MERCOURIOS GEORGIADIS
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS:
AN OUTLOOK FROM YALI
Summary. Obsidian is an important material circulated throughout the
Aegean. Melian obsidian is found at many sites and researchers have
concentrated on Melos without acknowledging obsidian from other sources.
This article endeavours to highlight obsidian of non-Melian origin, such as that
from the Carpathians, central Anatolia, Antiparos and, particularly, Yali. It is
demonstrated that this latter source was a central one for certain islands, with
its obsidian found at a number of sites in the Dodecanese and beyond. It is also
emphasized that there were circulation modes of obsidian parallel to that of
Melos, as well as different procurement conditions. Therefore, by including
important regional sources such as Yali in the obsidian discourse, our
understanding of obsidian exchange and circulation is enriched.
introduction
Flaked tools are very common in assemblages, but their study in Aegean prehistory has
only attracted serious attention in recent years. The most common materials for making these
tools are chert, which at many sites is available locally, quartz, radiolarite, rhyolite, hornstone,
chalcedony, jasper and obsidian (Carter andYdo 1996, 155; Karimali 2005, 182, 187; Kozlowski
et al. 1996, 296 7; Matzanas 2000, 1 2; Moundrea-Agrafioti 1997, 170 1; 2004, 473; PerlŁs
1992, 128; Rapp 2002, 71; Robb and Farr 2005, 28 9). Exploitable obsidian sources are rare
across the world; in the Mediterranean they are evident only on islands, while in the Carpathians,
central and eastern Anatolia, Ethiopia and Eritrea they are found in inland mountainous areas
(Karimali 2005, 182; Renfrew and Aspinall 1990, 259 61; Robb and Farr 2005, 35).
In the Aegean there are three sources of obsidian: Melos and Antiparos in the Cyclades
and Yali in the Dodecanese (Georgiades 1956, 151; Moundrea-Agrafioti 2005, 51; Renfrew
1972, 442; Renfrew and Aspinall 1990, 259; Renfrew et al. 1965, 229 32; Shelford et al. 1982,
183, 190 1). Obsidian from Melos has dominated discussions in Aegean prehistory and the
intention here is to trace the real extent of its distribution and the ways in which it was circulated.
At the same time, obsidian recovered from other sources in the Aegean will be discussed,
highlighting its range and significance. The most prominent of these is obsidian from Yali, which
will be treated in detail with respect to its use, distribution and procurement. In this way, it will
be possible to illustrate the importance of obsidian sources other than Melos and the extent of
their circulation in the Aegean.
OXFORD JOURNAL OF ARCHAEOLOGY 27(2) 101 117 2008
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK
and 350 Main Street Malden, MA 02148, USA. 101
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
the obsidian from melos
The volcanic island of Melos is located in the western part of the Cyclades and includes
two outcrops of obsidian in the north-east at Sta Nychia and Demenegaki (Barber 1987, 113;
Carter and Kilikoglou 2007, 115 16; Moundrea-Agrafioti 2005, 51; Torrence 1982, 193). The
exploitation of Melian obsidian began in the Upper Palaeolithic (UP), as the finds from Franchthi
cave in the Argolid indicate, and expanded thereafter both in range and quantity, reaching a peak
in the Early Bronze Age (EBA) (Moundrea-Agrafioti 2005, 51; Robb and Farr 2005, 36; Shelford
et al. 1982, 182 3). The wide distribution of Melian obsidian has created a magnified illusion
with regard to its role, importance and distribution, and causing sources on Antiparos and Yali to
be neglected because the quality of the obsidian was not good enough for tools (Mantzanas 2000,
2). Hence, all other sources and varieties have been overshadowed to such a degree that in
Aegean prehistory, reference to obsidian is almost automatically taken to mean Melian.
In the Aegean, sites can be divided into those where obsidian dominated the chipped
stone tools and those where other materials were prevalent. Throughout the Neolithic, obsidian
became progressively more popular and dominated the flaked tools assemblages in southern
mainland Greece (Barber 1987, 117; Renfrew et al. 1965, 238). However, this older model is
only partly correct, since it seems that there was a further division between the eastern and
western parts of the Greek mainland. It appears that the Pindos range, as well as its extension to
the Peloponnese which runs in a roughly north south direction, formed a formidable barrier
to the distribution of obsidian. The surveys conducted in Laconia, including Ayios Stephanos
(Carter and Ydo 1996, 141; Kardulias 1992, 427), southern Argolid, including Lerna (Kardulias
and Runnels 1995, 74 6; Runnels 1985, 357 8), and central Greece including Lithares in Boiotia
and Manika in Euboia (Hartenberger and Runnels 2001, 275), have confirmed that obsidian
prevailed in eastern mainland Greece (Cosmopoulos 1991, 76; Moundrea-Agrafioti 1997, 170).
In contrast to this is Messenia, located in western Greece, where flint predominated at all sites
during the Bronze Age (BA), except at Romanou, where obsidian blade production took place
during Late Helladic (LH) I or II (Blitzer 1992, 712; Matzanas 2001 2, 49 50, 54; Parkinson
1999, 76, 80).
Melian obsidian was dominant in the Cyclades, but the picture is not the same on Crete.
Although at Knossos obsidian was popular and has been found in large quantities since the Early
Neolithic (EN), the Western Mesara Survey suggests that in this area obsidian represents almost
2.5 per cent of the flaked stones (Blitzer 2004, 510; Hope Simpson et al. 1995, 395), at Gavdos
3.9 per cent (Mantzanas 2000, 9 n. 100) and the site survey conducted by Nowicki (2002) across
Late Neolithic (LN) Crete has shown that chert predominates in southern Crete. Exceptions are
LN Phaistos and BA Kommos, as well as 13 out of 40 tholos tombs in Mesara that contained
obsidian blades (Blitzer 1995, 488 9; 2004, 512 13; Branigan 1970, 66). At the Vrokastro area
chert was mainly used in the FN period, but it changed in favour to obsidian in the BA (Hayden
2003, 42). Thus it seems that the northern part of the island was more open to the obsidian
exchange network, with limited penetration into the south where local chert resources were
exploited.
In northern Greece, i.e. Macedonia and Thrace, chert was more prevalent, with obsidian
penetration being limited in size and extent, reaching as far as Dispilio, Yannitsa and Sitagroi
during the Middle Neolithic (MN) and LN periods (Kilikoglou et al. 1996, 343). In the east
Aegean the obsidian distribution is more varied, with chert being predominant on LN and EBA
Limnos (Mantzanas 2000, 9 n. 100; Moundrea-Agrafioti 1997, 168), EBA Lesbos (Moundrea-
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
102 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
Agrafioti 1997, 176 n. 24), LN and EBA Chios (Bialor 1982, 699 700; Hood 1982, 711), BA
Troad (Renfrew et al. 1965, 238) and most of the Ionian coast. At Ulucak H�y�k, for example,
obsidian was found in modest quantities in Neolithic and EBA phases (�ilingiro
lu et al. 2004,
52; Greaves and Helwing 2003, 485). Nonetheless, obsidian pieces prevail at LN and EBA
Miletos and comprise 50 per cent of the flaked stones from LN Aphrodisias, an inland south-
western Anatolian site (Greaves 2002, 44). Obsidian is the most common material for tools at LN
Samos (Felsch 1988, 134), LN Ikaria (Sampson 2006, 164 5) and parts of the LN Dodecanese
(Sampson 1987).
The processes of procurement, distribution and manufacture are important issues
closely related to the consumption of obsidian. Scholars accept that there was open access to
obsidian from both Sta Nychia and Demenegaki on Melos, at least from the UP until the
foundation of Phylakopi in Early Cycladic (EC) III. While some scholars believe that there was
some kind of control over the sources by Phylakopi from EC III, or even earlier, until the end of
the Late Bronze Age (LBA) (Mantzanas 2000, 4, 19), others favour a model of continuity for the
same manner of acquiring obsidian as in the Neolithic (Renfrew 1972, 442 3; Renfrew et al.
1965, 241 2; Torrence 1982, 197; 1984, 61 2; 1986, 170 1). This open access to material, at
least during the Neolithic and part of the EBA, was achieved either through special voyages or
as one function of a trip where other activities such as fishing were performed, with a down-
the-line reciprocal exchange taking place (Kardulias 1992, 441; PerlŁs 1992, 145; Torrence
1982, 220; 1986, 103 5, 135 6, 216). However, the presence of one production centre in a
region, as in Messenia, or the presence of ready-made products on a site with limited local
production, as in the case of Laconia, strongly suggests a different mechanism of obsidian
exchange (Carter 1994, 134; 2005, 303 5; Carter and Ydo 1996, 163; Kardulias 1992, 437;
Torrence 1986, 137, 221). Carter (1999, 330; Broodbank 2000, 297 8) underlines the quantity of
obsidian assemblages at Phylakopi on Melos and at Poros-Katsambas on Crete, reinforcing the
idea of a different production and distribution mode.
obsidian in the aegean from sources other than melos
Although in the past it was believed that no obsidian was imported from beyond
the Aegean (Renfrew 1972, 442), recent finds suggest otherwise (Fig. 1). Possibly, the most
astonishing was the presence of Carpathian obsidian at Mandalon in western Macedonia
(Kilikoglou et al. 1996, 349; Moundrea-Agrafioti 2005, 52 3). From the 11 analysed samples of
LN and EBA date, nine LN and one EBA had a Carpathian provenance and only a single one
from Melos of EBA date. The predominance of Carpathian obsidian at Mandalon is impressive,
especially when the distance from the source (more than 600 km) is considered. Mandalon is an
excellent example of a site that strongly suggests the presence of multiple exchange networks
and the range they may have covered. Moreover, it highlights the point that the proximity of a site
to a source of material is not necessarily a reason per se for importing the material from there.
Thus, social motives should be sought to explain a preference for a distant exchange network
rather than a closer one.
The most common exotic source for obsidian found at sites in the Aegean is central
Anatolia, more specifically �iftlik. Obsidian from here has been found as far north as Sitagroi
in inland Thrace, most probably of EBA date, as well as at Nova Zagora in Bulgaria
(Kilikoglou et al. 1996, 344; Moundrea-Agrafioti 1997, 175; Renfrew and Aspinall 1990,
266; Shelford et al. 1982, 191). At Troy, Melian obsidian predominates, but Anatolian is
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 103
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
Figure 1
Distribution of obsidian from the Carpathians (triangle), Anatolia (circle) and Antiparos (square) in the Aegean.
1 Mandalon (Macedonia); 2 Sitagroi (Thrace); 3 Mikro Vouni (Samothrace); 4 Troy; 5 Poliochni (Limnos); 6 Ulucak
H�y�k (Izmir area); 7 Emporio (Chios); 8 Tigani (Samos); 9 Halasarna (Kos); 10 Yali; 11 Rhodes town (Rhodes);
12 Kalythies (Rhodes); 13 Platanos (Crete); 14 Knossos (Crete); 15 Poros-Katsambas (Crete); 16 Malia (Crete);
17 Apandima (Antiparos); 18 Krassades (Antiparos); 19 Saliagos.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
104 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
represented in modest numbers from phase I until much later (Moundrea-Agrafioti 1997, 175
6). At Ulucak H�y�k in the Izmir area, the analyses conducted on the obsidian finds from this
site reveal that the material derives mainly from central Anatolia (�ilingiro
lu et al. 2004, 52).
Although there are no chemical analyses, the transparency of a few tools from Mikro Vouni
on Samothrace, Poliochni on Limnos, Emporio on Chios and Tigani on Samos suggests an
Anatolian provenance (Moundrea-Agrafioti 1997, 176 n. 24; 2005, 53). The same applies for
the Dodecanese, where a few pieces, including cores, have been recovered at Kalythies and
Rhodes town on Rhodes (Sampson 1984, 69, 72; 1987, 46 53, 96), on Yali (Sampson 1988,
205) and most probably one core from the Halasarna region in southern Kos (Georgiadis
forthcoming). In Crete, �iftlik obsidian comes from Knossos (Carter and Kilikoglou 2007,
128; Kilikoglou et al. 1996, 344; Renfrew and Aspinall 1990, 266; Shelford et al. 1982, 191),
from an Early Minoan (EM) II Middle Minoan (MM) II tholos tomb at Platanos (Carter and
Kilikoglou 2007, 128, 130; Moundrea-Agrafioti 2005, 53) and Middle Bronze Age (MBA)
Malia, where the obsidian comes from the east G�ll� Da
source (Carter and Kilikoglou 2007,
126 8). In the case of the central Anatolian obsidian, there is a wide distribution mainly
concentrated in the eastern Aegean from the LN until the MBA. So far, no obsidian from the
central Mediterranean has been recognized in the Aegean nor any Aegean obsidian in Italy and
the Adriatic.
The obsidian source on Antiparos is located at the heart of the Cyclades. The quality of
the obsidian is good, but it is found in nodules of less than 5 cm in length, making this material
inappropriate for producing blades (Cann et al. 1968, 105; Evely 1993, 119; Georgiades 1956,
160; Moundrea-Agrafioti 2005, 51; Shelford et al. 1982, 191). Owing to this fact, many scholars
disregard its existence and do not include it as a possible source of obsidian (Barber 1987, 113;
Cosmopoulos 1991, 76; Karimali 2005, 182; Robb and Farr 2005, 35). Nonetheless, obsidian
from Antiparos in the form of nodules and flakes has been recovered from LN Saliagos, a small
islet probably connected to Antiparos at the time of its occupation (Cann et al. 1968, 106;
Renfrew 1972, 442). Moreover, a nodule has been found in a grave at Apandima and Krassades
cemeteries on Antiparos dating to the EC period (Renfrew et al. 1965, 239). Although the nodule
must have had certain significance in order to be placed in the grave, there seems little practical
use for this material. Their small size and their limited distribution peculiar to Antiparos led
Renfrew and Aspinall (1990, 259) to consider this source as insignificant. However, recently
an Antiparos obsidian piece has been found at Poros-Katsambas (Dimopoulou 1997, 433 4;
Mantzanas 2000, 8) of Prepalatial date (EM I MM I), raising questions about the extent of its
distribution and whether or not the material can be worked.
the obsidian from yali
Yali, only 6 km2, is the smallest island in the Aegean to possess an obsidian source; it is
located between Nisyros and Kos in the south-east Aegean. The main outcrops, situated in the
north-eastern part of the island, are placed within perlitic strata (Sampson 1984, 63; 1988, 10).
The obsidian from Yali is more transparent than that from Melos and Antiparos, but it is not pure.
It has white or pink spherical spots known as spherulites (chalcedony crystals), distributed five
or six per square centimetre, causing irregular fracture when worked, especially when the
 pressure technique is employed (Georgiades 1956, 155; Evely 1993, 119; Katsarou et al. 2002,
111; Moundrea-Agrafioti 1990, 406; Renfrew et al. 1965, 232; Sampson 1984, 68). Obsidian
from Yali is mainly known for the stone vessels made from it in Minoan Crete from MM I Late
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 105
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
Minoan (LM) II (Barber 1987, 113; Betancourt 1997, 173 4; Carter and Kilikoglou 2007,
124 6; Renfrew and Aspinall 1990, 259; Renfrew et al. 1965, 239 31; Sampson 1984, 69 72;
1988, 220 chart 23; Warren 1969, 135 6). Recently, nodules of higher quality obsidian have been
identified within the volcanic layers of the island. Thus far, most of the recovered specimens are
rather small for making blades, but flakes and fine implements have been recognized (Bassiakos
et al. 2005, 18; Sampson 1988, 205; 2006, 241).
A few scholars have proposed a limited use of this source for tools at sites in the
immediate vicinity, always with reference to its low quality (Moundrea-Agrafioti 2005, 51;
Renfrew 1972, 442). The same belief was shared by Sampson (1984, 67, 69; 1987, 84), who
having excavated many prehistoric sites in the Dodecanese, noted the presence of flakes and
blade-like pieces both at Alimnia and at Kalythies on Rhodes. However, during his excavation at
Yali he did not recover typical tools from the local source, but instead found blades from Melian
obsidian (Sampson 1988, 205). He therefore concluded that the material was not worked, but
circulated as a raw material in a limited way. Based on these new finds, he revised his belief about
the workable flakes from Alimnia, declaring the use of Yali obsidian to be a myth (Sampson
1988, 217; 2006, 236). He admitted only a very limited use of Yali obsidian for tool making,
primarily when Melian obsidian was unavailable at both Alimnia and Yali (Sampson 1988, 217;
2006, 240).
The distribution of Yali obsidian in the Aegean (Fig. 2)
The earliest use of the Yali obsidian is indicated at the site of Kerame on Ikaria, which
is dated to the Mesolithic period (eighth millennium BC) (Sampson 2006, 240 1). The technique
for extracting blades from Yali obsidian on this site followed the typical Epigravettian tradition
(Kaczanowska and Kozlowski 2006, 80 1). Nonetheless, the present evidence argues against the
continuation of exploitation of this obsidian source into the LN.
At five sites on Leros, in the Dodecanese (Table 1), when the Yali variety is recovered
it is always found together with greater quantities of Melian (Dreliosi 1994, 798; Sampson 1987,
90 1, 111). On Kalymnos they are attested at Embolas (Betancourt 1997, 173; Hope Simpson
and Lazenby 1962, 172 3), but at Chrysochera the majority of the obsidian is from Yali
(Nowicki, pers. comm.).
On Kos a series of Neolithic and/or EBA sites have been recovered across the island of
which ten yielded pieces of obsidian. At only one site does Melian predominate over the Yali
variety; at five of them Yali obsidian dominates the assemblages, but Melian is also found, while
at four sites only Yali obsidian is attested (Betancourt 1997, 173; Georgiadis forthcoming;
Jacopich 1928, 100; Hope Simpson and Lazenby 1970, 55; Levi 1925 6, 278 9, 282; Sampson
1987, 109; 1988, 218, chart 22, 229; 2001, 37). This is further supported by the recent survey
conducted in the Halasarna area on the south side of the island, where Yali obsidian has been
recovered from 13 sites (Georgiadis forthcoming). TheYali variety predominates over the Melian
at three sites, while Yali is found on its own at the remaining ten. Kos is the first island where,
without doubt, Yali dominates the assemblages over the Melian variety or is found on its own.
The survey has yielded a fair number of obsidian pieces (386) from which only 13 are Melian
and 96 per cent is fromYali, thus providing quantitative confirmation of the above point. Near the
site of Kephalos hill B, on a geological layer of volcanic ash, small obsidian pieces were
observed that were similar in appearance to the Yali variety. They do not exceed 1.5 cm in length,
which means that they are not appropriate for making flaked tools.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
106 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
Figure 2
Distribution of Yali obsidian in the Aegean. 1 Leros; 2 Kalymnos; 3 Kos; 4 Yali; 5 Pyrgoussa; 6 Pachia; 7 Nisyros; 8
Tilos; 9 Symi; 10 Seskli; 11 Rhodes; 12 Alimnia; 13 Chalki; 14 Karpathos; 15 Crete; 16 Anthochori (Lakonia); 17
Akrotiri (Thera); 18 Grotta (Naxos); 19 Saliagos; 20 Ayia Irini (Kea); 21 Tinos; 22 Kerame (Ikaria); 23 Iasos (Anatolia).
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 107
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
table 1
The sites with Yali obsidian from the Dodecanese
Island Sites More Melian More Yali Yali only
Less Yali Less Melian
Leros Ayia Kioura Probably
Partheni West peninsula Probably
Partheni Yes
Ayios Konstantinos Probably
Drymonas Probably
Kalymnos Embolas Probably
Chrysochera Yes
Kos Kastro Palaiopyli Yes
Misonisi Yes
Mesaria settlement Yes
Aspri Petra cave Yes
Ayios Fokas Yes
SW of Ayios Fokas Yes
Troulli Yes
Vouno Yes
Kephalos hill A Yes
Kephalos hill B Yes
Halasarna Survey 3 sites Yes
Halasarna Survey 10 sites Yes
Yali Site I Probably
SW cemetery Yes
SW Hill Yes
Laimos Yes
Pyrgoussa Pyrgoussa Yes
Pachia Pachia Yes
Nisyros Dhali Yes
Krios Yes
Zotikou Yes
Khokhlakoi Yes
Paliokastro Yes
Nisyros town Yes
Tilos Lakia Probably
Kampos Yes
Plakes Yes
Ayioi Apostoloi Yes
Symi Panormitis Probably
Drakounta Yes
Drakou Yes
Seskli Seskli Probably
Rhodes Trianda Yes
Anagros Probably
Massari Yes
Kamiros Skala Probably
Peukoi Probably
Koukoumia Yes
Kakoskali Probably
Limani Papakonstanti Probably
Kolympia Probably
Ayia Varvara Yes
Rhodes Town Probably
Koumelo cave Yes
Kalythies cave Yes
Alimnia Exo Pontikovounaro Probably
Kastro Alimnias Yes
Chalki Tracheia Probably
Emporio Probably
Pefkia Probably
Karpathos Leptokoros Probably
Lefkos Yes
Moulas Yes
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
108 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
As mentioned above, Yali is a small island which has been inhabited only occasionally
over the centuries, mainly on a seasonal basis, but during the LN and EBA there was a permanent
occupation of the island (Sampson 1988, 211 12). Large quantities of Yali obsidian have been
recovered across the island, but among them no typical tool has been found, although there are
numerous flakes and blade-like items that could have been used as tools, all without retouch
(Sampson 1988, 205 10). In the north-eastern part of the island where the obsidian outcrops
exist, three LN sites were located (Sampson 1988, 14), which, given the topography of the island,
could control access to the obsidian outcrops. Laimos 2 is an important site on a low hill close
to the sea, located on the narrow channel connecting the south-western with the north-eastern
part of the island. A very large quantity of obsidian has been recovered, including large pieces,
suggesting that at this site a local workshop was active during the LN and EBA (Sampson 1984,
67; 1988, 61). Several Melian pieces have been reported, but some of them may represent pieces
of the second obsidian sources of this island with colours closer to the Melian (Katsarou et al.
2002, 111). Although their size does not allow their exploitation for tools, they may have been
attributed to the Melian variety, which in reality may have been imported in far more limited
numbers. This obsidian variety is also used as a tempering material in the local pottery
production (Katsarou et al. 2002, 112).
Nisyros is the closest large island to Yali, and Yali obsidian has been found there on
its own in large quantities at five sites, and on another Yali predominates over the Melian
(Betancourt 1997, 173; Hope Simpson and Lazenby 1962, 169; Melas 1988, 287 91; Sampson
1988, 218, chart 22). The same applies to Tilos where both varieties are attested at three sites and
on a further two only Yali is found (Hope Simpson and Lazenby 1970, 68; Melas 1988, 293 4;
Sampson 1987, 115). On the small islet of Pachia, which was most probably seasonally
occupied, only Yali obsidian has been found (Sampson 2006, 242). The islet of Pyrgoussa did not
have a permanent settlement, but Yali obsidian was found in large quantities with some Melian
pieces attested as well (Sampson 1988, 252; 2006, 242).
Nonetheless, at Symi the situation is not as clear, since at one site more Melian pieces
were recovered than Yali, but at two a single Yali piece was found (Farmakidou 2003, 292 3;
Melas 1988, 294; Sampson 1987, 106; 1988, 218, chart 22). The islet of Seskli lies to the
south of Symi and has produced unspecified quantities of obsidian from Yali and Melos
(Farmakidou 2003, 292; Sampson 1987, 106; 1988, 218, chart 22). For Symi and Seskli it is
difficult to define the proportion of Yali obsidian and whether it was more or less common
than Melian.
Yali obsidian occurs in modest numbers in the earliest phases of occupation in the
Kalythies cave on Rhodes at the end of the sixth millennium BC (Sampson 1987, 46). From that
period on and certainly for the entire LN and FN periods (fifth and fourth millennia BC), Yali
obsidian was imported in considerable quantities to Rhodes. It has been found on at least 13 sites
across the island, always together with Melian obsidian, which predominated in quantity in these
assemblages. Nonetheless, there are two notable exceptions to this pattern: the first is the FN
Koumelo cave and the latter the LB I settlement at Trianda (Betancourt 1997, 174; Mee 1982,
4; Monaco 1941, 76; Papazoglou-Manioudaki 1982, 179; Sampson 1987, 76). At Trianda it is
possible to suggest a working of this material for the production of stone vessels, such as were
popular at that time on Crete. On almost all sites there are amorphous nodules of Yali obsidian,
suggesting their importation as raw material for local manufacture. Nevertheless, there is no
concrete evidence for the circulation of this material on Rhodes during the EBA, mainly due to
a lack of excavated sites from this period.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 109
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
The pattern of the Melian variety predominating over the Yali is most probably observed
on two further islands: on Alimnia at two sites (Betancourt 1997, 173; Melas 1988, 302 4;
Sampson 1987, 83 5, 106) and on Chalki at three (Betancourt 1997, 173; Melas 1988, 304,
306 7; Sampson 1987, 107, 113 14).
Karpathos is the southernmost island of the Dodecanese where Yali material has been
collected. Yali obsidian has been found in small numbers at only three sites, along with the
predominant Melian (Betancourt 1997, 173; Melas 1985, 70, 76 7, 139; Sampson 1987, 107;
1988, 218, chart 22).
There are also 15 sites across the Dodecanese, where there are reports of obsidian
recovered, but its type is not specified (Melas 1988, 306; Sampson 1987, 98 9, 101, 104 5,
107 10, 114; 2006, 243). This indicates that even researchers who were active in this region and
were conscious of Yali obsidian, were not consistent in reporting it and/or by the use of the single
term obsidian automatically meant Melian, as discussed above. Nevertheless, we should always
bear in mind that obsidian from Melos is the commonest flaked stone found on Leros, Kalymnos,
Rhodes, Alimna, Chalki and Karpathos.
Moving beyond the Dodecanese, Yali obsidian has been found in some areas in small
numbers. On Crete there are three LN FN sites where Yali obsidian has been recovered in the
form of flake tools: Knossos (Evans 1928, 14, fig. 5b, 56; 1921, 86 7, fig. 55c, 178, fig. 127e,
412; Nowicki 2002, 66 n. 194), and of uncertain quantity and character at Drapanon Cape
and Itanos (Nowicki 2002, 66 n. 194; pers. comm.). Nonetheless, Yali obsidian vessels were
especially popular in MM I LM II Crete, found at 11 sites across the island (Table 2). Thus far,
there is only one site on the Greek mainland, at Anthochori in inland Laconia, where a few pieces
of Yali obsidian, of EBA date, have been recovered (Gallou, pers. comm.).
table 2
Yali obsidian items from Crete
Site Item(s) Reference
Ayia Triada vessel Betancourt 1997, 174; Renfrew et al. 1965, 239 40; Sampson 1984,
69 72; 1988, 220, fig. 23; Warren 1969, 136, pl. 497
Juktas vessel Betancourt 1997, 174
Kato Zakros vessel Betancourt 1997, 174; Renfrew et al. 1965, 239 40; Sampson 1984,
69 72; 1988, 220, fig. 23; Warren 1969, 36, 136, pl.195
Katsambas pieces Betancourt 1997, 173; Dimopoulou 1997, 434; Evans 1935, 895;
Mantzanas 2000, 8; Sampson 1984, 69 72; 1988, 220, fig. 23; Warren
1969, 136
Knossos pieces and 5 vessels Betancourt 1997, 173 4; Evans 1928, 14, fig. 5b, 56; 1921, 86 7, fig.
55c, 178, fig. 127e, 412; Nowicki 2002, 66 n. 194; Renfrew et al.
1965, 239 40; Sampson 1984, 69 72; 1988, 220, fig. 23; Shelford
et al. 1982, 191, table 15.6; Warren 1969, 37, 77, 102, 136
Malia pieces Betancourt 1997, 173; Carter and Kilikoglou 2007, 124 6; Sampson
1984, 69 72; 1988, 220, fig. 23; Warren 1969, 136, pl. 627
Mochlos sealstone Betancourt 1997, 174; Sampson 1984, 69 72; 1988, 220, fig. 23
Palaikastro pieces and vessel Betancourt 1997, 173 4; Renfrew et al. 1965, 239 40; Sampson 1984,
69 72; 1988, 220, fig. 23; Warren 1969, 36, 91, 136
Drapanon pieces Nowicki 2002, 66 n. 194
Pseira vessel Betancourt 1997, 173; Sampson 1984, 69 72; 1988, 220, fig. 23
Pyrgos Myrtou vessel Betancourt 1997, 174; Sampson 1984, 69 72; 1988, 220, fig. 23
Tylissos pieces Betancourt 1997, 173; Renfrew et al. 1965, 239 40; Sampson 1984,
69 72; 1988, 220, fig. 23; Warren 1969, 136
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
110 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
In the Cyclades there are a few islands that imported Yali obsidian, such as LN Saliagos,
where a few pieces have been found (Betancourt 1997, 174; Cann et al. 1968, 106; Sampson
1984, 72; 1988, 218, fig. 22; Shelford et al. 1982, 191). Yali obsidian has also been recovered at
Ayia Irini on Kea (Betancourt 1997, 173; Davis et al. 1983, 361 2; Renfrew et al. 1965, 240;
Sampson 1984, 72; 1988, 220, fig. 23) and Akrotiri on Thera (Moundrea-Agrafioti 1990, 391 2),
both of Late Cycladic (LC) I date, and at Grotta on Naxos (Betancourt 1997, 173; Cann et al.
1968, 105 n. 1; Davis et al. 1983, 361 2; Sampson 1984, 72; 1988, 218, fig. 22 ) and on Tinos
(Levi 1925 6, 270; Renfrew et al. 1965, 232 n. 41; Sampson 1984, 72; 1988, 218, fig. 22), both
of uncertain date. Finally, a few pieces of Yali obsidian have been found in the EB I II burials
at Iasos, in coastal Anatolia (Pecorella 1984, 98), while both Renfrew (1972, fig. 20.1) and
Sampson (1984, figs. 10 11) mark a site in the Halikarnassos peninsula with Yali obsidian on
their maps, without providing further information about it.
conclusions
Melian obsidian was the primary commodity for manufacturing chipped stones in the
Aegean. Nevertheless, its predominance was limited in the eastern part of central and southern
Greece, including Thessaly, central Greece, the Peloponnese, the Cyclades, northern
Crete, some of the south-eastern Aegean islands and certain coastal parts of Anatolia. The
hypothesis of Melian obsidian exchange beyond the direct access theory is confirmed by the
presence of parallel exchange networks through which other types of obsidian circulated in
the Aegean. This is supported by Carpathian obsidian coming to Mandalon through an
extensive inland web of interactions and by central Anatolian material reaching the Aegean
either through the Izmir region in central western Anatolia or through the Troad and the
Propontis to the south-eastern Balkans. Thus, exchange was both the only way by which
non-Aegean obsidian could be circulated in this region, and an important factor in the
distribution of the Aegean material. Furthermore, the exchange of non-Aegean obsidian allows
the mapping of some of the contact channels through which goods, ideas, techniques and
technologies may have been transported.
The non-Aegean and Antiparos obsidian circulated in small numbers in most cases, but
the extent of their circulation, especially from central Anatolia, is impressive. It has been
demonstrated furthermore that the Antiparos source was used on a small scale at a local level,
although there was some circulation beyond its immediate environs. Knossos and its port,
Poros-Katsambas, seem to have been rather cosmopolitan given that obsidian from Melos, Yali,
Antiparos and central Anatolia has been recovered.
There are 71 sites across the Dodecanese where Yali obsidian has been recovered.
Nonetheless, certain islands seem not to have participated in its exchange, preferring Melian
obsidian, such as Patmos (Sampson 1987, 113) and Pharmakonissi (Dreliosi 1994, 799) to the
north, Astypalaia (Hope Simpson and Lazenby 1973, 163, 165 7) and Syrna (Hope Simpson and
Lazenby 1973, 169) to the west, which are more Cycladic in character. On Karpathos, Yali
obsidian is found in small amounts at just three sites, whilst on Saria (Melas 1985, 79) and Kasos
(Melas 1985, 81 3) none has been recovered, suggesting that its circulation beyond Rhodes was
limited. On most of the Dodecanesian sites Yali and Melian obsidian coexist, but in more than 20
settlements, only pieces from Yali have been found. Those locations are concentrated on islands
where Yali obsidian predominates the local assemblages. Hence Kos, Yali, Pachia, Pyrgoussa,
Nisyros, Tilos and possibly Symi and Seskli used Yali as their primary source for flaked stones.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 111
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
All of them are in the immediate environs of Yali itself and form a core area for the circulation
and use of this obsidian. Other islands such as Leros, Kalymnos, Alimnia, Chalki and Rhodes
imported large quantities of Yali obsidian and used it at a number of sites, in almost all cases
side-by-side with Melian, which was more common. Karpathos and Crete seem to have imported
some limited quantities of Yali pieces to more than one site, but they seem to have been outside
the main circulation area and participated only sporadically, as was the case with the other sites
in the Cyclades, the Greek mainland, Ikaria and coastal Anatolia, at least for the LN and EBA.
The fact that on Rhodes, Karpathos and the surrounding islands, Melian obsidian was imported
in large quantities suggests two things. Firstly, that there was a conscious preference for Yali
obsidian in the core area outlined above, since there was an active network circulating Melian
obsidian through or adjacent to this area. Secondly, this preference may indirectly reveal that
Melian obsidian reached this area through exchange and not direct access. Overall, Yali obsidian
has been recovered on at least 91 sites across the Aegean, an important indication of its
consumption and significance.
In the Dodecanese, the great majority of the datable sites on whichYali obsidian is found
belong to the LN and EBA, while others have a continuous use into the LBA. This peak of
circulation corresponds with the popularity of flaked tools and, most importantly, of the possible
permanent occupation of Yali itself. This model fits well with the evidence from the earliest LN
phase, Kalythies I on Rhodes, when Yali obsidian reached this site, although in small numbers
and predominated by the local chert (Sampson 1987, 46). However, from the next phase onwards
more Yali obsidian has been recovered from several sites across the Dodecanese, which
correspond with the new dating attributed to the occupation of Yali by its excavator (Sampson
2006, 234 5). The presence of a site in the north-eastern part of Yali, where the obsidian sources
are located, and the site of Laimos 2, where core preparation was taking place during the LN and
EBA, strongly suggest restricted access to the material and control over it. This procurement
strategy fits well with a scenario of territorial claim from the island inhabitants, limiting direct
access (Ericson 1984, 7). The duration and function of this site is immediately associated with
the procurement and circulation of this material, supporting the idea of an organized exploitation
rather than the hypothesis of sporadic use of the sources and no exchange of the material
(Sampson 1988, 217; 2006, 236). Yali is a rather small island in size with limited resources in
land and water but wealthier in minerals. The control of the mineral resources like obsidian was
a strategy to ensure the exchange of goods that were necessary for the local population. Clay
analyses have shown that a number of pots were imported for local use, possibly from Kos, thus
strengthening the above point (Katsarou et al. 2002, 115). The LN and EBA pattern of direct
access and limited exchange attested at Melos is different at Yali where, in the same period, the
obsidian source was controlled, with pre-working at a different location with immediate access
to the sea, followed possibly by exchange, reciprocal or other, for its circulation. The presence
of large pieces or nodules at several sites throughout the Dodecanese reveals the form in which
they were exchanged, supporting the hypothesis of pre-worked cores on Yali itself. This contrast
between Melos and Yali in the procurement should be emphasized, as well as the production on
site. It is possible to argue that the access, the geographical position and the quality of the two
obsidian varieties were the basic factors for their distribution and preference. The more complex
circulation of Melian obsidian, along with the direct access, allowed the formation of different
exchange networks, often acting in parallel across the Aegean. This was not achieved at Yali
since the control of its source permitted more limited quantities to be exported. The further an
island was located from Yali the more impractical its acquisition was, which together with its
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
112 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
lower quality made the inhabitants of these sites turn to the better and more widely circulated
obsidian from Melos.
In the MBA and LBA, the obsidian from Yali became popular on Crete for making stone
vessels. The large LBA pieces found at Trianda on Rhodes, Akrotiri on Thera, Ayia Irini on Kea
and possibly Grotta on Naxos may have served the same function. This change in use can be
explained by the fact that in the second millennium, when metal tools became widely available,
obsidian was not as essential as before. In that context, Melian obsidian was occasionally used,
mainly in blade form, but while the quality of the Yali was a disadvantage for such tools, it acted
as an advantage in that it produced stone vessels in beautiful colours. In this period no settlement
or exploitation seems to occur on Yali itself, suggesting that the pre-working of the material may
have taken place at a different site in the immediate vicninity, possibly on Kos.
There is no attempt here to suggest that Yali obsidian was of excellent quality; on the
contrary, it was worked with difficulty and did not produce standardized flaked tools. At the same
time it has become clear from excavations that it could produce workable flakes and blade-like
tools. This is evident from sites that have produced large quantities of this material and have
been properly published, such as Alimnia, Kalythies and Yali (Sampson 1984, 68 9; 1987, 47,
84; 1988, 205, 217), and is confirmed by the recent finds in the Halasarna area (Georgiadis
forthcoming).
Exchanges played a vital role in the distribution of obsidian in the Aegean, both of
local and non-Aegean provenance. This is clearly demonstrated in the ways by which central
Anatolian and Carpathian obsidian came to this region. However, the material from these sources
and Antiparos is rather small and quantitatively unimportant for the Aegean context.
Nevertheless, this does not apply to the Yali obsidian, which was preferred as a material in a core
area and circulated in considerable quantities in a wider region. Interestingly, this distribution
constituted a regional phenomenon, largely confined to the Dodecanese. Furthermore, it
provided an alternative model of obsidian source management and local exploitation, associated
with an exchange distribution mode.
Melian is by far the most common obsidian provenance in the Aegean, but the evidence
from Yali has demonstrated that it was a source of regional significance. Moreover, the myth of
a Melian monopoly is undermined by the increasing number of other obsidian types recovered
from more than a hundred sites across the Aegean, especially from Yali, enriching our
understanding of obsidian procurement and circulation.
Acknowledgements
I would like to express my gratitude to Professor W.G. Cavanagh for his help and comments, to
Dr Nowicki for our discussions and the information he provided me with, and to Dr C. Gallou for her
support and data she has kindly allowed me to use.
Department of Archaeology
The University of Nottingham
University Park
Nottingham NG7 2RD
e-mail: (1) [e-mail]
(2) [e-mail]
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 113
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
references
barber, r.l.n. 1987: The Cyclades in the Bronze Age (London).
bassiakos, y., kilikoglou, v. and sampson, a. 2005: Yali island: geological and analytical evidence for
a new source of workable obsidian. International Association of Obsidian Studies Bulletin 33, 18.
betancourt, p.p. 1997: The trade route for Ghyali obsidian. In Laffineur, R. and Betancourt, P.P. (eds.),
Craftsmen, Craftswomen and Craftsmanship in the Aegean Bronze Age (LiŁge, Aegaeum 16), 171 5.
bialor, p.g. 1982: The chipped stone assemblages from Emporio and from the lower cave at Ayio Gala.
In Hood, S. (ed.), Prehistoric Emporio and Ayio Gala, Vol. II (London, The British School at Athens
Supplementary Volume 16), 699 711.
blitzer, h. 1992: The chipped stone, ground stone, and worked bone industries. In McDonald, W.A. and
Wilkie, N.C. (eds.), Excavations at Nichoria in Southwest Greece. Vol. II The Bronze Age Occupation
(Minneapolis), 712 56.
blitzer, h. 1995: Minoan implements and industries. In Shaw, J.W. and Shaw, M.C. (eds.), Kommos I:
The Kommos Region and Houses of the Minoan Town (Princeton), 403 535.
blitzer, h. 2004: Appendix C. The chipped stone and ground stone, raw material sources, production
sites, and finds. In Watrous, I.V., Hadzi-Vallianou, D. and Blitzer, H. (eds.), The Plain of Phaistos  Cycles
of Social Complexity in the Mesasa Region of Crete (Los Angeles), 509 24.
branigan, k. 1970: The Tombs at Mesara (London).
broodbank, c. 2000: An Island Archaeology of the Early Cyclades (Cambridge).
cann, j.r., dixon, j.e. and renfrew, a.c. 1968: Appendix iv. The sources of the Saliagos obsidian. In
Evans, J.D. and Renfrew, C., Excavations at Saliagos near Antiparos (London, BSA Supplementary Vol.
No. 5), 105 7.
carter, t. 1994: Southern Aegean fashion victims, an overlooked aspect of Early Bronze Age burial
practices. In Ashton, N. and David, A. (eds.), Stories in Stone (London), 127 44.
carter, t. 1999:  Through a Glass Darkly : Obsidian and Society in the Southern Aegean Early Bronze
Age (Unpublished Ph.D. thesis, University College London).
carter, t. 2005: Chipped stone. In Cavanagh, W., Mee, C. and James, P. (eds.), The Laconia Rural Sites
Project (London), 303 5.
carter, t. and kilikoglou, v. 2007: From reactor to royalty? Aegean and Anatolian obsidians from
Quartier Mu, Malia (Crete). Journal of Mediterranean Archaeology 20(1), 115 43.
carter, t. and ydo, m. 1996: The chipped and ground stone. In Cavanagh, W., Crouwel, J., Catling,
R.W.V. and Shipley, G. (eds.), The Laconia Survey, Vol. II (London, BSA Supplementary Vol. No. 27),
141 82.
�ilingiro
lu, a., derin, z., e_
ref, a., sa
lamtimur, h. and kayan, i. 2004: Ulucal H�y�k  Excavations
Conducted Between 1995 and 2002 (Louvain, Ancient Near Eastern Studies Supplement 15).
cosmopoulos, m.b. 1991: The Early Bronze 2 in the Aegean (Jonsered, Studies in Mediterranean
Archaeology 97).
davis, j.l., schofield, e., torrence, r. and williams, d.f. 1983: Keos and the Eastern Aegean  the
Cretan connection. Hesperia 52(4), 361 6.
dimopoulou, n. 1997: Workshops and craftsmen in the harbour-town of Knossos at Poros-Katsambas. In
Laffineur, R. and Betancourt, P.P. (eds.), Craftsmen, Craftswomen and Craftsmanship in the Aegean
Bronze Age (LiŁge, Aegaeum 16), 433 8.
dreliosi, a. 1994: L ro�  Drum� na�. rcaiologikon Deltion 49, roniką, 798.
�
ericson, j.e. 1984: Towards the analysis of lithic production systems. In Ericson, J.E. and Purdy, B.A.
(eds.), Prehistoric Quarries and Lithic Production (Cambridge) 1 9.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
114 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
evans, a. 1921: The Palace of Minos, Vol. I (London).
evans, a. 1928: The Palace of Minos, Vol. II, Part I (London).
evans, a. 1935: The Palace of Minos, Vol. IV (London).
evely, r.d.g. 1993: Minoan Crafts: Tools and Techniques  An Introduction (G�teborg, Studies in
Mediterranean Archaeology 92).
ą ą
farmakidou, e. 2003: Pro storik� per odo� sth S�mh. In Vlachopoulos, A. and Mpirtacha, K.
(eds.), rgonauth� (Athens), 292 9.
felsch, r.c.s. 1988: Das Kastro Tigani  Die Sp�tneolitische und Chalcolithische Siedlung (Bonn).

� �
georgiades, a.n. 1956: Petrologąa, �reuna epą twn ellhnik n oyidian n. Praktika th�
kadhmia� qhn n 31, 150 63.
w
georgiadis, m. forthcoming: The Prehistoric finds. In Kokorou-Alevra, G. and Kopanias, K. (eds.), The
Halasarna Survey Project (Athens).
greaves, a.m. 2002: Miletos: A History (London).
greaves, a.m. and helwing, b. 2003: Archaeology in Turkey, the Stone, Bronze and Iron Ages, 2000.
American Journal of Archaeology 107, 71 103.
hartenberger, b. and runnels, c. 2001: The organization of flaked stone production at Bronze Age
Lerna. Hesperia 70, 255 83.
hayden, b.j. 2003: The Final Neolithic Early Minoan I/IIA settlement history of the Vrokastro area,
Mirabello, Eastern Crete. Mediterranean Archaeology and Archaeometry 3, 31 44.
hood, s. 1982: Prehistoric Emporio and Ayio Gala, Vol. II (London, The British School at Athens
Supplementary Volume 16).
hope simpson, r. and lazenby, j.f. 1962: Notes from the Dodecanese. Annual of the British School at
Athens 57, 154 75.
hope simpson, r. and lazenby, j.f. 1970: Notes from the Dodecanese ii. Annual of the British School at
Athens 65, 47 77.
hope simpson, r. and lazenby, j.f. 1973: Notes from the Dodecanese iii. Annual of the British School at
Athens 68, 127 79.
hope simpson, r., with betancourt, p.p., callaghan, p.j., harlan, d.k., hayes, j.w., shaw, j.w., shaw,
m.c. and watrous, l.v. 1995: The archaeological survey of the Kommos area. In Shaw, J.W. and Shaw,
M.C. (eds.), Kommos I: The Kommos Region and Houses of the Minoan Town (Princeton), 325 402.
jacopich, g. 1928: Coo  grotta de Aspri Petra. Clara Rhodos I, 99 100.
kaczanowska, m. and kozlowski, j.k. 2006: Palaioliqik�� parad sei�, esoliqik �
o �
�
prosarmog � kai eoliqik�� kainotomąe� sto igaąo m sa apo to prąsma th� liqotecnąa�.
�
h- esoliqikh- eoliqikh (Athens),
In Sampson, A. (ed.), Proistor a tou iga ou: Palaioliqik
i i
67 87.
kardulias, p.n. 1992: The ecology of Bronze Age flaked stone tool production in Southern Greece,
evidence from Aghios Stefanos and the Southern Argolid. American Journal of Archaeology 96, 421
42.
kardulias, p.n. and runnels, c. 1995: The lithic artifacts, flaked stone and other nonflaked lithics.
In Runnels, C.N., Pullen, D.J. and Langdon, S. (eds.), Artifact and Assemblage, Vol. 1 (Stanford), 74
109.
karimali, e. 2005: Lithic technologies and use. In Blake, E. and Knapp, A.B. (eds.), The Archaeology of
Mediterranean Prehistory (Oxford), 180 214.
katsarou, s., sampson, a. and dimou, e. 2002: Obsidian as temper in the Neolithic pottery from Yali,
Greece. In Kilikoglou, V., Hein, A. and Maniatis, Y. (eds.), Modern Trends in Scientific Studies on Ancient
Ceramics (Oxford, BAR International Series 1011), 111 20.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 115
THE OBSIDIAN IN THE AEGEAN BEYOND MELOS
kilikoglou, v., bassiakos, y., grimanis, a.p., souvatzis, k., pilali-papasteriou, a. and papanthimou-
papaefthimiou, a. 1996: Carpathian obsidian in Macedonia, Greece. Journal of Archaeological Science
23, 341 9.
kozlowski, j.k., kaczanowska, m. and pawlikowski, m. 1996: Chipped-stone industries from Neolithic
levels at Lerna. Hesperia 65, 295 372.
levi, d. 1925 6: La grotta di Aspripetra a Coo. Annuario della Scuola Archeologica di Atene e delle
Missioni Italiane in Oriente 8 9, 235 312.
matzanas, c. 2000: ropoi proskthsh� kai katątmhsh� puritik�n petrwmątwn katą thn
i �
pr imh alkokratąa. rcaiologikon Delt on 55, el te�, 1 22.
�
�
matzanas, c. 2001 2: ukhnaąk liqotecnąa sto Yąri rijuląa�. Praktika tou S
Dieqnou� Sunedriou Peloponnhsiakwn Spoudwn (Athens), 49 64.
mee, c. 1982: Rhodes in the Bronze Age: An Archaeological Survey (Warminster).
melas, m. 1985: The Islands of Karpathos, Saros and Kasos in the Neolithic and Bronze Age (G�teborg,
Studies in Mediterranean Archaeology 68).
melas, m. 1988: Exploration in the Dodecanese, new prehistoric and Mycenaean finds. Annual of the
British School at Athens 83, 283 311.
monaco, g. 1941: Scavi nella zona micenea di Jaliso (1935 6). Clara Rhodos 10, 41 183.
moundrea-agrafioti, a. 1990: Akrotiri, the chipped stone industry, reduction techniques and tools of the
LC I phase. In Hardy, D.A. (ed.), Thera and the Aegean World III. Vol. 1 Archaeology (London), 390 406.
ą
moundrea-agrafioti, a. 1997: liqotecn a th� Poliocnh� kai h q� sh th� w� pro� ti�
��
ą � ą �
ergaleiotecn e� tou apokrousm nou l qou th� Pr� imh� poc tou alko . In Doumas,
u o
C.G. and La Rosa, V. (eds.), Poliocnh kai h Prwimh poch tou alko sto reio igaio
(Athens), 168 94.
� � ąontwn
moundrea-agrafioti, a. 2004:  grotik�� kalli rgeie�, metapoąhsh agrotik n pro kai
�
ląqina ergaleąa sto krwt ri th� Q�ra� . Altellenische Technologie and Technik (Weilheim),
469 84.
� ą
moundrea-agrafioti, a. 2005: a ergaleąaapo laxem no l qo katą thn poc� tou alko�, to
� � i
t lo� mąa� tecnologik � parądosh�. rcaiolog a kai ecne� 94, 49 57.
nowicki, k. 2002: The end of the Neolithic in Crete. Aegean Archaeology 6, 7 72.
�
papazoglou-manioudaki, l. 1982: naskaj tou inwiko� oikismo� sta riąnta th� odou.

rcaiologikon Deltion 37 elete�, 139 90.
parkinson, w.a. 1999: Chipping away at a Mycenaean economy, obsidian exchange, linear B, and palatial
control in Late Bronze Age Messenia. In Galaty, M.L. and Parkinson, W.A. (eds.), Rethinking Mycenaean
Palaces  New Interpretations of an Old Idea (Los Angeles), 73 85.
pecorella, p.e. 1984: La Cultura Preistorica di Iasos in Caria (Rome).
perlŁs, c. 1992: Systems of exchange and organization of production in Neolithic Greece. Journal of
Mediterranean Archaeology 5, 115 64.
rapp, g. 2002: Archaeomineralogy (Berlin).
renfrew, c. 1972: The Emergence of Civilisation  The Cyclades and the Aegean in the Third Millennium
BC (London).
renfrew, c. and aspinall, a. 1990: Aegean obsidian and Franchthi cave. In PerlŁs, C. (ed.), Les Industries
Lithiques Taill�es de Franchthi (Argolide, GrŁce), Vol. 2 (Bloomington), 257 70.
renfrew, c., cann, j.r. and dixon, j.e. 1965: Obsidian in the Aegean. Annual of the British School at
Athens 60, 225 47.
robb, j.e. and farr, r.h. 2005: Substance in motion, Neolithic Mediterranean  trade . In Blake, E. and
Knapp, A.B. (eds.), The Archaeology of Mediterranean Prehistory (Oxford), 24 45.
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
116 Journal compilation � 2008 Blackwell Publishing Ltd.
MERCOURIOS GEORGIADIS
runnels, c. 1985: The Bronze-Age flaked-stone industries from Lerna, a preliminary report. Hesperia 54,
357 91.
o ą
sampson, a. 1984: O oyian � th� is�rou kai h diądosh tou sto iga o. nqrwpologika 5,
63 74.
a
sampson, a. 1987: eoliqikh Periodo� sta Dwdek nhsa (Athens).
h
sampson, a. 1988: eoliqik atoikhsh sto Guali th� isurou (Athens).
ą ą
sampson, a. 2001: w� kai h nhs da Gual sth eoliqik� perąodo. In Kokkorou-Alevra, G.,
i o
Lemou, A.A. and Simantoni-Bournia, E. (eds.), Istor a, ecnh kai rcai thte� th� w (Athens,

ą
rcaiognws a 1), 37 48.
i
sampson, a. 2006: Proistoria tou iga ou: Palaioliqikh- esoliqikh- eoliqikh (Athens).
shelford, p., hodson, f., cosgrove, m.e., warren, s.e. and renfrew, c. 1982: The sources and
characterisation of Melian obsidian. In Renfrew, C. and Wagstaff, M. (eds.), An Island Polity  The
Archaeology of Exploitation in Melos (Cambridge), 182 92.
torrence, r. 1982: The obsidian quarries and their use. In Renfrew, C. and Wagstaff, M. (eds.), An Island
Polity  The Archaeology of Exploitation in Melos (Cambridge), 193 221.
torrence, r. 1984: Monopoly or direct access? Industrial organization at the Melos obsidian quarries. In
Ericson, J.E. and Purdy, B.A. (eds.), Prehistoric Quarries and Lithic Production (Cambridge), 49 64.
torrence, r. 1986: Production and Exchange of Stone Tools: Prehistoric Obsidian in the Aegean
(Cambridge).
warren, p. 1969: Minoan Stone Vases (Cambridge).
OXFORD JOURNAL OF ARCHAEOLOGY
� 2008 The Author
Journal compilation � 2008 Blackwell Publishing Ltd. 117