Naukratis: Greeks in Egypt

Alexandra Villing, Marianne Bergeron, Giorgos Bourogiannis, Alan Johnston, François Leclère, Aurélia Masson and Ross Thomas

With Daniel von Recklinghausen, Jeffrey Spencer, Valerie Smallwood, Virginia Webb and Susan Woodford

Search this Catalogue

Advanced search  

See all objects in this catalogue 

Supported by

The Leverhulme Trust
  • The Shelby White - Leon Levy Program for Archaeological Publications
  • Christian Levett and the Mougins Museum of Classical Art
  • Share this catalogue

The material culture of Naukratis - an overview

Alexandra Villing, with Marianne Bergeron, Alan Johnston, Aurélia Masson and Ross Thomas

 Download this chapter as a PDF


Research on finds from the early fieldwork undertaken at Naukratis in the late 19th and early 20th centuries by Flinders Petrie and others now makes it possible for material previously little known and studied to make a substantial contribution to scholarship, increasing our knowledge of the site of Naukratis and its history, our understanding of ancient Greek and Egyptian practices and of the interaction between these two cultures.

The aim of this brief chapter is thus threefold: firstly, to provide an introduction to the rich corpus of finds from the early fieldwork at Naukratis that sets out the nature of the assemblage and the difficulties inherent in its interpretation; secondly, to briefly characterize the main (and most revealing) groups that make up the assemblage and to set them within a wider archaeological and historical context; and thirdly, to highlight in particular some of lesser known and studied categories of material that open up new perspectives and insights for scholarship on Naukratis and beyond.

The insights summarized in this chapter are the result of research conducted by the Naukratis Project curators and academic collaborators and are set out by them in more detail also in the various individual chapters of this Online Catalogue and related publications.

1. Introducing the assemblage

1.1 Characterizing the assemblage

The ancient site of Naukratis is rich in material from diverse cultures and periods: Late Period Egyptian (664–332 BC), Archaic and Classical Greek (700–323 BC), Macedonian and Ptolemaic (332–30 BC), Roman and early Byzantine (30 BC–641 AD); Greek and East Greek, Cypriot, Levantine, Phoenician, Persian, North African, Etruscan and Italian. This diversity is clearly (though not entirely accurately, see below) represented in the multitude of preserved finds from the early excavations of 1884 to 1903 at the site, totalling some17,000 objects (with a further c. 1,500 currently not traceable). Petrie’s, Gardner’s and Hogarth’s publications (which referenced barely 20% of the excavated and retained material) had presented a rather selective picture, in which certain categories of material were favoured and highlighted.

Other groups were neglected or ignored – and by no means just the insignificant or uninstructive ones: for example, the approximate 1,600 stamped handles from (mostly Classical to Hellenistic) trade amphorae (see the chapter on Stamped amphorae), which were carefully collected yet never published, or the numerous ‘phallic figurines’, clearly deemed unsuitable for public consumption and only obliquely referred to as ‘rude’ or ‘indecent’ (Thomas forthcoming b). The work of subsequent scholars increased the body of known material only slightly, and their focus on selected groups of material and periods further contributed to creating a somewhat skewed picture of the site, notably a bias towards Archaic and Classical Greek material.

Chart 1 Approximate proportions of previously published (1884–2012) versus unpublished finds. For many finds the information given in publications is highly incomplete; they are subsumed here under ‘partially published.’

In actual fact, the preserved material from the site is far richer and more diverse than is often acknowledged. It attests to the site’s long and complex history from the late 7th century BC through to the mid-7th century AD, its wide network of contacts stretching areas across Egypt, Africa, the Mediterranean and Levant and to the daily lives and interaction of the inhabitants and visitors of this bustling multicultural city. Originating from sanctuaries and settlements, tombs and workshops, storerooms and river fronts, the objects range broadly in type and date, reflecting a wide range of activities, practices and spheres of life in which the inhabitants and visitors of Naukratis engaged: production and consumption, religious and domestic life, commerce and exchange.

Such insights are possible despite the fact that, for a number of reasons, precise findspot information is notoriously tricky to come by. Firstly, although the vast majority of finds were recovered during the four seasons conducted by Petrie, Gardner and Hogarth, a number of objects were also found and collected at other times and outside of the excavations (see the chapter on Discovery and excavations). Even during fieldwork, many finds were not ‘excavated’ but brought to the excavators by sebbakhin and other individuals from different parts of the mound. This is particularly true for finds from the ‘town’, which Petrie left to the locals to work through, recording only a few findspots (such as the house of the bronze cache: see the chapter on Bronze votive offerings) in detail.

Chart 2 Objects with recorded findspot, based on a sample of c. 3,000 objects

Of finds with a recorded findspot (some 3,000 pieces; however, for many others, a findspot can be reconstructed with considerable likelihood), about two thirds come from the Greek sanctuaries, notably that of Apollo. If likely (but not explicitly attested) findspots are included, the sanctuary of Aphrodite in particular gains in prominence.

1.2 Problems with the assemblage

Some 17,000 objects collected at Naukratis in the late 19th and early 20th century – mostly during the four main fieldwork seasons – are today still located in over 70 museums worldwide. In addition, well over a thousand objects are known from publications or notes and although not yet identified are likely to also be extant. The total number of 19,000 known, and over 17,000 extant, finds is impressive – and to record and study them has certainly proved a massive task. However, the material assemblage originally encountered by the early excavators at this substantial and long-inhabited site must have been considerably larger.[1] This is particularly true for the pottery finds, the largest group of material preserved, which makes up over half of the extant assemblage with more than 9,500 pieces (see below, Chart 4).

However, this figure is put into perspective when we read that up to 5,000 sherds could be found in a single day’s digging (Gardner 1888, 15) and that an estimated 150,000 ‘good fragments’ (presumably fine or semi-fine ware) were recovered from a single Archaic layer in the Aphrodite sanctuary (Gardner 1886b, 181). Even if this figure might be somewhat exaggerated, Greek sanctuaries are well known for their high density of pottery. To cite but two examples, excavations in the Argive Heraion in the late 19th century yielded a total of some 250,000 sherds, while in the Demeter sanctuary at Cyrene 4,014 fragments were counted in a single cubic metre.[2] During our own excavations in 2014, the sherd count was lower, but this area, beside what may be a large temenos wall, was probably not a central cultic area.[3] At Naukratis, at least five Archaic (and later) Greek sanctuaries, large parts of a town and a massive Egyptian sanctuary were excavated; even a conservative estimate of the total number of sherds turned over by early archaeologists and sebbakhin during the four seasons when archaeologists were present at the site would be in the region of several million in total.[4]

Chart 3 Proportions of decorated versus undecorated pottery among extant finds from early fieldwork at Naukratis (a) and recent fieldwork in the settlement at Archaic Miletos (b).

What is preserved today can thus be but a tiny fraction of the original assemblage. Nor is it, indeed, a representative sample: ‘coarse’ kitchen and storage pottery (not counting stamped amphora handles, which are far better attested) make up only some three percent of the surviving pottery evidence. In contrast, at Miletos – one of the Greek cities most closely linked with Naukratis from the beginning and thus an apt comparison – some 90% of pottery fragments encountered in the 7th–6th century BC settlement were plain or ‘coarse’.[5] Even in sanctuaries a substantial proportion of ‘coarse wares’ as well as simply decorated wares (banded/black-glazed) is to be expected, with plainer wares typically making up around a third of fine wares in Archaic and Classical sanctuaries. In the Demeter sanctuary of Cyrene, for example, over 60% of the sherds encountered in a cubic metre of sanctuary deposit were identified as plain or coarse ware; in sacred refuse and destruction contexts at Archaic Didyma (Taxiarches), 41% of vessels were plain or simply decorated.[6] Yet at Naukratis plain or simply decorated vessels account for no more than 12% of known finds from any (not just sacred) context. The fact that at least 90% of them carry an inscription makes it clear that only some ‘special’ pieces were kept, while regular, uninscribed plain wares – although undoubtedly common – were routinely discarded.

Figure 1. Pottery from a well excavated by Hogarth in 1903, including 6th and 5th century BC Egyptian and Syro-Palestinian vessels (Edgar 1905, 104 fig. 3)

Indeed, undecorated ceramics (Fig. 1) are recorded as being plentiful in the notes and diaries of Petrie and Hogarth: ‘The well unproductive & I shall not dig many. Any am[oun]t of mud & coarse kitchen ware – but nothing else’ (Entry in David Hogarth’s diary for Sunday 19 April 1903).[7] Undecorated wares were also recorded in large numbers for all periods of the site in the American fieldwork at Naukratis in the 1970s–80s,[8] and they make up a substantial proportion of finds in the British Museum’s own recent (2014) excavations, including in the 6th century BC layers. Here, a preliminary count of several thousand sherds is particularly revealing: the vast majority of finds are undecorated, comprising especially Egyptian vessels (nearly always entirely undecorated, and frequently quite ‘coarse’), but also numerous plain or banded wares of Greek, notably East Greek, manufacture. Leaving them out of the picture has major implications for statistical analyses: therefore, while Egyptian, Cypriot and Levantine wares make up by far the majority of finds from the 2014 excavations, much the reverse is true for a statistical analysis of known pottery from the early fieldwork.

Chart 4 Approximate proportions of pottery by place of production from a) fieldwork by Petrie, Gardner and Hogarth, as represented in the total extant/known assemblage (predominantly 6th century BC); b) excavations by the British Museum (2014) in 6th-century BC levels.

The above demonstrates beyond doubt not only that large amounts of plain household pottery (Greek or otherwise) must have been deliberately omitted from the 19th-century assemblage, but also that any local Egyptian pottery in particular, which was almost always undecorated, along with most of the trade amphorae of all periods and origins, including Levantine and Cypriot (except for stamped amphora handles and inscribed pieces), are heavily under-represented, to the extent of their presence having been almost entirely eradicated. With such an incomplete and biased part of the assemblage known today, any interpretation – certainly quantitative, but even qualitative – is clearly problematic. At the very least, any historical interpretation based on the absence or scarcity of certain undecorated wares – notably Egyptian pottery – must be severely compromised.

Unsatisfactory as this undoubtedly is, the situation does not devalue the assemblage completely as a historical document. Certainly, our understanding of certain topics necessarily remains restricted by the excavators’ limited focus and find selection. Yet many aspects of life in ancient Naukratis are nevertheless represented and indeed recoverable from it at least to some extent, on the conditions that due caution is applied, contexts are carefully considered and any information – comparative or otherwise – that may serve as a potential corrective to obvious biases is judiciously exploited.[9] In some instances – such as Petrie’s interest in iron tools or shells (see Introduction, section 3) – we even catch unexpected glimpses of rarely studied aspects of ancient reality. Modern fieldwork, conducted with new methods and a wider perspective, now crucially supplements early evidence, with both the American fieldwork of the 1970s and 1980s (Coulson 1996; Leonard 1997; 2001) and the British Museum’s recent three seasons of geophysical, geological and archaeological fieldwork (Thomas and Villing 2013; Thomas and Villing forthcoming) providing vital additional information to complement and correct the picture derived from the work of the pioneers, including allowing us to assess practices rather than just the mere presence of objects. These new insights have enabled us to radically revise longstanding notions of the site: its situation in the landscape, its size and layout, its ethnic make-up and its development over the approximate 1,300 years of its existence. Perhaps most importantly they challenge our idea of Naukratis as an essentially Greek foundation.[10] Instead, they present a complex picture of a far more Egyptian and indeed multi-ethnic town, with both Egyptian and Greek elements to the population present from its beginning, who participated not just in Greek–Egyptian trade but played a role in the variety of networks of contact and exchange that linked Egypt with the rest of the ancient world.

Chart 5 Categories of material represented among the known finds from early fieldwork at Naukratis (approximate percentages based on number of fragments).

2. The finds from Naukratis

2.1 Architecture

For a city the size of Naukratis, which covered over 60 hectares and with space to accommodate well over 10,000 people (Thomas and Villing 2013 and forthcoming), surprisingly few architectural elements are preserved. This is in part due to the scarcity of stone in the Nile Delta region, which meant that mudbrick was the building material of choice, sometimes with plaster or stone for cladding and additional elements such as columns and ornamentation. Over time, decommissioned buildings moreover tended to be stripped of stone for re-use or processing in lime kilns. Despite this, over 130 objects from the old excavations that can be classified as architectural elements still survive, many of them belonging to the once numerous temples, altars and public buildings of Naukratis (Koenigs 2007).

They include some important elements of early Ionic architecture in limestone and (Ephesian) marble (Fig. 2) with some preserving rich traces of original colouring.

Figure 2. Fragment of an Archaic Greek marble cornice from the sanctuary of Apollo, c. 520–490 BC. British Museum, 1886,0401.41.

The majority come from altars as well as the two successive temples (probably dating to around 560–550 and 530–510 BC, respectively)[11] in the sanctuary of Apollo, but there are also parts of Archaic votive column bases from the Aphrodite and Dioskouroi sanctuaries and possible architectural relief fragments depicting a Late Archaic hoplite and a walking man found reused in the Hellenion (Edgar in Hogarth et al. 1898–9, 65–7; Edgar 1905, 126–7 fig. 8). From the early Ptolemaic period, fragments of the elaborately decorated and inscribed low relief blocks (in basalt or dark grey granite) from the outside walls of the temple of Amun-Ra, (re)built by Ptolemy I Soter (323/305–283 BC), are preserved, representing a procession of regional personifications of the parts of Egyptian territory and its productive forces presenting offerings to the Pharaoh (for more details, see the chapter on the Decoration of the temple of Amun).[12]

Excavating the monumental pylon (entrance gate) to the temple precinct, Petrie discovered at least six foundation deposits of Ptolemy II Philadelphos (285–246 BC) (Fig. 3) one at each of the four corners and one at the northwest and southwest corners of the central hall (Petrie 1886, 28–31, pls XXV–XXVI; Weinstein 1973, 376–8, no. 158). Nearly 100 objects have now been re-assigned to these deposits, although a good quarter of them are lost or have not yet been located.

Figure 3. Foundation deposit from the pylon of the sanctuary of Amun-Ra, built under Ptolemy II Philadelphos (285–246 BC), excavated and photographed by Flinders Petrie in 1885. Photograph © Petrie Museum of Egyptian Archaeology, University College London, PMAN 2680.

Figure 4. Limestone model of Egyptian tower house from Naukratis, showing two windows and courses of brick, 7th–5th century BC. British Museum, EA68816.

In addition to the finds made by the early excavators, several limestone column drums are still today dotted across the village of El Baradany and were recorded by Coulson as part of the American survey at Naukratis (Coulson 1996, 14–16, fig. 7, pls 7–8). Column and possible pavement fragments were also collected by Leonard in the eastern part of Naukratis (Leonard 2001, 201–4, fig. 3.10, pl. 3.31). Of the painted wall-plaster encountered by the early (and later) excavators in many areas of the site (cf. Leonard 2001, 205–16), few examples survive, including some fragments of the meander and stars-within-square designs that E.A. Gardner (1888, 31) recorded as having been painted on fine hard stucco on the inside walls and pillars of the sanctuary of the Dioskouroi. We may also note that wells were excavated in different parts of the site, constructed in a number of different ways and using a variety of different materials (Hogarth et al. 1898–9, 31, 34, 41; Gardner 1888, 35).

Architectural models from Naukratis give a further impression of the architectural styles that might once have been present at the site, including Egyptian naos models as well as models of tower houses (Fig. 4), which recent magnetometry survey suggests was a common house form at Naukratis (Thomas and Villing 2013; Thomas forthcoming a; see also the chapter on Topography).

2.2 Pottery

2.2.1 Greek and Roman imports

Archaic and Classical painted Greek pottery, totalling over 7,500 objects, today makes up nearly half of the total known assemblage of finds from the early fieldwork at Naukratis, with the bulk of it dating to the 6th century BC.[13] Even though Greek pottery has long been a traditional focus of Classical scholarship at Naukratis,[14] only a part of the material has actually been available for inclusion in the academic discourse: the objects assembled in the present catalogue (the majority of which are sherds) more than treble the number of known pottery finds.

Figure 5. Corinthian column krater from the sanctuary of the Dioskouroi, c. 600–570 BC. British Museum, 1886,0401.1103.

Figure 6. Fragment of Attic olpe (jug) depicting a ram, attributed to the manner of the Gorgon Painter, c. 600 BC. British Museum, 1888,0601.566.

Moreover, the catalogue also provides new information regarding both provenance and findspot for some of the previously known material. This not only requires a number of revisions and corrections, but also furnishes the basis for in-depth analyses of regional and supra-regional networks of distribution as well as local patterns of consumption and ritual practice in particular.

Among mainland Greek pottery, Corinthian ware (Fig. 5) is chronologically the earliest, even if – with just over 300 pieces – it is not especially frequent, peaking in the first third of the 6th century BC (see the chapter on Corinthian pottery as well as Bergeron forthcoming a and c).

It is followed by Attic pottery (Fig. 6) – far more common (some 2,000 pieces are preserved) and present from around 600 BC through to the end of the Classical period, in black-figure, red-figure and black-glaze and some white-ground – and a very limited amount (some 70 pieces) of Laconian pieces (Fig. 7).

Figure 7. Laconian kylix attributed to the Naukratis Painter, with a votive inscription to the goddess Aphrodite by Ne(i)lomandros, c. 570–560 BC. Ashmolean Museum, Oxford, AN1896-1908-G.133. Photograph © Ashmolean Museum, University of Oxford.

Figure 8. Grand Style chalice, c. 580-570 BC, Ashmolean Museum, Oxford, AN1896-1908-G.134. Photograph © Ashmolean Museum, University of Oxford.

Overall, however, East Greek wares (around 4,000 specimens) dominate, confirming what we know from Herodotus about the involvement of major East Greek poleis in the emporion. From the early period of the site there are large numbers of Chian vessels (once thought locally produced because of their abundance at Naukratis), many of which are from the Aphrodite sanctuary, but also from other parts of the site (Fig. 8; for more details see the chapter on Chian pottery).

Aeolian, North Ionian – Clazomenian and likely Tean – as well as South Ionian – notably Milesian and Samian – vessels are decorated in the Wild Goat and later black-figure (north) and Fikellura (south) styles, but also include numerous ‘Ionian cups’ especially in the Milesian Apollo sanctuary (Fig. 9), as well as distinctive Samian-made sanctuary crockery for the Heraion (see Schlotzhauer 2006a, 311–13; Schlotzhauer 2012, 154–7).

The presence of East Greek grey wares from the region of the Aeolis and the Troad as well as of a few Carian fragments (Fig. 10) is remarkable as these wares are otherwise little exported.

Figure 9. South Ionian cup from the sanctuary of Apollo, c. 580–540 BC. Museum of Fine Arts, Boston, 86.638. Photograph © Museum of Fine Arts, Boston.

Figure 10. Fragment of Carian skyphos from Naukratis, late 7th century BC, made up of two joining sherds. Photomontage combining Greenock, McLean Museum & Art Gallery, 1987.464 and British Museum, 1888,0601.653. Photograph © McLean Museum and Art Gallery, Greenock / Inverclyde Council and © The Trustees of the British Museum.

Nearly as elusive is Rhodian and other East Dorian painted pottery, which is represented primarily by Vroulian cups and a few painted plates (for which Kos now seems to emerge as a prominent centre of production; Villing and Mommsen forthcoming). However, the picture may again be somewhat skewed, as Ionian cups of East Dorian production were presumably originally more frequent than can be gleaned from the extant record. Several figure or ‘plastic’ vases, mostly East Greek perfume containers, have also been found (see also Möller 2000a, 146–7). A few vessels reached Naukratis from further afield, including Italy. A handful of fragments of Etruscan kantharoi (Fig. 11) and oinochoai are not altogether surprising, given that a thin smattering of Etruscan bucchero is attested across the East Greek world (Naso 2006).

Figure 11. Fragment of Etruscan bucchero kantharos from Naukratis, 7th–6th century BC. Leiden University, Faculty of Archaeology, T.2930. Photograph © Faculty of Archaeology – Leiden University.

From a later date there are scarce examples of Sicilian red-figure and Apulian Gnathian wares. A Classical Syracusan visitor is also attested in person in the form of a graffito dedication. Finally, as discussed below, Greek style fine and semi-fine table ware was also locally produced at Naukratis, certainly by the 6th century BC, and apparently in some quantity.

Findspots of painted Greek pottery (where recorded) are primarily the Greek sanctuaries, although some painted wares have also been found in tombs (Fig. 12) and other parts of the site. As regards to the range of shapes, unsurprisingly for sanctuary pottery, symposion-related vessels dominate. However, a variety of storage vessels also feature, as do some shapes possibly used for food, and a number of special ritual vessels such as kernoi, notably from Chios (see the chapter on Chian pottery).

Figure 12. Attic red-figure acorn lekythos from the cemetery of Naukratis, c. 400–380 BC. British Museum, 1888,0601.716.

The wide variety of wares that were locally consumed, mostly in ritual contexts, which originate from many parts of the Greek world, reflect both the close links that Naukratis maintained with cities particularly involved in the emporion and more general patterns of trade and interaction that characterized the Mediterranean world in the 1st millennium BC. Milesian, Samian, Chian, North Ionian (Fig. 13) and Aeolian (Fig. 14) pottery at least in part mirror the involvement of Miletos, Samos, Chios, Teos, Phokaia, Klazomenai and Mytilene at the site without, however, supporting the theory (based on a note in Strabo 17.1.18 [C801]) of Naukratis originally being a Milesian foundation.

Figure 13. Fragment from the rim of a North Ionian black-figure dinos depicting a horse-rider, c. 560–550BC. Fitzwilliam Museum, Cambridge, GR.54c.1899. Photograph © The Fitzwilliam Museum, Cambridge.

Figure 14. Aeolian Wild Goat Style askos of the London Dinos Group, c. 630–600 BC. British Museum, 1888,0601.462.

The rarity or even absence of pottery from places such as Phaselis and Aigina should not necessarily be equated with an absence of traders, but is explained by a lack of notable fine ware production in those cities. By contrast, it is Aeginetan traders who are often credited with being responsible for bringing Attic and Corinthian wares to Naukratis, while Samian traders are often associated with distributing Laconian wares (although this picture may have to be revised on account of recent finds in Miletos), rather than necessarily pointing to Athenian, Spartan and Corinthian visitors to Naukratis. Equally, the few pieces of Etruscan bucchero pottery may well have come with the East Greeks or Aiginetans. It is these travellers – mobile elites, and especially the diaspora of traders operating around the Mediterranean including those that had now made Naukratis their main base – who most likely also ‘consumed’ large parts of the wares they brought to Naukratis at least in its early years, through dedication and feasting at the local sanctuaries (Villing 2013). This included people such as the Naukratite trader Herostratos who, after being saved from shipwreck by the goddess Aphrodite, ‘invited his relations and closest friends to a feast in the goddess’s temple at Naukratis’ (Athenaeus 15.675f–676c).

In contrast to decorated wares, plain Greek table wares are rare among the extant material; both (East Greek) banded wares (Fig. 15) and Attic and other black-glaze wares are attested.

Figure 15. East Greek banded oinochoe from Naukratis, 6th century BC. British Museum, 1886,1005.20.

Judging from what is known from other sites, notably in domestic but also sanctuary settings, their original percentage at all periods originally must have been far higher. This is now also confirmed by the British Museum’s excavations in 2014 in the area around the Hellenion, where simple banded Archaic East Greek wares were encountered in some quantity, as well as by the American survey and excavations in the 1970s and 1980s, which yielded large quantities of household pottery of all periods, especially Hellenistic/Ptolemaic and Roman. The same is true even more for Greek and other imported kitchen pottery. A handful of Corinthian mortaria (grinding bowls) and perirrhanteria (sacred water basins) are worth noting, as these are vessels known to have been widely exported in the Late Archaic and especially Classical periods (Villing and Pemberton 2010). Imported Greek cooking pots appear to be essentially missing from the 19th-century assemblage, but have now been attested for the Archaic period in new fieldwork at the site (Thomas and Villing forthcoming). A whole range of both decorated and plain ‘functional’ Greek wares from a variety of centres thus reached the city from an early period onwards.

During the Macedonian and Ptolemaic periods (332–30 BC) the vast majority of the pottery needs of the inhabitants of Naukratis were met by local potters, who produced a range of transport and storage vessels, cooking pots and the typical black and red-slipped table ware echinus bowls (Fig. 16) and fishplates commonly found at Hellenistic sites.

Figure 16. Fragment of locally made Hellenistic black ware bowl from Naukratis with the name ‘Kolios’ incised underneath the foot. British Museum, 1910,0222.225.a.

Figure 17. Fragment of relief-decorated goblet made in Alexandria showing a woman dancing, 2nd–1st century BC, from Naukratis. British Museum, 1886,0401.1544.

These are poorly represented in the early excavations, although they were well published by Andrea Berlin from Leonard’s subsequent excavations (over 2,000 examples in Berlin 1997a and 2001). Indeed, among the material from early fieldwork, Ptolemaic pottery is limited to complete, decorated or inscribed vessels with only a few exceptions. Less common table wares include imported Attic black-glazed pottery and Hellenistic mould-made bowls. A quite substantial group specific to Naukratis and Alexandria consists of relief-decorated and painted goblets (Fig. 17) and table amphorae produced in Alexandria and (probably) Naukratis (Bailey 2011).

Roman table pottery is very poorly represented in the corpus of early finds, comprising fewer than 100 examples; indeed, more Roman lamps were recovered during these early seasons than vessels. They are, however, well attested in subsequent fieldwork, particularly Coulson’s field survey (Coulson 1996; Thomas and Villing 2013; Thomas 2014b). As elsewhere in the western Delta (Wilson and Grigoropoulos 2009; Tomber and Thomas 2011), Roman pottery was largely locally made, consisting of cooking pots, casseroles, dishes and bowls with a drab red slip. Imported table wares included Arretine terra sigillatas of the Augustan period and the early 1st century AD; copies of terra sigillata from Ephesus and Syria (Fig. 18); and small quantities of glazed wares from Asia Minor. Subsequently, Egyptian copies of red slipped wares and barbotine thin-walled wares, as well as copies of terra sigillata forms in Egyptian faience became increasingly popular in the 2nd century AD.

Figure 18. Fragment of an Imperial Roman terra sigillata bowl or cup from Naukratis, made in Syria, c. 20 BC–150 AD. Musée du Louvre, Paris, AM1419. Photograph © Musée du Louvre.

By the 3rd and 4th centuries AD, Late Roman red-slip table ware imported from Tunisia, and later especially Cypriot Red slipped table ware, became popular in Naukratis, as well as at Alexandria and Mareotis (Tomber and Thomas 2011, 38; Thomas 2014b), whilst Egyptian variants made in Aswan, Alexandria and the Nile valley are also represented (Hayes 1972; Tomber and Thomas 2011).

2.2.2 Trade amphorae

The use of trade amphorae to transport a variety of products, not just wine and oil (for example, fig remains are preserved inside the neck of an Archaic East Greek amphora; Stacey et al. 2011), was well established in the Mediterranean world by the time Naukratis was founded, and Naukratis’ position as an international trade port is confirmed by the range and frequency of such containers in all periods (see the chapters on Greek transport amphorae and Stamped amphorae. It is clear from Petrie’s diary and the early excavators’ published accounts that many more finds were found than were kept, and only fragments that carry a painted, incised or stamped text were collected. Only one intact jar seems to be preserved.[15]

The few jars surviving from the Archaic period are largely from East Greece (especially Chios and Samos: Fig. 19) and Cyprus.

Figure 19. Fractional Samian trade amphora from a well in the sanctuary of Apollo at Naukratis, 6th century BC. British Museum, 1886,0401.1291.

Classical material is very thin, although we should note that marks on Classical amphorae are generally fewer in number, virtually disappearing in the Hellenistic period; therefore little can be said for the amphora material arriving at Naukratis in c. 500–350 BC. Stamped jars – of which some 1,600 have their handles preserved from the early excavations – have a reverse pattern of production: very rare in the Archaic period, although common in Hellenistic times, and this is reflected in the identifiable corpus from Naukratis. The earlier jars are from Thasos, dating approximately to c. 375 to 250 BC, while a lengthy Rhodian series began in the early 3rd century BC, joined in the later 3rd century by substantial numbers of Knidian and Koan pieces (Fig. 20).

Figure 20. Stamped amphora handles from Naukratis. From left to right: Thasian, c. 340–330 BC, British Museum, 1925,0119.464; Rhodian, c. 200–160 BC, British Museum, 2011,5003.50; Koan, c. 200–50 BC, Musées Royaux d’Art et d’Histoire, Brussels, A.1822 (Photograph © Musées royaux d’Art et d’Histoire, Bruxelles); Italian (Brindisi), c. 130–50 BC, British Museum, 1955,0920.89.

One should also stress a large number of stamps whose origin remains unclear. The overall pattern is reflected in the known corpora from other Egyptian sites. Stamps clearly dating to after c. 70 BC are extremely rare; among the later examples is a good range from the workshops in the Brindisi area. In addition, Ptolemaic Egyptian amphorae (only rarely stamped) are well represented among the material excavated during the American mission’s work (Leonard 1997, 2001).

During the Roman period Naukratites largely relied on Egyptian products transported in locally produced amphorae, such as fine wines from the Nile valley and Mareotis – areas that had become regional and international exporters of wine in the 1st century AD, when Alexandrian amphorae were exported to India, South Arabia and East Africa (Tomber 2008). During the 3rd and 4th centuries AD, the increase in North African and Cypriot/Cilician imports to Naukratis was largely driven by imported wine amphorae. Two stamped Tripolitanian amphorae from the reigns of Septimius Severus to Severus Alexander represent just a small proportion of the coarse, unstamped Roman amphorae sherds that must have been encountered in the early excavation. Large numbers of locally produced Roman and Late Roman amphorae were found by Coulson during his survey (Coulson 1996) and they have also been observed in recent surveys and excavations at the site, including Italian, Cypriot, Cilician, Gazan and Greek amphorae (Thomas and Villing 2013; Thomas and Villing forthcoming).

In addition to amphorae, a number of stamped terracotta, mud and plaster amphora stoppers of Ptolemaic to Byzantine date are preserved from the site (Fig. 21).

Figure 21. Plaster amphora stopper from Naukratis with impressed decoration including the Christian symbol chi-rho in the centre, 4th–7th century AD. British Museum, 1888,0712.43.

One example with a Latin inscription is an Italian import, while another is an Egyptian example from the Fayum with a representation of the local goddess Isis-Renenutet. Ptolemaic stamps used to impress the plaster seals of amphorae stoppers have also been found, indicating that Naukratis was also actively involved in wine production or sale. Amphora stoppers from the Ptolemaic period tend to be decorative, whilst those of the Roman period record the names of traders. Some Byzantine examples carry Christian symbols and/or prayers.

2.2.3 Cypriot and Levantine pottery

Cypriot and Levantine pottery is sparingly preserved at Naukratis and therefore has been little considered by scholars. This may be justified as far as fine wares are concerned, of which only very few examples are known and which in all likelihood were never particularly common. Cypriot mortaria (Fig. 22) and basket-handled amphorae, however, were noted by the excavators as being frequently encountered in the early levels of the site, an observation now confirmed by recent excavations in 6th-century BC layers, which revealed numerous fragments of Cypriot mortaria and basket-handled amphorae as well as Phoenician torpedo amphorae (Thomas and Villing forthcoming).

Figure 22. Fragment of Cypriot mortarium inscribed with dedication to Apollo, early 6th century BC. British Museum, 1886,0401.80.

Who carried these vessels to Naukratis is difficult to ascertain, but the dozen Greek votive inscriptions on mortaria, a rare phenomenon in the ancient world (Villing 2006), indicates their use by Greek speakers.

2.2.4 Egyptian pottery

As is now clear, Egyptian pottery was ubiquitous at Naukratis throughout its history. In terms of types and shapes it largely conforms to what is found at contemporary Egyptian sites across the Delta. As is discussed in greater detail in the chapter on Egyptian Late Period pottery, it seems to appear at Naukratis at about the same time as Greek pottery, thus further supporting a scenario also suggested by other categories of finds that the site was newly established in the later 7th century BC, with both Egyptian as well as Greek elements to its population from the start. Even though relatively few examples of Egyptian-made pottery are today identifiable among the known early finds, it was clearly encountered in quantity by the excavators. Hogarth reported a whole layer with early Egyptian pottery below a layer of Archaic Chian and other Greek pottery, and noted and published Late Period (664–332 BC) pottery from other areas as well.[16] A number of Late Period (664–332 BC) vessels (Fig. 23) are still preserved today from Petrie’s excavations. Late Period Egyptian pottery was also found by the American mission (Berlin 1997a; 2001 – although largely passed over or dismissed as ‘Ptolemaic’)[17] and was abundant in recent fieldwork in Late Period layers at the site, accounting for over three-quarters of pottery finds (Thomas and Villing forthcoming).

Figure 23. Egyptian offering dish, Late Period, c. 630–400 BC. Warrington Museum & Art Gallery, 1888.57.154. Photograph © Courtesy of Warrington Museum & Art Gallery.

Figure 24. Fragment from a plate made locally at Naukratis but East Greek in shape and style; from the sanctuary of the Dioskouroi, second quarter 6th century BC. Museum of Fine Arts, Boston, 86.533. Photograph © Museum of Fine Arts, Boston.

Finds moreover suggest that Egyptian-made pottery remained the staple of the town’s pottery consumption through Ptolemaic and Roman times, consisting of a wide range of shapes and forms for household, but also possibly sanctuary use (Berlin 2001). While most of the Egyptian pottery appears locally or regionally produced (Berlin suggests production for example at Kom Firin/Kom Dahab: Berlin 2001, 45–6), some imported pieces come from other Egyptian regions, such as Thebes, Memphis and later Alexandria, but also the Western Egyptian Oasis, a link that is rare in the Delta after the New Kingdom (British Museum, 1886,1005.15).

Of particular interest is the phenomenon of locally produced foreign-style pottery, discussed in more detail in separate chapter. From the early 6th century onwards a wide range of (East) Greek table and fine ware shapes with (East) Greek decoration are attested (Fig. 24), including Ionian cups, oinochoai, juglets, bowls and plates (as well as lamps).[18]

Made by (East) Greek-trained craftsmen (as betrayed by the often exceptionally finely levigated fabric and expert potting), most were probably destined for local consumption, although an amphora from the Naukratis workshop has been found as far afield as Tell Dafana/Daphnae (British Museum, 1888,0208.57). In the Hellenistic period a kind of black-glazed pottery (‘terra nigra’ or ‘black ware’), rather coarsely made from a Nile silt ‘grey ware’ fabric covered with a thin black glaze (above, Fig. 16), makes its appearance (Berlin 2001, 28–31), mirroring other similar wares such as Memphis black ware made elsewhere at the same time. In addition to Greek-style pottery, other imported wares were also produced as local imitations. Imitation Cypriot mortaria appear in Egyptian marl clay from at least the 6th century BC onwards; they are not restricted to Naukratis, but can be observed at numerous sites, possibly made in different local production centres (Villing 2006). Together with instances of local imitations of foreign Archaic trade amphorae (Villing 2013, 82–3) they are a surprising indication of the permeability of the Nile Delta to foreign ‘influence’ – or indeed the integration of Late Period Nile Delta society into Mediterranean networks of contact and exchange.

2.3. Lamps

Well over 300 terracotta lamps dating from the end of the 7th century BC to the 7th century AD are known from Naukratis (although a substantial proportion of these are today lost or unidentifiable), both imported and locally produced. They must be just a portion of the lamps originally encountered at the site, as we know that Petrie found 280 lamps in his first season alone (Petrie 1886, 45). Among the earliest examples are Ionian multi-nozzled sanctuary lamps (Fig. 25) found in the sanctuary of Aphrodite and perhaps indicative of ritual nocturnal gatherings.

Figure 25. Large multi-nozzled Ionian sanctuary lamp from Naukratis, first quarter 6th century BC. Museum of Fine Arts, Boston, 88.1039. Photograph © Museum of Fine Arts, Boston.

Simple single-nozzle lamps are more common, ranging from East Greek and Attic wheel-made lamps of the Archaic, Classical and Hellenistic periods to Egyptian mould-made lamps of the Ptolemaic, Roman and Early Byzantine periods. Local production is, however, attested to an earlier date, and includes local copies of simple pinched dish forms also imported from Cyprus, as well as copies of more complicated Archaic Greek types, such as a lamp with a 6th-century BC pre-firing votive inscription to the Dioskouroi (Johnston 2008, 116 and 120, fig. 3). Nearly 200 of the lamps are of Ptolemaic and Roman period Egyptian manufacture (Fig. 26) and it is likely that most were made at Naukratis itself, as confirmed by the presence of plaster moulds. Indeed, the almost 100 lamps that can be dated to the Roman and early Byzantine periods (a far greater number than the Archaic and Classical Greek/Late Period examples of the 7th to late 4th centuries BC) betray the early excavators’ collection bias towards complete objects.

Figure 26. Roman lamp with a representation of a Sothic dog in relief, probably made at Naukratis, 2nd century AD. Museum of Fine Arts, Boston, 88.1039. Photograph © Museum of Fine Arts, Boston.

It also confirms that there must have been a vibrant Roman and early Byzantine settlement at Naukratis, despite the apparent scarcity of pottery from these periods preserved from the early excavations. Two elaborate Roman bronze lamp stands and fragments of hanging lamps of 1st or 2nd century AD Egyptian manufacture presumably come from the higher status Roman houses at Naukratis. Lamps with traditional Egyptian motifs, such as the frog design associated with fertility and the inundation festival, continued to be produced locally until the 4th century AD. By the 7th century AD lamps were decorated solely with Christian iconography (Bailey 1988; 2008).

2.4. Stone vessels

Calcite vessels are one of several craft products locally made at Naukratis. Petrie discovered ‘many thousands of alabaster drill cores from tubular drilling’ (Fig. 27) as well as unfinished vessels in alabaster (calcite) in a layer in the temenos of Apollo, between the temple and the north-west corner of the temenos (Petrie 1886a, 15).

Figure 27. Drill cores from calcite alabastra from the sanctuary of Apollo at Naukratis, probably 6th century BC. National Museum of Scotland, Edinburgh, A.1886.518.27. Photograph © National Museums Scotland.

Production of these vessels probably dates back to at least the later 6th century BC, as the layer formed a pavement that (if we follow Petrie) appears to be associated with the second temple of Apollo, built in the second half of the 6th century BC. The bulk of the calcite vessels discovered in Naukratis are alabastra (Fig. 28), a shape that appears already in the late 25th Dynasty (712–664 BC) yet becomes popular only from the 26th Dynasty (664–525 BC) and continues into the Roman period.


Figure 28. Calcite (alabaster) alabastron from the area of the Great Temenos at Naukratis, 6th–5th century BC. Ashmolean Museum, Oxford, AN1896-1908-E.3695. Photograph © Ashmolean Museum, University of Oxford.


Figure 29. Gypsum ‘alabaster’ alabastron from Naukratis, terminating in the bust of a female figure, 2nd quarter 6th century BC. British Museum, 1888,0601.15.

Production is attested in a number of places in Egypt (such as Memphis) and beyond.[19] Given the shape’s popularity in the Mediterranean world, it is likely that Naukratis produced alabastra not just for local use but also for export, perhaps alongside or already filled with (local) perfume (cf. e.g. Faure 1987, 162; Shaw 2010). Egyptian perfume certainly appears to have been appreciated in Classical Greece as references to expensive Egyptian perfume and Egyptian perfume sellers in Athenian texts indicate (Athenaeus 12.553d-e; cf. Bäbler 1998, 69–77). Bowls and, unusually, squat lekythoi were also produced at Naukratis, which in addition to calcite alabastra were also recorded as finds in the site’s cemetery (Gardner 1888, 29).

A separate, special category of vessel are gypsum (calcium sulphate) alabastra the upper part of which takes the shape of a female figure; together with figured gypsum bowls (and indeed gypsum figurines most characteristically of the kouros type), they are most likely 6th century BC imports from Cyprus.[20]

Large stone vessels at Naukratis mostly had special cultic functions: in the Apollo sanctuary, fragments of several perirrhanteria were found, inscribed with dedications to Apollo (Fig. 30); similarly, Egyptian sha-basins were associated with sanctuary use. A number of stone mortars were also found. These and other stone vessels are discussed in grater detail in a separate chapter on Stone vessels.

Figure 30. Fragment of marble louterion from Naukratis inscribed with a dedication to Apollo, c. 550–450 BC. Sammlung des Ägyptologischen Instituts der Universität, Heidelberg, 3247. Photograph © Ägyptologisches Institut der Universität Heidelberg, Photo: Robert Ajtai.

2.5. Faience vessels

The 70 or more faience (glazed composition) vessels preserved from Naukratis comprise a large variety of types and dates. Egyptian New Year’s flasks (Fig. 31 discussed in greater detail in a separate chapter), which were linked to local Egyptian cults, were produced locally during the 26th Dynasty (664–525 BC), alongside scarabs and amulets.

Figure 31. Egyptian faience New Year's flask, probably made at Naukratis, late 7th–late 6th century BC. Fitzwilliam Museum, Cambridge, GR.76.1887. Photograph © The Fitzwilliam Museum, Cambridge.


Figure 32. Faience libation vase model from one of the foundation deposits of the pylon of the sanctuary of Amun-Ra, built under Ptolemy II Philadelphos (285–246 BC). British Museum, EA58329.

In contrast, aryballoi in faience (popular in the Mediterranean world especially during the 6th century BC and often believed to have been produced at Naukratis) are rare at the site, with only a few examples and fragments securely assigned and local production appears to have been limited (see the chapter on Archaic mixed style faience vessels and Webb forthcoming a). Ptolemaic and Roman faience vessels comprise a number of shapes, including several models of libation vases (Fig. 32) and offering cups from the foundation deposits associated with the monumental gate of the sanctuary of Amun-Ra.

Bowls and dishes copying Roman sigillata table ware forms produced in Memphis and popular in the 2nd century AD (Petrie 1909b, pl. 40, nos 1–4; Tomber 2006, 48, fig. 1.16, type 13-174; Nicholson 2013) are some of the chronologically latest glazed composition objects from Naukratis.

2.6. Glass

Early glass vessels include fragments of core formed closed vessels with polychrome trailed decoration (Fig. 33).

Figure 33. Top of core-formed glass flask alabastron, 6th–4th century BC. Petrie Museum, London, UC22093. Photograph © Petrie Museum of Egyptian Archaeology, UCL.

However, it seems that glass was rarely collected by the early excavators of Naukratis, as glass is exceedingly rare among extant finds: some half a dozen fragments of Roman and Late Roman glass vessels are preserved in addition to very few other glass objects.

2.7. Inscriptions

The inscriptional evidence from Naukratis was of great interest to the early excavators (Ernest Gardner in particular had an especial interest in the early history of the Greek alphabet) and so is preserved in very good numbers.

By far the largest number of inscriptions is the corpus of Greek inscriptions on pottery (Fig. 34): texts either painted (pre-firing) or incised (post-firing), often for the purpose of dedicating a vessel to a deity, but also to mark ownership or for other reasons.

Figure 34. Chian bowl from the Aphrodite sanctuary, decorated with an animal frieze and carrying a votive inscription (graffito) to Aphrodite by Sostratos: ‘Σωστρατος με ανεθηκεν τηφροδιτηι[’, late 7th century BC. British Museum, 1888,0601.456.

They are discussed in far greater detail in the chapter on Ceramic inscriptions, but a brief overview is given here. With some 2,800 preserved fragments, the body of material presented in this catalogue more than doubles what had previously been known to scholars (cf. Bernand 1970; Möller 2000a, 166–81) and for the first time the inscriptions are not presented in isolation, but as an integral part of their ceramic carriers. They constitute the largest such corpus to date known from any site, and – as even tiny scraps with minute letter traces are preserved – probably preserve the near complete body of the inscribed material as encountered by the excavators. In addition to such graffiti or dipinti (i.e. incised or painted) inscriptions that relate to a vessel’s use or trade, there are also those inscriptions that are integral to a vessel’s decoration, which are invariably painted and known as ‘vase inscriptions’.

The ceramic graffiti and dipinti can be broadly split into two equal categories: dedications and other. The dedications are for the most part to Apollo and Aphrodite, with very modest numbers to other deities. Most are simple formulaic texts, although they provide us with a range of dedicators’ names. The ‘other’ category is difficult to break down into sub-groups, although small numbers can be confidently assigned as owner’s inscriptions, commercial notations or ‘bon mots’. While dedications are mostly of the 6th century, with only a few from the 5th century, the bulk of the ‘others’ belong to the Classical and later periods, the majority on Attic or local vases. Where evidence exists Ionic script and dialect prevail, with rare exceptions. Some dedications are of particular interest by providing the name of the polis of origin of the inscriber; all those adequately preserved from the Archaic period reflect the range of poleis listed by Herodotus involved in the emporion. Particularly remarkable is the large number of dedicatory dipinti that were written before firing at the request of the dedicators, in addition to the incised dedications which could have been put on by anybody at any time before dedication. Many dipinti occur on vases made on Chios (Fig. 35), but they are also found on vessels from other North Ionian centres, and include the earliest attestation of the name ‘Naukratis’ in a 6th-century dedication to ‘Aphrodite at Naukratis’.

Figure 35. Chian pottery chalice fragment with pre-firing votive inscription naming the dedicant, Zoilos, c. 575–550 BC. British Museum, 1924,1201.820.



Figure 36. Attic black-glazed cup fragment with incised Cypro-syllabic inscription, c. 420–400 BC. British Museum, 1900,0214.17.

Such special commissions, which are also found at sites other than Naukratis (Cook and Woodhead 1952), clearly reveal the dedicator’s intention of bringing the vases with them to Naukratis for dedication to the local gods, and further underline the notion of Naukratis as a key node in well-established networks of targeted long-distance trade.

In addition to Greek, at least one ceramic inscription was Carian, incised on a now lost pottery fragment. Two Cypro-syllabic inscriptions were found on Attic pottery of the time around 400 BC (Fig. 36).

A possible Phoenician inscription appears on a 6th century BC East Greek cup while a Phoenician letter (painted pre-firing) is preserved on what is probably a trade amphora. The slimness of this crop, of course, can hardly be taken as evidence for the absence of people from these regions, as few peoples were as keen to mark their pottery as Archaic and Classical Greeks.

The pottery retained by the early excavators also includes an expected number of pieces with ‘vase inscriptions’ – pre-firing dipinto texts giving signatures of potters or painters, naming figures in the scenes (Fig. 37), or ‘kalos’ names, or indeed ‘nonsense’.


Figure 37. Fragment of Attic back-figured dinos depicting an Amazon fallen in battle, named ‘Lykopis’ in the painted inscription next to her shield. Probably from the sanctuary of the Dioskouroi, c. 575–560 BC. British Museum, 1888,0601.442.

Virtually all are on Attic vases, especially black-figured kylikes. The signed pieces of Ergotimos, the potter of the famous 6th-century Athenian ‘François vase’ that found its final resting place in an Etruscan tomb, are perhaps the most significant. Five Late Corinthian kraters with inscriptions naming painted figures are all of individual interest. A range of other objects also preserve minor inscriptions; of the approximate 1,600 amphora stamps with inscriptions, the great majority also feature alphabetic texts.

Nearly 70 Greek inscriptions on stone were recovered through excavation or by purchase from locals at Naukratis, some on small scraps of material and a few as dedications on stone objects (see the chapter on Stone inscriptions). Texts per se consist largely of epitaphs and personal or civic dedications, where enough of the text is preserved to make a judgement. The former are few, but range in date between the mid-6th century BC and the 1st century AD.

The latter includes the Hellenistic text (Fig. 38) which enabled Petrie to identify the site (see also the description of Petrie’s first excavation season) and confirms its status as polis, as well as a number of dedications to the Ptolemies notably Ptolemy II, Ptolemy XII and probably Ptolemy VI.

Figure 38. Inscription recording the polis of the Naukratites honouring Heliodoros, priest of Athena, reign of Ptolemy VI (181–146 BC); photographed by Petrie (Delta Series no. 315). Photograph © Petrie Museum of Egyptian Archaeology, University College London, PMAN2681.

Important Egyptian inscriptions are found at Naukratis on statues (Horemheb) and dedications (sha-basins), but also on stelae – although all were retrieved from the site outside of the ‘official’ excavations. The most famous is the version of the ‘decree of Sais’ that was erected at Naukratis, twin to the (virtually) identical stela set up at Herakleion-Thonis (Bomhard 2012). It regulates tax arrangements for both port cities at the beginning of the reign of Nectanebo I in 380 BC. Later examples include a fragmentary stela of year 23 of Ptolemy V Epiphanes (known as the ‘Damanhur stela’ although from Naukratis), which preserves a fragmentary version of the hieroglyphic text of the ‘Rosetta stone’,[21] as well as a stela erected in the 2nd century BC with a demotic decree of euergetism by a local community of sheep breeders (Fig. 39).

Figure 39. Demotic donation stela by the local community of sheep breeders attached to the temples of Naukratis, honouring Sheamenope for his services to the community and the divine ram, hypostasis of Amun-Ra lord of Baded; year 3 of Ptolemy II or III (281/280 or 245/244 BC). Kelsey Museum of Archaeology, Michigan, 0.2.5803. Photograph © Kelsey Museum of Archaeology, University of Michigan.

Further inscriptions appear in minor arts, such as on faience objects used for sealing or as dedications. In addition to definite finds from Naukratis, there are important inscriptions of unknown provenance that mention or are otherwise related to Naukratis, some of which at least may also derive from the site. Chief among them is a donation stele dating from 577 BC, which mentions the renewal of a donation to the local Naukratite cult of Amun-Ra Baded, thus clearly indicating the existence of the cult well before this date (see the discussion of the Egyptian presence at Naukratis in Chapter 2).[22] There is also a fragmentary donation stele from year 16 of the reign of Amasis (570–526 BC)[23] that records the (regular) donation by Neferibresaneith, a man of possible part Greek descent from Nokradj (the Egyptian name of Naukratis), of a ritual lamp to the chapel of Osiris, Horus/Harpokrates and Isis, potentially located in the area of Naukratis and Sais.

2.8. Sculpture

Some 400 pieces of sculpture are known from the site of Naukratis. They consist mostly of small scale locally made Egyptian and imported Cypriot votive figurines, with only a few pieces coming from Greece (see also the overview on Stone and terracotta figures). For the earliest period of Naukratis, it is Cypriot imports that dominate the picture. The 120 or more Cypriot figurines, made from (painted) limestone or alabaster/‘gypsum’ (calcium sulphate), mostly date from the end of the 7th to the mid-6th century BC and comprise a range of different types: male figures of the kouros type (Fig. 40), some with Egyptian features; standing male and female votaries with offerings including animals; animals such as falcons and lions and mixed beings; musicians; seated figures; and a number of intriguing groups including Isis with Horus and a bull being led to sacrifice (now studied in detail by Nick 2006 and Höckmann 2007).

Figure 40. Fragmentary Cypriot limestone statuette of a youth (kouros), c. 580–570 BC. Ashmolean Museum, Oxford, AN1896-1908-G.69. Photograph © Ashmolean Museum, University of Oxford.

Most were dedications in the sanctuaries of Aphrodite and Apollo. One, a base for a statue of Herakles with a makers’ inscription, is of a later date and seems to attest a Cypriot working at Naukratis (British Museum, 1900,0214.22). Whilst other Cypro-Classical and Hellenistic pieces are rare, their presence nevertheless suggests ongoing close links between Cyprus and Naukratis.

The 200 or more examples of local Egyptian sculpture include the spectacular double life-size statue (the largest of all known Egyptian non-royal statues) of Horemheb, ‘prophet of Min Lord of Baded’, a resident of Naukratis of mixed parentage who represents himself visually as Egyptian, but describes himself as a Greek in the hieroglyphic inscription on the statue’s back pillar.[24] The statue, presumably dating to roughly around 300BC, was apparently found in the temple precinct of Amun-Ra (although not during excavations), the same place that also yielded a large inscribed early Ptolemaic obelisk.[25] Further large-scale Egyptian-style sculptures include a sphinx (Fig. 41) and fragments of rams that probably flanked the sacred road leading from the precinct’s gate towards the sacred quay on the river (Petrie 1886, note inserted between pp. vii and viii; Gardner 1888, 13–14), as well as fragments of the extensive relief s that once decorated the Ptolemaic temple of  Amun-Ra, discussed in a separate chapter .

Figure 41. Red granite sphinx, probably Ptolemaic, from the processional road outside the Great Temenos gate, excavated and photographed by Petrie (Delta series no. 488). Photograph EES Archive © Egypt Exploration Society.

Figure 42. Limestone trial piece or sculptor's model depicting the pharaoh’s head in relief, 4th–early 3rd century BC. Museum of Fine Arts, Boston, 88.732. Photograph © Museum of Fine Arts, Boston.

Inside the precinct’s casemate building as well as elsewhere (Masson forthcoming b), a number of limestone and plaster Egyptian sculpture models or ‘trial pieces’ (Fig. 42; see Tomoum 2005) were excavated, suggesting the existence of a sculptors’ workshop on site between the 30th Dynasty (380–343 BC) and the early Ptolemaic period.

The vast majority of Egyptian sculptures from Naukratis, however, are small figurines, often quite crudely made from soft limestone, or more rarely in plaster or mudstone. Particularly common are phallic figurines representing the Egyptian child-god Harpokrates with an oversized phallus, or variations thereof (Fig. 43), but there are also reliefs of nude reclining female figures (sometimes with a child: Fig. 44), horsemen with riders, musicians, captives, animals and hybrid creatures, as well as reclining symposiasts and erotic groups (see the chapter on Stone and terracotta figures).

Figure 43. Limestone ithyphallic figure of naked youth with side-lock, probably Harpocrates (Horus), 4th–3rd century BC. Fitzwilliam Museum, Cambridge, E.85.1914. Photograph © Fitzwilliam Museum, University of Cambridge.

Figure 44. Limestone relief of reclining woman with child standing by her feet, from the town area of Naukratis, c. 6th–5th century BC. Bolton Museum, 1966.88.A. Photograph © Bolton Library and Museum Service.

Dating from the 6th–3rd centuries BC they are essentially Egyptian in type and developed out of earlier local traditions, but they also represent new developments in the Late Period (664–332 BC) of Lower Egypt types and designs influenced by foreign (Eastern Mediterranean and Levantine) motifs. It is probable that most were locally made, although some close parallels are also known from religious and domestic contexts at some other sites, notably Memphis and Saqqara, and were likely made at Saqqara (Jeffreys et. al 1988; Martin 1981; 1987; Thomas forthcoming b). Phallic figurines were particularly common in the area of Naukratis town, whereas some representations of females were also found in sanctuary contexts, notably in and around the Great Temenos, as well as some rider figures in the Hellenion. Together with (often closely related) Egyptian terracotta figurines, they indicate that a substantial proportion of the local population adhered to Egyptian religious practices.

An intriguing group among the Ptolemaic sculptures of Naukratis (although the provenance is not entirely assured) is that of some 20 carved cats of variable quality, some in the act of catching birds (Fig. 45), early Hellenistic in date and made from either marble or limestone; the group includes a base with a dedication to the Egyptian cat goddess Bubastis (Bastet).

Figure 45. Marble figure of female cat with bird in its mouth, probably from Naukratis, late 4th–3rd century BC. British Museum, 1905,0612.6.

Although seemingly playful in a Hellenistic ‘genre’ manner, the catching of the bird probably has a deeper religious meaning as a symbol for the destruction of the enemy by the goddess (Villing forthcoming b and c). Further Ptolemaic (to early Roman) sculptures include representations of kings, queens and priests.

Greek and Roman sculpture is relatively rare at Naukratis, but it does include some fine examples such as a Late Archaic relief of a hoplite, a Late Classical grave stela (although its provenance is not assured), a Hellenistic funerary banquet relief and a Hellenistic altar decorated with boukrania and leafy garlands. A number of small marble figures are also preserved, albeit fragmentary, including figures of Aphrodite Euploia, Herakles, Harpokrates and two fine representations of Eros.

2.9. Terracotta figurines

Terracotta figurines form one of the largest groups of objects from Naukratis with some 1,500 preserved fragments (mostly) of figurines, although only a very small number have been previously published and studied.[26] This is despite the fact that they provide crucial evidence for understanding the religious life of the town, the concomitant patterns of production and consumption engaged in by its mixed population and the networks in which its numerous visitors operated. As is set out in some more detail in the introductory chapter on Stone and terracotta figures, two thirds of the known figurines – over 1,000 – were locally produced, most probably at Naukratis itself, where the discovery of moulds and wasters confirm the existence of local workshops. The figurines fall into three main categories. Imported Cypriot figurines are frequent in the earliest period of the site from the late 7th century BC onwards, but very rare after the 6th century BC. They were largely replaced in the late Archaic and early Classical period by Greek figurines, in particular the popular East Greek types, rare examples of which were already present at the site from the time of its foundation. Finally, locally made figurines (which equally date back to the beginning of the site) flourished particularly from the 5th century BC and became dominant from the early Hellenistic period onwards; they include some Greek style figures, but mostly represent local Egyptian types.

Local Egyptian figurines appear at Naukratis throughout the site’s history, from the Late Period (664–332 BC), to which nearly a third of the overall 1,000 plus extant figurines can be dated, through to the Macedonian and Ptolemaic periods (332–30 BC) into the Roman period, at which point numbers drop significantly. Late Period figurines comprise a range of types often similar to limestone representations. Particularly common are figurines of ithyphallic Harpokrates and plaques depicting nude women, often standing in a shrine and probably associated with Isis-Hathor (Fig. 46); they seem to be linked with notions of fertility, the inundation festival and possibly also ‘magical’ rites, and are often found (as also elsewhere) in religious and domestic contexts.

Figure 46. Terracotta figure plaque with a nude woman in relief, c. 550–400BC, from Naukratis. McLean Museum & Art Gallery, Greenock, 1987.375. Photograph © McLean Museum and Art Gallery, Greenock.

Both types continue through to Ptolemaic times, with phallic figurines increasing in popularity and female types decreasing. The influence of wider geopolitics on local crafts can be traced in fashions such as the number of locally produced ‘Persian rider’ figurines, well known across the Delta and based on Persian period types found across the Levant and Middle East, which were replaced by riders wearing Macedonian dress at the end of the 4th century BC (Thomas forthcoming b).

From the beginning of the Ptolemaic period there is a proliferation of types, following innovations and developments in the technique and manufacture of terracotta figures. Naturalistic representations of Egyptian-style devotees, lay celebrants and priests, female figures in the popular mainland Greek Tanagra style (Fig. 47) and representations of Greek theatre actors were produced in addition to the usual deities already popular in the Late Period; mainly Isis, her son Harpokrates and his protector Bes. Other Greek and Egyptian deities and mythological creatures occasionally feature. There is a significant decrease in the quantity and variety of terracotta figures in the Roman period, with Isis, Harpokrates, Orans (woman praying) and a few other types represented.

Figure 47. Terracotta figure of a girl seated on a bench, possibly a local copy of a Greek ‘Tanagra’ type figurine, 3rd century BC, from a house in the town area of Naukratis. Museum of Fine Arts, Boston, 86.398. Photograph © Museum of Fine Arts, Boston.

Although many figurines have no recorded findspot, the ‘town’ area of Naukratis (where workshops were located on the eastern side) appears to have yielded the majority of finds; interestingly, some typical Nile Delta Egyptian plaques with nude female figures (associated with Isis-Hathor), probably dating to the 27th Dynasty (525–404 BC), have been found in the Hellenion. A few imports from Alexandria and other Egyptian sites are also identifiable.

East Greek figurines make up the largest group of imports, with over 380 fragments recorded, although as over 270 of these belong to relatively large and heavily fragmented female protomes, the number of individual pieces is actually lower. Many of the protomes were found in the Hellenion, where they were clearly popular offerings in the Late Archaic to Classical periods (Fig. 48).

Figure 48. East Greek protome of a woman, from the Hellenion at Naukratis, c. 450–350 BC. Fitzwilliam Museum, Cambridge, GR.3.1898. Photograph © The Fitzwilliam Museum, Cambridge.

Figure 49. Cypriot terracotta figure of horse and rider from Naukratis, 6th–early 5th century BC. British Museum, 1911,0606.1.

In as far as production places can be determined (including with scientific clay analyses), both Ionian and Rhodian imports are attested; a few examples were also locally made. Other East Greek figurines mostly date to the 6th century BC and include female figures and Egyptianizing figures of Ptah, which were also produced in Cyprus.

Imports from the Greek mainland as well as from other parts of the ancient world are attested, but relatively rare. Only Cypriot figurines (Fig. 49) are seen in sizable numbers, with over 100 examples known today.

Featuring among the earliest finds from Naukratis, they were produced in a number of different workshops on the island, including Achna, Arsos, Larnaka and particularly Salamis, mostly between the end of the 7th century BC and the mid to late 6th century BC. They have been found in particular in the sanctuaries of Aphrodite and Apollo, although three figurines (two of them Cypro-Classical) are also among the material from the Hellenion and at least one comes from the Hera sanctuary. Several pieces are also recorded to have been found in the ‘town’.

The picture that emerges confirms the impression of close links between Naukratis and the East Greek Mediterranean, particularly during the early period of the site, with Cyprus playing an important role as well. Greek and Cypriot figurines dominate in the site’s Greek sanctuaries, whereas Egyptian figurines were found especially in domestic contexts and around the Great Temenos, but also in the Hellenion sanctuary (Fig. 50), highlighting that Cypriot, Egyptian and even Phoenician figurines could be at home in a ‘Greek’ sanctuary.

Figure 50. Head from terracotta figurine or figure plaque of a woman, locally made in a Greek style, c. 600–450BC. Ashmolean Museum, Oxford, AN1896-1908-G.77. Photograph © Ashmolean Museum, University of Oxford.

Figure 51. Locally produced relief of Gorgon’s head for attachment to a wooden coffin, c. 332–250BC probably from the cemetery at Naukratis. Birmingham City Museum and Art Gallery, 1888A200.2. Photograph © Birmingham Museums Trust.

The diverse cultic practices at Naukratis include consistently present local elements and suggest a certain level of permeability between Greek and Egyptian religious practices. Indeed, the meaning of objects and their role in ritual practices must have been eminently fluid and negotiable in different contexts in this diverse multi-ethnic town. Local workshops appear to have flourished throughout Naukratis’ history, adapting to the specific demands of the city’s unique demographic.

A special group within the local terracotta production of the site is the colourfully painted and gilded terracotta coffin fittings (Fig. 51), of which some 170 examples are known today, which were recovered from excavations at the site’s cemetery. Mostly dating to the early Hellenistic period, they were once attached to wooden coffins, a widespread practice in Ptolemaic Egypt known also in Cyrenaica and other areas of the Hellenistic world.

The most prominent types are gorgoneia of the ‘beautiful’ Classical type, occasionally with wings attached to the head, as well as griffins, winged lions, boukrania and rosettes.

2.10. Egyptian bronzes and other votive offerings

Egyptian votive objects are among the least known and studied categories of finds from Naukratis, yet form a significant and substantial element of the site’s material assemblage. As is set out in more detail in the chapter on Bronze votive offerings,a large and important group among them are Egyptian bronze figures and votive boxes, which with over 160 figures or pieces thereof (including fittings for figures) and over 100 votive (or ‘relic’) boxes make up a significant proportion of the 1,200 or more known metal finds.[27] A range of Egyptian gods is represented in the bronze figures (Masson forthcoming b): with some 40% of figures, the Osirian triad (Osiris [Fig. 52], Isis, Harpokrates) dominates, followed by the local triad (Amun-Ra Baded, Mut Dame of Baded, Khonsu-Thot Lord of Baded), the Memphite triad (Ptah, Sekhmet, Nefertum), and Bastet and her son Mihos.

Figure 52. Bronze figure of Osiris, Late Period (664–332BC), from Naukratis. University of Pennsylvania Museum of Archaeology & Anthropology, Philadelphia, E139B. Photograph © Courtesy of the Penn Museum.





Figure 53. Bronze votive box of a lizard, probably from the bronze cache in the town area of Naukratis, late 5th–early 4th century BC. Bolton Museum, 1886.31.67.f. Photograph © Bolton Library and Museum Service.

Also represented are Horus, Anubis/Upuaut, Atum and Neith, whose cults are attested at Sais. While some figures lack a recorded findspot (and for some even the general findspot of Naukratis is not assured), many are known to come from a single location in the town, a cache of bronzes (and some other objects) discovered in a building in the southern part of the city, for which associated finds now confirm a late 5th to early 4th century BC date (Petrie 1886a, 41–2; see also Masson forthcoming a).

Like bronze figurines, votive boxes are typical Egyptian offerings from the Late and Ptolemaic periods (664–30 BC). They consist of a small bronze box topped with a figure of an animal, and are generally assumed to have contained the mummified body (or part) of the animals represented on them. The animals that appear on the boxes from Naukratis are sacred especially to Atum, such as lizards (Fig. 53), snakes, eels and cobras, with ichneumon (Egyptian mongoose) and falcon more rare.

Many boxes were found alongside bronze figures in the cache of bronzes, while others were discovered by Gardner in a trench outside the Great Temenos (Masson forthcoming b). Together with bronze models of ritual equipment (such as situlae, sistra and staffs), Egyptian offering spoons (treated in detail in a separate chapter) and inscribed votive sistra offered by Psamtik III (526–525BC) and later rulers (Fig. 54), they represent traditional Egyptian cultic activity at Naukratis at least from the later Saite, i.e. the 26th (664–525 BC) through to the 27th (525–404 BC) and later dynasties.

Figure 54. Faience sistrum handle fragment with hieroglyphic inscriptions mentioning Amun and Mut of Baded, reign of Ptolemy II (283–246 BC). Musée du Louvre, Paris, E6268. Photograph © Musée du Louvre/Georges Poncet.

The flourishing of the (originally Saite) local sanctuary of Amun-Ra of Baded and associated gods in the early Ptolemaic period is confirmed by a number of large offerings or pieces of cultic equipment in stone that were found associated with the Great Temenos, including fragments of libation basins (sha-basins) and a campaniform base (for a Hathoric basin?) with dedications to Mut and Hathor of Baded (Jansen-Winkeln 1997).

2.11. Scarabs, sealstones and amulets

Egyptian (and Egyptianizing) scarabs and amulets as well as their moulds form a large group of finds at Naukratis, totalling over 1,000 objects. The majority are faience amulets of Egyptian type, which represents a wide range of Egyptian deities and symbols (Masson forthcoming b). Excluding scarabs and scaraboids, the widely popular wedjat-eyes that make up about a quarter of the examples; a third represent divinities such as Bes (Fig. 55), Pataikos, Taweret, Nursing Isis or Harpokrates, all generally linked with fertility and/or protection of childhood or against dangerous animals.

Figure 55. Faience amuletic pendant with the head of Bes, probably Late Period (664–332 BC), from Naukratis. Petrie Museum, London, UC52848. © Petrie Museum of Egyptian Archaeology, UCL

Figure 56 Faience figure of a kneeling man, from the sanctuary of Apollo, possibly produced at Naukratis, early 6th century BC. British Museum, EA58307.

A second albeit smaller group are amuletic figures of Greek-Egyptian mixed style as they are also known from Archaic East Greek sanctuaries. They comprise primarily nude females and males, mostly in a standard ‘striding’ (walking) or ‘presenting’ (kneeling) pose (Fig. 56), sometimes represented as a musician (cf. Webb forthcoming a and b).

While the findspot is not known for most of these objects, those for which such information exists suggest that the non-Egyptian figurine types were concentrated particularly in the sanctuaries of Apollo as well as of Aphrodite, although some were also found around the scarab factory and in the town (Petrie 1886a, 38). A group of Egyptian amulets was also deposited, alongside other votive objects, near the Great Temenos (Masson forthcoming b).

The ‘faience factory’ excavated by Petrie just outside the Aphrodite sanctuary is the main findspot for the well over 100 scarabs in faience and Egyptian blue paste known today from the site (Fig. 57).

Figure 57. Scarab in blue paste, a typical product of the ‘scarab factory’ at Naukratis, early 6th century BC. Museum of Fine Arts, Boston, 88.866. Photograph © Museum of Fine Arts, Boston.

Figure 58. Terracotta mould for a scaraboid in the shape of a Black African head, early 6th century BC, from the ‘scarab factory’ at Naukratis. Redpath Museum, Montreal, 2484.03. Photograph © Redpath Museum, McGill University.

The products of the ‘factory’ were widely distributed across the Mediterranean and Black Sea region (Gorton 1996), but were also used locally, as is confirmed for example by one preserved seal impression. Hundreds of terracotta scarab-moulds were also found in the ‘factory’, of which some 200 are extant. While moulds for ‘regular’ scarabs are most common, there are also moulds for scaraboids in the shape of Black African heads (Fig. 58), rams or plain discs; well over 130 examples of scaraboids made in such moulds were found at the site.

The designs incised on the flat base of scarabs and scaraboids comprise a wide range of motifs, Egyptian, Egyptianizing, Hellenizing, ‘Orientalizing’ or hybrid in style. Some scarabs feature cartouches of pharaohs, suggesting a period of production over several decades in the early part of the 6th century BC. Some two dozen amulet moulds (comprising a variety of Egyptian types) are also preserved among finds in the ‘scarab’ factory, which indicate that scarabs as well as Egyptian style amulets were locally produced. Faience wasters further confirm the local production of falcons and other Egyptian amulets (Masson forthcoming b).

In addition to faience, many Egyptian amulets (Fig. 59) and scarabs in stone have also been found at Naukratis.

Figure 59 Stone amulet in the shape of a wedjat eye, Late Period (664–332BC), from Naukratis. Petrie Museum, London, UC52358. Photograph © Petrie Museum of Egyptian Archaeology, UCL.

Figure 60. Glazed steatite scarab with a stylized head of Hathor flanked by red crowns, from the ‘scarab factory’ at Naukratis, possibly Second Intermediate Period or Middle Kingdom (first half 2nd millennium BC). City Art Gallery & Museum, Bristol, H5115.8. Photograph © Bristol Museums, Galleries & Archives.

Over 70 scarabs and scaraboids in steatite (as well as some in other types of stone) are preserved from the site. Their interpretation remains problematic, as they comprise numerous different types characteristic of the Second to Third Intermediate Periods, particularly the Ramesside period (Fig. 60).

The fact that almost all these examples appear to predate the foundation date for Naukratis suggested by the majority of the evidence from the site – and in fact account for the majority of such early Egyptian objects found there – does not necessarily have to be taken as evidence for an earlier Egyptian presence at the site. They could have been brought to Naukratis from other sites either in antiquity (perhaps for use as models for local workshops?) or in modern times (a phenomenon not unknown at 19th-century excavation sites in Egypt) (Masson forthcoming c).

Similar scenarios could also be imagined with regard to some of the other small and highly portable objects, including gems and sealstones. Most notable among the small but very mixed corpus from Naukratis in this respect is a North Syrian cylinder seal of the 18th to 17th century BC (Fig. 61), which may have reached Naukratis as an heirloom, perhaps brought by a Persian official (Amiet 1994), thereby reflecting the colourful political and economic history of the site.

Other seals and seal impressions (Fig. 62) found at Naukratis provide a glimpse of specific individuals linked with Naukratis, from 26th Dynasty (664–525 BC) court officials at the capital Sais to Achaemenid administrators stationed at Naukratis during the Persian Periods (525–404 and 343–332 BC) or Roman and Byzantine wine traders and estate owners.

Figure 61. Syrian hematite cylinder seal from a house in the town of Naukratis, probably made in Aleppo, late 18th–early 17th century BC. British Museum, 1886,0401.1722.

Figure 62. Clay seal impression with hieroglyphic inscription naming the ‘prophet’ Hor, a high official in the Egyptian hierarchy of priests. From Naukratis, late 7th–late 5th century BC. British Museum EA27575.

2.12. Jewellery and mirrors

Jewellery (with the exception of amulets) does not constitute a particularly large group of finds at Naukratis, which is perhaps in part due to a scarcity of Greek females at the site in the early period, in addition to excavation practices and focus. However, the assemblage is remarkable for yielding some of the most interesting Roman finds from the site. A small number of ear-rings, finger-rings, necklaces, bracelets and other elements are all attested. A group that stands out particularly is a find made by locals digging independently of Petrie’s work in a house ‘in the south-west of the town’ (Petrie 1886a, 43–4, pls 27–8). In addition to pieces of silver bracelets, a silver mirror, a gold chain and other finds, the cache contained several pieces of gold (Fig. 63).

Figure 63. Roman gold jewellery and silver mirror found in a house in the town of Naukratis, 1st century AD. British Museum, 1886,0401.1749, 1756–1765.

The most spectacular among them is a diadem of sheet gold embossed with divine figures and a Greek inscription. The inscription names Tiberius Claudius Artemidorus, quite possibly the famous Roman period athlete born at Tralleis in Asia Minor, thus highlighting the site’s continuing significance in this period. In addition to the silver mirror from the cache, a small number of bronze mirrors (including Egyptian types) have been found elsewhere at the site, such as in the cemetery and the sanctuary of Apollo. A Phoenician-type glass bead depicting the head of a bearded male (Museum of Fine Arts, Boston, 86.216) stands out as distinctive, but many of the Naukratis glass beads are of more generic types and difficult to date precisely.

2.13. Coins

Naukratis yielded over a thousand coins from many regions of the ancient world. From Petrie’s first season alone, B.V. Head (1886a and b) lists over 900 examples. While over half of these are Imperial bronze coins of Alexandria, the rest is quite varied and reflects both the diverse commercial contacts the port entertained throughout its history, as well as more general patterns of coin circulation and local demand for bullion. Further coins were added to the corpus during Gardner’s and Hogarth’s seasons,[28] as well as by the recent American survey (Coulson 1996, 145–6). The spectrum ranges from Late Archaic and Classical East Greek, Lycian, Cyrenaican (Fig. 64), Aeginetan and Attic (Fig. 65) coins to Ptolemaic and (numerous) Roman issues as well as rare Byzantine and later examples. For a period in the late 4th century BC the city also issued its own coinage, presumably under Kleomenes, overseer of Egypt’s finances for Alexander the Great from 331 BC (Le Rider 1997, 91–3; Bresson 2000, 75).

Figure 64 Silver tetradrachm of Cyrene (obverse) showing the local nymph Cyrene and a silphium plant, c. 525–480BC. British Museum, 1886,0802.12.

Figure 65. Silver Athenian tetradrachm (reverse) showing the owl of Athens, 2nd half 5th century BC. Castle Museum, Nottingham, NCM 1888-61. Photograph © Nottingham City Museums & Galleries.

2.14. Weights

An important category among the finds of Naukratis is weights, which were a particular passion of Petrie’s and are therefore preserved in large numbers, with over 1,000 extant examples (Fig. 66)[29].

Figure 66. Weights from Naukratis, from left to right: Three Egyptian stone weights: 2 qedets and (twice) 1 deben; (Bolton Museum, 1925.20.1–3; photograph © Bolton Library and Museum Service); small discoid stone weight: half a shekel (Oriental Institute, University of Chicago, E2911; photograph © Oriental Institute Museum, University of Chicago); teardrop stone weight (Oriental Institute, University of Chicago, E2882; photograph © Oriental Institute Museum, University of Chicago).

The rich and varied assemblage ranges from Egyptian to Greek and Near Eastern weights in various materials (mostly different types of stone, but one quarter in metal) and forms vital evidence for the trade and exchange between Egypt and the Near Eastern and Mediterranean worlds that passed through Naukratis.

2.15. Weapons and armour

Weapons and armour are rare finds at Naukratis. If the Corinthian helmet in Brussels (Musées Royaux d’Art et d’Histoire inv. A.1948) were indeed from the site (which seems unlikely), it would be a unique find as otherwise the category is mostly confined to arrow heads. The nearly 100 known examples (most but not all of which are currently traceable) range from the 26th Dynasty (664–525 BC) Egyptian to Greek and ‘Scythian’ types (Fig. 67), with the majority dating from the Late Period (664–332 BC), and there are only a few from examples from the Macedonian and Ptolemaic (332–30 BC) and Roman periods.

Figure 67 ‘Scythian’ arrow-head from Naukratis, probably 5th century BC. British Museum, 1909,1201.10.

Some examples were found in the sanctuary of Apollo, and religious dedicatory practices may also be the most likely explanation for those without recorded context. This picture contrasts with the richer evidence from sites such as Thonis/Herakleion, and may in part at least be due to ancient recycling practices, find preservation conditions, excavators’ retention patterns as well as loss of provenance information in museums.

2.16. Tools and implements

Tools and implements make up a substantial, if highly diverse category of the finds from Naukratis. Many preserved examples are bronze, but iron would have been more prominent if many iron finds made by Petrie and brought to London had not apparently been disposed of by the British Museum upon receipt (see the discussion of the finds and their dispersal). Various bronze and especially iron tools for agricultural and craft activities – including saws (Fig. 68), chisels, sickles, hoes, picks and scrapers – were found across the site; according to Petrie (1886a, 39 with pl. 11) these were frequent particularly in the lowest levels of the town, alongside iron slag and specular iron ore, suggesting local production.

Figure 68. Iron saw from Naukratis, possibly 6th century BC. British Museum, 1888,0601.697.

A saw and several nails are among the examples still preserved today. More intricate tools include a rare and early example of a surgical probe, found in an Early Hellenistic burial by Gardner (1888, pl. 16, no. 17). Among terracotta finds, a group of Hellenistic kiln supports (Fig. 69) – found both in early and more recent fieldwork – is indicative of a local pottery workshop (cf. Leonard 2001, 191–3).

Figure 69. Ptolemaic kiln support fragment, terracotta with pre-firing inscription ‘Αχιλ’, possibly 2nd century BC. British Museum, 1910,0222.233.

Further tools and implements include Greek limestone sundials, bronze strigils (one example from the cemetery), (Roman) metal keys and a dozen bells (mostly made of bronze, but one also of silver) – several of typical Egyptian type – which may have had a ritual or amuletic function. A small yet highly diverse group of loomweights is particularly noteworthy as rare attestations of women’s work; they comprise both imported and locally made examples of Greek type (Fig. 70), to which the recent fieldwork now adds Egyptian-type loomweights as well as fragments of Archaic Greek pottery re-used as likely loomweights (Thomas and Villing forthcoming).

Figure 70. Greek-type loomweight, 6th–4th century BC. McLean Museum & Art Gallery, Greenock. Photograph © McLean Museum and Art Gallery, Greenock / Inverclyde Council.

Figure 71. Barbed iron fishing hook, probably 5th–4th century BC, from the town area of Naukratis. World Museum, Liverpool, 9,9,86,126.b. Photograph © National Museums Liverpool (World Museum).

Unsurprisingly for a port on the Nile, a variety of fishing equipment has also been preserved, including fish-hooks (Fig. 71), net-weights, sinkers and a netting needle. Of particular importance are (rare) elements of a ship’s sailing equipment in metal and other materials, such as horn brail rings (used to guide ropes operating a sail), indicating that sea-going vessels reached Naukratis (Thomas forthcoming a).

2.17. Shells and other faunal remains

About 100 animal remains, including shells and bones, are still preserved from the early excavations at Naukratis. It is clear that these represent only a small proportion of what was originally encountered based not only on what we know from other archaeological sites, but also from Naukratis itself, where more recent fieldwork (by the British Museum and the American mission) has yielded substantial numbers of such remains. The majority of the finds preserved from the early fieldwork consists of unworked faunal remains. This includes a number of mammal bones (Fig. 72) as well as a variety of shells from Petrie’s fieldwork at the site, including shells from the Mediterranean and Red Sea and land snails likely consumed as a delicacy in Egypt by the Roman period.

Figure 72 Leg bone (right humerus) from sheep or goat with butchering marks from Naukratis. British Museum, 2011,5009.3.

Figure 73 Tridacna shell fragment, engraved decorating featuring wings, lotus flowers and buds. 7th–6th century BC. Ashmolean Museum, Oxford, AN1896-1908-G.451. Photograph © Ashmolean Museum, University of Oxford.

A small number of fragmentary tridacna shells are noteworthy among the worked faunal remains (Fig. 73), at least one of them from the Hera sanctuary (Möller 2000a, 163–6).

Originating from the Red Sea they probably acquired their elaborate designs in the 7th century BC Syro-Palestinian realm and were widely distributed particularly in the East Greek world, including to Samos, Miletos, Rhodes and Cyrene (cf. Furtwängler 2011). Unworked tridacna shells are also reported among finds from the cemetery of Naukratis (Gardner 1888, 29). A fragment of a decorated ostrich egg (ultimately originating from Sudan) was recovered from the Apollo sanctuary, and several further undecorated fragments are preserved from unspecified findspots. They are rare evidence of the rich array of consumables that traders once brought to Naukratis from far and wide, from the Red Sea littoral, Arabia, Africa and the eastern Mediterranean. Many of these goods were perishable and are now lost, and the preserved material assemblage can only occasionally give a small glimpse of this once important aspect to life and trade at ancient Naukratis.

  • [1] - The problems presented by the assemblage of Naukratis are discussed in more detail in Villing forthcoming a.
  • [2] - Waldstein 1902, 57, 60–1; White in Schaus 1985a, xxi; cf. also Stissi, 2002, 243–4.
  • [3] - It can be estimated to at least 150 sherds per cubic metre: 4,000 sherds were counted in an excavation area of 50m3, a mixture of sherd-rich surface (re)deposits and less rich primary strata; however, in the (very sherd-rich) mixed surface-deposits only diagnostic sherds were counted. John Hayes (pers. comm.) estimates some 700–800 sherds per cubic metre on average for Greek habitation areas.
  • [4] - In total, of course, the number of sherds contained in that part of the settlement ‘mound’ which was dug up between the 19th century and now must be far larger: by comparing 19th-century with current topographical data, we can calculate that approximately 350,000m3 of earth was moved in this period; even a very low average sherd density of around 100 sherds per cubic metre would therefore give a total number of some 35 million sherds.
  • [5] - The figure is based on a complete sherd count of finds made in 1989 in the 7th–6th century BC habitation quarter on Kalabaktepe (pers. comm. M. Heinz). In Attica, Rotroff (1999) observed a ratio of 2.8 and 2.9% of decorated fine wares in the Attic Dema and Vari houses, respectively, and 8.4% in a domestic well in the Athenian Agora, with undecorated fine wares making up 36–44%.
  • [6] - Cyrene: White in Schaus 1985a, xxi; Didyma: Bumke 2013, 339. Stissi 2002, 246–7, notes slightly lower ratios in other sanctuaries.
  • [7] - Hogarth nevertheless included a whole – if brief – chapter specifically on ‘unpainted pottery’ in his published report of the 1903 season (Edgar 1903).
  • [8] - The vast majority are coarse wares of Late Period (664–332 BC) to Roman date, with a peak in the Macedonian and Ptolemaic periods (332–30 BC). The material recovered by the American mission comes mostly from surface survey and excavations in what the excavators described as a Ptolemaic pottery rubbish heap: Coulson 1996; Leonard 1997, 2001; cf. also Thomas and Villing 2013.
  • [9] - What is not particularly effective and might in fact compound existing biases, however, is the application of any of the standard methods designed to give a better understanding of the actual proportions of different types of vessels from recent fieldwork, such as calculating estimated vessel equivalents or weighing (cf. Verdan, Theurillat and Kenzelmann Pfyffer 2011), as the assemblage of Naukratis already represents a highly selective, and only in certain aspects representative, sample.
  • [10] - A view summarized by Möller 2000a, 118–19: ‘Naukratis was a purely Greek empórion, the assumption of an Egyptian quarter being based on misinterpretation. The Great Temenos was not built until the Ptolemaic period […].’
  • [11] - For the date for the first Apollo temple, see now Dirschedl 2013, 172. Koenigs in Höckmann and Koenigs 2007, 327and 340–1 is more cautious in his dating. See also Möller 2000a, 96–9.
  • [12] - Edgar 1922, 2–6; Munro 1973; Yoyotte 1993–4, 684–9.
  • [13] - Nearly all the pottery still extant from early fieldwork can be categorized at least approximately by means of the standard scholarly regional and chronological classificatory systems. Only a very small number elude classification altogether; among them are also some finds that although recorded in registers or publications are today either lost or have not yet been located.
  • [14] - For summaries and analysis of the evidence as known previously, see Möller 2000a, 119–47; Kerschner 2001; Schlotzhauer and Villing 2006; see also Venit 1982 and Schlotzhauer 2012.
  • [15] - Amphorae from Naukratis have previously been surveyed in Gantès 2007, although the find spectrum known then was still limited.
  • [16] - Hogarth 1905, 107; Edgar 1905, 25 fig. 5. See also the discussion in Möller 2001a.
  • [17] - Leonard 1997, 29; cf. Leonard 2001, 30, with the discussion in Thomas and Villing 2013, 99–101 and Spencer 2011.
  • [18] - Local production of Archaic Greek-style pottery was confirmed by clay analysis and is discussed for example in Schlotzhauer and Villing 2006; Mommsen et al. 2006; Schlotzhauer 2012; Mommsen et al. 2012.
  • [19] - Bissing 1939; 1940; Hölbl 1979, 240–55; Aston 1994, 50.
  • [20] - On the type, see Bubenheimer-Erhart 2012, 24–31; Höckmann 2007, 137–9; Jenkins 2001, 172–3; Riis 1956.
  • [21] - Gardner 1888, 83; Kamal 1904–5, 183–7, pls 62–3; Yoyotte 1993–4, 690–2.
  • [22] - Berlin, Ägyptisches Museum und Papyrussammlung Inv.-Nr. ÄM 7780; cf. De Meulenaere 1993, 16–18, pl. 1.
  • [23] - St Petersburg, Hermitage 8499; cf. Collombert 2004.
  • [24] - Borchardt 1911–36, vol. IV, 120–1 no. 1230, pl. 171; Baines 2004, 39, 49–50.
  • [25] - Edgar 1922, 1; Yoyotte 1982–3, 130, no. 2; Yoyotte 1993–4, 684.
  • [26] - Notably in the publications by Petrie 1886a and Gardner 1888 and some more detail by Gutch 1898–9, for material from Hogarth’s first season, notably in the Hellenion.
  • [27] - Masson forthcoming b. Included in the recent study of Late Period bronze figures by Weiss 2012 are 121 examples of these finds.
  • [28] - Unfortunately the state of preservation of this material is patchy: only some of the coins from Petrie’s season and few from Gardner’s season are traceable. Nearly 140 coins from Hogarth’s fieldwork are preserved in Oxford.
  • [29] - Petrie 1886a and Gardner 1888 give information and weights for only about half of the assemblage; the group is currently being restudied by Aurélia Masson.