t: +20 12 5113357 (mobile)
t: +20 27955666 (office)

The North Tombs Cemetery 2017

Anna Stevens and Gretchen R. Dabbs

A second season of fieldwork was undertaken at the non-elite cemetery at the North Tombs from 10 April to 25 May. The excavation team comprised Anna Stevens, Wendy Dolling and Melinda King Wetzel (site supervisors); and Gretchen Dabbs (bioarchaeological director), Nicholas Brown, Johanna Petkov, Sofie Schiødt, Kate Rose, Conni Lord, Kelly Accetta and Thais Rocha da Silva. Our Ministry of Antiquities site inspectors were Mr Mazher Khallifa and Mr Osama Azmi, our magazine inspector was Ms Shimaa Sobhy, and we were joined by MoA conservator Mr Mamduah Bushra Malaty, to whom we extend many thanks. For three weeks we were joined in the field by trainee inspectors Mr Remon Talaat Adly and Mr Ramy Ishak Lwez, and later by Mr Ibrahim Saleh Helal. Twelve workmen from the village of El-Hagg Qandil were employed to assist with the excavations.

This report is a preliminary summary of the work.

A report in Arabic can be downloaded here.


The North Tombs Cemetery is one of six non-elite burial grounds so far identified in the eastern bay of Amarna, lying within a broad bay that breaks the steep cliffs between North Tombs 2 and 3. The bay contains three sandy embankments, where simple pit graves were cut during the Amarna Period. There are also a few unfinished rock-cut tombs in the main bay and its side-wadis.

A first season of excavation was conduced at the North Tombs Cemetery in 2015, when 86 pit graves were excavated and 150 individuals recovered (Stevens et al 2015). The 2015 results were striking in a number of aspects, especially when contextualized against the results from previous work at the South Tombs Cemetery, tentatively identified as the main burial ground for the Main City:

  1. The individuals recovered during the 2015 season at the North Tombs Cemetery were almost all subadults and young adults, ie. between 7 and 25 years. This is the inverse of what would be expected of a typical urban population of this antiquity, of which the South Tombs Cemetery offers an example.
  2. The burials are particularly modest. Even allowing for the fact that the site has been badly robbed and subject to flash flooding, the rate of recovery of objects and pottery was extremely low. There was also no evidence for the use of wooden coffins, which occur at the South Tombs Cemetery in about 10% of graves.
  3. The number of burials containing more than one person was striking, with nearly half of the graves excavated in 2015 (c. 43%) in this category. The disturbance to the site makes it difficult to reconstruct the multiple burials, but the individuals here seem usually to be placed closely side-by-side and/or stacked on top of one another, in graves cut only slightly wider than single burials. There is little obvious sign that graves were re-opened to add interments, suggesting they were buried consecutively or almost so. Graves containing more than one individual do occur at the South Tombs Cemetery, but they are not typical of the site and mostly confined to one excavation area, the Upper Site, where they formed c. 21% of burials, usually as wide graves where the individuals seem to have been laid side-by-side.

The 2017 excavation season aimed to increase the burial sample, test for variation in demography and burial practice, and ascertain the size of the cemetery, both physically and as regards the number of dead.

Site setting and fieldwork methodology

The bay containing the North Tombs Cemetery is formed of two adjacent wadi mouths, lined by raised sandy embankments, which are separated by broad channels that have been created by flash floods. These channels were almost certainly present already in the Amarna Period. There are three main embankments, which we labeled in 2017 the North Bank, Central Bank and South Bank. The largest, by far, is the South Bank, which was the focus of excavation in 2015. All three are partly covered by open robbers’ pits and spoil mounds, often containing weathered human bone. These pits and spoil heaps date to before 1922, when they appear on an aerial photograph of the wadi. The South and Central Banks are flanked along their north and south sides respectively by a lower terrace, which is largely free of robbers’ pits.

Excavation is undertaken in 5 x 5m squares arranged in a site-wide grid. The excavation method is first to remove the bulk fill across the square, usually a crust of water-hardened orange silt, which transitions into softer gravel-rich sand. As this is removed, oblong patches of darker silt typically begin to emerge. These seem to be water-laid deposits that have settled into depressions of robbers’ pits, which were dug into underlying graves. At this stage, each pit is investigated individually.

Map of the North Tombs Cemetery, showing the areas  excavated in 2015 and 2017. Figure 1: Map of the North Tombs Cemetery, showing the areas excavated in 2015 and 2017.

In selecting study areas, modern robbers’ cuts are avoided, but excavation reveals a much earlier phase of looting, which took place so long in the past that the robbers’ pits have entirely filled again with sand and are now invisible at surface level. The robbers often rummaged around in the grave pit, leaving much of the interment still inside the grave, but sometimes threw burial materials up onto the surface of the site or into an adjacent grave. Materials left on the surface seem subsequently to have been washed away by flooding. There is no sign, within the stratigraphy, of the Amarna Period surface of the cemetery, which must have been heavily truncated during looting and probably also by flooding. The current surface of the site may approximate the level of the ancient surface, however, since grave pits often appear just below it.

In 2017, 22 grid squares were investigated, distributed across all three banks.

South Bank: Fifteen squares were investigated on the South Bank. Three of these, E7, V7 and V8, had been opened in 2015 and cleared down to the horizon at which the grave pits began to emerge. These graves were cleared this season. Additional areas of the bank were also sampled: squares N8/O8, AG6/AH6 and AM16–18, AN16–18, AO17. All contained graves, apart from AM18 and AN18. A further target area was the northern terrace of the bank, where one test square (X19) was opened. No graves were found here.

View  across the South Bank during the 2017 excavations.Figure 2: View across the South Bank during the 2017 excavations.

Central Bank: The Central Bank is almost completely covered by robbers’ pits, although these peter out along its western end. Three excavation squares were opened here (-J28, -G27, -E30). Two of these, squares -J28 and -E30, contained no graves. Square -G27 yielded one grave, which had survived unrobbed, along with a number of empty grave-like depressions. A test square, R34, was also set into the low terrace along the southern flank of the Central Bank, which again yielded no graves.

North Bank: Three contiguous squares, D58, E58 and F58, were opened towards the eastern end of the North Bank, in an area largely devoid of modern robbers’ pits. All three contained graves, and all of these remained unrobbed. The easternmost square, F58, also contained two empty grave-like depressions.

The North Bank during excavations in 2017.Figure 3: The North Bank during excavations in 2017.


A total of 55 graves were excavated in 2017, which contained at least 81 individuals. The latter have not yet been analysed, and the individual count may rise as co-mingled remains are sorted. This brings to a total 141 graves and at least 231 individuals that have been recovered from the North Tombs Cemetery during the 2015 and 2017 seasons. Field assessment of the 2017 remains as they were excavated suggests the age distribution is consistent with the individuals excavated in 2015, having high frequency of individuals between the ages of 7 and 25 years, but formal analysis is needed to confirm this.

Table 1. Graves excavated at the North Tombs Cemetery in 2017. The empty pits in squares -G27 and F58 are not included in the grave counts.

Area No. of Graves Number of individual numbers assigned Number of single burials Number of multiple burials
North Bank 9 16 4 5
Central Bank 1 2 0 1
South Bank        
- Square E7 6 12 3 3
- Squares N8, O8 8 11 5 3
- Squares V7, V8 7 8 6 1
- Squares AG6, AH6 11 16 7 4
- Squares AM16–18, AN16–18 and AO17 13 16 8 3
TOTAL 55 graves 81 individuals 33 single burials + 20 multiple burials + 2 unknown
Map of the North Tombs Cemetery, showing excavation  squares containing graves, and those where no graves were found. Figure 4: Map of the North Tombs Cemetery, showing excavation squares containing graves, and those where no graves were found.

Cemetery limits and size: The excavations provided important data regarding the size of the cemetery and allow for a preliminary estimation of its population. Several of the squares this season (and others in 2015) contained no graves, indicating that they lie beyond the cemetery limits. There was a strong correlation between the density of the modern robbers’ pits and the areas containing graves, the looters evidently homing in on areas where they found ancient burials. An important finding was that there were no graves in the two test squares on the low terraces associated with the South and Central Banks. It is possible that the cemetery did once extend here and graves have been washed away during flooding, but this seems unlikely as there is no eroding bone in the walls of the embankments, while the level of the South Bank terrace, in particular, seems high enough to contain the lower parts of graves had these once been present.

In squares where there were no graves, the sand tended to be compact and gravel-rich, which seems to have deterred the gravediggers. This was particularly evident on the Central Bank, the western spur of which seems to be composed of very ancient compact flood deposits of this kind, and avoided by the Amarna Period gravediggers. The North Bank was also compact and coarse, and while it did contain graves, these were shallow and irregular in shape, suggesting the ground was difficult to dig into.

In several areas, our excavation squares overlapped what appeared to be the original edges of the cemetery; namely, squares F58 (North Bank), -G27 (Central Bank) and AM18, AN18 and AO17 (South Bank). Interestingly, squares F58 and -G27 also contained a number of unusual pits that had the appearance of graves, but were quite short and contained no burial material at all. They are not robbed-out graves, since looters almost always leave something of the interment behind. It is possible that they are natural depressions, resulting perhaps from deep scouring by boulders during powerful floods, but they have the appearance more of man-made features in shape, depth and spacing. They are perhaps robbers’ pits into empty ground, although if so it is difficult to understand why they appear on the North Bank, where all of the adjacent excavated graves had escaped robbery. Could they be preliminary grave cuts, marking out where new graves would be placed? This seems consistent with their position only around the edges of the cemetery.

This year’s excavations suggest that the original cemetery occupied most of the three raised embankments, the exception being the western end of the Central Bank. Allowing for a few remaining uncertainties – how far east, for example, the graves extended on the South Bank – we can now fix the cemetery limits with some confidence. By extrapolating the numbers of individuals recovered from the 2015 excavations (now fully studied) across the approximate area of the cemetery, setting its eastern limit on the South Bank at the eastern edge of the robbers’ pits, where eroding bone can still be seen, a preliminary population estimate is given in the range of 3,500–5,000 people.

A shallow, undisturbed burial under excavation on the  North Bank.Figure 5: A shallow, undisturbed burial under excavation on the North Bank.

Grave layout and form: Most graves were simple oblong pits cut down into the sand, quite regular in shape with straight sides, gently rounded ends and vertical walls, the less regular pits on the North Bank already noted. The sand at the North Tombs Cemetery tends to be more compact than at the South Tombs Cemetery, and the graves hold their shape well during excavation. In terms of size, graves tend to fall in the range of c. 0.4–0.5m in width, 1–1.5m in length and 0.9–1.4m in depth. Two exceptions this year were pits <17263> and <17256> in squares AM16–18, AN16–18 and AO17 on the South Bank which are much larger, measuring c. 2 x 2.1m and 2 x 1.8m, and having a depth of c. 1.7m and 1.4m, respectively. These are the first large pits of this kind found at the North Tombs Cemetery; a small number were found during excavations at the South Tombs Cemetery, in the sample area notable for its larger numbers of multiple burials (the Upper Site).

In several squares this season the graves were packed closely together with very little space between them; those in squares N8/O8 and AG6/AH6 were particularly dense. The graves in squares AM16–18, AN16–18 and AO17, in contrast, were noticeably more spread out, perhaps due to their location on the edge of the cemetery. Even in areas where graves were very closely packed, however, none cuts into another, and we can assume that they were visible at surface level. Some may have been marked by simple stone cairns, as seems to have been the case often at the South Tombs Cemetery. Boulders and stones that might once have formed parts of cairns occur somewhat less frequently in the sandy fill at the North Tombs Cemetery, although some may have washed away during flooding. The two very large graves <17263> and <17256> yielded pieces of mud-brick that were perhaps parts of superstructure/s; a few pieces of mud-brick were also found during the 2015 season in grave <16753> in square X8. It is also conceivable that the graves were topped with low mounds of sand, or were visible simply as slight depressions as the grave fill settled into the pit. As in 2015, no evidence of funerary stelae was found.

A triple burial under excavation on the North Bank,  containing three individuals (Inds 1167, 1168 and 1169), who have been wrapped  together in a single burial mat before interment.Figure 6: A triple burial under excavation on the North Bank, containing three individuals (Inds 1167, 1168 and 1169), who have been wrapped together in a single burial mat before interment.

In terms of alignment, the graves tend to run in perpendicular directions; they are particularly orderly in squares V7/V8. The widely spaced graves in squares AM16–18, AN16–18 and AO17 might show that graves were initially quite spread out and the ground was then gradually filled in with new pits. The influence of local landscape and social patterning on the layout of the graves requires further investigation, but the care with which the graves are cut and spaced, and the possibility that pits were pre-cut initially around the cemetery edges, suggest that the site had one or more dedicated grave-diggers, as can also be suggested for the South Tombs Cemetery.

Treatment of the body: The bodies were wrapped in textile and in a mat before being interred, both of which were usually in a poor state of preservation. The matting remains to be studied, but seems often to be of reed. As in 2015, the thicker and more rigid gereed and tamarisk-stick matting that was common at the South Tombs Cemetery was rarely encountered, and there were again no traces of wooden coffins.

While many of the individuals were burial singly, the trend for large numbers of multiple burials continued this season. Around 36% of the 2017 graves contained more than one person, a figure that may change slightly as co-mingled burial groups are analysed. As in 2015, double burials were the most common, but graves containing, three, four and five individuals were also encountered.  Again, the multiple graves were usually cut to the width of a single burial, or just slightly broader, the bodies placed closely side-by-side and/or stacked in layers. This season yielded the first clear evidence, in undisturbed burials, that bodies were sometimes wrapped together in the same roll of matting before interment (Inds 1211/1212/1213, Inds 1173/1192/1218, Inds 1166/1167/1168, Inds 1171/1172); in at least one grave, they even seem to have been wrapped together in textile first (Inds 1174/1175). Here, there is no doubt that the bodies were buried at the same time. A slightly different style of multiple burial is suggested by the two very large graves <17256> and <17263>. In <17256>, the skeletons of three people were laid out side by side in the base of the grave, with a fourth individual probably placed on top. This approach is more reminiscent of the large multiple burials at the Upper Site at the South Tombs Cemetery; their significance at both sites is not yet understood. The second large grave <17263> was entirely empty of human remains, which were presumably removed by looters, who left a large quantity of what is assumed to be burial linen behind.

Burial goods and offerings: As in 2015, very few artefacts were encountered during the excavations. In the undisturbed burial of Ind. 1160 on the North Bank, a necklace (obj. 41418) of glass and faience beads, and one with gold-leaf finish, was found around the neck, and a steatite scarab with the name of Amenhotep II (object 41419) rested against the left wrist. This was the only one of the 11 intact burials encountered this season to contain grave goods, indicating how rare these were. Part of a disc bead (obj. 41421) was found in the upper fill of the large grave <17263> in squares AM17/18 and a group of three cylinder beads and a wedjat-eye bead recovered from the disturbed grave of Inds 1178 and 1181 in square N8. There was also very little pottery overall, although the two large graves, <17256> and <17263>, yielded several pieces of large vessels. These remain to be studied and it is not yet clear if these were connected with the burials as offerings, or were perhaps used by people who were working at the cemetery. A few re-used sherds were also found, possibly used for digging.

Ind. 1160, wearing a necklace of glass and faience beads  (obj. 41419).Figure 7: Ind. 1160, wearing a necklace of glass and faience beads (obj. 41419).

Skeletal analysis of individuals excavated in 2015
Following the excavation season, a short season of skeletal analysis was conducted, which allowed for the study of the remaining 40 individuals left unstudied from the 2015 excavations (for previous work, see Stevens et al 2015; Dabbs and Rose 2016). There was no time to undertake analysis of the individuals recovered in 2017 and this work will begin in Autumn 2017.

A major theme at the North Tombs Cemetery continues to be one of youth, with most individuals aged between 7 and 25 years at the time of death. No infants (0–2.9 years) were encountered during the analytical work. There was one individual aged as an early subadult (3.0–6.9 years) (2.5%). Thirteen individuals were classified as late subadults (7.0–14.9 years) (32.5%). Twenty-six individuals were classified as adults (over 15.0 years) (65.0%), but the vast majority of these were classified as young adults, with 24 individuals being aged between 15.0–24.9 years of age at death (60.0% of the analyzed sample). 

Sixteen of the 26 adult individuals (61.5%) were estimated to be female. Six individuals (23.1%) were identified as male and the sex of the other four individuals (15.4%) could not be estimated from the remains available for analysis. This disparity between males and females present in the skeletal sample is not as extreme as, but is consistent with, the findings from 2015. Initial assessments after the first season of analysis suggested this female bias in the sample may be due to the difficulties distinguishing young males from the skeletal sample.  However, if the young males were simply not being identified, the frequency of “indeterminate” individuals would be higher, accounting for those indistinguishable young males along with the individuals for whom it is truly impossible to estimate sex based on the available materials.  That the dearth of males is a consistent feature of all areas of the cemetery excavated in 2015, suggests that the population buried in the North Tombs Cemetery was largely female.

The same types of trauma and disease markers observed in previous seasons of the North Tombs Cemetery analysis were again noted during this season. Schmorl’s nodes and fractures of the spine remain common occurrences, with 10.3% and 17.9% presence in individuals, respectively. Trauma to the arms was observed in two individuals (5.4%). Evidence of nutritional insufficiency is demonstrated by the high percentage of the population with cribra orbitalia (42.1%). A high percentage of the sample (85.7%) exhibits a combination of lesions (cribra orbitalia or humeral cribra or femoral cribra and spinal porosity or periostitis of the lower limbs) that have been previously identified as indicators of malarial infection (see Smith-Guzman 2015). This frequency is consistent with findings during the 2015 skeletal analyses. 

The adults within the sample analyzed this year continue to maintain the overall short stature previously observed in both the North Tombs Cemetery and the South Tombs Cemetery. While the sample sizes are small, the average adult female in the 2017 analyzed North Tombs Cemetery sample stood 153.7cm, while males were slightly taller at 161.0cm. These figures are consistent with the previously analyzed North Tombs Cemetery individuals. Importantly, the North Tombs Cemetery samples are consistent with the short stature observed in the South Tombs Cemetery sample previously, where females stood 154.3cm tall and males at 163.6cm (all stature estimates based on sex-specific formulae produced by Raxter et al 2008).

The findings of the North Tombs Cemetery skeletal analysis season, which completes the analysis of the 2015 excavation season, suggests the skeletal remains from the South Bank are homogeneous in their general characteristics, which include young age at death, workload injuries, possible malarial infections, and overall short stature. With the exception of the short stature, which is seemingly a universal Amarna characteristic, the North Tombs Cemetery continues to stand distinct from the findings of the South Tombs Cemetery, supporting the suggestion that this cemetery was used by a different subset of the overall Amarna population.


The 2017 excavation and study season confirms the main results of the 2015 work. The North Tombs Cemetery is a burial ground primarily for young people, who were buried modestly in textile and basic matting, usually without any burial goods. There are large numbers of burials containing more than one person, and clear evidence that some of these individuals were interred at the same time. There is little obvious sign that people were buried with family, or at least not in complete family groups, given the limited age range of the deceased.

The larger excavation area, and discovery of a number of unrobbed graves, provides a more nuanced view of the site, and a reminder that it was not of uniform character. Ind. 1160, for example, buried with a necklace and scarab, had also been laid out quite carefully for burial, wrapped in not one but two layers of plant-fibre matting, suggesting a certain level of ‘care for the dead’. The two large graves on the South Bank, <17263> and <17256>, also show variation in treatment of the dead, although they are difficult to understand. It may be significant that they lie quite close to an unfinished rock-cut tomb at the end of the wadi: are they family graves for people connected with the owner of this tomb? Or are they simply larger-than-usual mass graves?

Overall, though, it is clear that the North Tombs Cemetery is a very unusual burial ground. It is tempting to see those interred here as the victims of an epidemic, but this is not a very satisfactory explanation, at least not on its own: if this were the case, we would expect to see a broader distribution of individuals across the age spectrum. Our working hypothesis is that these are burials of a workforce conscripted on the basis of their youth and subject to working/living conditions that saw their immune responses compromised, resulting in large numbers of deaths. Preliminary skeletal analysis supports this, with evidence of workload and traumatic injury at rates consistent with a population undertaking arduous work, although more analysis is needed to fine-tune these results. Many of the multiple burials might be understood as those of individuals who died at roughly the same time and were buried expediently in the same grave.

In trying to pinpoint what work they may have been doing, it may be noteworthy that the main limestone quarries of Akhetaten are also located to the north of the site. There is a range of tasks in addition to quarry-related work that would have been necessary to establish and maintain the new city, however, including the production of mud brick, carting water and supplies, and agricultural work, all of which would have absorbed a large labour force.

We remain in the early stages of studying this dataset and understanding its implications, but its potential for elucidating the lives of peripheral communities at Amarna is clear.


Dabbs, G. R. and J. C. Rose, 2016. ‘Report on the October 2015 skeletal analysis of the North Tombs Cemetery Project’ in B. Kemp, ‘Tell el-Amarna, 2016’, Journal of Egyptian Archaeology 102, 7–11.

Raxter, M. H., C.B. Ruff, A. Azab, M. Erfan, M. Soliman, A. El-Sawaf, 2008. ‘Stature estimation in ancient Egyptians: A new technique based on anatomical reconstruction of stature’ American Journal of Physical Anthropology 136(2), 147–55.

Smith-Guzman, N. E., 2015, ‘The skeletal manifestation of malaria: An epidemiological approach using documented skeletal collections’, American Journal of Physical Anthropology 158(4), 624–35.

Stevens, A., G. R. Dabbs, M. Shepperson and M. King Wetzel, 2015. ‘The cemeteries of Amarna’ in B. Kemp, ‘Tell el-Amarna, 2014–15’, Journal of Egyptian Archaeology 101, 17–27.





Website first posted September 2000; last updated October 2017 | enquiries concerning website: email bjk2@cam.ac.uk