the
true ancient appellation. [PLATE I., Fig. 2.] Mount Orontes is to
be recognized in the modern Elwend or Erwend--a word etymologically
identical with _Oront-es_--which is a long and lofty mountains s