Geographical Journal (Journal of the Royal Geographical Society): Volume VIII, No. 5 [‎31r] (64/154)

A JOURNEY IN THE VALLEY OF THE UPPER EUPHRATES.
467
theory is adopted, the omission can he explained by the proximity of Dascusa, for
the troops there stationed would serve to protect the two posts.
The other site is that seen by Bore * near Melik-Sherif, and identified by
Kiepert with Arauraca.
With regard to the other stations on these roads, the only suggestions that we
•can make are that Dascusa of the Table and Itinerary may have been near Korpanik
-(see p. 330), that the ruins at Zinika (p. 455) may represent the old Zimara, and that
Analibla may have been at Hassan Ova, where we saw some signs of a Byzantine
site; but none of these identifications can be made with any degree of certainty.
Two fixed points on the main road on the right bank of the Euphrates are given
<1) by the bridge at Korpanik (p. 330); (2) by the bridge over the Kara Budak
(Sabrina) (p. 456). Between Malatia and Korpanik the country along the river is
easy; and, though we saw no certain traces of any Roman road during this part of
our journey, it is highly probable that the old road kept close to the earthy bank,
and possible that it has been long ago washed away. Above Korpanik, the
•country through which the river passes soon becomes mountainous, and from Egin
to Pingan the path along the right bank is reported to be most difficult by the
natives, and all traffic goes by the route which we followed on the left bank.
Hence it seems likely that the Roman road between the bridge at Korpanik and
the probable site of Zimara took a line to the west of the route by which we
travelled. Taylor,f on his way from Zimara to Arabkir, between Gamkeui and
Arabkir, high up on the Sarichichek Dagh, saw a paved road, which he speaks of as
being “ solidly and ingeniously constructed,” and it is not improbable that this
may be a portion of the road for which we searched in vain in the river-valley.
Moreover, the further west that we place this portion of the road, the more easy it
is to understand how Hispa and Arangas—which belong to the Melitene—
Sebastea road—were introduced among the stations of road lb. in the Peutinger
Table. , , , ^
Beyond the bridge over the Kara Budak, my personal knowledge of the country,
through which the road passed in two branches to Nicopolis and Satala, ceases,
and I will merely make one criticism of the views advanced by Ramsay as to the
■course which it took. From the lists of the Table and Itinerary, it appears to be
fairly certain that in the district which lies south of Nicopolis and Satala, there
was a triangle of roads of which the direct road Nicopolis—Satala was the base.
The apex of this triangle is placed by Ramsay at Carsagis, by Kiepert at Analibla.
Two sides of the triangle are made by the important roads Nicopolis—Satala and
Melitene—Satala, and the third side by the cross-road from Nicopolis to a point in
the Melitene-Satala road between Zimara and Satala. Ramsay makes several most
ingenious conjectures with regard to the stations Ole-Oberda, Aladaleariza, and
<laltiorissa, and makes out what is on paper a very fairly satisfactory restoration of
the roads. But in deviating from the view upheld by Kiepert, that road III.
■struck into the road Meliteue—Satala at Analibla, he leaves out of consideration the
■existence of a natural route which appears to answer very exactly to the require
ments of a road from Nicopolis to Zimara. The object in laying out a road from
Nicopolis to strike into the Satala—Melitene road would naturally be to make it join
the latter at a point in as direct a line as possible between Nicopolis and Melitene;
for the road in question cannot have been intended merely to join Satala and
Nicopolis, between which points a direct road is given in the Peutinger lable and
the Itinerary (Ho. and Ub.) This purpose would be best secured by a road going
* Referred to by Ritter, vol. xviii. p. 215.
t Loc. cii., p. 309.