South Tombs Cemetery 2008

Wendy Dolling

Contents

1 Introduction

1.1 Background

1.2 Excavation aims

1.3 Site description

1.4 Methodology
2 The archaeological findings

2.1 Southern excavation area — Grid Squares G51–L51

2.1.1 Grid Square G51

2.1.2 Grid Square H51

2.1.3 Grid Square I51

2.1.4 Grid Square I50

2.1.5 Grid Square J51

2.1.6 Grid Square K51

2.1.7 Grid Square L51

2.2 Northern excavation area — Grid Squares F52–M52

2.2.1 Grid Squares G52/F52

2.2.2 Grid Square H52

2.2.3 Grid Square I52

2.2.4 Grid Square J52

2.2.5 Grid Square K52

2.2.6 Grid Square L52

2.2.7 Grid Square M52

2.3 New excavation areas

2.3.1 Grid Squares L53–M53

2.3.2 North-west sector of the wadi

3 Discussion of archaeological findings

3.1 Burial density

3.2 Burial orientation and spatial patterning

3.3 Preparation of the body for burial

3.4 Burial/funerary goods

3.5 Artefacts (Barry Kemp, Anna Stevens)

3.6 Concluding remarks (Barry Kemp)

3.7 Human bones from the South Tombs Cemetery (Melissa Zabecki)

4.0 Publications cited

Appendix 1 (PDF 580k):
Register of individuals excavated during 2008, and list of units according to grid square.

Appendix 2 (PDF 1.18Mb):
List of Units according to Grid Square.

Notes to units when citing excavations:
The following conventions have been used when citing excavation unit numbers:
(000) for fill units
<000> for cuts
[000] for structures and underlines for surfaces.

Acknowledgements

The personnel of the excavation and associated research were as follows:

Archaeological Site Supervisors
Barry Kemp
Wendy Dolling

Field Archaeologist
Anna Stevens

Excavation Team
Hosni Osman Mehanni
Yahya Sadek
Bakr Amin
Abuzeid Azedin Abuzeid
Abdel-Hafiz Abdel-Aziz
Saleh Osman
Ahmed Mokhtar Mahmud
Mahdi Ahmed Abdel-Nazir
Hillal Mohammed Omar
Mohammed Abdel-Alim Hussein
Gamal Abdel-Halim Hassan
Ayman Shahata Fahmy
Mostafa Rabia Fathy
Abdel-Aziz Abu Aleaqa

SCA Inspector
Mr Ali Mustafa Mohammed el-Bakri

Physical Anthropology
Prof. Jerome Rose
Melissa Zabecki

Objects Registrar
Dr Anna Stevens

Pottery Analysis
Dr Pamela Rose

Artefact Conservation
Julie Dawson

1 Introduction

1.1 Background

The Amarna South Tombs cemetery is located on the sloping banks of a large wadi that cuts down through a low plateau to the south-west of the main city site. The mouth of this wadi exits on to the plain in the vicinity of a group of rock-cut tombs belonging to nobles of the Amarna Period, the South Tombs. Excavation at the site commenced in 2006; following on from the discovery of surface skeletal remains during a survey in 2005. In 2006 a 5m x 35m area was excavated on the eastern slope of the wadi, and a series of disturbed burials were identified. It was clear from investigations during this season that the cemetery continued to the north beyond the excavated area. Time constraints meant the investigators were not able to determine if the base of the archaeological deposits had been reached in all grid squares. In 2007 another 5m x 35m strip was excavated directly to the north of the 2006 area. During this second season an Amarna Period surface was exposed in a portion of the excavation area with a series of burial pits cut into this surface. Several of these burials were excavated during the 2007 season though a significant number of probable burials remained unexamined. It also became apparent that the base of the archaeological deposits had not been reached in much of the 2006 excavation area. At the completion of the 2007 season the excavation area was backfilled to protect the unexcavated burials.

1.2 Excavation aims

As has been previously noted (Dolling 2007) the South Tombs Cemetery is of particular significance within the archaeological record of Amarna due to its potential to illustrate the burial practices of a previously unrepresented group, the general populace or non-elite of the city of Akhetaten. Prior to commencement of investigations at the site archaeological evidence for New Kingdom burial practices at Amarna was restricted to that of the elite or noble residents. The methods of interment, burial material and any cult or ritual practice associated with the burials of the general populace of Akhetaten were unknown.

In 2008 investigations at the South Tombs cemetery focused on completing the clearance of the 2006 and 2007 areas of excavation. The strategy employed aimed to expose the Amarna Period surface in the entire area; locating and investigating any burials exposed by this process. Continuing the main objectives of previous seasons, in 2008 we aimed to determine the nature of the burial practices evident at the site, establish the base level of archaeological deposits and recover skeletal material for osteological analysis. Body orientation and containment, utilisation of grave goods, patterns of cemetery disturbance and the temporal relationship of disturbance and cemetery utilisation are all questions that ongoing excavation at the site aims to answer. An understanding of the physical layout of the cemetery and how it may have functioned as a place set within the broader landscape and social activity of the city is a longer term objective of the investigations. Analysis of this skeletal material is ongoing providing evidence for population demographics, paleopathology, age of death and physical characteristics of individuals buried at the cemetery. For findings from the analysis of skeletal material collected prior to 2008, see Zabecki (2007) Rose (2006a and b).

1.3 Site description

The excavated portion of the site is located on the eastern slope of the large wadi at the more southerly end of the cemetery. Here the east bank of the wadi initially drops relatively sharply downwards from the plateau, but at a distance of approximately 40–60 m from the current wadi floor the degree of slope flattens out and a relatively gentle slope continues downwards. Approximately 2 m from the floor of the wadi proper the surface again drops sharply to base level. As such the embankment provides a large open area, which would have served as a suitable space for a cemetery of significant size. The question of to what degree the current topography varies from that of the Amarna Period is not easily answered. Excavation undertaken to this stage does seem to point to a series of depositional and erosion phases. While the evidence is inconclusive, deposition seems to be the greater factor at work so that preserved Amarna Period archaeological deposits are buried beneath accumulated sand and gravel. The depth of this deposition varies significantly over the site. Excavation in a portion of grid square F52 during the 2008 season revealed the western limit of the wadi bank at the time when the cemeteries was in use. This illustrated that the current limit of the eastern embankment varies only slightly from that of the Amarna Period, with a deposit of loose fine-grained sand accumulated over the sloping end of the wadi bank. Based on survey findings, topography of the landscape and excavation results it is likely that the cemetery extends over much of this eastern side of the wadi. There is also a smaller area on the western embankment which is likely to be part of the cemetery. The western embankment has been identified as a potential site for future excavation. (For survey results, see Kemp 2003, 2005.)

During the 2008 season an area of 420 square metres was excavated, consisting of a series of contiguous 5m x 5m grid squares. The main focus of the excavation was on areas of the site that had been partially excavated in previous seasons, including grid squares G51–M51, investigated in 2006, and squares G52–M52, where excavation was undertaken in 2007. In addition, small-scale clearance was undertaken in grid squares I50, F52, L53 and M53, all of them areas previously not investigated.

1.4 Methodology

Site supervisors were responsible for the recording of all archaeological findings from their specific areas of excavation. Kemp supervised squares G51–L51 and L53–M53; Dolling supervised squares F52–M52. The resultant report is based on field records prepared by both.

The site was excavated manually and all spoil was sieved prior to being discarded. Each recognized deposit and or feature was given a distinct unit number using a sequential numbering system that applies to the entire site of Amarna. The South Tombs Cemetery is designated as Grid 14 within the Amarna site numbering system. The excavation grid is aligned to a site-north established during the 2006 season rather than true magnetic north. General discussion of location in this report relates to site-north; the exception is the orientation of burials which are described in relation to magnetic north (for a discussion of burial orientation, see Kemp 2006).

In order to maintain consistency with methods utilised in previous seasons bone groups were assigned an Individual Number (no.) when at least 50% of a complete skeleton could be determined to be articulated or clearly associated at the time of excavation. If a group of articulated and associated disarticulated bone was excavated from an original burial context, that is from within a discernable burial pit, then less then 50% of an individual may still be identified by an Individual no. The sequence of numbering continued from that used in 2007, identified individuals in 2008 commencing with Individual no. 40.

At completion of the 2007 season it became apparent that in some instances more than one person was buried within a single pit. It was also apparent that disarticulated bone from within a burial pit may have originated from adjacent burials and been scattered during disturbance of the site. With these variables in mind, a discussion with the physical anthropology team was undertaken regarding at what point skeletal material from potential multiple burials should be assigned a specific individual number. It was decided that if articulated or in situ skeletal elements from more then one individual were discernable at the time of excavation, then each individual would be given a distinct individual number. However if, as was the case in several of the investigated burial pits, disarticulated bone from within a single burial pit could only be definitively assigned to more then one individual during post-excavation analysis, then the skeletons would be numbered using an individual number and a suffixed alphabetical designation. For example, three individuals were identified during post excavation analysis of disarticulated bone recovered from burial pit <12174> and were designated as Individuals no. 62a, 62b & 62c. By contrast, articulated and in situ portions of two individuals were identified during the actual excavation of burial pit <11694>; that is, the archaeological evidence clearly indicated that these two individuals had originally been interred in this pit. These two people were designated as Individuals no. 41 and no. 42. For the purposes of unit numbering, all recovered skeletal material was considered to be an inclusion within a fill unit so that the unit number assigned to bones, including identified individuals, relates to the fill from which the bones were recovered.

In addition to the numbering of skeletons, a register of skulls was maintained to aid in determining the minimum number of individuals represented at the site. Skull numbering in 2008 commenced at 73. Those skulls that were clearly associated with a specific individual were labelled with both a skull and an Individual no. At the commencement of the 2007 excavation season analysis of 2006 skeletal material was incomplete so that the exact number of skulls represented in the assemblage was uncertain. Following consultation with Prof. Jerry Rose the number 40 was chosen as a starting point for the numbering of skulls recovered during 2007. The sequence of numbers continued in 2008 directly from the 2007 numbering system. As a result, within the total skull numbering system so far used at the sites there is a break in the skull numbering sequence so that no skulls have been numbered from 24–39. This makes a total of 71 distinct skulls identified over three seasons by the excavation team

Disarticulated bone that was clustered together but not contained within a burial pit was not assigned an individual number at the time of excavation. These bone clusters were given a distinct unit number to locate them in relation to any surrounding fill and burials. During post-excavation analysis some of these clusters were found to contain bone of an identifiable skeletal age and sex but not always from a single person. The physical anthropology team designated these skeletal groups as cluster individuals according to their specific unit number when appropriate, but they were not assigned a distinct Individual no. as, in many cases, only a relatively small percentage of the total skeleton was recovered. A number of the bones recovered from such clusters have been able to be associated with the burial of a formerly identified individual. All discussion of post-excavation analysis within this report is based on information provided to the author by Jerome Rose and Melissa Zabecki.

In determining the degree of weathering of any skeletal material a visual inspection was undertaken by the archaeologist at the time of excavation. Bone described within this report as ‘weathered’ is bone that appeared completely desiccated and bleached of colour from exposure to sun and/or wind. In addition, some bone had deteriorated further to become extremely brittle with the bone losing its structural integrity. Bones that had not undergone significant exposure retained a degree of moisture visible as a brownish colour of the bone and were less brittle. The term ‘bone’ as used in this report relates to human skeletal material unless otherwise specified. It is possible that some fragments of bone that were identified as human at the time of excavation may on analysis prove to be from other mammalian species.

The term in situ is used in this report to describe all archaeological deposits that could be determined to have been deliberately placed within or in association with the graves during the burial or funerary phase. It excludes all material that may or may not have been deposited in antiquity but was clearly the result of another phase of activity such as robbing or destruction at the site.

2 The archaeological findings

The findings are here discussed according to identified archaeological units. As much of this season’s excavation took place in areas partially cleared in previous years a brief discussion of findings from these earlier investigations is appropriate as it necessarily contributes to the interpretations of the 2008 archaeological findings. Results are discussed according to independent grid squares followed by a general discussion of the site as a whole.

2.1 Southern excavation area — Grid Squares G51–L51

This area was first investigated in 2006 (Ambridge and Shepperson 2006). It is the southernmost of the excavation areas and also includes a small extension into grid square I50, a previously unexcavated portion of the site. In 2006 a series of burials were identified the bulk of which were located at the eastern end of the excavation area in grid squares L51 and M51. At the western end of the area (grid square G51) a small brick tomb was also revealed. None of the burials excavated in 2006 were intact; in fact, there was a significant degree of disturbance evidenced across the entire area. At the completion of the 2006 season the excavators raised the possibility that the base of the archaeological material may not have been reached. Following excavation in the adjacent area in 2007 (grid squares G52–M52) it became apparent that this was in fact the case in at least a portion of the 2006 excavation area (Dolling 2007).

Prior to commencement of excavation in 2008 an assessment was made of the potential for unexcavated archaeological material having been preserved in the 2006 area. This assessment considered the depth of excavation reached during 2006 and the excavation findings in adjacent grid squares in 2007. The presence of excavated in situ burials in grid square M51 and correlation with the base of deposits in M52 suggested that there was little likelihood of additional material being excavated from M51. Of the remaining grid squares it was determined that there was a high probability that grid squares H51–K51 would contain unexcavated material with a slight probability of unexcavated material being preserved in L51 and G51. As a result a decision was made not to undertake further excavation in M51 but to continue investigations in the remaining grid squares, initially focussing on squares H51–K51 and then, as time permitted, squares G51 and L51.

Excavation commenced with removal of sandy fill that had accumulated across the excavation area since the completion of the 2006 excavation. This fill included wind-blown sand and a degree of baulk collapse along the southern limit of the excavation area. When possible, the transition point between this accumulated fill and the underlying unexcavated deposits prompted a change of unit number. In practice, however, this was not always possible as the sandy fill level at which excavation ceased in 2006 was difficult to distinguish from the accumulated sand level. As a result, the uppermost fill units have a high probability of comprising a mixture of accumulated sand and unexcavated fill. Ongoing clearance across the entire strip of grid squares exposed a moderately compacted sandy surface into which a series of burial pits had been cut. This surface was clearly at a greater depth of excavation than that reached in 2006 and was contiguous with the cemetery period surface exposed in the adjacent 2007 excavation area. The overlying fill units and excavation of the revealed burials is discussed in detail according to grid square.

2.1.1 Grid Square G51

The uppermost fill covered the entire grid square and consisted of a combination of accumulated sand post-dating the 2006 excavation and backfill covering the brick tomb in the south-west corner of the square (12109). There was also a portion of the original unexcavated surface that was removed as the transition point was unclear. Following excavation of this unit, a natural sandy surface was exposed 12255 contiguous with the surface in to which burial pits were visible in the remainder of the excavation area. In the south-west corner of the square a cut <12256> interrupted this surface. This is an excavation cut made in 2006 to expose the brick tomb located in the south-west quadrant of the grid square. In 2008 some scattered bone could be seen protruding from the fill within this cut. It is unclear where this bone originated from but there had been a degree of disturbance at the site between the 2006 and 2007 excavation seasons, and the bone could have been displaced from adjacent areas at that time. It is also at least partially the result of ongoing baulk collapse along the southern edge of the grid square. As such, a decision was made to re-clear the area of backfill and any unexcavated bone. A series of loose sandy fill layers were excavated from the space surrounding the brick tomb: units (12139), (12163), (12166) and (12176), and a minor amount of bone was recovered.

In the north-east quadrant of the square a shallow cut or pit <12257> was exposed cut into the compacted surface 12255. This cut was rounded at the southern end but the northern end was poorly defined and may have been truncated. It is a shallow depression only 5–8 cm in depth with slightly inward sloping sides. It was filled by a continuation of the loose sandy fill that covered the entire grid square (12109). This cut does have the appearance of being a remnant of a grave pit but no in situ skeletal or burial material was found here so that it may not relate to an original burial. No further excavation was carried out in this grid square as, with the exception of the depression <12257>, there were no indications of unexcavated burials (Figure 55).

2.1.2 Grid Square H51

Clearance in this grid square commenced with the removal of accumulated sand (12073). There was a also a significant degree of baulk collapse along the southern side of the square that was cleared as part of this fill; material originating to the south of the excavation area proper is therefore likely to be included. At a point where there was a discernible increase in the degree of compaction of the sandy fill it was designated as a new unit (12082); this was felt to potentially mark the transition point to the unexcavated depth in the grid square. The excavation of this fill (12082) exposed a ridge of moderately consolidated sand (12091) running along the northern margin of the square, and patches of consolidated sand (12099) in the south east-and east of the grid square. Both these fill units were found to overlie a natural sandy surface 12134 into which at least one burial pit <12216> had been cut. Given the degree of compaction and lack of inclusions in units (12091) and (12099) it is possible that they are the upper horizon of the Amarna Period surface 12134.

At the junction of grid squares H51 and G51 a small cluster of bones (12140) was found surrounded by the fill unit (12091); only a portion of it excavated due to time constraints. In the north-east corner excavation of unit (12091) revealed a second larger cluster of disarticulated bone overlying remnants of plant-stem matting and a consolidated patch of mounded sand (12107). This deposit was seen to extend into the adjacent square H52 so that a decision was made to excavate this area as part of the same unit. Though not immediately apparent during the excavation of unit (12107), ongoing clearance in grid square I52 revealed a poorly preserved burial pit the western end of which is highly likely to have been represented by unit (12107); see square I51, fill (12160) and pit <12181> for additional details.

Pit <12216>

A distinct rectangular pit <12216> crossing the junction of grid squares H51 and I51 was revealed by the excavation of units (12082) and (12099). This pit cuts down into the surface designated as unit 12134. Two layers of loose sandy fill, (12215) and (12254), were cleared from this pit. There was a small amount of botanical material and a small fragment of painted gypsum plaster contained in the fill but no skeletal material or definite indications of a burial. The pit itself is clearly cut in a roughly rectangular shape with relatively vertical pit walls and flat base. It appears to be purpose-made rather than an intrusive cut and has a high probability of having been used or intended for use as a burial pit (Figure 1).

2.1.3 Grid Square I51

The entire grid square was initially cleared of a loosely compacted sand deposit (12044) that included wind-blown sand that had accumulated since 2006 and a portion of previously unexcavated fill. At the completion of excavation of this unit several burial pits at a depth of excavation not reached during 2006 were exposed cut into a natural sandy surface 12258 (Figure 55). In the north-east corner of the grid square a slightly mounded area of compacted sand was also exposed (12106) directly overlying the natural sandy surface.

Burial of Individual no. 69a & b — pit <12132>

In the north-east quadrant a large burial pit <12132> was found that extended slightly into grid square I52. The uppermost fill (12133) comprised loose sand with scattered fragments of desiccated bone and a small amount of botanical material possibly plant-stem fragments. Several successive layers of sandy fill were removed (12202), (12203) and (12212). Floating within this fill were a series of large limestone boulders and a minor amount of disarticulated and fragmented bone. There was also a cluster of bar shaped fragments of gypsum, possibly parts of a window grill (12207) (object 38816; Figure 2).

At this point in the excavation fragments of a decorated timber coffin also became apparent floating loose in the pit fill. Sandy layers containing decomposing organic matter, primarily termite-affected timber, were progressively excavated. Separate areas of the pit were designated with individual units in the hope of aiding in later reconstruction of the fragmentary coffin. These units include (12214), (12217), (12219), (12222), (12225) and (12227). Within fill unit (12219) the wooden face from this anthropoid coffin was found floating in loose sand (Figure 3). Though several cracks were present in the wood the face was well enough preserved to enable lifting of the artefact intact. On-site advice and assistance regarding lifting the fragmentary coffin was provided by the project’s conservator, Julie Dawson. The coffin clearly was anthropoid in form and though poorly preserved could be seen to have been decorated with hieroglyphic inscriptions that were painted in cream colour on a dark background. The coffin fragments and wooden face carry the object registration no. 38819.

Concentrated at the northern end of the burial pit was a cluster of disarticulated and partially articulated bone including an intact skull, no. 88. Though disarticulated the majority of the bones appeared to belong to one individual, and as they were clearly contained in the burial pit were designated as Individual no. 69a. The leg bones (femur, tibia and fibula) had been displaced but still remained partially articulated to each other presumably indicating that the body was not completely skeletonised when the burial was disturbed (Figure 4).

During post-excavation analysis of all bone from the pit a second individual was identified and designated as Individual no. 69b. It is probably that Individual no. 69a was originally interred in the coffin as the majority of this individual’s bones were recovered from the base of the pit partially overlying the coffin base. The bones of the second individual were scattered throughout the pit at various levels and cannot securely be associated with the coffin. This is supported by the fact that Individual no. 69a is an adult while Individual no. 69b is a child of approximately 5 years of age. The coffin appears, at least originally, to have been made for an adult though adaptation for a younger individual or multiple burials within the coffin cannot be excluded given the degree of disturbance evident in the pit. Only a small portion of the coffin remained intact in its original burial position: this was the curving head-end of the coffin that was decorated with white and black bands. In the space between this end of the coffin and the western pit wall an intact ceramic vessel [12266] was found contained in a gravel and sand fill (12265) (Figure 4). This bowl had a slight crack in the rim but was otherwise intact and contained botanical material, as yet unidentified but possibly grains or seed pods. The bowl appears to be located in its original burial position and as such would have been part of the burial equipment associated with the burial of Individual no. 69a.

Burial of Individual no. 62a, b & c — pit <12174>

Located in the south-east corner of the grid square and crossing in to J51 a second burial pit <12174> was found again cut into the natural sandy surface 12258 (Figure 5). Several loose sandy fill units were excavated from this burial pit. Unit (12175), the uppermost fill, contained a minor amount of desiccated bone and a small amount of fragmentary botanical material; there was also a minor amount of fragmentary gypsum plaster. Lower levels of fill (12185), (12195) and (12198) contained increasing amounts of scattered skeletal material though none of the bones were undisturbed and most exhibited a significant degree of weathering. There were also fragments of plant-stem or stick matting, several large ceramic sherds and a possible mud jar stopper scattered through the fill. Two skulls were recovered floating loose in these layers: no. 93 and no. 95. At completion of clearance a large fragment of plant-stem matting was revealed resting up against the pit wall in the south-west corner.

During excavation it was apparent that more then one individual was represented here as both infant and adult-size bones were identified on site. However, as none of the bone was undisturbed and could not be immediately associated with a specific body identification of individuals was based on post-excavation analysis by the physical anthropology team. Three distinct individuals were represented in the disarticulated bone: Individual nos. 62a, 62b and 62c. Of particular interest during ongoing analysis was that the skeletal remains designated as Individual no. 62a were in fact part of a previously identified Individual no. 7. Individual no. 7 was excavated in 2006 from square I50/I51 located to the south-west of pit <12174>. Though the skeletal evidence is less conclusive, Individual no. 62b has a high probability of being part of another skeleton, Individual no. 11. This burial was also excavated in 2006. In 2006 the partially preserved bodies of these individuals were excavated from what appeared to be their original burial position in a double grave lying side by side. Portions of the lower limbs of the two individuals were found in an articulated state wrapped in plant-stem matting, though the remainder of the skeletons had been disturbed. The process that resulted in the disturbed bones of these individuals being deposited in pit <12174> is uncertain. It is possible that the bones were incidentally deposited here during the disturbance phase at the site; if this is the case the pit must have been open at that time. The absence of any in situ skeletal remains within burial cut <12174> raises the possibility that this pit was never used for an original burial and itself may be the result of intrusive digging at the site. There is also the possibility that it represents a deliberate reburial of disturbed bone, given the well-formed nature of the cut (Figure 55).

2.1.4 Grid Square I50

Clearance of the overlying fill (12044) from the south-west corner of grid square I51 had exposed the head-end of two potentially in situ burials. These burials appeared to continue beyond the excavation area into square I50 and, as a result, a decision was made to extend the area of excavation into this square. A 3 m x 3 m square was opened in the north-west corner and loosely consolidated surface sand was removed (12131) followed by several lower layers of loose sand with variable amounts of limestone gravel, units (12136) and (12144). These lower fill levels contained minor amounts of desiccated bone and botanical material. A slightly more compacted sand and gravel layer (12150) was encountered at a depth of circa 80 cm below surface level. This thin deposit overlaid a moderately-compacted sand and gravel surface 12155 and two distinct burial pits cut into this surface. These pits were directly adjacent to each other and aligned approximately east–west.

Burial of Individual no. 57 — pit <12159>

The northernmost of these two burials was that of a 16–20-year-old male (Individual no. 57) partially disturbed and contained within a relatively shallow oblong pit <12159>. The pelvic area seems to have been the focus of destruction, a relatively common feature of burials at the cemetery. The pelvic bones, femurs and hands were absent and the radius and ulna of both arms, though present, were disarticulated and lying loose in the fill along with some phalanges (12158). The in situ elements of the burial were lying in the supine position with legs extended and feet parallel to each other. Around the head and feet of the skeleton remnants of plant-stem matting were preserved. Once this matting was lifted the head could be seen to be wrapped in textile, most likely a length of cloth that had been twisted to a knot on top of the head. Traces of textile were also found on the torso and lower legs of the skeleton (Figures 6 and 7).

Burial of Individual no. 59 — pit <12168>

The southernmost of the two pits <12168> contained another disturbed burial of an adult male, Individual no. 59, aged between 35–39 years. The pelvic area, femurs and lower torso of this individual were disturbed, some of the disarticulated bone lying loose in the fill (12167) within the burial pit. The head- and foot-end of the burial were undisturbed and remained wrapped in plant-stem matting which was particularly well preserved at the western or head-end (12170). In situ skeletal elements included the skull, thoracic vertebrae most of the ribs, upper arms, lower legs and feet (Figure 8). There were traces of textile surrounding the in situ skeletal elements. Post-excavation analysis indicated that the individual suffered from a significant amount of pre-mortem trauma with multiple healed and partially healed fractures present.

2.1.5 Grid Square J51

Unit (12042), a deposit of loose sand, was initially removed across the entire grid square exposing a lower level of sand with minor gravel and a small amount of scattered desiccated bone (12055). This lower deposit seems to mark the level at which the 2006 excavations ceased; unit (12042) comprising-wind blown sand and some baulk collapse that had accumulated over this level. Excavation of unit (12055) revealed a sand-and-gravel surface 12065 into which a large rectangular burial pit had been cut <12272>. A second burial pit <12123> was located in the north-east corner of the square continuing into grid square J52 (see section 2.2.4 for discussion of findings in grid square J52).

Burial of Individual no. 55 — pit <12272>

This disturbed burial was contained in a large rectangular pit <12272>. The north-east corner of the pit appears to have truncated another burial cut <12046> that extends into square K51 and will be discussed in later sections of this report. Three successive fill levels — (12057), (12084) and (12124) — were excavated from the upper level of the pit and comprised loose sand with minor-to-moderate gravel inclusions and a minor amount of desiccated bone. With an increasing depth of excavation a level was encountered (12127) exhibiting a slight increase in compaction of the sandy fill, though the transition point to this level was indistinct. Once unit (12127) was cleared there was a discernable increase in the amount of scattered fragmentary and disarticulated bone contained in the pit (12141). In both this unit (12141) and the underlying unit (12153) there was an increased gravel content and some scattered plant stems probably from burial matting. No in situ bone was recovered from this pit. However, at base level and lining the central area of the pit was a partially preserved plant-stem mat (12273). This matting is of the type associated with many burials at the site and presumably indicates that an individual was once interred within this pit. A minor amount of ash and charcoal was noted contained within the sandy fill surrounding the eastern end of the matting (Figure 9). During post-excavation analysis the scattered bone recovered from this pit was identified as belonging to a female aged between 45–50 years of age (Individual no. 55).

This burial pit varies from other pits so far excavated at the site primarily as it is significantly deeper, the base level being approximately 1.5 m below surface level. It also truncates another pit <12046>. This is unusual at the site as, despite the density of burials evident in other grid squares such as I52 and J52, no other burial pit cuts through an earlier burial. It was not possible to determine if the intersection of the two pits occurred during construction of pit <12272> or occurred during the later disturbance at the site.

2.1.6 Grid Square K51

Excavation in this grid square commenced with clearance of sand that had accumulated since 2006 and included a degree of baulk collapse along the southern edge of the grid square (12039). Though the transition point was not clearly evident, a thin layer of slightly more compact sand was identified (12050) which appears to represent previously unexcavated sandy fill. Clearance of this unit exposed a natural surface 12051, at which point two distinct burial pits became evident. Also exposed was an apparent straight cut edge or pit <12045> within the distinct surface 12051. This cut ran in an east–west line at the southern limit of the excavation area but could not be investigated further as it would be necessary to undertake significant clearance in the unexcavated grid square K50. It is potentially another burial pit as it is orientated approximately on the same alignment as two other nearby cuts that have been confirmed as burial pits: <12046> and <12063>. Without further excavation the nature of cut <12045> must remain uncertain (Figure 55).

In the north-east quadrant of the square an oval patch of slightly discoloured sand and an adjacent slight depression were also visible at the level of surface 12051. It was not possible to determine on surface examination if these anomalies represented a burial pit or some other feature. A decision was made to make a test cut into this quadrant, measuring 2 m x 3 m, to determine the nature of this discolouration and depression. Following excavation of c. 10 cm depth of moderately-compacted sandy fill (12121) there was no evidence of a burial cut or purpose-made depression so that excavation ceased in this area.

Burial of Individual no. 43 — pit <12046>

This burial was contained in an oval burial pit <12046> that was oriented approximately east–west (according to magnetic north rather then site north). The westernmost end of the pit extends slightly into square J51 where it is truncated by the eastern end of burial pit <12272>. This burial (Individual no. 43) exhibited a significant degree of disturbance. The uppermost fill of the pit (12047) consisted of loose sand with fragments of plant stem or reed and a small amount of desiccated loose bone and human hair. Once this loose fill was cleared the partially articulated skeleton was exposed surrounded by disturbed portions of the burial (12058) (Figure 10). The head and torso had been disturbed, many of the rib bones, having been pushed down into the base of the pit, were fragmented and the head had been disarticulated from the cervical spine (skull no. 73). In addition, the pelvic area with the femur, coccyx, the left tibia and left fibula were all disturbed and lying loose in the fill. Ongoing clearance of loose fill and the disturbed portions of the burial (12058) revealed that the body had been laid in the grave wrapped in a plant-stem mat, the remnants of which lined the base of the pit (Figure 11).

Though slightly misaligned, the right tibia and fibula of this body appear to be close to their original burial positions. The feet were absent though whether their destruction occurred during the grave robbing or during the cutting of pit <12272> is uncertain. Only the right humerus, left humerus and left scapula clearly remain in their original burial position. Impressions of the skull and body in the sandy base of the pit indicate that the body was interred with the head at the eastern end of the pit in an extended position (Figure 11). Post-excavation analysis determined that this individual is a 40–50-year-old female with at least 80% of the skeleton being recovered from the pit. Skull no. 73 was confirmed as belonging to this individual.

Burial of Individuals nos. 48, 49 and 50 — pit <12063>

A large rectangular pit containing the disturbed burials of three individuals was located approximately 1 m to the north of the burial of Individual no. 43. This second pit <12063> was also orientated in an east–west direction and was cut into the sandy surface 12051. The western end of the pit extended slightly into grid square J51. Fill within the upper level of the pit consisted of slightly compacted sand containing fragments of plant stems and some phalanges particularly concentrated at the eastern end of the pit. Clearance of this fill exposed two disarticulated skulls (no. 79 and no. 81), other scattered bone, significant amounts of plant-stem matting fragments, and a large mass of plaited hair lying adjacent to skull no. 79, all floating in a loose sand deposit (12071). During excavation, disarticulated bone from three individuals was noted; two adults and a probable infant. Once the disturbed levels of the pit were cleared the undisturbed portions of two adult skeletons were visible (12088). These individuals had been laid in the base of the pit upon plant-stem matting in the supine position, legs extended, and were lying next to each other with their heads at the western end. Individual no. 48, lying on the northern side of the pit, has been identified during post-excavation analysis as a 35–40-year-old female. Though the skull (no. 81) had been removed from this skeleton the torso and much of the lower limbs remained in their original burial positions. The left ulna and radius were articulated to each other and resting over the abdominal area. The left humerus was absent. The right humerus was present but the radius and ulna had been disturbed. Neither of the hands were preserved in situ. The coccyx and right pelvic bone had been removed as had the tarsal bones and phalanges of the left foot. The remaining bones of the lower limbs were articulated (Figure 12).

The second adult skeleton, Individual no. 49, lay in the southern half of the pit-base, also resting on remnants of plant-stem matting, and has been identified as a 40–45-year-old female. This skeleton exhibited what seems to be the most common pattern of destruction at the site focussed on the head, hands and pelvic area. The skull (no.79) had been disarticulated and was found lying loose in the fill with some of the cervical vertebrae. The lower arms and hands, lumbar spine and coccyx had all been removed during the disturbance but the pelvic bones, legs and feet remained in their original burial position, the right foot resting slightly over the left. Fragments of an infant skull (Skull no.84) were found resting over the pelvic area of this individual. A significant number of disarticulated bones from the same infant had were recovered from the upper fill levels and surrounding the in situ portions of the adult burials. Post excavation analysis indicated that at least 50% of an infant aged at approximately 9mths was represented by these bones and was designated as Individual no.50. Given the relatively delicate nature of infant bones and the significant amount of the skeleton represented within the pit it is likely that the infant was interred within pit <12063>; the absence of any undisturbed bones does make this interpretation tentative. During excavation of residual fill (12119) from the base of the pit a small amount of botanical material, seed pods and possibly desiccated fruit, were recovered. This material was probably part of the original burial goods it is awaiting botanical analysis for specific identification of plant species.

2.1.7 Grid Square L51

Loose sand accumulated since 2006 was removed across the entire grid square (12142) exposing a slightly compacted sand level (12149), the probable level at which excavations ceased in 2006. The excavation of this fill layer (12149) exposed a gravel-and-sand surface 12259 into which two potential burial pits had been cut (Figure 55).

Pit <12251>

The southernmost of these two pits was found to contain a deposit of loose sand with a minor amount of gravel (12230). No cultural material or bone was contained in this fill which directly overlies the base the pit. The pit itself is oval in shape and though, in its preserved state, is only a maximum of 10 cm in depth it does resemble burial pits as represented elsewhere at the site (Figure 13).

Pit <12253>

Located in the northern half of the grid square, pit <12253>, on excavation, was found to be a shallow depression filled with slightly friable sand (12252). Mixed throughout this sand was a powdery brown material, most likely decomposing organic mater. A few very small fragments of textile were also contained in the fill and a minor amount of plant-stem fragments of the type used in mats recovered from burials pits elsewhere at the site. The western end of this pit had an oval shape but the eastern end was poorly defined. This pit has been tentatively identified as the base of a burial pit, given the presence of decomposing organics, pit appearance and location, being cut into the Amarna-Period surface 12259. During the 2006 excavations Individual no. 19 was excavated in this approximate horizontal location but at a slightly higher level. It is probable that pit <12253> is the base of a pit that originally contained the burial of Individual no. 19 (Figure 55).

2.2 Northern excavation area — Grid Squares F52–M52

This area comprises the strip of grid squares running along the northern half of the excavation area. In 2007 this area, squares G52–M52, was partially excavated. During that season a compacted surface was exposed in parts of the excavation area, most notably in grid squares K52–M52. A series of distinct burial pits were revealed cut into this surface. Only a portion of these burials could be excavated due to time constraints, and in the remainder of the grid squares it was evident that the base of the archaeological deposits had not been reached. In 2008 excavation continued with the aim of exposing the Amarna-Period surface across all grid squares. The area of investigation was also extended slightly to include grid square F52, a previously unexcavated space.

2.2.1 Grid Squares G52/F52

During the 2007 excavations season this grid square was cleared down to a level where a distinct burial cut <11683> was apparent containing the disturbed burial of Individual no. 33 (11674). The burial was cleared during that season but time constraints prevented the completion of excavation in the remainder of the square. Some loose fill remained surrounding the exposed burial pit as well as a mounded deposit (11670) at the eastern edge of the square that was felt to potentially overlie a burial.

In 2008, following removal of backfill and some baulk collapse (12078) from the northern side of the square, a deposit of moderately compacted sand with a minor amount of gravel was removed across much of the square (12077) (unit (12077) is equivalent to unit (11681) noted in 2007 but unexcavated). This deposit lay over a more compacted deposit of sand and gravel, a distinct surface 12122. It is likely that both deposits are original surfaces that date to the Amarna-Period cemetery, the uppermost level (12077) having become slightly degraded from exposure to foot traffic during the 2007 excavation and therefore less compact. The burial pit <11683> was found to cut down through both these levels, so supporting their identification as an Amarna-Period surface. At this point a decision was made to extend the area of excavation immediately to the west into the eastern half of grid square F52 in order to determine if, as seemed the case, the cemetery surface dropped away into the wadi here, and to determine if any further burials remained unexcavated at the wadi edge.

In the south-east corner of grid square F52 a cluster of bones (12043) was visible at commencement of excavation eroding out of the sloping wadi edge. These bones were collected as a distinct unit number though the fill surrounding them is essentially equivalent to the sandy deposit (12087). This cluster of bones was located directly to the west of the position of a bone cluster (11643) excavated in 2007 and is likely to be an extension of this cluster. Given the degraded nature of the bones recovered as unit (12043) it may not be possible to establish a relationship with the bone recovered as unit (11643). Loose sandy fill (12087) was next removed from the eastern third of this grid square to a point where a continuation of the surface 12122 was encountered. As expected, this surface could be seen to drop away towards the wadi floor. There were no indications of burials in the exposed portion of the surface though it is possible that there may be burials at a greater depth within the wadi floor proper. A decision was made to cease excavation in grid square F52 at this point, as complete clearance of the unexcavated western portion of the square would have required the removal of a significant quantity of sand for which time did not allow.

In order to confirm the nature of the surface designated as unit 12122 a test cut measuring 3.5 m x 1.5 m <12267> was positioned in the south-west corner of grid square G52. Approximately 20 cm depth of this unit was excavated from within the test area and found to consist of moderately compacted sand with a minor-to-moderate amount of gravel. The gravel content varies through the deposit with some slightly more concentrated patches but no distinctly definable lenses. No cultural material or botanicals were contained in this fill. It was confirmed as a naturally-occurring deposit of sand and gravel. The excavation of this unit ceased at an arbitrary point when it was felt that adequate fill had been removed to determine the nature of the surface represented. The level at which excavation ceased was designated as unit 12271 but is essentially contiguous with and equivalent to unit 12122. (Note unit 12122 is also contiguous with and equivalent to unit 12268 in H52 and 12255 in G51).

Burial of Individual no. 58 — pit <12180>

In 2007 at the completion of excavation a mounded area of sand and fragmentary organic matter (11670) was noted at the junction of grid squares G52 and H52. At this time a skull (no. 59) was found protruding from the upper levels of the mounded fill and was collected to prevent deterioration. Time constraints prevented further clearance of the deposit so that the area was covered by cloth and then a layer of protective sand. Following the removal of this backfill in 2008 excavation of the deposit recommenced.

The upper level of fill was given a distinct unit number (12162) but essentially was equivalent to unit (11670) identified in 2007. On excavation this sandy fill was found to directly overlie a disturbed burial lying within a shallow cut <12180> (Individual no. 58). Once the limits of the pit were more clearly defined the fill was given a new unit number (12165) (Figure 14).

This upper fill layer contained disarticulated bone, fragments of plant-stem or reed matting and a small number of textile fragments. Disarticulated bone included multiple ribs: many fragmentary, a sternum, mandible, humerus, and some vertebrae; the bone generally clustered in the central area of the pit. Clearance of this disturbed bone and surrounding loose sandy fill exposed the undisturbed portions of the burial (12171). Articulated portions of the skeleton were resting on a plant-stem mat which also curved up the sides of the pit and presumably originally covered the top of the body, as fragments were recovered from the overlying fill. A deposit of slightly compacted coarse-grained sand surrounded the bones. The upper part of the skeleton from the level of the base of the thoracic vertebrae had been completely disturbed. On the left side the humerus, radius and ulna remained articulated but both the hands and the entire right arm had been removed from the base of the pit. The lower half of the body including the lumbar vertebrae, pelvis and both legs were found articulated and resting within the burial pit. The body was lying in the supine extended position but with the right hip slightly elevated so that the body was rotated to the left side. The feet were extended with the left foot resting slightly over the edge of the right foot. The pit in which the body was resting is cut down into the Amarna-Period surface 12122/12268 and has a relatively narrow rounded base (Figure 15).

The narrowness of the pit possibly explains the slightly rotated position of the body as there appears to be inadequate space in the base of the pit for the body to have been laid flat. All recovered bone was completely skeletonised the articulated bone being less desiccated than the disturbed bone. There were textile fragments overlying and underlying the articulated portions of the skeleton. The textile was very delicate and tended to crumble to powder. No obvious seams were visible in this cloth but the poor state of preservation means it was not possible to determine if it was a garment or a length of cloth. Post-excavation analysis confirmed that skull no. 59 and the disarticulated bone were from the same individual as the in situ skeletal elements. This individual has been identified as a 30–35 year old male.

2.2.2 Grid Square H52

During 2007 a similar depth of excavation was reached in grid square H52 as was reached in G52. Two relatively poorly defined burial pits were found here: pit <11707> containing the disturbed burial of Individual no. 29, and pit <11706>, also with a disturbed burial, Individual no. 35. At the completion of the 2007 excavation season there was some loose fill surrounding these burials that did not appear to represent the base of potential archaeological deposits. A third skeleton (Individual no. 23) was identified as in situ though no burial pit was discernable in the soft sand surrounding the remains. In 2008 the upper level of loose fill (12062) was removed across the majority of the grid square exposing a natural sandy surface 12268 in the western half of the trench and in patches along the eastern margin of the grid square. Overlying the remaining areas of the grid square was a deposit of slightly friable fine-grained sand (12148). It was not immediately apparent whether this deposit represented a naturally occurring deposit contemporary with the cemetery or post-dated burials at the site. The decision was made to excavate down into this fill, this process revealing a distinct burial pit <12161> in the north-east quadrant of the square, cut into a continuation of the surface already revealed in the remainder of the grid square 12268.

Burial pit <11707>

This pit was first noted in 2007 and was partially excavated at that time. A disturbed burial was recovered from the pit (Individual no. 29) consisting of a cluster of disarticulated bone and an associated partially articulated burial including the legs and pelvis of a 35–39 year old female. Partially overlying the skeletal remains were a number of relatively large fragments of painted gypsum plaster. During the 2007 excavation it was difficult to accurately determine the exact limits of the pit as the fill was very similar to the surrounding sand; this was particularly true of the northern end of the pit. Time constraints prevented complete excavation of this area and it was evident that further excavation should be undertaken in the vicinity of this pit.

In 2008 excavation of loose fill (12062) surrounding and partially overlying the northern end of pit <11707> more clearly demarcated the limits of this burial cut (Figure 16).

Some fill remained in the base of the pit and on clearance was found to consist of slightly friable sand with patches of discoloration (brown to grayish-brown) presumably from decomposing organic material (12143). There were also a few small fragments of decorated gypsum plaster comparable to that previously recovered surrounding the burial of Individual no. 29. Once the pit was completely cleared it was revealed to be an oblong shape with relatively vertical walls and a flat base (Figure 55). The pit depth as visibly preserved was only 8–10cm. Given the fact that the burial itself would have protruded from the top of a pit of this depth it is likely that at least a portion of the pit’s depth was not discernable in the loose fill surrounding the pit or had eroded away prior to deposition of this loose fill (12062).

Burial pit <11706>

Pit <11706> was identified in 2006 as a discoloured, slightly mounded patch of sand with a central depression, the remnants of a pit containing the disturbed burial of Individual no. 35. In order to confirm the identification of this feature as a purpose-made burial cut a decision was made to excavate this small mound in 2008. When excavated it was apparent that the fill comprising the edges of the pit was contiguous with, in effect equivalent to, the surface identified as unit 12268 so that the depression in the centre of the feature cuts down into the surface.

Burial of Individuals nos. 60 and 61 — pit <12161>

The excavation of unit (12148) revealed an oval pit <12161> which on excavation was found to contain the disturbed burials of two individuals. The upper level of fill within this pit consisted of loose fine-to-coarse-grained sand with a minor amount of gravel and some fragments of plant-stem matting (12169). At a depth of between 20 and 40 cm below the top of the pit a slightly friable sandy fill was encountered containing scattered disarticulated bone (12177). As excavation progressed it became apparent that an adult and an infant were represented in the disturbed skeletal material. The infant bones, though disarticulated, were generally clustered in the central pit area slightly more to the south than the north side of the pit. The main exception was the fragmented infant skull (Skull no. 91) which was scattered through the more easterly end of the pit. The adult bone was more randomly spread, though some bones such as the radius and ulna were resting next to each other. An intact adult skull (no. 92) was also found floating in this fill. Within the sandy matrix surrounding these disarticulated bones were scattered fragments of plant-stem matting, a few pieces of textile and some human hair. The removal of all disturbed bone and associated fill exposed the undisturbed portions of the two individuals. The adult burial (Individual no. 61) was preserved as the lower half of both legs including the patella and the left femur, and has been identified as 22–30-year-old female. This articulated portion of the skeleton was lying in the original burial position with feet extended and left toes resting slightly over the right toes at the eastern end of the pit. The limbs were wrapped in plant-stem matting though this mat was only partially preserved due to the burial’s disturbance and a degree of decomposition (Figure 17).

Of particular interest is the position of the infant burial (Individual no. 60). Post-excavation analysis has identified this individual as of approximately 2 years of age. The articulated lower legs (tibia, fibula and feet) of this infant were found tucked in against the southern edge of the pit. The legs were positioned with the toes pointed to the south-east in an extended position and the legs were aligned with each other. This appears to be the original burial position of the infant, bearing in mind the concentration of disarticulated infant bone directly to the north of the articulated portions of the skeleton. Remnants of plant-stem matting lining the wall of the pit surround the in situ elements of the infant burial. The matting appears to be one contiguous piece that lines the pit wall and continues down into the base of the pit. As such it underlies the bodies of both individuals. It is likely that these two individuals were interred at the same time, as later interment of the infant would have necessitated at least a partial unwrapping of the adult burial (Figure 17). Traces of textile were recovered in the immediate vicinity of both individuals but were too poorly preserved to identify a textile type at the time of excavation. All bone recovered from this pit was completely skeletonised, the disarticulated bone exhibiting variable states of weathering or desiccation. The articulated bone was only moderately desiccated.

It is also of interest that the position of this burial pit directly underlies the location of Individual no. 23. This disturbed burial was excavated in 2007 and consisted of the articulated coccyx, pelvis, legs and feet of a 20–35 year old male. No visible burial pit was discernable in association with this individual but it was surrounded by a cluster of large limestone boulders forming a rough perimeter around the body, and was identified as an in situ burial at the time of excavation. Also during excavation it was thought that the nature of the loose sand surrounding the burial may have prevented the identification of any distinct grave cut (Dolling 2007). Given that this burial can now be shown to directly overlie a burial pit containing the skeletal remains of Individuals no. 60 and no. 61 it is possible that all three individuals were interred within one pit, the upper levels of which were not visible in the less compacted sand layer. An alternative possibility is that the burial of Individual no. 23 represents a later burial phase.

2.2.3 Grid Square I52

This square was only partially excavated in 2007. The depth of excavation here was less than that achieved in adjacent grid squares due to time constraints. At the completion of the 2007 season a loose-to-slightly friable sandy fill covered much of the grid square (11684). At the western edge of the square a mounded area of sand, gravel and limestone boulders was exposed (11685). At the eastern edge of the square another mounded area of sand was exposed, the upper level of which could be seen to contain disturbed bone and some fragments of matting (11646). Both these mounded areas were identified as potential burial pits but could not be excavated, being exposed in the final days of excavation. They were covered by protective cloth and sand at the closure of the 2007 season. In 2008 excavation recommenced in this square removing the loose-to-slightly-friable sandy fill covering much of the grid square (11684). Scattered fragments of plant stems, some human hair and desiccated bone fragments were contained in this deposit. Following excavation of unit (11684) a distinct sandy surface 12217 was revealed in the western half of the grid square. In the eastern half of the square a slightly friable sandy deposit (12110) was exposed that seemed to fill a depression in the surface 12217 though the exact transition point between the two deposits was not easily discerned. Excavation of unit (12110) confirmed that it overlaid surface 12117 and also two previously unidentified burial pits: <12192> and <12206>. At this point in the excavation it was evident that the surface and its associated burial pits were contiguous with the cemetery surface exposed in adjacent squares. (Surface 12217 is contiguous with and equivalent to unit 12268 in H52). Excavation focused on clearance of the revealed burial pits.

Burial of Individual no. 51 — pit <12093>

In 2007 a concentration of gravel-rich sand and limestone boulders was noted at the western margin of the grid square but remained unexcavated. Following the clearance of sandy fill (11684) surrounding and partially overlying this deposit (11685) the limits of a distinct oval pit were revealed. It became apparent that unit (11685) was mounded directly over and continued down into this pit <12093>. Given the previously unidentified limits of this fill a decision was made to give it a new unit designation (12090) specifically related to the 2008 excavation season. On excavation this fill consisted of sand with a moderate gravel content and a large number of moderately large limestone boulders. These large stones had perhaps originally been used to cover or mark the burial pit. The gravel-rich fill and the majority of the loose floating boulders were mounded up at the northern end of the burial pit; the deposit did cover the whole pit space.

At approximately 25 cm below the top edge of the pit small patches of a brownish powdery material begin to be apparent. A decision was made to change the unit number at this level though the exact transition point between the pit fill units was not sharply defined. This lower fill level (12094) was a mix of slightly friable fine-grained sand with a minor amount of coarser sand and gravel-sized limestone fragments. Brownish smears of decomposing organic material were found throughout this fill, most likely fragments of termite-affected timber. Several limestone boulders were found floating within the sand. At the western end of the pit a particularly large boulder was revealed and a cluster of moderately-sized boulders were visible at the eastern end of the pit. Surrounding this boulder large fragments of timber or bark were lying in the fill (Figure 18).

At this point in the excavation the fill was designated as a new unit (12102) as it seemed to mark a concentration of disturbed stones and burial material in the form of decomposing timber, and as such may be separated from the higher fill levels that had perhaps accumulated after the main disturbance of the burial.

Ongoing clearance of the sandy fill and large limestone boulders revealed the remains of a disturbed skeleton (Individual no. 51) (12116) within a poorly preserved timber coffin [12118] (Figure 19). The largest boulder at the western end of the pit was resting directly on top of the skull of the individual possibly explaining the damaged state of this skull. In the small space between the western end of the coffin and the pit’s western wall a group of in situ ceramic vessels were found. These vessels were covered by and surrounded by sand and gravel fill (12102) indistinguishable from the fill surrounding and covering the remnants of the wooden coffin. Six vessels were recovered, each given a distinct unit number. Two small-to-medium sized jars (units [12111] and [12113]) were recovered upright with their rims covered by inverted small dishes (units [12105] and [12112]). These jars were standing side by side. Neither contained any obvious contents but a small amount of loose sand in the base of each jar was collected for analysis. On the north side of the group of vessels two shallow bowls were found, the uppermost [12104] was an intact vessel containing botanical material in the form of seeds and/or grain and possible fruit. Beneath and slightly to the north a second shallow bowl was found [12114]. Though this vessel was broken, the fragments were still aligned and it also contained botanical material that on visual inspection compares to the material contained in the first bowl (Figure 20).

A partially disturbed skeleton of a 10–14 year old individual of indeterminate sex was lying within the exposed coffin [12118]. The torso of the body had been disturbed with many of the ribs disarticulated and fragmented but still within the coffin (Figure 21). All the vertebrae remained articulated to the skull and along the entire length of the spine to the coccyx. The thoracic vertebrae do appear slightly pushed out of alignment probably during the disturbance of the burial. The head was at the western end of the coffin and was rotated to the right side of the body. It was not possible to tell if this rotated position was the original burial position or if the head had been slightly moved during the disturbance. Other bones lying loose within the coffin include the pelvic bones, the right humerus and some phalanges. Two phalanges aligned to each other lay against the right hip area probably marking the original position of the right hand. The radius and ulna of both arms and the entire left hand were disturbed. The legs remained in their original burial position with textile fragments adhered to the bone. This textile was clearly wrapped around and over the lower limbs. There were also fragments of poorly preserved textile underlying the skull and torso. Both of the legs were extended with the right foot partially overlying the left. Surrounding the articulated and disarticulated bone within the coffin [12118] was a deposit of slightly friable sand with a minor amount of gravel and a large amount of decomposing organic material, evident as brown flecks (12116). The coffin itself had been severely affected by termite activity. The wood structure had been lost so that it was primarily preserved as a powdery brown material. It quickly became apparent that it would not be possible to remove this coffin intact. As a result a decision was made to draw and photograph the coffin in situ and then backfill the burial pit. There was no indication that the coffin had been decorated but, given its poor state of preservation, there is a possibly that any such decoration has been destroyed. It seems to have been a simple rectangular box. Some of the large fragments of decomposing wood from the overlying disturbed fill are probably from the lid of the coffin.

Burial of Individuals nos. 64 and 65 — pit <12145>

In 2007 a mound of sand with some desiccated bone was partially excavated (11646). Fragments of bone including vertebrae and long bones were recovered but due to time constraints the complete depth of the deposit was not cleared. In 2008 excavation recommenced designating the unexcavated portion of the visible mound as unit (12120). Minor amounts of desiccated bone and fragments of plant material were contained in a matrix of fine-grained sand.

Once this deposit was cleared a distinct oval pit became visible <12145> (Figure 22). Upper fill level of the pit proper consisted of sand with a moderate amount of scattered gravel-sized limestone fragments (12182). A large amount of brown grass-like material, possible halfa grass, was contained in the fill, particularly concentrated in the south-east portion of the pit, and several large fragments from ceramic vessels. Beneath this fill a slightly more compacted deposit of gravel-rich sand was encountered (12186). The transition point between the two fill levels was not distinctly demarcated. Disarticulated and desiccated bone was found scattered through this fill level including a skull (no. 94), femur, scapula, coccyx and multiple ribs. At this point the presence of two sternums within the fill indicated that the pit potentially contained more then one burial.

At the southern end of the pit the removal of unit (12186) exposed the articulated lower legs and burial matting of two individuals (12189). The legs were lying side by side with the feet at the eastern end of the pit. Preserved bone of both individuals was wrapped in independent mats constructed of some type of plant fibre or plant stem. The plant material used to make these mats appeared to be more fibrous in nature then the plant-stem type used most commonly at the site. The matting surrounding the northernmost burial (Individual no. 64) slightly overlies that of the southern burial (Individual no. 65). However, as the pit is relatively broad in comparison to the majority of the pits at the site it was most likely originally constructed to contain the two burials so that both individuals were presumably interred at or around the same time. A minor amount of coarse-grained sand was preserved surrounding the bones and burial matting (12189). No in situ skeletal remains or burial material were preserved at the western end of the pit where unit (12186) directly overlaid the base of the shallow pit. Individual no. 64 was identified at the time of excavation as the articulated tibia, fibula and foot of both legs, the legs lying beside each other, the left toes slightly overlying the right. Post-excavation analysis has determined that the skull (no. 94), a significant proportion of the upper torso, the remainder of the leg bones and the arms of Individual no. 64 were included amongst the disarticulated bone floating loose in the pit. This individual has been identified as a 12-year-old juvenile.

Individual no. 65 was also lying in the base of the pit on the southern side and adjacent to Individual no. 64. The legs of this burial were extended; the right leg was crossed over the shin area of the left leg with feet flattened though the phalanges were bent under slightly, possibly due to lack of space at this end of the pit. Individual no. 65 has been identified during post-excavation analysis as a 40–45 year old female. A number of the disarticulated bones from the pit fill were able to be associated with this individual, including humeri, the left femur and pelvis, the coccyx, phalanges and some vertebrae (Figures 23 and 24).

Burial of Individual no. 66 — pit <12192>

The burial pit <12192> containing the disturbed skeleton of Individual no.66 was exposed in 2008 by the excavation of sandy fill from the eastern half of the grid square (12110) (Figures 25 and 26). The upper level of the pit was cleared of sand including a minor amount of scattered pebble sized limestone fragments (12191). Fragments of ceramic, plant stem matting and fibre rope were found floating in this fill. At the northern end of the pit a mass of disarticulated bones was also revealed surrounded by this fill. The bones were markedly desiccated some quite brittle and could not be lifted without breakage particularly the ribs. Other bones identified in this cluster were a clavicle, several vertebrae, a mandible, sternum, coccyx, and long bones of the arm. There were also a few irregular lumps of greyish brown mud possibly brick or mortar fragments though no original faces were visible. Underlying this disturbed level were the in situ portions of the burial. Only the legs were preserved in their original burial position located at the eastern end of the pit in a supine extended position with the feet lying side by side. Ongoing clearance of the in situ elements of the burial and surrounding fill (12199) revealed the body had originally been wrapped in a plant stem mat the impressions of which remained in the base of the pit, though much of the matting had decomposed. All bone recovered from this pit was completely skeletonised; in situ skeletal elements were only moderately desiccated in comparison with the disturbed bone. A small steatite cowroid bead (object no. 38645) was found floating in sandy fill in the base level of the pit.

Burial of Individual no. 68 — pit <12206>

The disturbed burial of Individual no. 68 was contained in an oval burial pit <12206> on a similar alignment and positioned between two other burials pits: <12192> and <12145> (Figures 27 and 28). Burial pit <12206> was first revealed in 2008 by the excavation of unit (12110). The uppermost level of the pit fill (12201) consisted of loose fine-grained sand with a minor amount of gravel and coarser-grained sand. A small amount of desiccated bone was found floating in this fill. Once this deposit was removed a cluster of disarticulated bone became apparent concentrated at the western end of the pit and surrounded by a similar deposit of sand with minor gravel (12205). Disarticulated bone included a skull (no. 96), ribs, scapula, a humerus, sternum, mandible, vertebrae and unidentified long bones. The skull was crushed on the temporal area but the damage here looks recent and may have inadvertently occurred during excavation from pressure of foot traffic on the overlying fill. There were also a large number of plaited braids in the vicinity of the skull. In the north-west corner of the pit a length of apparently hollow wood was recovered that was filled with an unidentified substance. This object tapers to a rounded point at one end; the other open end has a fine wooden stick protruding from it. This object has significant weight indicating that the substance contained in the hollow area may have a portion of a metallic element such as lead. Though it requires further study the object has tentatively been identified as a kohl-tube and applicator (object no. 38640). Beneath the disturbed level of fill portions of the skeleton remained articulated in the original burial position (12208). The body was found lying in the base of the pit on remnants of a decomposing plant-stem mat, this mat also wrapped around the lower legs and presumably originally around the entire body. Traces of textile were found over and under the articulated portions of the skeleton. Undisturbed portions of the skeleton include the lower spine: two thoracic and all of the lumber vertebrae articulated to the coccyx, the pelvic bones, both of the legs and the feet. The bones were only moderately desiccated but completely skeletonised; several have breaks or cracks in them. The skeleton was lying in the supine position, feet crossed at the ankle area, the right leg over the left with heels down and feet extended. The feet were located at the eastern end of the pit. During post-excavation analysis the disarticulated bone and the in situ bone were all found to belong to a single individual identified as a 25–30-year-old female.

Following excavation of the skeletal remains and matting a deposit of slightly compacted sand and gravel was revealed filling the base of the pit (12210) (Figure 29). During excavation this deposit was also visible filling the space between the edges of the burial mat and the pit wall and as such must represent purposely-laid fill placed in the pit at the time of the interment. No artefacts were recovered from this fill. At the northern end of the pit a small undercut area in the pit wall was exposed. This space was filled by a continuation of the disturbed fill unit (12205). It is a slightly arched space. It was not possible to determine if this cut was a deliberately-made space and part of the pit’s original construction or was the result of intrusive digging.

Burial pit <12181>

In 2007, during exactions in the south-east corner of square H52, a cluster of desiccated and disarticulated bone was recovered from loose sandy fill and designated as unit (11616). Post-excavation analysis identified most of these bones as belonging to a 17–20-year-old individual of indeterminate sex. Though an apparent base of the bone cluster was reached by the end of the excavation season a deposit of loose sandy fill continued to a greater depth and raised the possibility of additional archaeological material being preserved here. During 2008 in the adjacent grid square H51 clearance of sandy fill (12091) revealed a cluster of bone and plant-stem matting (12107) that continued slightly into grid square H52 directly beneath the location of the previously excavated bone cluster (11616). Subsequently excavation was undertaken in the south-west corner of grid square I52. Here a deposit of loose sandy fill (12160) was cleared containing a small amount of fragmentary bone and some fragments of plant-stem matting and was found to directly overlie the remnants of a burial cut <12181>. The cut was preserved only to an approximate depth of between 5 cm and 8 cm but had the impressions of matting and fibre rope in the base, clearly indicating its original function as a burial pit. At the time of excavation this pit was found to terminate at the junction of H52 and I52 (Figures 30 and 55). Reviewing the description of disturbed bone, matting and slightly consolidated sand excavated as unit (12107) from H52 and H51 earlier in the season it is highly likely that the pit originally continued into both these grid squares. It is probable that disarticulated bone from units (11616) and (12107) are from an individual/s that had originally been buried in the underlying pit <12181>. Post-excavation analysis indicated that at least a portion of the bone recovered from unit (12107) could be associated with the cluster individual from unit (11616).

Pit <12223>

Pit <12223> was located at the junction of four grid squares J51, J52, I51 and I52 but is discussed here as one unit. Edges of this pit were first noted in grid square I52 following the excavation of units (11684) and (12110) but it was not excavated until a comparable depth of excavation was reached in the adjacent grid squares. Loose sand overlying and directly surrounding this pit was then cleared in two levels — unit number (12220) and (12221) — to a point where the limits of the pit were clearly definable on all sides. In these upper fill layers there was a small amount of desiccated bone including a femur and a vertebral segment; there were also some fragments of wood with possible painted decoration, though these fragments were very poorly preserved. The upper fill level of the pit proper (12226) consisted of slightly friable fine-to-medium coarse-grained sand with a moderate amount of small limestone pebbles. A desiccated pelvic bone and a few other small bone fragments were found floating in this fill. The only other inclusions were a very minor number of plant-stem fragments and a few small probable seeds of an as yet unidentified botanical species. At a significant depth of excavation into this clearly defined pit there was an increased concentration of limestone pebbles, in particular at the more westerly end of the pit. It also became apparent that this deposit (12229) filled a recessed or undercut area at the eastern end of the pit. The transition point between units (12226) and (12229) was not clearly definable. Inclusions within the unit (12229) were a small amount of bone (vertebrae and phalanges), a lump of grayish-brown mud with no original faces and a cluster of botanical material which at the time of excavation seemed to be a mixture of seeds or fruit such as dom-nuts and possible grains. This botanical material was contained in the sandy matrix of unit (12229) but was clumped against the north pit wall approximately halfway along the pit’s length.

At completion of clearance the pit was found to be clearly defined with relatively vertical sides and a flat base. It is oblong in shape and has a distinct recess or undercut area at the eastern end. This undercut space is well formed, follows the line of the pit edge relatively smoothly and has slightly arched walls and roof (Figures 31 and 32). No artefacts were recovered from the fill here but it does appear to be a deliberately-made portion of the original pit. The very minor amount of bone recovered from this fill precluded the identification of an individual associated with this pit. In fact, it is by no means certain that the pit ever contained a burial. Its shape and size is indicative of a burial function but there are no in situ elements such as matting or timber coffin fragments within the pit. If it was used for a burial the entire burial including the body and any associated containment material such as matting or a coffin must have been removed in a relatively controlled manner so that fragments were not left behind in the pit.

2.2.4 Grid Square J52

During the 2007 excavation season this square was cleared to a point where a distinct sand and gravel surface 11686 became apparent. Within this surface a series of pits or depression were visible filled with loose sandy fill. These deposits were unable to be investigated due to time constraint but had the appearance of burial pits. In 2008 excavations in this square initially focussed on these patches of loose sandy fill with the aim of clarifying the location of any underlying burial pits. A total of six definite and two possible burial pits were revealed as well as an irregular depression of an indeterminate nature.

Burial of Individual no. 46 — pit <12081>

At the end of the 2007 season a small depression was noted within the exposed sandy surface 11686. Fragments of plant-stem matting were lying on the upper level of this depression (11673). These matting fragments were lifted but no further excavation was possible. At the time it was suggested that this may mark an infant burial. Recommencing excavation in 2008 a deposit of loose-to-slightly-friable, medium coarse-grained sand (12080) was cleared from within what was revealed as a small shallow pit <12081> (Figure 33). The fill surrounded a disturbed infant burial including a cluster of disarticulated and fragmentary bone and the partially preserved in situ bones of the lower leg, both tibia and one fibula. All skeletal material was moderately desiccated, the disarticulated bone was particularly fragmentary and brittle.

Lining the base and edges of the pit and underlying the in situ bones was a plant-stem mat, the stems woven together with fibre rope. In the base of the pit underlying the matting a compacted yellowish-brown deposit was noted (12083). Around 3 cm depth of this deposit was excavated and it was then determined to be a natural surface, in effect the pit base which had become discoloured by decomposing organic matter. Though less then 50% of the individual was identified at the time of excavation, the preserved bone was clearly contained within its original burial pit and as such was designated as Individual no. 46. Post-excavation analysis has assigned an age of 9 months to this infant.

Burial of Individual no. 47 — pit <11699>

In the central area of the grid square a probable burial cut <11699> was identified at the completion of the 2007 season but remained unexcavated until 2008. Unit (12086), the uppermost level of fill from this pit, was excavated consisting of fine-to-coarse-grained sand with a moderate amount of gravel-sized limestone fragments. A small number of plant-stem fragments were found floating in this fill. The removal of unit (12086) revealed a slightly more compacted sandy fill with several large limestone boulders embedded within it (12089). There was a discernable increase in the amount of gravel contained in this fill and a minor amount of fragmentary ceramic. This deposit continued down into the pit to a depth of approximately 35 cm below the pit edge. At this point a cluster of disarticulated bone and flecks of brownish organic matter contained within gravel-rich sand became apparent (12092). The bone was concentrated at the western end of the pit and included a skull (no. 84), two pelvic bones, radius, ulnae, scattered phalanges, vertebrae, scapulae and a sternum (Figure 34).

Though none of these bones were articulated at least 50% of an individual appeared to be represented here and was designated as Individual no. 47. Post-excavation analysis has identified this individual as a 20–25-year-old female. Removal of the sand-and-gravel fill surrounding these skeletonised bones exposed a poorly preserved mat or coffin-like object lining the base of the pit and surrounded by a thin layer of fine-grained sand (12100). This burial container was badly deteriorated by termite activity but appears to have been constructed of a combination of reed or hollow plant stems and a more fibrous grass-like material. There were also small fragments of what may be decomposing wood or bark in the base of the pit. Samples of the material were collected from the northern end of the pit for later analysis but the majority was so degraded that it crumbled to fine particles on any attempt to lift it. The decision was made to leave the remaining material in the base of the pit and to cover it with backfill.

Burial of Individual no. 53 — pit <12126>

An oval patch of loose sandy fill first noted in 2007 (11700) remained unexcavated until 2008 when the clearance of 2–5 cm of this fine-grained sand revealed a relatively narrow ovoid pit <12126>. The top of a weathered skull (no. 86) was also visible protruding from the upper level of pit fill (12125). This deposit was excavated and found to fall easily away from the pit edges. It consisted of sand with a minor amount of gravel and inclusions of fragmentary textile and plant-stem matting. This relatively shallow deposit overlaid a cluster of disarticulated bone concentrated at the southern end of the pit and surrounded by a slightly more compacted sand-and-gravel deposit (12135). The base of the disarticulated skull was resting in this deposit (Figure 35). Other disarticulated bone identified included humeri, two scapulas, fragmentary ribs, a mandible, cervical vertebrae and scattered phalanges. Once this deposit was cleared the undisturbed portions of the burial were visible (12138). The body was lying in the prone extended position; most of the spinal column was articulated to the ribs though the matting originally covering this portion of the burial had been removed. From the pelvis to the feet a plant-stem mat remained in situ wrapped around the body (Figure 36).

It is clear that the disturbance has focussed on the head area with minor disruption of the torso by removal of matting and the upper arms. The disturbance of the scapula, occurring on a body in the prone position, made these bones uppermost. The lower arms of the burial were in their original burial position, bent at the elbow, hands resting over the pelvic area; the right hand slightly over the edge of the left hand. The prone position of the burial is probably at least partially responsible for the non-disturbance of the hands. A small number of the right ribs had been disturbed but the left side of the rib cage was intact. The legs were lying side by side, the feet extended so that the heels were uppermost and the top of the feet were resting flat on the base of the pit, the left foot slightly over the right. Textile fragments were preserved overlying the skeletonised bones. This textile was in the best state of preservation around the feet and lower legs, where it could be seen to have originally wrapped around the lower limbs probably as a single length of cloth. There were also fragments preserved around the femurs, hip and torso. The textile was very desiccated, with much disintegrating to a powdery state. Some samples were recovered. Plant-stem matting was well preserved at the northern end of the pit where it could be seen to be constructed of a series of stems woven together with twisted fibre rope.

Burial of Individuals nos. 54a and b — pit <12128>

In the south-west corner of the grid square an irregular patch of loose sandy fill (11703) was visible at the completion of the 2007 excavation season. This deposit was excavated in 2008 and found to overlie a slight, probably natural depression in the extreme south-west corner of the square and a small rectangular pit <12128> to the north of this depression. An adult long bone and several small limestone boulders were visible protruding from the upper level of this pit (12146). On excavation a number of additional desiccated bones were recovered floating in this fill the majority of which appeared to belong to an infant. Some small fragments of wood and fibre rope were also contained in this layer. The lower level of the pit was filled by a slightly friable deposit with a moderate amount of gravel and an increased amount of scattered disarticulated bone (12147).

Although the majority of the recovered bone appeared to belong to an infant there were a small number of identifiably adult bones. During excavation at least 50% of an infant was represented amongst the bone, including a fragmentary skull (no. 90), bilateral femurs, multiple ribs and vertebrae, a fragmented scapula and bilateral tibias. This skeleton was designated as Individual no. 54a. Post-excavation analysis determined that the majority of this infant’s skeleton was represented here. The major missing elements were the coccyx, facial bones and some phalanges. The bones have been identified as those of a 1–1.5-year-old infant. The remaining bone belonged to an adult of indeterminate age and sex and was designated as Individual no. 54b. Only a minor amount of adult bone was recovered and it is extremely unlikely that an adult was ever buried in this pit given the pit’s dimensions. The more likely explanation for the presence of the adult bones is that they originate from another disturbed burial in the vicinity of the pit.

Once the disturbed bones and associated fill (12147) were excavated, the pit could be seen to narrow abruptly to form an inner section in which the impressions of a timber or bark coffin was evident. A small northern ledged area was present at the junction of the upper and lower portions of the pit. Some fragments of timber with white gypsum plaster adhering to them had been recovered from the disturbed fill (12146, 12147). Also visible pressed in against the edge of the inner pit were small traces of white gypsum. The sandy edges of the pit are stained a grayish-brown colour in this lower area. The walls of the lower portion of the pit have a slightly curving profile and a distinctly smoothed appearance from something adhering or fitting snugly into the pit. There were also impressions of rope or string pressed into the pit walls. The archaeological evidence suggests that whatever the form of burial container used it had been tied together with fibre rope and had at least slightly flexible sides. It also seems to have been constructed of a combination of materials including timber, possible bark and gypsum. In the base of the pit a loose deposit of gravel and limestone pebbles was encountered (12151) (Figures 37 and 38). Embedded slightly into this deposit in the north-west corner of the pit was a small ceramic dish [12156]. The fill (12151) does appear to be a deliberate deposit associated with the original burial but it was not possible to determine if the vessel was deliberately placed here at the time of the infant’s burial or had fallen into this location during the disturbance phase.

Burial of Individual no. 63 — pit <12129>

In the north-east quadrant of the grid square a roughly oblong area of loose sand had been identified in 2007 (11678). At this time a skull (no. 71) was exposed in the upper c. 5 cm of this mound. The skull was significantly weathered and a decision was made to collect it at the completion of the excavation season; the area was then backfilled. In 2008 this loose sandy fill was excavated (11678) and found to be a relatively shallow deposit of 3–5 cm that, once removed, exposed a distinct oval pit <12129>. The upper fill level of this pit consisted of fine sand with a large number of gravel-to-pebble-sized limestone fragments and a few larger limestone boulders (12172) (Figure 39). Fragments of plant-stem matting, textile, ceramic and bone were recovered from this fill. At a depth within the pit where there was a discernable increase in the amount of bone and plant-stem matting a decision was made to commence a new excavation unit (12178). This unit was found to fill the lower level of the pit and contained both disarticulated and undisturbed portions of an adult burial (Individual no. 63). Surrounding and underlying the disturbed burial was a deposit of slightly friable sand with a moderate amount of gravel. Fragments of plant-stem matting, textile and several large ceramic fragments were recovered. A matchstick found floating loose in this fill. Its presence here is difficult to explain as there are no other indications of the burial having been disturbed in recent times. The pit had been exposed over several days which raises a faint possibility of contamination of the fill. However, the match was at a significant depth within the pit, surrounded by the fill unit (12178).

Much of the body had been disturbed, only the left leg and pelvis remaining articulated. Unlike the majority of the bone recovered from the site the skin of this leg was preserved partially covering the bone. There was no underlying tissue preserved and the preservation of the skin is probably the result of natural desic