Mummy 1911-210-1

An ancient Egyptian mummy from the collections of the National Museums of Scotland was examined using computerised tomography (CT) scanning as part of the NMS mummy project. A facial reconstruction was produced from the CT scans for comparison with a painted ‘portrait’ which covers the face of the wrappings. The scans indicated the mummified wrapped body of an adult male 1.65m tall with excellent preservation of the body. An exact replica, translucent model of the skull was created from the CT scans and standard cephalometric analysis undertaken. The facial features were reconstructed onto a plaster model of the skull from known tissue depths using terracotta clay. The resultant face was compared with the portrait and an extremely close match obtained, suggesting this was an individual portrait painted around the time of death.

Keywords: mummy, portrait, computerised tomography (CT) scanning, facial reconstruction, cephalometric analysis

There can be few areas of archaeology which raise greater interest amongst the general public than Ancient Egypt, particularly Egyptian mummies. Although most of archaeology deals with artefacts left by our ancient ancestors, mummies bring us into direct contact with the people of the past. They are surrounded by myth and magic and are often buried with magnificent treasures, although unfortunately most of these have been plundered. Those that remain, as in the burial of Tutankhamun, only enhance the mystery of the mummies and illustrate the capabilities of this ancient civilisation.

Although much is known about the general history of ancient Egypt and about the royal families, often from inscriptions on various tombs and other megalithic monuments, little is known about the lesser individuals. For some individuals, the name of the deceased, their occupation and references to their family may be found inscribed on the sarcophagus. In many cases, however, even this information is lacking and only the remarkable relics remain. Modern scientific techniques can be used to discover something of the lives and identities of these people. This approach has been applied to the collection of mummies in the National Museums of Scotland to assess their preservation. Non-invasive investigative techniques have been applied to obtain as much information about the collection as possible non-destructively. The investigations included subjecting the collection to computed tomography and plain film evaluation, examination of the wrappings, cartonnages, associated artefacts, (such as jewellery), and artefacts not deliberately included (such as insect remains) and, for those mummies already unwrapped, examination of the remains from an anthropological and medical view point. This article reports the findings for one of these mummies, referred to simply as acquisition number 1911-210-1 (Figure 1).

Mummification is the preservation of the body and has a long history in Egypt, as such preservation was considered essential for the survival of the deceased in the afterlife. The first mummification was purely due to burial in the desert sand and natural dehydration, but from about 2600 BC the bodies were artificially prepared to ensure preservation. Mummification continued until the 4th century AD, when the traditional practices were forbidden by Emperor Theodosius in AD 392 following the adoption of Christianity as the official religion of the Roman Empire.

Mummification techniques developed over the millennia, and essentially involved the desiccation of the corpse with natron, a naturally occurring salt found in areas such as Wadi Natrun, which consists mainly of sodium carbonate and sodium bicarbonate, (with some sodium chloride and sodium sulphate). The lungs, intestines, liver and stomach were removed but the heart was retained, as this was essential for a successful afterlife. Brain tissue was removed via the nose by perforating the cribriform plates. The body was then wrapped in sheets of linen and the viscera preserved separately and buried along with the corpse.

In the later periods of Egyptian history, mummification was often inferior to that in earlier periods. Large quantities of resin were used rather than careful embalming methods and more care was applied to the external appearance than to the preservation of the corpse. Mummies were encased in painted shrouds, in decorated stucco cases, or in elaborate linen wrappings with complex rhombic patterns. A variety of headpieces were used, including cartonnage masks (often gilded), plaster heads and encaustic or tempura pictures of an individual’s face. Although some mummies were still enclosed in the traditional coffins, many were buried without coffins and other elements of traditional mummification, such as the canopic jars, shabti figures and the Book of the Dead, were also abandoned.

During the late periods of Egyptian history, Egypt was ruled first by a Graeco-Macedonian regime, after the conquest of Alexander the Great in 332 BC, and then by the Roman Empire, after 31 BC. The Greeks settled in Egypt after Alexander’s conquest and established a Greek-speaking elite society of businessmen and officials, who lived in their own communities, such as El-Fayum. The Romans were assimilated into this society after 31 BC and both groups of foreign settlers gradually adopted the Egyptian religion and burial practices, including mummification. The traditional idealised facial representations used on the Egyptain cartonnage masks were, however, too unrealistic for contemporary Hellenistic styles, so from the middle of the 1st Century AD were replaced with more realistic depictions based on Hellenistic artistic tradition. These pictures were first found in the Fayum area and , as they appear to be life-like representations of individuals, are termed Fayum portraits, with the best examples from Hawara.

The finest portraits could probably only be afforded by the wealthiest families of Greek or Roman origin. The majority of the paintings show only the head and shoulders and the different hairstyles, clothing and jewellery depicted are used to date the portraits as these followed the well-known and frequently changing Imperial styles. As the portraits generally depict fairly young people, it has been suggested that they were commissioned during the lifetime of the individual and hung in their home as decoration. The rectangular panels would then be cut for use as funerary portraits, when required, with the final shape dependent on the preference of the local embalmer, (for example, arched portraits are characteristic of Hawara). This idea is supported by the fact that portraits are popular in the Roman world and were often set up in public places or placed in tombs. It is also possible, however, that the portraits may have been painted from the corpse, as there are some of middle-aged and elderly people. The panel was carried in the procession of the deceased through the village or town and taken with the deceased to the embalmer for mummification, where it cut to fit the mummy. Hence, the generally young nature of those depicted could simply reflect the low life expectancy. In some cases this is supported by a correspondence of sex and age between the portraits and the individuals. The portraits persisted for 200 years and their style was retained in Christian art and eventually developed into the Byzantine icons.

Sir Flinders Petrie excavated this mummy, Petrie 3, in 1911 at Hawara in the El-Fayum area of Egypt, along with two other portrait mummies Petrie 2 (now in New York) and Petrie 4 (now in Seabury theological seminary), one gilt bust and one plain wrapped mummy. The upper part of mummy 1911-210-1 is slightly discoloured but both mummy and portrait are in good condition. There is a high quality Fayum portrait over the face of the mummy, showing the head and shoulders of a young man wearing a white mantle (Figure 2). This is in encaustic wax on very thin wood (probably lime, although it has not been tested), with the grain running vertically. The hairstyle and beard suggest the picture dates from the Flavian period at the end of the 1st to the beginning of the 2nd century AD (c.80-110 AD). The painting is held in place by a series of 12 bandages, which form an octagonal frame around the arched panel, characteristic of those from Hawara. The outer wrappings were arranged in the style characteristic of this time and form an elaborate rhomboid pattern at the front, with gilded stucco studs on each rhomboid and on a band across the chest. The feet are encased in cartonnage and the toenails gilded. Examination of this mummy is discussed below and was undertaken as part of a conservation evaluation of the entire mummy collection held by the National Museums of Scotland.

The unique feature of mummy 1911-210-1 is, as indicated above, the presence of a beautifully executed facial portrait. It was intended to compare the individual with that depicted to assess the degree of accuracy of this portrait and to determine as much as possible about the individual.

The entire collection of mummies was examined at the Edinburgh Western General Hospital, using the Siemens Somatom Plus Spiral Computerised Tomography (CT) Scanner. The technical data used for Mummy 1911-210-1 is shown in Table 1. Reconstructed images were transferred to an independent Sun workstation for segmentation and three-dimensional reconstruction and were saved on an optical disc for future analysis. Preliminary three-dimensional reconstructions using surface-shaded rendering were made of the segmented skull data prior to stereolithography.

Scanner	Siemens Somatom Plus spiral scanner
Subject position	Supine, head first
Acquisition	Spiral scanning for the head: conventional single-section scanning for the remainder of the body
mA	210 (spiral); 420 (conventional)
kV	120
Slice thickness	3 mm (spiral); 5 mm or 10 mm (conventional)
Table speed (spiral)	3 mm/s
Reconstruction	Spiral scans reconstructed at 1 mm increments
Total exposure	93.090 mAs
Recon algorithm	Siemes 75575 (spiral); 7555 (conventional)
Number of images	433

There were extensive concentric wrappings around the entire mummy. In the region of the head several interesting features were noted within the wrappings:

The skull was held in a slightly flexed position, with the head turned to the left. The brain was removed. The bones of the cranium appeared normal, with no evidence of any trepanation, although the skull was thin in places, probably due to de-mineralisation. Medium density material was noted in the dependent parts of the cranial cavity, presumably embalming artefacts.

There was reasonable preservation of the soft tissues over the face. The orbital contents were present and there appeared to be residual optic nerves and extra occular muscle. Anteriorly, in the orbital cavities, were plaque-like structures which, although they do not resemble the usual spherical configuration of the globes, may be native globes which have undergone postmortem-post embalming atrophy. The para-nasal sinuses were unremarkable. There was good delineation of the inner ear structures and teeth, both within the maxilla and mandible. No obvious dental disease was visible. A further metallic object was present within the mouth, posteriorly and to the right of the midline. This appeared flat on the saggital views, and was elliptical in the other plane.

There was good preservation of the cervical, thoracic and lumbar spine. The upper cervical spinal canal contained embalming residue which had spilled over from the skull base. More distally the remnants of the spinal theca were clearly visible. No significant abnormality was encountered in these structures. Other bony structures were well preserved and in good condition, with good cortico medullary differentiation. The upper limbs had a normal appearance and were positioned against the body with the palms turned upwards. There was good preservation of the soft tissue structures.

The thoracic, abdominal and pelvic cavities were eviscerated. The thoracic cavity was empty superiorly, although the lower chest/upper abdominal cavity contained some packing material posteriorly, with a more solid object, about 14cms in length on the right at the approximate level of the diaphragm. This feature was visible on the scout view. The packing extended down into the pelvic cavity, which was otherwise empty. The genitalia were disappointingly indistinct, although there was evidence of male genitalia, and the shape of the sub-pubic notch tended to confirm the male sex.

The lower limbs appeared unremarkable. There was good preservation of bony and soft tissues and no evidence of any significant derangement of pathology. The individual muscle groups in the thighs were particularly well shown. Harris lines were not definitively identified on the digital planning scans, indicating the absence of any disorder that may have caused the arrest of growth during development. The feet appeared unremarkable and it was possible to discern the gilded toes, which had been positioned over the anterior wrappings at the level of the feet. A metallic object was seen immediately posterior to the left thigh, some 20 cm above the knee joint. Multiple metallic studs were present over the anterior aspects of the wrappings. Further smaller metallic objects were seen at various points over the posterior wrappings, some of which appeared to represent pins.

It was difficult to establish the precise age at death of this adult mummy from the skeletal characteristics. However, the skeleton was that of a fully grown adult, measuring 1.65 m in height. Ossification centres were fully fused, in particular the centres of the medial ends of the clavicles vertebral bodies, pelvis and hips. These features, along with the general assessment of the development of the teeth, indicate an age of death at no younger than 25 years. There were no clues from the scan as to the cause of death.

In conclusion, the CT scan indicated the well preserved mummified wrapped body of an adult male. Numerous metallic artefacts were included in the wrappings, with an elliptical metallic disc-shaped object in the mouth, overlying the tongue (Figure 3).

Figure 3: Reconstructed computerised tomography data showing the shape of the metal object which lay over the tongue (exteriorised)

Rapid prototyping is a relatively new process, whereby three-dimensional models are built layer by layer directly from computer data. This technique has already found multiple applications in industry and design. It is now being used increasingly for prototyping of body parts, primarily as an adjunct to complex surgery, but also in the field of archaeology and palaeo-pathology. ‘Medical’ models are built in layers, based on the data from high definition spiral CT scans. Although there are several different techniques for producing models, the method used in this case is known as Stereo Lithography (SLG).

Initially, scans were taken of the entire skull as indicated in Table 1, using a 3 mm-slice thickness with a table speed of 3 mm per second. Following the scan, slices were reconstructed at 1 mm-slice increments, giving a total of approximately 230 overlapping slices available for prototyping. The reconstructed slices were transferred to a computer workstation (Sun sparc station 20) for segmentation and preparation for data transfer. Segmentation means preparation of the CT slices so that only the information needed, in this case bones of the skull, are used to make the model. To do this, any pixels less than a given density threshold are removed from the CT slice. In a living subject, it is usually fairly easy to remove soft tissues in this way, leaving the dense bone which then can be made into a model. In the case of the mummy, the bone was relatively demineralised, and therefore nearer in density to the soft tissue and wrappings. Hence, it was not easy to separate bone automatically and a degree of manual editing of unwanted features from each slice was necessary.

Once the data set consisted only of the bony features that were needed for the model slices, they were transferred to a commercial prototyping bureau (Materialise, Leuven, Belgium) for model production. The SLG process involves a high-powered laser travelling across the surface of the liquid medium, usually acrylic or epoxy resin. This is solidified in the path of the laser which moves according to the information presented to the controlling computer. In the case of a skull model, the surface solidifies in the shape of the segmented CT slices presented to the machine, and the full skull is built up slice by slice. The resulting life size model is translucent.

An interesting feature of this mummy was the metallic object in the mouth. It was possible to identify this object separately from the skull during the segmentation process because of its high density. As the prototyping bureau were experimenting with a new form of resin which was able to change colour following the additional pass of the laser, the metallic object in the mouth was made red whilst the rest of the skull was transparent. This allowed it be easily viewed and identified and it was concluded that it was a metallic tongue plate rather than a coin (Figure 4).

The production of an exact replica, translucent model of the skull, allowed for the standard digitised cephalometric analysis to be undertaken (Figure 5). This indicated a Class II Division I incisal relationship with increase in overjet and reduction of overbite. The upper incisors were proclined beyond the norm, the lowers retroclined. In addition, there was lower incisor imbrication. Skeletally, the base SNA/SNB difference, ANB was reduced at 0.7degrees, showing a mild underlying Class III tendency. This was supported by an increase in the lower facial height ratio 58.7% (norm 55%) giving an overall increase in the facial length of 135 mm (norm 129 mm) and in the height of the ramus at 50 mm (norm value 44 mm). In addition, the mandibular body length was increased at 84.3 mm (norm 70 mm).

From a frontal view utilising the Grummon’s analysis (Figure6), there was an asymmetry in the cranio-facial complex, with skewing of the mandible to the right and a reduction in the measurement RFL-RJP (0.4 mm below norm) and an increase in the measurement LFL-LJP (4.6 mm above norm). This was mirrored in the mouth by a crossbite tendency on the right hand side of the occlusion.

From the above analysis a picture presented itself of an individual with a long narrow face - the narrowness particularly focused on the intranasal - intramaxillary regions. This led in turn to the suggestion that there had been a tendency to a downward growth pattern, a result of either hereditary and/or of a functional disturbance in childhood. A possible disturbance would be a chronic inability to breathe through the nose. The reasons for this inability are numerous, varying from skeletal anomalies from bony tissue intranasally, adenoidal hypertrophy, to a simple year round nasal allergy. However, this scenario usually manifests itself intra-orally as a bilateral crossbite, which is not seen in the present case. Long term habitual digit sucking, however, while contributing somewhat less to the gross skeletal changes, often does result in a unilateral crossbite with mandibular displacement, as found here. In addition, the measured proclination of the maxillary incisors and retroclination of the lowers, together with the imbrication of the lower incisors, further strengthens the argument for a long-term digit sucking scenario being involved.

The concept of reconstructing facial features back onto the skeletal framework is not new. Modern methods used predominantly in forensic identification are based on the work largely of 19th century anatomists such as His, who demonstrated the similarity of soft tissue depth measurements over similar points of the skull between individuals of the same age, sex and race. Using this knowledge and normal dimensions and positions of the eyes and nose it is possible to reproduce facial characteristics using the skull as a scaffold and knowledge of applied anatomy. The technique usually involves copying the skull using a standard impression material such as alginate to produce a plaster cast, into which approximately 21 pins are driven at specified points to the known average tissue depths. Terracotta clay is overlaid onto the skull, mimicking the muscles of the face, building up to the skin surface, using the final depth (Figure 7). The resultant face is intended to be a likeness of the individual rather than an exact reproduction but many cases have shown how comparable the results are, both in a forensic and archaeological context.

Figure 7: Building the facial anatomy to the final skin depth as indicated by the wooden pins

The difficulty with archaeological material is that often no contemporary portrait or busts exist with which to compare the resultant face. Consequently, the Fayum portrait in Mummy 1911-210-1 allows us to assess the value both of facial reconstruction in an ancient context and the potential accuracy of the very life-like Fayum portraits.

The facial reconstruction was handed to a facial artist, who was able to recreate a similar hairstyle and beard as shown on the Fayum portrait (Figure 8). The reproduction was then compared using overlaid 1:1 photographs with the portrait so that a direct comparison could be made (Figure 9). The resultant images showed a remarkable likeness, indicating that the portrait was probably painted of the individual at or near the time of death.

Figure 9: An overlay of the portrait (textured areas) onto the terracotta model (smooth areas)

In the case of Mummy 1911-210-1, CT scanning confirmed that the individual was a young adult male, consistent with the portrait. The facial reconstruction showed a striking comparison to the portrait, which suggests that this was painted for this man, either at his death or very shortly before. The scans also showed that the mummification was of high quality with excellent preservation of the body, in contrast to that often assumed for Roman mummies. It is interesting to consider that he may have been a digit sucker when younger, leaving the classic tell tale changes to his facial growth and teeth. This may be the earliest documented case, although digit sucking is probably as old as man! Apart from that, little further can be ascertained about this man and certainly there were no clues as to why he lived such a short life. The prevalence of disease in those parts is the most likely explanation; few would have left any marks visible radiographically and none were found in this study. In the future, micro invasive techniques, such as DNA sampling, may give us further insights into the remarkable world of the mummies.

The authors would like to thank all those involved in this work for their help. The work was undertaken as part of the National Museums of Scotland Mummy Project, which involves Dr Jim Tate, Dr Elizabeth Goring, Lesley-Ann Liddiard and Brian Melville. The CT scanning was initially facilitated thanks to help from Dr Malcolm Merrick. For this specific project, the authors would particularly like to thank Martin Connell for his expertise in computer reconstructions, Douglas Hunter and Barry Wohlgemuth for the computerised overlay techniques, Brian Hill for producing the terracotta facial reconstruction and Helen Allison for secretarial help and typing. The funding for the scanning and reconstruction was provided by the Charitable Trust of the National Museums of Scotland and by the Russell Trust.

For further information on the NMS Mummy Project see the NMS web-site at http://www.nms.ac.uk

Correspondence: K Eremin, National Museum of Scotland, Edinburgh, UK.
Email: ke@nms.ac.uk

HISTORICAL REVIEW

Mummy 1911-210-1