Geographical Journal (Journal of the Royal Geographical Society): Volume VIII, No. 5 [‎33r] (68/154)

A JOURNEY IN THE VALLEY OF THE UPPER EUPHRATES.
471
uothing seems to be known; it does not occur in any of tbe early lists of legions,
and would seem not to have come into existence till the third century a.d.,
and to have been one of the lesser auxiliary legions.* In the reign of Septimius
Severus, asjiogarth has shown, the great road Caisarea—Melitene was improved on a
magnificent scale, but the only evidence for similar improvements due to this
emperor along and to the north of the valley of the Euphrates is given by the
inscriptions on the great bridge at Kiakhta. Of the later emperors, Decius
restored the bridge over the Kara Budak, and there is a milestone, probably the
second from Melitene, of the emperors Constans and Constantins, either on the
road Melitene—Sebastea or Melitene—Satala. The inscription of Aurelian at Satala
need not be any evidence of his having improved that camp, and may be merely an
inscription in honour of his great achievements in the East. In still later times,
we know that the fortifications at Melitene f and Satala X were repaired by
Justinian.
Of the arrangement of the posts and disposition of the troops along the frontier
very little is known. Between Samosata and Melitene, Claudiana and Metita are
mentioned as posts in the Notitia; Barzalo and Laudias are called castra prxsidiaria
by Ammianus Marcellinus,§ and may be the same as Barsalium, Claudia and Metita
of the Peutinger Table. Between Melitene and Satala, along the road of the
Antonine Itinerary, seven of the ten stations (all except Carsagis, Sinibra, and
Zimara) are assigned troops by the Notitia. || Between Nicopolis and Satala
Aladaleariza of the Notitia may be the same as the Olotoedariza of the Table and
Itinerary. North of Satala, on the road to Trapezus, Domana and Zigana were both
stations of troops. So that an unbroken line of posts at fairly regular intervals
must have existed in later times along the river as far as Satala, and been extended
northwards to Trapezus.
With regard to the strategical importance of these posts, Samosata and Melitene,
two of the three great legionary camps, guarded two important crossing-places of
the Euphrates. Barsalium and Claudia, as may be inferred from the passage of
Ammianus Marcellinus (loc. cit.), probably were chosen with the same object. If
Dascusa was situated opposite Pingan, it doubtless protected another crossing-place.
The river is narrow at this point, and a bridge could always have been thrown over
it with little trouble similar to the wooden bridge now existing there. It is probable
that another station was near the crossing at Keban-Maden, but no troops are
assigned to the other Dascusa {vide, p. 466), which should be near this point. The
importance of Satala seems to lie in its position on the route to the East, which was
still much used until the middle of this century,f and in the advantages which it * * * §
* The name is confirmed by ‘ C.I.L.,’ vol. iii. 306; vide also vol. iii. Suppl. 6/46 ,
but I do not know that it is otherwise mentioned. In earlier times, possibly the
garrison was furnished by the twelfth and fifteenth legions, as inscriptions of both these
have been found at Trapezus {vide ‘ C.I.L.,’ vol. iii. Suppl. 6745,6747) .
t Procop., ‘ De J3d.,’ iii. 5. + iii. 4.
§ XVIII. 7. 11.
|| ‘Notit. Dignit. Orient.,’ cap. 32 and 35. ;.
1 This route goes from the modern Sadagh to Erzerum, and passes Ilija, usually
identified with Elegeia. Trajan went from Satala to Elegeia (Dio. Cass., Ixviii.).
Ilija, it must be remembered, however, is a very common Turkish name for a natural
hot spring, and is used in many places all over the Ottoman empire. If its identity
here with Elegeia could be proved from other evidence, it would aftord an interesting
example of the common practice of preserving an ancient name, slightly altered so as
to bear a meaning in a modern language.