In designing our selection of areas to excavate we had three objectives: 1) To determine he depth of deposits and the extent of the site so that as a second stage of our research we could estimate the total volume of occupations to be sampled – and from this judge the likely accuracy of our conclusions; 2) To secure enough excavated materials to make preliminary estimates of he variability and density of floral, faunal, artifactual, and architectural remains; and 3) To sample sufficiently the margins and core areas of the settlement to determine the outlines of both the settlement areas and the non-cultural, Pleistocene sand-gravel formation (gezira) on which the settlement was located.

Given these objectives, our strategy was to locate some of our units by means of a stratified random sampling design, and to select other areas on the basis of specific questions about topography, stratigraphy, etc. [Set up of grid system here] Excavation units, unless otherwise noted (especially in 1988), are designated by the southwest corner of the square.

In the 1984 season we excavated mainly in 2x1m units, because this relatively small size permitted precise stratigraphic control and also allowed us to excavate many units in different areas of the site. A stratified random sampling design was used to locate six units within the central area of the site (see [Wenke, 1985 #288]:6). Three additional areas were also chosen for excavation based on features of interest: A small step trench (ST-1) was excavated in the southwest near the modern village where sebakhin had cut away a portion of the mound allowing us to expose a fairly deep stratigraphic section; an area where a wall was exposed by [Hamada, 1950 #248] in their excavations (“Test Square “); and a third – 1203S/1070E – a 1x2m unit adjacent to unit 1202S/1070E excavated to expose a large hearth located in 1202S/1070E. [Note: Both are located on the map in W&R 1985]

In 1986 we excavated in larger units, some of them 2x2m areas, again most placed according to a stratified random sampling design. In addition, two more step trenches were excavated (ST-1 and ST-2) to determine if a depression between the main area of our excavations and a modern village was formed by excavation or perhaps by a water course. A 2x4m unit was also excavated to the west of the main settlement mound in an area of newly discovered Old Kingdom occupations. Finally, an area of 72 m2 was excavated in an area where well-preserved architectural remains were very near the surface. This area, named the “Block Area”, was not excavated to great depth in 1986, but portions were more deeply excavated in the 1988 season.

The 1988 season concentrated on deepening and extending the Block Area excavations and also investigating other areas of interest revealed in previous seasons. Excavation methods changed somewhat for the 1988 season; spatial control was still maintained using the original grid system but excavation was generally carried out within boundaries defined by architectural features, usually 3-4 intact walls. These were named by “room” in sequential order (Room 1, Room 2 etc.). In some cases where walls were not present an arbitrary square (designated by its southwest corner) was used for unit boundaries. Two other areas were also cleared to reveal the architectural features, one a few meters northwest of the main block area (containing Rooms 17 and 20) and another in the far western portion of the site abutting agricultural fields (Rooms 22 and 23).

Excavations were carried out by natural stratigraphy to varying depths. The basic excavation unit over all three seasons was the “sedimentary unit” or SU. The SU was defined to be a sedimentary structure that differed from the surrounding sediments in compositional characteristics (color, content, structure, texture, etc.) and extent. Ideally, each SU could be considered a separate depositional event; in practice, the excavator often noted that what initially appeared to be separate units were actually variable portions of larger units. SU’s can thus represent either cultural (e.g., a wall) or natural (e.g., sheetwash) depositional events. Thus, the SU represents he basic unit of analysis for all of our subsequent work.

The scale profile drawings included below for the 1984 and 1986 excavation units are based on stratigraphic analysis carried out in the field by Paul Buck:

The individual deposits were distinguished in the field based on examination of the stratigraphy in the excavation squares, and boundaries are based on color, dominant particle size, composition, and texture, all field observable attributes. The resultant analytical units are called “profile units” and are assigned roman numerals. . .profile units are not always equivalent analytical units to SUs [Buck, 1990 #249]:113.

Thus, for the 1984 and 1986 season profiles shown below the strata shown are profile units, not SUs. Buck (1990) provides extensive discussion of many of the 1986 units, and the remainder are described here.

In 1988 we were more interested in recovering as much artifactual material as possible from secure architectural contexts rather than building up a depositional history of the site. Consequently, excavators tended to favor “lumping” as opposed to “splitting” the create SUs. (that was probably a useless observation) Also, profile units were not specifically defined and drawn on profiles since most often walls defined the unit boundaries and thus contained no interior deposits in the section.

Chronology
A more detailed description of the various methods used to generate a temporal sequence at Kom el-Hisn is in chapter XX. Given the nature of the excavation, there were very few cases outside of the Block Area where deposits in separate excavation units could be linked to create a sitewide stratigraphic chronology. However, [Cagle, 2001 #666; Cagle, 2003 #720] analyzed the stratigraphy and artifact contents of most 1986 and 1988 units and created a general chronology for those deposits. A total of six levels were defined numbered 0 (top) through 5. Level 0 was defined to be any deposit demonstrably later than Old Kingdom date. Levels 1-5 comprised most of the deposits with Level 3 making up the bulk of the exposed architecture and domestic deposits. Levels 4 and 5 occurred sporadically in some of the deeper units. Levels are noted for all deposits so classified.