Texts and Textiles: A Study of the Textile Industry at Ancient Kellis
By Gillian Bowen
Abstract: Excavations at Ismant el-Kharab, ancient Kellis in the Dakhleh Oasis, Egypt, have yielded a significant number of textiles and tools used in the weaving industry Textual evidence from the site has shown that the occupants of one of the houses were involved in a small weaving and tailoring business. The texts have thrown considerable light upon several aspects of that business: the products manufactured, clothing worn, the yarns and colours used. Scientific analysis of some of the fabric and yarn samples is currently being undertaken by the CSIRO, textiles division. This paper examines the current state of our knowledge of the textile industry at Kellis, by combining the artefactual, textual and scientific data.
In 1999, work commenced on the study of the textiles from the excavations at Ismant el-Kharab, Dakhleh Oasis, Egypt. Each fragment, no matter how small, was catalogued: the piece was described in full, taking note of the number of warps and wefts per square centimetre, and where possible, the fibre was identified. In 2000, the study of the Kellis textiles was expanded when three samples were analysed by Dr Jeff Church, Principal Research Scientist, Fibre Structure and Function, CSIRO Wool Technology Geelong. The analyses were undertaken in order to determine the yarns, dyes and mordants used, and to determine if possible, whether the raw materials were local or imported. That same year it was decided to broaden the study by incorporating data related to the tailoring and weaving industry contained in the texts which have been discovered at the site. The following is an interim report and represents work in progress.
Excavations at Ismant el-Kharab
Ismant el-Kharab, ancient Kellis, is a large late Ptolemaic-late Roman period village located in the Dakhleh Oasis, Egypt, some 800 km south-south-west of Cairo and approximately 300 km due west of Luxor in the Nile Valley (see C. A. Hope in this volume; Figure 1). The remains of the village are remarkably well preserved (Knudstad and Frey 1999). Excavations, under the direction of Dr Colin A. Hope of Monash University, have been carried out annually since 1986. Extensive excavation has been undertaken in the Temple of Tutu and three of its associated shrines (Hope 1998: 803-58; in press a), three churches (Bowen 1998: 243-78; 2000: 29-34; in press), four 4th century houses (Hope 1985: 114-25; 1986: 74-91; 1989: 1-12; 1990: 43-54), domestic structures in the north-east of the village dating to the 2nd and 3rd centuries (Hope 2000: 62-3; in press a), 4th century domestic structures to the north of the Temple of Tutu (Hope 2000: 57; in press a), a Christian cemetery and five mausolea (Hope and McKenzie 1999: 53-68; Hope 2000: 57-61; in press b) (Figure. 1). Textual and numismatic data have shown that the village was abandoned in the late 4th century and consequently, much of the material remains can be dated to that period (Hope 2001: 43-59; Bowen 1998: 3, 10-30).
Excavations have yielded a vast array of artefactual material amongst which are large quantities of textile. Textiles have been found within all the excavated structures in the village as well as the rubbish dumps, but the greatest concentration is in the cemeteries where the bodies of many of those interred were wrapped in plain linen shrouds. With a single exception, a fragmentary hooded child's tunic with embroidered elements retrieved from one of the pit graves, decorated textiles are only attested in the settlement. There are few substantial fragments surviving; most are scraps, some of which are no more than two centimetres square.
In many instances, the decoration had been cut from its twill, diamond twill and double-face. The finest cloth is original garment and crudely incorporated into a piece a tightly-spun, evenly woven fabric with a thread count of patchwork that may well have served as a blanket or a heavy outer garment. It is difficult to determine whether the scraps are from items of clothing, tablecloths, cushions, blankets or wall hangings, all of which are attested at Kellis, as the texts retrieved from one of the houses clearly indicate (Bowen 1999; in press).
The texts provide an invaluable insight into the types of garments worn, the variety of textiles manufactured in the village, and the yarns used. These documents, most of which date to the late 4th century, are written in both from Greek and Coptic; the former was the bureaucratic language of Egypt from the 4th century BCE until the Islamic conquest in the 7th century CE, and the latter is Egyptian written in Greek script with an additional six characters from the demotic script. An important document relating to commodities produced in the village and surroundings regions is the Kellis Agricultural Account Book (Bagnall 1997). This eight-leaf wooden codex was found, together with a literary codex was found, together with a literary codex (Rijksbaron And Worp 1997), in the kitchen of House 2 (Hope 1997: 5-16); it records the accounts of an estate over a four-year period.
The discovery of spindle whorls and loom weights in every domestic structure, as well as in some of the temple shrines that were occupied by squatters from about the middle of the 4th century, suggests that spinning and weaving was carried out within many households, no doubt to meet certain of the family's needs. The Kellis corpus however, attests a small textile and tailoring business organized by a female occupant of one of the excavated houses; it includes private letters, deeds of sale, contracts, and accounts. Numerous fragments from implements used in the weaving trade, as well as scraps of decorated fabrics and unwoven warp threads were retrieved from the same house.
The Kellis Textiles: weaves, decoration, fibres, dyes and mordants
In excess of 40 variations in weave and quality of fabric have been identified. The basic weave is plain or tabby, the quality of which varies from fine gauze to a heavy canvas. Basket weave in several variations is well represented as is tapestry, looped weave, boucle, soumak, twill, diamond twill and double-face. The finest cloth is a tightly-spun, evenly woven fabric with thread count of around 36 warp and 26 weft per centimeter; the more coarse and uneven the yarn the lower the count, with the loosely-woven Hessian-like fabrics having no more than five wefts and warp threads per centimeter.
Decorative fabrics from the late Roman period on are commonly known as Coptic. The ground fabric is linen and the decoration is woven with dyed wool. Decorated fabrics from Kellis may be classified into the following categories (Bowen 1999; in press):
Plate 1: Two fragments of purple-dyed, double clavi in rinceau design; wool on linen; upper sample from the North Structure, the lower sample from House 2.
These are vertical bands of tapestry weave that extend from the shoulder of a tunic. There are examples of both single and double clavi in varying widths at Kellis. The clavus often has a parallel false braid, or soumak weave, in linen on either side. Most clavi are solid blocks of colour but some are ornamented with what is termed rinceau, a continuous pattern in the form of a vine (Plate 1).
The majority found to date are woven with purple wool on a linen ground; the rinceau-type incorporate the ground fabric into the design. Clavi are well attested in garments found in Roman Egypt, and are depicted on so-called Fayum portraits (see especially Doxiades 1995; Walker and Bierbrier 1997). Purple clavi were used as a mark of rank in Rome itself; however, in the eastern Empire it may have had no such significance (Shore 1972:15) and could simply attest an affiliation to Roman customs (Walker and Bierbrier 1997: 16). The quantity retrieved from Ismant el-Kharab, and the context in which some were found, adds credence to the latter statement.
2. Geometric designs.
Plate 2: One of the two fragments of identical geometric design, wool on linen, from structure C/2/2.
This category comprises blocks of colour in a variety of square and/or rectangular patterns and stripes feature amongst the samples. Colour combinations include orange/green, red/green/brown and red/blue.
The illustrated fragment (Plate 2), which measures 27.0cm by 9.0cm, preserves four decorative bands, the uppermost of which is a false braid in red and blue contained within three wefts of blue. The second band has alternating blue, yellow, red and purple diamond motifs. The third band comprises solid blocks of natural, blue, red, purple, blue, yellow, red and natural; the natural blocks at each end have a red weft running through the centre. The lowest surviving decorative band is an alternating red and blue herringbone. The fragment was found in a domestic structure in C/2/2, Room 3 S. no. S97/124 in mud-brick collapse (Hickson in press; Hope 1999: 62). An identical piece of fabric was retrieved from the rubbish dump. It is not known whether the fragments were from items of clothing, wall hangings or cushions.
3. Complex geometrical designs.
Plate 3: Fragment of lozenge design, wool on linen, from House 3.
These are polychrome, decorative elements such as inter-woven lozenges and so-called 'yo-yo' design (Barber 1991: 317). Samples include alternating lozenges of yellow, purple, red and green (Plate 3), circles of dark green edged in yellow on a red ground, and ovals of yellow contained within an orange frame on a dark brown ground. All examples are from 4th century contexts. The fragments preserved are too small to determine their original purpose.
4. Plaids, checks and ginghams
Invariably in unbleached and blue linen, look modern (Plate 4). This category of decorative textile does not come under the classification Coptic. All examples found to date are in the same quality weave with a thread count of between 14-16 in the warp and 11-14 in the weft. It is not unreasonable to suggest that these fabrics were from clothing. A sample of the check design is found amongst the Egyptian textile collection in the Ashmolean Museum, Oxford, although without provenance. Other examples have been excavated at the Red Sea coastal site of Berenike (Dhaliwal 1996: 259, Plate 14-1). This suggests that the design enjoyed a wide distribution in Egypt during the 3rd and 4th centuries. This fabric has been found in conjunction with a plain blue linen textile that bears a remarkable resemblance to modern denim.
Linen, usually in its unbleached form, was the most common yarn used at Kellis. There is a variation in colour ranging from off-white to a light brown; two large pieces from a shawl have a golden-brown hue but whether this is the result of dyeing the flax will not be known until a sample has been analysed. It is presumed that the plants were grown locally and flax capsules have been recovered from the site (Thanheiser 1999; in press).
Cotton has been identified by infrared microscopy (Coombs, Woodhead and Church, in press). Entries for cotton, including, 'cotton for weaving' are recorded in the Kellis Agricultural Account Book (Bagnall 1997: lines 547, 556, 558-59, 720 and 1484). Two receipts for the sale of cotton were found amongst an archive of ostraka that dates to the second half of the 3rd century and belonged to Psais son of Soul who was presumably involved in the sale of this commodity. The archive was found in a domestic structure in Area Cof the site (Hope 1999: 64). Further evidence for the growing of cotton in the region is the presence of numerous cotton bolls and seeds found at the site (Thanheiser and Bagnall 1997: 39-40; Thanheiser 1999; in press). Coombs, et al. (in press) note that the dimension of the fibres of the analysed sample corresponds to the species grown in Egypt today: Gossypium barbadense.
Plate 4: A fragment of blue and natural linen, found during surface clearance of the walls in area B.
Wool, both natural and dyed, has been found in its spun and unspun state; scanning electron microscopy analysis has shown that it derives from sheep (Coombs et al. in press). Whether or not the wool was produced in the village is inconclusive; texts from the houses imply that it was, but sheep are almost absent from the faunal remains from the site (Churcher, in press) and wool is not attested amongst the commodities listed in the Kellis Agricultural Account Book. The latter may well be due to the nature of the accounts for, as Bagnall (1997: 24) cautions, it cannot be known whether all commodities or important agricultural or livestock activities concerning the estate are documented therein. As in the oasis today, the herding of a few sheep along with goats is likely; but whether there were sufficient sheep for the requirements of the village cannot be posited with certainty. Perhaps DNA sampling can provide a conclusive answer.
Textual evidence suggestive of wool production within the village may be found in the following private letters, two written in Greek and the third in Coptic; all were recovered from House 3.
1. Written by Pamouris to Psais. Pamouris is in Kellis awaiting Psais' return. The letter reads:
Give many greetings for me to mother Maria and the little Tsempnouthes. Please send the girl to me. I am giving you her travel money and each year I will give you a present of wool for a cloak as her hire (P. Kell I Gr. 71).l
2. This letter, addressed to Pamouris, is from Pekysis who is living away from Kellis; reads:
I wonder why you have not sent me even one fleece, though you know that we had no other one in hand; you did not send it, neither to me nor to your own son... Please buy for me the little amount of nicely coloured wool (P. Kell I Gr. 72).
3. The third letter was sent by Ammon living away from the village to Pshai in Kellis.
If you know that Louitoni has good wool for the colour of my black sticharion, take some for me (P. Kell. Copt. 37.28).
Other texts imply no shortage of wool, which was clearly available in a variety of colours and qualities. Coloured wool was incorporated into linen garments to form decorative elements and was also used for embroidery. From textual evidence it is known that wool was used in the manufacture of entire garments: 'Find 5 mna of use-able wool and send them; we can make them a garment for him' (P . Kell. Copt. 12.8), also for headscarves (P. Kell. Copt. 47.6; 48.13) and blankets: 'I gave you a share of the warp (and) made the wool for the blanket' (P. Kell. Copt. 44.6). Wool was sold both by the centenarium and by the mna: 'I have brought a centenarium and a half of wool, and six mna of dyed wool' (P. Kell. Copt. 47.1-2). The former is equivalent to 32.3 kg; the mna is probably the equivalent of 32.3 gm (Gardner et al. 1999: 64-5).
Several fragments of coarsely woven fabric are amongst the samples recovered. The colour is black/brown. Some of the fragments are of a substantial size; one sample was used as a lining in a composite garment. Initially it was considered to be animal hair (Bowen 1998), however, Jeff Church (personal communication, June 2001) has identified it as plant fibre, probably hemp.
Dyes and mordants
Certain of the linen samples were dyed, these are mostly the blue plaids and checks; the colours do not exhibit the same intensity as those of the dyed wool. This may be attributed to the fact that the process took place after the yarn was spun or, alternatively, that flax is a difficult fibre to dye (Barber 1991: 133). A fragment of green linen retrieved from North Tomb 1 (sample NT1/2000.188) was printed, not dyed (J. Church, personal communication, September 2001). As no mordant was used, and therefore the fabric could not have been laundered, it must have been produced specifically for funerary purposes. The sample is currently in the process of analysis.
All of the woven wool that has been retrieved is dyed. The process took place prior to spinning as numerous pieces of unspun wool attest. There is a wide range of colours that include various shades of green, blue, yellow, orange, red, and purple. Analyses carried out on three fabric samples have shown that the dyes are vegetable-based from plants that may well have grown locally. Through a series of extractions, a dark-green fabric (Sample A/6 S93.559) was shown to have been dyed first with orange, probably madder, and then blue, either woad or indigo; pyridine extraction of purple fibres, taken from a surface sample of cloth, resulted in a blue extract and brilliant red fibres (Coombs et al., in press). Coombs et al. (in press) concluded that the purple colour was achieved by a mixture of red and blue dye, again probably madder or woad/indigo.
Both madder ( Rubia tinctorum L. ) and woad ( Isatis tinctoria L. ) are indigenous to North Africa (Balfour-Paul 1998: 23); indigo derives from India (Pliny XXXV.27, 46) although indikon was traded into the west and was cultivated by the Romans (Balfour-Paul 1998: 23). Carroll (1986: 33), citing Pfister (1935), Forbes (1956: 98-126) and Lucas (1962: 150-59), states that with a single exception, madder, the dyestuffs used in Egypt in the Roman period were either not known or used in earlier times; if this is correct, woad can be excluded as a possibility from the Kellis fabrics. Lucas (1962: 151) actually states that woad was cultivated in the Fayum in the 1st to 4th centuries CE. The texts cited by Lucas however, all prohibit the cultivation of the plant ( Papyrus Oxyrhynchus I, II and X). Indigo resist-dyed woollen fragments have now been identified from Maximianon el-Zerga in the Eastern Desert; the wool and linen hangings of the Late Roman period from Akhmim are mostly dyed with indigo (Balfour-Paul 1998 24). Further analyses of the Kellis samples should determine which of the two blue dyestuffs was used.
The purple fibres from the surface sample were found to contain a large amount of aluminium; this indicates the use of alum as a mordant (Coombs et al., in press). Alum is well attested from the Dakhleh Oasis (Beadnell 1901; Lucas 1962 257-9); this provides further evidence that the wool was dyed locally.
Weaving and spinning implements
Figure 2: Wooden spindle whorls: three from area B (a, b and d) and one from House 2 (C), (drawings by Johnothan C. Howell)
Figure 3: Wooden comb used either for weaving or for combing the fibres; from outside entrance to House 2 (drawing by Johnothan C. Howell)
Numerous spinning and weaving tools have been found in the excavations. Some are easy to identify, such as hemispherical, wooden spindle whorls (Figure 2), wooden weaving and flax combs (Figure 3), shuttles, needles and loom weights; with some categories, an immediate identification is less certain.
Lengths of what appeared to be doweling are now recognised as shafts from spindle whorls; notched sections of palm probably functioned as heddles. Pegs and corresponding holes set in the walls may have served as warping frames; these have been observed in Houses 1-3. The peg holes are placed in such a fashion that the warp can be wound around to form a figure-of-eight shape; this process is well attested from pharaonic times and is perhaps best illustrated in a model of spinning and weaving workshop that was found in the tomb of Meketra and dates to the early Middle Kingdom (Cairo, JE 46723; Vogelsang-Eastwood 2000: 268). Sections of worked-wood also found in Houses 1-3 bear a striking resemblance to the horizontal, body-tension, ground looms used by the weavers of Lombok, Indonesia (personal observation). With body-tension looms, the weaver uses her body to hold the warp yarns in tension; such looms are well suited for domestic production due to their portability (Hitchcock 1991: 55). An advantage of the ground loom is that it does not restrict the length of cloth that can be produced; the warp is simply wrapped around the warp beam and unwrapped as required (Vogelsang-Eastwood 2000: 276). Ground, or horizontal looms are known from Egypt from Predynastic times and continued into the 20 th century (Vogelsang-Eastwood 2000: 277); the earliest attested is a painting on a pottery bowl from a woman's tomb at Badari (Tomb number 3802; Petrie Museum, UC 9547) and again in the model from the tomb of Meketra, as well as in several representations in tomb-chapel paintings of the pharaonic period (Vogelsang-Eastwood 2000: 268, 276-7).
Figure 4: Drawing of a Greek warp-weighted loom from a painting on an Attic red-figure skyphos dating to 450-20 BCE from D. Carroll (1988: figure 5).
The quantity of loom weights found in domestic structures however, suggests that a common type of loom at Kellis was the warp-weighted variety that derives from the Greek and Roman tradition, as illustrated on numerous vase paintings (Figure 4) (Carroll 1986: 23-25). This presumably replaced the Egyptian vertical, or fixed two-beam loom, attested from the Eighteenth Dynasty (Vogelsang-Eastwood 2000: 268, 278) and still in use in the 5 th century BCE (Herodotus: 2.81). Eleven large, clay loom weights were found in Rooms 2 and 3 of the structure C/2/1 (Hickson, in press), clear evidence that weaving took place on the premises. Examples of warp-weighted looms have not been found at Kellis although the remains of wooden wall fittings, set on a level with the tops of the doors in Room 4 of House 1 may well have been supports for such (Hope 1988: 168). Worked-wood elements from the houses that have been tentatively identified as parts of door or window frames, could equally have been from this type of loom. Textual evidence for the use of the weighted loom is also forthcoming. A letter from Pamouris in Kellis to Pekysis in Antinoopolis, has the following request written in the margin: 'Please get 10 loom weights from Kame... please send me... the iron ring because I want to put it on the loom' (P. Kell. I Gr. 71. 48, 50-1) The warp-weighted loom had a single major beam to which the warp was attached. The warp could be prepared by weaving a narrow strip leaving a long fringe of weft loops at one edge; the woven strip was then attached to the horizontal beam (Carroll 1986: 22). Two pieces of weft prepared in such a fashion have been retrieved from Room 5 in House 3.
The Textile Industry in House 3: textual evidence
A number of documents written in both Greek and Coptic indicate that in the latter half of the 4th century, some of the occupants of House 3 ran a small tailoring business. The evidence for such an enterprise is found in both the private letters and business documents. The full extent of the operations is difficult to determine because those involved were members of a religious sect, the Manichaeans, whose brethren in the Nile Valley required the Kellis community to supply them with a variety of commodities.
The woman who conducted the business is Tehat; she is mentioned in conjunction with Hatre, her husband or son, who is also involved in the business (Gardner et al. 1999: 46). The two items of apparel that feature prominently in the documents relating to these individuals, and which are produced in this cottage industry, are cowls and headscarves; members of the religious community may well have worn the former but this is uncertain (Gardner et al. 1999: 68). An order for one of the cowls reads:'Make them weave a cowl for the double-fringed gown' (P. Kell. V. Copt. 18). The business was not restricted to such items; requests for the tailoring of unspecified garments are frequent and four types are listed by name: a sticharion, which may be a shirt or linen tunic, a kolobion, a sleeveless tunic, a cloak, and an unidentified garment called a maphorion, 'for which I have bought the cloth and cut it out' (P. Kell. V. Copt. 46.6). The business clearly made blankets, sometimes from inferior wool: 'Five minus a share for the wool of the sticharion. What is bad we have left for the blanket' (P. Kell V. Copt. 44.24-5). With the exception of the cloak, sticharion and blankets, all of which were made of wool, none of the other fibres used are specified. Wool, in fact, is the only yarn to be specified in any of the Coptic texts pertaining to Tehat's business.
Details from the accounts indicate that the business was involved in almost every aspect of garment manufacture, from purchasing the wool, spinning the yarn, making of the warp and the weft, and cutting the garment (P. Kell. V. Copt. 44); no mention is made of the final sewing. Although Tehat paid two female weavers, Heni and Kame, to produce the fabric (P. Kell V. Copt. 44.5-6), some was purchased ready woven: 'I have bought the cloth and cut it out' (P. Kell. V. Copt. 46.7-8). The cost of preparing the warp was 600 talents per mna and 400 talents per mna for the weft. This again suggests that the former was more time-consuming, perhaps because of the woven strip. Both Heni and Kame spent three days weaving (P. Kell. V. Copt. 44.5-6); their wage was paid in coin, 1313 nummi, presumably for the finished length of fabric (P. Kell. V. Copt 44.23-5). By contrast, the wage for cutting a garment was 13 maje of wheat (P. Kell. V. Copt. 46.3); this was presumably for a cloak or a sticharion as she paid only 10 maje of wheat to purchase a cowl (P. Kell. V. Copt. 46.9-12). The texts twice mention fulling (P. Kell V. Copt. 44.23-5 and 48.5), and wool was probably sent out for this process which appears to have been carried out by men. The wage for fulling was given but the document is damaged at this point (P. Kell. V. Copt. 48.15).
The small business operated by Tehat at Kellis is not the only one known from Roman Egypt to have been operated by a female. Private letters from the archive of Apollonios, a member of a wealthy land-owning Graeco-Egyptian family in the Hermopolis region, show that the family had business interests in the weaving and textile industry. It is clear from the documents that it was the women of the family who supervised the weaving (Rowlandson 1998: 118). One letter sent by Apollonios' mother to her daughter-in-law, illustrates the gender and status of the individuals involved:
With difficulty I got the material [wool] from the dryer... I am working together with your slave women and to the best of my ability. I cannot find women able to work with us, for they are all working for their own mistresses. Our people have been walking around the whole metropolis, offering higher wages (C. Pap. Jud II.442; Rowlandson 1998: 122).
The status of the women involved in Tehat's cottage industry cannot be determined, although the letters suggest that both Kame and Heni were free rather than slaves. Female slaves, however, did work as weavers in Kellis as the following document indicates: 'I acknowledge that I have given to you... for the current year NN daughter of my house-born slave NN for learning the weavers trade' (P. Kell. I GR. 19 a.8-11). Although this letter was retrieved from House 3, it is dated at least sixty years before the archive of Tehat (Worp 1997: 55-9) and consequently the contract bears no relation to Tehat's business.
Weaving in Roman Egypt was not confined to women as an archive spanning three generations of weavers, all male, found at Oxyrhynchus in the Nile Valley attests (Rowlandson 1998: 112-113). In 54 CE Tryphon purchased a loom; two years later he was teaching one of his sons the weaving trade. Twelve years later his younger son is known to have been learning his craft from another weaver in 66 CE (Papyrus Oxyrhynchus II 275; Rowlandson 1998: 112-3).
The manufacture of fine linen
Textile manufacture was a major industry in Egypt, especially linen weaving. In the Ptolemaic and Roman periods (from 323 BCE) production was strictly controlled by the State, both for quality and the quantity produced. An illustration of the extent of this control in Ptolemaic times, which continued into the Roman period, may be observed in the following excerpt that lists the duties of one of the officials, the oikonomos:
Visit the weaving houses in which the linen is woven and devote the utmost care to ensure that [as many] of the looms are in use as possible. Show particular care that the linen is of good quality and has the number of threads prescribed in the ordinance. Visit also the washing houses in which the raw flax is washed... make a list so that the castor oil and natron might be supplied for washing (Papyrus Tebtunis 703; Austin 1981: 255-6).
Whilst there is no evidence for such large-scale, controlled linen production at Kellis, the Agricultural Account Book does list a weaving workshop amongst its tenants (Bagnall 1997: line 1266); this suggests that textile production in the village was not restricted to small businesses. Monasteries may well have functioned as major centres of linen weaving. A text from an archive found in House 2, located next to House 3, reads:
As I indicated to you concerning my son... put him into the monastery, where it teaches him the linen-weaving trade' (P. Kell. I GR. 12.17-21).
The Agricultural Account Book has a reference to a monastery (Bagnall 1997: lines 320, 513) that may well have been in the vicinity of Kellis. Excavations carried out by the Supreme Council of Egyptian Antiquities on the site of Ain al-Gadida, just a few kiometres west of Kellis, was identified by Colin A. Hope as a probable monastery (Bayumi: 1998). Further evidence to corroborate the involvement of monasteries in linen production may be found at the Monastery of Epiphanius at Thebes where excavators found the remains of foot-powered looms (Winlock 1926-33: 68-9, cited in Carroll 1986: 37). Numerous papyri and ostraka found at the site also attest linen weaving in the establishment (Crum and White 1973: passim).
Fine linen was certainly produced in the Dakhleh Oasis and transported to the Nile Valley as is evidenced by the following receipt for such:
...of Hermopolis Magna, registered in the East City quarter, to Aurelius Horos, son of Mersis, camel driver from the oasis. I acknowledge that I have received from you one camel load (consisting of) - dried figs and dried grapes and fine linen (P. Kell. I GR. 51.3-6).
The receipt, dated circa 320, was found yet again in House 3. Horos, son of Mersis, is attested in other documents from the house, and whilst this does not identify Kellis as the place of production, it seems likely that it was manufactured in the vicinity.
The study of the textiles and textile industry of Kellis has just begun. In the coming years it will be widened to incorporate the material from the cemeteries. It is intended that a comprehensive catalogue will be produced, possibly on CD ROM, with an extensive commentary that will incorporate the results of the scientific analyses and a discussion of the social milieu in which the textiles were produced. There are at least 20,000 Coptic textiles in museums and private collections throughout the world with some authorities estimating the number to be closer to 100,000 (Carroll 1988: 1). The vast majority was retrieved in the early 20th century; their provenance is uncertain but they are thought to have derived from cemeteries (Carroll 1988: 1). Many of the elaborately woven decorative elements were cut from their garments for purpose of display. The corpus is poorly dated with estimates ranging between the late 3rd and early 4th centuries to the 11th or 12th century; earlier samples are rare. The Coptic textiles from Ismant el-Kharab come from a sound archaeological context that can be dated to the 3rd and 4th centuries. The publication of the catalogue should prove to be a valuable asset to museums and researchers alike.
The site of Ismant el-Kharab is within the concession granted to the Dakhleh Oasis Project (Director, Anthony J. Mills), by the Supreme Council of Antiquities, Egypt. Special thanks are due to the director of excavations at Ismant el-Kharab, Dr Colin A. Hope (Monash University), for inviting me to join his team and to work on the textiles and other artefactual remains and for making available to me all field notebooks and unpublished material. Ialso owe him a debt of thanks for his support for the textile project and his willingness to offer his expertise on the archaeology of Ismant el-Kharab. Thanks are also due to Dr Jeff Church, Principal Research Scientist, and his assistants Christine Coombs and Andrea Whitehouse of the CSIRO, Wool Technology, Geelong, Victoria, for undertaking the initial analysis of samples textile. I thank Dr Helen Whitehouse for allowing me access to the Coptic textiles in the collection of the Ashmolean Museum, Oxford, and Wanda Summers for her generous donation towards the conservation of the textiles in memory of her late mother. The textile project has been greatly facilitated by a Monash Small Grant in 2000 which will finance the cost of further analyses and which has permitted the purchase of a digital camera. The initial excavations at Ismant el-Kharab (1986-88) were funded by the Royal Ontario Museum . Since that time, work has been funded by two Large ARC Grants (1991-93 and 1995-97), a Monash University Faculty of Arts Research Initiative Grant (2000), two Small ARC Grants, a Monash Small Grant, The Egyptology Society of Victoria and Rosemary and Eric Cromby and Colin A. Hope.
Herodotus. Histories. Translated by A. Selincourt. Penguin Books. London 1954.
Pliny. Natural History. IX Books XXXIII-XXXV, Translated by H. Rackham. Loeb Classical Library. William Heinemann Ltd. Cambridge, Massachusetts. 1952.
Bagnall, R. S. 1997. The Kellis Agricultural Account Book. Oxbow Books. Oxford.
Balfour-Paul, J. 1998. Indigo. British Museum Press. London.
Barber, E. J. W. 1991. Prehistoric Textiles. Princeton University Press. Princeton.
Bayumi, K. 1998. Excavations at Ain al-Gadida in the Dakhleh Oasis. in O. E. Kaper (ed.) Life on the Fringe: Living in the Southern Egyptian Deserts during the Roman and Early-Byzantine Periods. CNWS Publications. Leiden. 55-62.
Beadnall, H. J. L. 1901. Dakhla Oasis: Its Topography and Geology Egyptian Geological Survey Report 1899 Pt. 4.1. Geological Survey of Egypt. Cairo.
Bowen, G. E. 1998. The Spread of Christianity in Egypt in Light of Recent Discoveries from Ancient Kellis. Unpublished PhDthesis. Monash University.
Bowen, G. E. 1999. Textiles from Ismant el-Kharab. in Bulletin of the Australian Centre for Egyptology 10. 7-12.
Bowen, G. E. 2000. The Small Church at Ismant el-Kharab. in Bulletin of the Australian Centre for Egyptology 11. 29-34.
Bowen, G. E. in press. Textiles, Basketry and Leathergoods. in C. A. Hope and G. E. Bowen (eds). Dakhleh Oasis Project: Preliminary Reports on the 1994/5 to 1998/9 Field Seasons. Oxbow Books. Oxford.
Bowen, G. E. in press. The Fourth-Century Churches at Ismant el-Kharab. in C. A. Hope and G. E. Bowen (eds). Dakhleh Oasis Project: Preliminary Reports on the 1994/5 to 1998/9 Field Seasons. Oxbow Books. Oxford.
Churcher, C. S. in press. Faunal Remains from Kellis. in C. A. Hope and G. E. Bowen (eds) Dakhleh Oasis Project: Preliminary Reports on the 1995-99 Field Seasons. Oxbow Books. Oxford.
Coombs, C. E., A. L. Woodhead and J. S. Church. in press. Report of the Characterization of Three Fabric Samples from Ismant el-Kharab. in C. A. Hope and G. E. Bowen (eds). Dakhleh Oasis Project: Preliminary Reports on the 1994/5 to 1998/9 Field Seasons. Oxbow Books. Oxford.
Crum, W. E. and H. G. Evelyn White. 1973. The Monastery of Epiphanius at Thebes Part II: Coptic Ostraca and Papyri. Metropolitan Museum of Art publications. New York.
Dhaliwal, H. K. 1996. Dyed Textiles and Pigment Analysis. in Berenike 1995 Preliminary Reports on the 1995 Excavations at Berenike (Egyptian Red Sea Coast) and the Survey of the Eastern Desert. S. E. Sidebotham and W. Z. Wendrich. Research School CNWS Publications. Leiden. 257-67.
Doxiadis, E. 1995. The Mysterious Fayum Portrait Faces from Ancient Egypt . Thames and Hudson. London.
Gardner, I., A. Alcock and W-P Funk. 1999. Coptic Documentary Texts from Kellis. Volume 1. Oxbow Books. Oxford.
Gazda, E. K. 1983. Karanis. Ann Arbor. Michigan.
Hickson, K. in press. Excavations in Area C at Ismant el-Kharab in 1996-7. in Dakhleh Oasis Project: Preliminary Reports on the 1995-99 Field Seasons. C. A. Hope and G. E. Bowen (eds). Oxbow Books. Oxford.
Hitchcock, M. 1991. Indonesian Textiles. British Museum Press. London.
Hope, C. A. 1985. Dakhleh Oasis Project: Report on the 1986 Excavations at Ismant el-Kharab. in The Journal of the Society for the Study of Egyptian Antiquities vol XV. 114-25.
Hope, C. A. 1988. Three Seasons of Excavation at Ismant el-Kharab in Dakhleh Oasis, Egypt. in Mediterranean Archaeology Volume 1. 160-78.
Hope, C. A. 1991 The 1991 Excavations at Ismant el-Kharab in the Dakhleh Oasis. in Bulletin of the Australian Centre for Egyptology 2. 41-50.
Hope, C. A. 1997. The Find Context of the Kellis Agricultural Account Book. in R. S. Bagnall The Kellis Agricultural Account Book (P Kell. IV Gr. 96). Oxbow Books, Oxford. 5-16.
Hope, C. A. 1998. Objects from the Temple of Tutu. in W. Clarysse, A. Schoors and H. Willems Peeters (eds) Egyptian Religion The Last Thousand Years Part II Studies Dedicated to the Memory of Jan Quaegebeur Leuven. 803-58.
Hope, C. A. 1999. Excavations at Ismant el-Kharab in 1998/9: a Brief Report. in Bulletin of the Australian Centre for Egyptology 10. 59-66.
Hope, C. A. 2000. Excavations at Ismant el-Kharab in 2000: a brief report. in Bulletin of the Australian Centre for Egyptology 11. 49-66.
Hope, C. A. 2001 Observations on the Dating of the Occupation at Ismant el-Kharab. in C. A. Marlow (ed.) 2001. The Oasis Papers I: proceedings of the First International Symposium of the Dakhleh Oasis Project. Oxbow Books. Oxford. 43-59.
Hope, C. A. in press a. Excavations in the Settlement of Ismant el-Kharab 1995-1999. in C. A. Hope and G. E. Bowen (eds.) Dakhleh Oasis Project: Preliminary Reports on the 1995-99 Field Seasons. Oxbow Books. Oxford.
Hope, C. A. in press b. The Excavations at Ismant el-Kharab and Mut el-Kharab in 2001. in Bulletin of the Australian Centre for Egyptology 12.
Hope C. A., O. E. Kaper, G. E. Bowen and S. F. Patten. 1993. Dakhleh Oasis Project: Ismant el-Kharab 1991-92. in Journal of the Society for the Study of Egyptian Antiquities
Hope, C. A. and J. Mc Kenzie. 1997. Interim Report on the West Tombs. in C. A. Hope and A. J. Mills (eds.) Dakhleh Oasis Preliminary Reports on the 1992-1993 and 1993-1994 Field Seasons. Oxbow Books. Oxford. 53-68.
Knudstad J. E. and R. A. Frey. 1999. Kellis, the Architectural Survey of the Romano Byzantine Town at Ismant el-Kharab. in C. S. Churcher and A. J. Mills (eds.) Reports from the Survey of the Dakhleh Oasis 1977-1987. Oxbow Books. Oxford. 189-2 14.
Lucas, A. 1962. Ancient Egyptian Materials and Industries, 4th edition revised J. R. Harris. Edward Arnold. London
Rijksbaron, A. and K. A. Worp. 1999. The Kellis Isocrates Codex. Oxbow Books. Oxford.
Rowlandson, J. 1998. Women & Society in Greek & Roman Egypt. Cambridge University Press. Cambridge.
Shore, A. F. 1979. Portrait Painting from Roman Egypt. British Museum Press. London.
Thanheiser, U. 1999. Plant Remains from Ismant el-Kharab: First Results. in C. A. Hope and A. J. Mills (eds.) Dakhleh Oasis Preliminary Reports on the 1992-1993 and 1993-1994 Field Seasons. Oxbow Books. Oxford. 89-93
Thanheiser, U. in press. Roman Agriculture and Gardening in Egypt as seen from Kellis. in C. A. Hope and G. E. Bowen (eds.) Dakhleh Oasis Project: Preliminary Reports on the 1995-99 Field Seasons. Oxbow Books. Oxford.
Thanheiser, U. and R. S. Bagnall. 1997. Crops. in R. S. Bagnall (ed) The Kellis Agricultural Account Book (P Kell. IV Gr. 96). Oxbow Books. Oxford. 35-46.
Vogelsang-Eastwood, G. 2000. Textiles. in P. T. Nicholson and I. Shaw (eds.) Ancient Egyptian Minerals and Technology. Cambridge University Press. Cambridge. 268-298.
Walker, S. and M. Bierbrier. 1997. Ancient Faces: mummy portraits from Roman Egypt. British Museum Press. London.
Worp, K. A. 1995. Greek Papyri from Kellis: I. Oxbow Books. Oxford.
1 All P. Kell. I. Gr. documents are published in K. A. Worp 1995; P. Kell. V Copt. documents are published by I. Gardner et al 1999.