BookPDF Available

Despre insula Peuce, Genucla, Halmyris si cultura Murighiol, English Edition

Despre insula Peuce, Genucla, Halmyris
şi “cultura Murighiol”,
English Edition
Bucharest 2020
ISBN 978-973-0-31087-0
© Copyright: Elisie-Florin Scurtu
Preface to the English edition of the Romanian edition
(DR. ING. ELISIE-FLORIN SCURTU – Despre insula Peuce,
Genucla, Halmyris şi “cultura Murighiol”, Editura Universitară
Carol Davila, Bucureşti, 2016, ISBN 978-973-708-929-8) with
The presence of the Peuce (Peuke) Island in the Danube
Delta’s area was mentioned by many authors in Antiquity:
Herodotus (ca. 484 BC – ca. 425 BC), Apollonius of Rhodes (ca.
295 230 BC), Strabo (ca. 64 BC ca. 25 BC), Pomponius Mela
(approx. AD 43), Pliny the Elder (AD 23 - 79), Ptolemy (ca. AD
87 – ca. 165), Avienus Rufus Festus (ca. AD 305 - 375) and
others. Manius Laberius Maximus’ (Governor of the Moesia
Inferior) Horothesia from AD 100 is of significant importance, as
Peuce Island” is mentioned within it as a limit of Histria’s
territory, yet the “island’s” geographical position exists only
within the description of ancient authors and in less or more
adventurous presumptions made by certain contemporary
After having determined, since 2009 (SCURTU, F., 2009
– Imagini geofizice ale Histriei antice / Geophysical images of
Ancient Histria, ISBN 978-973-0-07257-0), the position of this
“island” based on a detailed analysis of the initial data
presented in ancient writings, as well as the current
geomorphological, geophysical and geological data (we
determined that Peuce Island” was an isle of approximatively
150 km2 area, on the Danube’s Sfântu Gheorghe arm, in the
Murighiol - Dunavăţ area), an other derived issue was that of
the ancient settlement from Murighiol, situated exactly on the
ancient Peuce “Island”. The authors of the archaeological
excavations carried out here between 1981 and 1987 and
subsequently, when Peuce’s position was yet to be specified,
gave this area the antique name of Halmyris. The analysis of
the area’s urban and historic conditions led us, however, to the
conclusion that the antique fort of Halmyris could not be at
Murighiol: it was situated (as it can be deduced from the
historic documents, Itinerarium Antonini and Tabula
Peutingeriana, corroborated with Google Earth images) in the
Sarinasuf-Colina area, close to Lake Halmyris and to the ancient
Peuce arm of the Danube, at approximatively 12 km west from
the position presumed by the aforementioned authors.
Also from the analysis of the two quoted classic
documents, corroborated with the region’s geomorphological
aspect, it resulted that the ancient Vallis Domitiana settlement
was, most likely, located within the Lăstuni village area, on the
Hagilar valley, immediately to the west of the current Tulcea-
Babadag road, and the ancient Ad Salices settlement was in the
Jurilovca village area (probably at Sălcioara or at Capul
The route of the Roman limes road between Aegyssus
(Tulcea) and Histria, as resulted from the author’s complex
investigations, is shown under figure 14 within this book.
The analysis of the information resulting from the
archaeological excavations (starting with those referring to the
LaTène culture and up to those pertaining to the 7th century
AD), presented in specialized publications which were
published between 1957 and 2003, corroborated with personal
observations and by mentioning certain situations and events
in the ancient writings and/or more modern (1906-1968) ones,
allowed for other conclusions, as well:
- the archaeologically explored ruins at Murighiol
correspond, certainly, to the ancient fort of Genucla, mentioned
under Cassius Dio (LI 26,5); this was the capital of the Peucini
Kingdom led in the year 29 BC by king Zyraxes;
- the name Vicus Classicorum, borne by the settlement
around the citadel in the 3rd century AD, according to the
inscriptions found during the excavations, was the translation
in Latin of the indigenous name Genucla;
- it is quite possible that Genucla was built on the ruins
of the ancient polis Orgame or that it was its descendant;
- it seems that, in the area of Movila Duna (the “Duna
hill”), 3 km west of Murighiol, there was an observation and
circulation post on the limes road between Salsovia and
Halmyris, maybe even the Gratiana fort;
- the Bastarnae probably occupied not only the Peuce
island, and self-entitled Peucini, as Strabo informs us, but the
entire area of current Dobrogea;
- the “Murighiol culture” is a reality: it is a particular
culture, predominantly Greek-Bastarnae;
- the settlement from Murighiol was not only a river
port, but probably a naval site, shipbuilder of ships of different
categories, even since the end of the past era and at least until
the 6th century AD;
- Cetatea Zaporojenilor (The Zaporozhians’ fortress),
from the Dunavăţu de Jos, was, most likely, a Peucini pirates’
Several less costly archaeological actions (reassessment
of the LaTène pottery in the area, aerial-photographic and
limited field surveys, soundings) may confirm and detail
certain conclusions of this book.
Carrying out certain soundings in the northern area of
the archaeological site at Murighiol would also be of great
importance in identifying the ruins of the over 2500 years old
ORGAME polis.
An other conclusion, which derives from the analysis of
the ancient documents, is that the correct name of the function
from Notitia Dignitatum “Praefectus ripae legionis primae Ioviae
cohortis ... et secundae Herculiae musculorum Scythicorum et
classis…, Inplateypegiis” would have been “…et classis Histricae,
In Plateypegiis”.
Publishing an English edition of the homonymous
Romanian book issued in 2016 was an imperative action: the
ancient Peuce Island’s exact geographic position has become
certain already in 2009 (Scurtu, 2009, “Imagini geofizice ale
Histriei antice”, ISBN 978-973-0-07257-0), but in the foreign
literature it is still considered uncertain, because the recent
information on this subject was published only in Romanian
(Scurtu, 20009, 2010, 2013, 2014, 2016 see the bibliography of
this book). The same is valid for the information concerning
the route of the Roman limes road between Noviodunum and
Callatis (published in 2013 and 2016) and also, of course, for the
spectacular information concerning the historical evolution of
the antique settlement from Murighiol.
The readers of the present edition have now the
opportunity to find out and analyse the arguments
substantiating the scientific conclusions enunciated in this
Bucharest, January 2020
The “Peuce island” was first spoken about (and written about!)
over two millennia ago, and the presence of the Peuce (Peuke) island
in the Danube Delta’s area was mentioned by many authors in
The oldest known document at the present time referring to the
Danube Delta is the work of Herodotus (ca. 484 BC ca. 425 BC),
who writes that the river Ister (= Danube) flows into the Euxin Pontus
at Istria, that ”the Ister..... has five mouths” (Histories,IV.47), that it
runs into the sea flowing towards the east and that the first
ramification of the river’s arms is at two days of sailing from its
outfall into the sea, where Darius had built a bridge over the river:
“So, the [Dareios’s] fleet sailed through between the Kyanean rocks
and made straight for the Ister; and then they sailed up the river a
two days' voyage from the sea and proceeded to make a bridge
across the neck, as it were, of the river, where the mouths of the Ister
part off.” (Histories, IV.89) and “the Ister is falling into the sea at this
point with its mouth facing the east” (Histories, IV.99);
The following useful information originates from Apollonius
Rhodius (ca. 295 – ca. 230 BC), who wrote: “For a certain island is
enclosed by Ister, by name Peuce, three-cornered, its base stretching
along the coast, and with a sharp angle towards the river; and round
it the outfall is cleft in two. One mouth they call the mouth of Narex,
and the other, at the lower end, the Fair mouth.” (Argonautica, 308-
In his Geographica , Strabo (ca. 64 BC ca. AD 25) wrote: “Near
the outlets of the Ister River is a great island called Peuce; and when
the Bastarnians took possession of it they received the appellation of
Peucini. There are still other islands which are much smaller; some
of these are farther inland than Peuce, while others are near the sea,
for the river has seven mouths. The largest of these mouths is what is
called the Sacred Mouth on which one can sail inland a hundred and
twenty stadia” (= circa 25 km) “to Peuce. It was at the lower part of
Peuce that Dareius made his pontoon-bridge, although the bridge
could have been constructed at the upper part also. The Sacred
Mouth is the first mouth on the left as one sails into the Pontus”. He
added: “According to Ephorus, however, the Ister has only five
mouths.” (Geographica VII.3.15, C305).
In his paper De Chorographia (approximatively in the year AD
43), Pomponius Mela mentions that “There are six islands among the
mouths of the Ister, of which Peuce is the most well-known and the
largest of these.” (De Chorographia II.87).
Plinius the Elder (C. Plinii Secundi, AD 23-79) wrote
(Naturalis historiae, IV, 12 (24), 79): “from the point where it first
enters Illyria it is called the Hister; after receiving 60 tributary rivers,
nearly half of which are navigable, it is discharged into the Black Sea
by six vast channels. The first of these is the mouth of Peuce, close to
the island of that name, at which the nearest channel, called the Holy
River, is swallowed up in a marsh 19 miles” (= ca. 13 km) “in extent.
Opening from the same channel and above Istere spreads a lake
measuring 63,000 feet” (= ca. 43 km) “round, named the Saltings
Ptolemy (ca. AD 87 –165) wrote that at Noviodunum (today’s
Isaccea), the Danube’s course splits in two, the most southern arm
going around an island called Peuce and pouring into the sea through
Hieron Stoma (the Holy Mouth) or Peuce Stoma (the two names being
considered equivalents), also offering geographic coordinates to the
different points mentioned under the text. He indicates that the
northern arm of the Danube consecutively splits into several smaller
arms, that finally pour into the Pontus; on one of the northern arms
(Thiagola or Psilon) in this area, at approximatively 0035’ (ca. 30 km)
west from its pouring into the Pontus, there is a lake called Thiagola
(Geographia, III.10.2). We will elaborate on this topic a little later.
Ptolemy also tells us (Geographia, III.10.4) that “The lands on
the western part of the Moesia Inferior are inhabited by the tribes’
people, and the eastern lands - lower of the Peuce mouth - by the
Troglodytes; and the mouths, by the Peucini”.
Figure 1 - Position of the Peuce Island in Tabula Europae from
Ptolemy’s Geographia
Figure 1 presents a detail from Ptolemy’s Tabula Europae IX,
which indicates the position of the Peuce Island, inhabited by the
Peuce Island made its way once again into the old historians’
comments as, in the year AD 370, this was the place where Alaric
was born. The future king of the Visigoths (between the years AD
395 and 410), after having commanded the Roman Gothic troops,
rebelled and invaded Italy, besieging Rome and even entering
Rome’s forum on the 24th of August 410, this being an event that
marked the fall of the Western Roman Empire: ALARIC, born
370, Peuce Island (now Romania) - died 410, Cosentia, Bruttium (now
Cosenza, Italy), chief of the Visigoths from 395 and leader of the army that
sacked Rome in August 410, an event that symbolized the fall of the
Western Roman Empire.” (Encyclopædia Britannica).
One of the sources of the information according to which Alaric
was born on the Peuce island, was, probably, Claudian, who, in his
panegyric dedicated to Emperor Honorius (384 - 423), De VI
Consulatu Honorii Augusti”, wrote: duo namque fuere Europae
Libyaeque hostes: Maurusius Atlas Gildonis Furias, Alaricum barbara
Peuce nutrierat, qui saepe tuum sprevere profana mente patrem.” (freely
translated: “two were the enemies of Europe and Libya: the ferocious
Gildo was fed by the Mauritanian Atlas, Alaric - by the barbaric
Peuce. They often defied your father with their respect-lacking
Peuce Island is also mentioned in the 4th century AD by
Avienus Rufus Festus (ca. 305 - 375) in Descriptio orbis terrarum (“The
Danube pours into the sea through 5 mouths, where the Peuce island
rises”) and later on (the middle of the 6th century AD) by Jordanes in
Getica: “Aside from these tribes, the Ostrogoths had Goths and
Peucini on the Peuce island, which is at the mouths of the Danube,
where it pours into the Pontus”.
A significant importance is held by the mention of the Peuce
Island in Manius Laberius Maximus’ (the governor of Moesia Inferior)
horothesia from the year AD 100 regarding the borders of Histria (we
underline the fact that this is a reference to the Peuce Island” made
by a Roman official from that time, it is not a quote by an ancient
author, therefore “Peuce Island” did certainly exist at that time!).
Establishing an exact location of the Peuce Island on the
current map of Romania is a preliminary condition for establishing
the extent of Histria’s territory in Antiquity.
In the inscription which comprises this document, discovered
by Vasile Pârvan in 1914, it is mentioned that (Pippidi, 1956): “We
have established the Histrian people’s borders as follows: ... Peuce, Halmyris
lake, beginning with the land of ... the people of Argamum, from here along
the ridge ... up to the joining of the Picusculus and Gabranus rivers, then
from the Grabanus river’s mouth up to its waterhead, from here ... along the
Calabaeus river (toward the sea?) approximatively ... thousands of feet ...
(the suspension points represent the illegible parts of the inscription).
Numerous attempts were made in time by professionals or
amateurs to specify the route of the former border on the current
map, including the localization of the mysterious Peuce Island; all of
these attempts had very different results.
The first interpretation of the horothesia text was given, of
course, by the discoverer of the inscription, Vasile Pârvan. For now,
we will not insist on the “continental” route of the border (this, too,
incurred different opinions); we will only quote the fact that Vasile
Pârvan admitted, with certainty, that the Peuce mouth was the
present day Sf. Gheorghe mouth, and the Halmyris Lake was
identified with the Razelm Lake, expanded through the Zmeica and
Golovița lakes.
In a manuscript (Contribuţiuni la istoria Histriei/Contributions to
the history of Histria), professor Paul Nicorescu from Iași, considers
that Peuce Island would have meant “the entire complex of islands
found between the exterior arms of the Delta and the sea, that is the
entire delta” (Pippidi, 1956).
Another author quoted at the same time (Pippidi, 1956),
Captain M.D. Ionescu-Dobrogianu, issued the hypothesis according
to which the “sacred” arm would have originated at Eski Kale, near
Isaccea, pouring into the Razelm Lake through the Babadag Lake
and the marshes called Coada Bălții. According to the same study,
the Peuce Island must have comprised, at the beginning of the 2nd
century, “not only the alluvial part ... but also a continental part, that
is the triangle between Isaccea, the Babadag Lake and the Razelm
Lake, up to Dunavăț”, an opinion previously issued (1851) in a
dissertation entitled De Istri ostibus (Kruse, Breslau 1851, p. 53) which
was “heatedly defended by a whole series of Russian, Bulgarian and
American researchers” (Pippidi, 1956).
In Polaschek’s (Austrian historian) opinion (Pippidi, 1956), in
the 5th century BC and later on, the Peuce arm would have
communicated with the Halmyris through a secondary arm, today
called the Dunavăț, therefore Peuce would have had two mouths
throughout time: one - the Dunavăț mouth (τό Πεύχης Στόμα – the
Peuce mouth, in the term’s strictest sense) and the other "’Iερψ
Στόμα” which became, at one point, the primary mouth of the river,
today called Sfântu Gheorghe. As the author we quoted also shows
(Pippidi, 1956), this hypothesis contradicts Herodotus’ text, which
entails the Danube pouring into the open sea, and that Halmyris was,
at that time, according to the horothesia, a lake.
More recently, a presumption was issued that the Peuce island
would probably correspond to the accumulative formation
(“grind/sand dune”) Caraorman, to which the relict elements of the
pre-delta terrain were added, thus being located between the current
Sulina and Sfântu Gheorghe arms, in the area of the Caraorman
“grind” (Panin, 1983).
In developing our scientific investigation concerning Peuce
Island, we will use geomorphological, geological and geophysical
arguments, starting from the following ancient documented records
previously detailed and which we consider most likely:
- the southern mouth of the Danube (Ister) was
simultaneously called Hieron stoma and Peuce stoma;
- the river poured into the sea flowing toward the east;
- at its mouth, there was a large island (Peuce), of a triangular
shape, with its base towards the sea and with a sharp angle towards
the river included in the Ister’s route, that splits into two arms
around it;
- the most southern arm was called Peuce and it went around
(on the south side, evidently) the homonym island;
- there was a distance of circa 25 km from the Hieron Stoma,
upstream to the Peuce island, and near this island stood the ship
bridge built by Darius;
- the first arm, Peuce, was absorbed by a marsh measuring
19,000 feet (circa 13 km), and from here, to Histropolis, a lake
measuring 63,000 feet (circa 43 km) in circumference unfolds, called
This information is credible, given that it was obtained by at
least five ancient authors, in six different periods of time spreading
over seven centuries, and that the respective information does not
overlap, referring to different aspects of the issue at hand, therefore,
it was not copied (possibly with errors) from one author to another.
There are small contradictions between these written records only
with regards to the number of mouths that the Danube had at the sea
(in the case of Herodotus and Avienus it had 5, Strabo - 7, and
Plinius, Pomponius Mela and Ptolemy - 6), situation which can be
explained through the subjective appreciation of classifying the small
local ramifications that poured into the sea as “arms”.
The results of our research, described hereinafter, confirm the
essential ancient information mentioned above.
The first punctual objective: the place where the Peuce arm
divided from the Sfântu Gheorghe arm (therefore, the most
upstream place in which one of the ancient Danube’s arms made a
large loop towards the south, as to go around the Peuce island).
The results of the international research on the geological
history of the Black Sea (Panin, 1983, citing the works of Almazov et.
al, 1963, Degens and Ross, 1972, Fedorov, 1972, Scerbakov et al, 1979)
indicate the fact that at the beginning of the Holocene (that started
approximatively 10,000 years ago, so around the year 8,000 BC - not
suddenly, of course, but through a vast transition period of the
glaciers’ melting after the Würm glaciation), the level of the Black Sea
was approximatively at the -50 m level, compared to the current one.
The sea level continuously increased afterward, reaching around the
year 3,000 BC up to the +5 m level compared to the current one, after
which it dropped to a minimum of approximatively -5 m around the
year 2,000 BP (the beginning of the first millennium AD) and then it
continued to rise to the current level (0 m). Certainly, these are
statistical values over great periods of time, but the sea level did
have smaller oscillations over smaller periods of time, as well. In any
case, grosso modo, we may admit the existence of a period of time in
which the Black Sea’s level was 5 m higher than it is now, as well as
the existence of a period of time (around the beginning of the first
millennium AD, probably even during the period of Histria’s
colonization) during which this level was approximatively 5 m lower
than it is today. Figure 2 presents the evolution of the Black Sea’s
level in the Holocene and the Upper Pleistocene (after Panin, 1983,
with simplifications).
Figure 2 - Variation of the level of the Black Sea in the Upper
Pleistocene (after Panin, 1983, with simplifications)
In order to have an image of the area during the Black Sea’s
maximum level, we started off from the levels of the current terrain,
based on which we built a sketch of the level curves that illustrate
the current geomorphology of the terrain (figure 3). Considering that
the shape of the land has not changed significantly over 5,000 years,
it therefore results that the Black Sea’s shoreline at around 3,000 BC
followed roughly the route of the +10 m level of the current terrain.
Figure 3 - Geomorphological sketch of the area NW from Razelm
Taking into consideration the aforementioned, we ascertain,
among other things, that during this period of maximum level, the
Black Sea occupied the entire area of the current delta, the Danube
flowing into the sea probably in the Galați - Măcin area, and the
Razelm Lake with the Babadag and Agighiol lakes made up a strait
of the Black Sea; the Taița, Telița and Valea Tulcea rivers poured
directly into the sea, as did Siret, Prut and Ialpug. It can also be
observed that in the Murighiol - Plopu - Sarinasuf area, there is an
arm linking the sea and the current Razelm Lake through the Fundea
gulf, this arm going around a high area corresponding to today’s
Dunavățu de Sus - Dunavățu de Jos area, which, at that time,
constituted an island near the Dobrudja’s shoreline. From the sketch
presented under Figure 3, we see that on the Danube, upstream from
Murighiol, there is no possibility of a river arm occurring south of
the Sfântu Gheorghe arm, as it would have to pass through the Telița
- Malcoci - Beștepe - Mahmudia alignment, with a minimum level of
at least 40 m, conclusion which cancels all hypotheses regarding the
existence of such an arm in this area.
Figure 4 - Variation of current terrain levels along a W - E profile
that passes through Poşta locality
Figure 4 presents the variation of the terrain alongside a west-
east profile that passes through the Poșta locality (see figure 3),
where it can be clearly seen that there in no area the current terrain
level is under 40 m.
Once with the gradual drop of the Black Sea’s sea level with
approximatively 10-15 meters up to around 0 m, a significant part of
the sea’s large gulf became land, the sea being forced to gradually
pull back to the east in depression areas on the floor of the former
sea, eventually leaving behind salt-water lakes in closed depressions.
The Danube’s course should also be subject to the same processes, its
mouth mostly tending towards the east, closer and closer, through
potential local erosion areas generated by the submarine course of
the river in the immediate vicinity of the shoreline. Simultaneously,
the marine areas near the seashore also became dry, with a rather
faded terrain due to the small inclination of the continental platform,
potentially presenting certain small local unevenness generated by
lithology and by the local dynamics of the water.
Under these conditions, during the period of maximum sea
level drop, the Razelm gulf became the Halmyris Lake and the
current Fundea gulf, less deep, became a marsh (see Figure 5, as
After the minimum period of the sea level, it, of course, begun
to increase again, so that the Halmyris Lake extended its limits on
land, simultaneously unifying with the marsh that became one of its
gulfs again.
Coming back to our island, the moment described in the
horothesia (the year AD 100) is found in the situations mentioned in
the two previous paragraphs: there was a natural channel that
started off from the Danube, passed through the Murighiol - Plopu -
Sarinasuf area and poured into the Fundea marshes.
With regards to the route of the Danube arm through the
Fundea gulf, this is emphasized within a magnetometric research
performed in this area as well, in which “the morphology of the
geomagnetic anomaly is discordant compared to the rest of the
surface occupied by the Razelm Lake”, in the sense that “the
orientation of the isolines is no longer on the direction known on the
rest of the lake, NW SE; instead, it gains a NNW-SSE tendency”
(Beșuțiu 1997, p.171). It is a matter of a complex geomagnetic
anomaly area, which, from the Sarinasuf locality, passes the entire
Fundea gulf on the north-south direction. It is constituted of a
minimum area in the center, flanked by two maximum areas on one
side and the other (on the west and the east). The anomaly
magnitude is of circa 15 nT. The author of the work advanced a more
complex geological interpretation, however, the real explanation of
this complex anomaly is the presence of the paleochannel of the
Danube’s arm (we may already call it the Peuce arm), which
generated a valley that is now filled with the gulf’s water, thus
constituting a (water) body with a practically null magnetic
susceptibility, and therefore smaller than any type of rock in which
this valley is cut, generating a central magnetometric minimum of
low intensity.
We also add a geological piece of information here:
immediately to the west of the Dunavățu de Sus - Dunavățu de Jos
localities, an area measuring approximatively 2 km x 2 km, covered
by “sands and brim” from the Upper Holocene period, (probably
residual deposits from the valley of the former Danube arm), was
mapped (geological map of Romania, 1:200,000 scale, the Sulina
The route of the Peuce arm from the Murighiol bifurcation
toward the south is marked by the current salt-water lakes Sărăturii,
Beibugeac and Sarinasuf (probably fossil lakes corresponding to
certain depression areas remaining from a time when the sea
expanded more to the west, up to Măcin).
The second punctual objective: the Peuce arm’s route from
Murighiol to the confluence with the Sfântu Gheorghe arm.
With regards to the Halmyris Lake (with a circumference of
approximatively 43 km, according to Plinius the Elder), it was much
narrower than the current width of the Razelm, which is circa 15 km
(the satellite images show that in the Sabangia - Sarichioi - Enisala
area the land drops slightly to the east, so the western shore of the
Halmyris lake was probably several kilometers to the east compared
to the current shoreline of the Razelm), thus it should be located on
the placement of the central part of the current Razelm Lake.
Probably, given the satellite images, the Popina Island, which has a
maximum level of +47 m, was linked to the shoreline through a neck
that tied it to the north, over the current Calica pond, with the Pietriș
hill, located immediately to the east of Valea Nucarilor. The Popina
Island (that was the “Popina Cape” at that time) is most likely the
northern limit of the lake, and the Iancina Cape area - the southern
limit. Immediately to the south there is a smaller lake, delimited to
the south by an isthmus that joined the Dolojman Cape with the
Bisericuța Island, as suggested by the allure of the gravimetric and
magnetometric anomalies in the area.
Figure 5 - Danube Delta in the area of the Razelm Lake -
Dunavăţ channel (Google Earth)
Figure 5 presents a Google Earth image of the Danube Delta in
the Razelm Lake - Dunavăț channel area, and figure 6 - the
paleogeographic aspect of an area probably at the beginning of the 1st
millennium AD. It can be observed that the former “Fundea” marsh
received water from the Danube (with great flow, given the width of
the paleochannel Murighiol - Sarinasuf - Fundea gulf and the current
flow of the Sfântu Gheorghe arm), it lost relatively little water
through evaporation (given its small surface area) and, as the current
tracks of the meanders show, it poured to the east through two
almost parallel mouths: a southern one, on the route of the current
Dunavăț channel and an intermediary one, several km north of the
first one. These two arms met at approximatively 3 km south-east of
the current Dunavățu de Jos locality and continued on together to
the east, probably on the present route of the Mustaca channel and
then further to the Danube, in the area of the current Belciugu Lake.
Figure 6 – Paleogeographic sketch of the Razelm – Sfântu
Gheorghe area at the beginning of the first millenium AD
Terra firma Open water surface Thatch covered water
S. = Sarinasuf D.S. = Dunavăţu de Sus D.J. = Dunavăţu de Jos
M. = Murighiol V.N. = Valea Nucarilor A. = Agighiol E. = Enisala
Sl. = Sălcioara VC = Vicus classicorumLM = Lake Murighiol
LSt = Lake Sărăturii LB = Lake Beibugeac LS = Lake Sarinasuf
Grindul Crasnicol, a narrowed bank of sand with a recent
history, resulted from the deposits generated by the sea currents, is
seen under figure 6 as well. Polibius (200 – 118 BC) mentioned the
presence of ridges (“Paps”) in the area in which the Danube flows
into the Black Sea:
“For the Danube flowing from Europe and falling into the
Pontus by several mouths, a bank formed of the matter discharged
from these mouths and reaching out to sea for a day's journey,
stretches for circa a hundred miles opposite them, and ships
navigating the Pontus, while still far out at sea, often at night when
sailing unwarily run aground on certain parts of this belt, which are
known to sailors as "The Paps." (Polibius, Historiae IV, 41, 1-2) and
“This is why in the case of large and swift rivers the deposits are
formed at a distance, the sea near the coast being deep, but in that of
small and sluggish streams the sand-banks are close to their
mouths.” (Polibius, Historiae IV, 41, 6).
Figure 7 - Google earth image of the eastern area of the
Fundea gulf
In the Google Earth image of the eastern area of the Fundea gulf,
at the joining with the Dunavăț Channel (figure 7), the numerous
meanders of the channel show that it is not an artificial, almost
rectilinear channel, created for easier fishing movement in the
respective area, but that it corresponds, evidently, to the course of
the Danube’s old arm!
The Halmyris Lake, which occupied only the central part of the
current Razelm Lake, was isolated from the sea and the current
Fundea gulf; it was supplied by current Taița and Telița valleys
(through today’s Agighiol Lake) and the Tulcea Valley.
Currently, the waters in the ponds and the channels south of
the Sfântu Gheorghe arm are practically almost stationary; however,
during the period in question, when one of the Danube’s arms (with
a rather high flow, probably half of the current flow of the Sfântu
Gheorghe arm) passed through the present Razelm Lake, the water
flow’s direction was certainly toward the east and north-east, as the
current course of the Danube was.
Histria’s boundary could, in consequence, have had the
following route (figure 8): on the Sfântu Gheorghe’s arm from its
mouth to the sea, upstream to the Belciugu Lake, then to the south-
west over the Razelm Lake (Lacus Halmyris), then to the west until
Slava Rusă and on the Slava Rusă Valley up to close to the water
heads, the Casimcea water heads, then south-west on the Casimcea
Valley, up to its mouth in the Tașaul Lake (we have thus admitted
the correspondence from horothesia Gabranus = Slava Rusă river,
Picusculus = Slava Chercheză river, as Paul Nicorescu stated before,
and Calabaeus = Casimcea River, as Vasile Pârvan had stated
previously), then to the north, alongside the ancient shore of the
Euxin Pontus up to the mouth of Sfântu Gheorghe. Sinoe Lake was
probably a part of the marine domain as well, which probably
included Grindul Lupilor. South of Grindul Lupilor the sea reached
up to several hundred meters from the current shore’s position.
Finally, in the solution proposed by us, the novelty compared to
Pârvan’s and Nicorescu’s solutions is represented by the
documented specification of the route of the northern part of the
Histria’s chora boundary (west of Sfântu Gheorghe - Belciugu Lake -
Fundea gulf - Slava Rusă), as a western limit of the territory, thus
adopting Vasile Pârvan’s version. Combining the two solutions has
led to a route that, from a geomorphological point of view, seems to
better correspond to the horothesia.
Figure 8 - The limits of Histria’s territory in the year AD 100
(after Pippidi, 1956, with amendments)
Other aspects also result from the corroboration of historic
information and the information resulted from the analysis of
geological, geomorphological and geophysical data from the area
located immediately to the south of the Sfântu Gheorghe arm:
1. from the geological research (Panin, 1983) - at the beginning
of the 1st millennium AD, Sfântu Gheorghe arm’s mouth to the sea
was approximatively 5-6 km further upstream compared to its
current position;
2. from the written (Herodotus) and numismatic (two Histrian
monetary issuances) documents, it results that the river Istros pours
into the sea on Histria’s territory, thus the “cadastral” territory of
Histria comprised Danube’s mouths into the sea, as well (or at least
the Sfântu Gheorghe arm), even if “the city was very far from the
river’s mouths”, as is mentioned by Governor Flavius Sabinus in his
letter to the “Histrian people and council”, included in the horothesia
3. the fact that the Danube’s southern mouth was
simultaneously called Hieron stoma and Peuce stoma, indicated that
the two arms surrounding the Peuce island were just as important as
to give any of their names to the common area up to the mouth to the
sea, and therefore, they likely had almost equal flows, each of them
carrying to the sea approximatively half of Sfântu Gheorghe arm’s
4. the pines which constituted a wood source for the Histrian’s
flares, likely grew on the Peuce island on the current hills of
Murighiol (maximum level of 65 m) and Dunavățu de Sus (51 m),
however maybe slightly uphill on the Beștepe-Mahmudia mound
(maximum level of 242 m), at approximatively 20 km NW from this
The conclusion of the aforementioned is that the Peuce Island
was not just a legend, but that it actually existed. In fact, it was an
isle with a surface area of circa 150 km2, surrounded by the Sfântu
Gheorghe arm to the north and the current Dunavăț Channel to the
south, which expanded to the east likely up to the area of the
Belciugu Lake.
Its characteristics, mentioned by the ancient authors
(approximatively triangular shape, with its base to the sea and its
apex inward, at approximatively 25 km upstream from Sfântu
Gheorghe arm’s mouth to the sea, near Halmyris lake, which had a
circumference of approximatively 43 km), are confirmed in their
entirety in the reconstituted paleogeographic situation from above,
presented under figure 6.
In time, the Peuce Island was sometimes confused with the
Leuce Island (both located near the Danube’s mouth in the Black Sea),
and other times, it was much more simply to state that Peuce Island
was just “a romantic invention”.
Recently (2015), a group of “authors” had “discovered” (from
the book “Imagini geofizice ale Histriei antice / Geophysical images
of ancient Histria”, ISBN 978-973-0-07257-0, author E.F. Scurtu, 2009,
which was very known to them, but which they failed to quote in
their paper), that the “Peuce island” was in the Dunavăț area!
Figure 9 - Danube Delta, the area of the Chilia arm, in the
year 1922 (after S.I. Cepleanu, 2012)
With regards to the Lake Thiagola, it could have been located
using the map under figure 9 on one of the northern arms (Thiagola
or Psilon) of this area, at approximatively 0035’ (circa 30 km) west
from its mouth to the Pontus (Ptolemy, Geographia, III.10.2). On this
map (edited in 1920), it may be seen that the northern arm (called
Chilia in the present day nomenclature) had at least three essential
moments of change in its configuration, reflected in successive
formation of areas with a delta geomorphology.
The most recent “delta” of the Chilia arm visible on the map
is constituted of nine primary obvious arms, which start from a
common node located at approximatively 5 km west from the Black
Sea’s current shore, in the Vâlcov locality area; the mouth to the sea
has a width of circa 15 km along the shore.
The previous position of the delta, at a point in time when the
Black Sea’s shoreline was at 10 km to the west compared to its
current position, indicates (between Chilia N and Vâlcov localities)
the presence of a delta with three primary arms and with other
numerous intermediary arms formed in between them.
Figure 10 - Map of the Galaţi - Tulcea - Sulina area
In the previous period, the seashore was at circa 30 km west
of its current position; the Danube Delta had at least four primary
arms that started from the area of the current Ismail city and ran for
circa 25 km to the east, pouring into the sea on a width of circa 25
km, in the Chilia N area.
Taking into consideration the increase of the Black Sea’s level
throughout the last 2000 years (see figure 2), it is probable that
another delta, the most recent one, is under the present Black Sea
level, in the Vâlcov area delta’s continuation to the east.
The comparison between Ptolemy’s map and this recent map
indicates that the delta’s geomorphology described by him
corresponds to the oldest stage of the four detailed above. In
consequence, the Thiagola Lake on Ptolemy's map is in the area of the
Danube Delta, on the Chilia arm. This lake was situated on the area
currently occupied by Limanul Catalpug (Catlabuga Lagoon) and by
the Chilia (Chitai) Lagoon, which at that time likely formed a single
lake with a total surface area of approximatively 200 km2 (see present
regional map of the area in figure 10). Today, both lakes are
characterized by the presence of a level of black brim containing
hydrogen sulfide. Their evolution is similar to that of other lakes in
the Danube Delta’s area (including Razelm Lake and the Beibugeac,
Sarinasuf and Sărăturii saltwater lakes).
The trade was able to develop in Roman Dacia and Moesia
Inferior, as in the rest of the Empire, due to the vast network of
communication ways built by the Roman State.
The immediate purpose for these paths was generally
military, having to ensure the fast and easy movement of troops,
including large pedestrian units that had to transport their entire
military equipment and personal goods.
We will make a small parenthesis as to present the structure
of the different military units in order to appreciate the technical
necessities of the roads that they had to use, sometimes in emergency
The army nucleus was represented by the legions, elite units
constituted of Roman citizens, each with a number of 5,600 soldiers.
The legion was divided into 10 cohorts.
Regular auxiliary units, cohorts and alae, were either
quingenariae (with a number of 500) or milliariae (with a number of
1,000 people); there were also cohortes equitatae, which had riders
among their numbers, as well: 120 in the case of quingenariae and 240
in the case of milliariae.
Along legions, cohorts and alae, which made up the regular
army, at the beginning of the 2nd century there were also non-regular
troops, with various numbers, which received, under Hadrian and
Antoninus Pius (AD 138-161), an organization closer to that of the
regular troops, with the denomination of numerus, having 500 - 900
Details regarding the organization of the Roman army in
Dobrudja can be found in one of Aricescu’s works (Aricescu, 1977).
There were three primary roads in Moesia Inferior. The first of
these roads came from the west through Oescus and Transmarisca,
passed through Durostorum, following the Danube downstream at
Cimbrianae, Sucidava and Altinum, then continued by Sacidava,
Axiopolis, Capidava, Carsium, Cius, Beroe, Troesmis, Arrubium,
Dinogetia, Noviodunum, Aegyssus, Salsovia, Halmyris,Vallis Domitiana,
Ad Salices. The second road continued along the shoreline through
Tomis, Callatis, Timogitia, Dionysopolis, passing into Moesia Secunda at
Odessos. The third one, starting from Marcianopolis (in Bulgaria)
passed the central area of Dobrudja from the south to the north,
passing through Tropaeum Traiani, Ulmetum and Libida, where it
divided to Noviodunum, Troesmis and Aegyssus.
The primary source of information regarding the routes of
these roads and the reciprocal distances between the intermediary
points (which constituted march stages of troops between
consecutive rest and supply stages) and the localization of fortified
settlements is constituted by the documents united under Itinerarium
Antonini (Cuntz, 1929), respectively Tabula Peutingeriana (Müller,
As one of the primary roads passing Moesia Inferior, on the
route Dinogetia - Noviodunum (= Isaccea) - Aegyssus (= Tulcea) -
Salsovia - Halmyris - Vallis Domitiana - Ad Salices Histria (Istria),
passed close by the Murighiol - Dunavăț area in which we located
the ancient Peuce “island”, we also approached the issue of
establishing the position of this road on the current map and in
relation to Peuce’s position.
For the segment between Noviodunum and Histria of this
strategic Roman road, on which we will focus our attention in what
follows, we will concentrate on the data from the two documents
mentioned above, included and noted in Fontes ad historiam
Dacoromaniae pertinentes vol.I, as well as in Istoria Românilor
(History of the Romanian people), vol. II (Editura Enciclopedică,
2001). We will also take certain observations and comments from A.
Aricescu’s works (Aricescu 1972, 1975).
In Tabula Peutingeriana, as well as in Itinerarium Antonini, the
distances are mentioned in roman numbers, in m.p. (thousands of
double paces, I m.p. corresponding to a distance of 1479 meters; we
note here that transforming the distances from m.p. to kilometers
must take into account the fact that the distances expressed in m.p.
must be considered as having an approximation of ±0.5 m.p., so ±740
The distances noted in Tabula Peutingeriana are the following:
Noviodunum-Salsovia - XLI m.p. = 60.6±0.7 km; Salsovia – Ad Stoma
XXIV m.p. = 35.5±0.7 km; Ad Stoma – Histria - LX m.p. = 88.7±0.7 km
(we mention here the relative distances at Ad Stoma, as we will make
a comment with regards to its probable localization, as well).
Itinerarium Antonini mentions the following distances:
Noviodunum – Aegyssus – XXIV m.p. = 35.5±0.7 km; Aegyssus
Salsovia XVII m.p. = 25.1±0.7 km; Salsovia – Salmorus (that is
Halmyris – Aricescu, 1975) IX m.p. = 13.3±0.7 km; Salmorus – Vallis
Domitiana XVII m.p. = 25.1±0.7 km; Vallis Domitiana Ad Salices
XXVI m.p. = 38.4±0.7 km; Ad SalicesHistriaXXV m.p. = 37.0±0.7
With regards to the current localization of the ancient
localities, in the year 1975, the only sure equivalences were
Noviodunum = Isaccea, Aegyssus = Tulcea, and Salsovia = Mahmudia,
and “researches up to now have not led to the certain identification
in the field of the following localities recorded in the two documents:
Salmorus = Halmyris, Vallis Domitiana, Ad Salices and Ad Stoma
(Aricescu 1975). In what follows, we will try to resolve this issue, as
The starting point is establishing, with maximum probability,
the position of the Halmyris settlement, which should simplify the
localization of Vallis Domitiana and Ad Salices taking into account the
reciprocal distances and the distance to Histria, whose position we
already know.
Figure 11 - Sketch of the ruins at Murighiol (after Polonic,
Given the fact that there already is an ancient settlement in
the area at the present time, which was archaeologically investigated
and was considered to be Halmyris (Topoleanu, 2000, Suceveanu et
al., 2003), we will make a note in the following as to analyze this
In the year 1898, the almost legendary engineer-topographer
Pamfil Polonic visited the ruins located between the Murighiol and
Dunavățu de Sus localities, situated approximatively 2 km south of
Danube’s Sfântu Gheorghe arm (Tocilescu, 1898). The sketch
developed with this occasion (presented under figure 11) shows that
it is a construction of an almost trapezoidal shape, with its sides
measuring 45 m (N), 177 m (E), 138 m (S) and 145 m (W), whose
walls, with a thickness of 2-3 m, are made of limestone blocks (the
total area of this fort was therefore of approximatively 16,000 m2 =
1.6 hectares).
In the year 1981, archaeologists initiated excavations in the
interior of the building’s walls and outside these walls, the
excavations being carried to this day. In the year 2003, a monograph
of the site was published (Suceveanu et al., 2003), comprising the
results of the archaeological research carried out up to that moment.
A difficult issue for the archaeologists was that of
establishing the ancient name of the settlement (Suceveanu,
Zahariade, 1987); the sole concrete information resulting from the
inscriptions discovered in the area was that the settlement was called
Vicus Classicorum (“the village of sailors”) in the 2nd century AD.
We shall not go into details regarding the authors and the
motivation behind the numerous contradictory proposals relative to
the ancient denomination of the Murighiol settlement, these being
described in much detail in the aforementioned works (Suceveanu,
Zahariade, 1987; Suceveanu et al, 2003); we shall only mention the
proposed options and several of the arguments of certain authors
with regards to these proposals. Beforehand, however, we will
specify that the equivalence Halmyris = Salmorus = Thalamonium was
unanimously accepted, therefore, the references to Halmyris imply
this equivalence.
As such, the settlement from Murighiol is successively
considered as representing the ancient Ad Stoma or Halmyris or
Gratiana or Ad Salices or Vallis Domitiana. With regards to Halmyris,
some authors (for example, Weiss 1911) say it was located at
Dunavățu de Jos, but this hypothesis is undoubtedly rejected by the
authors of the two papers, who believe that the argument of the need
to place Halmyris near the lake with the same name, is not necessarily
valid; furthermore, “how could a civitas (Acta Sanctorum) be a
sheltering of a “cuneus equitum (Notitia Dignitatum) - formation with a
number of at least 300 horse riders, becoming an episcopal
headquarter in the 6th century (Notitia Episcopatuum), in a burgus of
approximatively 2,000 m.p.?” (Suceveanu et al., 2003, p. 82).
We quote another reference to Halmyris from the same paper
(Suceveanu et al., 2003, p. 87) “if we can speak of control... in the
Dunavăț peninsula, then this is the control of Halmyris over the
settlements from Ad Stoma - Stoma Peuci (Dunavățu de Sus), Gratiana
(Dunavățu de Jos) and maybe Sarinasuf where, however, given the
extent of the ruins, the settlement seems very important, as well”.
After commenting upon all proposals existent in the
specialized literature, the authors of the two works decide to accept
the name of Halmyris for the settlement at Murighiol, the
determining argument being that it is at 13 km from Mahmudia =
Salsovia (Suceveanu et al., 2003, p. 24).
Our opinion is that Halmyris could not be located at
The first argument is the one used for rejecting Weiss’
hypothesis regarding its localization at Dunavățu de Jos (see above).
Another argument is that, as detailed in the first part of the
paper, the Murighiol, Dunavățu de Jos and Dunavățu de Sus
localities were, at that time, on the Peuce “island”, and the potential
military road that linked Salsovia (Mahmudia) to Murighiol would
have had to pass the Danube’s Peuce arm.
Taking into consideration the fact that it was a strategic road
that had to be constantly ready to support forced marches of
relatively vast military units, it could not pass the Danube neither on
a ship bridge (improvised means for short-term local actions), nor
through trans-shipment with river ships. Moreover, given that
Halmyris was just a stopover of a longer march, the Danube would
have to be passed using the same means twice: from Salsovia toward
Halmyris and then from Halmyris toward Vallis Domitiana.
A solution for this dilemma is that, in fact, Halmyris is located
near Sarinasuf, at approximatively 12 km west from Murighiol, and
“Halmyris” from Murighiol is, in Roman Dobrudja, just a vicus in the
Peuce isle: Vicus Classicorum.
The argument provided by the two works and cited above
(“given the extent of the ruins, the settlement” from Sarinasuf
“seems ... very important”) is supported still by the satellite image
provided by Google Earth (figure 12), where, immediately to the west
of Sarinasuf (respectively, immediately to the east from Colina
village) there is a rectangular zone with an area of approximatively
Figure 12 - Google earth image of the area SE from Colina
35 hectares (circa 700 m on the W-E direction, circa 500 m on the N-S
direction), in which a very regular rectangular network can be clearly
seen, corresponding to the streets of a former settlement, now
partially buried. A settlement of this size and organization also
corresponds to the exigences of a civitas (Acta Sanctorum),
sheltering a cuneus equitum (Notitia Dignitatum) - a formation of at
least 300 riding soldiers and that had become an episcopal
headquarters in the 6th century”, to which Weiss’ aforementioned
hypothesis does not apply. Furthermore, an archaeological research
performed in the Colina area between 1976 - 1980 by the
archaeologists at ICEM Tulcea led to unveiling Roman pottery in that
area dating from the 2nd - 3rd centuries AD and Roman-Byzantine
pottery (Ernest Oberländer, personal information).
The Halmyris settlement thus localized corresponds to several
other important characteristics mentioned in the historic document,
as well, which the Murighiol settlement does not possess:
- it is located in an area with a salty soil (halmyris = salting)
and near the saltwater lakes Sarinasuf, Beibugeac and Sărătura (=
salty soil) ;
- it is located at approximatively 7 km NE from the Halmyris
lake (which used to occupy the central area of the present day
Razelm lake) and at circa 10 km from the Danube (but only 2 km
away from the Sarinasuf lake, which, during that respective period
of time, was on the route of Danube’s Peuce arm - Scurtu, 2010) and
could thus have been considered, from a regional appreciation, both
“on the Danube” and “near Halmyris lake”.
Figure 13 presents a topographical sketch of the Mahmudia -
Colina - Murighiol area, as it is today, on which we also represented
the most likely route of the roman road between Mahmudia and
Colina village (respectively between Salsovia and Halmyris, in the
version proposed by us; hereinafter, when we shall refer to Halmyris,
we will understand the settlement at Colina).
This road starts from Salsovia to the SE, following the 25 m
level curve over a distance of approximatively 5.5 km, after which,
west of Colina (=hill) Duna, it changes its direction toward WSW as
to reach Halmyris, by descending gently. The total length of this road,
resulted from our map, is of 12,850 m, very close to that mentioned
in Itinerarium Antonini (13.3±0.7 km). Evidently, the 12,850 m
distance has a certain degree of relativity, as well, as we do not know
the exact position of the two milestone posts delimiting the
respective road segment on the ground, however, the potential
difference cannot exceed in total several hundred meters, more or
less (see the map under figure 13, as well).
Figure 13 - Topographic sketch of the Mahmudia - Colina -
Murighiol area and the probable placement of the Roman military
road on the Salsovia – Halmyris segment
A last ascertainment on the area’s map: at approximatively 3
km north of Halmyris, there are two Triassic limestone quarries.
Given the significant changes suffered by the terrain due to the
excavation of the materials in these quarries, it is to be expected that
they would have been operational in antiquity, providing the
necessary materials for the construction of the city of Halmyris, as
After those specified above, we shall concentrate on the
probable route of the Roman road between Halmyris and Histria,
comprising the Halmyris Vallis Domitiana, Vallis Domitiana Ad
Salices and Ad SalicesHistria stages.
The segment most easily identified seems to be the Histria -
Ad Salices one: the distance between these two reference points is,
according to Itinerarium Antonini, of XXV m.p., that is 37.0±0.7 km.
Given the fact that the Histria - Ad Salices (“La Sălcii / At the
Willows”) constituted the continuation of the road coming from the
south, from Tomis, it is to be expected that it would continue on the
same direction, along the Sinoe lake shore. At the present time, there
is a road heading in this direction, on the Nuntași - Istria - Sinoe
route, which then deviates to the west (as to avoid the ponds
between Golovița Lake and Ceamurlia Lake) to Mihai Viteazu - Baia
- Ceamurlia de Jos and comes back on the shore of the Sinoe Lake on
the Lunca - Vișina - Jurilovca - Sălcioara route.
Given that the sea level was 5 m lower during the period we
are discussing, the respective ponds probably did not exist yet, and
therefore, a direct road between Sinoe and Lunca was possible. The
same is valid for the area NW of Histria, so there could have been a
direct road between Histria and Sinoe, as well. Under these
conditions, a road alongside the Sinoe lake shore from Histria toward
the north, would pass through Sinoe (Vicus Quintionis), Lunca and
Vișina and would reach Jurilovca after approximatively 32 km. The
last 4-6 km could be made to the north (up to Sălcioara) or to the east,
up to the Razelm Lake, at Capul Dolojman (which is presumed to
correspond to the ancient Argamum, however without concrete proof
- see Avram, 1991, as well). With regards to the “La Sălcii / At the
Willows” toponym, it fits both localities, the willow being a well-
represented species in this area.
For the last two stages which have yet to be clarified
(Halmyris Vallis Domitiana XVII m.p. = 25.1±0.7 km and Vallis
DomitianaAd SalicesXXVI m.p. = 38.4±0.7 km), we have only one
indication: the fact that from Halmyris to the Vallis Domitiana, the
road went west (this is a fact already accepted by contemporary
authors, who consider that it is likely that Vallis Domitiana is located
at Agighiol, about 13 km west of Colina village). The total Halmyris -
Ad Salices distance must be 63.5±1.4 km, therefore a reasonable route
must be found, that respects this 62 - 65 km distance.
Under these conditions, the essential issue is that of finding
the probable position for Vallis Domitiana, which would implicitly
lead to finding a likely route for the Roman road between Halmyris
and Ad Salices.
Starting from Colina toward the west (following the 10 m
level curve), we reach the current day Agighiol locality after
approximatively 13 km. Going further west, the road should pass an
area with a rougher terrain, potentially passing through the Dealul
Regimentului and Furca Dealurilor points (there is an old, non-
modernized road on this route), reaching Valea Teliței and then
Valea Hagilar; from Agighiol to this point, immediately to the SE of
Lăstuni locality, the distance is of approximatively 13 km.
In this area, at circa 26 km west of Colina village, “near the
Deniztepe hill, at its foot, to the east, a Roman settlement and
numerous flattened funeral tumuli were found” upon an
archaeological cartography flight (Baumann, 1983, p. 86).
Taking into consideration the two pieces of information
presented above, we believe that we are entitled to presume that the
settlement emphasized by the aerial photographs taken over Hagilar
valley, SE of Lăstuni village, corresponds to the sought after Vallis
Domitiana. The Deniztepe hill, constituted of Jurassic sandstones, was
probably this settlements’ source of construction material.
The last stage, that which must establish the route of the path
between Vallis Domitiana and Ad Salices, becomes relatively simple:
this road should follow the route Lăstuni – Valea Teliței - Zebil
(approximatively 9 km), and then to head to Enisala, following the
northern bank of the Babadag Lake (8 km) and finally, to reach
Capul Dolojman (circa 22 km).
The final conclusion with regards to those presented above:
the Roman military road from Noviodunum to Callatis passed by in
the immediate vicinity of the current Colina (Halmyris?) – Lăstuni
(Vallis Domitiana?) Enisala Jurilovca (Dolojman cape, = Ad
Salices?) Histria localities. Certainly, although these localizations
present, in our opinion, a probability close to 100%, they shall
become a certainty only after material proof in this respect shall
validate them.
A short comment with regards to Ad Stoma’s placement
mentioned in the Tabula Peutingeriana, located, according to this
document, at a distance of XXIV m.p.= 35.5±0.7 km from Salsovia,
following, however, a trade route and not a military one.
As suggested by the name of the locality, Ad Stoma must be
sought for on the Danube, toward the river’s mouth to the sea, so
downstream of Murighiol. As the distance between Mahmudia and
Murighiol is of approximatively 8 km, it therefore results that Ad
Stoma would have been at circa 27 km SE of the latter. The distance
must be measured on land, not on water, as suggested by Aricescu
(Aricescu 1975, p. 318), given its expression in m.p.
Keeping in mind the last information regarding the Peuce
Island (Scurtu 2009, 2010, 2016 and the present work), Ad Stoma (=
Stoma Peuci?) could have been located at the eastern extremity of the
island (see figure 6), currently under the waters of the Danube Delta,
in the area of present day Belciugu Lake. The location was propitious
for such settlements, where larger, long travel ships that were
coming from the sea could meet with those that ensured the river
transportation (there evidently was a river connection between
Salsovia and Ad Stoma, probably with a stopover at Vicus
Classicorum): the Danube’s flow increased immediately to the east of
the Peuce isle, due to the intertwining of the two arms that went
around it.
The respective road, with a length of approximatively 36 km,
started from Salsovia and followed the right bank of the Sfântu
Gheorghe river arm downstream of Murghiol (probably right on the
current Murighiol - Dunavățu de Jos road).
Another road, most likely, went to the south from the
strategic military road Mahmudia - Sarinasuf immediately to the
west of Duna hill, passing through the Peuce arm through a shallow
portion between the current Plopu locality and the Sărăturii Lake
(the level curves of the present day terrain show that the Danube bed
was less deep in that area). After passing the Peuce arm, the road
would head south up to Cetatea Zaporojenilor (The “Zaporozhians’
Fortress”) and perhaps further up to the southern shore of the Peuce
island (the current Dunavăț channel). As it can be seen from the map,
this road was in fact the continuation toward the south of the road
that came from Salsovia and intersected, in the immediate proximity
of Duna hill, the road that came from the west, from Halmyris, and
that probably continued toward SE on the Sfântu Gheorghe arm up
to Ad Stoma.
Corroborating the existence of this intersection between the
two strategic roads with the fact that Duna Hill is with circa 10 m
higher than the other local tops of the summit south of Murighiol, we
can presume that there was a military construction on Duna Hill,
which must have held a control and supervision unit that watched
over the traffic on this road. The Gratiana citadel, mentioned by
Procopius (Book 4, Part 3, p.315) could have also been here.
A sketch of the probable placement of the Roman military
road between Salsovia and Histria, of Ad Stoma’s position and the
probable route of other roads in the area is presented under figure
An issue that remains unresolved to this day is that of the late
Roman name of the ancient settlement from Murighiol. Can we
accept the hypothesis that this name might have been Gratiana, as
certain authors dealing with Dobrudja’s history presume? In order to
answer this question, we will begin with the static aspect of this issue
- the geographic and geomorphological data of the area and the
description of the physical results of the archaeological excavations
carried out at Murighiol, and shall move on noting certain historic
situations that could be relevant to our issue (the dynamic aspect of
the issue) and try to use this data “cloud” to approximate the area’s
historic evolution.
Figure 14 - A sketch of the probable placement of the Roman
military road between Salsovia and Histria and of the ancient Ad
Stoma settlement’s position
Ae Aegyssus AS Ad Salices ASt Ad Stoma CZ Cetatea Zaporojenilor
H Halmyris Hs Histria Ge Genucla LB Lacul Belciugu
Og Orgame Sv Salsovia VC Vicus Classicorum VD Vallis Domitiana
Peuce Island
The limes road
Roman road
As we have shown before, the geomorphological data,
corroborated with the geophysical and geological data, have led to
the certain conclusion that “the Peuce Island“ was an isle with a
surface area of circa 150 km2 between the Sfântu Gheorghe arm and
a paleo-arm of the Danube (called “the Peuce arm“ in antiquity)
which followed the route set out today by the Murighiol locality, the
Murighiol – Sărăturii Beibugeac Sarinasuf lakes, the Fundea gulf
of Razelm lake, the Dunavăț channel and the Belciugu Lake in the
Danube Delta, where it met with Sfântu Gheorghe arm (see figure 6).
Peuce Island’s position is very correctly presented graphically (see
figure 1) in Ptolomy’s famous Geographia (Tabula Europae - tabula IX),
drafted around AD 150 (this map confirms Strabo’s statement as well
- Geogr. VII, 3, 163, 17 - with regards to this island being occupied by
the Bastarnae, a population of Germanic origin, which, as such, took
on the denomination of Peucini). However, the fact that the area’s
appearance changed as a result of the Black Sea’s level increasing,
confused the contemporary researchers who tried to find its
placement on the current topographic maps.
The presence of a population of Bastarnae origin, however
otherwise named - “Peucini”, in that area, was doubted by many
researchers studying Dobrudja’s history, believing that it is just a
name confusion (Babeș, 1977). The closest area in which Bastarnae
origin traces had been discovered was Poienești (Bacău County), in
the year 1936, as a result of excavations carried out by Radu Vulpe,
but only in the year 1955, after the investigation of 55 “graves with
black urns“, did he reach the point of defining the Bastarnae
intrusive culture. The archaeological excavations continued to be
carried out for many years by Mircea Babeș, who subsequently
published a synthesis of the results he had obtained (Babeș, 1985),
from which we present certain information necessary as to unfold
our demonstration.
The excavations from Poieneşti, correlated with results of
investigations performed over a large surface occupying the area
between the Dniester and the Eastern Carpathians, north of a line
that follows the route Piatra Neamţ - Bacău - Crasna - Leova (on the
Prut) - Tighina (on the Dniester) almost exactly, have led to the
conclusion that an allogenous culture was spread throughout this
area, receiving the name of the Poieneşti - Lukaşevka culture,
attributed to a foreign population - identified in the Romanian
specialized literature as the Bastarnae.
The material traces of this culture were researched, traces
dating back from the end of the 3rd century BC until the end of the 1st
century BC, that were represented mostly by a specific type of
pottery used both with household purposes and as incineration urns.
In the field of the Poienești-Lukașevka culture, there are four
different pottery groups: local Geto-Dacic pottery, Poienești-
Lukașevka Germanic pottery, and in much smaller quantities, the
Celtic and Hellenistic pottery. They often appear associated in the
same settlements, sometimes in the same complexes. The Poienești-
Lukașevka pottery is completely handmade, the most obvious being
the fine pottery, most often present in necropoli, but there is also
common use, gross pottery, found mainly in settlements.
The fine vessels are black, with a metallic lustre. In most
cases, they have thickened and faceted spouts, and their handles are
shaped into an X; rarely, there are embossed ornaments applied on
them or ornaments set through incision. As opposed to this type, the
common use pottery is characterized by the large granulation
degreasing agent (ground fire clay), as well as by the less careful
modeling and finish of the vessels. Figures 15 and 16 present several
specific types of Poienești-Lukașevka pottery (after Babeș, 1985).
Aside from pottery, there were also other types of objects that
where established, with typical aspect that is specific to the Poienești-
Lukașevka culture: spindle whorls, short handle spoons, rattles and
especially clay andirons (“fire dogs”), adornment objects: necklaces,
bracelets, fibulae, buckles, as well as tools: knives, scythes, axes,
chisels, rakes etc. Very important: all of the over 130 Poienești-
Lukașevka settlements researched up to the year 1985 are, without
exception, lacking fortification and express an evident preference for
low leveled lands, found near water.
Figure 15 - Several specific types of Poieneşti-Lukaşevka fine
pottery (after Babeş, 1985)
Figure 16 - Several specific types of Poieneşti-Lukaşevka
gross pottery (after Babeş, 1985)
The scene changed in the 1990’s, after the publication (Irimia
and Conovici, 1989, Conovici, 1992) of the results from the
archaeological research performed in Satu Nou – Oltina village,
Constanța County.
The Getic fortified settlement from Satu Nou - “Valea lui
Voicu (Voicu’s Valley)” had seen, according to the authors cited
above, two main stages of inhabitance, established based on the
analysis of amphora stamps and other Hellenistic imports: the first
stage was dated in the largest part of the 3rd century BC (circa 280
190 BC the latest), and the second stage was in the whole 1st century
BC (after the inhabitants had changed their location in the 2nd
century BC 800 m upstream, where they built a new, fortified
During the excavations, several handmade fragments of
pottery and a complete vessel were recovered, which “stand out
completely from the great mass of aboriginal and Hellenistic pottery,
however, they can be attributed with certitude to the Poienești-
Lukașevka culture” (Conovici, 1992, figure 4):
- brown dish, polished inside and out, with thickened, faceted
and flared spout on the interior and a small X shaped handle under
the spout;
- dish, tapered body, jutting shoulder, flared and faceted
spout on the inside, straight base, polished walls,
inside and out;
- fragments from dishes, pots, bowls, made of a paste
containing ground shards and fine sand, of a brown-gray or black
color, polished, with an usually faceted spout, rarely arched, found
together with the Getic and Hellenistic pottery dating back to the end
of the 3rd century BC.
Figure 17 presents pottery fragments and a Poienești-
Lukașevka dish from Satu Nou (after Conovici, 1992).
The Poienești-Lukașevka pottery from Oltina, the first of its
type in Dobrudja, “indicates the existence of this already well
established culture even from the last quarter of the 3rd century BC,
that is even before the Bastarnae had settled into Moldova,
simultaneously constituting the first archaeological attestation of the
existence of Bastarnae in Dobrudja in the 1st century BC” (Conovici,
Figure 17 - Pottery fragments and a Poieneşti-Lukaşevka dish
from Satu Nou (after Conovici, 1992)
Given that the the Piatra Neamț - Tighina barrier does not
exist anymore, today we can look for traces of the Bastarnae kitchen
in all of Dobrudja, including (and especially) on the Peuce island and,
thus, at Murighiol as well. The search must be extended
chronologically not only up to the first century BC, but to much later
as well, given the fact that the Peuce island, the birth place of the
Visigoth king Alaric (in the year AD 370), was still a famous
geographic entity in the Roman Empire even in the 5th century AD,
as shown by the Latin poet Claudius Claudianus in Panegyricus de
sexto Consulatu Honorii Augusti and the historic encyclopedias. The
late presence of Bastarnae (Peucini) in this area is, in consequence,
very likely and we shall try to offer additional arguments in this
respect throughout this paper.
The simplest action, which may precede a more detailed
research, could be the reassessment of the “Geto-Dacic” LaTène
pottery discovered throughout time in Dobrudja (it exists in
warehouses or was signaled during the field surveys or in detailed
archaeological researches), as to classify it, keeping in mind the fact
that attributing this pottery to the Bastarnae had probably been
considered slightly eccentric at the time of the respective initial
research. Gavrilă Simion’s paper “Les Gètes de la Dobroudja
septentrionale du VIe au Ier siècle av.n.è.” (Simion, 1976) could
constitute an important support element.
Let us now return to Murighiol.
The largest study carried out upon the settlement, before the
start of the systematic researches, belongs to Al. S. Ștefan (Ștefan,
1984, pp. 297-310), “even if the realities of the field did not confirm
all of his interpretations” (Topoleanu, 2000, p. 34, Suceveanu et al.,
2003, p. 22).
We will note the results of the archaeological excavations
carried out in the aforementioned area in enough detail for the
subsequent realization of the analysis (we will use the acronyms
“T2000” instead of “Topoleanu, 2000” and “S2003” instead of
“Suceveanu et al., 2003”, as we will be citing these papers often).
The actual excavations begun by creating two master
trenches, SI (220 m in length, oriented E-W) and SII (285 m in length,
oriented N-S), then stripping in surfaces of 4 m x 4 m squares. Figure
18 presents the detailed topographical map of the ruins’ area and the
placement of the two master trenches. The map emphasizes the
presence of at least two gates (placed on the western and northern
sides of the dwelling) and at least 11 towers. The total surface area
comprised between the triangular shaped building’s (120 m x 175 m)
walls is of approximatively 10,500 m2 that is 1.05 ha.
Figure 18 - Detailed topographic map of the area of the ruins
at Murighiol and the placement of the two master trenches (after
Zahariade and al., 1987)
The results of these excavations were made public through
the works of Zahariade (1986), Suceveanu and Zahariade (1986),
Zahariade et al. (1987), Opaiț (1991), Topoleanu (2000), Suceveanu et
al. (2003) and Zahariade, Topoleanu et al. (1987-2015, in Cronica
Cercetărilor Arheologice din România/The Chronicle of
Archaeological Research in Romania). To our knowledge, there is no
synthesis of these results, nor is there a graphic presentation of the
planimetry of the buildings discovered as a result of the excavations
carried out until now at Murighiol.
The master trenches led to establishing a very complex
stratification, within which (S2003, T2000) “the Getic settlement”
(“Pre-Roman period”), “early Roman castra” (“Early Roman
period”) and the “late Roman castra” (“the Roman-Byzantine
period”) were individualized.
We continue by briefly presenting the primary results of the
excavations, results extracted from the two works above mentioned.
The pre-Roman period (up to the 1st century AD).
The oldest artifacts discovered (in secondary position - S2003)
in the researched area, are represented especially by the fragments of
kitchen vessel and Hellenistic amphora, and a fragment of an
indigenous vessel, probably dating back to the 6th-4th centuries BC,
corresponding to the “Getic settlement” - level A (basic), which “it
seems ... was concentrated to the north of the future late Roman
citadel” (S2003, p. 29). Among these, there is also mention of a bowl
fragment, dating from the 3rd century BC, “made from gray paste
with a black core, which is fine, compact, and presents exterior
polishing” (Opaiț, 1991). Other pottery fragments, originating from
Rhodian amphora and from handmade vessels, corresponding to
level B (superior), date back to the 2nd and 1st centuries BC. Level B,
largely extended outside the future Roman fortification, could imply
the existence of a dava in that respective period, whose fortification
elements would have been overlapped with those of the Roman fort
(T2000, S2003).
The Early Roman period (1st-3rd centuries AD).
Level N1, from the 1st century AD - dated with late
Hellenistic, early Roman and Getic materials, “presents an expansion
in time, but also a continuity of the old settlement, having, however,
a predominantly civil character” (T2000, p. 36).
At the beginning of the 2nd century AD, a Roman fort is built
on top of the Getic settlement (N2), and during the reign of Caracalla
(212-217), a stone fort is operational at Murighiol, as certified by a
coin from this period found in the walls, however, without being
able to specify the date of its construction (N3). The following level,
N4, seems to indicate a destruction, as results from the investigation
of the western area beyond the city walls (extramuranus), where the
defense moat’s filling is observed.
The following step (level N5, the last level of the “early
Roman citadel”) is the reconstruction of the settlement (“violently
destroyed in the year AD 295, event which was experienced by other
Dobrudja forts too”), including the retracing of the cardo and of the
decumanus maximus and the reconstruction of the praetorium.
The Roman-Byzantine period (4st – 7th centuries AD).
The first level of inhabitance (N6, from between AD 308-348)
of this last period of existence of the Murighiol citadel, represents an
organic continuation of the previous. After the aforementioned
destruction, the reconstruction is attested in the foundation’s
inscription, on which the name of the emperor Galerius Valerius
Maximus can be found, located near the western gate. In its new
form, the fort had a triangular plan, following the shape of the rocky
promontory on which it was placed, with its long sides slightly
arched, the confined wall’s thickness of 3.40 m, with 16 towers and
two gates instead of four, probably in the early age - one at the
northern top (offering immediate access to the Danube) and one
almost in the middle of the western side. In the middle of the citadel,
at the intersection of the two main arteries oriented east-west and
north-south, there was an important building (“the central edifice”),
in the proximity of which a thermal complex was identified. This is
also the period during which the defense system beyond city walls
(extramuranus) was built, having two dikes and moats to the west
and three dikes and moats to the south.
The citadel witnessed significant bloom (level N7) in the 4th
century AD (approx. between AD 351-402).
The Vth century AD, to which the levels N8 and N9
correspond, ends with a long period of relinquishment of the citadel,
or, if not, of long abandonment in any case, after the violent
destruction of some of the primary urban or strategic monuments.
The next level (N10), dated by coins from Justinus I (AD 518-
527) and Justinianus (527-565) proves a spectacular restoration of the
settlement. The archaeological researches have shown not only the
reconstruction of old monuments, but also the construction of others,
of an impressive quality (the northern thermae, the north-eastern
gate, the fountain in the central area). “This last period of prosperity
is temporarily interrupted by the Kutrigurs raids in AD 559, event
archaeologically emphasized by the traces of violent destruction on
this level”.
The following level (N11) shows that the settlement is quickly
rebuilt, none of the previous monuments being abandoned. This
level is inhabited until sometime toward the year 584/585, after
which a new destruction indicates the beginning of the citadel’s
urban decline, as well.
“At the end of the 6th century and the beginning of the 7th
century AD (levels N12 and N13), the citadel maintains all of its
strategic, urban and economic attributes: the premises, the
extramuranus defense system, the edifices and the street network, a
lamps workshop that sent its products at least to the surrounding
areas, and a large quantity of imported pottery types, especially from
the oriental provinces. All of these elements, to which we add several
monetary discoveries from the beginning of the 7th century AD,
exceed the dark year of AD 602, considered until not long ago the
date of the final fall of the Danube limes. The citadel continues to be
inhabited even after the fall of the city’s defense system. The
masonry elements from the previous edifices are used and semi-
buried houses and surface hovels are created at a time which we can
presume was toward the middle of the 7th century AD, after the fall
of the limes.
In a special chapter regarding the monuments discovered as a
result of the archaeological excavations, there are details described
regarding the probable evolution of the northern gate and of “edifice
no. 1” = “central edifice” (located on the NW side of the fortified
area), which functioned, as resulted from the numismatic data, both
during the Early Roman age, probably as of level N2, as well as
during the Late Roman age. “In the current state of the research, we
can only presume the existence of a building, within this researched
central perimeter, that had functioned for a long period of time and
had been of a significant importance ... used as a home and not as a
public edifice” (S2003, p. 77). The coping of a 5 m deep well was
discovered in the immediate proximity of this edifice, on the N10
During the archaeological research performed on the area,
almost 500 coins were discovered (the oldest dating back from the 2nd
century BC, the newest from the 5th century AD). Over 2,000 whole
vessels and pottery fragments that were found at Murighiol (898
amphora, 537 food serving dishes, 416 kitchen vessel pieces, 70
pieces intended for serving fluids - jugs, mugs, cups and goblets, 182
lamps and 3 toilet bowls) were analyzed and published (T2000; here
we specify that in T2000, only the Roman and the Roman-Byzantine
pottery was taken into consideration, that is centuries 1st to the 7th
AD; several older pieces of pottery, as of 6th century BC, are
presented in the S2003 paper and in that of Opaiț, 1991).
For the present study, special interest is expressed, among
others, towards the opinion regarding the “provincial west-Pontus”
lamps discovered at Murighiol: “the great diversity of shapes and the
relatively small number of the pieces is due to ... a diversified
production, with short series, as to satisfy the needs of a small
community” (T2000, p. 194). Furthermore, “the presence in Halmyris
of certain types which are rare in Dobrudja or, to the contrary, the
rarity of certain lamps which were very common in other centers”
(T2000, p. 194) and “Although any of the provincial west-Pontus
lamps previously described, and not just them, could have been
modeled at Halmyris, this can be said with certainty only about two
types” (in total, 53 samples). Their common element is represented
by the same production defect, present on the pieces of the two
types, defects which are hard to accept by a producer looking for an
outlet. Finally, most of the first type lamps found next to the traces of
a kiln, do not present signs of use, proof of their in-situ production.”
(T2000, p. 208). Of the lamps presented in the work, we mention the
inv. 505 “west-Pontus” lamp, who has the letter Π (from Πεύχη –
Peuke?) written on its base, dating from the second half of the 3rd
century AD, and that with inv. no. 508 (from 4th century AD), whose
base has the potter’s mark, PE (from “Peuce”?).
Figure 19 - Two lamps with inscriptions from the
archaeological excavations at Murighiol (after Topoleanu, 2000)
With regards to the inscriptions discovered at Murighiol (in
total 21 pieces), we offer new details about some of them which can
be implicated in our demonstration.
One of these has the votive inscription with the catalog no. 2,
with the translation To Hercules, the vexillatio of the Legio I Italica and
of the Legio XI Claudia Pia Fidelis”, dating from the beginning of the
2nd century AD (Zahariade, 1986). The existence of a vexillatio
constituted of army men from two different legions, stationed in two
different locations and the association of Hercules with a vexillation,
lead to the hypothesis that the activity of this mixed vexillation was
not a routine one, but a special one which required significant effort
(Zahariade, 1986). Here is an appreciation (Vulpe, Barnea, 1968, p.
157) of the situation, from the respective period of time, at the Lower
Danube: “The number of troops left on the Lower Danube’s limes
had decreased significantly. In the place of the legions that had left,
now stood only mixed squadrons. An inscription from that time
found in Băneasa (former Ghiuvegea), west of Adamclisi and
pledged to the god Neptunus Augustus, mentions such a squadron,
entitled vexillatio Legionum I Italicae Moesicae et V Macedonicae Dacicae
Tropaei agens. The epithet Dacica clearly shows that at that time, after
the year 167, Legio V Macedonica was no longer part of the troops”.
The vexillatio that made the dedication to the god Neptunus Augustus
was on the Danube’s bank.
We also mention the votive inscription with catalog no. 4, 5,
6, 7 and 8, dedicated to Jupiter by the “Roman citizens residents of
the vicus classicorum”, as well as the funeral inscription no. 14
(Zahariade, 1990, with the translation: “To the Mani gods. To
Valerius Ponticus and Valeria Nene, well-deserving parents. Valerius
Valens, son, officer in the Legio I Italica, set for them”) and the
inscription no. 21 (dating from the 3rd and the 4th centuries AD) on a
fragmented tile, which includes the text “Valerius Valerinus from the
Legio I Iovia urges Valeria ...”. With regards to the five inscriptions
mentioning Vicus Classicorum, inhabited by “Roman citizens
residents” (set between AD 136 and 200), it was observed
(Suceveanu, Zahariade, 1986) that no inscription mentions the
emperor’s name, not even the exact day of the dedication, as it is
usually done in the case of such inscriptions, the implicit
determination of the date resulting only from the names of councils,
mentioned by some of the inscriptions; therefore, we are dealing
with a particular situation. With regards to Vicus Classicorum, this
could have originated from many possible situations, however, “the
most plausible would be that this denomination corresponds to a
vicus (canabarum) stationis classis, which indicated a civil settlement
that appeared alongside a military unit (usually called vici, besides
the auxiliary, and canabae, besides those of the legion), in our case a
fleet unit (classis)” (S2003, p. 98).
We also note the fact that the brackets of inscriptions 12 and
19 appear as a grape vine decoration, and those with no. 5 and 17
appear as a grape vine and pinecones decoration. The vine appears
also as an adornment on a funeral stele found on the wall of Cetatea
Zaporojenilor near Dunavățu de Jos locality.
The archaeological researches in the perimeter of the
Murighiol fort continued after the drafting of the monograph from
2003 (S2003), by means of a 4 m x 4 m squares grid with a control of 1
m between them, their results being published annually in the
Cronica Cercetărilor Arheologice din România / Chronicle for
Archaeological Research in Romania (CCAR). They mention, as is
natural for the specific of this publication, very many details
necessary for a subsequent synthesis of the results and as to explain
certain situations that are visible today in that respective area or that
have been deduced from the historic literature. We will continue to
mention only certain passages from those papers (including those
from 1999 and 2000), which we will use for our analysis in the
present case study.
As such, in the description of the archaeological results from
Murighiol from 1999 (CCAR 2001, pp. 154-155), (in square N27) a
surface dwelling (2,25 m x 1 m) is mentioned, with stone walls
cemented with earth, which had functioned on the 13th level of the
site, in the second decade of the 6th century AD. Stone walls
cemented with earth, from the 6th century AD, appeared in the N28,
O27, O28, P27 and P28 squares, as well. Common kitchen dishes,
fragments of luxury pottery, fragments of dishes with the Christian
symbol of the fish, several fragments of lamps and a coin from
Mauricius (582-602) were found in the last four squares mentioned
Other walls associated to those from the year 1999 were
discovered in the year 2000 (CCAR 2001, pp. 155-157), in the P28,
Q28-31, R27-31, S27-31 squares, that emphasized the northern side of
the basilica which was to be completely unveiled in the following
years, as well as the southern side of a large building (“edifice A”).
Within the archaeological researches from the 2001 campaign
(CCAR 2002, pp. 211-213) carried out in the perimeter of edifice A,
measuring 42 m x 17.7 m, (according to the surveys performed in
previous years), in the L15 square, at a depth of 1.55 m, “a massive
piece was uncovered, made from basalt that seemed to be a grinder
or an oil press” (it was later ascertained that it stood on a pedestal
built from the remains of an amphora). The earliest level ascertained
in edifice A pertains to the 4th century AD. Investigations continued
in the Northern Gate sector on a large building called “Domus
Nord” (discovered in the year 1994), probably built in the 6th century
AD, which also had a private bath (“that is why the edifice to which
the bath was attached could only pertain to an important civil or
military individual”, according to Madgearu and Dvorski, the
authors of the respective text).
From the results of the researches carried out between 2002-
2003 (CCAR 2004, pp. 207 - 214), we note the fact that, in the first
stage of construction, the towers at the west gate (as well as those
from the northern gate), were constructed using a “cyclopean”
technique, possibly between AD 101 and 107, having as base yellow
limestone blocks measuring 1-2 m in length and 0.3-0.5 m thick,
cemented with a fine layer of mortar. We also note the fact that a
semi-circular construction (covering a 3000 circumference of a circle)
was uncovered behind the western tower, with an interior diameter
of 1.60 m and an exterior diameter of 2.70 m (therefore, the wall’s
thickness is of 55 cm) sealed with clay at 80 cm from the last
remaining layer, with burn traces on the interior walls, and
fragments of pottery, numerous bovine, sheep and pig bones inside
it and near it; it is presumed that this was a supply storage space
(probably, we might add, having the role of smokehouse, as well).
The excavations within the basilica revealed, among other
things, traces of a pre-Roman settlement at approximatively 1.5 m
deep, represented by hovels, from which fragmentary pottery pieces
were collected dating back to the 4th-3rd centuries BC, together with
Greek pottery fragments and roof tiles.
The excavations carried out between 2004 and 2010 are not of
use for our objective.
Within the excavations carried out in 2011 (CCAR 2012, pp.
87-88), the primary objective was that of investigating “the access
road to the Danube”. With this purpose, 6 squares were opened (S1 -
S6) measuring 3 m x 2 m each, on the SW-NE axis, outside of the
postern and of the line of the two walls that create it. The results
have been extremely interesting, yet complicated to interpret by
authors due to certain erroneous premises regarding the place’s
topography and history; we will analyze them later in this study.
The results from the years 2012 and 2013 are not indicative
for our study, and there were no reported excavations carried out in
Besides the excavations performed between 1954 and 1958 as
to investigate the fort, Murighiol already benefited of excavations for
the archaeological research of certain areas where the presence of the
incineration necropolis was signaled (Bujor, 1957, 1959, 1961, Simion,
1995). The results of these researches were not included in the
volume “Halmyris, monografie arheologică / Halmyris, an
archaeological monograph” (S2003), which referred solely to the fort
Figure 20 presents the placement of the areas partially
investigated in the year 1955 by E. Bujor.
In the case of the area noted with 1 on the map, located north
of the locality (on the “Moroianu Ridge“), the first research had
already been carried out in the year 1949 by V. Zirra, with the
occasion of certain construction works performed in the area. In the
report submitted to the National Museum of Antiquities, it is
mentioned (after Bujor, 1957) that “pottery fragments were seen
among adobe and clay, most of them amphora, of which, several had
a stamp. Aside from these, there were also fragments of local pottery
- gray paste dish - similar to those from Zimnicea. The same works
brought to light an incineration urn as well, with handles underneath
its belly, fragmented, made of gray paste, with burnt bones inside.
Among the small objects, there were: a large bronze arrow with three
edges, a glass pearl and an indigenous lychnos of primitive
craftsmanship”. The two survey trenches of 10 m each, performed in
1955 (Bujor, 1957) on the same area, revealed few pottery fragments
from the superficial part of the trenches (up to the depth of 60 cm):
pot spouts and dishes made from a paste with small gravel or
ground shards. The dig stopped at approximatively 90 cm, where
researches stumbled upon a yellow-white soil (probably a Tertiary
clay) that was extremely hard, in which it was not possible to
Figure 20 - Placement of the areas researched at Murighiol in
the year 1955 (after Bujor, 1957)
Concurrently, in the year 1955, archaeological researches
were performed in the sector placed on the northern side of the
Murighiol hill (three trenches measuring 50 m x 2 m each, with a
maximum depth of 90 cm), noted with 2 on the map (figure 18).
Numerous fragments of Hellenistic amphora were found, as well as
six incineration graves. The urns are represented by vases of a local
fashion, made from impure, crumbling paste, with thick walls of a
brown color, with alveolated belts or from Hellenistic amphora
(“making possible the idea of a social differentiation within the Geto-
Dacian population in the 3rd century BC”).
During the research continued in that sector in 1956, 18 new
incineration graves were discovered. The numerous urns found here
were constituted largely of pots made of brown paste with
impurities, insufficiently kneaded, incompletely burnt, friable, with
the spout slightly heightened and tilted to the outside. The urns’
covers were usually made from a gray dish, with clean paste, well
cured, in certain cases with a twisted handle, other times with
handles and lateral spurs or without handles. As E. Bujor underlines
(Bujor, 1957), “At Murighiol, an aspect of the Geto-Dacic culture was
identified characteristic of the end of the 4th century and for the 3rd
century BC”; subsequently, the “Murighiol culture” is recognized in
other areas of Dobrudja, as well: Satu Nou, Canlia, Enisala, Telița,
Bugeac (Preda, 2000).
Other 13 incineration graves were discovered in the other
three areas surveyed in the year 1956 (in front of the elementary
school, in the Christian cemetery and on the other side of the road
from it), graves with the same characteristics as those described
above, but also fragments of stamped amphora and a Thracian fibula
(Bujor, 1959a). In the year 1957, another 9 graves were discovered in
the incineration cemetery uncovered in front of the elementary
school, having the same characteristics mentioned above.
The author (Bujor, 1959b) underlines the fact that none of the
graves in the two cemeteries from Murighiol held urns that were
Hellenistic amphora, and the offering vessels were local. In the year
1958, another two graves were discovered in the incineration
cemetery from the village center (Bujor, 1961).
Gavrilă Simion (Simion, 1995) also carried out researches
during the year 1983 on the south side of the Murighiol locality,
previously investigated by E. Bujor (no. 2 under figure 20). His
works concentrated on two of the funeral tumuli out of the 6 (M1-
M6) located on the Murighiol hill ridge (figure 21) with an average
diameter of 23 m (M1), respectively 17 m (M2).
The morphology of the two researched tumuli includes a
burnt layer with ash and coal (a result of the fire lit to purify the
respective area), with a thickness of about 2 cm and incineration
graves (including the urn with burnt bones, alongside offerings and
very numerous fragments of “sacrificed” amphora - probably broken
in a ritual at the place of incineration and transported in tumulus
with the deceased’s remains).
Figure 21 - Placement of the area researched at Murighiol in
the year 1983 (after Simion, 1995)
The M1 grave contained, besides the urn-vessel (a mug with a
spheroidal body, made from a fine, sandy, fragile and permeable
clay, cured at high temperatures), a bowl, the remains of a bronze
mirror twisted as a result of burning, several smaller sized objects -
including over 30 glass beads and numerous amphora fragments -
including 40 handles or fragments of handles of stamped amphora,
made from a fine, pale-pink paste with a yellow engobe.
In the M2 grave, traces of two graves were discovered, one of
them completely destroyed, containing fragments of a handmade
vessel made from a gross paste, decorated with alveolated bands,
and the second - pottery fragments originating from Rhodian
amphorae, of which 8 were stamped, with the same characteristics as
those from M1. According to the date of the amphorae mentioned in
the catalog presented in the work, “these funeral monuments could
have been constructed approximatively toward the end of the first
quarter or toward the middle of the second century a. Ch.”.
The presence of such a large number of elements coming
from the same Aegean production center “attracts our attention to
the economic role that Rhodes played at the end of the 3rd century
and the beginning of the second century a. Ch., ... which implies the
existence of an important center that would facilitate the promotion
of an active transit trade” (Simion, 1995, p. 279). “Then again, in the
stratigraphy of the citadel” (from Murighiol) “there were sufficient
amphora fragments and Rhodian stamps which could support the
idea that the settlement here had played such a role” (of transit trade
center) “before the arrival of the Romans”. (Simion, 1995, p. 280).
In order to better understand the funeral ritual, we must
realize the fact that the urn represented the “heart” of the grave, that
it contained the earthly remains and the entire personality of the
deceased. It practically always was a local, popular production
vessel, regardless of the deceased’s rank, made from the earth from
which he/she came and into which he/she returned. The other
objects accompanying the urn, constituted objects that would create
the “spiritual” comfort, as well as offerings for the gods with which
may be the deceased shall meet in the future life.
We will continue by reproducing a very important paragraph
from the mentioned paper (Simion, 1995, pp. 283-284): “All studies
that have approached, in one way or another, the issue of spirituality
in the case of the Geto-Dacians have commented upon the
phenomenon of the disappearance or lack of necropoli of the locals
during the period from the end of the 3rd century BC up to the arrival
of the Romans, issuing several hypotheses in this respect. In the case
of all comments written with regards to the phenomenon which
occurred over such a large space and over an entire people, how
come the Greek-Roman antiquity did not note it in any of the age’s
documents? And if that’s how things are, then which is the factor
that put to rest a spiritual tradition known up to the half of the 3rd
century a. Ch. and well documented through numerous discoveries?
To all hypotheses that attribute the phenomenon to certain
measures or new spiritual orientations dictated by the Buresbista -
Deceneus organizations, we can answer with a simple question that
we ask ourselves with regards to the recounts sent by an eye witness
about the Getae groups that invaded the Tomis market on trade
days, with their ever-present knife at their belt and drinking iced
wine from jugs. How is it that the measure taken by the great
organizers of the centralized state as to eliminate the vines from all
over the country had as a result the frequency of such scenes that
horrified the delicate poet of Tristia and Epistulae ex Ponto? How
can an issue so important to the life of an entire people, as that of the
cult of death, be so generally and strongly made to disappear?
Certainly, only the lack of information, written or archaeological,
regarding the Geto-Dacian funeral rite and ritual is at the base of
such a diversity of ideas and hypotheses. All of these issues allow us
to justify them due to not having discovered their archaeological
source yet and, as such, the need to look for it”.
In the course of the present work, we will try to reach closer
to the correct answer to these questions. However, in an anticipatory
manner, we can affirm that the “phenomenon” was the Bastarnae’
invasion, mentioned by historic documents.
Geophysical research was also performed in the area of the
Murighiol citadel in 2004 (Scurtu, 2005). The geophysical
(magnetometric) research was carried out on two distinct areas: on
the exterior of the citadel, immediately south-west of this, on the
presumed placement of the late Roman settlement, as well as in the
interior of the citadel, near the south-eastern corner.
The mapping of the geomagnetic field from the southern
sector (south of the Murghiol-Dunavățu de Sus road, on the northern
side of the Dunavăț hill) started through very detailed measurements
(50 cm equal distance between stations) performed along certain
profiles oriented almost north-south, with an equal distance of 20 m
between them and with a length of 160 m each (with several
exceptions), their northern limit reaching up to approximatively 10
m from the road. The reason for which we first performed the
reconnaissance profiles was that, without any other archaeological
excavations in this area, there was no information with regards to the
details of the presumed settlement, its localization being performed
solely through the presence of certain pottery fragments at the
surface of the soil.
Figure 22 - Regional magnetometric map of an area located
SW of the archaeologically researched area at Murighiol
The magnetometric map performed based on the profile
measurement results (figure 22) does not present many details on the
west-east direction, given the large distance (20 m) between the
neighbouring profiles and therefore, it is impossible to realize an
efficient interpolation. A certain zoning is however observed, as well
as the presence of several abnormal tendencies in the north-west area
(towards the foot of the hill, several tens of meters from the road).
Furthermore, in the south-west corner of the map, we can see an area
with values of the geomagnetic field that are greater than in the rest
of the investigated area.
Figure 23 - Detailed magnetometric map of an area located
SW of the archaeologically researched area at Murighiol
The next step in the northern area research was that of
performing even more detailed measurements (2 m distance between
profiles, 50 cm distance between the stations) in the perimeter
selected from the map of results on the reconnaissance profiles.
The characteristic of the magnetometric map thus obtained
(figure 23) is constituted by the fact that, in general, there are no
elongated anomalies, the image being dominated by several bipolar
anomalies (with a maximum to the south and a minimum to the
north), disparate, in the B3, B4, C4, D3, E1, E2 squares, indicating
disturbing bodies with an almost circular horizontal section, with an
average diameter of 2 m, located at a shallow depth. These do not
seem to be parts of an organized building, regardless of the fact that
they do have a tendency toward alignment. In our interpretation,
given the shape and size of these anomalies, it is very likely that they
might correspond to certain incineration graves grouped in that
respective area, including a local agglomeration of pottery fragments
(sacrificed urns and vases) and a burnt area for soil purification
through fire.
The strong anomaly from square D2 represents the parasite
effect generated by a small iron object, buried close to the surface.
The very small amplitude anomalies scattered over the map’s surface
do not have a material correspondent, they constitute a parasite
effect of the interpolation of data under the conditions of a network
with different equidistances on the two directions.
By comparing the two maps discussed above, we observe
that the placements of the abnormal areas detected on the general
map (figure 22) are very precisely found on the detailed map (figure
23), and the far denser measurement network allows for a more
explicit and detailed view of the local anomalies.
No excavations as to verify the geophysical maps presented
above were performed.
Interpreting the magnetometric anomalies described above
(mapped on the northern side of the Murighiol hill, on the opposite
side of the area where excavations are performed for the
archaeological documentation of the ancient fort) as an effect of
certain incineration graves is justified firstly due to the discovery of
the two necropoli dating from the 4th to the 3rd centuries BC in its
proximity (Bujor, 1957, 1959a, 1959b, 1961, 1971), but also due to the
information according to which “there are true archaeological
treasures on the hill platforms near the ponds, and on the hills in the
ancient city’s vicinity” (“Halmyris from Murighiol), “whose
documentation sends us back in time to the Indo-European
migration” (Simion, 1995, p. 265).
The existence of certain rather extensive necropoli (probably
at least over the entire northern side of the Murighiol hill, but also on
the flat area near the Murighiol locality) in the 4th and 3rd centuries
BC implies the existence in this area of a large LaTène settlement,
during the same period of time. The variety of manners in which the
deceased were buried (from sumptuous tumuli, with various and
rich, imported, offerings, up to the graves placed in common
cemeteries, with more modest offerings or even without any
offerings) indicates a society in which, besides the common people,
there were persons in economic and/or political positions, that were
very important for the community.
The archaeological excavations (level A, basic, 4th to 3rd
centuries BC) confirm the existence of such a settlement here, which
“it would seem focused more on the northern side of the future late
Roman citadel” (S2003, p. 29), so approximatively along the Danube.
Moreover, at the archaeological level B (attributed to the second and
first century BC and located immediately above level A), ”largely
extended outside the future Roman fortification, could imply the
existence of a dava in that respective period, whose fortification
elements would have been overlapped by those of the Roman fort”
(S2003, p. 29).
We mention that this archaeological information was
obtained only based on two master-trenches created in the first stage
of the excavations at Murighiol, therefore they are very delicate with
regards to the horizontal expansion of the observed archaeological
Let us stop here briefly and try to reconstitute the area’s
historic evolution.
We will not take into consideration the beginning period of
the human civilization in the area, as during that respective period,
the potential differences between Dobrudja’s different areas were not
that great and in any case, these potential initial differences could not
significantly influence the subsequent historical evolution of the
A reference point in the area’s historical evolution was the
period during which the Greek colonists started to create stable
settlements on the Black Sea’s western shore, in the 7th century BC,
the most famous being Histria, Tomis and Callatis.
According to archaeological research, it would seem that, for
example, in Histria, populated around 630 BC by the colonists from
Milet, the inhabitants had collaboration relationships with the
indigenous, pre-colonial, Hallstattian population, pertaining to the
Babadag culture, for half a century (or had no relevant contact with
it), as the first fortification of the city dates back to the years 580-570
BC, so more than 50 years after the settlement of the Greek colonists
(Maria Coja, 1986). Building a precinct wall (the settlement’s
“archaic” precinct wall, expanding throughout the entire western
and southern limit) with an overall length of 1,000 m and a width of
several meters, indicated the subsequent appearance of certain
relatively severe antagonizing situations between the newcomers
and the local populations.
The founding of the city of Tomis, also a Milesian colony, took
place later, probably in the first half of the 6th century BC, and
Callatis, the sole Dorian colony, was founded by the colonists from
Heracleea Pontica in the last quarter of the 6th century BC.
These three Greek colonies are well known and mentioned in
the documents of the age, constituting both trade nodes between
Greek cities and indigenous population, as well as foci for the
ancient Greek civilization in the Geto-Dacian world. They were, of
course, the largest and most known; however, aside from them, there
were also commercial branches and “secondary colonies”,
subsequently founded, as a result of the arrival of new waves of
immigrants after the creation of the basic colony (Istoria Românilor /
The History of the Romanians, vol. I, p. 543). One example, among
others, is the fortified settlement at Albești, Constanța County, which
was included in the Callatis’ territory (Buzoianu, Bărbulescu, 2008)
and constituted both a trade area for the locals and the Greeks
(emporion) and a strategic monitoring and defense point (phrourion);
however, its ancient name is not known.
Our geophysical research carried out at Albești (Scurtu, 2014),
has shown that, in the immediate vicinity of the fortified area, there
are large quantities of sedimentary, syngenetic, disseminated iron,
inside the chemical precipitation limestones of Khersonian age. This
iron mineralization, easy to exploit (it lies immediately under the
soil) and to prepare (it is constituted by iron oxides and hydroxides,
which formed local accumulation of red and yellow ochre - figure 24)
could have constituted an important reason for creating this ancient
Figure 24 - Red ochre and yellow ochre sedimentary deposits
at Albeşti
Figure 25 presents the southern part of the geomagnetic map
of the geophysically researched area at Albești. The chaotic iron
mineralization distribution (the red colored areas) from the northern
part of the map (A-C squares lines) can be observed, and in the
southern part of the map, on column 4/(F-J) squares lines, a row of
intense geomagnetic anomalies can be seen; each of them could
represent either a local iron enrichment center, or, most likely, the
traces of former iron ore processing kilns, or local pottery production
With regards to the settlement from Murighiol, it was indeed
located, sometime prior to the year AD 100, within the territory of
the Istros citadel, as is deduced from Laberius Maximus’ horothesia
which, in the year 100 AD, confirmed the fact that “the ancestors’
borders” reached up to the Danube’s Sfântu Gheorghe arm (Pippidi,
Figure 25 - Geomagnetic map of the archaeological
excavation area at Albeşti
As the incineration graves, flat or tumular, from Murighiol
contained numerous fragments of Hellenistic amphorae, especially
Rhodian and Thassian ones, it is a logical deduction that the
settlement at Murighiol was supplied with Hellenistic pottery and
other Greek goods through an emporion in the 2nd century BC.
Logically, this would have recorded maximum efficiency if it were
exactly in the respective place, having a most numerous clientele and
benefiting from a port on the Danube. The large quantity of traded
Hellenistic products entitles us to presume a great number of
indigenous inhabitants from the Greek cities in Murighiol, who had
already arrived “in a wave of epoikoi, who would have debarked,
upon request from the Istrians, at the bridge between the 7th and the
6th centuries BC, given that this is the moment in time when the
massive penetration into the rural territory begins”. (Istoria
Românilor / The History of the Romanians, Vol. I, p. 543).
On the other hand, we also have the previously presented
archaeological information at Murighiol, regarding the existence of a
rather numerous population in the 4th – 2nd BC centuries (we mention
the fact that excavations were carried out only with regards to the
necropoli of the civil settlement and we have no information upon
the actual settlement) and other information, resulting from the
archaeological excavations in the fortification area, regarding the
probable existence of a dava, with a surface of approximatively 1,5
hectares, in the 2nd and 1st centuries BC.
This information should be completed with new information
(Conovici, 1992) regarding the presence of Bastarnae in Dobrudja
and with other, more recent, information regarding the Peuce island
history (Scurtu, 2009, 2010, 2013).
As shown before, the archaeological researches carried out at
Satu Nou – Oltina village, Constanța County (Irimia and Conovici,
1989, Conovici, 1992) have indicated the presence of the Bastarnae in
this area from the last quarter of the 3rd century BC and up to the 1st
century BC. The Poienești-Lukașevka pottery, representative for the
Bastarnae population, discovered in great quantities at Satu Nou,
constitutes a proof for the aforementioned.
On the Peuce island (whose geographical position on the
current map is between the Sfântu Gheorghe arm to the north-east,
the alignment of lakes Murighiol-Sărăturii-Beibugeac-Sarinasuf-
Fundea gulf of the Razelm lake to the north-west and the Dunavăț
channel to the south, including the current Murighiol, Dunavățu de
Sus and Dunavățu de Jos localities - see Scurtu 2009, 2010, 2013), the
presence of the Bastarnae and their auto-entitlement as “Peucini”
was signaled from the first decades of the 1st century AD by Strabo
(as we mentioned before), and the discoveries at Satu Nou confirm
the veracity of this information. We consider that it is also confirmed
by the presence of the Bastarnae-type pottery (see its description in
the reference to the Satu Nou settlement) in the Murighiol graves
and even in the level A (basic) discovered in the excavations carried
out around the fort, at least as “a fragment of a bowl dating back to
the 3rd century BC made from gray paste with a black core, which is
fine, compact, and presents exterior polishing” (Opaiț, 1991)
mentioned before in the text concerning the description of the
The work (T2000), referring solely to the Roman and Roman-
Byzantine pottery (1st to 4th centuries AD) cannot represent a
potential evidence about the state of things prior to this age.
However, given the fact that the Bastarnae domination in the area
would have had to last for a longer period of time, we reproduce
below several passages from this paper as well, which could be
reinterpreted in the sense above, in light of the excavations from Satu
- “The common elements of all pottery types pertaining to the
kitchenware are the uneven rough paste, with iron oxide, small
rocks, limestone granules and mica in their composition, as to
enhance resistance to thermal shock; the shape of the spouts that
would allow for the vessels to be covered, as is the case of bases,
rounded or concave, often with thinner walls than in the upper part,
as to enhance thermal conductibility. A characteristic of the
kitchenware is the long life of the shapes, as certain types may be
observed without any essential changes, over hundreds of years.”
(T2000, p. 100);
- “A number of 354 pots, in one piece or fragmented, have
been modeled in the west-Pontus provincial workshops. Aside from
the paste’s aspect, that of the execution and analogies, the great
quantity of the collected material (Table XII, Graph 7) also plead in
favor of this attribution, which, given the specific of this pottery
category, could not be produced anywhere else than the local
workshops. In this respect, we mention that out of the 15 types of
identified pots, only six have less than 10 pieces.
A great deal of these dishes were probably shaped in
Halmyris, the kiln for lamps constituting a precedent in this respect.
With regards to pot production in the citadel, an argument is the
piece pertaining to type XIV (catalog no. 293), found complete,
without engobe or traces of use” (T2000, p. 106);
- “The (west -) Pontus provincial amphorae ... represent 29.1%
of the total number of the amphorae determined in the Halmyris
settlement (232 specimens). These represent the absolute majority in
the early Roman period (1st to 3rd centuries AD). On the Roman-
Byzantine levels, their share is in constant decrease: 37% in the fourth
century AD, almost 36% in the fifth century circa 26% in the sixth
century AD, 16% in the first few decades of the seventh century AD.
With regards to the certificated types, only two of these - Kuzmanov
1985, types 15 and 16, represent together 58.6% of the number of
Pontic amphorae (Table XXII, Graphs 10, 11).” (T2000, p.149);
- “This time, the vast term of West-Pontus provincial pottery
may, we believe, be easily replaced with that of local pottery, as the
analogies and the specific of the kitchenware limit the area of the
production centers in space, sometimes even up to the workshops
which must have produced it in Halmyris.
The rarity of the oriental imports for this pottery category
results from the diminished number of pots, only 28 pieces
representing 6.73% of the total kitchenware and 7.32% of the total
pots. We also observe that of the 15 distinguished types, 13 are
represented solely by one vessel each. “
From the aforementioned, we may conclude that, probably
even in the 3rd century BC, in the southern area of the current
Murighiol locality, there was a settlement with a Greek-Bastarnae
population, spread across the right bank of the Danube’s Sfântu
Gheorghe arm (in accordance with the already certified preferences
of the known Bastarnae settlements). This settlement would have
been previously developed (probably in the 6th century BC) by Greek
colonists, over which a mass of Bastarnae migrants would have
overlapped, in a more or less peaceful manner, probably in the 3rd-
2nd centuries BC (as in the Satu Nou area).
With regards to the those mentioned above, the Murighiol
amphorae” (Meljukova, 1975, Mateevici, 2008) could once again be
brought into discussion. “It is possible that the Murighiol amphorae
(with a 5.5 liter capacity) was destined for transporting and
maintaining wine made in a center that was certainly sufficiently
known and popular in the Aegean region. It is not unlikely, as it can
be seen in the number of archaeological complexes in which this type
of amphorae has been attested, that the wine brought in this
packaging was of a good quality and made its way clearly (even if
for a very short period of time, estimated at a half of a century) to the
North-West Pontus barbarian markets”. “In certain sites, the
fragments of these amphora were attributed to certain unknown
centers or wrongly attributed to Thassos or Heracleea” (Mateevici,
The advantages of the Murighiol area, placed on the Peuce
Island, came from its position and the economic and military
opportunities offered by this position.
The Peuce Island terrain was moderate (the current average
level is of 15-20 m), with the exception of the Dunavăț Hill (with a
maximum current level of 64 m) and of the hill (current maximum
level of 75 m) on which stands Cetatea Zaporojenilor. We also have
to mention the presence of the two fresh water lakes in the vicinity of
the Sfântu Gheorghe arm, dry today, Crulic Mic (recent surface 40
hectares) and Crulic Mare (84 hectares) that had a connection to the
Danube through small channels, registered on the topographic
surveys from the beginning of the twentieth century (S2003, p. 13). In
antiquity, the terrain had the same aspect as the today’s one, only
that the absolute nominal levels were with 5-7 meters greater than
they are at the present time (v. figure 3).
The palynological analysis of the excavated material from the
archaeological excavations indicated the presence of wheat, barley,
oats, rye, flax, hemp, vine crops in the ancient vegetation, as well as
fruit-bearing trees and certain oak, pine, poplar species; the
examination of the osteological material has led to the conclusion
that there were sheep herds and a rich wild pig population (S2003,
pp. 17-18). Of course, the fish in the Danube constituted a basic
element in the local population’s nutrition, however it was also an
important item for export.
Figure 26 - Funeral stele from Cetatea Zaporojenilor (after
Suceveanu et al., 2003)
Of the Murighiol settlement’s economic resources, it seems
that the most important in the second century AD were vine (which
appears as adornment in the inscriptions noted with 5, 12, 17 and 19
discovered in the crumble of the northern gate) and pine wood
(inscriptions no. 5 and 17 were represented in the brackets, alongside
vine, pine cones, these two thus appearing as emblematic elements of
the settlement.). A stela made of limestone ornate with vine was
discovered also on the walls of Cetatea Zaporojenilor, near the
Dunavățu de Jos locality (figure 26), and a fragment of another
funeral monument decorated with vine wreaths was discovered at
Murighiol in the area of the western gate during the excavations
from 2001 (CCAR 2002, p. 212).
On the subject, we once again mention the discovery, in the
area of the “edifice from the NW area”, on a level corresponding to
the 7th century AD, of a massive piece made from basalt, that seemed
to be a grinder or an oil press” (CCAR 2002, p. 212), subsequently
(CCAR 2004, p. 214) ascertained to have been “standing on a
pedestal made from the remains of an amphora”. As the use of local
oleaginous plants (sunflower?, rye?) for oil extraction at Murighiol
has yet to be ascertained, we presume that it might be a cereal
grinder for the settlement’s local consumption.
As in the case of the neighboring Histria, where cattle, honey,
wax, salted fish, pine wood for lamps were exported, it is expected
that the same articles (and perhaps wine) might have constituted the
export base through the Murighiol port, but there is a very important
difference: the presence of pine and oak forests on the Peuce Island!
It should be implied that this characteristic of the area did not
go unnoticed by the colonists: the pine wood was the base material
for making ships, and oak - the base material for masts, therefore, the
situation constituted a direct invitation to action in this respect!
Let’s try to build a scenario as logical as possible, taking into
account the historic realities of the respective age - revealed by the
documents and results of the archaeological researches.
In the first stage of the development of the future economic
and military center in Murighiol the naval construction activity
started with the production of small ships, probably plateypegia, flat
belly vessels, therefore with low sea gauge, adequate for fishing in
the Danube Delta and for local trade navigation. Subsequently, with
local tradesmen and others, prisoners or those attracted by the
substantial material advantages, work begun on producing larger
ships, maybe even military ships.
The ship building activity continued (for a very long period
of time, as we will try to prove in this study) under changing
historical conditions.
It is probable that a difficult situation appeared when the
Bastarnae extended their inhabitance area up to Dobrudja (certainly
up to Satu Nou in the 3rd century BC, as well as, during a still
unspecified period of time, on the Peuce Island - in accordance with
Strabo’s statement and with characteristic artifacts discovered in the
excavations from Murighiol); however the ship production could
have continued, producing even a part of the thousands of ships
subsequently used by the Bastarnae in military operations.
We do not have details about the actual events that took place
during this period of time, however, in the archaeological
excavations from Murighiol, the 2nd and 1st centuries BC are
represented by “level B, largely extended outside the future Roman
fortification, (which) could imply the existence of a dava in that
respective period, whose fortification elements would have been
overlapped by those of the Roman fort” (T2000, S2003). In the 1st
century AD, dated with late Hellenistic materials, early Roman and
Getic materials, the settlement still had a “predominantly civilian”
character, after which, at the beginning of the 2nd century AD, the
first Roman fort was built on the settlement from stone, certified with
a coin issued between 212-217; subsequently, it seems that “a
destruction took place, probably in the year AD 295, an event
experienced also by other citadels in Dobrudja”.
Slightly detailing the data presented in the previous
paragraph, we ascertain that, in the 1st century AD, the inhabitants of
the settlement had already built the stone fort near the Danube’s
Sfântu Gheorghe arm, at the base of the Murighiol hill. This fort was
not “Geto-Dacian”, it was not on the ridge of Murighiol hill, from
where the surrounding area could have been monitored, instead it
was in a lowland, along the Danube, where, according to the rules
presented before, had to exist a Bastarnae settlement.
The settlement could not have been a vicus (canabarum)
stationis classis (as presumed in S2003, p. 98), as it (later called vicus
classicorum”) existed much earlier than a potential Roman military
unit near which it should have appeared.
From the statistics regarding “minor fortifications” (“dava
from Murighiol falls under this category), it results that these can
also represent “fortified rural settlements”, including even some
trade and production centers and various fortified residences
(Dinčev, 2007, Băjenaru, 2010).
But why was such a construction, costly from all points of
view, necessary?
It probably had the role of protecting the (initially civilian)
settlement later called Vicus Classicorum, which must have been a
very important strategic, economic and political and, implicitly,
military objective for the community (including the “Peucini”) that
populated the area.
The “Bastarnae and Peucini” are mentioned in many
documents, which is not a pleonasm, because, as can be seen, the
Peucini were a special category of Bastarnae, both from the point of
view of their territory, as well as, probably, that of other social,
economic and military aspects; the presence of the Bastarnae on the
Danube is mentioned by Strabo (64 BC - AD 24), in Geografia, VII,3,2:
“These peoples - “Hippomolgi” and “Galactophagi” and “Abii” - as
the Bastarnae, are even today mixed with the Thracians, especially
with those beyond the Ister, but also with those from the other side,
which are mixed with the Celtic peoples”, and later, in the 5th century
AD, by Avenius Rufus Festus in Descriptio orbis terrarum 435-440:
“The Danube pours into the sea through five mouths, where the
Peuce island rises. From here on out, where the breath of the frozen
north wind whistles, the Sarmatians, Germans, Getae, ferocious
Bastarnae and the Dacian peoples reign”.
One of the reasons for building this fort could be that here
was the place were river transport ships were built, which could
have been equipped for military conquests or pirate actions in the
numerous invasions upon neighboring or distant provinces.
Maybe one of the most important reasons, enhanced by the
“industrial” and potential commercial activity, originates from the
development of certain events that unfolded in the year 29 BC,
described by Cassius Dio and recounted and commented (in 1926) by
Vasile Pârvan (Getica, p.78 and pp. 88-90, of which we cite several
passages below):
- “we ask ourselves if in this very year, 61 a. Ch., Burebista
himself had been the commander of the Getae-Bastarnae troops that
crushed Antonius and in taking his flags ... took them to the Genucla
Getic citadel from where, thirty years later, Crassus would come (a.
29) to reconquer them (Cassius Dio LI 26.5) from the Getae King of
the Lower Danube land, Zyraxes, who, this time as well, turned for
help to the “Scythians” in Bessarabia, that is to the Sarmatian-
Bastarnae in the steppe” (p. 78);
- “While he punished and organized Thracia, the Getae Roles,
the Romans’ ally, send word for help to Crassus, being attacked by
another Getae king, Dapyx” (p. 88);
- “The circumstances recounted now (26, 1-6) not only
confirm the localizations offered above for the campaign of a.29, they
also offer us precious details about the Getae in Dobrudja. The
kingdom of Dapyx appears to be neighboring and, more precisely, to
the North of that of Roles, as we see Crassus after passing Haemus
defeating Dapyx first in an open field, then besieging him in a
fortress where he had taken refuge, more inside the country, of
course, then conquering Genucla, which was exactly toward the
mouths of the Danube. Thus, we’d have Roles’ kingdom between
Durostorum-Abritus-Axiopolis, Dapyx’s kingdom to Capidava-
Carsium, with the interior toward the Ulmetum-Histria (this is in fact
the second greatest Dacian center of Dobrudja: territorium
Capidavense), and finally, Zyraxes’ kingdom in the well-known,
from Ovidius too, Getic quadrilateral: Troesmis-Dinogetia-Aegyssus-
Buteridava.” (p. 89);
- “Once arrived here, Crassus believes it’s his duty to attack
the other Getae neighbors, although they did not pertain to Dapyx,
nor did they do anything to him, and “he starts against Γένουκλα
citadel, the strongest in Zyraxes’ kingdom, as he had heard that there
lay the flags taken by the Bastarnae from Gaius Antonius, when they
had defeated him near Histria” (26, 5).
As such, the contiguity of Roles’, Dapyx’ and Zyraxes
kingdoms is out of the question, and the settlement right near the
Danube’s mouths belonging to the state of the latter is supported
even more by the following.
Crassus conquers Genucla surrounding it on land and on
water - as it was up on the Danube’s bank not without difficulty,
but relatively quickly. Zyraxes, however, had fled beyond, taking the
thesaurus with him, even before Crassus had caught up to him: he
headed to the “Scythians“ as to ask for help: yet, these “Scythians“
were nothing else than the Getae-Sarmatian-Bastarnae from Moldova
and Southern Bessarabia (26, 5-6)” (pp. 89-90).
From those described above, we can draw a very important
conclusion, even a spectacular one: Genucla (“the strongest one in
Zyraxes’ kingdom” and its capital) was in fact the settlement at the
current Murighiol, the sole settlement that meets all conditions
described by Cassius Dio, supported by the archaeological
discoveries made in the area. It was “the strongest” (the kingdom
probably including other auxiliary, non-archaeologically researched
fortifications as well); however in Zyraxes’ kingdom (Peuce island?)
there was at least one more citadel, whose vestiges are called Cetatea
Zaporojenilor today, and there could have also been another citadel,
now submerged, near the Dunavăț channel (see figure 14). The fact
that the Murighiol citadel was the residence of the Peucini king
explains the security measures and the special comfort ensured for
the inhabitant of the “central edifice”.
With regards to the citadel’s name, “Genucla”, the linguistic
analysis of its name leads us to the conclusion that it is a proper noun
originating from the Vulgar Latin common nounsnucla” and “ge”.
As it is mentioned in Appendix Probi (Baehrens, 1922), some
diminutives from Classical Latin were different from those used in
current speech - Vulgar Latin. An example of the 227 mentioned in
Appendix Probi (“auris non oricla”):
auris: diminutive → auricula; Vulgar Latin → oricla.
By analogy,
nava: diminutive → navicula; Vulgar Latin → naucla.
Nucla” is probably a corruptela of naucla, originating from the
initial pronunciation by the local Bastarnae population (of Germanic
origin) of the sequence “au as a diphthong, as it happens in the
contemporary German Language.
With regards to the other component of the name Genucla,
ge”, as results from the paper “An Inventory of archaic and classical
poleis (M. H. Hansen, T. H. Nielsen, 2004, pg. 39 - Synonyms for
Polis), “In the sense of territory, polis is used synonymously with ge”,
as such, “ge may be translated through polis. We also add the fact
that “the Vulgar Latin already had Greek loanwords before the
Roman Empire” (Transactions and Proceedings of the American
Philological Association. American Philological Association. 1870.
pp. 8–9).
From the aforementioned, it results that Genucla was a polis in
29 BC or it had kept the name of a former polis that had existed in
past in the area of the citadel and that the population was mainly
represented by sailors.
This conclusion, correlated with the information coming from
Hecateus (550 - 470 BC) according to which “Orgame was a polis near
the Danube” („Orgame polis epi to Istro”) becomes very important,
given the fact that there are doubts with regards to the correctness
associated to attributing the name Orgame to the settlement from
Capul Dolojman, cf. Avram, 1991, Damyanov, 2005; Alexandrescu,
1999 stated that “Orgame” (= Argamum from Capul Dolojman)never
attained the polis status”. Therefore, it is very likely that the ancient
settlement at Murighiol, presently wrongly named “Halmyris” (as we
have demonstrated before - pages 26-28) to have been, in fact, the
ancient Orgame, which subsequently became Genucla. We also add
the observation that if Orgame was already a polis at the end of the
6th century BC, it had been founded at about the same time as the
other great Greek colonies at the Euxin Pontus (Histria, Tomis,
Callatis), that is, 7th to 6th centuries BC.
These conclusions are also supported by the results of the
archaeological excavations carried out here throughout the years,
which confirm the existence of an ancient settlement, represented
mostly by fragments of kitchenware and Hellenistic amphora from
the pre-Roman period (6th to 1st centuries BC up to 1st century AD ;
see page 43 under the present work).
An essential argument is “Horothesia ... ”, which refers to “the
Histrian people’s borders, as follows: ... Peuce, Halmyris lake,
beginning with the land of ... the people of Argamum....”. It results,
doubtfully (see figure 8), that Argamum = Orgame was a very
important Greek city (a polis), whose territory included the island of
In anticipation, we shall note that the auxiliary Roman naval
units in Scythia (milites navclarii at Flaviana) and respectively Moesia
secunda (milites tertii navclarii at Appiaria and milites navclarii
Altinenses at Altina) had (at the beginning of the 5th century AD,
according to Notitia dignitatum) the name of milites navclarii. This
denomination shows, in accordance with the interpretation of
nauclathat there were troops with small numbers that used small
ships (naviculae) and were dedicated to certain small, concrete and
local objectives. Small ships, musculus, were used by milites
muscularii, the sailor combatants (marines) of the legio II Herculia
Inplateypegiis, as well.
Furthermore, the use of Vulgar Latin by the Roman
administration for the denomination of certain local Roman military
units (milites navclarii) should also be noted.
An observation with regards to the Vulgar Latin - Classical
Latin relationship: in our opinion, Vulgar Latin could not originate
from Classical Latin! The correlation must be reversed: Classical
Latin was probably an artificial language, as is today’s “Esperanto”
language. In order for it to have become an used language, it could
only have been learned in an organized manner, in schools, by a
mass of students who (perhaps) spoke Vulgar Latin (which had no
official orthography of its own and it was, may be, a mixture of
“coterritorial languages”) fluently.
A contemporary example is Australia, where more than
250 Indigenous Australian languages including 800 dialectal
varieties were spoken on the continent at the time of the European
settlement in 1788, only 13 traditional Indigenous languages are still
acquired by children and approximately another 100 or so are
spoken to various degrees by older generations, with many of these
languages at risk as Elders pass away ( AIATSIS, Aboriginal Institute
of Aboriginal and Torres Strait Islander Studies 2019. Indigenous
Australian Languages).
From the quote on pages 89-90 of Pârvan’s Getica, we note
that Genucla was located on the territory occupied by the Bastarnae
and that units of the Roman fleet also participated to its conquest in
the year 28 BC, and that from the previous page (page 88), it results
that king Roles was the ally of the Romans, and therefore a Romans’
client king.
In moving forward, let us briefly look over other events
interesting to our issue that took place in Dobrudja.
In the year AD 6, Octavianus Augustus founds the new
province of Moesia on the Morava and Timoc valleys, and in the year
AD 46 Thracia is transformed into a Roman province, however,
without the northern area of the Balkans, to the east of which a
praefectura ripae Thraciae continues to exist. Nothing changes until
Nero’s reign between AD 54 and 68 (Istoria Românilor/Romanian’s
History, vol. II, pp. 291-292).
The eulogy of Tiberius Plautius Silvanus Aelianus, the
governor of Moesia between AD 56 and 66 states that he “had
brought in ripam, quam tuebatur (as opposed to Moesia, which he
officially governed: praefuit) over 100,000 Trans-Danubian
individuals. This ripa must have been equivalent to Ripa Thraciae, i.e.
Dobrudja, as it shall be subsequently mentioned in the Roman
customs system” (Istoria Românilor/The History of the Romanians,
vol. II, p. 292). The respective inscription shows, among others, that
he “had brought kings, unknown until then, or enemies of the
Roman people, on the Danube’s bank, which he kept an eye on, as to
make them bow to Roman flags; he sent back the Bastarnae and the
Roxolani their sons held captive or rescued from their enemies, and
to the Dacian kings he sent back their brothers; however, from some
of them, he took hostages. Thusly he established peace in the
province and expanded its borders.” (Pârvan, Getica, p. 103). Of
those mentioned above, we take that Plautius Aelianus founded
client kingdoms here, in Dobrudja, intentionally. Upon this occasion,
the odris kingdom was given the task of keeping guard and of acting
police over the high shore, from Asam’s mouth to the west and up to
the Danube’s mouths to the east. However, “The Getae and the
Sarmatians were absolute rulers over Dobrudja, exactly as invaders
from the North, stronger than the locum tenens of Romans in the
Danube police”, as it is seen from Ovidius’ writings, Ex Ponto and
Tristia (Pârvan, Getica, p.97).
In the age of the Flavian dynasty (AD 69 to 96), the first
governor of Moesia during Vespasian’s time, Rubrius Gallus,
fortified the Danube border, so that the barbarians could not cross
the river. This is the moment in time when the first Roman auxiliary
troops appeared in Durostorum, Sacidava, Carsium, Arrubium, Troesmis
and Noviodunum (Istoria Românilor / The History of Romanians, vol.
II, p. 294).
Coming back to Murighiol-Genucla, we must presume that
Plautius Aelianus founded a client kingdom here as well, in the
Peucini-Bastarnae territory, in the 6th decade AD. The annual
amounts granted to this kingdom would have compensated the
loyalty of locals towards the Roman Empire, following that they
“bow to the Roman flags”, yet, perhaps more than that: provision of
ships for the Roman fleet and maybe even a potential cooperation on
the activity of building said ships at Genucla.
Let us try to explain further the uncommon situation of the
nine fragmented votive inscriptions (set between AD 136 and 200), of
which five mention vicus classicorum, inhabited by “residing Roman
citizens”. They do not mention the emperor’s name, not even the
exact day of the dedication, as it is usually done in the case of such
inscriptions, the actual date resulting only from the names of
councils, mentioned on some of the inscriptions.
The absence of the emperor’s name and the exact day of the
dedication could be explained by the fact that Vicus classicorum
Genucla was within an autochthonous, probably client territory,
which was not directly ancillary to the emperor. In this respect, we
must, however, note the fact that, in the 3rd century AD (according to
Itinerarium Antonini), Vicus Classicorum stayed behind the Roman
territory’s limes at Lower Danube, marked by the strategic military
road Aegyssus - Salsovia Halmyris –Vallis Domitiana Ad Salices
Histria (see figure 14). Furthermore, the dedications were symbolic
and not tied to a certain date.
The respective “residing Roman citizens” were cives Romani
consistentes, that is “persons pertaining to certain communities,
usually of Italic origin, which the Roman politics used as to learn all
that there was to learn about Barbarian territories (natural resources,
their intrinsic value, and implicitly, the political organization of the
respective Barbarian kingdom) and as to enforce the Roman system
of economic exploitation of the respective land; they occupied the
best places for carrying out their economic activities” (Benea, 2002).
Aside from the almost standard content of the inscriptions,
we must note that these inscriptions, along with another three
funeral inscriptions, were found under the rubble of the “northern
tower”, in a position that proved that they had been cemented in the
tower’s construction. The respective inscriptions sheeted the tower,
all having been set facing outward, the funeral ones being set
towards the “fort’s” exterior, and the votive ones - towards the
interior (Suceveanu, Zahariade, 1986).
The logical analysis of this exceptional situation leads us to
the following:
- the respective plaques were collected all together (probably
in the 3rd century, after the year AD 200, indicated by the no. 9
inscription) from a place where they already were in a fragmented
- moving them to the wall of the new building, facing the
exterior as to be permanently visible, constituted a symbolic act, with
the purpose of honoring those that had set the first inscriptions;
- remembering the dead on the exterior of the fortification
and placing the votive inscription on the inside represented an
optimistic message: “It is a new beginning. We honor the Mani Gods
and our predecessors; we ask the all-mighty and all-merciful Jupiter
to grant protection over the inhabitants of the vicus”;
- a coin issued at Histria between AD 238-244 was found on
the stratigraphic level N3 of the archaeological excavations; level N4
(cca 250 - cca 270) seems to indicate a state of destruction, and level
N5 (between 270-295, according to two coins from Probus, AD 276-
282) indicates the reconstruction of the citadel’s walls during the age
of Emperors Aurelian and Probus (S2003), perhaps after the final
defeat of the Goths, Peucini and Herules in 269, at Naissus, and
Claudius II is declared Gothicus Maximus (Istoria Românilor /
Romanians’ history, vol. II, pp. 303-304); therefore, the inscriptions
were moved during the last quarter of the 3rd century AD;
- with regards to the situation of the citadel at Murighiol
between 250-295 ascertained by the archaeological excavations
described above, it might also be linked to the events that occurred in
the area during that time: “In the spring of the year 256, the Goths,
Carpi and other Trans-Danubian peoples, after having built ships
(probably on the Dniester bank, with the aid of professional sailors
taken prisoners or commercially attracted), unleashed the invasion
upon the Empire, moving to the south, in two lines: one on the sea,
and the other one on land alongside the left Pontus shore, keeping in
touch with one another at all times... After which, they took the same
road home, filled with riches.” (Vulpe, Barnea, 1968, p. 254);
“In the spring of 269, enormous masses of Goths, Heruli,
Gepids, Bastarnae, Sarmatians, approximatively 320,000 men, had
gathered on the Dniester’s bank (Tyras) and boarding 2000 ships
(6000 according to Zosimus), they headed along the Scythia Minor’s
shore to the south, surrendering to the sea current, with the wind at
their backs. The warriors had their families, cattle, carriages with
them. It was no longer a matter of a simple marauding expedition,
but that of a massive migration, with a large proportion of civilians.
Exactly as in 267, the ships convoy was doubled by a line on
shore that moved forward, in parallel. It would have been impossible
that the thousands of cattle of all kinds that represented the
emigrants’ wealth to have remained crammed between the ships’
walls indefinitely. And their food and transportation required
passage on land... In their journey along the Dobrudja shore, the first
locality upon which the invaders stopped was Tomis, where they
debarked, trying to take the city’s walls by assault, however without
victory... Gallienus’ military reforms worked splendidly. The cavalry
organized by him decided the victory. The Roman fleet made by
him defeated the masses of enemy ships...” (Vulpe, Barnea, 1968, p.
263) and “Probus returned to the Lower Danube in 280, where he
received the allegiance of 100,000 Bastarnae (Zosimus I, 71)” (Vulpe,
Barnea, 1968, p. 286).
- the inhabitants, including the Bastarnae population
colonized by Probus in the year 280, shall remain, probably, of
Germanic origin in its majority (Alaric, the future king of the
Visigoths was born here in 370), however, the Roman authorities
probably took administrative and military control of the settlement
(given the inhabitation density, it results that probably almost the
entire population in that area was of Bastarnae origin). About the
professional Bastarnae mercenaries we learn from Plutarch: “For
there came to him [Perseu] from the Basternae, at his request, ten
thousand horsemen with ten thousand men to run at their sides, all
professional soldiers, men who knew not how to plough or to sail the
seas, who did not follow the life of herdsmen, but who were ever
practising one business and one art, that of fighting and conquering
their antagonists.” (Plutarch, Vitae parallele, Aemilius Paulus 12, 4).
The options proposed in time for the ancient name of the
fortification at Murighiol have been commented by the initiators and
executioners of the archaeological excavations in this area. As we
know, after having analyzed the proposals, they opted for
Halmyris” (Suceveanu, Zahariade, 1987, Suceveanu et al., 2003).
Before, Gratiana had also been one of the researchers’ choices:
after Vasile Pârvan (1913, 1924), who admits that the settlement at
Murighiol was called Gratiana, the issue of the Murighiol settlement’s
name is once again brought under discussion in 1968 (Vulpe, Barnea,
1968), and the Polish researcher Helena Gajewska, after a
documentation period in Romania, published a paper in Warsaw on
the Roman fortifications in Dobrudja, in which she also opts for the
name Gratiana (Gajewska, 1974, p. 154).
The Gratiana citadel’s name appears in two documents that
are essential for locating ancient forts in Dobrudja: Procopius of
Caesarea’ De Aedificiis (written approximatively in the year 550) and
Notitia dignitatum (approximatively in the year 400). Gratiana and
Valentiniana appear among other citadels mentioned on the
Procopius’ list of forts in Scythia. These two names suggest that they
were created or rebuilt (and renamed) after Valens’ victory against
the Goths, when Valens, and with him, all of its co-Augustians in the
west, Gratian and Valentinian, received the title of Gothicus maximus
in the same year (369).
Figure 27 – Models of shields specific to the milites primi