The Chronology of the Last Phases of Occupation at Akrotiri in the Light of the Evidence from the West House Pottery Groups

The West House pottery group includes all the various types of local and imported wares of the Volcanic Destruction Level, which date the final phase of the house to the 'mature' Late Minoan IA.

The West House is the only building in which systematic stratigraphic research is being carried out. The pottery deposits below the floors of the final phase indicate the first phase of the building. This earlier phase overlaps chronologically with an early stage of Late Minoan IA.

The rather fragmentary evidence from the other buildings on the site, when looked at in the light of the research in the West House, suggests that the whole settlement of Akrotiri entered its last period of occupation, divided into two phases, in the Late Middle Cycladic era.

INTRODUCTION

Scholars have been interested in the chronology of the closed pottery group of the Volcanic Destruction Level (VDL) at Akrotiri, from the first excavation seasons, in order to discover the exact moment of the settlement's destruction and also to correlate that with devastations on Crete. The preliminary assessment of the material for this purpose was confined to only a few characteristic vases and sherds presented in the excavation reports (Marinatos 1968-76, II, 36; Hood 1971, 379). Now, that the systematic study of the Akrotiri material is more advanced, the chronology of this closed group can be reviewed. This group is unique because it contains whole vases and can provide a guide for a more closely defined classification of other Aegean closed deposits in the late Middle and early Late Bronze Age and hence improve our general knowledge of Aegean relative chronology at the time of the Thera volcanic destruction. Therefore, one of the two main aims of this paper is to present all the wares of this group so far recognized.

Study of the pottery from the whole settlement has shown it to be largely homogeneous, whether it is from a detached house or houses that are part of larger complexes. Even though there are differences in percentages from structure to structure the wares themselves are similar. The West House (WH) is the most fully excavated building, therefore I shall give more weight to its material as this may be considered representative.

Apart from the chronology of the above closed group, other problems of a chronological nature have emerged as the excavation has continued. Material in pits dug for the roof supports and finds embedded in floors and walls of the VDL have yielded evidence of human presence at the site from at least the Early Cycladic period (Doumas 1978, 777-778; Sotirakopoulou 1986). These occasional finds demonstrated the need for a methodical investigation of the early history of the settlement. In the light of the excavations at Phylakopi and Ayia Irini, where three and six periods of occupation have been identified respectively till the early LBA (Renfrew 1978, 404-407; Caskey 1979, 412), it does not seem unreasonable to expect that Akrotiri had a similar history. Consequently, systematic excavation under the floors of the Late Cycladic town was necessary in order to define the stratigraphy and thus the habitation periods of the site.

The first excavations at Akrotiri concentrated on the structures revealed immediately under the volcanic layers, although some tests were carried out beneath them (Marinatos 1968-76, V, 27, II, 17, sections VIII and IX). From 1978 onwards a system of trenches has been laid out in the open areas between Complex Delta to the west and Xeste 2 and Xeste E to the east in order to investigate earlier occupation levels (Marthari 1984, 119-126; Doumas 1985, 64-65). However, these trenches were of no value in dating the buildings of the Late Cycladic town. For this reason it was deemed necessary to open test-pits against the interior and exterior sides of walls in various houses. This action opened up certain difficulties. The houses are preserved to an impressive height, up to two storeys, with intact frescoes, slab floors, shelves, cupboards and benches with jars set into them even at first floor level. Two problems were faced. Firstly the technical task of supporting the walls and fittings in such a way as to allow safe excavation and secondly an issue of archaeological ethics: is it right for morphological elements of such well-preserved buildings to be sacrificed in order to study the early history of the settlement?

After much deliberation, a compromise was arranged which, although not ideal, permitted work to commence by the walls. It was decided to open small trenches in positions that would create the fewest technical problems and restrict damage to a minimum. These keyhole tests provide results of a limited value but can give useful indications regarding the construction of the buildings. The West House was the first structure to be investigated in the above fashion and the presentation of results from this particular work, the examination of earlier phases, forms the second main aim of this paper.

THE WEST HOUSE EXCAVATION DATA

The West House has been explored between 1973 and 1984 (Marinatos 1968-76, V, VI; Doumas 1976-84). With the removal of the ash and pumice it was revealed to be standing to its first floor. Only its two staircases remain partially unexcavated for technical reasons at ground-floor level. Therefore, almost the entire pottery content of the building is available for study.

On the surviving part of the first floor a considerable number of vases were found in situ. These were discovered mainly in cupboards, on window-sills and in one case embedded in a bench. From the ground floor (Fig. 1) a far larger quantity of ceramic material was recovered in situ. Most of these vases stood on the floor in Rooms 5 and 6. On both the ground- and first-floor levels pots were found, not in situ, but fallen from the collapse of a possible second floor and of a large part of the first floor. Vessels were often found entirely intact and some have preserved residues in the form of cereals, pulses and other organics. From such remains a vivid picture of life in the building may be built up. The position of pithoi against walls and the great number of storage vessels in the ground-floor Rooms 5 and 6 bear a resemblance to the picture we have of the cellar in the palace of Odysseus as described by Homer (Odyssey II, 337-342).

There is a difference in floor level from east to west which takes the form of steps between the rooms. Thus, the floor of the vestibule at the east high end is level with the Triangle Square outside of it whilst Rooms 4 and 5 at the west end are in reality semi-basements. During the 1984 season trenches were opened in the floors of Rooms 3A, 3C, 4, 5, and 6 (Fig. 1). These trenches revealed that there was a difference in stratigraphy between the east and west parts of the house.

In 3C, 4 and 5 at the west end (trenches WHe, c, b respectively: Fig. 1) the following picture emerges (Fig. 2). Under the clay floor of the house, which is 0.10 m thick, there was a fill which differed from room to room (0.40 m, 0.30 m and 0.65 m for 3C, 4 and 5 respectively). This fill consisted of large stones, broken stone tools and vessels, pieces of white plaster from floors and walls, broken small objects such as loomweights and ceramic material. This seemed to be debris brought in from elsewhere to level the floors. This debris was deposited as two layers. The pottery of these layers and in the clay floor above contains joins across all three strata and it is clear that this material comes from a single operation.

The fill rested on a second clay floor, 0.20 m thick, made from two successive clay layers. This was laid onto bedrock and also concealed pits which had been cut into the bedrock. The pits were circular, elliptical, or irregular in shape and were covered by either a light yellow clay, discarded quernstones, large body sherds, stone slabs or crushed bedrock excavated from the pit itself. In all of them were vessels, the open shapes usually turned upside down and the closed ones standing upright.

In some of the pits were small stone balls, in one case a stone bead and in three of the closed vases, that is nippled jugs, obsidian flakes were found. The stepping of the final floor (see above) can be seen, with differences, in the earlier floor beneath the fill. It is now clear that it is due to the slope of the bedrock from east to west in this area.

The trenches in Rooms 3A and 6 (WHd and WHa: Fig. 1) yielded a quite different and less clear picture. In Room 6 under the 0.10 m clay floor was a fill 0.40 m thick similar to that noted above. In Room 3A there was no fill but successive clay layers 0.50 m thick. However, both the fill of 6 and the clay layers of 3A stood on a clay floor at the same depth. This particular floor was a thin layer of clay, just 0.04 m thick, on top of a layer of small pieces of red lava 0.50 m thick, collected from nearby Mavro Rachidi, forming a prepared platform for the clay. These lava pieces filled up and covered over the remains of the earliest walls under the West House (walls a and b, Fig. 1). These walls had been cut by the later foundation walls of the West House. Unfortunately, the need to keep this floor intact for further study, the existence of roof support pit 24 in Room 6 and the small size of the tests made it impossible to establish the exact date of these early walls.

THE WEST HOUSE POTTERY GROUPS

• The West House Group A - Pottery from the ground and first floors:

The pottery from the ground and first floors (Fig. 3) consists of c. 1000 entire or almost complete vessels and some sixty boxes of fragmentary sherd material. It is characterized by a great variety of both local and imported wares.

The imported wares represent no more than 10% of the total assemblage, but they do demonstrate the wide range of contacts across the Aegean that Thera enjoyed at this time. The most numerous imports are Minoan. LM IA style is the most frequently found, across a broad range of shapes and sizes. The semi-globular cup is most common while the Vapheio cup is very rare. Other common types are the askos, the bridge-spouted jug and jar, the whole-mouthed jug, the piriform jar and many different forms of rhyta. Less frequent are the alabastron, the feeding bottle, the cut-away neck jug, the oval-mouthed amphora and the stirrup jar. Of the motifs, the first in popularity is the running spiral and the second, the foliate band. Rosettes and leaf-like tendrils are common as well. There are just a few examples of tortoise-shell ripple and reed pattern while conglomerate pattern and the double-axe motif occur on only a single vase each. A similar picture emerges from the LM IA style pottery in other buildings with the addition of a few other motifs; the crocus, the ivy leaf, the figure-of-eight shield and stipple pattern.

The LM IA style wares of the West House provide parallels with pottery from central and eastern Crete. Comparanda from Knossos are the most valuable. There are close affinities with the Gypsades Well Deposit (Evans 1921-35, II, 549, Fig. 349; Popham 1967, 338-339) and with other deposits that are contemporary with it such as the fill below the East-West stairs of the Domestic Quarters of the Palace and the later deposits of the Unexplored Mansion (Popham 1976, 194-195; 1984, 1154, 156-158). The Gypsades Well Deposit formed Evans's main group for the LM IA period which he felt represented a mature stage for that period. The following comparisons may be made. A series of cups decorated with running spiral in both its retorted and central disk varieties may be compared with examples from the Unexplored Mansion (Popham 1984, Pl. 130 a-g). The Vapheio cup, no. 3191, with a double spiral band and rows of white disks (Fig. 4b) finds an exact counterpart in the Gypsades Well Deposit and also in other Knossos deposits (Popham 1967, 337, Pl. 76d; 1984, 157, Pl. 133c: 1st; Warren 1980-81, 75, Fig. 5). The lustrous whole-mouthed jug, no. 4675 (Fig. 5a), with spirals and reeds, finds parallels in the later Unexplored Mansion deposits where the combination of this shape with reeds is usual (Popham 1984, 157, Pl. 131g and 135b). From this it is possible to conclude that the West House material and that of other houses at Akrotiri was contemporary in Minoan terms with the mature stage of the LM IA style at Knossos.

East Cretan parallels are provided by several different urban centres. Three bridge-spouted jars with disk spiral decoration, subsidiary bands of red and applied white enrichments (Fig. 6a-b) from the West House are strongly reminiscent of a vase in the Ashmolean Museum to which Popham assigned an East Cretan provenance (Popham 1967, 339, Pl. 77b). A whole mouthed jug (Marinatos 1968-74, VI, Pl. 78b), a peg top rhyton (Fig. 5b) and a small jar decorated with bands of connected rosettes, foliate bands and rows of disks may be equated with East Cretan wares, particularly from Gournia, Room C58 (Boyd Hawes 1908, 40, Pl. vii nos. 25-32, 34-41, F). This deposit is considered as characteristic LM IA by several scholars (Hood 1978, 685; Niemeier 1980, 63-65). Finally a large jar, no. 5997 (Fig. 7), with double-axe motif recalls similar East Cretan jars (Seager 1910: Pl. VII).

Examples of Minoan dark ground wares also form a part of this imported material. They include three intact vases, an alabastron rhyton with reed decoration, a small amphora with toliate branches and a large ewer with running spirals on the shoulder (Fig. 9). Their white painted decoration with a single red band on just two of them points to a late MM III date. The ewer is very similar to ewers from the Temple Repositories at Knossos, considered by Evans to be MM IIIB deposit (Evans 1921-35, I, 557, Fig. 404, f). In addition to this West House ewer, other large storage vessels have been found in both the light-on-dark and dark-on-light (Fig. 10)techniques at Akrotiri, which also find a good match in the Temple Repositories material and other contemporary Knossian groups. It has been suggested that these groups could overlap with LM IA to some extent (Platon 1973, 248-253); however, this has proved difficult to establish because of the lack of fine wares which are necessary for chronological correlations to be achieved (Popham 1976, 192).

Vases from the Greek mainland also form a part of the Group A pottery. They number far fewer than the Cretan imports. Half of them are characteristic of the LH I style and are mainly Vapheio cups decorated with linked spirals flanked by blobs or other motifs of this repertory (Fig. 4a). Mainland Vapheio cups have been found in large quantities at Akrotiri in comparison with their Cretan counterparts. Mycenaean semi-globular cups have not been found in the West House whereas their related Cretan types are common although this is the second most common Mycenaean form at the town as is quite clear from other house assemblages. The LH I material permits the correlation of the West House VDL material, as well as that of other houses with closed Helladic groups such as Korakou East Alley, levels XII-XVI (Davis 1979, 234-240, Fig. 3). One interesting vase from the West House is a bridge-spouted jug with double axes and dotted crosses (Fig. 8). Its decorative character could suggest a possible mainland origin although it is considered that this form first appears in LH IIA deposits (Mountjoy 1981, 26, Fig. 24). This matter needs further consideration and space than can be given here, and should be examined in another paper.

The other half of the mainland material consists of large containers which continue the Middle Helladic stylistic traditions. There are three matt-painted vases (one amphora and two hydrias decorated with double circles) and two kraters of Polychrome Aigina ware (Marthari 1980, 182-192, Fig. 1, 4, Pl. 68α-β, 70). Other such wares, including Polychrome Mainland, have been found in other houses (Marthari 1980, 132-198). Very few of these MH-tradition wares could be survivals from an earlier period. Recent studies have shown that pots with double circles were produced continuously from the end of the Middle Helladic to LH IIIA1 (Mountjoy 1981, 75-77; Cummer and Schofield 1984, 46) whereas the Polychrome Mainland and Aigina wares were being produced in LH I alongside LH I style pottery (Davis 1979, 256-259; Dietz 1980, 139).

The third large group of imported vessels comes from the other Cycladic islands. Both form and fabric support this observation, but for most examples they do not provide an exact provenance. Given that only two other Cycladic settlements of MBA and LBA date have been systematically excavated, Ayia Irini and Phylakopi, this is entirely understandable. More recent work on Naxos and Ios has shown that there were developed settlements at this time on other islands with their own local wares. An eyed jug, no. 5126, decorated with spirals, the fabric of which suggests a Cycladic provenance has been found in the West House (Fig. 11). At Akrotiri, a ware easily distinguishable by the micaceous dark red fabric, thick yellow slip and red and black decoration could be of Naxian or Kean provenance (Caskey 1972, 381; Davis 1986, 4; Cummer and Schofield 1984, 48).

Relations with the Dodecanese and south-west Asia Minor are reflected in the settlement pottery. A fabric of red-brown colour and golden mica inclusions decorated on its surface mainly in light-on-dark must have its origins there (Morricone 1975, 269-326). An amphora and some sherds of this style come from the West House (for this ware, see in detail, Marthari et al. 1990, 171-175).

Local wares constitute the majority of the West House Group as is the rule for the settlement as a whole. Local vases are made from a buff, semi-coarse, fabric of Theran origin. (For chemical and petrographic analysis see Jones 1978; Einfalt 1978; Williams 1978; Kilikoglou 1988; Vaughan, volume one of this Congress; for Williams's 1980 and Pittinger's 1979 unpublished works see Jones 1986, 275.)

There are three main classes of this pottery defined by surface treatment (see Marthari 1990, 450 Fig. 1). Firstly, the most numerous are those with a self slip in the same colour as the buff clay or most usually a wash. A number of these are plain, but most are decorated. The decoration is monochrome, rarely bichrome (red and black) and usually monochrome or bichrome with white enrichments (polychrome). The second most numerous are dark coated and usually burnished vessels either plain or with white painted motifs. The third and least numerous category is formed by those vases which are white coated and burnished, plain or with dark decoration. Within these three classes there are also variations of form and motif.

The variety of this pottery can be explained in part by the continuing local tradition but also by its adoption of Minoan elements which permits some of this material to be termed Minoanizing (Marthari 1987, 362, 366, 373, 376). But despite this apparent variety there is, nonetheless, strict standardization with a motif or sets of motifs following a form consistently. This regularization of Theran pottery production involves large numbers of vessels to the extent that vases occurring in ones and twos in the Late Cycladic town might be considered survivals as a rule. This would seem to be the case with a trefoil-mouthed jug with pomegranate decoration (Marthari 1984, 129, Fig. 8d, right) and a pithos with faunal and marine motif decoration (Doumas 1976-84 in 1980, Pl. 178, 179). Parallels have been found amongst the earlier deposits, but there is nothing of comparison from the VDL.

This local pottery is quite similar to the Kean pottery of Period VI and the Milian of Period III, even if there are sharp differences between the various Cycladic workshops.

It has been shown that the West House Group A and more generally the Akrotiri VDL closed group are contemporary with many Aegean groups of the early phase of the Late Bronze Age. The Minoan material in particular declares that it is contemporary with mature LM IA groups. It also contains some imported and local vases which survived from earlier occupation of the West House and the settlement in general.

• Group B -The pottery from the levelling fill and in the final floor of Rooms 3C, 4 and 5: 

As mentioned above, this material has been shown by study to form a coherent, homogeneous group (Fig. 3). However, because of the small number of sherds recovered, results from these trenches should be treated as indicative, but not conclusive.

The ratio of imported to local wares is similar to that inside the house. The imported material includes some shiny LM IA sherds including pieces of Vapheio and semi-globular cups decorated with tortoise-shell ripple pattern and spirals. The only other recorded motif is the ivy leaf seen on one sherd. Mainland wares are represented by hydrias and jars of MH style. The local wares have a limited range of motifs when compared to the Group A material. Tortoise-shell ripple is the dominant design, followed by circles with in-filled disks in the interior (Marthari 1987, 361, Fig. 1a, 2-4). Pictorial designs are rare. Several sherds from jugs with pomegranate motifs were found and, like the jug of Group A, they recall Milian black and red style jugs.

This material may be correlated with pottery from Trench A to the east of Complex Delta and from Pit 35 (Marthari 1984, 129-131). Both trenches contained part of a debris layer that has been isolated under the floors of the settlement in many open areas. This layer has been related to severe seismic destruction which has to have occurred when samples of an early stage of LM IA style had arrived on the island. There is evidence of large-scale damage and rebuilding operations, part of which consisted of the deliberate accumulation and covering over of debris in the open areas. This was accompanied by levelling operations within houses in order to raise the floor levels to a height comparable to that of the external areas (Marinatos 1968-76, V, 9, 44; Doumas 1978, 780; Marthari 1984, 119-133; Palyvou 1984, 134-147).

The Seismic Destruction Level (SDL) deposits have been previously compared with a stratified deposit in Trench PLa from the modern excavations at Phylakopi assigned by Renfrew to a very early phase of the LBA (Marthari 1984, 131; Renfrew 1978, 407). The sherd material of Group B in the West House contains similar parallels. Comparisons with Cretan material had been avoided because of uncertainty regarding the distinctions between MM IIIB and LM IA. The recent publication of the Unexplored Mansion, which detailed early LM IA deposits, stratified in the south corridor below mature LM IA deposits (Popham 1984, 155-156), allows tentative correlations between the early LM IA at Knossos and the West House Group B material. The decoration of early LM IA sherds is restricted to tortoise-shell ripple pattern and spirals in both deposits while tortoise shell ripple is the dominant motif of the local material in this West House group.

• Group C - The pottery from the pits in bedrock:

In the pits dug into the bedrock under the West House a total of fifteen intact vases have been found and together with some sherds (Fig. 3). All fifteen vases are local, made from the characteristic buff Theran fabric. The following wares may be distinguished:

1. Plain: one conical cup with a ledge rim and squat Cycladic bowls and cups.
2. Cycladic White: panelled cups (Fig. 14b) and one jug.
3. Polychrome (with red and black decoration and white enrichments): one cup in the Cycladic White panelled cup shape but with all-over decoration (Fig. 14a) and three nippled jugs with swallows (Fig. 13).
4. Red Burnished: a tall Cycladic cup (Fig. 12). The forms and motifs are largely of clear Cycladic (Cycladic bowls and cups) or more specifically Theran origin (swallow nippled jugs). Minoan influence is obvious in the conical cup and the zonal decoration of the polychrome cup.

Two of the sherds found with the vases are Minoan in the light-on-dark style. One is a cup rim decorated with white spots, the other is from a larger vase decorated with a polychrome floral pattern. Both belong to Pit WHe2 which yielded two of the swallow jugs but may be earlier than these.

The absence of imported wares prevents straight correlations with Cretan pottery. However, the local vessels seem to be of a late Middle Cycladic date, a conclusion based on comparison with material from the rest of the Cyclades and from Thera. The production of panelled cups, for instance, is thought to have commenced in the Cyclades during the late MC period (Barber 1974, 33; Davis 1970, 218, n. 7). The pit assemblages form a part of a local sequence, the earlier phases of which are provided by material from two other Theran sites, those of Phtellos and the Karageorgis Quarry (Marthari 1987). There is a typological evolution of forms, motifs and decoration techniques across these three sites in which the pit pottery occupies the last stage. For example, polychromy occurs only in the West House vases and not on those from the other two sites.

No room will be given here to the interpretation of these intriguing pits as it is sensible to await results from tests in other houses and open areas before attributing the term foundation deposit or anything else. At the moment their main value lies in supplying a terminus post quem for the earliest construction of the house which can be assigned to the late MC period.

CONCLUSION - THE WEST HOUSE AND THE SETTLEMENTS

The sequence at the West House reflects important events in the history of the settlement (Table 1).

Under floors and fillings of Rooms 3A and 6 cut walls have been found. Their construction technique and orientation leaves no doubt that they belonged to buildings which had existed in the area before the West House was erected. Many such or other types of walls have been occasionally noticed under other buildings as well. They are proof of earlier occupation. Further research is necessary to date these walls and define the extent and plans of these earlier cities.  

The evidence suggests that the West House was founded on earlier walls not before the late MC period. Sporadic finds show that this is also the case for most of the buildings of the town. At a time when the vases of early LM IA style had reached the island, the West House suffered serious damage from seismic destruction in common with much of the site. Extensive rebuilding followed, the products of which are the impressive and sophisticated buildings, with frescoes, visible today. The reconstruction activities were combined in the West House and other houses with the levelling of their interior floors. This seems to have been carried out to facilitate entry into the buildings from the open areas, the level of which had also been raised with debris from the seismic destruction. Most of these floors served the West House and the other structures until their final abandonment at a time when mature LM IA was being imported to the town. This completes the picture of two successive phases, A and B, of the last period of occupation at Akrotiri.

Detailed examination of the buildings has shown that small-scale changes were made during Phase B. Minor earth tremors, which may have been common, could be one reason for them. The inhabitants, probably familiar with such events, seem to have been well prepared for them and carried out repairs. More building repairs and deliberate piling of debris around the West House and other buildings had been started as an organized programme after serious earthquakes, in the period before they abandoned their settlement for the last time. These were never to be completed (Marinatos 1968-76, III, 7-8, 28-31; Doumas 1974, 111-112). The building changes in Phase B seem to have occurred faster than the pottery evolved and it is impossible to date them.

If pottery cannot help to date building changes made over so short a span of time, it is still the best tool we possess to penetrate the darkness of Akrotiri's past. The fragmentary architectural remains under the impressive ones of the latest phase offer only a few data. Frescoes were a recent innovation and stone vessels and tools, even if they have a long tradition, are not as informative as decorated pottery. Ceramic wares related to stratigraphy enable us to follow both internal developments of local culture and external relations during the site's long history.

Each pottery form, with its decoration, has its own specific developmental trajectory. Each form has its origin in one or more previous types, develops its own characteristics, often under foreign influence, reaches a zenith before decaying and finally disappearing. During its life, other types are born, some created from this original which will overlap and continue after its extinction on their own independent trajectory.

The West House material, when considered stratigraphically, demonstrates this organic type of development. It is possible to trace this in the progression of the ceramic types from the Group C pit deposits to the Group A house deposits. In the pits there are two types of the Cycladic cup, plain squat and red-burnished tall (Fig. 12), whereas in the house only the latter survives, with modifications, as one of the most popular forms. In the pits panelled cups (Fig. 3, 14b) coexist with a cup of similar shape but with all-over polychrome decoration (Fig. 3, 14a). The Minoan influence on the latter is obvious. This traditionally Cycladic and Mainland shape survives in a few examples only in the Group A material which are decorated in a Minoan fashion with white reeds on a dark surface (Fig. 3, 14c). The polychrome swallow nippled jugs in the pits (Fig. 13) have an evolution that may be traced via the SDL deposits of other areas into Group A and VDL deposits where they are almost replaced by eyed jugs of Minoan origin, but which are decorated with swallows in the Theran manner (Marinatos 1968-76, VI, Pl. 74a, left).

It is difficult to track these stylistic changes, yet through them it is possible to follow the course of Akrotiri in its last phases of existence. It is a settlement that succeeds in maintaining its identity whilst accepting elements of Minoan origin. The pottery sequence echoes this battle for supremacy between local and foreign elements that is concluded with a peculiar co-existence and balance.

Akrotiri is a Cycladic settlement with permanent habitation from at least the EC period. In the two final phases of its occupation it took the form of a very well-developed urban centre which can only be compared to the dominant urban sites of the first stages of the Neopalatial era on Crete. The close proximity of Thera to Crete must be one of the important determining factors in this development; however their relationship is not an obstacle for the self-determination and individuality in the largest and possibly the most powerful Cycladic town of that time.

------------------------------------------------

------------------------------------------