4.4 Assessment of Soils

by Richard I Macphail, BSc, MSc, PhD *(1st draft December 1995; redrafted June 1996)

4.4.1 Introduction

A detailed assessment of the soils excavated during the period 1986-1992 was carried out in 1993 (Macphail 1994a) and reported in Powlesland (1994). More recently, considerable advances have been made in investigations of the local soils and sediments. This comes from:

• Dr Maureen McHugh's preliminary findings from the off-site area (1986-1992)
• 1995 borehole information collected by Matthew Canti and Jennifer Heathcote and from the 1995 excavation.

The last period of fieldwork was carried out in the context of the 1986-1992 findings and our improved understanding of rural and occupation soils in general (Fisher and Macphail 1985; Gé et al. 1993; Macphail 1994b; Matthews and Postgate 1994; Macphail and Goldberg 1995; Gebhardt 1995; Engelmark and Linderholm 1996). Given the scale of excavation and related bio-artefact recovery, it is clearly possible to apply these systematic advancements in our scientific microstratigraphic methodologies to the soils at West Heslerton in accordance with the archaeological objectives of the site. The role of these microstratigraphic studies at West Heslerton, as demonstrated in this assessment, is all the more crucial now that pollen and land snail evidence has proven to be insufficient for recovery of usable data (see also 4.7 Pollen, 4.5 Land Snails Assessment).

4.4.2 Aims and objectives

The sampling strategy during the 1995 excavation campaign was focused upon a number of objectives identified in 1993 (Macphail 1994a) and then flexibly applied to newly discovered archaeology as it was revealed. During the 1993 assessment a number of thin sections were scanned and a few bulk samples were analysed, these soil studies being driven by the archaeological objectives of the time. The primary aim was to characterise the soils of a number of specific contexts (see Figure 4.5) so as to aid two of the primary archaeological research themes, which concerned:

1. The settlement morphology (6.5.6 Settlement morphology and planning) and
2. The environment and economy of the settlement (6.5.3 Reconstructing the environment, 6.5.4 Agriculture, economy, production and exchange).

Within the context of these objectives, special attention was paid to a third research objective, relating to the construction, use and disuse of Grubenhäuser (6.5.2 Anglo-Saxon settlement evolution and decline). During 1995 Grubenhäuser, as an established area of research, continued to be sampled as new areas of the settlement were uncovered, to fulfil objectives related to 1 and 2. As fieldwork progressed, however, further topics of study were soils associated with new aspects of the settlement morphology and its relationship to the environment and economy of the site. These basically revolved around domestic activities, industrial activities and animal husbandry. The last had already been indicated by the abundance of animal bones on the site and the presence of deposits in Area 11BA that were indicative of an ashed dung-rich farm mound (Powlesland 1994). The assumed importance of animal stocking activities, in the as yet unexcavated areas of Areas 12AA and 12AB, led to the involvement in 1995 of Ms Jennifer Heathcote (UCL) in the geoarchaeological sampling programme. This was in respect of her CASE (EH and NERC) research into the effects of domestic animal activity on soils, Ms Heathcote being supervised by the present author, along with Matthew Canti (AML) and Ken Thomas (UCL). Ms Heathcote's involvement led to an expanded sampling strategy as regards stocking practices, animal management, animal movement and dung deposition (stabling) etc., as part of the investigation into the settlement's organisation and economy. For example, buried soils, the 'farm mound' deposits, linear feature fills (ditches, etc.) and colluvium sealing worn pebble surfaces, were sampled. On the other hand, Grubenhäuser, linear features and specific contexts such as cesspits were also sampled in order to study domestic waste disposal, areas influenced by industry, etc., as additional elements of the investigations into themes 1 and 2. These studies would also yield data that would contribute to the identification and management of similar sites (2.2 Project Objectives).

Additional aims of the soil sampling programme.

4. To aid the interpretation of the transitions and changes in land-use and settlement patterns during the Roman-Saxon and Early Saxon-Middle Saxon periods (2.2.1 Primary objectives). This aim was to be tackled not only from the main sampled contexts (see above), but also from specific soil sequences identified by Powlesland (see Table 4.11, e.g. samples 72, 73 and 74).
5. To build upon the reconstructed prehistoric soil history of the sandy area to the north (Site 1), by a minimal investigation of soils pre-dating the Roman and Anglian settlement. This was to permit interpretations of soil conditions relating to the Roman and Saxon periods to be based upon a reasonably well-understood Roman soil progenitor. At Site 1, this approach allowed a robust soil model covering the prehistoric to Saxon periods to be constructed (Macphail 1982; Fisher and Macphail 1985; Powlesland et al. 1986). It would be difficult to achieve fully objectives 1, 2 and 4 without this basic soil background data.
6. Full excavation of this settlement and the three-dimensional recording of bio-artefacts provide a welcome opportunity to test methods of recovering soil data. This opportunity is unique in terms of the site being an Anglian rural one and unusual in terms of geoarchaeological studies generally. It will provide highly valuable analogue data to the geoarchaeological fraternity. This is because methodologies for analysis and prediction can be fully assayed, advancing the discipline in a much more effective way than could be achieved by investigations of only partially excavated sites.

4.4.3 Samples and methods

4.4.3.1 Field

During the period 1986-1993 the sampling strategy was designed to sample a representative cross-section of the many contexts present according to the research objectives of the time (see above).

In 1995 the sampling strategy was two-fold in its approach to the main soil objectives. Firstly it was designed to permit the investigation of the research topics identified in 1994 (Powlesland 1994), such as the continuing study of industrial zones as identified from deposits in Grubenhäuser and linear features. The case for this had been strengthened by the results of the laboratory assessment (see below). Secondly, the approach permitted newly discovered archaeological contexts to be built into the sampling strategy (see above) and in this way particular attention was paid to stocking areas which were absent from the previously excavated areas.

During the 1986-1992 investigations, 42 thin section samples were taken by McHugh and Macphail from Sites 2 and 11 (Table 4.7; Figure 4.5). The 1995 excavation yielded another 78 undisturbed samples, as collected by Heathcote and Macphail. All are conserved by having been impregnated with crystic resin (Murphy 1986).

The proportion of samples per feature type reflects the nature of site, the soil research objectives outlined above and the archaeological objectives (Powlesland 1994, 9-10, and 2.2 Project Objectives).

These are approximately (Figure 4.5 and Table 4.7):

Coprolites were also recovered during the excavation and a selection is to be studied. Various types of smithing waste have also been received from Jane Cowgill to act as reference material for the identification of industrial spreads. It is also intended to collaborate with Alan Vince and others over the identification of building and raw materials used for ceramics, spindle whorls, daub, mortar, etc., used on the site.

Fig. 4.5 Soil micromorphology: Proportion of soil samples per feature/theme

4.4.3.2 Laboratory

Of the 1986-92 samples, 35 were manufactured into thin sections (at the Universities of Stirling, Newcastle and the Institut National Agronomique, Grignon, France) and assessed employing Bullock et al. (1985) and Courty et al. (1989). Selected complementary bulk samples were analysed for grain size and chemistry (see below).

Soil micromorphology:

The 1986-92 thin sections were observed under the polarising microscope using plane polarised light, crossed polarised light, oblique incident light and ultra violet light (UV). UV is useful in identifying modern roots and phosphatic residues which fluoresce under UV. The last include fine fragments of bone, some ash residues, chalk stained with phosphate and specific anthropogenic materials such as human coprolites, which are particular indicators of human occupation (Courty et al. 1989; Macphail et al. 1990; Macphail and Goldberg 1995).

Chemistry and geophysical studies:

During 1986-92 selected bulk samples were analysed for grain size, loss on ignition, organic carbon, calcium carbonate and magnetic susceptibility (Tite and Mullins 1971; Longworth and Tite 1977; Avery and Bascomb 1974). These bulk data were targeted to support views on textural differences between sandy Grubenhäuser fills, daub-rich fills, colluvial and blown sand and the prehistoric soil cover. Carbonate and organic matter, together with magnetic susceptibility, were measured in order to characterise more clearly occupation soils and fills in comparison with the natural deposits across the site so these could be identified. A series of samples also underwent non-quantitative elemental analysis through SEM/EDXRA. This was carried out to examine fills that may contain cess or industrial waste products from hearths, for example. Lastly, as a pilot study, the fine fraction (<500 µm) of four samples was analysed for loss on ignition (LOI), organic phosphate (Pº), total phosphate (Ptot), magnetic susceptibility (MS) and magnetic susceptibility after ignition (MS 550) at the Laboratory for Environmental Archaeology, Department of Archaeology, University of Umeå, Sweden (Table 4.12). This combined chemical and geophysical method, which employs standard techniques but on the fine fraction of the soil, was shown to differentiate clearly dwelling soils from manured fields (Engelmark and Linderholm 1996) and described at the 6th Nordic Conference on the Application of Scientific Methods in Archaeology at Esjberg, Denmark in 1993. Modern manured soils, for instance, are also clearly distinguishable from ancient soils (Linderholm, personal communication). The four samples selected for testing by this method were 10 (Table 4.8, farm mound), 55 (Table 4.11, farm mound), 102a (Table 4.11, animal trample over pebble surface) and 141 (Table 4.11, red soil layer in Grubenhaüs - ?industrial waste). This was carried out to test the chemical signatures of these major soil contexts at West Heslerton as a basis for selecting methods for the analytical stage and as a future predictive/survey method as it is most commonly used in Sweden.

At the analytical stage it is intended to integrate as much numerical and semi-numerical soil micromorphological data as possible with soil chemical and geophysical data. Both bulk samples and impregnated blocks, and in some cases the thin sections themselves will be used for microstratigraphic SEM/EDXR and microprobe analyses. Some aspects relating to animal stocking will be guided by the results of Ms Heathcote's current research both into ethnographic/experimental sites and archaeological sites such as West Heslerton. It can be noted that Macphail (1982) (Fisher and Macphail 1985) employed point counting of thin sections and iron and organic matter chemistry on the podzols of Site 1. The new work on the more calcareous sands of the Anglian settlement will employ methods appropriate both to the new site and to the techniques of the 1990s, as listed above and next.

4.4.4 Summary of results and discussion

Results from the 1986-1992 period of fieldwork and laboratory assessments and the findings from the field study of the 1995 excavated areas are presented in detail in Results 2.

These preliminary soil studies at West Heslerton need to be viewed in the context of the present expanding archaeological soil database. For example, analogue, experimental and collaborative soil microstratigraphic studies (soil micromorphology, physical, geophysical and chemical properties) have been permitting increasingly accurate identifications of many features typical of rural and occupation soils (e.g. buried soil taphonomy, Crowther et al. 1996; animal and human occupation, Matthews and Postgate 1994; Matthews 1995; Macphail and Goldberg 1995; Engelmark and Linderholm 1996; cultivation, Gebhardt 1992, 1995; see reviews in Macphail et al. 1990 and Macphail 1992; site formation and deformation processes, Gé et al. 1993; Macphail 1994b). The accuracy of many of the interpretations, that were first independently based upon microstratigraphic analyses, has also been tested successfully against complementary archaeological and environmental findings (Matthews and Postgate 1994; Crowther et al. 1996; Macphail 1990a; Macphail and Goldberg 1995; Macphail et al. in press). Moreover, the exact correlation of microstratigraphy as observed under the optical microscope and its quantification where appropriate, with physical, geophysical and chemical (both standard chemistry and SEM/EDXR and microprobe analyses) properties can greatly contribute to the confidence of interpretation (Fisher and Macphail 1985; Courty et al. 1989; Goldberg and Whitbread 1993; Canti 1995; Crowther et al. 1996). This approach has been increasingly utilised by the present author and is the principal component of Jennifer Heathcote's CASE (NERC/EH) research into animal effects on soils, both at experimental/ethnographic and archaeological sites such as West Heslerton (Macphail et al. 1996; Heathcote in preparation).

At West Heslerton, it is intended to develop this multifaceted approach, for example, by developing on the soil micromorphological and chemical techniques employed at Site 1, where grain size, organic matter and iron chemistry analyses were the appropriate techniques complementary to the quantitative soil micromorphological analysis of the site (Macphail 1982; Fisher and Macphail 1985; Powlesland et al. 1986). As shown below, a suite of microstratigraphic techniques appropriate to the 1990s, and its expanded database (see above), will be applied to West Heslerton to tackle a series of scientific objectives that both relate to specific archaeological questions (see below) and to the development of the discipline (soil science in archaeology) itself.

In the context of these related studies, the sampling strategy and proposed methodologies, as demonstrated by the assessment results at West Heslerton, can be viewed as likely to achieve the aims and objectives as outlined above.

To summarise:

1. A minimum number of prehistoric soil samples have been taken, some of which have been scanned, to demonstrate the potential of producing information on soil conditions prior to the Roman occupation on this south side of the A64. This is an absolute requirement to fulfil the objectives of understanding the environment and identifying transitional phases, especially between the Roman and Saxon periods. Equally, sufficient samples have been taken to characterise the Roman occupation in order to recognise the transformation of the settlement that marks the arrival of the Saxon culture.
2. Results from a small portion of the samples taken from the Saxon settlement show a very high potential for the study of animal stocking, animal movements, animal husbandry (e.g. pounds, droveways, farm mound, respectively), land-use organisation, domestic waste disposal and industry (cess pits, linear features, Grubenhäuser).

Thus:

• the typical deposits of this settlement be characterised to achieve the primary research objectives (2.2.1 Primary objectives)
• soil studies will be major contributors to achieving the primary archaeological objectives of recovering the settlement plan, permitting in addition large amounts of data to support investigations into the environment and economy of the settlement (2.2.1 Primary objectives and 7.1 Research Objectives). The lack of usable evidence from land snails and pollen makes the role of soil studies all the more crucial.

3. Lastly, the breadth and depth of soil sampling at this fully excavated and potentially well-understood site (in terms of its archaeology and environment), along with the methodologies available, provides an absolutely unique opportunity for geoarchaeological techniques to contribute to the reconstruction of a rural Saxon settlement. The study will also greatly advance the soil micromorphological and soil chemical disciplines in geoarchaeology, improving these as investigative, predictive and monitoring tools.

4.4.5 Detailed results

4.4.5.1 Sites 2 and 11 soil studies (1986-92)

Two broad topics have been under investigation. These are:

• the soil landscape development in relationship to pedogenesis, soil erosion and redeposition, as both background and response to human activities on the site.Investigations attempt to link the general inferred sequence of pedological processes active under human influence at Site 1 during prehistory through to the Anglian period, with on-site data from Site 2. Such processes that are being identified are: early Holocene decalcification and brown soil formation, Beaker and Bronze Age podzolisation, which continued into the early Anglian period; soil erosion, including aeolian soil loss and deposition, and colluviation from the Wold.
• the detailed examination of soils and anthropogenic deposits occurring as a result of on-site occupation. (In the context of the last, taphonomy and post-depositional reworking are important elements of the study.)

Most emphasis has been given to the examination of fills of Roman and Middle Saxon ditches, Middle Saxon hearths and the fills of Grubenhäuser (chosen from over 70). A major archaeological question being researched is the identification of Grubenhäuser use, how this varied across the site, and is there any way to identify a progression of usage through the Anglian occupation across the site? The biggest problem with this topic is taphonomic changes and post-depositional biological reworking of anthropogenic soils and materials, although they are considered archaeologically as well-sealed contexts (Powlesland, pers. comm.).

4.4.5.2 Site 12 soil studies (1995)

The 1995 soil studies developed the themes a) and b) outlined above, as highlighted by Powlesland (1994) from the full range of artefactual and ecofactual assessments, including soils. The sampling strategy also took cognisance of the AML geophysical surveys that had identified, for example, possible stock enclosures and the many archaeological features that were exposed throughout the long excavation period. The fieldwork strategy and sampling also took into consideration the increase of slope towards the Wold edge, the increased thickness of deposits in 'dry valley' gullies running upslope which entailed the understanding of superimposed stratigraphy. The site was visited and sampled by Macphail and Heathcote on four occasions, with the last few samples being taken by the excavators at the end of the dig. In addition, a major gully was studied by means of eight boreholes, running at right-angles to the slope (Canti and Heathcote, pers. comm.). During 1995, the following features came under special scrutiny:

1. The prehistoric to Saxon soil cover (erosion and colluviation). Soils dating to these periods were examined from the colluvial footslope to the edge of the Wolds and included Neolithic, Bronze Age, Roman and Saxon fills and buried soils, one example being the c. 3.5 m thick Roman colluvial deposits that pre-date the Roman temple. Sampling reflected results from the 1995 borehole data towards the top of the gully and McHugh's preliminary findings (in Macphail 1993) (see also 7.1.1.1 Re-Assessment of the prehistoric landscape).

2. The Roman-Saxon interface. The so-called 'turf hollow' (12AC) was of great importance because the soil here seems to have developed during this period (2.2.1 Primary objectives and 7.1.3 Transitional). Other features, such as soils developed over Roman pebble surfaces, were also targeted.

3. Four Grubenhäuser of special interest were examined; this work was selectively based upon findings from the large variety of these structures studied from Sites 2 and 11. This included the examination of a particularly large one, a particularly deep one and a one that had been excavated and sampled in great detail for complementary archaeological research (doctoral studies by Jess Tipper, University of Cambridge) (2.2.1 Primary objectives).

3. Features of stock management. Stock enclosure features, which had been predicted by the 1994 assessment, were of particular interest to Ms Jennifer Heathcote's (EH CASE/NERC) researches into the effects of domestic herbivores on soils. In addition, soils associated with a major pebble surface sloping towards the well and palaeochannel in Site 11 were targeted as likely drove-way deposits (6.5.1 Topography and landscape management, 6.5.4 Agriculture, economy, production and exchange and 7.1.4.2 Reconstruct the environment and economy of the settlement).

4. The 'farm mound' deposit; this had been tentatively identified in 1986-1992 (see below) and was a significant feature (as predicted) at the north end of the new site (12AA). Such a feature seemed to have arisen from the accumulation of domestic/industrial/stock management waste.

4.4.6 Results

These are dealt with in two parts. First (4.4.7), detailed (and briefly updated) results are given from the 1993 assessment and where applicable comments are added in italic on their relationship to the 1995 findings and sampling strategy. Second (4.4.9), the additional new approaches and findings of the 1995 work is given.

4.4.7 Results 1

Prehistoric to Roman periods

At Site 2, early soils were studied from firstly, a Neolithic tree-hollow feature (2CB 2069/2061) with burned soil (enhanced mag. sus. Table 4.10, samples 24 and 25; thin sections 29, 30, 31) and associated charcoal (identified?, radiocarbon dated?), and secondly, from the base of a Neolithic pit, partially sealed by a large piece of pot (pit 273, thin sections 27 and 28). The soils which are sandy loams (with 17% clay) to sands (8% clay), show evidence of the decalcified and partially decalcified nature of the Neolithic soil, and its brown soil status, although there are indications of possible tendencies of the soil to develop argillic fabrics. The probable process of tree-killing and tree-toppling disrupted the soil profile in the tree-hollow feature, bringing up chalk clasts from the subsoil. Similar features have been identified at other prehistoric burned tree-hole contexts (Macphail 1990a; Macphail and Goldberg 1990; Macphail 1992). In situ burning not only produced burned soil and charcoal, but even a residual patch of wood ash was found within a chalky (and thus calcareous) part of the disturbed soil. Interesting microfabrics are also visible in the Neolithic pit, and include burned daub/pot, charcoal, fine bone and ash, and amorphous organic substances.

It would be interesting to phase these Neolithic activities in relationship to Site 2, Site 1, and the surrounding area. Was the clearance activity which produced the burned tree-hollow in any way related to the soil erosion which produced the colluvium that buried a soil with a surface organic matter date of about 6690 years BP (M. McHugh, pers. comm.)? (A calculated age of burial would be about 5,300 years BP (Macphail 1987, 360).) At Site 1 there was evidence of Neolithic arable farming, whereas in the Late Neolithic there was in increase in activity across the Wolds, producing a number of pits across Site 1. These features have not yet been dated by radiocarbon assays (Powlesland et al. 1986).

In 1995, a possible Neolithic charcoal-rich buried loamy soil was sampled (Table 4.11, samples 106 and 107) in Area 12AA, the study of which may give clues on Neolithic activities, such as cultivation on the Wold edge (cf. nearby Kilham, Dimbleby and Evans 1974; Macphail et al. 1990). In Area 12AC a Bronze Age pit was sampled, and this may provide data on Bronze Age soil movement and the possible junction between the prehistoric soils of this area and the Saxon colluvium which overlies it (see also 4.8 Prehistoric Ceramics Assessment, 4.13 Lithics Assessment, 6.1 The Prehistoric Landscape and 7.1.1 Prehistoric).

Iron Age and Roman colluvium and ditchfills

At Area 11BD prehistoric colluvium (Sample 17) is very contaminated by Saxon material, as is the so-called Roman layer (Sample 16). Prehistoric and Roman colluvium thus could only be accurately characterised in Area 2CB. Here the prehistoric colluvium (Sample 33) is a loamy sand (Table 4.8 and Table 4.9, Sample 33), composed of typical iron-stained fine and medium sand, of original wind-blown origin (Macphail 1982). Its magnetic susceptibility is only weakly enhanced compared with the parent materials and modern topsoils on site (Table 4.8, Samples 11, 12, 13, 14, 15, 16, 29).

Generally, the Roman colluvial and ditch fill deposits differ little from the prehistoric sediments (Sample 32; Table 4.8 and Table 4.9, Sample 30), and are similarly well-sorted sands, but are rather more organic, calcareous and have a slightly higher magnetic susceptibility. Although similarly sterile, the Roman layer at Area 2CB, and the ditch fill at 2CE, do contain much more gravel-size chalk clasts, perhaps suggesting more active disturbance of the chalk substrate at times by tillage, than had occurred earlier.

Thus in summary these findings may possibly suggest that soils were coarsened in prehistory, through such processes as clay translocation (hastened by agriculture), podzolisation and aeolian and rill/sheet wash winnowing (agricultural effects) (Macphail 1982). At Area 2BB there is an indication that podzolisation was still active on these sands during the Anglian period (Macphail 1991). Closer to the colluvial footslope of the Wolds, Roman agricultural soils became slightly finer and more calcareous, and possibly more stable, allowing a slightly more humic soil to form.

In 1995 a Roman ditch fill was sampled from Area 12AA (Table 4.11, sample 43), and the huge quantities of colluvium examined at the top of the gully (samples 142-143; 144A-161A) may provide answers concerning Roman land-use and soil stability on the sloping Wold edge. It is intended to study some details of this latter material through two thin sections that target ephemeral (plough?) topsoils and through a series of loss on ignition and magnetic susceptibility assays. Such high-energy (stony) gully fan-like deposits have been successfully characterised by similar investigative methods before (e.g. Allen 1988; 1991; 1992). Boreholes (Canti and Heathcote, pers. comm.) found between 0.5-3.5 metres of deposits across the top of the dry valley in front of the Roman ?shrine.

Soil development between major occupation activities of the Romans and Saxons, which is one way to monitor 'continuity' (2.2.1 Primary objectives), is to be studied from a soil hollow (samples 71-4) and from soil formed over a Roman surface (sample 135). Experience with dark earth and dark earth-like soils has shown that the investigation of post-abandonment soils can be highly informative about their history (Macphail 1994b). With this in mind, the excavators collected reference samples of Roman building materials, so these too can be studied in this context. A similar strategy proved successful at the Courage's Brewery sites (Macphail 1994c, Macphail, draft report to MoLAS) and a study of mortar weathering from flint walls at Fishbourne Palace, West Sussex, has also been initiated (see also 6.2 The Changing Roman Landscape, 7.1.2 Roman and 7.1.3 Transitional).

Saxon colluvium (northern limit of 'farm mound')

A Saxon colluvial sequence was examined from Area 11BA, between 0.5 and 1.2m beneath the modern-day surface (Table 4.8, Samples 10, 11, 12 and 13). The lowest layer examined is a very organic loam, containing few coarse chalk fragments, abundant finely fragmented charcoal and amorphous organic matter, occasional slag (fused phytolith-rich material), and has very abundant textural features and juxtaposed areas of calcareous and non-calcareous soil. The soil itself may be weakly phosphatic, based on the observation that it is weakly fluorescent. Some of the amorphous organic matter is of a character typical of highly humified herbivore dung/stabling material (Macphail and Goldberg 1995), which is suggestive of either intensive manuring or a dung heap or stock concentrations; the textural features being a result of animal trample. Perhaps the deposit is colluvial only in the broadest sense of the term.

Above (Sample 12), the loamy colluvium is quite different, being poorly humic, totally calcareous and rich in rounded chalk gravel. It also contains few burned soil fragments, hammerscale?, rare fine charcoal and some of the chalk looks a little heated - discoloured, fissured. In the fine fabric, which is very thin, calcite crystals of wood ash may be present in addition to fine fossils liberated from the weathered chalk. Thus this calcareous deposit looks as though it could be a chalky wood ash residue, possibly in part from some kind of industrial activity, such as smithing. The chalk does not look so heavily burned that lime production can be envisaged. It is more likely that some chalk became accidentally used during such activities, whereas most was probably incorporated later by ploughing or 'kicked-in' by animal trample.

The next layer above (Sample 11) is similar in that it is a calcareous loamy colluvium, containing chalk stones and gravel-size material. It has weakly formed intercalatory textural features, further indicating that it is a ploughsoil colluvium, the high amounts of chalk suggesting the formation of a chalk gravel fan through a high energy erosional episode (Allen 1992). In addition, the fine fabric contains very abundant fine charcoal and heavily burned amorphous organic matter, amorphous organic staining, probable grass ash residues, some phytoliths and diatoms. All these last inclusions indicate that the ploughsoil could have been very heavily manured with burned herbivore stabling material (mainly of a grass origin?) (Macphail and Goldberg 1995).

Above, the next layer (Sample 10) appears to be more of a dumped deposit than a colluvium, because it retains layering. It also contains very little non-calcareous soil material (i.e. very little quartz sand). In fact, it is primarily made up of ash layers. Some of the layers are of pure wood ash, whereas others also contain high amounts of spherulites and calcium oxalates, which suggest the burning of large quantities of leaves/sheep dung (e.g. Brochier 1983; Courty et al. 1992). As these phosphate-rich (UV fluorescent) layers contain sheep/goat coprolites, possibly composed of woody and leafy residues, these layers indicate the dumping of burned stabling residues. Macrobotanical evidence from waterlogged sites (Robinson and Rasmussen 1989) and soil micromorphological data from well protected environments, such as caves (Courty et al. 1992), have demonstrated the use of leaf hay as a stabling fodder as early as the Neolithic. This unique material from the colluvial deposits at 11BA may indicate that the Saxon people were stabling sheep/goat probably nearby. (Coprolites occur rarely in grub fills, see below.) As the deposit also contains possible human coprolitic material, this farm mound deposit probably acted as a midden heap too. It may also be noted that the deposit was weathered, and there is secondary calcium carbonate formation resulting from decalcification of upper deposits. The layer was also affected by very thick dusty clay soil inwash, presumably dating to the period when the ash midden was buried by later, more mineralogenic ploughsoil colluvium.

The discovery of this sequence of manured colluvial ploughsoil, mixed ash residues, chalky colluvium and, lastly, an ash midden resulting from the dumping of burned herbivore stabling deposits, is very important, and as far as the author knows, quite unique. The deposits are in fact the best stratified archaeological sediments on site, and their proper dating, phasing and comprehension is probably crucial to the full study of the Saxon occupation at West Heslerton.

Buried soils, probable colluvial occupation deposits and probable 'farm mound' deposits (and associated ditch and pit fills) that were presumed to mirror those same deposits characterised from Area 11BD, were examined in Area 12AA. At five locations this feature, broadly termed 'farm mound', was sampled (Table 4.11, Samples 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55 and 56) (4.4.9). Macrobotanical studies by Carruthers (in Powlesland 1994) suggest that crops were mainly grown towards the fen edge. The high amount of post-Saxon colluvium burying the 'farm mound' and resulting from ?medieval to modern ploughing should not therefore too strongly encourage the view that the 'farm mound' results from an important arable colluvial component if seed recovery is indicating Saxon cultivation at the fen edge. The possibly dung-rich deposits found at its base in Area 11BD may have features of animal trample (Beckman and Smith 1974), and animal activity as well as dumping and cultivation colluviation has to be considered as a mechanism for the accretion of the 'farm mound' and its chalk stone component, i.e. trampling rather than cultivation may be the major mechanism behind colluviation. Saxon soil accretion/formation, sensu stricto, is thus being studied from samples taken from above a pebble surface in Area 12AD (Table 4.11, Samples 108 and 109) for this very reason (see also 6.5.2 Anglo-Saxon settlement evolution and decline, 6.5.3 Reconstructing the environment, 6.5.4 Agriculture, economy, production and exchange and 7.1.4 Anglo-Saxon).

Area 2CB

Three fills of the Grubenhäuser were studied from this area just to the west of the area of rectangular houses. These are 2CB3 (Samples 39, 40 and 41), 2CBl261 (Samples 34, 35 and 36) and 2CB1324 (Sample 37). Bulk analyses show that these fills are more calcareous, much finer in texture (sandy clay loams), and are a little more organic and have much higher magnetic susceptibility readings (Table 4.9 and Table 4.10, Samples 4, 5, 6, 27 and 28) compared with the natural soils and the prehistoric and Roman soils. The micromorphology of the seven thin sections shows a variety of inclusions that help explain the bulk analyses. Firstly, many burned organic residues, e.g. charcoal, ash, bone, burned soil and burned daub, slag and burned chalk are present. Non-sandy materials that occur are:

• clay loomweight fragments
• daub made from argillic brown soils (from upper slope of Wolds)
• daub manufactured from a mixture of chalk soil and organic matter, possibly including dung.

Some of these have been weathered and broken down biologically within the grub fill. Other major inclusions are bone and probable human coprolites. The phosphatisation of some chalk clasts (UV fluorescent margins etc.) suggest staining either by human or animal liquid waste (Macphail and Goldberg 1995). This staining may relate to dumping within the sunken feature, or be residual of chalk soil floors from animal or human occupations. There are also indications that where no calcareous material is present, non-UV fluorescent phosphate compounds occur which form more typically in a rather acidic and ferruginous soil environment. Soil acidity is also indicated by the partial decalcification of Arionid (slug) granules and the weathering of chalky daub.

In Grubenhaüs 2CB 1261 in particular, the abundance of phytoliths can be noted throughout. The bottom of the Grubenhaüs was infilled with calcareous daub and organic debris, some presumably from cereal processing, but some phytolith types (dumbells) suggest other plant types were present (wetter ground types?). This rather organic deposit broke down and was reworked by earthworms and slugs, the weathering of the calcareous daub resembling the similar breakdown of insubstantial building materials in dark earth sites (Macphail 1994b). Some of the weathered calcareous daub/soil still retains pseudomorphic void space relic of original plant temper. Coating features indicate some wetness, probably due to liquid waste being thrown in, as this gave rise to yellowish ferruginous staining. Upwards in the fill, phytoliths remain in high numbers, although mineral matter such as sand is also in increasingly higher quantities. Towards the top of the fill, the presence of phosphatic burned cereal waste, many bone and probably human coprolitic remains, confirm the theory that much midden waste was being dumped. Rare herbivore coprolitic material was also being included, as well as ash waste, some calcite ash still persisting. Magnetic susceptibility readings and oblique incident light observation clearly show that increasing amounts of fine burned soil were included in the uppermost fill. The weathering of the fill as a whole has given rise to secondary calcium carbonate formation too, e.g. around root channels.

There is therefore evidence from the Grubenhäuser in Area 2CB that the original fills could have been made up of unburned organic waste, e.g. from cereal and other plant storage, and the decay of building materials, including calcareous daub. Later in the fill's history, general midden dumping of domestic refuse - ash, cereal processing waste, bone and coprolites - occurred.

Area 2CA

Samples from 2CA60 (Sample 38) and 2CA70 (Sample 42) were studied alongside a series of bulk samples from 2CA60. Bulk analyses showed the fill to be a rather homogeneous, weakly humic, calcareous sandy loam, with increasingly very high magnetic susceptibility readings towards the top of the fill (Table 4.9 and Table 4.10; Samples 1, 2 and 3). The last indicate the increasing inclusion of highly burned materials into the grub fill. In thin section, the sands are shown to include many fragments of daub, which include burned soil and charcoal. Again, chalk clasts have often been weakly phosphatised. Pieces of loom weight, sometimes made up of 'washed' (alluvium?) fine silt, are interesting because planar voids within them are relic of plant temper. In addition, weathering cracks within them may be coated by very dusty fine soil containing charcoal, suggesting the surface disturbance (ploughing?) of soils affected by burning. The Grubenhäuser in this area may have had the same kind of history as those in Area 2CB.

Area 2DA

Here a number of Grubenhäuser, 2DA84 (Sample 23), 2D-125 (Samples 24 and 25) and 2DA186 (Sample 26), a malt kiln (Samples 18, 19 and 20) and a metallurgy hearth (Sample 22), were examined. The grub fills are very dominantly sandy, non-calcareous, with a meagre fine fraction, in part relic of possible liquid waste dumping, and some dominated by the input of fine burned organic matter. These Grubenhäuser also contained fewer chalk clasts than were found in those of Areas 2CA and 2CB. In addition to the usual inclusions of bone, daub (soil and organic matter mixture) and probable coprolites, some fragments of various types of slag occur. High temperature homogeneous iron slag and vesicular silica slag can be compared with more heterogeneous types of slaggy fragments. A possible piece of burned horn occurs too.

The malt kiln, as expected, is characterised by fragments of burned argillic silt loam daub soils, high amounts of phytolith-rich burned cereal waste, and even a small quantity of surviving ash. Likewise, the metallurgy hearth is characterised by a very high amount of burned 'red chalk' and 'red' chalky soil daub, and some weakly cemented phosphatic wood ash remains and few fine sand-size round iron slag.

Again, all these features have undergone a long period of weathering and biological activity, but they show that this area was an industrial one, where dominant natural sand blowing into the Grubenhäuser was accompanied by inputs of local burned fine soil, slag and burned organic debris. The latter were in part the result of such activities as metal-working and malting.

Little can be said about the original use/fill of the Grubenhäuser, but it seems likely that infilling was probably rapid due to sand blowing. A few residual materials, from metallurgy, malting and other industrial activities, became included during this period of infilling.

Area 11CE

A so-called 'furnace associated ditch fill' (11CE532) and possible cess-filled grub (11CE529; SEM/EDXRA) were investigated from this area for this assessment. The 'furnace associated ditch' was infilled by sandy soils containing much heavily burned cereal matter rich in phytoliths. Some of the cereal waste was still phosphatic, and suggests these organic residues derive from low-temperature cereal-matter fires. These light materials would have easily blown in, together with sand, and the blowing in of fine burned soil material.

SEM/EDXRA analyses of grub 11CE529 shows the presence of Ca, K, and P, in addition to the expected high amounts of Al, Si and Fe. As well as the possibility of cess being dumped, these data suggest an important ash input (Wattez and Courty 1987).

Here ash was probably being dumped into Grubenhäuser, whereas the 'furnace associated ditch fill' was probably infilling at the time that the local furnace was in use. In Area 2DA, although industrial remains became incorporated into Grubenhaüs fills, these were highly secondary to infilling by natural sand blowing.

Area 11CD

Two dark reddish ditch fills were studied, 11CD/5115 (Sample 5) and 11CD/5112 (Sample 9), the last probably dating to the Middle Saxon period (Powlesland, pers. comm.). Both contain abundant fine sand and many chalk clasts, although the fragments tend to be coarser in Sample 9. They contain an important quantity of fine material in addition to this sand and chalk, and thus differ from fills noted in Area 2DA. Most important is the type of fine fabric which they share. These reddish (see SEM sample WH 55) fills are made up of highly burned organic matter/sesquioxidic(?) reddened soil, phytoliths and high amounts of residual ash. Although some wood charcoal is present, most of the ash seems derived from Gramineae material. Close scrutiny of the charred organic matter show that it is highly humified, and reddened probable herbivore excrement was noted. This suggests that the fuel being burned was possible dung. There was also a burned fragment of possible horn. Again, some of the chalk appears to have been weakly phosphatised, perhaps when it was part of the stable floor or because the fill is generally phosphatic. It seems likely that local burning of dung, possibly only at low and medium temperatures (no vesicular ash or slag found) was carried out, and this material was blown into local ditches, alongside local sand and relics of soil fauna such as Arionid (slug) granules.

As in Area 11CE, these ditch fills seem strongly contemporary with industrial activities.

Area 11BD

SEM/EDXRA analyses were carried out on selected grub-fill material (sampled by M. McHugh) from this area. A reddish brown ash(?) grubfill deposit (WH 55) contains high amounts of Al, Si, Ca and Fe, with sometimes very high Fe. These results suggest highly burned ash residues (Si and Ca), background sesquioxides (Fe and Al) and the presence of aluminium silicates (clay). This interpretation is consistent with the soil micromorphology of thin section 9 examined in adjacent Area 11CD (see above). Here it is worth making the general comment that oak ash, for example, also contains Al, and cereal ash is high in Si (Wattez and Courty 1987, 678). The peak of iron may possibly suggest ironworking, although iron slag appeared to be infrequent.

Sample WH 56, a grey ash layer within a Grubenhaüs, appears to be a low-temperature burned ash. Al and Si (clay), manganese (Mn) and iron (Fe) are probably at background soil levels.

Sample WH 59, is another grey ash(?) Grubenhaüs fill, associated with charcoal. Again, it is probably a mixed wood and cereal ash (Ca) residue, with residual amounts of magnesium (Mg) and phosphorus (P).

Samples WH 56 and WH 59 appear to be typical ash residues, whereas sample WH 55 could well be a much more highly burned ash residue (see Sample 9), relating to some industrial process such as ironworking.

4.4.8 Analysis of soils from Grubenhäuser at West Heslerton, North Yorkshire

Excavation of the Early and Middle Saxon site of West Heslerton has found a large area of rectangular buildings and across the rest of the site some 70 Grubenhäuser occur. Detailed soil and soil micromorphological investigations have been carried out on nine of these to help characterise different uses of space across the site.

Basic grain size analyses show that the fill of the Grubenhäuser is finer than the local blown sand soils of Saxon date. Other analyses indicate that the fills contain little organic matter now, but may be more calcareous than the surrounding soils. Magnetic susceptibility can be low or strongly enhanced.

Soil micromorphology permitted a more detailed comprehension of the grub fills and how these differed across the site. In Areas 2CA and 2CB, fills seem to reflect the close proximity of domestic rectangular dwellings (Area 2CC). The basal fills (see 4.4.9) indicated that chalky building daub, probably used in the construction of the building, and organic debris from a thatched roof or stored cereal material, comprised the first infill material. This was biologically worked in situ by slugs, earthworms etc., just as in a tree hollow. The upper fills reflect greater amounts of domestic dumping - cereal processing waste, burned soil from hearth sweepings and cess such as coprolites - liquid waste soaking into the lowest deposits, and causing some phosphatisation of chalky material.

In contrast, in Area 2DA and Site 11, Grubenhäuser and ditches were very much more rapidly infilled with blown sand, and background material from industrial activities such as malting and ironworking.

Principal inclusions found within Grubenhäuser at West Heslerton were:

• Ubiquitous material in addition to blown sand
• Biologically worked soil
• Fine charcoal
• Phytoliths
• Chalky daub
• Decalcified soil daub
• Coprolitic material
• Phosphate stained chalk
• Slug (arionid) granules

Material specific to Grubenhäuser in Areas 2CA and 2CB (Grubenhäuser possibly used as grain stores and later for domestic waste disposal).

Upper and middle fills:

• Abundant phytoliths
• Gramineae ash residues
• Gramineae charcoal
• bone
• human coprolites
• fine burned soil
• wood charcoal
• weathering chalk daub
• rare sheep/goat coprolites
• biological soil fabrics
• secondary calcium carbonate root pseudomorphs

Lowest fill:

• Abundant phytoliths
• organic fragments and weathering chalky daub
• earthworm soil fabrics and yellow void coatings

Material specific to Grubenhäuser and ditches in Area 2DA and on Site 11 (industrial area; malt kilns, metallurgy hearths)

• Dominant blown sand content
• Iron and other slag
• Strongly burned soil
• Wood and Gramineae charcoal
• Heavily burned/ashed ?dung
• Phytoliths and vitrified ash
• Burned bone and ?horn
• Human coprolites

The Grubenhäuser assessed in 1993 show evidence of a tripartite fill (discussed in Powlesland 1994 as relating to primary use and construction, secondary use and later infilling) and like some of the Saxon ditch fills were useful in supporting the theory of zones of activity in Sites 2 and 11. Sampling in Site 12 examined in detail the fills of two structures, namely 12AD5009 (Table 4.11, Samples 114, 115, 116) and 12AC107 (samples 121, 122, 123, 124, 125, 163 and 164) to test further, after full phasing and finds analysis, the tripartite fill theory and the use of these structures as grain stores and/or possibly small workshops. In addition, fills from a particularly deep example were examined (Table 4.11, 126, 127, 128 and 129) alongside burned fills from others (samples 137, 140 and 141) and pits (samples 111, 112 and 113). These last resemble industrial red soil waste found in Area 2DA and Site 11, suggesting industrial activity in Area 12AD. Various hearths and kilns were examined in 1986-1992 and to complement the study of these a sample was taken from the roof of an on-site experimental pottery kiln (Table 4.11, sample 117). The study of these types of features will contribute to the zoning of the whole Saxon settlement. Possible cess inputs into Grubenhäuser was identified in Sites 2 and 11, and a feature termed a latrine in Area 12AA (Table 4.11, samples 85 and 86) was additionally sampled. A recent assessment of a latrine pit at Monkton, Kent, found interesting coprolitic remains rich in bran and fine bone (associated pollen, nematode egg, macrofossil recovery; Macphail 1995). At the same Kent site, Romano-British Grubenhäuser contained strong evidence for the use of turf as a constructional material.

4.4.9 Results 2

The major new study theme for soil investigations of Site 12 that developed from the 1993 assessment is related to domestic animal stocking activities. Geophysical survey found a number of possible stock enclosures, whereas the assessment of thin sections 10, 11, 12, 13 (Area 11BA) suggested the input of herbivore manure into accreting soil deposits at the north end of Area 12AA. At the base of the sequence (4.4.7) manure was mixed with domestic and possible industrial waste, while the upper part was ash-rich containing evidence of dumped burned herbivore dung (including sheep/goat coprolites). The buried soils of the stock enclosures and their associated features, such as entrance ways, were sampled by Ms Heathcote as part of her CASE doctoral studies at the Institute of Archaeology, UCL (Recognition of the effects of intense vertebrate activity - structural, geochemical and mineralogical signatures; Table 4.11, samples 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 80, 81, 82, 83, 84, 92, 93, 94, 95, 96, 130, 131, 132, 133, 134, 138 and 139). Preliminary phosphate studies showed a pattern of phosphate enhancement across the site. It is also probable that samples taken from soil associated with the pebble surface (?drove-way) may also be included in her studies (Table 4.11, samples 102, 103, 104, 105, 136, 162). Selective soil micromorphology, phosphate analysis and SEM/EDXRA/microprobe analyses are to be carried out by Ms Heathcote on vertically and laterally taken samples to provide archaeological analogue data for her research. She is also studying other archaeological sites, experiments and ethnoarchaeological examples. Although her researches will greatly aid the present study of West Heslerton they are not designed per se to be an integral part of the West Heslerton project, but rather must in the first instance support her doctoral remit.

The 'farm mound', which in part may develop from soils formed in stock enclosures through trampling, is apparently stratified. This was sampled from five locations to give both a lateral and vertical understanding of its formation and composition (Table 4.11, samples 45, 46, 47, 48, 49, 52, 53, 54, 55, 56). In addition, Dr David Weir (AML) took complementary undisturbed Kubiena and column samples for pollen analysis and the excavators took associated bulk samples for macrobotanical remains recovery (see 4.7 Pollen). As stated earlier, the presence of ashed herbivore dung remains as seen in this 'farm mound' is rare in this country, and their investigation is likely to aid the reconstruction of Saxon animal husbandry. Detailed studies of the ashed dung components (calcium oxalates, spherulites, ash, charred organic remains etc.) may be able to provide new information on grazing/foddering/stabling regimes in a way so far only possible on well-protected (prehistoric cave) continental sites (Brochier 1983; Brochier et al. 1992; Courty et al. 1992). In these it has also been possible to identify (with animal bones, pollen and macrobotanical remains) seasonality, use of space, and other elements of animal husbandry (Macphail and Goldberg 1995; Macphail et al. 1996; Maggi 1997). This highly unusual archaeological deposit also deserves specific studies of the spherulite component, and this is to be undertaken by Matthew Canti (AML) who has researched their presence in modern dung (Canti 1997).

The 1995 sampling also increased the number from Grubenhäuser, pits (and a latrine and selected coprolites) and soil examples (4.4.7) to widen the geographical/settlement component study. Added to the data from the animal bones, structures, stock enclosure soils etc., these microstratigraphic investigations offer a great deal to the better understanding of Saxon rural life. This is in addition to the full series of questions being asked of the site (4.4.13).

4.4.10 Chemical and geophysical signatures

Ash and spherulite-rich farm mound material identified in thin section (sample 10) has a similar chemical and geophysical signature as farm mound material located in the field (sample 55) (Table 4.11 and Figure 4.6). Equally, supposed animal trampled soil overlying a pebble surface (sample 102a) and a red layer within a Grubenhaüs (sample 141) have different signatures, as expected (Figure 4.6). The red Grubenhaüs fill is of interest because not only has it a high MS, but also contains plenty of material still to be enhanced by ignition. This perhaps indicates industrial raw materials being present alongside burned soil, etc. All samples have Ptot/Pº ratios over 1.0 suggesting the presence of manure in all these samples, which is consistent with the working interpretation of the site soil's make-up (Engelmark and Linderholm 1996; Linderholm, pers. comm.).

Fig. 4.6 Soil micromorphology: Chemical signatures from 'farm mound' and daub enriched deposits

4.4.11 Comparison with other sites

These studies on Grubenhäuser are within a wider research framework. At a workshop on Sunken Featured Buildings (Norwich, April 23rd 1993), at which soil data from West Heslerton was presented, the shortage of knowledge concerning the use and history of Grubenhäuser across the country was highlighted. At present the author is involved in the study of Saxon Grubenhäuser from Bedfordshire and Norwich, and as noted earlier (4.4.7) a variety of fills of Romano-British Grubenhäuser at Monkton, Kent, have recently come under scrutiny. The working group on Archaeological Soil Micromorphology has produced discussions on a number of themes relevant to West Heslerton. For example, colleagues in Basel, Switzerland, discussed very well-preserved phytolith-rich deposits at the base of a 'Saxon grub'. The author is being kept up to date with their findings at this and other related sites.

Other sites of Middle Saxon age have been under study from London (Jubilee Hall, Shorts Gardens, Bruce House) and podzols around West Stow, Suffolk, were also investigated. The processes affecting the transformation of rural buildings and occupation and associated debris, and how the evidence of these remains in the soil, is well understood from studies of the dark earth (Macphail 1994b). In fact, it is likely that continued work on the West Heslerton material will enhance our understanding of (late Saxon and Norman) dark earth-like sequences, such as found at Greyfriars, Norwich, and Pot Row, Grimston, Norfolk, and new London sites such as Guildhall Yard East. In addition, international experimental research data (Courty et al. 1992; Macphail and Goldberg 1995) has helped in the study of soil micromorphological fabrics and chemistry (e.g. on chalky colluvial soils at Butser Ancient Farm, Hampshire) which relate to rural occupation, cultivation and the presence of animals. The inclusion of the CASE student Jennifer Heathcote in the West Heslerton project has already been discussed. EH support for her CASE award was partially based upon the potential of her contribution to the understanding of West Heslerton. It is also worth noting background research into erosion and cultivation colluvium on downland sites (e.g. Allen 1992; Bell and Boardman 1992) and ongoing research on the breakdown of Roman constructional materials (Macphail 1994b; Fishbourne Palace reference studies etc.). The continuity of Roman through to Saxon/medieval occupation has been studied from a number of dark earth sites by the author, and for example new work is being carried out at Pevensey Castle, East Sussex, on this very topic.

4.4.12 Potential for analysis

Quality of material:

Long experience with the West Heslerton project has shown that although some features, such as ditch and Grubenhaüs fills, can be highly diluted with sand, included anthropogenic materials can be identified and characterised. In fact, the very amount of sand can be quantified and has led to preliminary interpretations on how features infilled. Findings from 1986-1992 also permitted the more accurate targeting of features to be sampled, and encouraged in many cases the taking of vertical and lateral sampling series to allow better interpretations about fills, use of space and comparability. There are plenty of samples to permit specialised work on SEM/EDXRA, microprobe, additional bulk sample analysis of spherulites (by Canti), organic matter, magnetic susceptibility and phosphate (with Heathcote; see Table 4.11 and Figure 4.6) to complement optical microstratigraphic investigations. In many cases, special attention was paid in order to ensure complementary bulk samples were taken for finds, macrobotanical and bone recovery, which will act as independent controls and sources of information. Pollen sampling was specifically co-ordinated with soil studies of the 'farm mound'. Another avenue of study will be the characterisation of selected coprolites because these, besides preserving pollen, nematode eggs and other microfossils, may provide data on ancient diet and lifestyle. Most of the 1986-1992 thin sections have been made, whereas over 50% of the new Kubiena samples have been conserved through resin impregnation.

There is also the added benefit of such a potentially wide data-set (Table 4.8 and Table 4.11) enabling a whole range of archaeological and related environmental questions to be tackled (4.4.13).

4.4.13 Conclusions

Details of the conclusions of the 1986-1992 assessment are presented in Macphail 1993. It is quite clear from those results and the field evaluation (and sampling) of the 1995 excavation of Site 12, that there is a wide range of soil topics to be studied that will greatly contribute to the understanding of the history, landscape context and agrarian economy of the occupation at West Heslerton. The work may help in answering a large number of questions, such as:

• What kinds of soils were present in prehistory from the edge of the Wold to the fen edge?
• What was the landscape like when the Romans arrived and how did they transform it?
• How can the period between the major Roman and Saxon occupations be characterised through the soil evidence?
• How can the Saxon settlement plan be reconstructed - domestic areas, storage areas, industrial areas, stock areas, 'farm mound', routes - and phased alongside the archaeology?
• What are the mechanics of Anglo-Saxon farming and what role does the settlement perform in managing the agrarian resource?
• is production purely a function of the hinterland or does the space in the settlement play a role?
• To what degree does the soil evidence support the plant macrofossil indications of cropping on the sandy soils towards the north of the site?
• Is there any evidence of horticulture/in field farming?
• Can soil studies add some details on animal husbandry - stocking, overwintering practices etc. - to this debate which otherwise would be based purely upon faunal remains and archaeological reconstruction?

All the above can be attempted through the study of the soils, which on its own will characterise the nature of the archaeological deposits - one clear possible result being the better comprehension of the function/s of Grubenhäuser.

Tables

© Internet Archaeology URL: http://intarch.ac.uk/journal/issue5/westhes/4-4.htm
Last updated: Tue Dec 15 1998