Journal of Archaeological Science: Reports 12 (2017) 876–894

Contents lists available at ScienceDirect

Journal of Archaeological Science: Reports

journal homepage: www.elsevier.com/locate/jasrep

Life cycle of estuarine islands — From the formation to the landlocking of

former islands in the environs of Miletos and Ephesos in western Asia
Minor (Turkey)
Helmut Brückner a,⁎, Alexander Herda b,c, Michael Kerschner d, Marc Müllenhoff e, Friederike Stock a
a
University of Cologne, Institute of Geography, Germany
b
Humboldt University of Berlin, Institute of Classical Archaeology, Germany
c
National & Kapodistrian University of Athens, Faculty of History and Archaeology, Greece
d
Austrian Archaeological Institute, Austrian Academy of Sciences, Vienna, Austria
e
Geo-present, Korbach, Germany

a r t i c l e i n f o a b s t r a c t

Article history: The focus of this article is to link historical accounts about former islands of the Anatolian gulfs of the Aegean Sea
Received 11 March 2016 to geoarchaeological evidence. During the Holocene, prominent environmental and coastline changes have taken
Received in revised form 9 November 2016 place in many tectonic grabens of western Asia Minor, today's Turkey. The Büyük and the Küçük Menderes fault
Accepted 14 November 2016
systems are excellent examples for deciphering these changes. Since mid-Holocene times, the eponymous rivers
Available online 12 January 2017
have advanced their deltas, silting up marine embayments which had once reached inland for tens of kilometres.
Keywords:
To describe this terrestrial–marine–terrestrial evolution of estuarine islands we coin the term “life cycle of estu-
Estuarine islands arine islands”. Besides other factors, such as natural erosion, sea-level changes, and tectonic activities, the delta
Ancient historians progradation was mainly governed by riverine sediment load, which, in turn, was to a great extent dependent
Geoarchaeology on human impact on the vegetation cover of the drainage basins. Based on historical accounts as well as modern
Miletos geoarchaeological research it is possible to reconstruct the spatio-temporal evolution of the landscape.
Ephesos For Miletos and the Büyük Menderes (Maiandros, Maeander) graben, remarkable transformations have been re-
Maiandros river vealed: the metamorphosis of the marine gulf into residual lakes (Lake Azap, Lake Bafa), the landlocking of
Kaystros river
islands (Hybanda, Lade, Asteria, Nergiz Tepe), the transition of the Milesian archipelago to a peninsula and finally
Büyük Menderes
to a part of the floodplain. A dramatic effect of the ongoing accumulation of fine-grained sediments was the sil-
Küçük Menderes
Asia Minor (Turkey) tation of harbours – a major reason for the decline of the once flourishing coastal cities of Myous, Priene,
Herakleia, and finally Miletos, today some 8 km inland.
For Ephesos and the Küçük Menderes (Kaystros) graben, the research focused on the former island of Syrie. Pliny
the Elder (Naturalis Historia, c. CE 77) attributed the landlocking of Syrie to the Kaystros River – a scenario which
has been verified by our geoarchaeological research and 14C–dated to the 5th century BC. The local foundation
myth according to which an island, presumably Syrie, was the location of the first settlement of immigrants
from the Greek mainland in the 11th century BCE can neither be proven nor disproven for lack of archaeological
evidence. The delta advance was the main reason why the settlement sites and the harbours of Ephesos were
relocated several times from the Early Iron Age to the Middle Ages.
© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Life cycle of an estuarine island Ensuing global warming caused a rapid marine transgression at an
average speed of more than 10 m in 1000 years (between 17 and 7 ka
During the Last Glacial Maximum (LGM) c. 20,000 years ago, when BP), which was even more rapid when so-called meltwater pulses oc-
worldwide the ice sheets as well as the permafrost regions had evolved curred. Palaeolithic and Mesolithic humans experienced this rapid
to their maximum extent, sea level had glacio-eustatically fallen ~120 m sea-level rise by the extensive drowning of parts of their game areas
below its present position (e.g., Waelbroeck et al., 2002; Peltier and and the interruption of their migratory routes.
Fairbanks, 2006). At this time, more than half of the continental shelves One effect of the marine transgression was that islands were “born”
were exposed and comprised dry land. The effect was especially dramatic when, depending on the topography, some parts of the land became
in regions with shallow shelves (cf. Sakellariou and Galanidou, 2016). separated from the mainland. This holds true for many islands of the
Aegean Sea, a prominent case being Samos.
⁎ Corresponding author. Another effect of the sea-level rise was the marine flooding of river
E-mail address: [email protected] (H. Brückner). mouths. Thus, gulfs evolved which reached far inland, sometimes for

http://dx.doi.org/10.1016/j.jasrep.2016.11.024
2352-409X/© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 877

tens of kilometres, giving birth to another series of islands, those in estu- progradation of the deltas of these rivers and their tributaries has con-
aries. According to Klug's (1985) typology of islands, they belong to the tinuously shifted the coastline westwards. The outcome was the prom-
group of “Drowning Islands”, a subsection of the “Ingression Islands”. If inent metamorphosis from a ria coast to a delta coast: the present-day
they form part of a ria coast, we suggest the term “Estuarine Islands”. bird's foot delta of the Büyük Menderes, and the blocked delta of the
The considerable deceleration of the transgression around Küçük Menderes. Important harbour cities, such as Miletos and Ephesos,
7000 years ago led to the formation of deltas (see also Stanley and which were located in the graben systems, fought against siltation with
Warne, 1994), thus reversing the direction: from the landward to the different weapons (dredging, damming, construction of a canal, reloca-
seaward shift in the shoreline. The speed of the delta advance was de- tion of harbours) – it was, however, a battle that could not be won.
pendent from the configuration of the embayment with its specific ac-
commodation space, mid- and late Holocene sea-level changes and
tectonics. The major factor, however, was the availability of sediments, 3. Methods
which, in turn, was dependent from the erodibility of the rocks in the
drainage basin, the carrying capacity of the rivers, the sediment input For reconstructing the former geographies of the environs of Miletos
by coastal currents, and, last but by no means least, human impact on and Ephesos, i.e. the Milesia and the Ephesia, we combined historical
the vegetation cover in the drainage basins as the major trigger of ero- accounts and evidence from archaeological excavations with
sion (Brückner, 1986). geoarchaeological results (including geophysical images, geological
Delta progradation led to the silting up of the marine embayments. data, interpretation of maps and satellite images). Since the beginning
In many cases this caused the landlocking of the estuarine islands. of the 1990s, many drill cores have been retrieved from the geo-
“Death” came, when the delta front had reached the island, thus turning bio-archives, such as alluvial plains, deltas, harbours, silted-up
it into a peninsula. Landlocking could also happen when littoral currents lakes, and swamps. The samples were analysed according to their
created a tombolo, which connected the island with the mainland. The geochemical, sedimentological, faunal and floral properties (cf. Stock
island's “life cycle” was completed when the delta front had passed et al., 2014; for the geoarchaeological research design see Brückner,
by, causing its total integration into the floodplain of the river. Already 2011). In many cases, coring in the study area followed the geophysical
in the 5th century BCE, the historian Thucydides correctly identified sil- research. The georadar, geomagnetic, and geoelectric images provid-
tation by the river Acheloos as the main reason for the landlocking of the ed first impressions of the subsurface strata. The interpretation of
Echinades islands: these data was then tested by coring. At very promising sites archae-
“Opposite to Oeniadae lie most of the islands called Echinades, so ological excavations were carried out; this is of course the “master
close to the mouths of the Achelous that that powerful stream is con- discipline”, but also the most expensive and time-consuming effort,
stantly forming deposits against them, and has already joined some of and in cases where groundwater table is high, it affords a costly drain-
the islands to the continent, and seems likely in no long while to do age system.
the same with the rest.” (History of the Peloponnesian War 2.102.3; The synopsis of all of these data leads to the spatio-temporal re-
transl. R. Crawley, 1874). construction of the landscape evolution, and sheds light on the
This palaeogeographic scenario was confirmed by geoarchaeological human-environment interactions – and in particular on the life
evidence (cf. Vött et al., 2007). That siltation was the fate of several cycle of the former islands. When describing them in detail, it is ev-
islands was also observed by Pliny the Elder (CE 23–79). In his encyclo- ident that the explanatory notes on those of the Büyük Menders gra-
paedic “Natural history” he writes: ben are richer than those of the Küçük Menders graben. Due to the
“Again she [natura, ‘nature’] has taken islands away from the sea and much bigger size of the Maiandros graben, it once hosted more
joined them to the land―Antissa to Lesbos, Zephyrius to Halikarnassus, than a dozen islands, while the Kaystros graben hosted only one
Aethusa to Myndus, Dromiscos and Perne to Miletus, Narthecusa to prominent island (Syrie). This imbalance is reflected in the following
Cape Parthenius. Hybanda, once an Ionian island, is now 200 stadia dis- sections.
tant from the sea, Ephesus has Syrie as part of the mainland, and its
neighbour Magnesia the Derasides and Sapphonia.” (Naturalis historia
2.204; transl. H. Rackham, 1967). 4. The Büyük Menderes (Maiandros, Maeander) and Miletos – delta
For this Holocene process – the “birth” of an island in the course of progradation, demise of islands, silted-up harbours
the postglacial sea-level rise, its flourishing time (“life span”) as an is-
land, the shores of which are exposed to marine waves, and its The tectonic graben of the Büyük Menderes (Maiandros, Maeander;
“death” when fluvial siltation becomes dominant with the result of see Herda, 2013a) is flanked to the north by the nearly 1300 m high
landlocking – we coin the term: “life cycle of an island”. This is especially mountain range of the Mykale (modern Dilek Dağları, formerly also
the fate of estuarine islands, which are part of the coastal metamorpho- Samsun Dağı), consisting of limestones and schists, to the south by Neo-
sis from a ria coast to a delta coast. gene marl- and limestones, and to the southeast by the granite and
gneiss massif of the Latmos Mountains (modern Beşparmak Dağları),
2. The study area which reaches up to 1332 m. In the river's drainage basin loosely consol-
idated Neogene rocks and weathered Palaeozoic mica schists outcrop;
Due to the westward drift of the Anatolian Microplate and the they have a high erosion potential, wherefore the Menderes alluvia
availability of sufficient accommodation space, Western Anatolia is are very rich in mica.
tectonically characterized by horsts and grabens. The latter render the Geoarchaeological research in the Milesia started in the 1990s: for
natural depressions for the river courses, e.g., the Büyük Menderes, Miletos and the lower Maeander valley see Brückner, 1995, 1996,
the ancient Maiandros (Nomen est omen!) and the Küçük Menderes, 1997a, 1998, 2003; Brückner et al., 2002, 2006, 2014a, 2014b;
the ancient Kaystros. Schröder, 1998; Bay, 1999; Müllenhoff, 2005; Müllenhoff et al., 2009;
The dramatic coastline changes described above holds particularly Herda et al., 2017). In this article we focus on the former islands that
true for the Mediterranean, and especially for western Anatolia had existed in the Ikarian/Karian gulf. The diverse and manifold material
(Kayan, 1999). About seven millennia ago, the maximum transgression of the Büyük Menderes graben and Miletos will be presented in three
of the sea had inundated the Büyük Menderes (Maiandros) graben for at sections: the Maeander delta progradation as a whole (Section 4.1);
least 25 km inland (Müllenhoff, 2005: 8, 187 f.; even for 60 km accord- the islands which became landlocked (Section 4.2); and new insights
ing to Bay, 1999: 27 f.), and the Küçük Menderes (Kaystros) graben for c. into the palaeogeography of the former Milesian archipelago
20 km inland (Brückner, 2005; Stock et al., 2015). Thereafter, the (Section 4.3).
878 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

4.1. Reconstructing the delta advance inhabitants of Myous had given up their city and had re-settled in
Miletos: the deposits of the Maeander had cut off a former marine
4.1.1. Delta advance according to geoarchaeological evidence inlet so that a coastal lake evolved; after freshening it became the breed-
The typical coring profile – as documented, for example, in the later ing place for mosquitoes, which made living in Myous unbearable.
discussed Figs. 7 and 15 – starts with transgression facies (pebbles, “The Ionians who settled at Myous and Priene, they took the cities
coarse sand), lying with erosional disconformity on the pre-Holocene from Karians. (…) The people of Myous left their city on account of
bedrock. These deposits were accumulated in a littoral environment, the following accident. A small inlet (κόλπος) of the sea used to run
when the post-glacial marine transgression had reached the coring into their land. This inlet the river Maeander turned into a lake
site (1st transition of the shoreline). With the continued rise in sea (λίμνη), by blocking up the entrance with mud (ἰλύς). When the
level, the environment turned into shallow marine, as evidenced water, ceasing to be sea, became fresh, gnats (κώνωπες) in vast swarms
by finer-grained strata (fine sand, silt) and a milieu-specific fauna. bred in the lake until the inhabitants were forced to leave the city. They
The fining-upward sequence may be reversed during the following departed for Miletos, taking with them the images of the gods and their
regression. The 2nd transition of the shoreline may be expressed by other movables; and on my visit I found nothing in Myous except a
coarse-grained material (pebbles, coarse sand), indicating a littoral white marble temple of Dionysos.” (Pausanias, Description of Greece
environment. If, however, a sandbar has evolved seawards, a lagoonal 7.2.10–11; transl. Jones and Ormerod, 1918).
environment is created at the coring site with very fine-grained deposits A few passages later the same author describes the major environ-
(clay) and a typical faunal association. In most cases, the top of the pro- mental change from a former marine gulf into arable land resulting
file comprises alluvium, deposited by the river and its affluents during from the siltation of the river (for the position of the contemporaneous
floods. shoreline see Fig. 1, scenario between the turn of the eras and CE 300):
Fig. 1 shows the scenario for the spatio-temporal advance of the “My reasoning is confirmed by the fact that the Maeander, flowing
Maeander delta in eight time slices: 1500 BCE, 800 BCE, 300 BCE, turn through the land of the Phrygians and Karians, which is ploughed up
of the eras, CE 300, CE 100, CE 1500, and the present situation. This in- each year, has turned in a short time the sea between Priene and
terpretation is based on c. 300 percussion corings similar to the one de- Myous into solid land.” (Pausanias, Description of Greece 8.24.11).
scribed above. The depositional environments and their changes are To whom did the newly created land belong? An inscription found in
deduced from the faunal associations (esp. ostracoda and foraminifera), Miletos sheds light on this important issue: in the time of Roman em-
which in turn leads to the reconstruction of the coastal changes. The peror Augustus (31 BCE – CE 14), the Milesian Caius Iulius Epikrates
chronostratigraphy relies on radiocarbon (14C) age estimates, archaeo- gained permission from Augustus to take possession of “the land
logical as well as historical evidence (details in Müllenhoff, 2005). We newly turned to earth by the Maeander, and the sandbanks” (τὴν
shall come back to Fig. 1 several times when discussing the many former ἀπ[ο]γ̣α̣ι̣[ου]μένην χώραν ὑπὸ τοῦ Μαιάνδρου | καὶ τοὺς
estuarine islands of the Ikarian/Karian gulf. γαι̣εῶ̣ νας).
He did this in the name of the city of Miletos, who honored him with
4.1.2. Delta advance according to literary sources a statue in the gymnasium in return (Herrmann, 2006; Brückner et al.,
Deltaic growth has been described in ancient literary sources. Pausa- 2014a: 86). This new land as well as the profitable fisheries in the la-
nias (later 2nd century CE) correctly explains the reason why the goons along the coastline, were either located north of Miletos, in the

Fig. 1. Progradation of the Büyük Menderes (Maiandros, Maeander, Meander) delta since 1500 BCE As for the general topographic situation see inserted map in Fig. 10. The advancing delta
front has continuously shifted the shoreline towards the southwest, infilling the former marine embayment. Bafa Gölü has not been silted up since the delta bypassed this region and
sediment input from the adjacent mountains is low. The former islands of Lade, Hybanda, and Nergiz Tepe (N) are noted. Source: Müllenhoff, 2005.
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 879

Fig. 2. Historical topography of southern Ionia with adjacent parts of Lydia and Karia. The shoreline is that of the Late Geometric Period (c. 700 BCE), the lifetime of Homer. Light grey:
approximate area of the later Kaystros and Maiandros deltas still occupied by the sea c. 1500 BCE, at the end of Minoan dominated Miletos IV and the beginning of Mycenaean
dominated Miletos V. Source: Herda, 2009, 45 Fig. 3; supplemented by Herda et al., 2017.

delta area of the “Old” Maeander approaching from Priene, wherefore BCE, see Müllenhoff, 2005) (Fig. 2). By then, Homer performed his
both cities had been in rivalry about their ownership (Thonemann, Greek-Ionian epic Iliad, singing of the Karians of Miletos who own the
2011: 322 f., 334), or it had accumulated at the mouth of the “New” mountains of Mykale as well as Latmos (Phthiron oros) and “the streams
Maeander west of Myous and became disputed between Miletos and of Maeander” (Maiandrou te rhoas; Homer Iliad 2.869; cf. Herda, 2009:
its other adversary in the lower Maeander plain, Magnesia. 37–45) – the plural may hint to bifurcation, a typical phenomenon of
deltas.
4.2. Former islands in the gulf of Miletos It is amazing that the tradition of this former island survived for such
a long time. The reason may have been that the winter floods of the
4.2.1. Hybanda Maeander turned Hybanda into an island again; this was still the case
Hybanda bears a pre-Greek, indigenous Anatolian-Luwic name until the early 20th century CE (see Wiegand, 1929, Beilage). Plus, the
(Zgusta, 1984: 647 § 1395 Hybanda). Already in 1775, Richard Chandler Hellenistic designation Hybandis for the territory (chora) around the
convincingly identified it with a 40–70 m high group of four hills at the former island (Lohmann, 2006: 199) points to an unbroken settlement
modern village of Özbaşı, c. 13.5 km south of today's Söke and 4 km activity in ancient Hybanda, a few traces of which were found west of
northeast of the ancient city of Myous (Chandler, 1775: 178). the modern village of Özbaşı (Lohmann, 2006: 198, 254 Fig. 3
Hybanda (Figs. 1, 2, 3) is among the islands enumerated by Pliny in “Felsbettung”). The knowledge about the former island of Hybanda
his famous account mentioned above. It became landlocked by the will therefore have survived as part of a local tradition, Pliny could
Maeander alluvium already in Late Geometric times (late 8th century rely on.
880 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

Nevertheless, the 200 stadia (36 km) the author mentions as being 1 F 241). The name Lade (Λάδη) or in its elder form Late (Pliny, Naturalis
the distance between Hybanda and the open sea are “l'exagération Historia 5.135) may be, like Hybanda and Maiandros, of indigenous
rhétorique” (Robert, 1959: 23 f.): in reality it amounted to only 112 sta- Anatolian origin (Bürchner, 1924: 381; Herda, 2009: 99).
dia (20 km) during his lifetime (Müllenhoff, 2005: 197 Fig. 51). Even Lade, today's Batmaz Tepeleri (“the unsinkable hills”, see Brückner
today the said distance is only c. 150 stadia (27 km) (Müllenhoff, et al., 2014a: 86 with n. 155; Brückner et al., 2014b: 795 with n. 65), is
2005: 214 Fig. 56) (Fig. 1). nowadays part of the delta plain, up to 99 m high and 3 km long. It is sit-
The “Hill of Hybanda” (〈Hyb〉andios lophos) was the site of a battle in uated 2 km east of the coastline and 2.5 km west of Miletos (Fig. 4). In
428/427 BCE. An Athenian expeditionary force under the command of ancient times, the closest distance between the island and the coast
Lysikles had landed in the harbour of Myous and started to march up was c. 1.3 km (Tuttahs, 2007: 339). Geologically, Lade consists of fresh-
the Maeander plain to collect the tribute from their Karian ‘allies’, water limestone (Neogene Akbük formation); it is a tectonic block
when they were attacked and defeated by a united Karian-Anaiïtian which had slid into the Maeander graben (Müllenhoff, 2005: 23;
army near the hill (Thucydides 3.19.2 has Sandios lophos; for emenda- Tuttahs, 2007: 30–32). Due to its prominent position at the southern en-
tion see Rayet and Thomas, 1877: 25 n. 1. p. 27; Herda et al., 2017: trance to the marine gulf as well as its proximity to Miletos, Lade was
chap. 5 with n. 198 f.). By then, the former island must already have the strategic key to control the access to Miletos and the other harbour
been at least partly integrated into the delta plain by the sediments of cities of the region. So far, archaeological work has been restricted to an
the southern branch of the Maeander (today's Büyük Menderes) and extensive survey, and nearly nothing is known about the settlement his-
its tributary, the Hybandos (today's Kısır Çay) which joins the main tory of Lade (Lohmann, 2006: 204 f. s.v. Lade).
river 4 km to the west of Hybanda. In several naval battles, the island played a decisive role. In the fa-
After the city of Myous had been given up by her inhabitants (cf. mous “Battle of Lade” in 494 BCE, the victorious Persians ended the
chap. 4.1.2), the Hybandos river became the border between Magnesia Greek-Ionian rebellion against their rule. The mid-5th century BCE his-
and Miletos in c. 185/4–180 BCE. The surrounding land of the Hybandis, torian Herodotus reports this battle as follows:
formerly in possession of Myous, fell to Miletos which integrated the “These [the Ionians], when they came to that place [the Panionion]
Myousians into her own citizen body. A peace treaty ended the long peri- and there consulted, resolved to raise no land army to meet the Persians,
od of ‘alluvium wars’ between Magnesia and Miletos for the newly accu- but to leave the Milesians themselves to defend their walls, and to man
mulated lands (A. Rehm in: Kawerau and Rehm, 1914: 341–349 no. 148 their fleet to the last ship and muster with all speed at Lade, there to
lines 28–38 [peace treaty]; cf. Herda et al., 2017: chap. 5 with n. 189–194). fight for Miletos at sea. This Lade is an islet (nesos mikri) lying off the
city of Miletos.” (Histories 6.7; transl. Godley, 1975).
4.2.2. Lade In 334 BCE, the occupation of Lade by the Macedonian fleet pre-
Lade is first mentioned by the Milesian geographer and ethnogra- luded Alexander's conquest of Miletos, as the 2nd century CE histori-
pher Hekataios in his “Journey round the world” (Periegesis/Ges an Arrian describes it in his Anabasis (1.18.4–5; cf. von Graeve, 2000:
Periodos) c. 500 BCE (Die Fragmente der griechischen Historiker (FGrHist) 118. 121. 124 f.). Alexander's troops occupied the island of Lade and

Fig. 3. View from today's Güllübahçe (village in the foreground) at the southern slope of Dilek Dağları (Mykale Mountains), immediately east of ancient Priene, towards southeast via the
floodplain of the Büyük Menderes (Maiandros) (M). On the southern side of the valley the former island of Hybanda (H) is visible in front of Beşparmak Dağları (Latmos Mountains) (LM).
The area of the valley floor, which has been intensively cultivated (green and yellow fields) since the installation of an irrigation-drainage system in the 1950s, was once part of the
extended marine embayment of the Ikarian resp. Karian Sea. The course of the “Old” Maeander (OM) is seen in the middle- and foreground (greenish yellow meandering (!) line).
Photo: M. Müllenhoff, August 1999.
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 881

the strait between Lade and the opposite mainland to cut off the Per- middle hilltop of former Lade at 72 m a.s.l. (Wilski, 1906: map, quadrant
sians and Phoenicians in besieged Miletos from help by the Persian C4 “m-a Rundbau”; Thonemann, 2011: 264 n. 66).
fleet: The Pyrgos estate which became more and more integrated into the
“Nicanor, however, bringing up the Greek fleet, reached Miletus newly accumulated terrain, is described in the monastery's acts as being
three days ahead of the Persians, and anchored at the island of Lade “the land which lies between the two rivers, the Maeander and the river
with 160 ships. Lade is over against Miletus. The Persian fleet were too flowing from Palatia [i.e. Miletus]” (Patmos II.61 line 19; cf. Ragia, 2009:
late, and when their commanders learnt of Nicanor's arrival already at 42. 198 f. with n. 847; Thonemann, 2011: 263 f. with n. 66). This is a
Lade, they anchored under Mycale.“ (Anabasis 1.18.4–5, transl. P.A. clear indication that the main branch of the northern “Old Maeander”
Brunt, 1976). emptied north of former Lade. The second river “flowing from Palatia”
In 200 BCE, another naval battle was fought at the island of Lade: the may likely have been a minor branch of this “Old Maeander” which
one between the Rhodians and Philip V (Polybios 16.15.6; cf. Brückner had found its way between the Milesian peninsula and Lade, and started
et al., 2014a: 68). to close the strait between the former island and the mainland, once
The landlocking of Lade started in the Late Roman period, when the called in the acts also “the narrow of Palatia” (Patmos I.14 lines 22 f.
northern branch of the Maeander, the Eski Menderes (“Old Maeander”), ἐγγὺς τῶν Παλατίων; cf. Thonemann, 2011: 264 n. 66).
approached the Milesian peninsula (Fig. 1). In the mid-4th century CE, Patmian ownership of the Pyrgos-lands around Lade was acknowl-
the governor of Asia, Skylakios, had to build “a canal through the edged by the Ottoman emperor Mehmet II in CE 1454 (Ragia, 2009:
plain”; he also “turned the Maeander back into a natural course” and 199 χωριό της Πάτμου), who also granted the monastery, called
“restored the harbours to the city” of Miletos (Himerios Orationes Batnos by the Turks, administrative independence and immunity from
25.73–95; cf. Thonemann, 2011: 318 f.; Brückner et al., 2014a: 87, attacks (Thonemann, 2011: 290 f.). Even when the extremely profitable
Brückner et al., 2014b: 796). By then, Miletos' access to the open sea salines, located in the lagoon northwest of former Lade and the mouth
was already reduced to the small strait between Lade and the mainland of the “Old Maeander”, which had been part of the Pyrgos estate, became
in the south (cf. Fig. 4). In the following half-millennium, a lagoon devel- property of the Ottoman state in the 16th century CE, they kept their old
oped northwest of Lade, and by c. CE 1000 the sediments of the “Old” owners name, “Batnos saltpans” (Thonemann, 2011: 328 f.). The 19th
Maeander (the term is first used in 13th century CE acts of the monas- century CE Ottoman Greek village Pat(i)niotiko equally preserved the
tery of St John the Theologian on Patmos; Tomaschek, 1891: 36) land- Patmos-link (Lohmann, 2006: 235 f.), as does the modern Turkish
locked the eastern part of the island (Müllenhoff, 2005) (cf. Fig. 1). Batmaz Tepeleri, a folk etymology related to the fact that the hills of for-
Around that time, in CE 1073, the island together with a village on it mer Lade “do not sink” (Turkish batmaz) during the winter floods
called Galaidai as well as other estates in the lower Maeander plain had (Brückner et al., 2014a: 86 n. 155).
been given by the Byzantine emperor Michael VII Doukas to his cousin
Andronikos Doukas. In this context the name Lade appears for the last 4.2.3. Asteria and the grave of Asterios
time in the acts of the monastery of St John the Theologian on Patmos According to the ancient travel guide Pausanias (mid-2nd century
(Patmos II.50 line 167; cf. Thonemann, 2011: 263. 260 map 11). In CE CE), Lade was surrounded by some detached islets. One of them was As-
1216, the monastery of St John successfully addressed the emperor teria, “the Starry” or “Comprised of Stars”. It gained some fame as being
Theodore I Lascaris to get the lands around Pyrgos, “tower”, as is was the place of the tomb of the autochthonous Asterios, the son of the giant
called by then after a medieval round watch tower located on the Anax (Wernicke, 1896; cf. Herda, 2013b: 90 f.):

Fig. 4. View from the theatre of Miletos on the up to 99 m high hills of the former island of Lade (Batmaz Tepeleri) (L). To the right (north) the Maeander plain, to the left (south) the 1.3 km
broad former strait between Lade and the mainland. Photo: H. Brückner, 25.09.2015.
882 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

“Before the city of the Milesians is an island (nesos) called Lade, and and that life on Earth, including humans, originally stemmed from sea
from it certain islets (nesides) are detached. One of these they call the creatures (Aetius Placita Philosophorum 5.19.4; Kirk et al., 1994: 153
islet of Asterios, and say that Asterios was buried in it, and that Asterios on no. 133; Mayor, 2000: 73 f. 214 f.; Boardman, 2002: 33–43. 217 no.
was the son of Anax, and Anax the son of Earth. Now the corpse is not 159).
less than ten cubits.” (Description of Greece 1.35.67; transl. Jones and
Ormerod, 1918).
Pausanias seemed to have visited the site personally, where he was 4.2.4. Nergiz Tepe
shown the remains of Asterios. The small size of the corpse (nekros), ob- Nergiz Tepe, the “Narcissus Hill”, surfaces in the southern Maeander
viously too small for the son of a giant, provoked his ironic remark “not graben some 7 km east of the former Milesian peninsula, which itself
less than ten cubits (5 m)”. forms part of the so-called Nergiz Tepe formation (Miocene freshwater
Asteria is likely to be identified with a 22 m high rock less than 1 km limestone; Schröder et al., 1995: 239) (Fig. 6). Today, the highest point
west of the village of Pat(i)niotiko, today's Batı Köy (“West Village”) on lies 6 m above the surrounding alluvial plain, the level of which is 2 m
the eastern slopes of the former island of Lade (Fig. 5), an identification a.s.l., i.e. about 1 m lower than the plain's average level in this region
already proposed by Olivier Rayet, the first excavator of Miletos (Müllenhoff, 2005: 206). Corings have shown that Nergiz Tepe formed
(1873–1874), in 1877 (Rayet and Thomas, 1877: map. pl. 2 “I. an island in ancient times (Müllenhoff, 2005: 158–162). Lohmann
Astérion”). The distance to Miletos adds up to 2.5 km. In early 20th cen- (2006: 186 s.v. Dromiskos, 237 s.v. Perne) proposes to identify the hill
tury maps, this small hill is named in Turkish Mezar Tepe, “Hill of the with the island of Dromiskos, mentioned by Pliny (Naturalis historia
Graves” (Wilski, 1906: map, quadrant C3 “Mesartepe”; Philippson, 2.204) as already landlocked in his time (cf. chapters 1 and 4.3). But
1936: map, “Mesartepe”). It seems, however, as if the graves on Mezar this does not fit with the dating for the landlocking of Nergiz Tepe ac-
Tepe are modern and not ancient. They are not distinguished from cording to Müllenhoff (2005), which Lohmann admits. As yet, the
those in the graveyard of today's Batı Köy. Furthermore, ancient ruins island's ancient name is unknown to us. It still existed around CE
or graves are otherwise designated in red on the map. Today, no re- 1000, when it was situated in the western part of a gulf which extended
mains of these graves have survived. The limestone of the hill has from Miletos as far southeast as Herakleia under Latmos, and was
been extensively quarried for building a 15 km long asphalt road named “Milesian Lake” (Milesie Limne) or “Gulf of Melamitorum”
through the Maeander plain, connecting Miletos-Balat with Tuzburgazı (Sinus Melamitorum) in late Byzantine texts (Herda et al., 2017:
at the foot of the Mykale Mountains. The landlocking of Asteria by the chap. 6 Fig. 12).
alluvia of the “Old Maeander” occurred roughly contemporaneously The island's landlocking occurred during the progradation of the
with that of eastern Lade, i.e. c. CE 1000. southern branch of the “New” Maeander (today's Yeni or Büyük Mende-
It has convincingly been assumed by classicists as well as res) which bypassed it in the 14th or 15th century CE, thus cutting off
palaeontologists that the bones shown to Pausanias as those of the gi- the former Gulf of Herakleia/Melamitorum which turned into a brackish
ant's son Asterios were actually Miocene fossils of marine animals. interior lake, today's Lake Bafa (Müllenhoff, 2005: 203 Fig. 53; Herda
These fossils, which are abundant in the Neogene limestone of the west- et al., 2017) (cf. Fig. 1). Due to the lower level of the alluvial plain sur-
ern coast of Asia Minor and the adjacent islands, e.g. Samos, may have rounding Nergiz Tepe (Müllenhoff, 2005: 162), the winter floods re-
inspired the Milesian philosopher Anaximandros (c. 635–545 BCE) to peatedly turned the hill into an “island”, until the Maeander was
develop his theory that the sea once covered what is now dry land, regulated in the 1950s.

Fig. 5. The former island of Asteria (Mezar Tepe) seen from the southeast. According to a Milesian myth, here the grave of the giant Asterios, son of Anax, the first founder of Miletos, was
situated. The 22 m high hill consists of freshwater limestone. It has been extensively quarried for road construction in recent times. Photo: H. Brückner, 26.09.2015.
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 883

4.3. Miletos: from archipelago via peninsula to floodplain This story is of special interest as the normal strategy for founding a
colony is contrary: for security reasons the newcomers usually first
4.3.1. The Milesian peninsula, a former archipelago choose an island close by, before they enter the mainland (see
Our geoarchaeological studies revealed that the Milesian peninsula chap. 5.1). The story, therefore, likely preserves a historic kernel. The
as a site of the famous Ionian metropolis developed from an archipelago, landlocking of the former island of Miletos is a definite historic fact;
settled since at least the 4th millennium BCE (Brückner, 1996, 2003; the “dam” (gephyra) may actually be the natural sandbar (tombolo) be-
Brückner et al., 2006, 2014a: 56–64; Müllenhoff et al., 2009). During tween the mainland and the ‘Athena Temple Island’, which our corings
the time of the maximum Holocene transgression c. 2500 BCE, which revealed. This tombolo is the result of coastal accumulation caused by
equals the Early Bronze Age settlement period Miletos II (cf. Niemeier, longshore currents and adequate sediment supply. It can be dated to
2007), this archipelago consisted of two larger and two smaller islands the 15th century BCE at the latest, when Miletos became dominated
(Fig. 7). The two smaller islands, supposedly named Dromiskos and by Mycenaean Greeks (Miletos V–VI, c. 1450–1220 BCE; cf. Brückner
Perne, will be dealt with later. The two larger islands lack any ancient et al., 2014a: 57 ff.).
name except for ‘Miletos’. We call the one to the west ‘Athena Temple A similar tombolo was detected between the ‘Kale Tepe–Humei Tepe
Island’, and the one to the east ‘Kale Tepe–Humei Tepe Island’. They Island’ and the mainland. But it developed much later, during
were separated from each other by a 150 m broad strait. Their distance Protogeometric to Geometric times (Miletos VIII a/b, 11th–8th centuries
to the mainland amounted to only 300–350 m (Figs. 7-8). BCE; cf. Brückner et al., 2014a: 64), when the site was settled by Ionian
The flat, up to 9.60 m a.s.l. high Athena Temple Islands, measuring c. Greeks. In the course of this process, the strait between the Athena Tem-
600 m E–W and 400 m N–S, seems to have been the centre of Minoan ple Island and the Kale Tepe–Humei Tepe Island was closed and turned
Miletos (Miletos III–IV, c. 1900–1450 BCE; cf. Niemeier, 2007: 8–13). into a well-protected embayment. By then the beach harbour on the
It provided a small indented harbour embayment to the north, the eastern shore of the Athena Temple Island became part of the ‘Theatre
‘Athena Harbour’ (Brückner et al., 2014a: 92 f.), as well as a large Harbour’, one of the two closable harbours of Ionian Miletos, where
beach harbour in the leeward position to the prevailing northerly part of her fleet was stationed – the backbone of Milesian colonisation
winds (“Etesia”, “Meltemi”) along its eastern and southern coastline. ventures from the 7th century BCE (Brückner et al., 2014a: 64. 69 f. 89
A local Milesian myth, preserved in a Late Antique scholium on the ff.; Brückner et al., 2014b: 775. 782 f.) (Fig. 8). The other closable har-
early 2nd century CE geographer Dionysios Periegetes, tells us that the bour, the ‘Lion Harbour’, was located between Kale Tepe and Humei
Minoans, arriving under the leadership of name giving Miletos, first set- Tepe, forming a 300 m long and 100 m broad indentation in the Kale
tled on the mainland in a place called Oikous, “houses”, a later suburb of Tepe–Humei Tepe Island (Brückner et al., 2014a, 2014b).
the Greek city of Miletos, before Miletos' son Keladon moved to an is- During the following 350 or so years, the newly developed Milesian
land nearby (cf. Herda, 2005: 288 f. 291): peninsula, consisting of two former islands and two connecting tomboli,
“But Keladon, who ruled over Oikous, buried his father on an island eventually grew together due to continued littoral deposition, as well as
nearby where he himself went to dwell because of an oracle. He called human-induced erosion and denudation processes from the adjacent
the island ‘Miletos’. A dam (gephyra) connects today's Oikous with hills to the south, which in turn caused increased siltation. Parallel to
Miletos.” (Scholium on Dionysios Periegetes 825). this development, the size and structure of the settlement grew. Besides

Fig. 6. Nergiz Tepe, presently surrounded by fields, was an island in the marine gulf (Ikarian/Karian Sea) and even later in the “Milesian Lake” (Milesie Limne). Its freshwater limestone is
heavily quarried. The present name means “Narcissus Hill”, the ancient name is unknown to us. View towards northwest, in the right background the Dilek Dağları (Mykale Mountains).
Photo: H. Brückner 26.09.2015.
884 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

Fig. 7. The Milesian archipelago during the time of the maximum marine transgression c. 2500 BCE. By then it consisted of two larger islands: the so-called ‘Athena Temple Island’ in the
west, where the mythical founder Miletos of Crete was buried, and the extended ‘Kale Tepe–Humei Tepe Island’ in the centre. East of the latter, two smaller islands were located in
antiquity, called Dromiskos and Perne. They were connected to Humei Tepe via sandbars (tomboli) in the life time of Pliny (CE 23-79) or even before. Source: Brückner, 2003: Fig. 4
(slightly modified).

the old Bronze Age centre on the former ‘Athena Temple Island’, The landlocking of formerly marine areas of the peninsula continued
Palaimiletos (“Old Miletos”), as it was called by the 4th century historian through Classical, Hellenistic and Roman times, increasing significantly
Ephoros (FGrHist 70 F 127 = Strabo Geographica 14.1.6), new districts when the northern “Old” Maeander reached the waters north of Miletos
developed from modest beginnings: first of all around the Lion Harbour, around CE 300. The ‘Kalabak Tepe Harbour’ to the SW was silted up, as
where Apollo ‘Delphinios’, the “Dolphin”, the main god of Iron Age was the ‘Athena Harbour’. The ‘Theatre Harbour’ had lost most of its
Miletos (Miletos VIII, c. 1050 BCE–CE 400), was said to have landed in capacity already in Late Classical times, while the ‘East Harbour’ may
mythical times, wherefore his sanctuary, the Delphinion, was located simply have moved further to the east. During the time of the geogra-
there (Callimachus Branchos fr. 229.12–13; cf. Herda, 2005, 2008, pher Strabo (first half of the 1st century CE), four harbours were still
2011: chap. 2.7; Brückner et al., 2014a: 65 ff.). in use, one of them, the ‘Lion Harbour’ or ‘Harbour of Dokimos’, func-
A new factor in changing the landscape of the Milesian peninsula tioning as a lockable harbour for a fleet (Strabo Geographica 14.1.6; cf.
was systematic artificial infill of formerly swampy areas as a means of Brückner et al., 2014a: 64 ff., Brückner et al., 2014b: 774 ff.).
urban development since the early 6th century BCE, detected by our
corings (Fig. 8). Affected were the major public areas of the southern 4.3.2. Finally connected: Dromiskos and Perne
‘Lion Harbour’, the ‘North Market’ and the Delphinion sanctuary, as In close vicinity to Miletos two islands had been situated, which in
well as the eastern ‘South Market’. These measures formed part of the the lifetime of Pliny (CE 23–79) or even before had become the most re-
implementation of a new orthogonal insula-street grid which optimized cent part of the peninsula. In the already quoted passage (chap. 1), Pliny
the organisation of traffic in between the six harbours of Miletos, the mentions that
city area, and its countryside. It can be argued quite convincingly that “nature” (natura) “has taken islands away from the sea and joined
the planners behind were men like the philosophers Thales and his them to the land (...) Dromiscos and Perne to Miletos” (Naturalis Historia
companion Anaximandros, who founded the Milesian school of natural 2.204; transl. H. Rackham, 1967).
philosophy, the first Greek philosophical school (Herda, 2005, 2013b: The two islands are not mentioned by name in any other preserved
84 ff.). ancient source, so we have to rely on their names as primary
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 885

Fig. 8. The Milesian peninsula and its harbours in Archaic–Roman times. The position of the shoreline is indicated for three time slices (maximum marine transgression, Archaic time,
Roman Imperial time). Source: Brückner et al., 2014a: Fig. 10.
886 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

information. The Latin name Dromiscos, easily recognisable as Greek It has been overlooked so far that in 1877 Olivier Rayet (1847–1887)
Dromiskos (Δρομίσκος), is etymologically clear: it is the diminutive of in his groundbreaking geoarchaeological reconstruction of the Maean-
dromos, the Greek word for “race”, “course”, or a “place for running”, der progradation proposed to identify these two small rocks 300 to
“race-course” for athletes, horses, or chariots, as well as a “broad, 350 m east of Humei Tepe with Dromiskos and Perne (Rayet and
straight street” resp. “avenue” (Liddell et al., 1996: 450 s.v. δρόμος). Thomas, 1877: 30 f., map pl. 2 “Iles Dromiscos et Perné”). The only
As a topographical term it can designate a long, small peninsula, like scholar who ever recognized the discovery of Rayet was Carl Humann
the Achilleos Dromos at the mouth of the Borysthenes river in the north- in his map of 1891, which has never been properly published, but only
ern Black Sea. Therefore, it seems likely that Dromiskos was originally a in a small copy (Weber, 2007: 328 Fig. 1; to be read with a magnifying
small, longish island (Bürchner, 1905a assumed “a narrow tongue of glass: “Ins. Dromiskos?” below “Kütschük Tschakmak-tepe”, and “Ins.
land which was deposited by the Maeander“; Lohmann, 2006: 186 s.v. Perne?” below “Böjük Tschakmak-tepe”). The two islands have been
Dromiskos). neglected in recent studies. The only exception is Tuttahs (2007: 358 f.
Perne (Πέρνη) is likely a Pre-Greek Anatolian-Luwic place name, re- Fig. 388). This is partly depending on the task of how to map the large
lated to Hittite parn-, pir-, “house” (Zgusta, 1984: 485 § 1045 Περνις), area of ancient Miletos properly. In the first years of excavation reports,
which would be a suitable name for a settled island. Zgusta lists a Lycian the two islands still appear in the maps depending on that of C. Humann
site named Pernis (Περνις), perhaps located in the neighbourhood of of 1891 (see Weber, 2007: 328–331 Figs. 1–4), while they are left out
Limyra. If the name is originally Greek instead, or was later furnished after 1906 (see Weber, 2007: 331–335 Figs. 5–6. 8). They reappear as
with a Greek ‘folk etymology’, it can designate two quite different islands on the maps drawn by Armin von Gerkan (1924–1935) for
things: either it is the Ionian dialectal form of perna (πέρνα), “ham”, 2nd century CE Miletos (Weber, 2007: 336–339 Figs. 9–10). After WW
or the present participle form of the verb pernemi (πέρνημι), “export II, they are left out once again, first of all in the influential new topo-
for sale (slaves)”, “sell” (Liddell et al., 1996: 1394 s.v. πέρνα, 1394 f. graphical map by Walter Bendt of 1959–1965 at a scale of 1:2000
s.v. πέρνημι). The latter case seems most likely. The participle form (Bendt, 1968).
pernas (περνάς) of the verb pernemi, used in the epics as terminus According to our reconstruction both islands were landlocked be-
technicus for “’exporting’ captives to foreign parts ‘for sale’ as slaves”, tween the 1st and 2nd centuries CE, when the Maeander had not yet
is attested in Homer (Iliad 22.45; cf. Chantraine, 1968–1980: 888 s.v. reached Miletos. Therefore, the reasons for the landlocking are denuda-
πέρνημι; Liddell et al., 1996: 1394 f. s.v. πέρνημι), where Priamos be- tion from the mainland, and/or the formation of sandbars (tomboli) due
moans on the walls of Troy that Achilles has “killed or sold (pernas) to to longshore drift in the leeward position of the peninsula. The already
distant islands” (κτείνων καὶ περνὰς νήσων ἔπι τηλεδαπάων) sev- discussed results for the Maeander delta progradation (Müllenhoff,
eral of his sons, and is now going to kill Hector, too. Did the Milesians 2005) largely confirm this reconstruction (cf. Fig. 1), which nevertheless
have this famous passage of Homer in mind, when they named the is- still has to be demonstrated by corings in the area of the assumed
land, or took over the Pre-Greek place name by ‘folk-etymologizing’ tomboli. That Rayet also gave preference to this scenario becomes
it? Possibly, because it functioned as a place, where slaves were traded evident from his reconstruction of the changes of the eastern shoreline
in reality. At least the Homeric epics were well known to the Milesians of the Milesian peninsula in three steps: 500 BCE, 1st century CE, end of
as their performance formed the climax of the Greek-Ionian national 2nd century CE (Rayet and Thomas, 1877: 30 f., map pl. 2 “Iles
festival in the Panionion. That Perne could have served as the place of Dromiscos et Perné”; see Herda et al., 2017: chap. 4 Fig. 7).
a slave market seems all the more likely when we take into consider- The landlocking of Küçük and Büyük Çakmaklık, respectively
ation its location as well as that of Dromiskos. Dromiskos and Perne, substantially improved the conditions for the
Today, two adjacent small rocks are visible on the Maeander plain, ‘Humei Tepe Harbour’, the facilities of which (warehouses, quay construc-
separated from the northeastern flanks of Humei Tepe by the stream tion etc.) have recently been detected with geophysical methods outside
of the southern “New” or “Large” Maeander which circles around the the fortification wall running along the foot of Humei Tepe (Tuttahs,
former Milesian peninsula since the 14th/15th century CE (Fig. 9; 2007: 356 ff.; Bumke and Tanrıöver, 2012: 76 Fig. 75; Brückner et al.,
Fig. 7 shows them as part of the Milesian archipelago). Same as the 2014a: 71 Fig. 16, 91 f.). Three geoarchaeological corings in the harbour
Milesian peninsula, they are built up of freshwater limestone (Nergiz basin (Mil 91–93) have shown good conditions for landing ships, which
Tepe formation) (Müllenhoff, 2005: 23; Tuttahs, 2007: 357 ff.). The lasted until at least early Byzantine times if not longer (Müllenhoff,
smaller northern one is c. 100 m in diameter and up to 12 m a.s.l., the 2005: 93; Tuttahs, 2007: 358 f. Fig. 388). The presence of a commercial
larger one c. 500 m to the south is c. 150 m in diameter and up to harbour at the eastern flank of Humei Tepe was confirmed in 2011
11 m a.s.l. On the topographical map of Wilski of 1906 they are named when H. Bumke, Halle, excavated the harbour gate, and when an inscrip-
Kütschük and Böjük Tschakmaklyk, equaling Küçük and Büyük Çakmaklık tional copy of a letter of the Roman emperor Hadrian dated to CE 131 was
in modern Turkish, meaning “small” and “large place with flint”. Carl discovered, allowing the foundation of a corporation (oikos) of Milesian
Humann, the author of the first accurate map of Miletos of 14.06.1891, shipowners (naukleroi) (Bumke and Tanrıöver, 2012: 78 Fig. 80;
i.e., before the German excavations started in 1899, has Kütschük and Ehrhardt and Günther, 2013; Brückner et al., 2014a: 92).
Böjük Tschakmak-tepe (“Small” and “Large Flint-Hill”) (Weber, 2007: The historian Arrian of Nicomedia (2nd century CE) mentions a re-
328 Fig. 1). On Kütschük Tschakmaklyk some modern graves are marked markable event taking place during the siege and conquest of Persian-
(Wilski, 1906: 21, map, quadrant 3E). They are indicated as medieval in dominated Miletos by Alexander the Great in 334 BCE (Arrian, Anabasis
the map of (Wulzinger et al., 1935: pl. 48; reprinted in: Weber, 2007: Alexandrou 1.19.4–6). The main sources Arrian's account relied on were
339–341 Fig. 11, “Friedhof”). Wulzinger et al. (1935: 60) remark the works of Alexander's companions Ptolemy and Aristobulus
that the graveyard is still in use in their time by nomadic Yürüks. (Strasburger, 1934: 24 f.; Brunt, 1976: XVIII–XXXIV; Bosworth, 1980:
According to them the same situation, medieval as well as present 16–34):
Turkish graves, was the case also on Tschakmak-tepe (Wulzinger et al. “Then, as the Milesians and mercenaries were hard pressed on all
leave out the designation Böjük, “large”; this island is not included in sides by the Macedonians, some threw themselves into the sea and
their map). inverting their shields paddled over to a little nameless island (nesida
Today, both hillocks, which lack a modern name, are occupied by anonymon) off the city (…). With the city now under control, Alexander
farm-houses (Fig. 9). The southern, former Böjük Tschakmaklyk, is now sailed in person against those who had fled to the islet, ordering ladders
integrated into the 3.53 m a.s.l. high dam protecting the plain to the to be brought to the bows of the triremes, so as to disembark from the
north from the winter floods of the “New” resp. “Large” Maeander ships on the cliffs (apotoma) of the island as if on a city wall. (…)
(Tuttahs, 2007: 446 with course of “Hochwasserdeiche”, flood protec- There were about 300 of these Greek mercenaries.“ (Anabasis
tion dams, in Fig. 479). Alexandrou 1.19.4–6; transl. Brunt, 1976).
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 887

Fig. 9. View from Humei Tepe towards the northeast on two former small islands, presumably Dromiskos and Perne in antiquity. One is visible to the right with houses and trees, the other one in
the middle part to the left with a house and a group of trees. Their modern names are Büyük (B) and Küçük (K) Çakmaklık. Between the two, where the Humei Tepe forms an embayment, the
‘Humei Tepe Harbour’ (HH) was located. Today this area is crossed by the “New” Maeander River which separates the two hillocks from Humei Tepe. Photo: H. Brückner, 26.09.2015.

Von Graeve (2000: 126 f.) identified the small “nameless” of alluvium overlie the marine strata (Kraft et al., 2000, 2005;
(anonymon) islet, where 300 mercenaries found shelter, with either Brückner, 2005). Fig. 10 is a synoptic view of the scenario of the delta
Küçük or Büyük Çakmaklık. Bosworth, in his commentary on Arrian, in- progradation.
geniously identified Arrian's nameless island with either Pliny's Among others, the former island of Syrie (Kuru Tepe; Figs. 10–11,
Dromiskos or Perne (Bosworth, 1980: 139). 13–14) and the two lakes at the northern flank of the graben Gebekirse
To sum up: Küçük and Büyük Çakmaklık both match Arrian's descrip- Gölü and Çatal Göl (Fig. 10; names according to the modern Turkish
tion of the nameless cliffy (apotoma) islet. As there were no other two map, Harita Genel Komutanlığı, 1996; on Schindler's map of 1906:
islands at such a close distance to Miletos which could easily be land- Göbek-Kilisse-Gjöl and Tschakal-Boğhaz-Gjöl) bear witness to the for-
locked, Küçük and Büyük Çakmaklık must be consistent with Pliny's mer marine embayment. The Küçük Menderes was meandering until
Dromiskos and Perne, though it seems impossible to decide which of the beginning of the 20th century CE. Due to hydraulic measures the
the two islands can be precisely attributed to one or the other. Arrian river is today regulated and an irrigation-drainage system is installed
delivers us as terminus post quem for the landlocking the end of the in the floodplain (Güldali, 1979). As for the neighbouring Büyük Mende-
4th century BCE. The terminus ante quem is given by Pliny's Naturalis res valley, the valley floor is nowadays used for agricultural purposes.
Historia (mid-1st century CE). The geomorphological setting of the outer delta plain, into which at
least two former islands have been integrated, is clearly visible in Fig. 12.
5. The Küçük Menderes and Ephesos – delta progradation and the
demise of Syrie 5.1. Historical accounts and sedimentary evidence

Since Neogene times, the W–E striking Küçük Menderes graben has In the already quoted passage of his encyclopaedic “Natural history”
formed; it is filled with Miocene to Quaternary deposits (Rojay et al., (Naturalis historia 2.201), Pliny the Elder (CE 23–79) adduces Ephesos
2005). The northern mountains are built up by various kinds of marble as an example for the “increase of the land … by what is brought
(dominant), phyllite, carbonaceous schist and paragneiss, while the down the rivers” or “by the receding of the sea”. Further on he writes
south and east of the study area as well as large parts of the Küçük Men- (5.115):
deres catchment mostly consist of easily erodible, deeply weathered “From these [rivers] comes a quantity of mud which advances the
mica schist (Philippson, 1912; Vetters, 1989; Brückner, 1997b; coastline and has now joined the island of Syrie on to the mainland by
Çakmakoğlu, 2007; Rantitsch and Prochaska, 2011). The ancient city of the flats interposed.” (transl. H. Rackham, 1969).
Ephesos is located at the southern flank of the graben, c. 6 km east of Thus, the siltation of the narrow strait between Syrie and the main-
the Aegean Sea. Same as for the Milesia, geoarchaeological research in land, which had already occurred during the latter part of the Classical
the Ephesia started in the 1990s (Brückner, 1997b, 2005; Kraft et al., period (Stock et al., 2014: 57), was still present in the collective memory
2000, 2001, 2005, 2007, 2011; Brückner et al., 2008; Stock et al., 2013, of the Roman era.
2014, 2015, 2016; Delile et al., 2015). To verify the historical accounts and reconstruct the siltation process
Due to the continued delta advance of the Küçük Menderes and its of the marine embayment, many percussion cores have been retrieved
tributaries since the 5th millennium BCE, the Ephesian gulf has nearly from the modern floodplain. The scenarios for the delta advance
completely been infilled with fluvial sediments, so that several metres (Figs. 10–11) are the result of many years of geoarchaeological research.
888 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

Fig. 10. Scenario of delta progradation for the Küçük Menderes (Kaystros) since the max. transgression 5000-7000 years before present. The former ‘Ephesian gulf’ hosted several islands;
Syrie (now called Kuru Tepe) is the most prominent. The positions of the shoreline are indicated for different time slices. Note the geomorphological change from the bird's foot delta of the
inner gulf to a beach ridge – swale dominated coast with a blocked delta when coastal longshore currents became dominant. Due to the continued delta advance the harbour of the
settlement had to be relocated several times.
Source: Brückner, 2005, modified.

Among others, a geological transect between Panayırdağ and Kuru Tepe presumably at a depth of 20 or so metres below present mean sea
(Syrie) had revealed the stratigraphic architecture of this geoarchive, level (b.s.l.). The typical stratigraphy shows slope debris, overlain by
especially the thickness of the marine facies, while 14C age estimates transgressive facies (beach deposits with abundant littoral to sublittoral
provided the chronological framework (Brückner, 1997b; see Fig. 11 macrofauna) and shallow marine silts. The transgression reached this
for location of drill cores). Meanwhile this information has been supple- area during the 6th millennium BCE (cf. Brückner, 1997b).
mented by further corings. A typical sediment sequence, representing In Eph 237, the base of the core at 9 m b.s.l. consists of sandy silts
the subsurface strata of the Menderes floodplain can be demonstrated (Fig. 15), deposited in a shallow marine environment (foraminifers:
by drill core Eph 237 close to the former Koressos harbour, the main Adelosina sp., Ammonia beccarii, A. parkinsoniana, Elphdium sp.; ostra-
harbour of the area until Hellenistic times (Figs. 10–11; Steskal, 2014, cods: Acathocythere sp., Ponticythere turbida, P. sp.). Typically brackish
331 Figs. 7–8; Stock et al., 2014, 37–38. 51. 55. 57 Fig. 8). species (foraminifer: Ammonia tepida; ostracod: Cyprideis torosa) are
In drill core Eph 237 (Fig. 15), located 200 m to the north of quite common, and it is likely that fluvial input already dominated the
Panayırdağ and the Roman Vedius gymnasium (cf. Fig. 11), the pre- area. This interpretation is supported by the high K/Ca ratio.
Holocene bedrock was not encountered, due to the steep fault zone The marine strata are overlain by coarser deposits, mostly sands of
bordering the Küçük Menderes graben to the south. Its position is marine to fluvial origin (8.77–2.89 m b.s.l.). The increasing influence
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 889

of the Küçük Menderes and its tributaries is apparent in the geochemis- 5.2. Islands of the Küçük Menderes graben
try as well as the microfauna (decline of marine species; they are absent
from ~5 m b.s.l. upwards). This coarsening upward sequence with the Syrie is definitely the most famous example of a former island in the
described faunal inventory can be correlated with many other cores Ephesian gulf. Nonetheless, there are more island candidates as evi-
from the delta plain. Geoarchaeological research attests to a rapid denced by a view on the topographic map: while it is questionable for
delta advance, especially since the 1st millennium BCE (Stock, 2015; Köprü Tepe, it is sure for Ada Tepe (Fig. 12) and an unnamed islet east
see also Eisma, 1978; Erinç, 1978; Meriç, 1985; Hess, 1989; Brückner, of the modern holiday village of Pamucak (see also Figs. 10–11). These
2005). names were adopted from Schindler's (1906) map; their ancient
A mollusc shell from the transition unit (2.89–2.52 m b.s.l.) dates to names are unknown to us. Ada Tepe (in Turkish: island hill) is still in
480–333 cal BCE. This layer represents a lagoonal environment, indicat- an amphibic environment.
ing the final siltation of the area. It confirms the results from the geolog- In the foundation myth of Ephesos, as related by the historian
ical transect between Panayırdağ and Kuru Tepe (Syrie) which dated Kreophylos in his “Annals of the Ephesians”, a small offshore island is
the siltation to Classical and Hellenistic times (details in Brückner, mentioned. Kreophylos' book, written around 400 BCE, is lost for the
1997b). most part. In the excerpt quoted by Athenaios (8, 361c–e [62] = FGrHist
The following light greyish clayey silts (2.52–1.40 m b.s.l.) with low 471 F 1) the name of the island is not given, albeit it is very likely to be
values of electrical conductivity and a lagoonal fauna (mollusc shell: identified with Syrie:
Cerastoderma glaucum; foraminifers: A. tepida, Haynesina germanica; “The people who were trying to found Ephesus had a great deal of
ostradod: C. torosa), indicate a freshening of the depositional environ- trouble, because they were unable to locate a site. Finally they sent to
ment from the 5th century BCE on. By then, the coastline had already the god's oracle and asked where they should put their city, and he
bypassed the coring site. The top of the sedimentary column consists prophesied to them that they should found a city in a place a fish
of floodplain deposits with abundant evidence of human influence. would show them and to which a wild boar would lead the way. […]
The 14C–age inversion indicates a reworking of the dated piece of wood. The Ephesians crossed over from the island where they had been living

Fig. 11. Scenario of the delta advance ~600 BCE and ~400 BCE. The reconstruction is mainly based on percussion corings. The coring sites of the geological transect from Panayırdağ via Kuru
Tepe to Cevasırdağ are indicated (details in Brückner, 1997b). Around 600 BCE, Syrie was still an island. After 400 BCE, Syrie became landlocked due to the progradation of the Küçük
Menderes delta. The island was transformed into a peninsula. A few centuries later it was totally integrated into the floodplain of the river. Then the “life cycle” of this estuarine island
was completed. Besides Syrie there was at least one more island in the marine gulf (on Schindler's map of 1906 called Ada Tepe). Based on Brückner, 2005 (modified and supplemented).
890 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

for 20 years, and settled Trecheia and the area around Coressus for a sec- No archaeological traces of the Early Iron Age have been found on
ond time […].” (transl. Olson, 2008). Kuru Tepe – the period when the so-called Ionian Migration is said to
It is controversially discussed, if this foundation legend contains an have taken place according to the ancient tradition. This ‘argument
authentic nucleus embedded in a mythical narrative (Kerschner, from silence’, however, cannot be used to contradict the narrative that
2006), or if it is in its entirety a later retrospective construction of the the Greek migrants settled first on the island of Syrie. On the one
past (Bintliff, 2006: 111). A very similar pattern of Greek emigrants – hand, no intensive survey nor systematic excavation has yet been car-
settling first on an offshore island for a short period before founding a ried out on this hill; on the other hand, a temporary settling of the
permanent settlement on the opposite coast under the guidance of an Early Iron Age would have left behind only scarce residues. If this settle-
oracle – appears in the narrative on the foundation of Cyrene in north- ment were situated at the shore of that period, it would have been bur-
eastern Libya in the second half of the 7th century BCE. Herodotus (4, ied meanwhile under several metres of alluvia of the Kaystros river, thus
150–158) relates that Theraens settled in a first, unsuccessful attempt being invisible in a modern survey.
on the small island of Plateia off the Libyan coast, before changing to If the narrative of an initial settling on Syrie constitutes an authentic
the mainland. nucleus within the foundation myth, the water supply must have
The ancient literary sources suggest that Syrie was located close to caused a major problem for the Greeks living on the island. Nowadays,
the city in the inner part of the now silted-up gulf of Ephesos. In this no spring exists on the limestone hill, a fact also expressed by the mod-
area, there is only one feature which can be interpreted as a former is- ern Turkish name “Kuru Tepe” meaning “dry hill” (cf. Scherrer, 2007:
land: the hill of Kuru Tepe (Figs. 10–11, 13–14). This identification 30; Meriç, 2009: 43; the name “Korudaǧ Tepe”, alternatively used on
dates back to the 19th century (Guhl, 1843, map; Bürchner, 1905b: modern maps, meaning “wooded hill”, is not consistent with the con-
2779–2780; Benndorf, 1906; Schindler, 1906; Honigmann, 1932); it temporary vegetation of macchia; cf. Figs. 13–14). In contrast, D.
has never been seriously contested. Kuru Tepe is a hill of elongated Crouch (2003: 235) assumed: “Kuru tepe […] lies in the path of this
shape, with two summits of 86 and 89 m a.s.l. and an extension of aquifer. It is a horst, part of the natural system for storing water in the
1.6 km in its main axis (SW–NE). It is situated at a distance of c. valley.” W. Prochaska (pers. comm.), however, considers the existence
1.7 km northwest of Ayasuluk hill, where the settlement was located of springs on Kuru Tepe highly improbable, as the former island was
during the Late Bronze Age and Early Iron Age. too far away from the mainland to produce sufficient pressure for an
The earliest archaeological traces from Kuru Tepe date back to the artesian spring. The apparent lack of a local water source on the former
Archaic period. Scherrer (2007: 330 with n. 47) reports surface finds island leads to the conclusion that it is quite unlikely that people had ac-
of Archaic and Classical pottery from an extensive survey in 1997. At tually settled there for several years, as stated in the foundation myth
the southern tip of the hill there was a tumulus with a polygonal grave according to Kreophylos.
chamber, and a dromos built in ashlar masonry. The latter was
unearthed in a rescue excavation by the Efes Müzesi in 1985; it dates 6. Discussion
to the late 6th/1st half of the 5th century BCE (unpublished, mentioned
by Bammer, 1988: 17 Fig. 14; Meriç, 2009: 43; Mohr, 2015: 320). A The Küçük and Büyük Menderes rivers reveal the typical delta ad-
watchtower, presumably of Hellenistic age, is reported by Meriç vance which can also be observed in other major river systems. Two
(2009: 43). phases of delta evolution can be distinguished (Anthony et al., 2014):

Fig. 12. An isolated hill, once an island of the former Ephesian gulf. The ancient name is unknown to us. On Schindler's famous map of 1906, the hill is identified as Ada Tepe (AT), “Island
Hill”. Meanwhile, it was “captured” by the Küçük Menderes river, which integrated the island into its floodplain. The outer delta plain, in parts still amphibic, shows a distinct ridge-and-
swale topography. Note the delayed river mouth due to the strong southbound coastal current, which is typical for a blocked delta. Nowadays the millennia-long delta advance is
terminated; over the last few decades a landward shift in the shoreline has taken place, resulting in coastal erosion. This has been caused by ongoing sea-level rise associated with
global warming, and limited fluvial sediment supply due to hydraulic measures (irrigation of fields, building of dams). View towards the north. Oblique aerial photograph from a
power glider, H. Brückner, 13.09.2011.
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 891

Fig. 13. View from Bülbül Dağ looking north on the former island of Syrie (S), now called Kuru Tepe. Clearly visible are prominent buildings and streets of the Roman city of Ephesos, such as
the straight street Arkadiane which connected the theatre with the Roman harbour (to the left, partly filled with water). A Artemision near the modern city of Selçuk; AH Ayasuluk Hill
with the medieval castle; KT Köprü Tepe; ÇG Çatal Göl, a remnant of the former marine embayment. Oblique aerial photograph from a power glider, H. Brückner, 13.09.2011.

With the sea level stagnation from Mid-Holocene times on (Stanley and Guillén, 1998) built out their deltas, strongly impacted by human
Warne, 1994; Kayan, 1999), the river systems around the Mediterra- influence.
nean Sea slowly started to prograde their deltas. Since the Bronze age, Other than Grove and Rackham (2001) who strongly emphasized
especially during the last 2500 years, scientists have proven a faster the climatic influence, most of the researchers attribute the landscape
delta advance, often along with a stronger human impact (deforestation changes in the Mediterranean to human impact, in many cases even
and subsequent erosion), and thus a larger sediment supply (see e.g. since Neolithic times. This is evident from a compilation by Brückner
Arnaud-Fassetta, 2002; Zolitschka et al., 2003; Bellotti et al., 2011; (1986) and has been confirmed by several other authors in more re-
Maselli and Trincardi, 2013). E.g., the Rhône (Vella et al., 2005), Po gional studies, for examples lately by Psomiadis et al. (2013) in the
(Machetti, 2002), Tiber (Bellotti et al., 2011) and Ebro (Palanques and Thessaloniki plain (Greece).

Fig. 14. View from Ayasuluk hill on Kuru Tepe, the former island of Syrie (in the centre of the background). To the left in the foreground a tower of the medieval fortification on Ayasuluk
hill. Photo: M. Kerschner, 30.08.2012.
892 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

Fig. 15. Percussion core Eph 237 with sedimentological, geochemical and microfaunal data, as well as the interpretation of the sedimentary environment. For the location of the drill core
see Fig. 11. Source: new data.

Human impact may also have the effect of a delta retreat. The Nile islands. The most prominent of them are Hybanda, Lade, Asteria and
delta shows an advance until 4000 BP, followed by a decline of sediment Nergiz Tepe in the Büyük Menderes (Maiandros, Maeander) graben,
supply due to drainage of the different river channels (Marriner et al., and Syrie in the Küçük Menderes (Kaystros) graben. Even the famous
2012). Since several decades, most of the deltas worldwide suffer Milesian peninsula evolved from an archipelago; it started with two
from erosion due to a deficit in sediment supply and the ongoing sea- main islands that were later complemented by two smaller ones,
level rise, wherefore the Küçük Menderes delta is a typical example which we identify as the islets Dromiskos and Perne. Even today, several
(cf. Fig. 12 with its caption). residual lakes attest to their former marine past: Lake Bafa and Lake
Azap are remnants of the Ikarian/Karian gulf, and Gebekirse Gölü and
7. Conclusion Çatal Göl of the Ephesian gulf.
Since mid-Holocene times, the Büyük and Küçük Menderes rivers
The early Holocene sea-level rise created many marine gulfs. Where have prograded their deltas and silted up these marine embayments,
the sea transgressed into river mouths, ria-type coasts evolved. Former- despite the continued rise in sea level and occasional co-seismic subsi-
ly dry land drowned and, depending on a favourable topographic set- dence of the grabens. One of the main factors for the increased siltation
ting, islands were born (“estuarine islands”). When the sea-level rise since Roman times was deforestation as a trigger for erosion processes
decelerated seven millennia ago, the rivers started to build out their in the drainage basins; this resulted in a high sediment load and in ac-
deltas and infill the embayments. Thus, many of these estuarine islands celerated delta advance. In principal, these erosion/accumulation pro-
became landlocked, were transformed into peninsulas and finally inte- cesses were already understood in ancient times, as demonstrated by
grated into the floodplain. To describe this geo-metamorphosis we statements of the historian Thucydides (5th century BCE), the geogra-
coin the term: “life cycle of an island”. pher Strabo (1st century BCE – early 1st century CE), the natural histo-
This article demonstrates many examples for these profound land- rian Pliny (1st century CE), and the travel writer Pausanias (2nd century
scape changes from western Anatolia. It focuses on the identification CE).
and the fate of former islands and demonstrates the added value gener- Our research results demonstrate the importance of the
ated by interpreting ancient sources in a synoptic view that links ancient interplay between social sciences on the one hand and natural
history and archaeology with geosciences. sciences on the other. Historical records should therefore be used
Our long-term geoarchaeological research of the Büyük and Küçük more intensively, and, wherever possible, be verified by geological
Menderes grabens has proven the landlocking of former estuarine records.
 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894 893

Acknowledgements Brückner, H., Kraft, J.C., Kayan, İ., 2008. Vom Meer umspült, vom Fluss begraben –zur
Paläogeographie des Artemisions. In: Muss, U. (Ed.), Die Archäologie der ephesischen
Artemis. Gestalt und Ritual eines Heiligtums. Phoibos, Wien, pp. 21–31.
The research in the Maeander plain and Miletos was rendered possi- Brückner, H., Herda, A., Müllenhoff, M., Rabbel, W., Stümpel, H., 2014a. On the lion and
ble through generous funding by several institutions, namely the other harbours in Miletos: historical, archaeological, sedimentological, and geophys-
ical investigations. Proc. Danish Inst. Athens 7, 49–103.
German Research Council (DFG, Bonn), the Gerda Henkel Stiftung Brückner, H., Herda, A., Müllenhoff, M., Rabbel, W., Stümpel, H., 2014b. Der Löwenhafen
(Düsseldorf, Germany), and the Center for Hellenic Studies (Trustees Von Milet – Eine geoarchäologische Fallstudie. In: Ladstätter, S., Pirson, F., Schmidts,
of Harvard Foundation, Washington D.C.). The fieldwork in Miletos T. (Eds.), Harbors and Harbor Cities in the Eastern Mediterranean from Antiquity to
ByzantiumRecent Discoveries and New Approaches, International Congress Istanbul
and the Milesia was supported by the Miletos Excavation, namely its 30.05. −01.06.2011, Byzas 19. Ege Yayınları, Istanbul, pp. 773–806.
long-term director Volkmar von Graeve and its new director Philipp Brunt, P.A., 1976. Arrian. With an English Translation in two Volumes. Anabasis Alexandri,
Niewöhner, the Priene Excavation, and the staff of the Miletos Museum. Books I–IV vol. I. Harvard University Press, Cambridge, MA.
Bumke, H., Tanrıöver, A., 2012. Milet 2011 – Neue Forschungen am Humeitepe im Norden
The research in Ephesos and the Ephesia was facilitated by the long-
der Stadt. In: Ruhr-Universität Bochum (Ed.), Jahresberichte des Instituts für
lasting financial and logistical support by the Austrian Archaeological Archäologische Wissenschaften für das Akademische Jahr 2010–2011, pp. 75–78
Institute (ÖAI, Vienna) and the Ephesos Excavation, namely its director (Bochum).
Sabine Ladstätter. Our fieldwork has been backed by the Ephesos Bürchner, L., 1905a. Dromiskos. Paulys Realencyclopädie der classischen
Altertumswissenschaft vol. V 2. Metzler, Stuttgart, col. 1715.
Museum (Efes Müzesi) at Selçuk, in particular its director Cengiz Topal. Bürchner, L., 1905b. Ephesos. Paulys Realencyclopädie der classischen
The T.C. Kültür ve Turizm Bakanlıǧı, Ankara, kindly granted the research Altertumswissenschaft vol. V 2. Metzler, Stuttgart, col. 2773–2822.
permits. Anna Pint determined the microfauna. Nick Marriner is Bürchner, L., 1924. Lade. Paulys Realencyclopädie der classischen Altertumswissenschaft
vol. XII 1. Metzler, Stuttgart, col. 381.
thanked for language editing. Çakmakoğlu, A., 2007. Pre-Neogene tectonostratigraphy of Dilek peninsula and the area
surrounding Söke and Selçuk. Miner. Res. Explor. Bull. 135, 1–17.
References Chandler, R., 1775. Travels in Asia Minor: or, an Account of a Tour Made at the Expense of
the Society of Dilettanti. Marchbank, Dublin.
Anthony, E.J., Marriner, N., Morhange, C., 2014. Human influence and the changing geo- Chantraine, P., 1968–1980. Dictionnaire étymologique de la langue grecque. Histoire des
morphology of Mediterranean deltas and coasts over the last 6000 years: from mots. Librairie Klincksieck, Paris.
progradation to destruction phase? Earth Sci. Rev. 139, 336–361. Crouch, D.P., 2003. Geology and settlement. Greco-Roman Patterns. Oxford University
Arnaud-Fassetta, G., 2002. Geomorphological records of a ‘flood-dominated regime’ in the Press, Oxford, New York.
Rhône Delta (France) between the 1st century BCE and the 2nd century CE. What Delile, H., Blichert-Toft, J., Goiran, J.P., Stock, F., Arnaud-Godet, F., Bravard, J.P., Brückner,
correlations with the catchment paleohydrology? Geodin. Acta 15, 79–92. H., Albarède, F., 2015. Demise of a harbor: a geochemical chronicle from Ephesus.
Bammer, A., 1988. Ephesos. Stadt an Fluß und Meer. Adeva Reisebegleiter. Akademische J. Archaeol. Sci. 53, 202–213.
Druck- und Verlagsanstalt, Graz. Ehrhardt, N., Günther, W., 2013. Hadrian, Milet und die Korporation der milesischen
Bay, B., 1999. Geoarchäologie, anthropogene Bodenerosion und Deltavorbau im Büyük Schiffseigner. Zu einem neugefundenen kaiserlichen Schreiben. Chiron 43, 199–220.
Menderes Delta (SW-Türkei). Dissertation. Universität Bochum. Eisma, D., 1978. Stream deposition and erosion by the eastern shore of the Aegean. In:
Bellotti, P., Calderoni, G., Di Rita, F., D'Orefice, M., D'Amico, C., Esu, D., Magri, D., Preite Brice, W.C. (Ed.), The Environmental History of the Near and the Middle East Since
Martinez, M., Tortora, P., Valeri, P., 2011. The Tiber river delta plain (central Italy): the Last ice age. Academic Press, London, pp. 67–81.
coastal evolution and implications for the ancient Ostia Roman settlement. The Holo- Erinç, S., 1978. Changes in the physical environment in Turkey since the end of the last
cene 21, 1105–1116. glacial. In: Brice, W.C. (Ed.), The Environmental History of the Near and Middle East
Bendt, W., 1968. Topographische Karte von Milet (1:2000) mit erläuterndem Text. Milet II Since the Last ice age. Academic Press, London, pp. 87–110.
4. Walter de Gruyter, Berlin. Godley, A.D., 1975. Herodotus. With an English Translation in 4 Volumes. Harvard Univer-
Benndorf, O., 1906. Zur Ortskunde und Stadtgeschichte. In: Benndorf, O. (Ed.), sity Press, Princeton, MA.
Forschungen in Ephesos I, Hölder. Vienna, pp. 9–110. Grove, A.T., Rackham, O., 2001. The Nature of Mediterranean Europe. An Ecological Histo-
Bintliff, J., 2006. Multi-ethnictiy and population movement in ancient Greece: alternatives to ry. Yale University Press, New Haven.
a world of ‘Red-Figure’ people. In: Olshausen, E., Sonnabend, H. (Eds.), “Troianer sind Guhl, E., 1843. Ephesiaca. Libraria Friderici Nicolai, Berlin.
wir gewesen” — Migrationen in der antiken Welt. Stuttgarter Kolloquium zur Güldali N. 1979. Geomorphologie der Türkei: Erläuterungen zur geomorphologischen
Historischen Geographie des Altertums 8, 2002. Geographica Historica 21, pp. 108–114. Übersichtskarte der Türkei 1:2000000. Beihefte zum Tübinger Atlas des Vorderen
Boardman, J., 2002. The Archaeology of Nostalgia. How the Greeks Re-Created Their Orients. Reichert, Wiesbaden.
Mythical Past. Thames and Hudson, London. Harita Genel Komutanlığı. 1996. Aydın, K - 816, M18 - b4, 1:25,000.
Bosworth, A.B., 1980. A Historical Commentary on Arrian's History of Alexander. Com- Herda, A., 2005. Apollon Delphinios, das Prytaneion und die Agora von Milet. Neue
mentary on Books I-III vol. 1. Clarendon Press, Oxford. Forschungen. Archäologischer Anzeiger 2005 (1), 243–294.
Brückner, H., 1986. Man's impact on the evolution of the physical environment in the Herda, A., 2008. Apollon Delphinios – Apollon Didymeus: Zwei Gesichter eines
Mediterranean region in historical times. GeoJournal 13 (1), 7–17. milesischen Gottes und ihr Bezug zur Kolonisation Milets in archaischer Zeit. In:
Brückner, H., 1995. Geomorphologie und Paläo-Environment der Milesia. Milet 1992– Bol, R., Höckmann, U., Schollmeyer, P. (Eds.), Kult(ur)kontakte. Apollo in Myus,
1993. Vorbericht über die Grabungsarbeiten und Geländeerkundungen, die Milet/Didyma, Histria, Myus, Naukratis und auf Zypern, Akten des Table Ronde in
Denkmälerrestaurierung und die naturwissenschaftlichen Begleitprogramme der Mainz vom 11.–12. März 2004. Marie Leidorf, Rhaden/Westfalen, pp. 13–86.
Miletgrabung in den Jahren 1992 und 1993, von Graeve, V. Archäologischer Anzeiger, Herda, A., 2009. Karkiša-Karien und die sog. Ionische Migration. In: Rumscheid, F. (Ed.),
pp. 195–333 (esp. 329–330). Die Karer und die Anderen, Internationales Kolloquium an der Freien Universität Ber-
Brückner, H., 1996. Geoarchäologie an der türkischen Ägäisküste. Landschaftswandel im lin, 13.-15. Oktober 2005. Habelt, Bonn, pp. 27–108.
Spiegel geologischer und archäologischer Zeugnisse. Geogr. Rundsch. 10, 568–574. Herda, A., 2011. How to run a state cult: the organisation of the cult of Apollo Delphinios
Brückner, H., 1997a. Coastal Changes in Western Turkey – Rapid Delta Progradation in in Miletus. In: Haysom, M., Wallensten, J. (Eds.), Current Approaches to Ancient
Historical Times. In: Briand, F., Maldonado, A. (Eds.), Transformations and Evolution Greek Religion, Papers Presented at a Symposium at the Swedish Institute at Athens,
of the Mediterranean Coastline. CIESM Science Series 3, Bulletin de l'Institut 17–19 April 2008. Swedish Institute, Stockholm, pp. 57–93.
océanographique, numéro spécial 18. Musée océanographique, Monaco, pp. 63–74. Herda, A., 2013a. Maeander. In: Bagnall, R.S., Brodersen, K., Champion, C.B., Erskine, A.,
Brückner, H., 1997b. Geoarchäologische Forschungen in der Westtürkei — das Beispiel Huebner, S.R. (Eds.), The Encyclopedia of Ancient History. Blackwell, Chichester,
Ephesus. Passauer Schriften zur Geographie 15, 39–51. pp. 4214–4215.
Brückner, H., 1998. Coastal research and geoarchaeology in the Mediterranean region. In: Herda, A., 2013b. Burying a sage: the Heroon of Thales on the Agora of Miletos. With re-
Kelletat, D.H. (Ed.), German Geographical Coastal Research — The Last Decade, marks on some other excavated Heroa and on cults and graves of the mythical foun-
pp. 235–258 (Tübingen). ders of the city. In: Henry, O. (Ed.), Le mort dans la ville. Pratiques, contextes et
Brückner, H., 2003. Delta evolution and culture − aspects of geoarchaeological research in impacts de inhumation intro-muros en Anatolie, du début de l'’Âge du Bronze à
Miletos and Priene. In: Wagner, G.A., Pernicka, E., Uerpmann, H.P. (Eds.), Troia and l'époque romaine, 2èmes recontres d'archéologie, IFÉA-Ege Yayınları, Istanbul,
the Troad. Scientific Approaches. Springer, Berlin, Heidelberg, New York, pp. 121–144. 14–15 Novembre 2011. Zero Prod. Ltd., Istanbul, pp. 67–122.
Brückner, H., 2005. Holocene shoreline displacements and their consequences for human Herda, A., Müllenhoff, M., Knipping, M., Brückner, H., 2017. From the Gulf of Latmos to
societies: the example of Ephesus in western Turkey. Z. Geomorphol. N. F. Suppl. 137, Lake Bafa. New historical, geoarchaeological and palynological insights into the
11–22. anthropogeography of the Maeander Delta region (Turkey). Hesperia 86 (in press).
Brückner, H., 2011. Geoarchäologie — In Forschung und Lehre. Tagungen des Herrmann, P., 2006. N. 1131 Postume Ehreninschrift für C. Iulius Epikrates. In: Herrmann,
Landesmuseums für Vorgeschichte Halle (Saale) 6, 9–20. P., Günther, W., Ehrhardt, N. (Eds.), Inschriften von Milet, Teil 3: Inschriften n.
Brückner, H., Müllenhoff, M., Handl, M., van der Borg, K., 2002. Holocene landscape evo- 1020–1580, Milet VI 3. Walter de Gruyter, Berlin – New York, pp. 82–85.
lution of the Büyük Menderes alluvial plain in the environs of Myous and Priene Hess, G., 1989. Die Entwicklung des Küçük-Menderes-Deltas in historischer Zeit. Essener
(western Anatolia, Turkey). Z. Geomorphol. N. F. Suppl. 127, 47–65. Geographische Schriften 17, 203–215.
Brückner, H., Müllenhoff, M., Gehrels, R., Herda, A., Knipping, M., Vött, A., 2006. From ar- Honigmann, E., 1932. Syrie. Paulys Realencyclopädie der classischen Altertumswissenschaft
chipelago to floodplain — geographical and ecological changes in Miletus and its en- vol. IV A 2. Metzler, Stuttgart, col. 1776.
virons during the last six millennia (western Anatolia, Turkey). Z. Geomorphol. N.F. Jones, W.H.S., Ormerod, H.A., 1918. Pausanias Description of Greece With an English
Suppl. 142, 63–83. Translation in 4 Volumes. Harvard University Press, Cambridge, MA.
894 H. Brückner et al. / Journal of Archaeological Science: Reports 12 (2017) 876–894

Kawerau, G., Rehm, A., 1914. Das Delphinion in Milet. Milet I 3. Georg Reimer, Berlin. Sakellariou, D., Galanidou, N., 2016. Pleistocene submerged landscapes and Palaeolithic
Kayan, I., 1999. Holocene stratigraphy and geomorphological evolution of the Aegean archaeology in the tectonically active Aegean region. In: Harff, J., Bailey, G., Lüth, F.
coastal plains of Anatolia. Quat. Sci. Rev. 18, 541–548. (Eds.), Geology and Archaeology: Submerged Landscapes of the Continental Shelf.
Kerschner, M., 2006. Die Ionische Wanderung im Lichte neuer archäologischer Geological Society, London, Special Publications 411, pp. 145–178.
Forschungen in Ephesos. In: Olshausen, E., Sonnabend, H. (Eds.), “Troianer sind wir Scherrer, P., 2007. Von Apaša nach Hagios Theologos. Die Siedungsgeschichte des Raumes
gewesen” – Migrationen in der antiken Welt. Stuttgarter Kolloquium zur Ephesos von prähistorischer bis in byzantinische Zeit unter dem Aspekt der
Historischen Geographie des Altertums 8, 2002. Geographica Historica 21, maritimen und fluvialen Bedingungen. Jahreshefte des Österreichischen
pp. 364–382. Archäologischen Institutes 76, pp. 321–351.
Kirk, G.S., Raven, J.E., Schofield, M., 1994. Die vorsokratischen Philosophen. Einführung, Schindler, A., 1906. Bemerkungen zur Karte. In: Benndorf, O. (Ed.), Forschungen in
Texte und Kommentare. Ins Deutsche übersetzt von Karlheinz Hülser. J.B. Metzler, Ephesos I. Vienna. Hölder, 235–236 with map.
Stuttgart – Weimar. Schröder, B., 1998. Mittel- bis jungholozäne Landschaftsgeschichte am Unterlauf des
Klug, H., 1985. Eine geographische Klassifikation der Inseltypen des Weltmeeres. Berliner Großen Mäander (W.-Anatolien). GeoArchaeoRhein 2, 91–101.
Geographische Studien 16, 191–216 (Berlin). Schröder, B., Brückner, H., Stümpel, H., Yalçın, Ü., 1995. Geowissenschaftliche
Kraft, J.C., Kayan, İ., Brückner, H., Rapp, G., 2000. A geological analysis of ancient land- Umfelderkundung. In: von Graeve, V. (Ed.), Milet 1992–1993. Vorbericht über die
scapes and the harbors of Ephesus and the Artemision in Anatolia. Österreichische Grabungsarbeiten und Geländeerkundungen, die Denkmälerrestaurierung und die
Jahreshefte 69, 175–233. naturwissenschaftlichen Begleitprogramme der Miletgrabung in den Jahren 1992
Kraft, J.C., Kayan, İ., Brückner, H., 2001. The geological and paleogeographical environs of und 1993. Archäologischer Anzeiger, pp. 195–333 (esp. 238–244).
the Artemision. In: Muss, U. (Ed.), Der Kosmos der Artemis von Ephesus. Stanley, D.J., Warne, A.G., 1994. Worldwide initiation of Holocene marine deltas by decel-
Österreichisches Archäologisches Institut, Wien. Sonderschriften 37, pp. 123–133. eration of sea-level rise. Science 265, 228–231.
Kraft, J.C., Brückner, H., Kayan, İ., 2005. The sea under the city of ancient Ephesus. In: Steskal, M., 2014. Ephesos and its Harbors: A City in Search of its Place. In: Ladstätter, S.,
Brandt, B., Gassner, V., Ladstätter, S. (Eds.), Synergia. Festschrift Friedrich Krinzinger Pirson, F., Schmidts, T. (Eds.), Harbors and Harbor Cities in the Eastern Mediterranean
1. Phoibos, Wien, pp. 147–156. From Antiquity to ByzantiumRecent Discoveries and New Approaches, International
Kraft, J.C., Brückner, H., Kayan, İ., Engelmann, H., 2007. The geographies of ancient Ephe- Congress Istanbul 30.05.−01.06.2011, Byzas 19, 773–806. Ege Yayınları, Istanbul,
sus and the Artemision in Anatolia. Geoarchaeology 22 (1), 121–149. pp. 325–338.
Kraft, J.C., Rapp, G., Brückner, H., Kayan, İ., 2011. Results of the struggle at ancient Ephe- Stock, F., 2015. Ephesos and the Ephesia — Geoarchaeological and Palaeogeographical Re-
sus: natural processes 1, human intervention 0. Geol. Soc. Lond. Spec. Publ. 35, 27–36. search About a Famous City in Western Anatolia. Dissertation. University of Cologne
Liddell, H.G., Scott, R., Jones, S.T., 1996. Greek-English Lexicon. Clarendon Press, Oxford. (http://kups.ub.uni-koeln.de/6368/).
Lohmann, H., 2006. Zur historischen Topographie des südlichen Ionien. Orbis Terrarum 8 Stock, F., Pint, A., Horejs, B., Ladstätter, S., Brückner, H., 2013. In search of the harbours:
(2002), 163–272. new evidence of late Roman and byzantine harbours of Ephesus. Quat. Int. 312,
Machetti, M., 2002. Environmental changes in the central Po plain (northern Italy) due to 57–69.
fluvial modifications and anthropogenic activities. Geomorphology 44, 361–373. Stock, F., Kerschner, M., Kraft, J.C., Pint, A., Frenzel, P., Brückner, H., 2014. The palaeo-
Marriner, N., Flaux, C., Kaniewski, D., Morhange, C., Leduc, G., Moron, V., Chen, C., Gasse, F., geographies of Ephesos (Turkey), its harbours and the Artemision — a
Empereur, J.-Y., Stanley, J.-D., 2012. ITCZ and ENSO-like pacing of Nile delta geoarchaeological reconstruction for the timespan 1500–300 BCE. Z. Geomorphol.
hydrogeomorphology during the Holocene. Quaternary Science Reviews 45, 73–84. N. F. 58 (Suppl. Issue 2), 33–66.
Maselli, V., Trincardi, F., 2013. Man made deltas. Sci. Rep. 3, 1926. Stock, F., Ehlers, L., Horejs, B., Knipping, M., Ladstätter, S., Seren, S., Brückner, H., 2015.
Mayor, A., 2000. The First Fossil Hunters: Paleontology in Greek and Roman Times. Neolithic settlement sites in western Turkey — palaeogeographic studies at Çukuriçi
Princeton University Press, Princeton, N.J. Höyük and Arvalya Höyük. J. Archaeol. Sci. Rep. 4, 565–577.
Meriç, R., 1985. Zur Lage des ephesischen Außenhafens Panormos. In: Komitee Stock, F., Knipping, M., Pint, A., Ladstätter, S., Delile, H., Heiss, A.G., Laermanns, H., Mitchell,
Festschrift Hermann Vetters (Ed.), Lebendige AltertumswissenschaftenFestgabe zur P., Ployer, R., Steskal, M., Thanheiser, U., Urz, R., Wennrich, V., Brückner, H., 2016.
Vollendung des 70. Lebensjahres von Hermann Vetters. Adolf Holzhausen Nfg., Human impact on Holocene sediment dynamics in the eastern Mediterranean —
Wien, pp. 33–37. the example of the Roman harbour of Ephesus. Earth Surf. Process. Landf. 41,
Meriç, R., 2009. Das Hinterland von Ephesos. Archäologisch-topographische Forschungen 980–996.
im Kaystros-Tal. Ergänzungshefte zu den Jahresheften des Österreichischen Strasburger, H., 1934. Ptolemaios und Alexander. Dieterich'sche Verlagsbuchhandlung,
Archäologischen Institutes 12 (Wien). Leipzig.
Mohr, E.-M., 2015. Eisenzeitliche Nekropolen im westlichen Kleinasien. Struktur und Thonemann, P., 2011. The Maeander Valley: A Historical Geography From Antiquity to By-
Entwicklung zwischen dem 9. und 6. Jh. v. Chr. Byzas 21. Ege Yayınları, İstanbul. zantium. University Press, Cambridge.
Müllenhoff, M., 2005. Geoarchäologische, sedimentologische und morphodynamische Tomaschek, W., 1891. Zur historischen Geographie von Kleinasien im Mittelalter. Die
Untersuchungen im Mündungsgebiet des Büyük Menderes (Mäander), Westtürkei. Küstengebiete und die Wege der Kreuzfahrer, Sitzungsberichte der Kaiserlichen
Marburger Geographische Schriften 141 (Marburg/Lahn). Akademie der Wissenschaften zu Wien, Philosophisch-Historische Classe, 124 Vol.
Müllenhoff, M., Herda, A., Brückner, H., 2009. Geoarchaeology in the City of Thales. 1. F. Tempsky, Wien.
Deciphering palaeogeographic changes in the Agora area of Miletus. In: Mattern, T., Tuttahs, G., 2007. Milet und das Wasser – ein Leben in Wohlstand und Not in Antike,
Vött, A. (Eds.), Mensch und Umwelt im Spiegel der Zeit. Aspekte geoarchäologischer Mittelalter und Gegenwart. Schriften der Deutschen Wasserhistorischen Gesellschaft,
Forschungen im östlichen Mittelmeergebiet. Philippika. Marburger Altertumskundliche Sonderband 5. Print Weilburg, Siegburg.
Abhandlungen 1, 97–110 (Wiesbaden). Vella, C., Fleury, T.J., Raccasi, G., Provansal, M., Sabatier, F., Bourcier, M., 2005. Evolution of
Niemeier, W.-D., 2007. Milet –von den Anfängen menschlicher Besiedlung bis zur the Rhône delta plain in the Holocene. Mar. Geol. 222–223, 235–265.
Ionischen Wanderung. In: Cobet, J., von Graeve, V., Niemeier, W.-D., Zimmermann, Vetters, W., 1989. Geologische Übersichtskarte der Umgebung von Ephesus 1 col. 25.000.
K. (Eds.), Frühes Ionien. Eine Bestandsaufnahme, Panionion-Symposion Güzelçamlı von Graeve, V., 2000. Die Belagerung Milets durch Alexander den Grossen. In: Avram, A.,
26. September — 1. Oktober 1999. Milesische Forschungen 5, 3–20 (Mainz). Babeş, M. (Eds.), Civilisation grecque et cultures antiques périphériques. Hommage à
Olson, S.D., 2008. Athenaeus. The Learned Banqueters. Books 8–10.420 E. The Loeb Classi- Petre Alexandrescu à son 70e anniversaire. Editura Enciclopedică, Bucarest,
cal Library 235. Harvard University Press, Cambridge (Mass.), London. pp. 113–130.
Palanques, A., Guillén, J., 1998. Coastal changes in the Ebro delta: natural and human fac- Vött, A., Schriever, A., Handl, M., Brückner, H., 2007. Holocene palaeogeographies of the
tors. J. Coast. Conserv. 4, 16–26. central Acheloos River delta (NW Greece) in the vicinity of the ancient seaport
Peltier, W.R., Fairbanks, R.G., 2006. Global glacial ice volume and last glacial maximum du- Oiniadai. Geodin. Acta 20 (4), 241–256.
ration from an extended Barbados sea level record. Quat. Sci. Rev. 25, 3322–3337. Waelbroeck, C., Labeyrie, L., Michel, E., Duplessy, J.-C., McManus, J., Lambeck, K., Balbon, E.,
Philippson, A., 1912. Reisen und Forschungen im westlichen Kleinasien. Petermanns Labracherie, M., 2002. Sea-level and deep water temperature changes derived from
geographische Mitteilungen, Ergänzungsheft 172. benthic foraminifera isotopic records. Quat. Sci. Rev. 21, 295–305.
Philippson, A., 1936. Das südliche Ionien, Milet III 5. Walter de Gruyter, Berlin. Weber, B., 2007. Der Stadtplan von Milet. Einhundert Jahre Stadtforschung. In: Cobet, J.,
Psomiadis, D., Ghilardi, M., Demory, F., Sabatier, D., Bloemendal, J., Yiu, C., 2013. Late Pleis- von Graeve, V., Niemeier, W.-D., Zimmermann, K. (Eds.), Frühes Ionien. Eine
tocene to mid-Holocene landscape reconstruction in the western part of the Bestandsaufnahme, Panionion-Symposion Güzelçamlı 26. September — 1. Oktober
Thessaloniki plain, Greece: evidence for environmental changes, and implications 1999. Milesische Forschungen 5, 327–362 (Mainz).
for human occupation. Z. Geomorphol. 58 (Suppl. Iss. 2), 67–87. Wernicke, P., 1896. Asterion 10. Paulys Realencyclopädie der classischen
Rackham, H., 1967. Pliny. Natural History in ten Volumes. Vol. I: Praefatio, Libri I, II. The Altertumswissenschaft II 2. Metzler, Stuttgart, col. 1785.
Loeb Classical Library 330. Heinemann, London. Wiegand, T., 1929. Die milesische Landschaft, Milet II 2. Verlag Hans Schoetz, Berlin.
Rackham, H., 1969. Pliny. Natural History in ten Volumes. Vol. II: Libri III-VII. The Loeb Wilski, P., 1906. Karte der milesischen Halbinsel (1:50000) mit erläuterndem Text, Milet I
Classical Library 352. Heinemann, London. 1. Georg Reimer, Berlin.
Ragia, E., 2009. Η κοιλάδα του κάτω Μαιάνδρου ca. 600–1300. Γεωγραφία και Wulzinger, K., Wittek, P., Sarre, F., 1935. Das islamische Milet, Milet III 4. Walter de
Ιστορία [the Lower Maeander Valley, c. CE 600–1300. Geography and History]. Byz- Gruyter, Berlin.
antine Texts and Studies 51. Centre for Byzantine Research, Thessaloniki. Zgusta, L., 1984. Kleinasiatische Ortsnamen. Beiträge zur Namensforschung, Beiheft 21.
Rantitsch, G., Prochaska, W., 2011. Die hydrogeologische Situation des Panayırdağ als Carl Winter, Heidelberg.
Bewertungsgrundlage für die Wasserversorgung der vorlysimachischen Siedlung. Zolitschka, B., Behre, K.-B., Schneider, J., 2003. Human and climatic impact on the environ-
Jahreshefte des Österreichischen Archäologischen Institutes 80 col. 243–254. ment as derived from colluvial, fluvial and lacustrine archives — examples from the
Rayet, O., Thomas, A., 1877. Milet et le Golfe Latmique. J. Baudry, Paris. Bronze Age to the Migration period, Germany. Quat. Sci. Rev. 22, 81–100.
Robert, L., 1959. Philologie et géographie II. Sur Pline l'Ancien, livre II. Anatolia 4, 1–26.
Rojay, E., Toprak, V., Demirci, C., Süzen, L., 2005. Plio-quaternary evolution of the
Küçük Menderes graben southwestern Anatolia, Turkey. Geodin. Acta 18 (3–4),
317–331.