The site of Old Smyrna lies on a low spur at the foot of the Yamanlar Daği, or Mountain of Unco Things, at a distance of 450 metres east of the present coastline. The spur is 365 metres long on a north-south axis, with a maximum breadth of 250 metres. On the north-east and east its outline is marked by a bank, revetted in large part by a high terrace wall of small stones. The elevated north-west corner of the site is encircled by superimposed terraces leading up to a circular platform or belvedere, which dominates the site, at an altitude of 21·3 m. above sea-level (Plates 2c, 6a, Squares G–Hix). On the west side the spur is bounded by a cart road leading inland to Bornova, and at the south tip it merges into the low-lying pasturage by a sluggish stream. The terracing of the hillock, which in the north-west presents a picturesque aspect, is believed in the main to be the work of a landowner named Turlita in the nineteenth century. A zone along the north side of the site, corresponding in width to the main sector of our excavations there, is planted with olive trees, of which we were obliged to cut down a score; there is also a narrow fringe of olives along the eastern edge of the site. The greater part of the surface of the hillock is divided between two vineyards, the upper on the north and the lower on the south, separated by a bank which is bordered near its east end by a row of fig trees (Squares Nxvi to Gxix); there is also a smaller vineyard of triangular outline on the slope between the upper vineyard and the Bornova road.

The occupational history of the site, like its name Smyrna, goes back beyond Hellenic times. The earliest observed prehistoric habitation, dating to the third millennium B.C., and contemporary and culturally akin to that of the First and Second Cities of Troy, has been encountered only on the rocky core of the peninsula where occupational strata of this era were revealed in a trench dug down the face of the rock (Square Nxiv). Deep soundings at other points (Squares Exii–xiii, Jxviii–xix) yielded no trace of third-millenium occupation, and it seems unlikely that the occupation in this period extended far to the east. The peninsula in fact seems to have been much smaller at that time. The lowest occupation in the trench in Square Jxviii–xix, in the third metre below modern sea-level, seems to be of about the beginning of the second millennium; and since it is unlikely, assuming a fairly steady rate of submergence of the coast (cf. n. 13), that prehistoric occupation could lie much deeper than this, it must have been about the end of the third millennium that the east shore of the peninsula advanced to this point. A considerable upwards slope to westward from this point in early times may be inferred from the fact that a stratum of early Geometric pottery was cut in works of field improvement in 1951 about the 8-metre contour in Squares L–Mxvi (i.e. about 2 metres higher than in Square Jxviii). A similar series of second-millennium levels in Square Exii, not explored to the bottom, attests the growth of the peninsula on the north-east. The gap in time between the second-millennium and the third-millennium levels revealed in these trenches has not been closed, though isolated fragments of pottery found in the course of field improvement north-west of the trench in Square Nxiv may belong to this intermediate phase. The second-millennium occupation, of which a number of successive levels were exposed in the deep soundings, seems perhaps to be more akin to Anatolian than to Aegean cultures. The expansion of the habitable peninsula, assisted by the action of streams flowing from the mountain-side into the embracing arm of the sea, was more rapid in the second millennium than at any other time, and the settlement here in the advanced Bronze Age may have been a not inconsiderable one by the standards of this coast.

Traces of fortifications around the area apparently once occupied by the city of Old Smyrna were observed by Louis Fauvel, and our first detailed description of them is that of Prokesch von Osten, who accompanied him there on a second visit in 1825. As we shall see later, it seems likely, though proof is no longer possible, that most of the circuit wall around the tell, as well as that on the low spur to the west of it on which the modern village now stands, as described by Prokesch, may have belonged to the defences of the classical city. Nothing today survives of these above ground, owing to extensive stone-plundering in the interval; and it is to be feared that the fate of much of this rather exposed classical enceinte has been to provide masonry either for the houses of the modern village or for the terrace walls which today encircle the tell.

The plundering of this outermost circuit probably left the earlier ones inside it rather more exposed to view. I have not been able to verify which of the city walls it was that was photographed by Keil in 1911, but when Franz and Helene Miltner excavated here in 1930 a part of the late-seventh-century B.C. circuit was visible on the east side of the city. Here they cleared about 80 metres of its face, for the most part to no great depth, then picked up its line again with a small probe some 20 metres farther north. Two further small trenches seem to have located more of this late-seventh-century wall-line south-south-west of their long cut, in addition to traces of yet other circuits. Besides this they report sinking two shafts into the mound dominating the north-west corner of the tell and making two small probes in occupation levels within the city itself.

The following catalogue is divided into groups according to stratification. Brief descriptions of each piece and reference to similar vases from other sites are given in the catalogue. The chronological significance of the vases from the destruction level is discussed at the end of Section H.

A.Occupation Phase before the Curved Buildings

1. Plate 21. Fragment of rim of large krater.

Diameter about 0·30.

Very pale clay; dark-brown glaze; on the rim groups of short vertical lines; on the shoulder, panel containing hatched maeander; inside, rather streaky glaze varying from dark to light brown.

Weinberg holds that this type of krater developed during the Late Geometric period, but the term ‘Late Geometric’, as applied to Corinthian pottery, wants further definition. This piece must belong to the first half of the eighth century.

Both high- and low-footed examples are found.

Context: Trench H (Plate 74, Squares E-Gxi-xiii), Room XLI between floors at 8·90–8·40.

The earliest fine Attic pottery (excluding scraps of Geometric ware) found at the site of Old Smyrna was made in the first quarter of the sixth century (see no. 2 in the Catalogue below), although it is not until the second quarter that it appears on the site in any appreciable volume. One of the earliest, and incidentally most complete vases among otherwise very fragmentary material, is a lebes gamikos (no. 1) from the workshop of Sophilos, and it is accompanied by near-contempory lekanai by known painters. From this time on into the first decade of the fifth century fragments of almost every known variety of Attic cup are found, some of the highest quality, and at the end of this period skyphoi and lekythoi also appear while larger vases are conspicuously few. In the first quarter of the fifth century there is a marked falling off in the import of Attic pottery and it is noteworthy that no vases later than the first decade of that century were found in the main excavated area of the sixth-century houses at the north end of the site, and only isolated fragments from Trench C to the West. The possible historical significance of this is pointed out elsewhere by Mr. Cook. By the middle of the fifth century import is resumed on a small scale, but it is of high quality, including as it does a volute crater by the Niobid Painter (no. 118). The volume increases to the beginning of the fourth century, dying away again by the middle of it, and in this period fine black pottery, some of it with impressed decoration, appears beside the figured vases. The Attic black pottery of the fourth century is intimately associated with its numerous Ionic imitations and the presence of some of the commoner shapes is noted in the publication of the Ionic ware in a later volume of this Annual, to which the reader is referred.

In the summer of 1951 N. Dadoudis, while digging a deep trench to bury stones from a plot of land belonging to him on the east edge of Makritikhos village, about 250 metres north of the Palace of Minos, found a complete Minoan amphora (Plan, Fig. 5, no. 3). He reported this discovery to the Ephor of Antiquities for Crete, Dr. N. Platon, who invited Piet de Jong, then the School's Curator at Knossos, to examine the area. Excavation revealed a small room, against the east wall of which the amphora had evidently once stood. The room was cleared by Piet de Jong assisted later by Sinclair Hood. To judge from the character of the vases found in it, the room might have been used as a kitchen.

The ground on the edge of Makritikhos village here slopes steeply down through a series of terraces to the bed of the Kairatos stream about 40 metres to the east. The ‘Kitchen’ lay just below a high bank, forming the western boundary of the plot of ground, with the natural rock exposed at the north end and Minoan house walls showing in it to the south. There seems to have been a marked slope down towards the north as well as to the east here in Minoan times. The wall a–a (Fig. 2) at the north end of the original trench dug by Dadoudis, of squared limestone blocks measuring up to about half a metre in length and 0–35 thick, lay at a lower level than the Kitchen, although it appeared to be of the same period with it. A roughly constructed wall b–b south of a–a may have supported a terrace marking this change in ground level. The corner d–d of another, presumably contemporary, house built of squared limestone blocks was exposed in the south part of the original trench.

The existence of Late Minoan tombs somewhere in this area was noted by Evans, and several of those found during our excavations had been partly destroyed or plundered. Nineteen tombs were discovered in all. One of them (XVIII) was a Middle Minoan tomb (see p. 220). The rest (with the possible exception of tomb VIII which was found empty, and which from its shape and character may be Middle Minoan) all belonged to the Late Minoan III period, from the time of the destruction of the Late Minoan II Palace about 1400 B.C. (tombs I, II, XV), till the end of Late Minoan times (tomb VII), when iron had begun to come into use (c. ?1150 B.C.), a range of some 250 years or more. It is possible that tomb I, with its finely decorated clay vases and interesting bronze knives and razors, may go back within the limits of the early part of Late Minoan IIIA (Furumark's IIIA 1) before the destruction of the Palace.

Tomb I (Plan, Fig. 2; Plate 49b). This was the original tomb whose collapse led to the discovery of the cemetery. It was also apparently the earliest in point of time, containing four clay vases (1–4) decorated in a fine style which may go back before the destruction of the Last Palace. The main chamber was more or less rectangular (L. 1·90, W. about 2·50), approached by a level dromos (L. 3·00, W. 0·85–90), on the left side of which was a walled recess (L. 1·78, W. 0·75) exactly like the chamber in graves of the pit-cave type as found in the Zafer Papoura cemetery (PT 15 f., and fig. 11a, b). The dromos was entered by an unusually steep ramp descending at an angle of nearly 45 degrees.

The tomb was quite intact and unplundered. But the roof of the main chamber had partly collapsed, together with the top of the entrance, so that the upper part of the blocking wall had fallen into the chamber. The solitary burial had been contained in a wooden coffin, set against the south wall of the chamber, of which traces survived together with some indications of the blue paint with which it had been decorated (Plate 49b). The dusky patch indicating the decayed wood of the coffin lay above the floor of the tomb to a depth of between 5 and 10 centimetres over an area of about L. 1·50 m. and W. 0·50 m. The coffin traces began immediately on the floor over most of the area where they occurred, but in some places they were a little above it.

The numerous pots found in the Gypsades cemetery are interesting both for their range in time and for the evidence they provide concerning the occupation of Knossos in the second half of the Late Minoan period. The earliest pots found, the L.M. IIIA vases, are the natural successors of the Palace Style at Knossos, and the Sub-Minoan vases are the precursors of the earliest pottery in the Fortetsa cemetery, where the ceramic development into Protogeometric may be followed. The Late Minoan pottery from the tombs amounted to thirty whole vases of twelve basic shapes. By far the most common vase was the stirrup jar, of which most examples belonged to the end of the L.M. IIIB period. Of the twelve examples in the collection six (VII. 3, 4, 5, 6, 7, 8) came from tomb VII, the latest tomb; with them were found the two belly-handled amphorae (VII. 1 and 2), which appear late both in Crete and on the mainland of Greece. Only one stirrup jar (I. 4) earlier than L.M. IIIB in date was found. None of the other shapes was represented by more than two vases. The absence from the cemetery of the tall-stemmed kylix and of the deep bowl, which are common in Mycenaean IIIC tombs of Mainland Greece, is noteworthy.

Some sixteen clay coffins or larnakes of the well-known Late Minoan type with gabled lids were recovered from nine tombs (IV, VI, VII in the first group, all the tombs IX–XIV of the second, and XVI of the third). They were made of coarse gritty pithos fabric, with the surface orange, paler orange or buff, or greenish, due to wet-smoothing or the application of a thin slip. The chests were between 0·75 and 1·20 long, standing to a height of about 0·50–75 including the legs but without the lids. One larnax (larnax II from tomb VI) had never apparently had any legs. One very small larnax (larnax II from tomb XII), only 0·67 long, was evidently for a child (Fig. 24aa). The sides normally had panels, inlet or suggested by incised lines, no doubt imitating panels of the wooden originals from which the larnakes were clearly copied. There were normally four vertical handles on the chest, two each side, and corresponding handles on the lid. But in some cases (e.g. larnax IV from tomb VI and that from tomb VII) the handles were replaced by perforations for ropes through the corners of the chest. The lids of these larnakes had curious small ribbed handles set on the four corners.

All the larnakes were plain and unornamented, except for that from tomb VII, with a simple decoration of straight and wavy stripes in red paint; together with larnax IV from tomb VI, and the fine example unfortunately much destroyed from tomb XIII, which both had designs of spirals and papyrus flowers of a type classified by Furumark as Late Minoan IIIA 2.

The Knives. Three complete one-edged knives, and one fragmentary, were found in the tombs. The bronze knive (I. 5) belongs to a small group in which the whole of the handle including a T-shaped pommel extension was cast in one piece with the blade, while inset panels of perishable material were secured by rivets and deep flanges in the metal. Apart from this pommel-flange it is similar to a large contemporary group of bronze knives to be found throughout the Aegean.

One-edged knives with flanged handles are at least as old as the Shaft-grave period at Mycenae. In Crete a riveted handle without flanges was preferred from Middle Minoan times, but the flanged handle appears at the end of Late Minoan II or the beginning of IIIA. The pommel-flange is rare, though found occasionally on both riveted and unriveted knives. The earliest dated example is probably that from the Palace of Mycenae found with other objects of Late Helladic II or early III date; one comes from Dendra, Chamber Tomb 7, and one unriveted from Chamber Tomb 2, dated L.H. IIIB. The Mycenae knife has not the true T-projection but only a broadening at the end of the handle which is in shape very like handles of Levantine daggers and two-edged knives found in the so-called ‘Hyksos tombs’ of Ras Shamra and at Tell el Ajjul, Tell Fara, and elsewhere in Palestine.

It has long been axiomatic that amber is seldom found in Crete, and it is therefore of some interest to be able to record amber beads in two of the Gypsades tombs, Nos. II and VII. In Mycenaean sites on the mainland of Greece amber beads of different shapes are found in great numbers and so frequently as to appear a fairly constant feature of these sites; certainly as much as faience and rather more so than lapis lazuli, an exotic from Asia.

In the shaft-graves of both the new and old grave circles at Mycenae amber is plentiful and thereafter at Prosymna in graves dating from Late Helladic I to Late Helladic III; and this is characteristic of all the principal sites, though the number of beads found may be much smaller. Amber is also particularly plentiful on the west coast of the Peloponnese: at Kakovatos in Elis, in tholos A (dated L.H. IIA); at Epano Englianos in western Messenia, in the tholos excavated by Lord William Taylour, close to the palace, and dated to the fifteenth century or earlier, in which about 360 beads were recovered; and in the second tholos recently excavated by Professor Marinatos at Myrsinochorion in the same area. Farther north there was much in the Ionian Islands, particularly in Cephalonia; and a little has been found in Epirus.

I. 1. Rhyton (Fig. 26, Plate 54c). Ht. 0·364. Ht. incl. handle 0·405. Diam. at tip 0·104. Diam. of spout 0·02. Broken but complete; the paint is slightly worn and the rim chipped. The painting is in orange and brown, with linear decoration on the lower part of the cone and interlocking zigzags above. The lip and the outer edges of the handle are outlined. The handle is of the raised metallic type and the lip overhangs slightly; the tip of the spout is offset a little to the rear. The buff clay without slip forms the ground for the decoration.

The shape of the vase corresponds with Furumark's Type 199 (MP 71–72). The rhyton may be compared with the example from Palaikastro (Palaikastro 104 fig. 87a) and belongs to the L.M. IIIA 1 phase. For the zigzags cf. MP 182.

I. 2. Piriform Jug (Fig. 26, Plate 54a, d). Ht. 0·29. Max. diam. 0·18. Diam. of base 0·06. Broken but complete, with well preserved black and brown decoration. The base is splaying and flat. A repeated S-pattern is incised at the base of the neck moulding. The lip is cut away above the handle. Argonauts are arranged in series around the shoulder, and a fish has been painted below the base of the handle. Bands around the neck join to form a chevron at the front. The edges of the handle are outlined. The argonaut is a development of the L.M. IA type and lacks tentacles.

The amber beads were examined with a view to determining the origin of the amber—in particular to distinguish between Prussian and Sicilian amber. The samples were compared with specimens of Prussian and Sicilian amber of authenticated origin as controls.

A. Appearance under ultra-violet light

All the samples and the controls showed the same superficial dull-yellow fluorescence except samples 1 and 6, which were darkish brown in colour. When a fresh surface was exposed by removal of the outer brittle crust, all samples showed a greyish-blue fluorescence. Finally, when the samples (with the exception of no. 6) were cut in half, the central areas showed a milky white fluorescence. This area was particularly prominent in the case of the Sicilian sample, in which the greyish-blue fluorescence was confined to a very thin zone.

The phenomena observed in the case both of samples and of the control specimens were so similar that no conclusions could be drawn about the origin of the amber beads.

In 1955 it was decided to transfer the School's undersea activity from Chios to Crete during the time that excavations on land were being continued at Knossos. The Greek Ministry of Education most generously granted permission for the School to carry on diving operations at selected points off the coast of Crete in co-operation with the Ephor of Antiquities, Dr. Nicolas Platon.

The campaign lasted three weeks, from August 9th to 30th, with a party of ten volunteers under the leadership of John Leatham. The sites explored were chosen by Dr. Platon, who gave the party much of his time and interest, and made repeated visits to the places where they were working. The party had its headquarters in an empty house which belonged to the School's foreman, Manoli Markoyiannakis, conveniently situated upon the main road in the suburb of Poros (Katsamba) just east of Herakleion. This was used as a central depot, and the compressor for filling the aqualungs was established here. The work was unfortunately hampered throughout the three weeks by high seas owing to the prevailing north winds (Meltemia). It was difficult therefore to make great use of the aqualungs.

The slopes of Ailias beyond the Kairatos stream to the east of the Minoan city have so far largely escaped the intensive vineyard cultivation that has swallowed so much of the land round Knossos. Isolated tombs and cemeteries (e.g. Knossos Survey 58, 95, 96) have been explored on these slopes due east and north of the Palace of Minos, but virtually nothing has yet been recorded from the area to the south. Minoan sherds on the surface of the fields opposite the Temple Tomb, however, suggest the possible existence of tombs in this area.

In 1951 on the lower slopes here on land belonging to Evstratios Sarikis a hole was dug to plant an olive tree, leading to the removal of some large stones and disclosing a right-angled cut in the rock, which was noticed by the sharp and practised eye of Spiro Vasilakis. Two years later with the permission of Dr. Platon, Ephor of Antiquities for Crete, I cleared the cutting during the course of the School's excavations in the Middle Minoan cemetery higher on the slopes of Ailias to the north. The cutting, which was rectangular, measuring 1·90 × 1·40 at the bottom, and 1·20 deep from the surface of the rock at the highest point in the south-east corner, is perhaps best explained as a plundered Minoan shaft-grave, although no sign of a burial or of any grave goods was found in it. It was entirely filled below the level of the rock surface with large blocks of the local limestone, several of them worked, and including a slab which might have served as the covering for a shaft-grave, and a pyramid with a square socket in the top (A on the plan, Fig. I ), evidently the base for some ritual object like a double axe.

What seemed to be the covering slabs of a tomb were noticed by the School's foreman, Manoli Markoyiannakis, exposed in the low field bank which forms the southern continuation of the high bank with the Geometric tombs dug by Hogarth and Payne, a couple of metres north of the path running eastwards across the base of the Kefala ridge past the Hellenic tower (BSA 52 (1957) 244ff.). During the course of the School's excavation of the early Christian building on the neighbouring Sanatorium site in 1953 Dr. N. Platon, Ephor of Antiquities for Crete, kindly gave me permission to examine the slabs, which were found to cover a grave roughly rectangular in shape and measuring 1·60 × 0·70 at the bottom (Fig. 1). The grave's floor was only 1·15 at the west end and 1·60 at the east below the modern surface of the rock; but the surface here must have once stood much higher, and have worn away owing to erosion.

The three large slabs over the grave were blocks of dressed limestone. That at the east end (A) had a ‘branch’ sign (L. 0·25, max. W. 0·13), boldly carved with broad shallow V-shaped grooves, on the upper exposed face in the north-east corner (Fig. 1, Plate 66d). The joints between the three slabs were carefully wedged with small stones. In the grave below them was a clean fill of kouskouras, the soft white chalky rock of the area, containing a few nondescript Minoan sherds. The cover slabs with the smaller stones wedged in the joints between them seemed to be in place; but the earth below contained small lumps of rock, as if the grave had been deliberately filled before the cover slabs were laid in position. This agrees with what has been observed in the case of other Minoan shaft-graves in the Gypsades cemetery (see p. 219). At the bottom of the grave were the scanty remains of a skeleton, lying on its back with the knees flexed and originally perhaps raised in the usual manner (see p. 218). Although the cover slabs appeared to be in position, and the grave undisturbed, nothing was found with the skeleton. The grave is, however, like other similar shaft-graves at Knossos, presumably Late Minoan, and perhaps early rather than late in the period. The cover slabs may well be old building blocks, and they and the ‘branch’ sign carved on one of them may therefore be considerably earlier than the grave.

Among the local finds displayed in the Museum of Antiquities at the Piraeus are six Roman portraits, all of Pentelic marble and all unrestored. Two of them, a colossal head of Trajan (no. 270) and a colossal statue of Balbinus (no. 278), have already been fully edited. But according to the Director of the Piraeus Museum, Dr. Threpsiades, and to the best of the present writer's knowledge, of the other portraits two are still unpublished, two described without any illustration; and it is with Dr. Threpsiades' kind permission that all four are published here.

I. No. E 4. Head of Claudius(Plate 67a–b)

Total height: 48 cm.

Height from crown of head to bottom of chin: 26 cm.

Width across at greatest extent: 20 cm.

Width from back to front at greatest extent: 24 cm.

The sculpture depicts the head and neck, slightly over life-size, of an elderly man, in whom we can immediately recognize the Emperor Claudius. At the base of the neck is a rounded ‘tenon’, designed for insertion into a cavity between the shoulders of the now vanished body of a full-length statue. Claudius' face has sustained considerable damage. The nose has practically gone, only part of the side, and the hollow interior, of the right nostril remaining. The chin, lips, and right ear are bruised, the brow is marred by several abrasures, and the left ear is lost.

The site to be described lies on a low hill a little to the north of the small village or suburb of Nemesis, about 150 metres to the west of the main road from Patisia to Koukouvaounes, and 1 kilometre to east-south-east of the Mycenaean tholos tomb at Λυκὸτρυπα, which was excavated by the German Institute in 1879, and is usually called the Menidi Tomb. The site at Nemesis is visible from the tholos tomb, and is separated from it by a gentle valley through which run, in a southerly direction, two streams with steep banks. The eastern stream is the river Kephissos, whose name goes back at least as far as the classical period.

The hill of Nemesis stands about 15 to 20 metres above the level of the surrounding land, and measures about 160 m. north-west to south-east × 120 m. north-east to south-west. The hill is an isolated outcrop of conglomerate rock, thinly covered with stony brown earth. It has been eroded over an area about 250 m. north-south × 50 m. east-west, so that its original size was considerably larger than at present, in all about 30,000 square metres. Mycenaean sherds were found over the whole of this area, though mainly in the eroded part, among the lumps of fallen earth and rock. Remains of rubble walling together with several Mycenaean sherds were found here, and also in the steep cliffs formed by the erosion on the west and south sides (this part of the hill has been undermined by recent excavation of the beds of grey clay, which here lie at between 2 and 3 metres below the original ground level). The ancient remains are particularly noticeable in the south-west angle of the cliffs (roughly in the centre of the part of the hill shown on Plate 71a), where there is a greater depth of earth above the rock than is visible elsewhere on the hill.

Pindakas is a small and low hill lying about 1·5 km. from the harbour at Emporio in South Chios. It is at the foot of the hills which divide the modern town of Pyrgi from the Emporio valley, but it also commands access to the adjacent valley to the north, that leading to modern Kalamoti (see Figs, 1 and 2). The position is, however, one which is better called convenient than strategic. The name must derive from πίδαξ, ‘spring’ or ‘fountain’, and, although there is no water on the hill itself, an excellent well lies close to its slopes on the west. Ano- and Kato-Pindakas have been distinguished, but the distinction is lost today, and Ano-Pindakas, rather nearer Emporio, is as bare of human habitation as Kato-Pindakas, which is our site.

Today the flat top of the hill is bare and rocky with scattered olive trees, though the fields around are rich in corn and mastika. Zolotas had noted antiquities there but it is the great polygonal walls at the west which make two terraces of the hill-top that are the most conspicuous remains (Plate 72a–b). It was on these two terraces that excavations were conducted by the writer in July 1954, in the course of the British School at Athens' excavations at Emporio. Sections were cut east–west through the upper and lower terraces, and north–south through the upper terrace (Figs. 3 and 4, AA and BB) and these, with minor trenches and some ground observation, tell a clear story of the site's history and buildings.