Copyright
Henry Fanshawe Tozer.

Researches in the highlands of Turkey; including visits to mounts Ida, Athos, Olympus, and Pelion, to the Mirdite Albanians, and other remote tribes (Volume 1) online

. (page 29 of 31)
Online LibraryHenry Fanshawe TozerResearches in the highlands of Turkey; including visits to mounts Ida, Athos, Olympus, and Pelion, to the Mirdite Albanians, and other remote tribes (Volume 1) → online text (page 29 of 31)
Font size
QR-code for this ebook


innermost angle of the Thessalian plain. From this
point the rivers of Albania radiate in different directions
to the north-west, the Viosa towards Tepelen and Avlona,
to the south the Arta flowing into the Ambracian Gulf,
and the Aspropotamo (Achelous), which waters Acar-
nania. Westward, in the heart of the mountains, lies the
lake of Yanina, whose waters have no visible outlet ; and
from the groups of mountains in its neighbourhood, the
Kalamas descends to the sea opposite Corfu."

The city of Uskiub lies in the slope of a low valley
reachino- to the river, having on one side of it the castle
hill, on the other a lower hill covered with gravestones,
of which also a vast number extend round the upper
part of the place, testifying to the large population it
must once have contained. The present inhabitants are
said to amount to about 21,000, of whom 13,000 are Ma-
hometans, 7000 Christians of the Greek Church, and 800
Jews f but there is a look of decay about the city, and
its glories are of the past. As compared with the other
cities of this part of Turkey, it has a great look of anti-
quity, which is especially apparent in the baths and
minarets. The khan in which we lodged was a fine
specimen of these old buildings — a brickwork structure,

7 See on this subject Grisebach, chap. x\., passim; and on the distribu-
tion of the tribes of this district in ancient times, sec chap. xxv. of Grote's
* History of Greece. '

« Hahn, ' Reise,' p. 64.



Chap. XVI. Tlic Kurschumli-khan. 367

built round a spacious quadrangle, with two rows of
stone arches and pillars, one above another, supporting
its corridors and galleries. It is entered by a gateway
secured with strong iron-bound doors ; in the centre
stands a large stone basin, which once contained a
fountain, and at the back of the building are excellent
stables. The whole place is massive, and very pictu-
resque. In former days, when there was an extensive
trade between Ragusa and Uskiub, it was a great resort
for Ragusan merchants, and in one place a Slavonic
name, which is thought to have belonged to one of them,
is inscribed in large red letters on the wall. From the
arched gallery of the upper story, which is reached by two
stone staircases, doors open out into square apartments,
which were occupied by these merchants ; most of them
are now left to decay, but a few we found in repair, and
still tenanted. On the outside are seen the g-rated
windows of the upper story, together with the dome-
shaped attic roofs, covered with lead, from which it gets
its name of Kurschumli-khan, or Lead Hotel. The bridge
by which the Vardar is crossed was originally composed
of nine arches, seven of which still remain, and the piers
of two others, over which woodwork is now thrown ; the
stones of which it is built are very large, and the piers
very strong to resist the force of the rushing stream. The
appearance of the workmanship, and the level roadway
which passes over it, so different from the steep ascent of
ordinary Turkish bridges, leaves little doubt that it
dates from Roman times. In the castle walls there
is also work which is evidently Roman.

In the upper part of the city there is a lofty clock-
tower, the lower part of which is of stone, — the upper, in
which the clock is contained, of wood ; it is mentioned



368 Prisrend to Uskiiib. Chap. XVI.

by the English traveller Brown,^ who passed by here in
the seventeenth century, and before his time by the
Mahometan writer Hadji Khalfa, in his description of
Uskiub.^° The last-named writer speaks of the clock as
dating from the time of the " unbelievers," and as being
famed for its size and sound, so that its striking was
heard at the distance of three hours' journey from the
city. As it had not been investigated since that time, we
were anxious to know whether it still remained, and
accordingly laid siege to the tower, using a big stone in
default of a knocker, for the door was fastened, and we
could hear the keeper moving about in the upper story.
At first he pretended not to hear us, and when at last he
descended, a long palaver ensued before we were ad-
mitted ; for, as he told our dragoman privately, he could
not understand why we should want to examine the
country, unless it was with the view of coming to conquer
it afterwards. When the door at last was opened, we
ascended the steep wooden staircase, which had been
rendered neither cleaner nor safer by the multitude of
pigeons that tenanted the tower, until we reached the
clock ; from this to the bell, and to an opening above it,
the stairs were very rotten and rickety, but we were
rewarded for our trouble by a superb bird's-eye view
over the country, including the city itself, the river and
its plains, the Karadagh, about ten miles off to the north,
and the distant range of Scardus to the west. The bell,
which we examined, had no marks to explain its origin
or date, and the old clock has unhappily been broken
and removed (so the keeper told us), and has been
replaced by a new one.

9 ' Travels,' p. 33.

i» ' Rumeli and Bosna,' p. 95, quoted by Hahn,' ' Reise,' p. 63.



Chap. XVI. Justiiiiaiis Aqueduct. 369

Early on the morning after our arrival we visited an
aqueduct, which lies a mile and a half beyond the
northernmost extremity of the city, and is still used for
its original purpose, to bring water into the city from the
lower slopes of the Karadagh. This structure is specially
interesting, because in all probability it was erected by
the Emperor Justinian, when he adorned the city with
buildings, in commemoration of his having been born in
the immediate neighbourhood." It crosses a depression
in the ground over which the watercourse had to be
carried, and is composed — not, as Brown says, of 200
arches, nor, as Hahn computes it, of 120, but of 53
round arches supported by strong piers, over which, in
the intervals of the main arches, the masonry is pierced
by small arches mostly pointed, though some here and
there are round. In the lowest part of the depression its
height is about forty feet, and at one point in its course
it makes an angle ; the material is mixed brick and
stone, rather roughly put together, except that of the
arches, which are entirely of brick ; the whole structure
supports the watercourse, which is composed of stones
and rubble, and covered in at the top. The architecture
throughout is Byzantine.

As we returned, we noticed at the door of a public
building in the suburbs two armed men, who wore a
peculiar head-dress of a rough brown material, in appear-
ance something between our guards' bearskin and the
chimney-pots of the dervishes, and ornamented with a
knob at the crown. On enquiry we found they belonged
to a Circassian colony which had lately been established
in this place by the Turkish government. In doing so

" See Appendix E, on The Birthplace of Justinian.
VOL. I. 2 13



370 Prisrend to Uskiub. Chap. XVI.

they have followed the example of their forefathers, for,
as I have already mentioned, when the Turks first con-
quered the city, they placed an Asiatic colony there. By
some persons the planting of these colonies of Circas-
sians, the most fanatical of Mahometans, in the inland
districts is regarded with great suspicion, as being in-
tended as a demonstration against the Christian popu-
lation.



( 371 )



CHAPTER XVII.

THE VARDAR VALLEY.

Justinian's Birthplace — Kiuprili — Unexplored Route to Salonica — The
Site of Stobi — Negotin — Banja — -Demirkapu or Iron Gate of the
Vardar — Boats Shooting the Rapids — Traffic to Perlepe — Lower
Coiu-se of the River — Ardjen Lake — Avret Hissar — Arrival at Salo-
nica — Railway Route across Turkey — Lines to India — Migrations of
Labourers — Commercial Treaty with England — The Eastern Ques-
tion — Greek and Slavonic Races — Future Prospects of Turkey.

On leaving Uskiub we rode for four hours over a long
plain which skirts the Vardar, with fine views of the
majestic Liubatrin behind us, while at one point a lofty
distant peak was seen through the nearer mountains to
the west, perhaps one of those on the hither side of the
Perim-dagh or Orbelus. A great quantity of corn is
grown in the plain ; and where the land was left uncul-
tivated, large herds of cows, horses, and buffaloes, were
feeding on the rank grass. These last-named animals
are also used here for drawing cars ; and when left at
liberty may often be seen immersed in muddy pools, to
which they betake themselves as a refuge from the flies ;
occasionally I have seen their whole bodies encrusted
with a coating of grey slime, bearing evident traces of
their mud bath. In one part of our route we passed a
large building, which serves as a factory or storehouse
for saltpetre, which is collected in the neighbourhood and
carried to Constantinople to be used in the manufacture
of gunpowder. At the southern end of the plain the
river makes a sudden bend and enters a narrow defile, in
which the rapids must be very considerable, to judge by

2 B 2



372 The Vardar Valley. ' Chap. XVII.

the difference of elevation between the level of Uskiub
and that of Kiuprili in the plain below it ■} just at its
commencement are the villages of Taor and Bader, which
are believed to correspond to the Tauresium and Bede-
riana of ancient times." The former of these villages, as
having been Justinian's birthplace, was visited by Von
Hahn in the winter of 1858. He found it to be situated
on a small spur projecting towards the plain from the
rocks which form the defile, and overhanging the river ;
above it rises a small plateau, well suited to be the posi-
tion of the Tetrapyrgion, or square castle, which Pro-
copius describes as having been built by Justinian ; and
here, the peasants told him, in ploughing they meet with
traces of old masonry, and in one part there are remains
of a watercourse of brickwork covered in with tiles.

Above the defile the river is joined by a tributary
flowing from the east, the Egri-su, which even at this
time of year had a considerable stream ; but throughout
this plain and the Tettovo we had observed, what we
before noticed on the other side of Scardus, the remark-
able freshness of all the vegetation, the explanation of
which was to be found as well in the amount of rain that
had fallen during the spring, as in the upland character
of these districts. Close to this stream we stopped
during the heat of the day at the Kaplan Khan ; and on
resuming our journey left the river and made our way by
a pass in the mountains through the spur which forms
one side of the defile ; from this we descended into a
lower and smaller plain, at the further end of which lies
the town of Kiuprili, or, as the Christians call it, Velesa.
The scenery of this part was extremely wild and barren ;
a small quantity of corn was grown on the nearer slopes,

> Uskiub is by the barometer 855 feet, Kiuprili 565 feet above the sea.
' See Appendix E., on The Birthplace of Justinian.



Chap. XVII. Kiuprili. 373

.but all the mountains beyond were wholly uncultivated.
Just before entering the place a single plane-tree greeted
our eyes, the first we had seen throughout the whole of
this journey, giving signs of the approach of more luxu-
riant growths and more varied foliage than those to which
we had lately been accustomed. Mulberry trees also are
grown here in considerable numbers for the sake of
■the silkworms which are reared by the inhabitants, and
their produce sold to merchants, principally Italians, who
come hither to fetch it for exportation.

The appearance of Kiuprili surprised us. We had
expected to find it an insignificant place, for it is never
named among the more important cities of Turkey ; but
instead of this, it presents a very imposing aspect from
its numerous well-built houses, and has an excellent
khan, which, from the blue and yellow with which its
front is decorated, is known as " the Painted Khan." We
were told that it is the first town, as you emerge from the
central part of European Turkey, where the Christians
have comparative liberty, and enjoy something like pros-
perity. During our stay we often heard Greek spoken in
the streets ; and though the population is estimated at
25,000,^ only six or seven minarets were to be seen. It
would almost seem as if the former greatness of Uskiub
had migrated here. Its position on sloping hills on both
sides of the Vardar, at the entrance of a narrow defile, is
extremely striking ; in places the steep cliffs rise close
.above it, and at one point there is a nook far up on the
hill-side to the west, in which stands a newly-built church
in a pretty position, though from being closed in by
•bare rock it must have the temperature of a furnace.
The two parts of the town are united by a long wooden

' i.e., 5000 houses, as given by Grisebach, ii. 223.



374 The Vardar Valley. Chap. XVII.

bridge supported on stone piers, from which its Turkish
name is derived. The Christian name of Velesa is pro-
bably a corruption of Bylazora, the name of the old
Greek town which occupied the same site ; it is one of
those important positions which are naturally occupied
and defended from an early time, as it commands the
entrance of the defile, the key of the lower country, from
the north, the side from which barbarian invaders would
be most likely to come.

From Kiuprili our journey became one of exploration,
for information was wholly wanting about the lower part
of the Vardar valley, as no traveller seemed hitherto to
have taken this route. Strange though it may appear,
since this is the direct way, and presents no difficulties to
the traveller, yet the post road or track from this place
to Salonica, which was followed by Grisebach and Hahn,
makes a long detour westwards into the mountains to
Monastir, and from thence descends by Vodena to the
sea ; the line of telegraph wires from Belgrade follows
the same direction. As might be expected, we had not
proceeded far from the beaten path before we found our
maps — even Kiepert's, usually so accurate — quite at fault ;
so that, though we had no great expectation of finding
many objects of interest, and could look forward to
increasing heat as we got further south and descended
into the plains, we had at least the satisfaction of breaking
new ground. Over this region I hope to carry my
readers somewhat rapidly, but I would advise those who
do not care for topographical details to avoid it alto-
gether.

The market of Kiuprili, which is held in an open space
close to the bridge, presented a busy scene as we passed
through it early in the morning on the ist of August, in
order to cross to the right bank of the river. Arrived



Chap. XVII. Unexplored Route. 375

there, we turned southwards through the defile, and after
passing the last houses of the city, noticed some old By-
zantine churches, now in ruins, which stand on a pro-
jecting spur beneath the lofty rocks overlooking fine
reaches of the river. Following its stream for some dis-
tance, we reached the point where a good-sized tributary,
the Babuna, flows in, descending from the mountains of
the same name ; through the deep valley in which it runs
lies the regular route to Monastir. This we crossed, and
leaving the Vardar, which here makes a bend, pursued
our course over country undulating in dull, desert-like,
stony plateaux. The only human being we saw in this
part was a shepherd-boy, playing shrilly on a curved
pipe, and followed by his flock feeding on the scant dry
herbs. At length we again descended towards the river
and more cultivated regions, where the barley had only just
been cut, and still lay out in sheaves. The maize grew
to an immense height, so high that our baggage-horse,
unable to resist the temptation, became almost invisible
in the green thicket. The river's course to our left could
be traced by a partial line of poplars and willows, and
between our track and it we passed scattered villages,
miserable hovels of unbaked brick. Throughout our ride
we passed excellent springs ; but requiring shade for our
mid-day halt, we struck off a little way from our path,
and stopped at the small village of Gratschan. Both the
houses here and the neighbouring land belong to a Turk, at
whose steward's house we stopped in the middle of the
day : the fields are cultivated by Bulgarian peasants, who
have half the produce, according to the metayer system,
which is common in Turkey. They were a heavy-looking
set of people, like most of their race ; but whether from
natural dulness or from oppression, or both, it would be
hard to say. The upper room, where we rested, was



3/6 The Vardar Valley. Chap. XVII.

open on two sides to the air ; and in it were two speci-
mens of the giizla — one of the same kind as w^e had seen
in Montenegro, the other much smaller, and shaped like
a guitar, with several jingling wire strings : this is the
most common form of the instrument among the Bul-
garians. At the foot of the wooden staircase which led
up to this chamber outside the building, was a slice of
the capital of a marble column, which, the proprietor told
us, was brought from the neighbouring village of Czerna
Gratzko. This place is situated about two miles off, on
a hill rising above the river Czerna, the ancient Erigon,
which carries off the waters of the plain of Monastir, and
is the largest tributary of the Vardar : here it is a wide
stream, and we had to search for some time before we
could discover a safe ford. The course of this river,
which passes between Mount Nidje and the end of the
Babuna chain, has never yet been explored, and well
deserves the attention of future travellers.

The position of Czerna Gratzko is precisely such as the
Greeks were accustomed to choose for their towns — a lower
height at the end of a range of mountains, separated'by a
depression from those behind, and projecting into a plain
which it thus commands, while a river makes a bend under
its walls. The site is now occupied by a walled Bulgarian
village, the house of the chief man being placed at the
angle which overlooks the river, and supported on high
stone foundations. There can be little doubt that this
place represents the important town of Stobi, wdiich in
Roman times was the meeting-point of four great roads ;
one from the Danube by Scupi (Uskiub) ; another from
Serdica, near the modern Sophia, to the north-east ; a
third from Heracleia (Monastir), to the south-west, thus
forming a line of connection with the Egnatian Way ;
and a fourth to Thessalonica. It is mentioned in the



Chap. XVII. The Site of Stohi. 377

Tabular Itinerary* as being 47 Roman miles from Hera-
cleia, and 55 from Tauriana (Doiran), which is 33 from
Thessalonica : these distances, as far as we can at
present judge, agree very fairly with this position ; and
still more exact is the distance of 23 Roman miles from
the Stena, or Iron Gate of the Axius, at which we arrived
on the following day. Again, as it was on the road
from Thessalonica to Scupi, we should expect it to be
near the Axius ; and it is described by Livy as being a
town of Pseonia, in the district Deuriopus, which was
watered by the Erigon :^ the importance of this position
also at the junction of two considerable rivers is in its
favour, for the Czerna joins the Vardar less than a mile
below. That it is an ancient site is shown by the piece
of a column which had been brought from thence. I
inquired for coins, but could not hear of any. After
crossing the river, we ascended the height behind the
village, which is the highest point of these hills, thinking
that possibly the city might have stood there instead of
being on the lower spur, but we found no traces of ruins ;
and I have little doubt that Czerna Gratzko itself is on
the site of Stobi.

The heights just mentioned belong to one of the
numerous spurs which from time to time are thrown out
towards the Vardar, in this part of our route, from the
loftier chains, which run parallel to it at a distance of
about 20 miles to the west, and from 10 to 15 to the
east : the still higher range which at intervals appeared
far away to the west was probably the Babuna. From
Czerna Gratzko we rode over a succession of low hills to
Negotin, or Tikvesh, which was to be our resting-place
for the night — a poor country-town, though far superior

* 6"^^ Leake's 'Northern Greece,' iii. 441.

* Livy, xxxiii. 19, xxxix. 53, xlv. 29.



378 The Vardar Valley. Chap. XVIL

to all the other places between Kiuprili and Salonica : it
is divided into a Turkish and a Christian quarter, the
former of which is distinguished by its mosque, the latter
by its clock-tower. This place is situated at some dis-
tance from the river, which here makes a considerable
bend to the east ; we were much puzzled in finding it, as
the two names are marked in Kiepert as representing-
two separate places. To us, both in the town itself and
in the neighbouring village of Islam-Koi, the natives
declared that they were two names for the same place ;
at a khan near Marvinsta, however, a day's journey fur-
ther south, the khanji said that Tikvesh is the name of
the district, Negotin of the town ; and the same account
is given by Dr. Barth," whose route through Turkey, in
1862, here cuts across ours. These conflicting statements
can only be reconciled by supposing that while the town
is called Negotin, of which there is no doubt, the name of
Tikvesh is applied both to it and to the district. Certain
it is that double names are frequently found in these
parts, both for places and features of the country ; and
though in many cases this arises from the mixed Turkish
and Bulgarian population, yet apparently it does not
always proceed from this cause.

At half-past five the next morning we were on our
way, still keeping on the low hills, from which we occa-
sionally obtained peeps of the river, until we gradually
descended to it near the miserable village of Banja, which
lies about a mile-and-a-half above the Iron Gate. As
the city of Antigoneia, according to the Tabular Itine-
rary, was situated about half-way between Stobi and the
Stena, its remains, if any exist, ought to be found some-
where in this part of the country ; however, notwith-

^ 'Reise durch das Innere der Europalschen Tiirkei,' p. 120.



Chap. XVII. Iron Gate of tlic Vardar. 379

standing many inquiries, we could hear nothing of anti-
quities, either at Negotin or Banja, or anywhere else
along our route, nor did we see any squared stones or
pieces of marble in the villages. Shortly before reaching
Banja there is a hill suitable for the site of a city, in a
position somewhat resembling that of Czerna Gratzko,
and commanding the entrance to the defile : this, how-
ever, is too far south for what is required, and at Banja
we were informed that there were no old walls on its
summit. Throughout this part of the country the soil is
composed of a sandy clay, which is very friable and pow-
dery, and consequently ill-suited for preserving the traces
of ancient cities. Below Banja, and just above the defile,
is a wide river-bed, in which a narrow stream of clear
water was flowing ; all about its bed dwarf plane-trees
were growing, and this was the first place where we had
seen any number of them ; below the defile they appear
at once to have entered on a different temperature, for
there they form the principal vegetation, and grow most
luxuriantly.

The Demirkapu, or Iron Gate of the Vardar, resembles
in its main features the more famous defile of the same
name on the Danube, being a passage between steep
walls of rock, through which the river forces its way in a
series of rapids. The first of these commences just above
the pass, and makes a sharp turn on entering it ; then
succeeds a long reach of calm water, until another rapid
is formed towards its exit. On the two sides, lofty cliffs
of grey limestone patched with red rise almost precipi-
tously over the river, bearing feathery trees and shrubs
in their crevices. The path lies on the right bank ; just
at its entrance, at the angle formed by the first rapid, a
Turkish guard-house is built over an archway through
which the road passes, and shortly after this, where a



38o The Vardar Valley. Chap. XVII.

large rock on your left hand lies detached from the pre-
cipices, a passage has been cut through, traces of the
work being left in the ribbed lines and grooving with
which its sides are marked. This is probably Roman
work, and we saw exactly similar marks on the rocks in
the pass of Tempe ; but it may date from a still earlier



Online LibraryHenry Fanshawe TozerResearches in the highlands of Turkey; including visits to mounts Ida, Athos, Olympus, and Pelion, to the Mirdite Albanians, and other remote tribes (Volume 1) → online text (page 29 of 31)