Minoan Paintings in Avaris, Egypt

It dates to the late Hyksos period and was reoccupied by the early Eighteenth Dynasty. From the Hyksos period we have evidence of a massive fortification wall with buttresses and monuments with royal inscriptions (Bietak 1981; Görg 1981; Bietak, Hein et al. 1994, nos. 126, 130, 131, 133). Especially interesting is the jamb of a monumental doorway with the full titles of the hitherto unknown Hyksos Seker-her = Sikru-Haddu ('Memory of Hadad' according to an interpretation of T. Schneider (1994, 275)). Equally interesting are remains of gardens and possibly vineyards within the citadel, which remind us of the gardens of the Hyksos Apophis, mentioned in the second Kamose stela (Habachi 1972, 36).

This communication focuses, however, on the vestiges of a citadel of the early Eighteenth Dynasty. Using the fortified area of the Hyksos citadel the early Eighteenth Dynasty built new monumental installations. The most important ones are a platform construction of mudbrick, measuring 130 x 90 cubits (70.5 x 47 m.), which bears a close resemblance to the so-called 'Southern Palace' at Deir el-Ballas (Stevenson Smith 1958, 156-159, figs. 51-52; Lacovara 1990, 5, 26, 30, 40, pl. VII; 1993, 27), and an even larger palatial compound, not yet completely excavated, but which already shows close similarities to elements of the 'Northern Palace' at Deir el-Ballas (Bietak 1999, figs. 10, 15, 19, pl. II; cf. Stevenson Smith 1958, 156-159, fig. 52; Lacovara 1990, 2-3, 28, 31; 1993, 27).

The close resemblance of the installations of the early Eighteenth Dynasty at Avaris to the 'Northern' and 'Southern Palaces' at Deir el-Ballas (Stevenson Smith 1958, 156-159, figs. 51-52; Lacovara 1990, 2-3, 28, 31) is an issue of some importance. Also important to this discussion is their near contemporaneity. According to Peter Lacovara, Deir el-Ballas should be considered as a campaign residence of the late Seventeenth Dynasty during their warfare against the Hyksos. It was abandoned by Ahmose as Avaris was taken (Lacovara 1993, 27). This new discovery at Avaris suggests that a new residence was built for the king, probably as the headquarters for his subsequent campaigns against Sharuhen (most probably Tell el-'Ajjul south of Gaza) and other strongholds of the Hyksos in southern Palestine.

The platform construction, of which only the foundations are preserved, reveals an inner division in casemate fashion similar to that of the 'Southern Palace' of Deir el-Ballas. According to the division of space, we can recognise a large central court and staircases near the north-eastern, north-western and south-western corners. The building cuts into the fortification walls of the late Hyksos period, and was accessible by a ramp, 6.40 m. (12 cubits) wide, which was attached to the eastern flank of the platform. This ramp must have cut through the Hyksos fortification wall without necessarily destroying it (Bietak 1996a, 67-72; 1996b, 8). Blocks belonging to a granite doorway of King Amenemhet I, found about one hundred years ago by E. Naville (Jánosi 1994, 22-27, figs. 5, 6), were probably set up secondarily at the start of the ramp (Bietak 1999, pl. I:b). In view of the parallel provided by the 'Southern Palace' at Deir el-Ballas, which still stands about 8 m. high and was identified by P. Lacovara as a monumental watch tower (Lacovara 1993, 27), our platform most probably also served as a similar substructure, which can be reconstructed as a military building (Bietak 1999, fig. 16). The position of the stairways suggests corner towers. The dense gridwork of foundation walls in the western part of the building looks like the foundation of a columned hall. In parallel to the 'Southern Palace' at Deir el-Ballas, the palatial character of the platform building can also be recognised by fragments of column bases and lintels with torus and cornice, a limestone sculpture of a lion (Bietak et al. 1994, pls. 5a, 6b) as well as fragments of a magnificent glass vessel, so far the oldest found in Egypt and most probably of Mesopotamian origin (Hein in Bietak, Hein et al. 1994, no. 323).

To the east of the platform were found numerous dumps of wall plaster with Minoan wall paintings, which are discussed below. Most probably they originate from this building. Similar military buildings constructed on raised platforms of this sort were found on top of ramparts at Middle Bronze Age Ebla and Gezer (Kempinski 1992, 132-133, figs. 13, 14). They provided excellent views over the environment and the defence systems of towns.

The date of the construction of the platform can be narrowed down to some extent. On the one hand, it cuts into the defence walls of the late Hyksos period (see above); on the other hand, it conforms to the orientation of buildings of the Eighteenth Dynasty. Besides this evidence, we have the remains of a humble settlement of mudbrick attached to the east of the ramp. From this settlement (H/I) numerous scarabs, many of them with inscribed royal names, were retrieved. Their stratigraphic position

appears to accord with the succession of the kings (Jánosi 1994, 32f.). They range from the time of Ahmose to Amenophis II (Hüttner in Bietak, Hein et al. 1994, nos. 324-338; Jánosi 1994, pl. 7a; Bietak 1996a, pl. 31a). As scarabs often remain in use for a long time they should be used only as termini post quos. A more reliable means of dating is provided by the pottery from this settlement, which seems to originate from the time of Tuthmosis II-III according to Irmgard Hein (study in progress).

The compound of the main palace H/II-III has not yet been fully explored. It is oriented parallel to the river and extended for at least 150 m. from east to west (local north to south). The palace was surrounded by an enclosure wall with a portico entrance in the north-eastern corner where there are many remains of painted plaster. An entrance to the palace itself was found continuing from the portico at its north-eastern corner, leading into a vestibule with two washing installations. Evidently nobody was admitted without a proper bath.

Further to the north-west is a secondary entrance leading into a long corridor, probably a staircase. Towards the south-west we encountered long corridor-like magazines. The walls were 4.3 m. (10 bricks) wide and left only limited space for rooms. From a constructional point of view, one must expect the more impressive state rooms to be in an upper storey.

Against the outside of the north-eastern enclosure walls were found workshops which produced, among other things, numerous finds of stone missiles for slings. This can be taken as evidence for the military function of the citadel. Besides those within the magazines of the palace, arrow tips of bone, flint and copper appeared in some quantities along the enclosure wall and in the area north-east of the platform H/I (Schwab and Tillmann in Bietak, Hein et al. 1994, nos. 344-349). The typology of the flint tips points towards the Sudanese Kerma culture (Tillmann in Bietak, Hein et al. 1994, 348, 349), and indeed pottery of the Kerma culture, with Kerma beakers and also household ware, was retrieved at several spots within the citadel. The conclusion that Nubian archers were stationed here seems justified, as we know that the early Eighteenth Dynasty engaged in warfare in Nubia against the kingdom of Kush, first recapturing Lower Nubia and finally conquering this powerful African kingdom. Such events gave the opportunity for recruiting archers from prisoners of war. ⁽²⁾

In the area north-east of the palatial building H/II-III and the above mentioned workshops, numerous postholes were discovered which look like the remains of tents erected again and again on this site. It seems not unlikely that these are the remains of temporary shelters for troops.

A surprise were two bundles of arrowheads with indented base, made of copper or bronze (Schwab in Bietak, Hein et al. 1994, no. 347). Remains of fibres from a bag which contained one of the bundles were still visible. Both were found along an enclosure wall of the eastern extension of the palace of the early Eighteenth Dynasty. The typology of these arrowheads appears very similar to that of Late Helladic arrowheads (Demakopoulou et al. 1990, no. 163). A typological link of this sort is not impossible, as other connections with the Aegean may also be recognised in the find material. There is the neck of an amphora with oval mouth of Trickle Decorated ware (MM III/LM IA). ⁽³⁾ On another amphoriskos

we have the painting of a leopard in flying gallop, chasing an ungulate (Hein in Bietak, Hein et al. 1994, no. 359). The handles are outlined in the pattern of the Aegean figure-of-eight shields. While the amphoriskos is probably of Levantine origin, the painting looks definitely Aegean. In areas H/I, II and III, red burnished rhyta were found, which were made locally after LM I prototypes (Hein in Bietak, Hein et al. 1994, no. 314).

In the magazines of the big palatial structure of area H/II were found large amounts of broken pottery, among them Late Cypriot Bichrome ware, White Slip I ware, Red Lustrous Wheelmade ware and Base Ring I ware (Hein in Bietak, Hein et al. 1994, no. 359). Evidently under Ahmose or his successors the intensive trade relationship with Cyprus of the late Hyksos period continued unbroken, and there is every reason to believe that the import of copper from that island accompanied other commodities sent in the pottery containers. ⁽⁴⁾ The same conclusion can be reached from the imports from Canaan, primarily amphorae (Hein 1994). The difference in comparison to the material obtained from the late Hyksos period lies in the fact that a much wider variety of fabrics now appeared, which shows that the trading network had become much more complicated and varied than before. In the late Hyksos period nearly all imports from the Levant came from southern Palestine (McGovern 1999). In the Eighteenth Dynasty a much wider spectrum of trading patterns can be recognised, and this conclusion fits well with the contacts with the Aegean now emerging.

The most puzzling finds from the citadel of the early Eighteenth Dynasty were, however, thousands of fragments of wall plaster with Minoan wall paintings (preliminary reports: Bietak 1992; 1994a; 1994b; 1995; 1996b; 1997; Bietak and Marinatos in Bietak, Hein et al. 1994, 50-52, nos. 219-230; Bietak and Marinatos 1995; Marinatos 1998; Marinatos forthcoming; Morgan 1995; Palyvou forthcoming; first final publication: Bietak, Marinatos and Palyvou forthcoming; see also Bietak, Marinatos and Palyvou this volume).

The paintings were found in connection with at least two building compounds. The most interesting ones appeared in debris east of the platform H/I and it is possible that they were dumped from there (for the stratigraphic situation see Jánosi 1994, 32; 1995b; a new assessment in Bietak 1996a; 1996b and forthcoming; Bietak, Marinatos and Palyvou forthcoming). ⁽⁵⁾ Other paintings were uncovered in the palatial area of H/II-III, and it is possible that the dumps in H/I originate from here. In the eastward extension of the palace the paintings were found partly in situ on the walls, partly half in situ fallen to the ground at the foot of the walls. A particular concentration showed up around a doorway. Among those immediately restorable, ornamental designs of ivy and loop patterns, and also figural motifs, such as the feet of a large female representation with two anklets, were observed. Besides figural and ornamental paintings, numerous plaster fragments from both findspots of frescoes (H/I and H/II-III) showed patterns imitating ashlar masonry with thick reddish brown horizontal and vertical zones, probably giving the effect of wooden beam elements in the walls. It is obvious that an attempt was made to produce the illusion of Aegean walls in some parts of the palace, although the layout of the architecture and the building materials were Egyptian.

It is difficult to recognise and conserve plaster fragments with paintings which have lain for a long time in moist soil, under perennial irrigation for over a century. Due to recent brick pits to the south of platform H/I a large part of the original paintings can no longer be found. The reconstruction of the original wall programme in this sector is therefore limited.

The mixed technique of buon fresco and tempera painting, executed on a polished lime plaster surface, is particularly Minoan (see Seeber this volume). While lime plaster is also known in Egypt, but without the high degree of polishing, wall paintings are more commonly executed there on gypsum plaster in pure secco technique (see El Goresy this volume). The planning of borders and patterns with string lines impressed on the still wet surface is also typical of Minoan art. So is the choice of colours, such as, for example, blue for grey in the rendering of horns and hooves of bulls in blue, ⁽⁶⁾ or the shaved scalps of youths - a remarkable iconographic parallel for the representation of boys and girls in the wall paintings of Thera (Davis 1986). The details in the painting of griffins, including the pattern and colour spots on their wings, also show close parallels to the griffin behind the goddess in Xeste 3 on Thera (Doumas 1992, figs. 122, 128 (Pl. 12)). This does not mean that the artists came from Thera, since other Thera conventions, such as painting figures primarily against a light background, cannot so far be observed at Tell el-Dab'a. The iconographical similarities should rather be seen as evidence that our paintings belong approximately to the same period as those from Thera (LM IA). ⁽⁷⁾ The quality of the plaster and the themes point rather towards Minoan palatial centres, in particular to Knossos, as the origin of the mural paintings at Avaris.

A major surprise was the discovery of fragments with bull leaping scenes -a motif restricted in Minoan wall paintings so far to the palace of Knossos. Such scenes at Knossos, however, date primarily to the LM IIIA period, while the paintings from Tell el-Dab'a, according to their style, iconography and context, date to the LM I period. Only small quantities of early paintings, dating possibly to MM III/LM I are preserved in the major palatial centre of the Minoan world. This makes our discovery particularly valuable.

Of great interest are the remains of a tableau with at least four bulls, two blue speckled and two reddish yellow speckled ones against the background of a maze pattern (see Bietak, Marinatos and Palyvou this volume, fig. 1). The maze has a more elaborate parallel from the excavations of Sir Arthur Evans at the palace of Knossos (Evans 1921, fig. 256). This is, however, the first time that representations of bulls appear in combination with a maze pattern. This should not encourage speculation about the sagas of the labyrinth. The pattern can be seen as an ornament rendering of the pavement of a court on which bull games were performed (Shaw 1995, 108). Near the base of this tableau a half-rosette and triglyph frieze can be reconstructed, which should be seen as an emblem of the palace, most probably of the palace of Knossos itself (convenient summary in Hallager 1985, 18, fig. 21; Immerwahr 1990, 193, 199,204; Press 1967, 101, fig. 44/103, fig. 45/214, fig. 92; for function, see Hägg 1987). Only later was it taken over by Mycenaean palaces such as that of Pylos. This motif appears as an architectural decoration on stone friezes, as found in the West Court of the palace of Knossos. It has been plausibly suggested that the half-rosette frieze should be identified as an emblematic architectural decoration under the 'window of appearance' of the palace, situated in a prominent position facing the West Court of Knossos (Hägg 1987, 129-134).

The upper end of the maze pattern is in the form of an outline of an undulating hill silhouette against a red background, indicating that the arena bordered open country. The hooves of the hindlegs of the bulls were placed on a ground line, which is perhaps the upper part of the half-rosette frieze; one or two other bulls were placed partly against the hilly silhouette and the red background. The question is, what was shown on this tableau?

To this, the answer is acrobatic scenes in connection with bulls. One of the largest fragments of the wall plaster shows a yellow skinned bull leaper grabbing the neck of a bull which is shown en face. He is evidently a young leaper, because he has partly shaved hair, showing the blue scalp of a youth (Davis 1986). His yellow skin is unusual for Minoan art, but the youngest of the Xeste 3 boys from Thera also has yellow skin (Doumas 1992, figs. 109, 112). As only a few paintings have survived from this period in Crete, it may have been a colour convention which has hitherto escaped recognition by scholars. Yellow may mean very young.

Although the fragment under discussion is not well preserved, it is obvious that one of the best artists of his kind was at work here. Not only is the drawing superb, but the most minute details, like the eye of the leaper and the strands of hair hanging from his temples, were executed in a fine naturalistic style. Of special interest is the representation of his jewellery. An armlet in blue with a yellow running spiral decoration is most probably a rendering of a silver armlet with a gold and niello spiral inlay. On his wrist one recognises a cushion-shaped seal on a leather string. Even the indentations at both sides of the pierced seal, caused through wear and tear by the leather strap, have been rendered by the artist. Only artists living in the Minoan world could be familiar with such minute details. A similar amazing treatment can be recognised in the paintings of the bulls with their speckled skins, rendered in a whole spectrum of shades.

Bulls en face are only very rarely represented in painting (Evans 1930, 208, fig. 142), but they are well known from other works of art such as seals and seal impressions (Morgan 1995, 43 n.140), and from the Vapheio cup A (Davis 1977), which in several respects provides a good comparandum for our bull and maze tableau. For example, the bull hunting scene on the Vapheio cup is accompanied by palm trees, perhaps considered as holy trees and as symbols of sacrifice (Marinatos 1984).

Likewise, our bull and maze tableau is also associated with at least one and probably more palm tree representations. One palm tree defines the left margin of the tableau and connects it with another bull leaping and bull grappling frieze, painted by another artist in slightly less careful style. There is evidence again in this scene of at least four bull representations associated with taureadors. Two bulls are again reddish yellow speckled, two have black speckles against a whitish skin. It seems that a taureador frieze can be reconstructed along similar lines to the famous taureador frieze from the palace of Knossos, only in this case less than half its size. The best preserved fragment shows a side leap (Bietak 1999, pl. VII:a). Several other fragments belong to a black speckled bull, probably with a falling leaper. The last scene which joins the bull and maze tableau shows a bull with collapsed forelegs, looking upwards with his tongue hanging out of his mouth (Bietak 1995, pl. 2.2; see Bietak, Marinatos and Palyvou this volume). There are two taureadors in this scene, one in front of the animal and the other on top of him. Both seem to have wrestled the bull down. It is probably a scene from the end of a bull leaping event, or possibly shows the capture of the bull for use in a bull leaping ceremony.

As at Knossos, we also have evidence of plaster reliefs of bulls and other horned animals at Tell el-Dab'a, found near the platform construction, east of the ramp and also in the other two excavation areas H/II and III. There is also a fragment of a white skinned human representation in life size against a red background with some vegetation (Bietak and Marinatos 1995, 54, fig. 5). This is reminiscent of the so-called 'Prince of the Lilies' from Knossos. All this should be seen as strong evidence of the Knossian, or at least Minoan, origin of the paintings at Avaris (Morgan 1995, 40-44).

Besides the bull scenes, fragments of floor acrobats with plumed headdresses have also been retrieved, performing beside palm trees (Bietak 1994a, pl. 17b; Bietak 1995, pl. 3A). Such representations are familiar from seals and seal impressions (Evans 1935, figs. 443, 444). They probably relate to festivities connected with the bull leaping events. The attitude of the acrobats is identical to such an extent that we may assume the use of templates. A similar observation has been made in connection with the bull drawings of the bull and maze tableau (see Bietak, Marinatos and Palyvou this volume).

Of special interest are hunting scenes, depicting hunters with dogs on leashes, in running motion (Bietak and Marinatos 1995, 55, fig. 6; Bietak 1997, fig. 4.32; Marinatos forthcoming). Another fragment of such a scene shows a white dog with a red collar biting into the flanks of a mountain goat or similar ungulate which is running away (Bietak 1995, pl. 4.1; Marinatos forthcoming). Besides human hunters we also have representations of lions and leopards chasing fallow deer (Bietak 1999, pl. VIII:a) and mountain goats (Bietak 1994a, pl. 19a; Marinatos 1998).

It is not clear if the fragments retrieved of this genre belong to one and the same scene or if there were several scenes like this. Such hunting scenes are known from Kea (Morgan forthcoming), while hunting felines are represented on the ship frieze on the south wall and on the exotic landscape on the east wall of the West House of Thera (Doumas 1992, figs. 33, 36 (Pl. 2, 3-3.40 m., Pl. 3, 0-1.80 m.)). They fit into the ideology of the hierarchy in nature in the Minoan world.

There are other themes, such as representations of an almost life size bearded man, probably a priest, or a robed man in front of an architectural representation (Bietak 1999, pl. VIII:b), probably part of a ritual scene.

To sum up, the technique, the composition, the themes, the style and the iconography of the wall paintings from the early Eighteenth Dynasty citadel at Tell el-Dab'a/'Ezbet Helmi are evidence that several Minoan artists were at work there, some of them artists of the highest standard. All specialists who have studied the originals of these paintings are unanimous in this finding. ⁽⁸⁾ The style is naturalistic and reminiscent of early neopalatial art in Crete, such as, for example, the representations from the 'House of the Frescoes' at Knossos or the paintings from Ayia Triada. Bull leaping and bull wrestling are themes which can be considered as so far particularly characteristic of Knossos. The maze pattern and the half-rosette frieze can also be seen as emblems of this chief centre of the Minoan thalassocracy. There are fragments of griffin representations from Tell el-Dab'a (Bietak 1999, pl. IX) which are as big or even bigger than the griffin representations from the throne room at Knossos. As at Knossos, the background consists of aquatic plants. It seems not unlikely that there was such a throne room, decorated with heraldic griffins, in the platform building H/I at Avaris.

According to Reusch (1958), Niemeier (1986), Marinatos (1993, 151-155, figs. 122, 128-131, 134; and 1995), the throne room in the palace of Knossos was not for a king but for a queen, as griffins are depicted in painting and in the minor arts as companions of a goddess or queen who should be considered as the 'mistress of animals'. If this also applied at Avaris, we might expect to find a throne room for a queen there, decorated in the tradition of Knossos. Such a scenario would endorse the working hypothesis of a dynastic connection between the courts of the early Eighteenth Dynasty and Knossos as an explanation for the Knossian royal symbolism and all the Minoan wall paintings at Avaris (Hankey 1993, 13f.; Bietak 1994a, 58). From the beginning, however, I have with all caution made it plain that this is a hypothesis, and that there is no direct evidence for it. On the other hand, to those familiar with Egyptian history the assumption of such a royal marriage does not appear at all unlikely (see already Meyer 1932, 54-57, and recently and independently of my hypothesis Hankey 1993). Whenever we have direct evidence of correspondence concerning Egyptian foreign policy, we can see that marriages with foreign princesses were an important part of it (Amarna tablets, Boghazköy tablets, see also Amasis's princess from Cyrene). Of course, there was no kind of romance connected with such arrangements. They were merely instruments for establishing international connections in the Old World.

Whether the throne room was for a king or a queen is a question that can be left open. The abundant presence of Minoan wall paintings with the emblems and the symbolism of the palace of Knossos, most likely executed by artists from Knossos, is a strong indication of direct links between the court of Knossos and that of the newly emergent superpower of Egypt at the beginning of the New Kingdom.

These contacts occur in a period when the Minoan thalassocracy, the major seapower of its time, was expanding to the Greek mainland and to Asia Minor. We also have Minoan paintings of approximately this time or even slightly earlier from other royal centres such as Alalakh and Kabri (Niemeier 1991; Niemeier and Niemeier here, vol. II), though without the royal Knossian symbolism which appears only at Avaris.

The connections with the Minoan world would fit well in the time of King Ahmose, whose ceremonial weapons display distinctly Aegean, if not Minoan, motifs, such as the Aegean griffin inlaid in niello on his ceremonial axe (bibliography in Saleh and Sourouzian 1986, no. 121). The blade of one of his daggers shows typical Minoan symbolic motifs in the same technique, including inverted landscape and a lion in flying gallop chasing a calf (bibliography in Saleh and Sourouzian 1986, no.122). ⁽⁹⁾ The base of the hilt consists of a bucranium worked in gold.

Puzzling is a title of the mother of Ahmose, Queen Ahhotep II: hnwt idbw H3w-nbwt (i.e. 'Mistress of the shores of Hau-nebut') (Lacau 1909, 3f.; recently Jánosi 1991-1992, 99-101 and Hankey 1993). This is a geographical term which was originally associated with the Aegean islands (Gauthier 1927, 12; Gardiner 1947, 206f). ⁽¹⁰⁾ Such an identification was rejected by Jean Vercoutter and Claude Vandersleyen who preferred to see it as a more general term for one of the nine bow-countries, focusing on a seabound region, particularly in Asia, especially Phoenicia (Vandersleyen), in opposition to t3, the land of the Nile valley, and h3st, the hill country or the desert, in consideration of the full epithet of this queen (Vercoutter 1946, 137-140, 156f.; 1948, 107-209; 1954, 43-47; 1956; Vandersleyen 1967, 168-174). In view of the strong evidence for the Minoan connections of the early Eighteenth Dynasty at Tell el-Dab'a, one should perhaps reconsider the problem of explaining this unique title. Ahhotep was without doubt an Egyptian queen. If the title had connections with the Aegean, these were of purely honorary character. How she might have acquired such a title, and whether the griffins decorated a throne room for her in Avaris, are at present subjects of pure speculation, but we cannot ignore the feeling that there was something important going on between Egypt and the Minoan world at the time of Ahmose, which is otherwise difficult to explain.

If we consider the basics, we have a strong link between the major seapower of its time, the Minoan thalassocracy, and a rising land power, Egypt at the beginning of the New Kingdom. What was the advantage of such a policy? For the Minoans, Egypt was a valuable partner because it had access to gold after Ahmose was able to retake Lower Nubia from the kingdom of Kush and because it was an intermediary in the trade in luxury goods from the Sudan. What, however, was the benefit to Egypt? Trade with the Minoan world could have been only a minor factor in this deal, as we have only very limited evidence of Minoan pottery at the time of the early Eighteenth Dynasty. What other reason could be found for Egypt to approach the Minoan thalassocracy? The motives are limited. Egypt lacked a seagoing fleet during this time and may have felt the necessity of protection from the sea. Against whom? one may ask. We should not forget that Ahmose had just expelled his former overlords, the Hyksos, from Egypt. There is the possibility that they still posed a potential threat from their bases in southern Palestine (Redford 1992, 129), in particular from Sharuhen, most likely Tell el-'Ajjul which had a harbour (Kempinski 1974). ⁽¹¹⁾ We should not forget that the young state had at the same time to grapple on its remote southern border with the kingdom of Kush.

(1). The excavations are organised by the Austrian Archaeological Institute, Cairo, the Institute of Egyptology of the University of Vienna and the Austrian Academy of Sciences. The work is financed by the Austrian Federal Ministry for Science and Research and the Austrian Funds for Scientific Research. For general information and bibliography on the site of Tell el-Bab'a see Bietak 1975; 1986; Bietak in Bietak, Hein et al. 1994, 17-58, 290-293; Bietak 1996a; Bietak 1997. For preliminary reports of this new site see Bietak et al. 1994; Bietak 1996b; Jánosi 1995a; 1995b; Dorner 1993; 1994a; 1994b; 1996.

(2). Janine Bourriau (1991) has concluded in a study of Kerma ware found in Egypt that most of this pottery dates to the time of the Eighteenth Dynasty and could be explained by the presence of soldiers with their families.

(3). Still unpublished. I owe its identification to Stefan Hiller and Peter Warren.

(4). According to an unpublished lecture given in Vienna in December 1995 by Edgar Peltenburg of the University of Edinburgh.

(5). The paintings were originally dated to the late Hyksos period because a small Eighteenth Dynasty settlement attached to the platform from the time of Ahmose onwards seemed initially to provide a terminus ante quem for the platform. The platform is, however, younger than the late Hyksos defences (see above), and shares the orientation of the New Kingdom remains. The date of the frescoes should therefore be revised, the more so as in area H/IIII such paintings were found in situ in early Eighteenth Dynasty contexts. While plaster made of lime was also found in late Hyksos strata, plaster with Minoan paintings has been found so far only in Eighteenth Dynasty contexts. That such paintings already appeared in the late Hyksos period cannot be ruled out; in chronological terms this would represent only a minor difference (a decade or two). However the political hiatus from the Hyksos period to the time after the conquest of Avaris would be very significant.

(6). Bull horns are depicted at Knossos not only in yellow (Morgan 1995, 42) but also in blue, and one may suspect that among the early frescoes blue was more frequently used for this purpose: see, for example, the bull from West Magazine XIII (Herakleion Museum, Room XVI, vitrine 174/no. 57; see Evans 1921, fig. 385; Cameron 1975 vol. 1, 168, fig. 24, pl. 74A); bucrania between two sphinxes in miniature painting as a part of a dress (Herakleion Museum, vitrine 174/no. 51) and unpublished fragments of paintings in boxes nos. 20, 105.

(7). Such a parallelism in dating can also be supported by the presence of Late Cypriot White Slip I hemispherical bowls both in the early Eighteenth Dynasty stratum at Tell el-Dab'a and on Thera (cf. Hein 1994, 42, fig. 12d, pl. 13b; Hein in Bietak, Hein et al. 1994, 258, no. 352; Maguire 1992; 1995). I should take this opportunity to correct a mistake in Bietak 1991, 310-312; 1995, fig. 1, where I stated that White Slip I ware already appeared in the late Hyksos period. A check of all occurrences of this ware in areas A/V and H/I-V at Tell el-Dab'a has revealed that the only secure contexts are in early Eighteenth Dynasty strata. For Thera, cf. Niemeier 1980, 72-74, fig. 44.

(8). I am indebted to Ellen Davis (New York), Stefan Hiller (Salzburg), Nanno Marinatos (Athens, iconographist of the Tell el-Dab'a team), Lyvia Morgan (Cambridge) and restorer Iakovos Michaelides (Hellenic Antiquity Service) who has twenty-five years of experience in working with frescoes from Thera and Knossos. Maria Shaw (Toronto) proposed before she had seen the originals that the artists were not Minoans but were trained at Knossos (Shaw 1995).

(9). The weapons were found together with other precious objects of Kings Ahmose and Kamose in the coffin of Queen Ahhotep I Naga, perhaps the wife of King Senakhtenre' of the Seventeenth Dynasty. As the objects do not bear the names of Ahhotep I it is perfectly possible that the priests who later hid the half plundered royal burials from tomb robbers put objects from the burials of Ahmose and Kamose into this coffin.

(10). The famous German historian Eduard Meyer even translated this toponym as Crete, and suggested, on account of the title of Queen Ahhotep, that she had established special ties with the Minoans during a regency for her sons Kamose and Ahmose. With Minoan help the Hyksos were finally overthrown. Meyer (1928, 54-57) even went as far as to explain the title by postulating a marriage between the widowed Queen Ahhotep and the Minoan king. Of course this is stretching the evidence too far, but it shows that the idea of a dynastic link was already suggested many decades before the discovery of the Minoan frescoes at Avaris.

(11). Oren (1997) has recently suggested Tel Haror, inland from Tell el-'Ajjul.

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