Version classiqueVersion mobile

AGRUMED: Archaeology and history of citrus fruit in the Mediterranean

 | 
Véronique Zech-Matterne
, 
Girolamo Fiorentino

The earliest evidence for citrus in Egypt

Charlène Bouchaud, Jacob Morales, Valérie Schram et Marijke van der Veen

Texte intégral

We would like to thank Hélène Cuvigny and Adam Bülow-Jacobsen for their kind permission to refer to an unpublished ostracon, their invaluable help in sourcing the papyrological literature and for their comments on earlier drafts of this paper. Thanks are also due to Alan Clapham and Ursula Thanheiser for sharing with us their unpublished archaeobotanical data, to Anne Boud’hors for giving us useful information on the Coptic documentation, to Sean Goddard for creating figure 1, and to Reinder Neef and Wendy Smith for their help in accessing literature.

  • 1 Amigues 2005; Zohary et al. 2012: 146-47.
  • 2 Langgut et al. 2013; Langgut 2014.
  • 3 Van Zeist et al. 2001.
  • 4 Bui Thi Mai, Girard 2014: 174.
  • 5 Coubray et al. 2010; Pagnoux et al. 2013.
  • 6 Botto, Salvadei 2005.

1While the cultivation of citrus fruits in Egypt is thought to have a long history, there is surprisingly little known about its earliest introduction. It is generally assumed that citron (Citrus medica) is the first of the citrus fruits to reach the Mediterranean, probably via Persia and Mesopotamia in the late 4th century BC, associated with Alexander’s campaigns.1 Recent pollen evidence from the Mediterranean produces similar dates, such as at the 5th or 4th century BC Royal Persian garden at Ramat Rehel, near Jerusalem,2 and at Carthage, Tunisia, in deposits dated to the 4th century BC.3 Citrus pollen finds from the Greek colony at Cumae near Naples are less precisely dated. Indeed, the first citrus pollen in this core derive from an undated deposit (level -641 cm), with radiocarbon dates below this level (at -656 cm) giving an 896-657 cal BC date and above it (at -575 cm) an 38 cal BC-123 cal AD date.4 According to some scholars, the proximity of the layer containing Citrus pollen to the oldest of these two dates suggest an early distribution of citrus into the Mediterranean.5 We are, however, of the opinion that this record cannot be used to accurately date the earliest introduction of citrus fruits into the Mediterranean, other than to say, as the authors of the original report do, that it was present by the 1st century BC. Several radiocarbon dates from layer -641 would be needed to establish the real antiquity of this record. Finally, chemical analyses of organic residues in a 6th century BC wine jug at the Punic necropolis of Monte Siraï, Sardinia, identified citrus as one of the components,6 though this does not prove local cultivation of the tree.

  • 7 Loret 1891: 247-249.
  • 8 Germer 1985: 106.
  • 9 Keimer 1984.
  • 10 Andrews 1961.

2This raises the question of when the fruit was introduced into Egypt. While Loret7 suggested the 15th century BC (18th Dynasty), he offers no convincing evidence for this. Germer,8 referring to work by Keimer,9 puts forward a possible Ptolemaic date, on the basis of a description of a complete fruit and longitudinal section thereof from Thebes, which Keimer describes as of uncertain date, but possibly Ptolemaic. Finally, Andrews10 suggests that the cultivation of citron trees may have been introduced into Italy by gardeners from Alexandria, after the annexation of Egypt into the Roman Empire in 30 BC, and that it was Jewish people who brought the citron to Egypt prior to that (there was a large Jewish colony in Alexandria). Much new archaeological and papyrological evidence has come to light since then, and here we review this evidence in order to determine the likely date for the introduction of citrus species into Egypt and Egypt’s possible role in the dispersal of the fruits into the Mediterranean.

1. Archaeobotanical remains of citrus

3Due to the arid climatic conditions in Egypt the preservation of botanical remains of foods and other plant material at archaeological sites is extremely good. Unlike most of North Africa, the Middle East and Europe, botanical remains are found in both charred and desiccated form, the latter often including many vegetative parts and looking almost modern. All the remains of Citrus found in Egypt to date have been recovered in desiccated state. They were derived from eight archaeological excavation projects, representing eleven chrono-cultural contexts or ‘sites’, including seeds at seven different sites, rinds at three, fruits at two and leaves at one (table 1, fig. 1).

Table 1 - Citrus finds at archaeological sites in Egypt.

Table 1 - Citrus finds at archaeological sites in Egypt.

Numbers refer to the minimum number of items present.
References: a) Van der Veen 2011; b) Van der Veen 2001; c) Van der Veen, Tabinor 2007; d) Germer 1988; e) Thanheiser 2002; f) Täckholm 1961; g) Clapham personal communication; h) Thanheiser personal communication.

Fig. 1 - Map of Egypt listing the sites mentioned in the text.

Fig. 1 - Map of Egypt listing the sites mentioned in the text.

4The earliest archaeological remains recovered so far date to the 1st century AD and the latest to the 14th-15th century AD. The earliest and most common citrus species is citron, Citrus medica, recovered from ten sites, while lime, Citrus cf. x aurantifolia, was found at only one site, dated to the 11th-13th century. Possible remains of one or two additional species of Citrus are discussed below. Most of the citrus remains originate from three distinct regions, namely the Eastern Desert, the Western Desert and Nubia. Here the sites at which they were found are discussed in chronological order.

1.1. Myos Hormos – Quseir al-Qadim (Red Sea coast/Eastern Desert) - Early Roman

  • 11 Whitcomb, Johnson 1979; 1982.
  • 12 Peacock, Blue 2006; 2011.

5The archaeological site of Quseir al-Qadim is located on the Red Sea coast of Egypt, some 8 km north of the modern town of Quseir. In antiquity the archaeological site was known as Myos Hormos, one of the two principal ports for Indian Ocean trade. By the medieval Islamic period the site was known as Kusayr, here discussed as a separate site, see below. The harbour was located in an ancient lagoon, created by the drainage of two wadi systems. The site functioned as a transport hub for Eastern trade in spices and other luxury goods coming from India and further afield. An ancient road through the Wadi Hammamat connected the port with Coptos in the Nile Valley, and from there the spices and other goods were transported to Alexandria and onwards to Rome. The site was excavated twice, first by an American team between 1978-198211 and, again, between 1999-2003, by a British team from the University of Southampton.12 The excavations revealed a large settlement, remains of the harbour, as well as a vast array of artefacts, texts (ostraca and papyri), biological remains of foods, textiles, ropes and much more. The preservation of organics at the site was excellent and an extensive sampling strategy was implemented to recover the botanical and archaeozoological remains.

  • 13 For further detail, see Van der Veen 2011.
  • 14 Van der Veen 2011; Van der Veen, Morales 2014.

6During the British excavations 43 samples were collected from the Roman refuse and midden deposits found across the settlement. From each securely dated context two litres of sediment were dry-sieved and subsequently floated over a 0.5 mm mesh and 20 litres were dry-sieved over a 2 mm mesh, the latter to ensure that rarer larger items were recovered. Additionally, botanical remains visible with the naked eye were collected by the workers, the so-called ‘handpicked’ material.13 Most of the botanical material represents the left-overs of food and fodder consumed by the people and animals working at the ports. Food and fodder was brought to the site from the Nile Valley on a regular basis (see section 3.2). Most of the imports from India were for onward transport and were not consumed in the port, though small amounts of black pepper (Piper nigrum), rice (Oryza sativa), mung bean (Vigna radiata), coconut (Cocos nucifera) and belleric myrobalan (Terminalia bellirica) were recovered. The latter four probably represent some of the foods the Indian sailors took with them on the journey. In total some sixty economic plants (mostly food plants) were found, including staples such as cereals, pulses and dates, but also luxuries such as nuts (e.g. Juglans regia, Pinus pinea), fruits such as peach (Prunus persica) and pomegranate (Punica granatum), vegetables and herbs such as artichoke (Cynara cardunculus), coriander (Coriandrum sativum), fennel (Foeniculum vulgare), dill (Anethum graveolens), cumin (Cuminum cyminum) and aniseed (Pimpinella anisum).14

  • 15 Moore 2001; Watson 1983.
  • 16 See also Van der Veen 2011: 83-89.

7The citrus remains from Myos Hormos are composed of twelve desiccated seeds, eight of which have been identified as citron, Citrus cf. medica (fig. 2a). The other four seeds were poorly preserved and thus have not been identified beyond genus level, Citrus sp. They were found at three separate parts of the site, all containing domestic refuse: in Trench 2D (1st-2nd century AD), Trench 6GHJ (late 1st-mid 2nd century AD) and Trench 8 (2nd-early 3rd century AD). Radiocarbon dates on two of the seeds (from Trenches 2D and 8) gave similar dates of 1-130 cal AD and 70-240 cal AD (Table 1). The seeds are in both shape and size similar to those of the sour and sweet orange (C. aurantium and C. sinensis), but the presence of C. medica rind in a contemporary site nearby (Mons Claudianus, below) and the fact that the sour orange is not thought to have been introduced into the Mediterranean until the medieval period and the sweet orange not until the 16th century,15 makes the identification of citron the most likely. Note that the full identification criteria for remains from this site are listed in Table 2 and discussed below, under section 2.6 Kusayr.16

Fig. 2 - Archaeobotanical remains of citrus seeds.

Fig. 2 - Archaeobotanical remains of citrus seeds.

a. Type A - citron, Citrus cf. medica, first two from Myos Hormos, 1st-3rd c. AD, second two from Kusayr, 11th-13th c. AD (from Van der Veen 2011, fig. 3.4); b. Type A - Citrus cf. medica from Mons Porphyrites, 2nd century AD (from Van der Veen, Tabinor 2007, fig. 4.9); c. and d. Type B, cf. lime/lemon, Citrus cf. x aurantifolia/limon from Kusayr, 11th-13th c. AD (from Van der Veen 2011, fig. 3.4). See Table 2 for identification criteria (photographs: Jacob Morales).

1.2. Mons Claudianus (Eastern Desert) – Early Roman

  • 17 Maxfield, Peacock 2001a; Peacock, Maxfield 1997.
  • 18 For more details, see Van der Veen 2001.
  • 19 Bülow-Jacobsen 1997; 2003; Van der Veen 1998; 2001.

8Mons Claudianus is located in the heart of the Red Sea mountains, about 700 m above sea level. The site is a quarry settlement occupied from the 1st to the 3rd century AD, known for its granodiorite, which was used for imperial building projects in Rome (e.g. the Pantheon, Trajan’s forum). The site is composed of the quarry field itself, extending over some 750 ha and with 130 individual quarries, as well as a fortified settlement, the main residential and administrative centre, which contained accommodation, a communal kitchen, bath house and temple, stables, a granary and a well. Both inside and outside the settlement were large refuse or midden deposits, full of artefacts, texts (ostraca), biological remains of food, textiles, etc.17 Both the types of deposits sampled and the sampling strategy are similar to those described above for Myos Hormos. Preservation of the archaeobotanical remains was excellent and some 57 food plants were recovered, comprising a similar range to that found at Myos Hormos. Most of these foodstuffs were brought in from the Nile Valley, in antiquity a journey of ca. seven days, and these include both staples (cereals, pulses, dates), and more expensive items, the latter probably luxuries for the higher ranking officers, such pine nut (Pinus pinea), walnut (Juglans regia), hazelnut (Coryllus avellana), black pepper (Piper nigrum), artichoke (Cynara cf. scolymus), etc.18 Remarkably, there is both archaeobotanical and textual evidence that small gardens were kept near the site for the growing of green vegetables.19

  • 20 Van der Veen 2001: 220.
  • 21 For more details, see Van der Veen 2001.

9One desiccated fruit and twenty-three seeds were identified as citron remains (Citrus cf. medica) according to their size and shape, and they are classified as potential luxury items.20 The fruit is not complete, but rather represents half a fruit with the seeds visibly embedded in the fruit pulp (fig. 3a). It was not possible to determine with certainty whether it was deliberately cut in half or damaged after deposition. The remains have been found in six separate midden deposits, both inside and outside the main fort. Most of the seeds (N=17) come from a large midden dated to the early 2nd century that contained much kitchen waste (onion skin, vegetables, burnt bread), snack foods and table waste (fruit stones) and possible fecal material.21 Finding fruit as well as seeds indicates that fruits were brought to the site, not dried segments of rind (see discussion in section 4.2 below).

Fig. 3 - Archaeobotanical remains of citrus fruits and rind.

Fig. 3 - Archaeobotanical remains of citrus fruits and rind.

a. near-complete fruit of citron, Citrus cf. medica, with seeds still in situ, from Mons Claudianus, 2nd century AD (from Van der Veen 2011, fig. 3.5); b. cluster of seeds embedded in fruit pulp, Citrus cf. medica, from Mons Porphyrites, 2nd century AD; c. rind of half a citron, Citrus cf. medica, from Kusayr, 11th-13th century AD (from Van der Veen 2011, fig. 3.5). See Table 2 for identification criteria (photographs: Jacob Morales).

1.3. Mons Porphyrites (Eastern Desert) – Early Roman

  • 22 Maxfield, Peacock 2001b; Peacock, Maxfield 2007.
  • 23 For more details, see Van der Veen, Tabinor 2007.

10Mons Porphyrites is another quarry complex, this one dedicated to the extraction of purple porphyry, used in imperial sculptures, sarcophagi and small columns. It differs from the Mons Claudianus complex in that it lies deeper into the mountains and is more difficult to reach from the Nile valley than Mons Claudianus (but is closer to the Red Sea), and the spatial organization of the site is more dispersed. The complex was in use during the early 1st to late 2nd/early 3rd century, and again in the 4th to early 5th century.22 The archaeobotanical sampling strategy was identical to that at Mons Claudianus and Quseir al-Qadim. The majority of the samples originate from domestic refuse deposits in some of the rooms, from large midden areas outside the main fort, from an earlier settlement, as well as from a way-station (Badia) on the route to the Nile Valley, and a quarry village high in the mountains.23 The late Roman remains are discussed below, as a separate chrono-cultural context (or site).

  • 24 Van der Veen, Tabinor 2007.

11The plant remains were not as well preserved as those at Mons Claudianus or Myos Hormos, but still mostly preserved by desiccation. Some fifty food plants were recovered, including the usual staples (cereals, pulses, dates), as well as luxuries, in this case including fruits such as persea (Mimusops schimperi) and peach (Prunus persica), nuts such as hazelnut (Coryllus avellana), almond (Amygdalus communis) and pine nut (Pinus pinea), and vegetables such as lettuce (Lactuca sativa). Thus, the more remote location had not affected the range of foods available to the workers, soldiers and officers.24

12The citrus remains comprise eleven desiccated seeds and a cluster of seeds still embedded in fruit pulp (figs. 2b and 3b), found in four different rubbish deposits both in and outside of the main fort; all dated to the second half of the 2nd century AD. They were all identified as remains of citron, Citrus medica.

1.4. Antinoë (Nile Valley) – Mid Roman

  • 25 Keimer 1984: 1-2, and endnote 6, which reads: Bot. Museum Berlin-Dahlem, Sammlung Schweinfurth, Lis (...)
  • 26 de Vartavan, Asensi Amorós 2010: 80.

13Antinoë, also called Antinopolis or Antinoopolis, was a city founded by the emperor Hadrian in AD 130-132 on the east bank of the Nile, opposite Hermopolis and the modern city of el-Ashmunein. Excavations by Albert Gayet during the late 19th and early 20th century revealed a necropolis amongst other remains. Fragments of leaves found there were identified by Schweinfurth and deposited in the Botanical Museum in Berlin-Dahlem. They concern two 6 cm long leaf segments of Citrus medica, part of a wreath around the head of a 3rd century AD mummy.25 The Codex of Ancient Egyptian Plant Remains lists several further finds of leaves of Citrus medica from this site, apparently all associated with mummies in the necropolis, some dated as Greco-Roman (not further specified), others as Roman (3rd century AD) or Coptic (4th century AD).26 As their authenticity is not secure (they are not listed in Germer 1985), these are not listed in our Table 1.

1.5. Kellis (Western Desert) – Mid-Late Roman

  • 27 P. Kell. IV Gr. 96, Bagnall 1997; Thanheiser 2002.

14Kellis or Ismant el-Kharab lies in the centre of the Dakhleh Oasis, and represents a large settlement. Excavations here started in 1986 and have revealed occupation from the 1st-4th centuries AD, though there is a cemetery dating to the Ptolemaic period, suggesting the site must already have been occupied then. The village (or town) grew up following the introduction of advanced irrigation technology by the Romans, and comprises residential areas, as well as several temples, early Christian churches, Roman baths and extensive industrial areas. It derived its livelihood from agriculture, and traces of ancient fields can still be seen west and south of the settlement. The excavations have yielded a rich archive of ceramics, papyri, textiles, basketry and food remains, as well as two wooden books, including the now famous Kellis Agricultural Account Book.27

15Archaeobotanical samples were taken within the habitation areas, as well as in the temple complex and the churches. Most of the samples were sieved through a 0.5 mm mesh and all significant matrices were also screened through bigger meshes to recover rarer items. Approximate 700 samples, representing 10,000 plant remains, have been analysed to date.

  • 28 For more details, see Thanheiser 2002; Thanheiser et al. 2016.

16The plant macro-remains found, mainly desiccated, show that the agricultural life during the 1st-4th centuries was based on various winter and summer crops, including wheat, barley and pearl millet (Pennisetum glaucum, ssp. glaucum), pulses, especially lentil (Lens culinaris), oil and fiber plants, such as safflower (Carthamus tinctorius), flax/linen (Linum usitatissimum) and cotton (Gossypium sp.). The fruit tree production is dominated by olive (Olea europaea), fig (Ficus carica), grapevine (Vitis vinifera), and also includes almond (Prunus amygdalus), pomegranate (Punica granatum) and peach (Prunus persica). Vegetables and herbs include artichoke (Cynara cf. scolymus), onion (Allium cepa), garlic (Allium sativum), fennel (Foeniculum vulgare) and coriander (Coriandrum sativum). The majority of these crops are listed as locally grown in the Kellis Agricultural Account Book mentioned above.28

  • 29 Thanheiser 2002.

17Several desiccated rind fragments of Citrus were found in the domestic areas of the last period of occupation, in the 3rd-4th centuries AD. They are identified as citron (Citrus cf. medica). Citron is not mentioned in the Kellis Agricultural Account Book, and it is thus difficult to prove that citron was locally cultivated in the Dakhleh oasis, though it is certainly likely. There are imported food items at Kellis, e.g. hazelnut (Corylus avellana), pine (Pinus pinea) and walnut (Juglans regia),29 but these are all nuts that can easily be transported long distances in dry form. This would have been more difficult with fresh fruits of citron.

1.6. Mons Porphyrites (Eastern Desert) – Late Roman

18The quarry complex of Mons Porphyrites is described above (section 2.3). During the Late Roman period most of the activity appears to have been concentrated at the small quarry village of Lykabettus, high in the mountains; occupation of the main fort was restricted at that time. Additionally, the first way-station on the road to the Nile Valley, Badia, was still in use, with a high mountain pass connecting it with Lykabettus. The plant assemblage at this time was more restricted than during the heyday of the quarry complex, but with twenty-seven food plants it still suggests the workers and military personnel had access to a varied diet. Apart from the usual staples (cereals, pulses and dates), they had grapes (Vitis vinifera), olives (Olea europaea), figs (Ficus carica), sebesten (Cordia myxa), linseed (Linum usitatissimum), garlic (Allium sativum) and coriander (Coriandrum sativum).

  • 30 For more details, see Van der Veen, Tabinor 2007.

19While occupation was on a much smaller scale than during the early Roman phase of the complex and while only very small-scale excavations were possible at Lykabettus due to the logistics of reaching the village and bringing samples back down the mountain, three desiccated citrus seeds were found here (none were found at the way-station Badia). They came from three separate rooms/buildings, all dating to the 4th-5th century AD. All three belong to citron, Citrus cf. medica.30

1.7. Phoebammon (Nile Valley) – Late Antique

  • 31 Täckholm 1961.
  • 32 Täckholm 1961: 23-24.

20The monastery at Phoebammon is located on the west bank of the Nile, more or less opposite Luxor, and ca. 8 km west of the Valley of the Queens. It was excavated in 1948. The monastery was founded in the 4th century AD. Botanical remains were collected, but not in any systematic way and no context information for the remains is available. The plants were identified by Täckholm31 and include one half fruit of Citrus medica dating to the Late Antique period, 4th or 5th century AD. The precise description made by Täckholm32 leaves no doubt about the identification:

Half of a young fruit. 8.5 cm long, 4 cm broad, of somewhat pointed shape. Skin blackish-brown with knots of glands. The inside smooth, with the dried-up fleshy part preserved. Two kernels were removed from it. They are 10 mm long, 5 mm broad, obliquely pointed, slightly wrinkled and of brown or blackish colour.

  • 33 Germer 1985.

21A second item concerns a fragment of a fruit, “3.5 cm long, 2 cm broad, wrinkled inside. The identification of this piece remains uncertain.” This latter item is not included in our table 1. Surprisingly, the citron from Phoebammon is not mentioned in Germer33.

1.8. Hisn al-Bab (Nile Valley/Nubia) – Early medieval Islamic period

  • 34 A. Clapham, personal communication; Gascoigne, Rose 2012; Rose, Gascoigne 2013.

22The site of Hisn al-Bab is located on the eastern bank of the river Nile, near the first cataract, between the High and Low Dams, ca. 5 miles south of Aswan. The fortifications on a rocky outcrop above the river controlled the main branch of the Nile. The area has always been a frontier zone between Egypt and Nubia and a military presence has been essential through time. The site was occupied during the Late Antique and early medieval period, with at least three phases of fortifications identified, dating from the 5th to the 10th centuries AD. The most abundant occupation dates to the late 6th-early 7th century AD, during which time the fort appears to have served as a Nubian frontier post. The site is best known for its role in the baqt, a negotiated agreement between Egypt and Nubia shortly after the Arab conquest of Egypt in 651-652 AD. The agreement concerns peaceful relations between the two powers, with exchanges of slaves from Nubia in return for foodstuffs from Egypt. Remains of the earliest fortress comprise stone walls, a tower, a triangular enclosure, traces of structures in a smaller enclosure, as well as midden deposits.34

  • 35 Clapham, Rowley-Conwy 2009.
  • 36 For further detail, see Clapham unpublished data.

23Botanical remains, charred and desiccated, from the middens are currently under investigation; they include both summer and winter crops. Cereals comprise wheat and barley, sorghum (Sorghum bicolor ssp. bicolor), pearl millet (Pennisetum glaucum), and foxtail millet (Setaria italica). Other crops are lentil (Lens culinaris), cow pea (Vigna unguiculata) and cotton (Gossypium sp.). The assemblage is not dissimilar to that from Qasr Ibrim, located about 120 km south of Hisn al-Bab, though no citrus remains were found there.35 One desiccated seed of citron (Citrus medica) was recovered, from a deposit dating to the 6th-7th century AD.36

1.9. Kusayr (Red Sea coast/Eastern Desert) – Islamic period

  • 37 Peacock, Blue 2006; 2011.

24The Roman occupation of the site of Quseir al-Qadim is described above (section 2.1 Myos Hormos). After a long hiatus the site was reoccupied during the Islamic period and was then known as Kusayr. Occupation spans the 11th to 15th centuries AD, but the most intense occupation was during the 11th-13th centuries. The site functioned again as a transport hub for the Indian Ocean trade, but additionally it acted as a disembarkation point for pilgrims and as a transshipment point for grain intended for Mecca and the Hijaz. By this time the lagoon had largely silted up and the harbour consisted of a small bay at the entrance of the former lagoon. A number of residential complexes were excavated as well as some buildings thought to be associated with the trade.37 Samples were collected from the many refuse/midden deposits. Sampling was as for the Roman remains, see section 1.1.

  • 38 For further details, see Van der Veen 2011; Van der Veen, Morales 2014.

25The botanical assemblage is extremely rich, with remains of 81 food and other economic taxa in the 11th-13th century assemblage and 49 in the 14th-15th century assemblage, the latter range possibly affected by the very limited excavations in these later medieval deposits. The most remarkable aspect of the assemblage is evidence for the wide range of spices and other foodstuffs originating from the Indian Ocean world, now consumed in the port. Eastern spices include black pepper (Piper nigrum), ginger (Zingiber officinale), turmeric (Curcuma sp.), and cardamom (Elettaria cardamomum). Other new introductions include banana (Musa sp.), sugarcane (Saccharum sp.), aubergine (Solanum melongena), taro (Colocasia esculenta) and coconut (Cocos nucifera).38

  • 39 Pagnoux et al. 2013, appendix 3.

26At least two species of Citrus were found. A total of 70 seeds and 42 large rind fragments were recovered from the 11th-13th century AD, as well as two seeds from the 14th-15th century deposits. Identification criteria are listed in table 2. The seeds could be divided into two types. The first type (seed type A) is identified as citron (Citrus cf. medica) (fig. 2a), although sour and sweet orange (Citrus x aurantium and C. x sinensis) cannot be ruled out on morphological grounds. Indeed, the average measurements of the seeds of these three taxa overlap, as observed on modern material.39 The seeds of type B are in morphology and size closest to those of both lemon (Citrus limon) and lime (Citrus aurantifolia) (fig. 2c and d). Because of the large number of rind fragments of lime in the same deposits, they have been cautiously identified as mostly belonging to lime (Citrus cf. x aurantifolia).

  • 40 For more details, see Van der Veen 2011.

27The rind fragments are divided into four types, see identification criteria in table 2. Type 1 belongs to a large oblong fruit (ca. 70-x30 mm) with a thick and furrowed skin, tentatively identified as citron (C. cfmedica) (fig. 3c). Type 2 (38 specimens) consists of small globular fruits (diameter varying between 25-35 mm) and many of these represent fruits cut in half cross-sectional (fig. 4). These have been identified as belonging to lime (C. x aurantifolia). Type 3 rinds (2 specimens) are of similar size to those of type 2, but have a larger pedicel scar and were not cut in half (fig. 5a). These could belong to either lime or a small type of lemon. Finally, type 4 represents one fragment of a much larger fruit (fig. 5b); it could not be identified, though might belong to sour orange (Citrus x aurantium), but no secure identification can be offered.40

Table 2

Table 2

Identification criteria for the citrus remains found at Quseir al-Qadim, i.e. Myos Hormos and Kusayr, from Van der Veen 2011: 83-89. See also figs. 2-5 in this paper.

Fig. 4 - Archaeobotanical remains of citrus rind.

Fig. 4 - Archaeobotanical remains of citrus rind.

a. modern lime, Citrus x aurantifolia; b., c. and d. Type 2 rind, cf. lime, Citrus cf. x aurantifolia, all from Quseir al-Qadim, 11th-13th century AD (from Van der Veen 2011, fig. 3.6); central column of rows a, b, and d show the small pedicel scar; row c shows the small nipple. See Table 2 for identification criteria (photographs: Jacob Morales).

Fig. 5 - Archaeobotanical remains of citrus rind.

Fig. 5 - Archaeobotanical remains of citrus rind.

a. Type 3 rind, cf. lemon/lime, Citrus x limon/x aurantifolia (11th-13th century AD); b. Type 4, indeterminate, both from Kusayr, 11th-13th century AD (from Van der Veen 2011, fig. 3.6). See Table 2 for identification criteria (photographs: Jacob Morales).

1.10. Syene (Nile Valley) – Mamluk period

28The city of Syene is located under modern-day Aswan, situated on the first cataract of the Nile. It was the southern border town of Egypt and an important garrison town. At the frontier with Nubia, Syene was also an important trading centre and famous for its granite deposits. The Swiss Institute started work on documenting and investigating individual monuments in the city of Aswan and nearby area during the 1970s. Since 2000 they have worked together with the Egyptian Supreme Council of Antiquities on systematic rescue excavations across the town. The objective is to understand the history of colonization of the eastern shore of Aswan, where in Greco-Roman times the city of Syene was founded (the twin city of Elephantine), representing the southernmost city of the Roman Empire and the most important city of Upper Egypt between the Greco-Roman and the early Islamic periods.

  • 41 Thanheiser, personal communication; see also <http://swissinst.ch/html/forschung_neu.html>.

29Excavations have recovered some prehistoric evidence and occupation of Ptolemaic, Roman and Islamic date. The citrus remains were derived from Mamluk deposits (mid 13th-early 16th century AD); no systematic archaeobotanical sampling was carried out in these deposits, and only handpicked material is available, amongst which were two large fragments of Citrus medica rind, as well as two fragments of cotton balls (Gossypium sp.), doum palm (Hyphaene thebaica) and date stones (Phoenix dactylifera). No citrus remains have been recovered from Ptolemaic or Roman levels.41

1.11. Finds of uncertain provenance

  • 42 Darby et al. 1977: 704-706, fig. 18.2.
  • 43 Initially identified as lemon, but subsequently identified as citron; Loret, Poisson 1895: 194-196; (...)
  • 44 Keimer 1984: 1.
  • 45 de Vartavan, Asensi Amorós 2010: 80.

30There are several further finds of citrus available in the literature, but none of these are precisely dated and are thus not included in table 1. For example, there are several fruits of citron in the Dokki Agricultural Museum at Cairo, ascribed to the Greco-Roman period; one of which is illustrated by Darby,42 and represents a half fruit with seeds still embedded in fruit pulp. Note that while the text refers to it as of Greco-Roman date, the caption indicates that it comes from Luxor and dates to uncertain period but probably New Kingdom. Similarly, there is one citrus fruit in the Louvre Museum, from Egypt but otherwise of unknown origin and date.43 There is also a citron fruit and half of one longitudinally cut from a grave near Luxor (Thebes), which are assumed to be of Ptolemaic or Roman date, despite the fact that they were presented by members of the local population without information about the exact spot it was found.44 In the Codex of Ancient Egyptian Plant Remains there is mention of two finds of Citrus aurantium, both ascribed to the Pharaonic period, both almost certainly representing medieval or modern contamination. And finally, there is mention of Citrus wood from Antinoë, again concerning old finds, never recently corroborated.45

1.12. Summary of the botanical evidence

31To date, the earliest secure archaeobotanical finds of citrus in Egypt belong to the 1st and 2nd centuries AD, from three separate sites, i.e. Myos Hormos, Mons Claudianus, Mons Porphyrites, and these all belong to citron, Citrus medica. Mid and later Roman evidence comprises leaves of citron which formed part of a 3rd century AD funeral wreath from Antinoë, rind fragments from 3rd-4th century AD Kellis, seeds from the late phase of Mons Porphyrites (4th-5th century AD) and one fruit from the 4th-5th century AD monastery at Phoebammon. Medieval Islamic records of citron consist of a desiccated seed from Hisn al-Bab (6th-7th century AD), seeds and skin/rind fragments from Kusayr (11th-13th century AD) and rind fragments from Mamluk Syene (13th-16th century AD).

32One further species of citrus was identified in Egypt, i.e. lime, Citrus x aurantifolia, from 11th-13th century Kusayr, while a further two species (lemon, Citrus cf. x limon, and possibly sour orange, Citrus cf. x aurantium) may be present (also at Kusayr), though current morphological criteria are not sufficient to offer a definite identification for these. Other scientific methods, e.g. aDNA, may ultimately allow verification.

2. Text references to citrus

  • 46 Habermann 1998.
  • 47 E.g. Bingen et al. 1992; 1997; Bülow-Jacobsen 2003; Cuvigny 2000; 2003b; 2005; 2012.

33The arid conditions in Egypt are also responsible for the excellent preservation of ancient papyri and ostraca, and there is a considerable corpus available, comprising some 25,000 texts spanning the Ptolemaic and Roman periods.46 Many originate from the early 20th century excavations at Oxyrhynchus, an ancient city located on the western edge of the Nile valley, in Middle Egypt, but others were derived from sites in the Fayum and other parts of Egypt. An extensive corpus of texts were recovered at the excavations of Roman sites in the Eastern Desert.47 While the majority of the Oxyrhynchus texts are papyri, the majority of the Eastern Desert ones are ostraca, the latter used primarily for local communications and documents that do not need to be kept for the longer term.

2.1. Greek terms for citrus fruits

  • 48 André 1985: 68; Amigues 2005: 364.
  • 49 I.109.3, ed. Wellmann.
  • 50 Harrauer et al. 1986: 164.
  • 51 Daris 1989: 93; Bülow-Jacobsen 2003: 420; Armoni et al. 2005: 375.
  • 52 Grenfell, Hunt 1920: 19 and 183; Sijpesteijn 1977a: 83-84.
  • 53 LSJ; Fournet, Magdelaine 2001; Amigues 2005.
  • 54 André 1985: 68.

34Greek was the principal written language in Egypt during the Ptolemaic and Roman periods, and the Egyptian papyrological documentation contains several Greek terms for the citrus plant, though as discussed below, these differ from those used by classical authors. The papyri use the terms κίτριον (kitrion) – also commonly spelled κίτριν (kitrin) – referring to either the tree or the fruit, and κίτρον (kitron) referring to the fruit; these terms correspond to the Latin equivalents of citrus, citrea and citrium.48 The term κεδρόμηλον (kedromèlon), which is given as a synonym of κίτριον (kitrion) by Dioscorides (De Materia Medica I.115.5.3),49 is found in just one, fragmentary, papyrus of the 7th century (CPR X 138).50 Two translations of the Greek terms have been offered: lemon51 and citron.52 The latter one, citron, is now the most commonly accepted translation,53 as is the case for the Latin terms.54

  • 55 I.109.3-4, ed. Wellmann.

35In contrast, Greek writers of the 4th century BC to the 1st century AD outside of Egypt often refer to the tree with the periphrasis “Median apple” or “Persian apple” (e.g. Theophrastus, Enquiry into Plants I.11.4, I.13.4 and IV.4.2). Dioscorides mentions “the Median or Persian apples or kedromèlon, called kitrion in Latin” (τὰ δὲ Μηδικὰ λεγόμενα ἢ Περσικὰ ἢ κεδρόμηλα, Ῥωμαιστὶ δὲ κίτρια; Dioscorides, De Materia Medica I.115.5.3-4).55 Similarly, Pliny refers to the fruit as “the citron or Assyrian apple, called by others the Median apple” (Natural History XII.7.15-16) and “the Greeks, however, call them Medic [Median] apples” (Natural History XV.14.47). As mentioned, these designations referring to the origin of the citron are not encountered in the papyrological corpus from Egypt, but, to highlight the complexity of identifying plants from their ancient names, the term “Persian apple tree” is, in the Greek papyri from Egypt, used to refer to the peach tree.

  • 56 Andrews 1961: 41-42.
  • 57 Something also mentioned by Galen (late 2nd century AD): “Its fruit is no longer called ‘Median app (...)
  • 58 Loret 1891.

36Andrews56 highlights that during the early Empire (late 1st century BC) the references to “Persian” and “Median” apple began to be replaced by a specific term for citron: citreum and citrium, with later variants citrum, cetreum and cedrium, and citrus for the tree,57 which matches what we see in the Greek papyri from Egypt. This name change, he suggests, is evidence that only then citron cultivation was introduced into Italy, with the successful acclimatization – meaning the production of flowers and fruits – not completed until the middle of the 1st century AD. He also points out, following Loret,58 that these terms may go back to an ancient Egyptian name, with kitr- as the root, hence his suggestion that citron may have been introduced from Egypt into Italy.

2.2. The earliest Egyptian papyri referring to citrus (1st-3rd centuries AD)

  • 59 The papyrological abbreviations follow the Checklist of Editions of Greek, Latin, Demotic, and Copt (...)
  • 60 Guéraud 1942: 168-169, no. 13.
  • 61 Bülow-Jacobsen 2003: 420.

37The first papyrological evidence of citrus comes from the Eastern desert (table 3).59 One ostracon found at the Roman fort of Persou, a way-station in the Wādi Fawākhir, on the route to Myos Hormos, concerns a private letter acknowledging receipt of syrmadia (horseradish?) and sending back one kitrin (κίτριν), citron, along with some kolokynthia (gourds?) (SB VI 9017 Nr. 13 (=O.Faw. 13), line 10).60 The ostracon itself is not dated, but it is part of an assemblage of ostraca belonging to the 1st-2nd centuries AD. There is no indication of the geographical origin of the letter, but while the horseradish is thought to have been grown at Persou, the gourds are listed among the food items mentioned in the letters from this site as coming from the Nile valley. It is thus likely that the citron did too.61

  • 62 H. Cuvigny and A. Bülow-Jacobsen, personal communication.

38The as yet unpublished 2nd century ostracon from Dios (inv. 749), another Eastern Desert way-station (on the route to Berenike), refers to kitrion (κίτριον) among other food items. It is a fragmentary private letter, containing the text “receive 1 citron”. The indication “from Coptos” is found elsewhere in the text, but it is not clear whether this refers to all or just some food items on the list.62 Nonetheless, most of the foods consumed at the Roman sites in the Eastern Desert originated from Coptos in the Nile valley, see discussion in section 3.2.

  • 63 Shelton 1971: 57-61.

39In addition to the two letters from the Eastern desert, a household account of the late 2nd century AD (AD 182), probably coming from Oxyrhynchus in the Nile valley (P.Mich. XI 619), twice refers to the citron fruit (κ̣ι̣τρίο(υ), the genitive singular form of κίτριον, line 3 and κιτ(ρίου), line 27) among other foodstuffs, such as asparagus, fava bean, lupin, leaf beet, bread, meat, etc.63

  • 64 David et al. 1941: 44-45.

40Finally, a private letter (P.Warren 15 (=P.L.Bat. I 15)), of unknown origin, dated to the 2nd century, concerns a papyrus written by Arsinoos who asks his ‘brother’ Alkimos for an agricultural (?) instrument and tells him that he has dispatched clothes and fruit, namely 1 artaba of dates and 70 citrons (κίτρια, line 15)64. Note the unusually large number of citron fruits mentioned in this text.

41At least five 3rd century AD records, papyri and ostraca, refer to the citron plant or fruit. Three of them were found at Oxyrhynchus, one of which clearly demonstrates the cultivation of the tree. Indeed, the fruits (κιτρίων, the genitive plural form of κίτριον, line 24) are mentioned in a contract established in 280 AD for labour in a vineyard and lease of a fruit garden. While it is clear that the citrons represent only a very small part of the rent compared to the other fruit items, the text confirms that citron was cultivated in Egypt at that time. In it, among other requests, three persons offer:

  • 65 Grenfell, Hunt 1920: 19.

to lease for one year the produce of the date palms and all the fruit trees which are in the old vineyard, for which we will pay as a special rent 1½ artabae of fresh dates, 1½ artabae of pressed dates, 1½ artabae of walnut-dates, 1½ artabae of black olives, 500 selected peaches, 15 citrons, 400 summer figs before the inundation, 500 winter figs, 4 large white fat melons (P.Oxy. XIV 1631).65

  • 66 Schuman 1980: 42-43.
  • 67 Grenfell, Hunt 1920: 183.

42Two other private letters found at Oxyrhynchus refer to the citron plant. The first one, dating to the first quarter of the 3rd century, mentions – although the context is not clear – the delivery of “four other citron trees for a total of 42” (kitria, κί|τρ̣ια, accusative plural form of κίτριον, P.Wash.Univ. I 29, lines 13-14).66 The second one, incomplete, from the 3rd century, refers to the “sending of a donkey in order to […] young plants of peach and citron”. Even though the papyrus breaks off and leaves the sentence unfinished, this letter makes clear that the peach (here persikôn, περσικῶν, line 18) is well distinguished from the citron tree (κιτρίω̣[ν], line 19) (P.Oxy. XIV 1764).67

  • 68 Armoni et al. 2005: 375.

43One ostracon of the second half of the 3rd century AD (O.Heid. 396), possibly coming from the Fayum oasis, ca. 100 km north of Oxyrhynchus, corresponds to a list of payments in kind which contains a reference to a large citron (κίτριν μεγάλην, line 20), as well as other foodstuffs such as eggs, chicken, oil, etc.68 Another 3rd century papyrus of unknown origin shows us a list of foods sent or given to a high official (the hypomnematographus). Note that the number of citron fruits is modest in this case:

  • 69 Sijpesteijn 1977a: 83-84.

to the (servant) of the hypomnematographus: 16 eggs, 1 pig, 4 cocks, vegetables, 2 citrons; to Hegoumenos: 2 honey-cakes, 3 duck-eggs; to Phanias: fish, 4 duck-eggs, 2 citrons; to Ioustos: 1 jar, 30 eggs, 4 cocks; to the man from Antaeopolis: 1 jar (P.Wisc. II 60).69

  • 70 Lewis 1964: 32-35.
  • 71 Papathomas 1996: 175.
  • 72 Maspero 1911; Russo 1998: 150.
  • 73 Harrauer 1986: 165.
  • 74 Wessely 1921: 128.
  • 75 Armoni et al. 2005, see comm. ad l. 20 p. 375.

44Only one text is dated to the 4th century (AD 388). It is a papyrus that describes an offer to lease an orchard at Hermopolis Magna in Middle Egypt (SB VIII 9907), containing date palm and citron trees (φύνικες καὶ κίδρια, l. φοίνικες; κίτρια).70 There appear to be no Greek texts referring to citron (or other citrus species) dating to the 5th or 6th centuries AD. However, it is worth highlighting five mentions of ‘citron-colour’, dated to the 4th-7th centuries, with two different derived adjective forms, either kitrinos (κίτρινον in P.Heid. VII 406;71 P.Cair.Masp. I 6700672 and probably CPR X 139)73 or kitreos – which is borrowed from the Latin citreus (κίτρεον in SPP XX 24574 and SB III 6024).75 All these references to citron colour are associated with clothing (table 3), which suggests that the fruit is sufficiently well known to be used to name a colour.

  • 76 Sijpesteijn 1977b: 81; concerning the date, see Fournet 2016.
  • 77 Kovarik 2007; Schram 2014-.
  • 78 Harrauer 1986: 108-109; Schram 2014-.

45In terms of later texts, at least five Greek texts from the 7th and one from the 7th-8th century refer to the fruit or the tree (table 3). Among them, three leases of orchards or land show that the tree was still cultivated then, in Middle Egypt (P.Mich XIII 666)76 and in the Fayum oasis, together with other fruit trees such as peach, pomegranate, apricot (?) and date palms (SB I 4483, line 12 and SB I 4485, lines 4 and 5).77 It is worth mentioning here that there is another 7th century papyrus from the Fayum that lists many of these fruits in very large quantities. The text concerns a delivery order for a large estate and it lists several fruits, followed by numbers. Although the edition of the papyrus interprets the numbers as prices, the lack of any monetary unit might, instead, be taken to mean that the numbers refer to quantities. These include 500 apricots, 2000 pomegranates and 1000 citrons (κίτρ(ων), CPR X 62, line 5).78 Unfortunately, the text does not divulge what all these citron fruits were to be used for.

  • 79 Boud’hors, personal communication; see also Förster 2002: 417.
  • 80 Boud’hors, Heurtel 2010: 354.
  • 81 Boud’hors, Heurtel 2010: 21.

46As for the Coptic texts, several mention citrons and the citron-tree. Here (in table 3) we include three noteworthy ostraca from the Theban region and dating to the 7th or first half of the 8th century, which derive from a tomb linked to the ascetic Frangé.79 These three ostraca refer to the fruit (borrowing the Greek words κίτρον or κιτρέα, in Coptic: k'idre and kitra), which suggest that citron was used there on a regular basis. One of these texts, O.Frangé 645, specifically refers to the medical purpose of citron. Indeed, an unknown sender asks for a little bit of citron to be given to his men who are sick.80 Thus the fruit may join the list of particular foods that were regarded appropriate for people who were ill or poorly, such as wine, honey and maybe fish, according to other ostraca from the same context.81

  • 82 Watson 1983: 45.
  • 83 Müller-Wodarg 1957: 73.

47Finally, we have not been able to determine whether any of the surviving Arabic texts from the period mention citrus fruits, but we do know that the fruit is mentioned by Islamic authors as early as the 7th century AD.82 None of these specifically refer to its cultivation in Egypt, but al-Hamdani (AD 967-1007) mentions that is was available in Egypt in his time.83

Table 3 - Papyrological evidence for citrus fruits from Egypt.

Table 3 - Papyrological evidence for citrus fruits from Egypt.

References: a. http://papyri.info/​search; b. Bülow-Jacobsen 2003; c. Cuvigny and Bülow-Jacobsen, personal communication; d. Shelton 1971: 57-61; e. David, Van Groningen, Van Oven 1941: 44-45; f. Grenfell, Hunt 1920: P.Oxy. XIV 1631 p. 15-25; P.Oxy. XIV 1764 p. 183 (descr.); g. Schuman 1980: 42-43; h. Armoni, Cowey, Hagedorn 2005: 368-376, esp. 375; i. Sijpesteijn 1977a: 83-85; j. Lewis 1964; k. Pintaudi 1981: 98-99; l. Schram 2014-; m. Maspero, 1911: 22-33; n. Wessely 1921: 128; o. Harrauer et al. 1986: CPR X 62 p. 108-109 ; CPR X 138 p. 164 ; CPR X 139 p. 165; p. Kovarik 2007; q. Sijpesteijn 1977b: 81-87; r. Boud’hors, Heurtel 2010: 46-47; s. Papathomas 1996: 175-190; t. Daris 1989; u. Guéraud 1942; v. Papathomas 1994 (SB XVI 12254, 8 and 11); w. Russo 1998; x. Boud’hors, Heurtel 2010: 354; y. Boud’hors, Heurtel 2010: 360; z. Boud’hors, personal communication.

2.3. Summary of the papyrological evidence

  • 84 H. Cuvigny, personal communication.

48As far as we know, citron is not mentioned in the Egyptian papyri dating to the Ptolemaic period. Noteworthy is the fact that citron is not found in the Zenon archive, dated to the 3rd century BC, in which luxury upper class foods are often mentioned,84 which might suggest, but does not prove, that it was not yet known in Egypt then. The earliest known records of citrus date to the 1st-2nd century AD, and these concern sites in the Eastern Desert, that were supplied from the Nile valley (see discussion below). The number of papyrological records increases somewhat during the end of the 2nd century/3rd century AD and these show clear evidence for the local cultivation of the tree in the Nile valley and perhaps in the Fayum oasis. During the Byzantine and early Islamic periods (4th-8th centuries), the growing number of records among the Greek and Coptic papyrological corpus suggests wider cultivation and availability of citron by then (table 3).

3. Discussion

3.1. Limitations of the evidence

  • 85 Habermann 1998.

49Both archaeobotanical and papyrological data referring to citrus species in Egypt known today are restricted to the Roman and later periods. It is clear, however, that the distribution of the finds reflects the research investment. For example, the number of excavations of Ptolemaic sites where botanical remains have been collected, is extremely limited compared to Roman ones, but even Roman sites with analysed botanical remains are not numerous, especially in the Nile valley, Delta and Mediterranean coastal strip. In contrast, the size of the Ptolemaic papyrological corpus, while somewhat smaller than that of the early Roman period, might appear to be of sufficient size to allow a good chance of rare foodstuffs to be mentioned (ca. 7,000 texts dating to the 3rd-1st century BC, compared with ca. 11,000 texts dating to the 1st-2nd century AD).85 Nonetheless, while the corpus is substantial, it contains few texts from settlement sites, limiting the chance of finding references to the growing of trees and foodstuffs (see below). Thus, an absence of references to citron does not mean it was not grown then. The discrepancy between the large numbers of papyri from the Ptolemaic period compared with the rarity of botanical remains can be explained by the fact that early excavations tended to focus on the collection of texts, but largely ignored botanical remains.

  • 86 Pagnoux et al. 2013: 424.

50A further factor to consider is that of preservation. Citrus remains, like many other fruits, will not normally come into contact with fire, and are thus rarely found in charred form. No charred remains of citrus have been found in Egypt; they are all desiccated. Only one charred seed is reported from the Mediterranean region (i.e. from Pompeii); all the others are mineralized.86 The lack of desiccated midden deposits in the Delta, the Mediterranean coast of Egypt and some parts of the Nile valley means that the preservation of both the seeds/fruits (and papyri/ostraca) is largely restricted to the Eastern and Western deserts, the dryer parts of the Nile valley and the Fayum.

  • 87 Van Zeist et al. 2001; Langgut et al. 2013; Langgut 2014.
  • 88 Bui Thi Mai, Girard 2014; papers this volume.
  • 89 Langgut et al. 2013.

51Furthermore, the earliest records of citrus in the Mediterranean consist of pollen evidence (at Carthage and Ramat Rehel87 and possibly at Cuma),88 but virtually no palynological work has been carried out in Egypt, due to the rarity of suitable deposits. Future excavations of Ptolemaic sites and/or pollen analysis covering the Ptolemaic and earlier periods may, therefore, change the current picture. Pollen analysis of mudbricks and plaster in rural settlements and near gardens may be particularly profitable, as the evidence for citrus pollen from Ramat Rehel was derived from a plaster layer in one of the garden pools.89

  • 90 Bagnall 1995: 13-14.
  • 91 Bagnall 1995: 27.

52Finally, it is worth remembering that the papyrological data also contain biases. This line of evidence records what needed to be written down, not everything that happened. Thus, it does not contain a full list of all the plants that were cultivated or consumed. We can see this in the texts referred to above. Some are letters asking for or promising foods to be delivered to the sites in the Eastern Desert, some are contracts concerning the sale of agricultural land, and others concern transactions of various sorts, that is, all involve activities that require a written record. Government records concerning taxation and property ownership, as well as individual transactions and contracts concerning land, houses, slaves, and animals make up the majority of the papyrological corpus.90 Additionally, while the papyri from Roman Egypt comprise a wide range of documents, those from Ptolemaic Egypt often concern matters of state; few concern life in ordinary settlements and Bagnall91 warns that silences in the documentary record from this period should be treated with great caution. Thus, the absence of references to citron in the Ptolemaic papyri cannot be taken to mean that the tree was not cultivated in Egypt.

3.2. Local cultivation or import

53One of the key questions to be addressed here concerns the timing of the dispersal of citron and other citrus species into Egypt and the moment they became locally cultivated. In order to correctly interpret the botanical and papyrological evidence, it is, of course, essential to appreciate the nature of the evidence and the nature of the sites from which the evidence derives. Apart from the limitations on the evidence described above (section 3.1), there are three aspects that need consideration.

54First of all, the sites at which we have found evidence of local citrus cultivation are all located in agricultural regions, that is, Oxyrhynchus and Hermopolis Magna in the Middle Nile valley, the nome of Arsinoites in the Fayum and possibly Kellis in the Dakhleh oasis. All the other sites with citrus remains represent non-agricultural settlements, namely two quarry settlements (Mons Claudianus and Mons Porphyrites), two ports (Myos Hormos and Kusayr), two way-stations (Persou and Dios), one necropolis (Antinoë), one monastery (Phoebammon) and one fort (Hisn al-Bab). Thus, it is precisely at agricultural sites that we find evidence for the local cultivation of citron in Egypt and the earliest evidence we have from such sites dates to the end of the 3rd century (Oxyrhynchus). This highlights the need for great caution in interpreting the current evidence as pointing to an absence of citrus cultivation in Egypt before the end of the 3rd century AD. Until we acquire more botanical and papyrological evidence from agricultural settlements of Ptolemaic and early Roman date in the Nile valley, Fayum, Delta and Western oases, we are not going to be able to determine definitively when the cultivation of citron started here.

  • 92 Van der Veen et al. in press.
  • 93 Sidebotham 1986; Bingen et al. 1992; 1997; Cuvigny 1996; 2000; 2003a; 2012; Bülow-Jacobsen 2003.
  • 94 Bülow-Jacobsen 1997; 2003; Van der Veen 1998.
  • 95 Bülow-Jacobsen 2003; Strabo, The Geography 17.1.45.

55Secondly, to understand the remains of citrus from the Eastern Desert sites, it is important to realize that the Roman sites in the Eastern Desert concern very specialized settlements, i.e. ports, quarries, and the way-stations that served them. None of them are ‘normal’ villages or cities and none have an agricultural component. They are all part of a tight and integrated network of industrial and commercial activities providing the Roman Empire with luxuries, such as spices and textiles from India and prestigious building stone and emeralds from the Eastern Desert itself.92 The importance of these products to the Roman Empire meant that there was a highly organized supply system in place, provisioning these remote settlements with food, water and other essentials. This organization was largely government-controlled, but included some private enterprise. The foodstuffs were brought in from the Nile valley, especially from the Coptos region, the starting point of the roads leading to the ports and quarries. At each of the excavated sites large numbers of ostraca and some papyri describe this organization and these transactions in some detail.93 Additionally, notwithstanding the fact that most of the foods come from the Nile valley, there is both botanical and textual evidence that some green vegetables were produced in small garden plots associated with the quarries and some of the way-stations, as a private enterprise.94 While water in the desert was very scarce, small plots could be watered by hand to grow vegetables with short growing seasons providing fresh greens, something difficult to achieve via the routes from the Nile valley. For example, in antiquity the trip from Qena (Kainepolis) to Mons Claudianus would have taken five days when carrying supplies, and that from the Nile to Myos Hormos six or seven days,95 making delivery of fresh leafy vegetables difficult. Taking these economic considerations and the local conditions into account, we conclude that, in all likelihood, the citrons found at these sites were brought there from the Coptos region, though they could have been grown in the Delta, the Fayum, in the Nile valley or in one of the oases, rather than necessarily near Coptos itself.

  • 96 Van der Veen 2001; Tengberg 2011; Van der Veen et al. in press.
  • 97 I.e. Periplus Maris Erythraei, Casson 1989.
  • 98 Theophrastus’ Enquiry into Plants IV.4.2; Pliny’s Natural History XII 7.15-16.
  • 99 Pagnoux et al. 2013; chapters of this volume.
  • 100 Bülow-Jacobsen 2003.

56It is, of course theoretically possible that the citron remains we found in the Eastern Desert came directly from the Indian subcontinent, via the ports of Myos Hormos and Berenike. This is implausible, however. Indeed, very few Indian foods were found at the desert sites (e.g. two peppercorns pepper at Mons Claudianus, one grain of rice at Didymoi),96 though some were found at the two ports. The latter (rice, mung bean, myrobalan) are interpreted as foodstuffs brought and consumed by Indian sailors that temporarily lived in the ports, rather than by the Roman residents. Concerning citron, there is no reference to this fruit in the mid-1st century AD text describing the trade,97 suggesting that it had already become available in the Mediterranean prior to the writing of that text, something we know to be true from historical texts,98 as well as from pollen and other evidence.99 Moreover, the ostraca from Persou and Dios that mention citron, concern lists of foodstuffs coming from the Nile valley, rather than from one of the ports. In fact, the ostraca from these way-stations do not mention the caravan supplying the ports or the one carrying the goods from the ports to the Nile valley.100

  • 101 Macaulay-Lewis 2006.

57Finally, the nature of the finds (fruits, seeds, and rinds) indicates that fruits, not just dried rind segments, were brought to the desert. The journey from India to either Myos Hormos or Berenike, would have taken months, making the survival of whole, fresh fruits on that journey unrealistic. The original introduction of citron into the west would probably have been as seeds or as small plants. The latter is described by Pliny: “[...] importing it [the citron plant] in earthenware pots provided with breathing holes for the roots...” (Pliny, Natural History XII.7.15-16), and such pots are, indeed, known from several Mediterranean archaeological sites.101 The fruits could, of course, have been preserved in brine and transported in ceramic vessels from India, or from other, Mediterranean, regions. However, we currently do not have any evidence of such a practice during the Greco-Roman period, though a section on citrus cultivation and grafting in Geoponica – a 10th century compilation of earlier Greco-Roman literature – refers to the preservation of the citron fruit (κίτριον). It advises that by covering the fruit with kneaded gypsum it can be ‘kept throughout’, and that ‘citrons stored under barley do not go rotten’ (Geoponica, X, 7-10, ed. Beckh; for translation, see Geoponika 2011: 204-205). The transportation of dried fruits or dried fruit halves within Egypt, not from India to Egypt, does need consideration (see the discussion in section 3.3).

58To sum up, the evidence from the Eastern Desert can give us no conclusive answer to the question of whether citron was cultivated in Egypt during the 1st and 2nd centuries AD. What it does indicate though is that citron was available in the Nile valley at that time, but whether it was cultivated in Egypt, or imported from elsewhere in the Mediterranean, will currently remain unanswered. However, even a journey from somewhere in the Mediterranean to Alexandria, from there to Coptos, and from Coptos to Mons Claudianus or Mons Porphyrites, would in all probability have taken too long to allow the survival of fresh fruits, though is possible as dried fruits. Thus, in all likelihood the citron was already cultivated in Egypt by the late 1st century AD, but we cannot offer definitive proof at this stage.

3.3. Consumption

  • 102 André 1981; Andrews 1961; Flower, Rosenbaum 1958; Zohary et al. 2012.

59Records of citron in Egypt, both archaeobotanical and textual, are modest in number, especially when compared with those of other foodstuffs. This can easily be explained by considering the way citron fruits were used and consumed. The fruits contain a small sour pulp which is very acidic and not eaten raw, but the rind is highly aromatic and rich in essential oil. This oil is used in perfumery and medicines, today and in antiquity,102 and the quantities of the fruits found should, thus, not be compared with those of fruits commonly consumed raw. Classical authors suggest that the medicinal use of the fruit was especially connected to its role as an antidote to poisons and as an insect repellent, as the following references highlight:

The ‘apple’ is not eaten, but it is very fragrant, as also is the leaf of the tree. And if the ‘apple’ is placed among clothes, it keeps them from being moth-eaten. It is also useful when one has drunk deadly poison; for being given in wine, it upsets the stomach and brings up the poison; also for producing sweetness of breath; for if one boils the inner part of the ‘apple’ in a sauce, or squeezes it into the mouth in some other medium, and then inhales it, it makes the breath sweet (Theophrastus, Enquiry into Plants IV.4.2; 4th century BC).

The citron or Assyrian apple, called by others the Median apple, is an antidote against poison. [...] For the rest, the actual fruit is not eaten, but it has an exceptionally strong scent, which belongs also to the leaves, and which penetrates garments stored with them and keeps off injurious insects. [...] Because of its great medicinal value various nations have tried to acclimatize it in their own countries, [...]. It is this fruit the pips of which, as we have mentioned, the Parthian grandees have cooked with their viands [food] for the sake of sweetening their breath... (Pliny, Natural History XII.7.15-16; 1st century AD).

[...] there are three parts to this fruit: the sharp centre, the fleshy pith and the outer skin. The skin is fragrant and aromatic, not just in smell but also in taste. It is naturally difficult to digest, as it is hard and knobbly. If it is used in medicine, it helps somewhat with the digestion, just like many other things that have a harsh quality. By the same token when a little is eaten it strengthens the stomach, so that after being cut open and squeezed, its juice is combined with medicines taken in pill form that act as a laxative or purge the whole body. Vinegar mixed with the inedible part of the fruit in which the pips are found is used for some other purposes, flat vinegar being made sharper through its addition. The middle of both the parts which give nourishment to the body, whilst containing neither harsh nor sharp qualities, are difficult to digest because of their hardness. Anyone who wants to liven up their dull taste, therefore, eats them with vinegar and fish-sauce. It is quickly found, either through experience or on the advice of a doctor, that they are digested better when taken like this (Galen, On the Powers of Food II.37.2; late 2nd century AD).

I am well aware, too, that when the citron is eaten before any food, dry or liquid, it is an antidote to every poisonous ingredient; I learned this from a townsman of mine who was entrusted with the governorship of Egypt. He had sentenced some convicted criminals to be the prey of wild beasts, and they were to be thrown among the creatures called asps. As they were entering the theatre assigned for the punishment of the robbers, a peddler-woman in the street gave them in pity some of the citron which she was holding in both hands and which she was eating. They took it and ate, and when, after a short time, they were thrown among those cruel and monstrous creatures, the asps, they received no injury when bitten. Perplexity seized the magistrate, and finally he questioned the soldier who guarded them to see whether they had eaten or drunk anything; when he learned that the citron had been given them, he ordered next day that a piece of citron should be given, exactly as before, to one convict, but not to the other, and the one who ate suffered no injury when bitten by the reptiles, but the other died the moment he was struck. And so, since the same result has been attested in many instances, the citron has been proved to be an antidote to every poison. Again, if one boil a whole citron in its natural state, seeds and all, in some Attic honey, it is dissolved in the honey, and anyone who takes two or three 'fingers' of it in the morning will not be harmed in any way by poison (Athenaeus, Deipnosophistae or Scholars at the Dinner Table III.84.D-F – 85.A; early 3rd century AD).

60All four texts provide some indication of how the citron was consumed. As a fumigant amongst clothes the fruits were used, though it is not clear from any of the texts whether these were fresh or dried fruits. As an antidote to poison the fruits were either soaked, boiled and/or dissolved in a liquid (e.g. wine or honey) and ingested as such, with the text by Athenaeus suggesting that the rind still existed within the liquid (reference to ‘fingers’), though the ‘two or three fingers’ may, equally, be a reference to the quantity of liquid to be taken. Also, the juice was squeezed from the fruit and ingested to aid the efficacy of medicines taken in tablet form. And there is the suggestion that the fruit was also consumed without boiling (the pedlar-woman was holding the fruit in her hands and eating it). The text from Athenaeus is particularly interesting in this context, as he was a native of Naucratis in Egypt. His text indicates that citron was easily available in Egypt by the early 3rd century AD, as well as highlighting that the magistrate knew the value of scientific methods.

61Finally, there is one reference to citron used in cuisine. In a collection of recipes credited to Apicius (late 4th-early 5th century AD) there is one mentioning the use of citron peel in a meat stew, not unlike the use of preserved lemons in Middle Eastern cooking today. The cook is instructed to clean the citrons inside and out and then dice them, before adding them to the stew; note that the reference is to complete fruits:

  • 103 Giacosa 1992: 95.

Sweet stew with citrons: In a pot put oil, garum, broth, head of leek; finely chop coriander, cooked pork shoulder, small meat patties. While this is cooking, grind pepper, cumin, coriander or coriander seeds, fresh rue, silphium root; pour on vinegar, defrutum, cooking broth, mix with vinegar. Bring to the boil. When it boils, put into the pot citron that have been cleaned inside and out, diced, and boiled. Break in tracta [thickening agent] and thicken, sprinkle with pepper, and serve (Minutal dulce ex citriis, Apicius De Re Coquinaria 169;103 5th century AD).

62The Greek papyrological corpus from Egypt, unfortunately, does not mention how the fruit was used. The texts mention citron fruits, but none of them indicate whether these are fresh or dried fruits, or how they were consumed. However, one Coptic text, O.Frangé 645, from a monastic context in the Theban region dated to the 7th-8th century, provides us with a specific mention to the medical use of the fruit. Unfortunately, we have no way of verifying whether the citron found at the 5th century monastery Phoebammon was also used for medicinal purposes.

63The archaeobotanical evidence does give some clues. Firstly, the presence of seeds at almost all the sites suggests that entire fruits had been present at each. Secondly, the fragments of rind at several sites indicate that citron rinds were used/consumed (see below), and, thirdly, the presence of near-complete fruits at both Mons Claudianus and Phoebammon, as well as at the Louvre, the Dokki Museum and in Keimer’s records (see section 1.10 above), suggests that the fruits had a particular role. In fact, remarkably, at least four of these concern half fruits, cut longitudinally, with the seeds still embedded in the fruit pulp (fig. 3a). This might indicate that the fruits were halved and dried, either for use as a fumigant in amongst clothes, in the preparation of an antidote to poison or a breath freshener, or in cuisine. Certainly, one of us (MvdV) has seen dried slices of citrus fruits for sale in herbalist shops in Iran; in this form they would be more easily kept longer term, and transported over long distances.

  • 104 Van der Veen 2011: 185-187.

64At the ports of Myos Hormos and Kusayr (the archaeological site of Quseir al-Qadim) the number of remains of citron and lime was large enough to allow a study of their comparative uses in one location.104 The remains of citron (Citrus medica) at both Roman Myos Hormos and Islamic Kusayr concern mostly seeds, with just one rind fragment found, representing half a fruit, cut lengthwise, but with no seeds or pulp still inside (fig. 3c). In contrast, the remains of lime (C. x aurantifolia), only found at medieval Islamic Kusayr, comprise many seeds, as well as some 40 large rind segments, many representing half fruits cut in cross-section, but without remains of fruit pulp or seeds (fig. 4b, c, and d). As preservation conditions and sampling strategy were identical for both sites/periods, the low occurrence of rind fragments of citron cannot be a product of either preservation or recovery, but, instead, points to differences in the way these fruits were consumed.

65The near absence of citron rind at Myos Hormos and Kusayr suggests that it was the rind that was consumed, which matches some of the uses of the fruit described above, i.e. the boiling of the fruit in a liquid (e.g. wine or honey) to soften or dissolve it before consumption – to counteract the ingestion of poison or improve bad breath –, as well as the consumption of the rind in meat stews. In all these cases the seeds would be discarded, which matches the archaeological findings. Only in the description by Theophrastus is there mention of the use of the fruit pulp (see above).

  • 105 Ibidem.

66In contrast, the lime fragments found at Kusayr comprised mostly fruits from which the juice is extracted, that is, the fruits were cut in half, squeezed and the rinds and seeds were discarded (fig. 4). This matches the use of limes today.105

  • 106 Van der Veen et al. in press.
  • 107 Cuvigny 2003a: 314-315, 339-340.

67In terms of social access to the fruit, current evidence from Egypt suggests that citron may have been restricted to the more well-to-do sectors of society in early Roman Egypt. It certainly was found primarily at sites where high ranking officials would have been present, such as at Mons Claudianus, Porphyrites and Myos Hormos, where it was part of a suite of ‘luxuries’, including various nuts (walnut, hazelnut, almond, pine nut), certain fruits (peach, pomegranate, persea), vegetables (artichoke) and herbs (anise, amni, fennel). This is in contrast to the way-stations, where such luxuries were not found and where such officials would not have resided, but only have passed through.106 Whether the citrons mentioned in the two texts from Persou and Dios, both way-stations, were intended for such personnel is not clear. They may have been intended for high-ranking individuals passing through, such as the curatores praesidiorum, in charge of the administration of the way-stations, or a centurion or Decurion.107

Conclusion

68Citrus fruits have been found at 11 chrono-cultural contexts (representing eight archaeological projects) and are mentioned in at least 20 papyri or ostraca from Egypt. In antiquity the only citrus fruit recorded and/or mentioned is citron, Citrus medica. By the 11th century AD we have evidence for a second species, lime, Citrus cf. x aurantifolia, while a further two species (lemon and possibly sour orange) may be present by then as well, though current morphological criteria are not sufficient to offer a definite identification for these rind fragments. Other scientific methods, e.g. aDNA, may ultimately allow verification.

  • 108 E.g. Andrews 1961.

69Thus, despite the fact that citron is present in North Africa and Israel from the 4th century BC (pollen evidence from Carthage and Ramat Rehel) and despite the fact that some authors108 assume the tree was introduced into Italy from Egypt, we have found no evidence for citron in Ptolemaic Egypt. The earliest evidence for citron in Egypt, that we have been able to find, dates to the 1st century AD, and the first specific mention of cultivation of citron in Egypt dates to the late 3rd century AD. However, as we have argued above, the presence of citron in the Eastern Desert sites indicates that citron was available at Coptos and the Nile valley more widely by the 1st century AD, which is highly suggestive of local, Egyptian, cultivation of citron at that time. It was probably a plant product reserved for the wealthy and officials of high rank, but may have become more widely available by the early medieval period. Historical sources emphasize the medicinal value of citron, including its value as an antidote to snake poison, its role in perfumery and, to a much lesser extent, in cuisine; the Greek papyrological sources are silent on the use of citron, but a Coptic ostracon mentions the use of citron for medicinal purposes, while the archaeobotanical evidence is suggestive of the consumption of the rind and the use of complete or half fruits, the latter possibly used in dried form, to ease transportation and storage and/or to be used as a fumigant.

  • 109 Andrews 1961.

70Our evidence may mean that citron was a Roman period introduction into Egypt, rather than the other way round as Andrews109 had suggested. Certainly, the Romans did introduce many new foodstuffs into Egypt and citron may have been one of them. Caution is needed however. The lack of archaeobotanical assemblages from Ptolemaic sites and the near absence of papyri from ordinary settlements of this period mean that there has been little opportunity to find evidence for citron cultivation in Ptolemaic Egypt. The results from this survey should not be taken to mean that citron was definitely not cultivated in Egypt during the Ptolemaic period; rather that we have, to date, not found evidence for it. Future excavations of Ptolemaic sites, if combined with systematic sampling for archaeobotanical remains and papyri, may recover earlier records.

Bibliographie

Amigues 2005: S. Amigues, Végétaux et aromates de l’Orient dans le monde antique, Topoi, 12, 1, p. 359-383.

André 1981: J. André, L’alimentation et la cuisine à Rome, 2nd ed., Paris.

André 1985: J. André, Les noms de plantes dans la Rome antique, Paris.

Andrews 1961: A. Andrews, Acclimatization of Citrus fruits in the Mediterranean region, Agricultural History, 35, 1, p. 35-46.

Armoni et al. 2005: C. Armoni, J.M.S. Cowey, D. Hagedorn (eds.), Die griechischen Ostraka der Heidelberger Papyrus-Sammlung, Heidelberg.

Bagnall 1995: R.S. Bagnall, Reading Papyri: Writing Ancient History, London-New York.

Bagnall 1997: R.S. Bagnall, The Kellis Agricultural Account Book, Oxford, Oxbow Books, (Dakhleh Oasis Project).

Bingen et al. 1992: J.A. Bingen, A. Bülow-Jacobsen, W.E.H. Cockle, H. Cuvigny, L. Rubinstein, W. Van Rengen, Mons Claudianus. Ostraca Graeca et Latina I (O. Claud. 1-190), Cairo.

Bingen et al. 1997: J.A. Bingen, A. Bülow-Jacobsen, W.E.H. Cockle, H. Cuvigny, L. Rubinstein, W. Van Rengen, Mons Claudianus. Ostraca Graeca et Latina II (O. Claud. 191-416), Cairo.

Botto, Salvadei 2005: M. Botto, L. Salvadei, Indagini alla necropoli arcaica di Monte Sirai. Relazione preliminare sulla campagna di scavi del 2002, Rivista di studi fenici, 33, 1/2, p. 81-167.

Boud’hors, Heurtel 2010: A. Boud’hors and C. Heurtel, Les ostraca coptes de la TT29, autour du moine Frangé, vol. 1. Textes, Brussels, CReA-Patrimoine.

Bui Thi Mai, Girard 2014: Bui Thi Mai, M. Girard, Citrus (Rutaceae) was present in the Western Mediterranean in antiquity, in A. Chevalier, E. Marinova, L. Peña-Chocarro (eds.), Plants and people: Choices and diversity through time, Oxford, p. 170-174.

Bülow-Jacobsen 1997: A. Bülow-Jacobsen, The correspondence of Dioscorus and others (224-242), in Bingen et al. 1997, p. 43-68.

Bülow-Jacobsen 2003: A. Bülow-Jacobsen, The traffic on the road and the provisioning of the stations, in Cuvigny 2003b, p. 399-425.

Casson 1989: L. Casson, The Periplus Maris Erythraei, Princeton.

Clapham, Rowley-Conwy 2009: A.J. Clapham, P.A. Rowley-Conwy, The archaeobotany of cotton (Gossypium sp. L.) in Egypt and Nubia with special reference to Qasr Ibrim, Egyptian Nubia, in A.S. Fairbairn, E. Weiss (eds.), From Foragers to Farmers: Papers in Honour of Gordon C. Hillman, Oxford, p. 244-253.

Coubray et al. 2010: S. Coubray, V. Zech-Matterne, A. Mazurier, The earliest remains of a Citrus fruit from a Western Mediterranean archaeological context? A microtomographic-based re-assessment, Comptes Rendus Palevol, 9, p. 277-282.

Cuvigny 1996: H. Cuvigny, The amount of wages paid to the quarry workers at Mons Claudianus, Journal of Roman Studies, 86, p. 139-145.

Cuvigny 2000: H. Cuvigny, Mons Claudianus. Ostraca Graeca et Latina III. Les reçus pour avances à la familia (O. Claud. 417-631), Cairo (DFIFAO, 38).

Cuvigny 2003a: H. Cuvigny, Le fonctionnement du réseau, in Cuvigny 2003b, p. 295-359.

Cuvigny 2003b: H. Cuvigny (ed.), La route de Myos Hormos. L’armée romaine dans le désert oriental d’Égypte. Praesidia du désert de Bérénice I, 2, Cairo (FIFAO, 48).

Cuvigny 2005: H. Cuvigny, Ostraca de Krokodilô: la correspondance militaire et sa circulation, O. Krok. 1-151, Cairo (FIFAO, 51).

Cuvigny 2012: H. Cuvigny (ed.), Didymoi. Une garnison romaine dans le désert oriental d’Égypte (Praesidia du désert de Bérénice IV), 2. Les textes, Cairo (FIFAO, 67).

Darby et al. 1977: W.J. Darby, P. Ghalioungui, L. Grivetti, Food: The Gift of Osiris, 2, London.

Daris 1989: S. Daris, Spigolature documentarie, Aegyptus, 69, p. 85-93.

David et al. 1941: M. David, B.A. Van Groningen, J.C. Van Oven, The Warren papyri (P. Warren), Leiden.

de Vartavan, Asensi Amorós 2010: C. de Vartavan, V. Asensi Amorós, Codex of Ancient Egyptian Plant Remains, London.

Dioscorides, De Materia Medica: Dioscorides Pedanius, Pedanii Dioscuridis Anazarbei de materia medica libri quinque, M. Wellmann (ed.), 3 vols., Berlin, 1: 1907; 2: 1906; 3: 1914 (repr. 1958).

Flower, Rosenbaum 1958: B. Flower, E. Rosenbaum, The Roman Cookery Book : A Critical Translation of the Art of Cooking by Apicius, London.

Förster 2002: H. Förster (ed.), Wörterbuch der griechischen Wörter in den koptischen dokumentarischen Texten, Berlin, W. de Gruyter.

Fournet, Magdelaine 2001: J.-L. Fournet, C. Magdelaine, Liste de vins aromatisés à usage médical (réédition du PAlex 36), in F. Andorlini (ed.), Greek Medical Papyri 1, Florence, p. 163-170.

Fournet 2016: J.-L. Fournet, Sur les premiers documents juridiques coptes (suite): les archives de Phoibammôn et de Kollouthos, in A. Boud’hors, C. Louis (eds.), Études coptes XIV. Seizième journée d’études (Genève, 19-21 juin 2013), Paris, p. 115-141.

Galen, On the Powers of Food: transl. M. Grant, Galen on Diet and Food, London, 2000.

Galen, De simplicium medicamentorum temperamentis ac facultatibus, in C.G. Kühn (ed.), Claudii Galeni opera omnia, vols. 11-12, Leipzig, 1826 (repr. Hildesheim, 1965).

Gascoigne, Rose 2012: A.L. Gascoigne, P.J. Rose, The forts of Hisn al-Bab and the First Cataract Frontier from 5th to 12th centuries AD, Sudan & Nubia, 16, p. 88-95.

Geoponica, sive Cassiani Bassi Scholastici de re rustica eclogae, ed. H. Beckh, Leipzig, Teubner, 1895.

Geoponika 2011: Geoponika: Farm work. A modern translation of the Roman and Byzantine farming handbook, transl. A. Dalby, Totnes (GB), Prospect Books.

Germer 1985: R. Germer, Flora des pharaonischen Ägypten, Mainz.

Germer 1988: R. Germer, Katalog der altägyptischen Pflanzenreste der Berliner Museen, 47, Wiesbaden (Ägyptologische Abhandlungen).

Giacosa 1992: G.I. Giacosa, A taste of Ancient Rome, Chicago-London.

Grenfell, Hunt 1920: B.P. Grenfell, A.S. Hunt, The Oxyrhynchus Papyri. Part XIV, London.

Guéraud 1942: O. Guéraud, Ostraca grecs et latins de l’Wâdi Fawâkhir, BIFAO, 41, 13, p. 168-169.

Habermann 1998: W. Habermann, Zur chronologischen Verteilung der papyrologischen Zeugnisse, Zeitschrift für Papyrologie und Epigraphik, 122, p. 144-160.

Harrauer et al. 1986: H. Harrauer (ed.), with M.H.M. Müller, B. Rom, W. Hameter, B. Palme, H. Taeuber, J. Diethart, K.A. Worp, Corpus Papyrorum Raineri, X. Griechische Texte, VII, Vienna.

Keimer 1984: L. Keimer, Die Gartenpflanzen im alten Ägypten, Mainz am Rhein.

Kovarik 2007: S. Kovarik, Von Zitronen, Melonen und Pfirsichen. Zum Obst- und Gartenbau im spätantiken Ägypten. Neuedition der Pachtvertrage SB 14483 und P.Ross.Georg, III 55 + SB 14485, Archiv für Papyrusforschung und verwandte Gebiete, 53, p. 152-181.

Langgut 2014: D. Langgut, Prestigious fruit trees in ancient Israel: First palynological evidence for growing Juglans regia and Citrus medica, Israel Journal of Plant Sciences <doi 10.1080/07929978.2014.950067>.

Langgut et al. 2013: D. Langgut, Y. Gadot, N. Porat, O. Lipschits, Fossil pollen reveals the secrets of the Royal Persian Garden at Ramat Rahel, Jerusalem, Palynology, 37, 1, p. 115-129.

Lewis 1964: N. Lewis, Four Cornell papyri, Recherches de Papyrologie, 3, 33, p. 32-35.

Loret 1891: V. Loret, Le cédratier dans l’Antiquité, Annales de la société de botanique de Lyon, 17, p. 225-271.

Loret, Poisson 1895: V. Loret, J. Poisson, Études de botanique égyptienne, Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes, 17, p. 177-199.

LSJ: The Online Liddell-Scott-Jones Greek-English Lexicon, <http://stephanus.tlg.uci.edu/lsj/>.

Macaulay-Lewis 2006: E. Macaulay-Lewis, The role of ollae perforatae in understanding horticulture, planting techniques, garden design, and plant trade in the Roman World, in J.-P. Morel, J.-T. Juan, J.-C. Matamala (eds.), The Archaeology of Crop Fields and Gardens, Bari, p. 207-220.

Maspero 1911: J. Maspero (ed.), Papyrus grecs d’époque byzantine, Cairo (Catalogue général des antiquités égyptiennes du Musée du Caire, 1).

Maxfield, Peacock 2001a: V.A. Maxfield, D.P.S. Peacock (eds.), Survey and Excavation at Mons Claudianus, 1987-1993, II. The Excavations, Part 1, Cairo (FIFAO, 43).

Maxfield, Peacock 2001b: V.A. Maxfield, D.P.S. Peacock (eds.), The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998, 1. Topography and Quarries, London.

Moore 2001: G.A. Moore, Oranges and lemons: clues to the taxonomy of Citrus from molecular markers, Trends Genet, 17, 9, p. 536-540.

Müller-Wodarg 1957: D. Müller-Wodarg, Die Landwirtschaft Ägyptens in der frühen Abbasidenzeit 750-969 n. Chr. (132-358 d. H.): 3. Nutzplanzen und Baüme, Der Islam, 32, p. 14-78.

Pagnoux et al. 2013: C. Pagnoux, A. Celant, S. Coubray, G. Fiorentino, V. Zech-Matterne, The introduction of Citrus to Italy, with reference to the identification problems of seed remains, Vegetation History and Archaeobotany, 22, p. 421-438.

Papathomas 1994: A. Papathomas, Überlegungen zu dokumentarischen Texten aux Ägypten, Zeitschrift für Papyrologie und Epigraphik, 103, p. 86-87.

Papathomas 1996: A. Papathomas (ed.), Veröffentlichungen aus der Heidelberger Papyrussammlung, VII, Heidelberg.

Peacock, Blue 2006: D.P.S. Peacock, L. Blue, Myos Hormos-Quseir al-Qadim, Roman and Islamic Ports on the Red Sea, 1. The Survey and Report on the Excavations, 1, Oxford.

Peacock, Blue 2011: D.P.S. Peacock, L.K. Blue (eds.), Myos Hormos-Quseir Al-Qadim, Roman and Islamic ports on the Red Sea, 2. Finds from the excavations 1999-2003, Oxford.

Peacock, Maxfield 1997: D.P.S. Peacock, V.A. Maxfield (eds.), Survey and Excavations at Mons Claudianus 1987-1993, 1. Topography & Quarries, Cairo.

Peacock, Maxfield 2007: D.P.S. Peacock, V.A. Maxfield (ed.), The Roman Imperial Quarries. Survey and Excavations at Mons Porphyrites 1994-1998, 2. The Excavations, London.

Pintaudi 1981: R. Pintaudi, PSI V 481 verso: conto di spese, Aegyptus, 61, p. 93-102.

Pliny: Natural History, vol. 4, transl. H. Rackham, London-Cambridge M.A., 1952 (Loeb Classical Library).

Rose, Gascoigne 2013: P.J. Rose, A.L. Gascoigne, Hisn al-Bab: More symbol than substance, in F. Jesse, C. Vogel (eds.), The Power of Walls. Fortifications in Ancient Northeastern Africa. Proceedings of the International Workshop held at the University of Cologne 4th-7th August 2011, Köln, p. 1-18.

Russo 1998: S. Russo, Il corredo dotale di una giovane antinoita, in L. Del Francia Barocas (ed.), Antinoe cent’anni dopo. Catalogo della mostra, Firenze, Palazzo Medici Riccardi 10 luglio-1o novembre 1998, Firenze, p. 149-153.

Schram 2014-: V. Schram, L’arbre et le bois dans l’Égypte gréco-romaine, PhD thesis in Greek papyrology, in preparation under the supervision of J.-L. Fournet, École pratique des hautes études, Paris, 2014-.

Schuman 1980: V.B. Schuman, Washington University papyri I. Non-literary texts: nos. 1-61, Chico.

Shelton 1971: J.C. Shelton, Papyri from the Michigan collection, Toronto.

Sidebotham 1986: S.E. Sidebotham, Roman Economic Policy in the Erythrea Thalassa 30 BC-AD 217, Leiden.

Sijpesteijn 1977a: P.J. Sijpesteijn (ed.), The Wisconsin Papyri II, Zutphen.

Sijpesteijn 1977b: P.J. Sijpesteijn (ed.), The Michigan Papyri XIII, The Aphrodite Papyri in the University of Michigan Papyrus Collection, Zutphen.

Strabo, The Geography, transl. H.L. Jones, London-Cambridge M.A, 1969 (The Loeb Classical Library).

Täckholm 1961: V. Täckholm, Botanical identification of the plants, in C. Bachatly (ed.), Le monastère de Phoebammon dans la Thébaïde, 3. Identifications botaniques, zoologiques et chimiques, Cairo, p. 1-38.

Tengberg 2011: M. Tengberg, L’acquisition et l’utilisation des produits végétaux à Didymoi, in H. Cuvigny (ed.), Didymoi. Une garnison romaine dans le désert oriental d’Égypte (Praesidia du désert de Bérénice IV), 1. Les fouilles et le matériel, Cairo (FIFAO, 64), p. 205-214.

Thanheiser 2002: U. Thanheiser, Roman agriculture and gardening in Egypt as seen from Kellis, in C.A. Hope, G.E. Bowen (eds.), Dakhleh Oasis Project: Preliminary reports on the 1994-1995 to 1998-1999 field seasons, Oxford, p. 299-310.

Thanheiser et al. 2016: U. Thanheiser, S. Kahlheber, T. Dupras, Pearl millet, Pennisetum glaucum (L.) R.Br. ssp. glaucum, in the Dakhleh Oasis, Egypt, in U. Thanheiser (ed.), News from the Past: Progress in African Archaeobotany. Proceedings of the 7th International Workshop on African Archaeobotany in Vienna, 2-5 July 2012, Groningen, p. 115-125.

Theophrastus: Enquiry into Plants, transl. A.F. Hort, London-Cambridge M.A, 1916 (Loeb Classical Library).

Van der Veen 1998: M. Van der Veen, Gardens in the Desert, in O.E. Kaper (ed.), Life on the Fringe: Living in the Southern Egyptian Deserts during the Roman and early-Byzantine Periods, Leiden, p. 221-242.

Van der Veen 2001: M. Van der Veen, The botanical evidence, in Maxfield, Peacock 2001a, p. 174-247.

Van der Veen 2011: M. Van der Veen, Consumption, Trade and Innovation: Exploring the Botanical Remains from the Roman and Islamic Ports at Quseir al-Qadim, Egypt, Frankfurt.

Van der Veen, Morales 2014: M. Van der Veen, J. Morales, The Roman and Islamic spice trade: New archaeological evidence, Journal of Ethnopharmacology, 67, p. 54-63 <doi 10.1016/j.jep.2014.09.036>.

Van der Veen, Tabinor 2007: M. Van der Veen, H. Tabinor, Food, fodder and fuel at Mons Porphyrites: The botanical evidence, in Peacock, Maxfield 2007, p. 83-142.

Van der Veen et al. in press: M. Van der Veen, C. Bouchaud, R. Cappers, C. Newton, Roman life in the Eastern Desert of Egypt: Food, imperial power and geopolitics, in J.-P. Brun, Th. Faucher, B. Redon, S. Sidebotham (eds.), Le désert oriental d’Égypte durant la période gréco-romaine: bilans archéologiques, Collection électronique de l’Institut des civilisations du Collège de France, série ‘Mondes méditerranéens antiques’ <https://books.openedition.org/cdf/4673>.

Van Zeist et al. 2001: W. Van Zeist, S. Bottema, M. Van der Veen, Diet and Vegetation at Ancient Carthage. The Archaeobotanical Evidence, Groningen.

Watson 1983: A.M. Watson, Agricultural Innovation in the Early Islamic World: The diffusion of crops and farming techniques, 700-1100, Cambridge.

Wessely 1921: C. Wessely (ed.), Studien zur Palaeographie und Papyruskunde, XX, Leipzig.

Whitcomb, Johnson 1979: D.S. Whitcomb, J.H. Johnson, Quseir al-Qadim preliminary report 1978, Cairo.

Whitcomb, Johnson 1982: D.S. Whitcomb, J.H. Johnson, Quseir al-Qadim 1980: Preliminary report, Cairo.

Zohary et al. 2012: D. Zohary, M. Hopf, E. Weiss, Domestication of plants in the Old World: The origin and spread of domesticated plants in Southwest Asia, Europe, and the Mediterranean Basin, Oxford.

Notes

1 Amigues 2005; Zohary et al. 2012: 146-47.

2 Langgut et al. 2013; Langgut 2014.

3 Van Zeist et al. 2001.

4 Bui Thi Mai, Girard 2014: 174.

5 Coubray et al. 2010; Pagnoux et al. 2013.

6 Botto, Salvadei 2005.

7 Loret 1891: 247-249.

8 Germer 1985: 106.

9 Keimer 1984.

10 Andrews 1961.

11 Whitcomb, Johnson 1979; 1982.

12 Peacock, Blue 2006; 2011.

13 For further detail, see Van der Veen 2011.

14 Van der Veen 2011; Van der Veen, Morales 2014.

15 Moore 2001; Watson 1983.

16 See also Van der Veen 2011: 83-89.

17 Maxfield, Peacock 2001a; Peacock, Maxfield 1997.

18 For more details, see Van der Veen 2001.

19 Bülow-Jacobsen 1997; 2003; Van der Veen 1998; 2001.

20 Van der Veen 2001: 220.

21 For more details, see Van der Veen 2001.

22 Maxfield, Peacock 2001b; Peacock, Maxfield 2007.

23 For more details, see Van der Veen, Tabinor 2007.

24 Van der Veen, Tabinor 2007.

25 Keimer 1984: 1-2, and endnote 6, which reads: Bot. Museum Berlin-Dahlem, Sammlung Schweinfurth, Listen-Nr. 47; and Germer 1988: 14-20, and figure 11 (the caption of figure 11 reads: S. Schw. Nr.47).

26 de Vartavan, Asensi Amorós 2010: 80.

27 P. Kell. IV Gr. 96, Bagnall 1997; Thanheiser 2002.

28 For more details, see Thanheiser 2002; Thanheiser et al. 2016.

29 Thanheiser 2002.

30 For more details, see Van der Veen, Tabinor 2007.

31 Täckholm 1961.

32 Täckholm 1961: 23-24.

33 Germer 1985.

34 A. Clapham, personal communication; Gascoigne, Rose 2012; Rose, Gascoigne 2013.

35 Clapham, Rowley-Conwy 2009.

36 For further detail, see Clapham unpublished data.

37 Peacock, Blue 2006; 2011.

38 For further details, see Van der Veen 2011; Van der Veen, Morales 2014.

39 Pagnoux et al. 2013, appendix 3.

40 For more details, see Van der Veen 2011.

41 Thanheiser, personal communication; see also <http://swissinst.ch/html/forschung_neu.html>.

42 Darby et al. 1977: 704-706, fig. 18.2.

43 Initially identified as lemon, but subsequently identified as citron; Loret, Poisson 1895: 194-196; Darby et al. 1977: 704-706; Keimer 1984: 1; Germer 1985: 106.

44 Keimer 1984: 1.

45 de Vartavan, Asensi Amorós 2010: 80.

46 Habermann 1998.

47 E.g. Bingen et al. 1992; 1997; Bülow-Jacobsen 2003; Cuvigny 2000; 2003b; 2005; 2012.

48 André 1985: 68; Amigues 2005: 364.

49 I.109.3, ed. Wellmann.

50 Harrauer et al. 1986: 164.

51 Daris 1989: 93; Bülow-Jacobsen 2003: 420; Armoni et al. 2005: 375.

52 Grenfell, Hunt 1920: 19 and 183; Sijpesteijn 1977a: 83-84.

53 LSJ; Fournet, Magdelaine 2001; Amigues 2005.

54 André 1985: 68.

55 I.109.3-4, ed. Wellmann.

56 Andrews 1961: 41-42.

57 Something also mentioned by Galen (late 2nd century AD): “Its fruit is no longer called ‘Median apple’ but kitrion by everyone”. Galen, De Simplicium VII.12.19.2 (XII, p. 77.4, ed. Kühn). Note that the word κιτρέα, borrowed from the latin citrea, and referring to the citron-tree, also appears in the Geoponica, a collection of agricultural lore, compiled during the 10th century, Geop. III.13.6 and X.7.8-11 (ed. H. Beckh; for translation see Geoponika 2011: 110; 204).

58 Loret 1891.

59 The papyrological abbreviations follow the Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca and Tablets, available at: <http://papyri.info/docs/checklist>.

60 Guéraud 1942: 168-169, no. 13.

61 Bülow-Jacobsen 2003: 420.

62 H. Cuvigny and A. Bülow-Jacobsen, personal communication.

63 Shelton 1971: 57-61.

64 David et al. 1941: 44-45.

65 Grenfell, Hunt 1920: 19.

66 Schuman 1980: 42-43.

67 Grenfell, Hunt 1920: 183.

68 Armoni et al. 2005: 375.

69 Sijpesteijn 1977a: 83-84.

70 Lewis 1964: 32-35.

71 Papathomas 1996: 175.

72 Maspero 1911; Russo 1998: 150.

73 Harrauer 1986: 165.

74 Wessely 1921: 128.

75 Armoni et al. 2005, see comm. ad l. 20 p. 375.

76 Sijpesteijn 1977b: 81; concerning the date, see Fournet 2016.

77 Kovarik 2007; Schram 2014-.

78 Harrauer 1986: 108-109; Schram 2014-.

79 Boud’hors, personal communication; see also Förster 2002: 417.

80 Boud’hors, Heurtel 2010: 354.

81 Boud’hors, Heurtel 2010: 21.

82 Watson 1983: 45.

83 Müller-Wodarg 1957: 73.

84 H. Cuvigny, personal communication.

85 Habermann 1998.

86 Pagnoux et al. 2013: 424.

87 Van Zeist et al. 2001; Langgut et al. 2013; Langgut 2014.

88 Bui Thi Mai, Girard 2014; papers this volume.

89 Langgut et al. 2013.

90 Bagnall 1995: 13-14.

91 Bagnall 1995: 27.

92 Van der Veen et al. in press.

93 Sidebotham 1986; Bingen et al. 1992; 1997; Cuvigny 1996; 2000; 2003a; 2012; Bülow-Jacobsen 2003.

94 Bülow-Jacobsen 1997; 2003; Van der Veen 1998.

95 Bülow-Jacobsen 2003; Strabo, The Geography 17.1.45.

96 Van der Veen 2001; Tengberg 2011; Van der Veen et al. in press.

97 I.e. Periplus Maris Erythraei, Casson 1989.

98 Theophrastus’ Enquiry into Plants IV.4.2; Pliny’s Natural History XII 7.15-16.

99 Pagnoux et al. 2013; chapters of this volume.

100 Bülow-Jacobsen 2003.

101 Macaulay-Lewis 2006.

102 André 1981; Andrews 1961; Flower, Rosenbaum 1958; Zohary et al. 2012.

103 Giacosa 1992: 95.

104 Van der Veen 2011: 185-187.

105 Ibidem.

106 Van der Veen et al. in press.

107 Cuvigny 2003a: 314-315, 339-340.

108 E.g. Andrews 1961.

109 Andrews 1961.

Table des illustrations

Titre Table 1 - Citrus finds at archaeological sites in Egypt.
Légende Numbers refer to the minimum number of items present.References: a) Van der Veen 2011; b) Van der Veen 2001; c) Van der Veen, Tabinor 2007; d) Germer 1988; e) Thanheiser 2002; f) Täckholm 1961; g) Clapham personal communication; h) Thanheiser personal communication.
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-1.jpg
Fichier image/jpeg, 900k
Titre Fig. 1 - Map of Egypt listing the sites mentioned in the text.
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-2.jpg
Fichier image/jpeg, 968k
Titre Fig. 2 - Archaeobotanical remains of citrus seeds.
Légende a. Type A - citron, Citrus cf. medica, first two from Myos Hormos, 1st-3rd c. AD, second two from Kusayr, 11th-13th c. AD (from Van der Veen 2011, fig. 3.4); b. Type A - Citrus cf. medica from Mons Porphyrites, 2nd century AD (from Van der Veen, Tabinor 2007, fig. 4.9); c. and d. Type B, cf. lime/lemon, Citrus cf. x aurantifolia/limon from Kusayr, 11th-13th c. AD (from Van der Veen 2011, fig. 3.4). See Table 2 for identification criteria (photographs: Jacob Morales).
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-3.jpg
Fichier image/jpeg, 2,3M
Titre Fig. 3 - Archaeobotanical remains of citrus fruits and rind.
Légende a. near-complete fruit of citron, Citrus cf. medica, with seeds still in situ, from Mons Claudianus, 2nd century AD (from Van der Veen 2011, fig. 3.5); b. cluster of seeds embedded in fruit pulp, Citrus cf. medica, from Mons Porphyrites, 2nd century AD; c. rind of half a citron, Citrus cf. medica, from Kusayr, 11th-13th century AD (from Van der Veen 2011, fig. 3.5). See Table 2 for identification criteria (photographs: Jacob Morales).
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-4.jpg
Fichier image/jpeg, 2,3M
Titre Table 2
Légende Identification criteria for the citrus remains found at Quseir al-Qadim, i.e. Myos Hormos and Kusayr, from Van der Veen 2011: 83-89. See also figs. 2-5 in this paper.
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-5.jpg
Fichier image/jpeg, 816k
Titre Fig. 4 - Archaeobotanical remains of citrus rind.
Légende a. modern lime, Citrus x aurantifolia; b., c. and d. Type 2 rind, cf. lime, Citrus cf. x aurantifolia, all from Quseir al-Qadim, 11th-13th century AD (from Van der Veen 2011, fig. 3.6); central column of rows a, b, and d show the small pedicel scar; row c shows the small nipple. See Table 2 for identification criteria (photographs: Jacob Morales).
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-6.jpg
Fichier image/jpeg, 1,5M
Titre Fig. 5 - Archaeobotanical remains of citrus rind.
Légende a. Type 3 rind, cf. lemon/lime, Citrus x limon/x aurantifolia (11th-13th century AD); b. Type 4, indeterminate, both from Kusayr, 11th-13th century AD (from Van der Veen 2011, fig. 3.6). See Table 2 for identification criteria (photographs: Jacob Morales).
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-7.jpg
Fichier image/jpeg, 480k
Titre Table 3 - Papyrological evidence for citrus fruits from Egypt.
Légende References: a. http://papyri.info/​search; b. Bülow-Jacobsen 2003; c. Cuvigny and Bülow-Jacobsen, personal communication; d. Shelton 1971: 57-61; e. David, Van Groningen, Van Oven 1941: 44-45; f. Grenfell, Hunt 1920: P.Oxy. XIV 1631 p. 15-25; P.Oxy. XIV 1764 p. 183 (descr.); g. Schuman 1980: 42-43; h. Armoni, Cowey, Hagedorn 2005: 368-376, esp. 375; i. Sijpesteijn 1977a: 83-85; j. Lewis 1964; k. Pintaudi 1981: 98-99; l. Schram 2014-; m. Maspero, 1911: 22-33; n. Wessely 1921: 128; o. Harrauer et al. 1986: CPR X 62 p. 108-109 ; CPR X 138 p. 164 ; CPR X 139 p. 165; p. Kovarik 2007; q. Sijpesteijn 1977b: 81-87; r. Boud’hors, Heurtel 2010: 46-47; s. Papathomas 1996: 175-190; t. Daris 1989; u. Guéraud 1942; v. Papathomas 1994 (SB XVI 12254, 8 and 11); w. Russo 1998; x. Boud’hors, Heurtel 2010: 354; y. Boud’hors, Heurtel 2010: 360; z. Boud’hors, personal communication.
URL http://books.openedition.org/pcjb/docannexe/image/2179/img-8.jpg
Fichier image/jpeg, 1,4M

Auteurs

Departamento de Ciencias Históricas, Área de Prehistoria, Universidad de Las Palmas de Gran Canaria, Spain (present address)

École Pratique des Hautes Études; UMR 8167 - Orient & Méditerranée, Paris

School of Archaeology & Ancient History, University of Leicester, UK; mvdv1@le.ac.uk

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search