New Light on Ancient Egyptian Prosthetic Medicine
Nicholas Reeves
Studies in Egyptian Antiquities. A Tribute to T.G.H. James (British Museum, 1999), pp. 73-77


(p. 73) In 1881, the British Museum (1) acquired by purchase from one of its principal `scouts`, the Rev. Greville Chester (Dawson, Uphill and Bierbrier 1995, 96-7), a large group of miscellaneous antiquities collected during the course of his previous round of travels in Egypt. One item in the group (EA 29996) was entered in the Egyptian accessions register (81.6.14, 77) as a `Leather artificial toe for the right foot, nail wanting, from a mummy. 5 in. From Thebes` (Fig. 1, Colour Pl. XVII,1-2). This description is expanded by Samuel Birch in his manuscript slip catalogue of the BM Egyptian collection under the old (and now superseded) running number 7029a:

"Object modelled in shape of the great toe and portion of the right foot but no other toes, the nail which has been of some other material has been taken out. It is said to have come from a mummy probably from one the toe of which was wanting. At the left side of the toe are 8 small circular holes to sew it on or attach it to some part of a mummy. It is of a light brown colour. 4 1/2 in[ches] l[ong]. Leather. Thebes"

Chester’s curious find, as perhaps both he and Birch sensed, was far more than a simple cosmetic restoration applied by the embalmer to make the body whole for the hereafter (2); it is, as we shall consider, an artificial limb employed during life. As such, it represents one of the earliest working prostheses to have been identified from the ancient world.

That such a rare and important object should have escaped previous published notice serves to demonstrate the extraordinary potential of the British Museum’s rich Egyptian holdings - holdings which the recipient of this volume, T.G.H. James, with characteristic generosity, made freely available to all interested parties during the course of a long and distinguished curatorial career. This brief note is offered to Mr James with warm and appreciative thanks - for first introducing the writer to the British Museum collection, and for the rare privilege of studying and working with it over several happy years as a junior curator under his charge.

Physical Description

As an examination under good light establishes, the material of the Greville Chester toe is not leather, as Birch had indicated, but cartonnage - a composite (p. 74) material made from fabric (linen) impregnated by a binding and setting medium (animal glue and gesso). The overall appearance is similar to that of a relatively modern artificial limb, the outer surface, together with a small part of the inner, having a smooth, tan-coloured coating (3) which obscures the texture of the underlying fabric. This coating is generally well-preserved, but on the outer top surface of the toe and at various points including the exposed upper right edge, there are noticeable signs of wear (Fig. 1, Colour P1. XVII, 1-2). On the inner surface, the cartonnage is covered by a much thinner coating of a distinctly red hue (4); this coating too is rather worn, and the underlying texture of the cartonnage base is again exposed. As Birch noted in his `slip` description, the toe appears originally to have been fitted with a separate toenail, most probably (given the depth of the recess) in an artificial material such as faience; this inlay is now lost, revealing the ancient adhesive below (5).

The design of the prosthesis appears to have been cleverly thought out. The longer free edge, pierced by a row of eight holes and distorted as a result of use, follows closely the line of a typical Y-thong sandal strap and, when in place, would have been effectively concealed by it (6). Along the shorter free edge may be discerned four further holes, subsequently filled (7) at the time of the last refurbishment (for funerary use) and now visible on the inner surface only (8). Both sets of holes were evidently employed for attachment - by a system of lacing around the foot itself, to a flesh-coloured sock, or by stitching directly to the previously mentioned Y-strap sandal (9) with which the prosthesis would have been worn and, indeed (in the absence of the original toe), which it would have served to hold in place. It may be speculated that the manner of attachment imparted a limited but crucial degree of flexibility while walking - permitting the upward motion of the artificial toe as it came into contact with the ground, while at the same time ensuring that the prosthesis would revert to its original position as this ground pressure was released.


Details of the precise circumstances under which the toe was first encountered are sadly lacking, but the claim that it was found in position on a mummy seems to be confirmed by the final, non-functional (p. 74) and presumably funerary refurbishment which obliterated the attachment holes required for daily use. The results of the scientific examination of the piece offer no contradiction to the stated Theban provenance (10).


Given its unique character and the absence of detailed archaeological context, the precise dating of this object remains problematic. Its discovery in position on a mummy seemingly confirms that it was made before Islamic times, while style and the quality of manufacture are perhaps reminiscent more of dynastic than of Graeco-Roman work. Technological considerations confirm this impression. Hero Granger-Taylor, who was kind enough to examine those areas of the textile core now visible, writes as follows:

"A fine plain ‘tabby weave’ linen cloth can be seen which has two notable characteristics. Firstly, the textile is `warp-faced`, that is, with warp threads predominating and lying over the weft threads (11). This is a feature that was typical of Egyptian linen throughout the dynastic, Hellenistic, Roman and Byzantine periods. Secondly, the threads of both warp and weft at intervals have a (p. 75) cord-like appearance, where the yarn, otherwise `single`, can for a short distance be defined in terms of structure as `plied` or `doubled` (12). Yarns of this type are found in all extant Egyptian linen textiles of the dynastic period up until about 600 BC, when they are superseded by more familiar `draft spun` yarns. (Granger-Taylor 1997) (13).

Unless, therefore, the British Museum prosthesis had been manufactured employing much earlier linen, the probability, on present evidence, is that it was produced no later than the first half of the first millennium BC; it could, conceivably, be earlier than this ceiling date by several centuries.


The principal aim of this short note is to draw the British Museum prosthesis to the attention of a wider audience, and to present the basic documentation relating to it; more specialist discussions of the piece will doubtless follow, but a few preliminary thoughts might and perhaps ought to be ventured.

Evidently dating to a period before c. 600 BC, the Greville Chester toe represents rare and early evidence for the use of artificial limbs (14), supplementing and predating the rather more extensive documentation available from Greek and Roman times (Bliquez 1983; Bliquez 1996; Jackson 1988, 68). Significantly, like so much else in ancient medical science (cf. Weeks 1980, 106-8 for the dental evidence), limb prosthetics can now be shown to trace their origins back to Egypt.
Despite its antiquity, the British Museum toe is a sophisticated production, beautifully designed, skilfully and strongly made and clearly a special order; it is reasonable to assume that the owner - male, to judge from the size, and if flesh-tone conventions were here observed - was a person of considerable rank and influence, to whom perfection in physical appearance was not only important but practically and financially achievable.

Distinct signs of wear and evidence of subsequent refurbishment indicate that the appendage had actually been worn in life before the toe was finally consigned, with its owner, to the grave. There can be little doubt, however, that the toe`s primary role was cosmetic, in line with its naturalistic modelling and colour. Despite the fact that 40% of the walking weight is normally borne by the big toe, with the loss of this member the weight tends to be transferred as a matter of course to the end of the first metatarsal; the toe`s absence incurs no residual disability beyond a limp when hurrying or attempting to run (Andrews 1988, 1245) - a minor inconvenience which the carefully considered manner of attachment may, nonetheless, have sought to rectify.

Finally, any number of suggestions might be ventured to explain how and why the natural toe which this prosthesis was intended to replace came to be lost. Trauma (as a result of accident or war) represents perhaps the most likely cause, followed by gangrene and ulceration from diabetes or arteriosclerosis, particularly if the person were middle-aged or beyond; leprosy, too, is a possibility. Less likely instigators of the condition are frostbite, ergotism (if present in ancient Egypt) or ainhum: while all of these go for the toes, it is chiefly the little toes - with ainhum taking them both off.

(1) Thanks are due to Prof. John R. Harris, Dr Ralph Jackson, Dr Walter Y. Loebl, Dr John Nunn and Dr John H. Taylor for their comments, suggestions, references and practical assistance. For their illuminating scientific reports, extracts from which make up much of the present text, I am particularly grateful to Hero Granger-Taylor, Janet Lang, and Dr Andrew Middleton. For the excellent and informative drawing, I am indebted to Annie Searight.

(2) Embalmers’ restorations of this sort, generally dating from the Ptolemaic and Roman periods, are well known; notable instances are the restored forearm on the mummy currently on loan to the Oriental Museum, Durham, from Darlington Museum (without number), and the fake penis and feet on a mummy in the Manchester Museum (no. 1770). See variously Gray 1966; Loffler 1984, 4-5; Tapp 1979; Tapp 1986.

(3) X-ray diffraction (XRD) analysis of this coating suggests that it is composed of a mixture which includes calcite, dolomite, halite and quartz. In the scanning electron microscope (SEM) (equipped with an energy-dispersive x-ray analyser - EDXA) calcite was observed as a finely granular phase, while dolomite occurred as rhombic crystals. Quartz occurred as rare, isolated grains. Halite (common salt, which is easily soluble) was not observed, probably because it had been dissolved during sample preparation. Grains of iron oxide (typically <10 µm diameter) were observed, scattered through the sample. A fine-grained, magnesian clay was also observed to be present as an interstitial phase. The assemblage of minerals observed suggests that this material was essentially crushed, dolomitic limestone, probably with a small proportion of ochre added to give the tan colour. It was probably mixed with an organic binder and applied as a gesso layer, although this hypothesis was not tested analytically, The presence of dolomite suggests that the material was not a lime plaster, as dolomite would have been destroyed during calcination and, unlike calcite, would not be expected to reform (Middleton 1992).

(4) XRD analysis of this material indicated only the presence of calcite, which may have been derived from the substrate. However, analysis by Fourier-transform infrared spectroscopy (FTIR) suggested that it is essentially a gum resin, possibly Dragon’s blond gum. Prnteinaceous material was also detected, but ,t is thought that this was probably derived from the underlying cartonnage. Dragon’s blood gum - a (p. 76) dark red exudation from a palm was widely used in the ancient world as a red varnish; it is also mentioned by Pliny. Interestingly, perhaps, in the present context, it has also been held in high esteem for its medicinal qualities (Middleton 1992).

(5) XRD analysis of a sample of this material suggested that it is mainly organic in nature. This was confirmed by FTIR analysis which showed that itis an animal glue (Middleton 1992).

(6) No definitive study of Egyptian footwear has yet been published, but cf. Brovarski, Freed and Doll 1982, 177-8.

(7) This fill was identified by XRD as consisting mainly of calcite. The nature of the diffraction pattern suggested that the calcite is relatively coarse-grained and unlikely to be a lime plaster. The possibility that it includes an organic binder was not investigated (Middleton 1992).

(8) Radiography of the prosthesis revealed little constructional detail, but did indicate an area of increased density at the toe. It is uncertain, however, whether this is to be interpreted as an indication of fill within the toe-cavity, or merely a reflection of the increased thickness of the artifact at this point. The Newton Victor radiography set provided too high a potential (? 52 kV) to be able to radiograph successfully the thinner parts of the section (Lang 1989).

(9) A suggestion I owe to Robin Cooper, communicated by Dr R. Redhead of the Disablement Services Centre, Queen Mary’s University Hospital, London (Redhead 1990).

(10) The morphology of the dolomite present in the tan-coloured surface and the presence of a magnesian clay are typical of some limestones from the Theban area (Middleton 1992).

(11) For weaving and basic spinning terms see Burnham 1980.

(12) As Hero Granger-Taylor points out, in practice the plied sections of ancient Egyptian linen yarns mark where two bundles of linen fibres have been spliced end to end with two new bundles, and the splices then twisted back on each other to form the plying (cf. most recently Cooke, el-Gamal and Brennan 1991) - a technique still used in parts of the Far East for the production of yarns from hemp and ramie fibre.

(13) The dating of these characteristics is based upon Granger-Taylor’s own study of Egyptian textiles, particularly in the British Museum. As she notes, because of a comparative shortage of independently dated textiles from around the time of the changeover in spinning techniques, that is, from c. 600 BC, it is impossible for the moment to date this change more precisely, or to estimate the length of time over which the two techniques coincided. Granger-Taylor’s report continues:
"... The different layers of the textile visible on the various areas of the toe are of similar quality, with approximately the same number of warp and weft threads per centimetre. They therefore may represent different areas of a single piece of cloth. But it is just as likely that the cartonnage was made up from two or more similar but different cloths".

The technical data may be summarized as follows: Undecorated textile in fine warp-faced simple tabby weave, visible where the surface of the cartonnage is worn away, on the inside of the prosthesis and around the edges of the lacing holes. Warp: undyed linen, zzS spliced-plied and S-twisted, c. 40 ends per cm. Weft: undyed linen, zzS spliced-plied and S-twisted, c. 16 threads per cm.

(14) The intriguing possibility exists that a second artificial toe, noted in place on a mummy in the Albany Institute of History and Art, Albany, NY (x1940.58.11), represents an even earlier, articulated prosthesis. The mummy, that of a woman, reached the United States in the base (the lid is said to have been discarded in Cairo in 1909) of a coffin belonging to the wab-priest and sankh of the temple of Mut, Ankhefenmut; for other elements of the latter’s burial cf. Niwinski 1988, 177, no. 416. The few details I have of this mummy and coffin are due to John H. Taylor and Tammis Groft, Chief Curator of the Albany Institute of History and Art. A preliminary report on the discovery of this second artificial toe appeared in the local Albany medical newspaper (Anonymous, 1989):

"X-rays taken of the female mummy showed an artificial great toe, according to William Wagle, MD, director of neuroradiology and magnetic resonance imaging at the Medical Center. `On the plain X-rays we saw an unusual artifact in the great toe of her right foot and the CT scan clearly showed it to be an artificial toe replacing the phalanges (the two distal bones at the tip of the toe)`, Dr Wagle pointed out. `This most unusual prosthesis was constructed in two parts, the proximal (closest to the foot) part was made of a high density material having the same density as bone, perhaps some form of ceramic. This was neatly aligned with the distal end of the metatarsal bone (end of the foot)`, he added. `The second part had a much lower density, fit exactly into the first part and its distal end had the form of a great toe. The prosthesis appeared well crafted and exactly aligned`. The X-rays also confirmed her sex, showing a female pelvis and bone structure, he noted. In addition, the X-rays showed that her spine was quite curved, most likely from the mummification process itself. The researchers estimated that the woman died in her 40s or SOs, but they were unable to determine how she died. There was no evidence of major trauma to the skeleton ..."

Further information on this important find is eagerly awaited.


Andrews, B.G. 1988. Amputations, in B. Helal and D. Wilson (eds), The Foot. Edinburgh, 1245-52.
Anonymous 1989. X-rays and CT Scans help Center staff to uncover secrets about Egyptian mummies, Center News (Albany, NY), January, 1 and 6.
Bliquez, L.J. 1983. Classical prosthetics, Archaeology, September/October, 25-9.
Bliquez, L.J. 1996. Prosthetics in classical antiquity: Greek, Etruscan, and Roman prosthetics, in W Haase and H. Temporini, Aufstieg und Niedergang der Roemischen Welt, II. Principat, 37.3. Berlin and New York, 2642-74.
Brovarski, E., Freed RE. and Doll, 5K. 1982. Egypt’s Golden Age. The Art of Living in the New Kingdom. Museum of Fine Arts, Boston.
Burnham, D.K. 1980. Warp and Weft: A Textile Terminology. Royal Ontario Museum, Toronto.
Cooke, WD., Mohamed el-Gamal and Brennan, A. 1991. The hand-spinning of ultrafine yarns, part 2: the spinning of flax, Bulletin de Liaison du Centre International d’Etudes des Textiles Anciens (Lyon) 69, 17-23.
Dawson, WR., Uphill, ER and Bierbrier, M.L. 1995. Who Was Who in Egyptology. 3rd edn, revised. London.
Granger-Taylor, H. 1997. Pers. comm. 4 November. Granger-Taylor, H. 1998. Pers. comm. 1 April.
(p. 77) Gray, P.H.K. 1966. Embalmers’ restorations, Journal of Egyptian Archaeology 52, 138-40.
Jackson, R. 1988, Doctors and Diseases in the Roman Empire. London.
Lang,J. 1989. Internal British Museum Department of Scientific Research report no. 5905, dated 31 October.
Loeffler, L. 1984. Der Ersatz fuer die obere Extremitaet. Stuttgart.
Middleton, A. 1992. Internal British Museum Department of Scientific Research report no. 6042, dated 26 June.
Niwinski, A. 1988. 21st Dynasty Coffins from Thebes. Chronological and Typological Studies. Mainz.
Redhead, R. 1990. Pers. comm. 18 September.
Tapp, E. 1979. The unwrapping of a mummy, in A.R. David (ed.), Manchester Museum Mummy Project. Multidisciplinary Research on Ancient Egyptian Mummified Remains. Manchester, 83-93.
Tapp, E. 1986. The unwrapping of 1770, in A.R. David (ed.) Science in Egyptology. Manchester, 51-6.
Weeks, KR. 1980. Ancient Egyptian Dentistry in J.E. Harris and E.F. Wente, An X-Ray Atlas of the Royal Mummmies. Chicago and London, 99-121.


Shortly after this article had been submitted for publication, it came to my notice that a further prosthetic toe, naturalistically carved in wood and very similar in form to British Museum EA 29996, had recently been discovered in position by a mission of the Deutsches Archaeologisches Institut on an intrusive mummy recovered from Theban tomb TT 95 at Sheikh Abd el-Qurna by Andrea Gnirs of the Aegyptologisches Institut of Heidelberg University. I am indebted to Vivian Davies and Renée Friedman for drawing my attention to this remarkable and timely find, and to Dr Daniel Polz for initial information on it; and I am particularly grateful to Prof. Dr Andreas Nerlich of the Pathologisches Institut of Ludwig Maximilians Universitaet, Munich (who is preparing the TT 95 prosthesis for publication), for most generously providing further details and a photograph. As Prof. Nerlich’s photograph clearly shows, the TT 95 artificial toe had been sewn on to a short linen sock for use in life; initial indications are that `the [TT 95] individual suffered from severe arteriosclerosis and ... that chronic ischemia may have been the cause for amputation` (Andreas Nerlich, pers. comm. 22 June 1998).

Fig. 1 Cartonnage toe-prosthesis for the right foot, EA 29996. Length 11.9 cm
Pl. XVII, 1 Front view
Pl. XVII, 2 Back view