WILTSHIRE HERITAGE STUDIES

The Wiltshire Archaeological

and Natural History Magazine

Volume 93 2000

77 = ieee eee 7 Pa : ; o, aa) a . } an at i i ty : a ; at i _ re ; wo = { 7 + hs = ee fc 7 : Me = 7 ‘ = - 7 ' : ¢ a oe Ko ns < zi ; . : =a fA a i : q 7 z 7 7" + Ti), S eel ee + n 7 . a) ia heey F = , we aX a 3 4 ’ eee . *, \ a = : : i a ne - 7 7 f 7 + - 7 a my 3 i 1 4 a ' ' . 7 7 : nl = ; ' : ae ‘ re : : i .'? a : : 1 a ’ 7 “ 1 r ror Z 7 = i ' : ' : : a a t ; 7 “ry > ge 7 — - f : _ F - - : - $ : - oan ‘ , A ; = 7 : 7 a od fi aT ee : i i : ' _ —_— oO | 1 ‘ 7 a ‘ ' ‘ 7 - 3 es ‘ D : = : { ‘ My -., ; ' ow ' i : fan ' ; bo, oF i = ee 5 w : a x sy ' 7 ‘

ote

- - , “3 a i ee - 7 aan - "a - “ ; o 4 f . ; f ; 3 : A ; i : : i : P ' "4 : tC i j ay F m f i C 7 “y ‘ F i 4 a hee 7 —— . u ‘ - i 7 2 * a pes eo =| - _ yey ee ne 7 : : oe 7 2? a -- : « : =. : } +

= es ie : sO woah 2 ie arr “ + ; 2 a 2 ; eh: : _ oo we ow i oo 7 : # : Deas - : _— 4 a } — : a (ok oa ! Se 7 ee 4 e c _— = - Z ei : ' ‘ ‘hes , a rs 7 ¥ = % { = — . - = ‘ ‘ ~ , -* 7 + ‘ ' ‘ ‘ : . ’ +. = i t 7 \ i my ° 1 7 - re a a r 7 7 7 zs 7 oe f : - | Ls : oe - 7 a aie = 7 a a ‘ a - : - 7 , 7 a nN os i 5 ans a bo po - ee , : ae - = ' - _~ i} = = - a ‘ 7 7 i e =~ *. dng ' F 7 i - 7 a i - —— rc - - : 7 7 € a i = eS a = = ' a = - “<> 4 o~ : ‘ 7 q i = Sots ’ . 7 ' © = 3 oA 2 of ef ; ‘ r cf ‘ S ; 1! ¥ r : 1 ‘ip 2) oa : a " ‘ 7 & _ : = a, f : ; : ote toa i ‘ ‘ a 2 1 1 - - a 74 y 7 a q 7 se - : ‘ = “

The Wiltshire Archaeological and Natural History Magazine Volume 93

2000

NORMAN DAVEY. This volume is dedicated to one of our Vice-Presidents, Dr Norman Davey, who celebrated his hundredth birthday 1n January this year, and who has had a long and distinguished career as a civil engineer and amateur archaeologist. Born and educated in London, he served in the Royal Flying Corps towards the end of World War I and later became an associate fellow of the Royal Aeronautical Society. For many years he worked at the Buildings Research Establishment, and designed the model dam on which the bouncing bomb was tested by Barnes Wallis. On his retirement, while continuing to work for the BRE, he devised new methods of restoring fallen Roman wall and ceiling plaster, using expanded aluminium mesh; he has also written several books on Roman wall plaster and the history of building materials. For many years associated with the St Albans and Hertfordshire Archaeological Society, of which he 1s Vice-President, he worked with Tessa and Mortimer Wheeler on the Verulamium excavations in the early 1930s, making detailed drawings of the tessellated floors.

Published by

Wiltshire Archaeological and Natural History Society 41 Long Street,

Devizes, Wilts. SN10 1NS

~ Telephone 01380 727369

Fax 01380 722150

Registered Charity Commission No. 309534 V.A.T. No. 140 2791 91

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE VOLUME 93 (2000)

ISSN 0262 6608

© Wiltshire Archaeological and Natural History Society and authors 2000

Hon. Editors: Joshua Pollard, BA, PhD, and John Chandler BA, PhD.

Hon. Local History Editor: James Thomas, BA, PhD, FRHistS.

Hon. Natural History Editor: Michael Darby, PhD, FSA, AMA, FRES, FRGS Hon. Reviews Editor: Michael Marshman, ALA.

Editorial Assistant: Lorna Haycock, BA, Dip.ELH, Cert Ed.

We acknowledge with thanks publication grants for this volume from the following bodies: Salisbury District Council, for the paper, ‘Excavations at Ivy Street and Brown Street, Salisbury, 1994’, by Mick Rawlings; the Trustees of St John’s Hospital, Wilton, the Trustees of the Matrons’ College, Salisbury, and Dr. A. J. Hall, for the paper, ‘Excavations in Wilton, 1995-6: St. John’s Hospital and South Street’, by Phil Andrews et al.; ‘Transco, for the paper, ‘Excavations along the Littleton Drew to Chippenham Gas Pipeline’, by Clifford Bateman; Wiltshire County Council, for the note, ‘Excavations of Bronze Age and Romano-British sites along the Chippenham Western Bypass A4 to A350 Link’, by Clifford Bateman and Dawn Enright; and Lovell Partnerships (Southern) Ltd, for the note, ‘Excavations at Vale’s Lane, Devizes, 1996-7, by Phil Andrews and Lorraine Mepham.

The journals issued to volume 69 as parts of The Wiltshire Archaeological and Natural History Magazine (Part A Natural History; Part B Archaeology and Local History) were from volumes 70 to 75 published under separate titles as The Wiltshire Natural History Magazine and The Wiltshire Archaeological Magazine. With volume 76 the magazine reverted to its combined form and title.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the Society and authors.

Typeset in Plantin by John Chandler

and produced for the Society by

Avonset, 11 Kelso Place, Upper Bristol Road, Bath BA1 3AU Printed in Great Britain

=~,

-2 MAR 205i

| GENERAL LIBRARY |

SSE EN

Contents _GENEF

[THE NATURAL | HISTORY MUSEUM i

i f

The Beckhampton Avenue and a ‘new’ Neolithic enclosure near Avebury: an interim report on the 1999 excavations, by Mark Gillings, Joshua Pollard and David Wheatley

‘A Very Pretty Seat’: Erlestoke Park, 1780-1999, by Isabel Ide

Excavations at Ivy Street and Brown Street, Salisbury, 1994, by Mick Rawlings, with contributions by Michael J. Allen, John Chandler, P. Hinton, S. Hamilton-Dyer, Emma Loader, Lorraine Mepham and Sarah F. Wyles

The Great Bustard in Wiltshire: Flight into Extinction? by James Thomas

Beaker Pits at Crescent Copse, near Shrewton, Wiltshire, and the Effects of Arboreal Fungi on Archaeological Remains, by Michael Heaton and Rosamund M J Cleal, with contributions by Peter Higgins, Peter Bellamy, Sheila Hamilton-Dyer and John Wilson

A ‘Perillous, Covetous Man’: the career of Thomas Tropenell Esq. (c. 1405-88), a Wiltshire lawyer, Parliamentary burgess and builder of Great Chalfield, by J. T? Driver

Excavations along the Littleton Drew to Chippenham Gas Pipeline, by Clifford Bateman, with contributions by Fiona Roe, Jane Timby and Tracey Stickler

Early Castles in the Medieval Landscape of Wiltshire, by Oliver H Creighton

Looking for Dr Ingen Housz: The evidence for the site and nature of the burial, in Calne, of the famous Dutch physician and scientist of the eighteenth century, by Norman and Elaine Beale

Neolithic activity and occupation outside Windmill Hill causewayed enclosure, Wiltshire: survey and excavation 1992-93, by Alasdair Whittle, Jessica J. Davies, Ian Dennis, Andrew S. Fairbairn and Michael A. Hamilton, with a contribution by Joshua Pollard

Excavations in Wilton, 1995-6: St John’s Hospital and South Street, by Phil Andrews, Lorraine Mepham and Rachael Seager Smith, with contributions by Michael J. Allen, Sheila Hamilton-Dyer and Pat Hinton; illustrations by S. E. James

Names on the Path to Remembrance: the building of Marlborough College Memorial Hall, by Brian Edwards

- Investigation of a Roman villa site at Euridge Manor Farm, Colerne, by Larry Luckett, with contributions by D F Mackreth, T S N Moorhead, Judith Roseaman and Bryn Walters

Notes Excavations of Bronze Age and Romano-British sites along the Chippenham Western Bypass A4 to A350 link, by Clifford Bateman and Dawn Enright, with contributions by Jane Timby and Graeme Walker

63

71

82

90

105

131

181

A miniature flat axe or chisel from Broad Town, North Wiltshire, by Bob Clarke

A gold finger ring from the Rudge Romano-British villa site, Froxfield, by Bernard Phillips and Martin Henig

Excavations at Vale’s Lane, Devizes, 1996-7, by Phil Andrews and Lorraine Mepham A monkey’s head knife finial from near Trowbridge, by Paul Robinson Yet more about Cumberwell, by Kenneth Rogers Excavation and Fieldwork in Wiltshire 1998 Reviews S. E. Kelly (editor). Anglo-Saxon Charters V; Charters of Shaftesbury Abbey, by R. Harvey Peter Tolhurst. Wessex, a Literary Pilgrimage, by Michael Marshman Mere Papers, Numbers 1 -9, edited by M. F. Tighe, by Steven Hobbs Hazel Gifford. The Biography of a Country Church; Berwick St. John., by Michael Marshman Gwyneth F. Jackson (compiler). A tale of two manors; Zeals, a Wiltshire village, by Steven Hobbs Books also noted Obituaries Desmond Hawkins Eve Machin

Michael Lansdown

Index, by Philip Aslett

The Wiltshire Archaeological and Natural History Society

The Society was founded in 1853. Its activities include the promotion of the study of archaeology (including industrial archaeology), history, natural history and architecture within the county; the issue of a Magazine, and other publications, and the maintenance of a Museum, Library, and Art Gallery. There is a programme of lectures and excursions to places of archaeological, historical and scientific interest.

The Society’s Museum contains important collections relating to the history of man in Wiltshire from earliest times to the present day, as well as the geology and natural history of the county. It is particularly well known for its prehistoric collections. The Library houses a comprehensive collection of books, articles, pictures, prints, drawings and photographs relating to Wiltshire. The Society welcomes the gift of local objects, printed material, paintings and photographs to add to the collections.

The Wiltshire Archaeological and Natural History Magazine is the annual journal of the Society and 1s issued free to its members. For information about the availability of back numbers and other publications of the Society, enquiry should be made to the Secretary.

Publication by the Wiltshire Archaeological and Natural History Society does not imply that the Society endorses the views expressed; the factual content and the opinions presented herein remain the responsibility of the authors.

Notes for Contributors

Contributions for the Magazine should be on subjects related to the archaeology, history or natural history of Wiltshire. Whilst there is no fixed length, papers should ideally be under 7,000 words, though longer papers will be considered if of sufficient importance. Shorter, note length, contributions are also welcome. All contributions should be typed/ word processed, with text on one side of a page only, with good margins and double spacing. Language should be clear and comprehensible. Contributions of article length should be accompanied by a summary of about 100 words. Please submit two copies of the text (with computer disk if possible) and clear photocopies of any illustrations to the editors at the Museum, 41 Long Street, Devizes, Wiltshire, SN10 1NS.A further copy should be retained by the author. The editors will be pleased to advise and discuss with intending contributors at any stage during the preparation of their work. When submitting text on disk, Word or Rich Text Format files are preferred.

Referencing: The Harvard System of referencing (author, date and page, in parentheses within the text) is

preferred: e.g. ‘... one sheep and one dog lay close together (Clay 1925, 69)’. References in footnotes should be avoided if at all possible. Only give references which are directly applicable, repeating as little as possible. All references cited in the paper should be listed in the bibliography using the following style (with the journal name spelled in full, and the place and publisher of books/ monographs given):

For a paper:

ROSS, A.J., and JARZEMBOWSKI, E.A., 1996. ‘A Provisional Checklist of Fossil Insects from the Purbeck Group of Wiltshire’, Wiltshire Archaeological and Natural History Magazine 89, 106-115

For a book or monograph:

SMITH, LF., 1965, Windmill Hill and Avebury: Excavations by Alexander Keiller, 1925-39. Oxford: Clarendon Press

For a paper in a book or monograph: FITZPATRICK, A., 1984, “The deposition of La Tene metalwork in watery contexts in Southern England’, in B. Cunliffe and D. Miles (eds), Aspects of the Iron Age in Central Southern Britain, 178-90. Oxford: University Committee for Archaeology

Endnotes can be used for specific information that cannot otherwise be comfortably incorporated in the main body of the text.

Illustrations need to be clear and easily reproducible, the format following that of the Magazine. If possible, all original artwork should not exceed A3 before reduction. Drawings should be produced on drafting film or high quality white paper using black ink. Detail and lettering should not be so small that it will become lost in reduction. Mechanical lettering (dry transfer or computer generated) is preferred over hand lettering. Photographs should be supplied as good quality black and white prints, and transparencies and colour prints avoided wherever possible. Original illustrations and photographs should only be sent once a contribution has been accepted.

Offprints: Ten offprints of each article will be given free (to be shared between joint authors). Offprints are not given for notes and reviews.

Editorial Note

Readers will be aware of certain typographical changes to this volume of the Magazine, which have been introduced in an attempt to improve its design. It is hoped that these changes, which principally affect the hierarchy of headings and the layout of text at the beginning of articles, will make the Magazine easier to use and more attractive. The page size, print area, typeface and point size of the main text remain unchanged, as do the Magazine’s title and scope.

WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY SOCIETY COUNCIL AND OFFICERS (October 1999)

VICE-PRESIDENTS R.G. Hurn N. Davey, OBE, PhD, DSc, FSA N.J.M. Anderson EK. Annable, BA, MA, FSA H.F. Seymour, BA

Chairman

FOUNDATION TRUSTEES R.G. Hurn H.F. Seymour, BA Dr T.-K. Maurice, OBE J.-F. Phillips, BSc.

Col. D. Part, OBE (Military), TD, DL (London) (1997)

D.N. Shelton, BA, BSc., Dip.Ed., FRGS (1993) Lt. Col. C. Chamberlain (1994)

M. Darby, PhD, FSA, AMA, FRES, FRGS (1995)

B. Coupe, LDS, RCS, MSc. (Dent.) (1996) D. Lovibond, BA (1996)

M.J. C. Smith, BA, FICE (1996)

C. A. Shell, MA, MMet, PhD (1997)

R. Sneyd (1997)

M. Corney, BA (Hons)

D.N. Shelton, BA, BSc, Dip.Ed., FRGS C.J. Perraton, MA, CBiol, MIBiol.

D.J. Williams, MA

A.G. Lansdown

Mrs L. Bennett, Mrs P. Rugg, Mrs M.S. Groom i. Price

T-R. O’Sullivan

Mrs M.F. Lloyd

Mrs J. Brunt

Ms A. Cutforth, BA, Dip.AGMS

P.R. Saunders, BA, FSA, FMA, FRSA

Chief Executive

Curator

Deputy Curator

Assistant Curator (Natural Sciences) Sandell Librarian

Education Officer

Deputy Chairman P. Taverner, MA

Elected Members Mrs G.

Swanton, BA, Dip.Ad.Ed. (1997)

D.J. Williams, MA (1997)

D. Field (1998)

T. Schadla-Hall (1998)

Maj. Gen. G. M. G. Swindells CB (1998)

E. Stanford ARBS (1998)

Mrs P. Sneyd, PhD, BSc., CBiol, MIBiol (1999)

Ex-officio Members

Nominated Members

Chairman, Archaeology Committee

Chairman, Buildings and Monuments Committee Chairman, Natural History Committee Chairman, Programme Committee

Hon. Treasurer

Members, Wiltshire County Council Member, Devizes Town Council

Member, Kennet District Council Member, North Wiltshire District Council Member, Swindon Borough Council

County Museums Officer, Wiltshire County Council

Officers

Director, Salisbury & South Wiltshire Museum

G. Chancellor, BSc, PhD, AMA, MIMgt P.H. Robinson, PhD, FSA, AMA

Mrs A.J. Rawlings, MA, AMA

A.S. Tucker, BSc., AMA

Mrs L. Haycock, BA, Dip ELH, Cert.Ed. Ms J. Harvest, Cert.Ed.

Wiltshire Archaeological & Natural History Magazine, vol: 93 (2000), pp. 1-8

The Beckhampton Avenue and a ‘new’ Neolithic enclosure near Avebury: an interim report on the

1999 excavations

by Mark Gillings', Joshua Pollard? and David Wheatley’

Excavations to the west of Avebury have led to the discovery of the remains of a second megalithic avenue leading from the henge monument, and an unusual earthwork enclosure of probable middle Neolithic date. The existence of the second avenue (the so-called Beckhampton Avenue) had been mooted by the 18th-century antiquarian William Stukeley, though doubts about its existence had subsequently developed. Excavation revealed both buried stones and post-medieval stone destruction pits along its course, together with original stone sockets. Oval, and up to 140m in diameter, the enclosure pre-dates the avenue. Consisting of a shallow, semi-segmented ditch broken by a wide entrance, it shares morphological similarities with the first phase of Stonehenge. The work, undertaken during the late summer of 1999, is part of a collaborative project between researchers at the University of Leicester, the University of Wales College, Newport, and the University of Southampton.

BACKGROUND TO THE PROJECT

The preliminary results described here are from an ongoing project within the Avebury World Heritage Site, designed to develop a detailed understanding of the dynamic of monument construction in the later Neolithic of the region, and of changing configurations of landscape perception and encounter. In part, it builds upon the recent work of Alasdair Whittle, John Evans and others, in investigating the Neolithic sequence, environment and context of the region (cf. Whittle 1993; Evans et al. 1993).

The work so far has included topographic survey and stone recording at Avebury, the first stages in the production of a series of Virtual Reality simulations of the monument complex (cf. Pollard and Gillings 1998). Developing from this, the 1999 field season involved excavation 1.3km to the west of Avebury in Longstones Field, Beckhampton (SU 089693). The excavations were intended to explore

two features; a cropmark enclosure, visible on aerial photographs taken by the RCHME in 1997 (RCHME 1998), and a section along the course of a putative second megalithic avenue (the ‘Beckhampton’ Avenue) leading from the Avebury henge monument. In both aims, the fieldwork proved highly successful.

Previous research

Longstones Field derives its name from two substantial megaliths, colloquially known as ‘Adam’ and ‘Eve’. Set c. 27m apart and at right angles to each other, both are massive blocks of local sarsen, standing c. 4m high and comparable in bulk to some of the larger stones within Avebury (Smith 1965). Located c. 100m to the east-south-east of the Longstones are the extensively plough-damaged remains of the South Street long barrow (Ashbee et al. 1979), a second long mound (the ‘Longstones’ barrow (Barker 1984, 23)) being situated 200m to the south-west.

1 School of Archaeological Studies, University of Leicester, Leicester LE1 7RH; 2 Dept. of Humanities & Science, UWCN, Caerleon Campus, PO Box 179, Newport NP18 3YG; 3 Dept. of Archaeology, University of Southampton, Southampton SO17 1BJ

The first systematic recording of the Beckhampton complex was undertaken by the antiquarian William Stukeley, who considered the extant Longstones to be part of a putative cove and stone avenue (the ‘Beckhampton Avenue’) leading from the western entrance of the Avebury henge (Stukeley 1743). Little work was undertaken on the complex until 1913, when the fall of one of the stones (‘Adam’) led to its re-erection and a limited excavation by Maud Cunnington (Cunnington 1914). Although no conclusive evidence of date was forthcoming, a Beaker inhumation burial uncovered at the foot of the stone was evidently secondary to its erection. An extensive excavation of the adjacent South Street long barrow was undertaken by John Evans in 1966-67 (Ashbee et al. 1979). In addition to revealing the form of the barrow (constructed in the mid-4th millennium BC), this work provided a lengthy environmental sequence, and evidence for pre-barrow and Beaker episodes of cultivation. A programme of extensive surface collection within the Avebury environs, undertaken by Holgate and Thomas in 1983 (Holgate 1987), included the Longstones field, and demonstrated the presence there of very low lithic densities.

Geophysical survey of the field was undertaken by Andrew David of the Ancient Monuments Laboratory during 1989, in an attempt to demonstrate the existence or otherwise of Stukeley’s Beckhampton Avenue (Ucko et al. 1991, 195-9). Whilst the results of this survey were somewhat inconclusive, it did hint at the existence of archaeological features (many perhaps pits or buried stones) around and to the east of the Longstones. Additionally, one of these anomalies appeared to describe part of an oval ditched or palisaded enclosure adjacent to the standing stones, a finding that was confirmed by aerial photographs taken by the RCHME in 1997 (RCHME 1998). The newly discovered enclosure forms a flattened oval, 140 x 100m, with a wide entrance to the east, and encloses the eastern-most of the Longstones within its circuit. Aligned north-east - south-west, its north-western edge appears to run along the present field boundary and that on the south-east along the line of the Beckhampton Avenue. Its course marked by a thin (<2m wide) regular line of more luxuriant crop, the form of the cropmark bore a strong resemblance to those observed marking the late Neolithic palisade enclosures at nearby West Kennet (Whittle 1997).

2 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

RESULTS OF THE 1999 FIELDWORK

Geophysical survey

Prior to the commencement of the excavation, the Ancient Monuments Laboratory of English Heritage undertook a detailed geophysical survey of part of the area to be investigated. Resistivity and magnetometer survey were employed on a 60 x 60m area, centred c. 100m to the north-east of the Longstones, over the location of three weak anomalies detected during geophysical work in 1989 (Ucko et al. 1991, pl. 63). The resistivity survey was successful in locating a regular pattern of four anomalies (positive and negative) that subsequently proved to indicate the position of buried and destroyed stones (David et al. 1999).

Excavation The Enclosure

The cropmark enclosure was examined by means of five trenches (nos. 11, 12, 13, 14 and 15) placed at intervals around its circuit (Figure 1). In trench 14 the northern terminal of the enclosure entrance was exposed and excavated, confirming that the original eastern entrance is of the order of 60m wide. In all the sections the ditch was shallow, steep-sided and flat bottomed, up to 2.10m wide and 0.80m deep. Despite the regularity in profile, it was markedly uneven in plan. This was seen particularly well in the 5m long section excavated in trench 11, where the ditch sides bowed at various points, and the base became correspondingly deeper. A semi-segmented form to the ditch can also be discerned in the re- worked data from the 1989 geophysical survey. The impression is of the ditch having been dug as a series of separate segments, subsequently joined by removing intervening causeways of un-dug chalk.

A uniform sequence of ditch filling was encountered in each trench. A primary fill of chalk rubble was sealed by a secondary fill of silty clay and chalk fragments. In places, this was overlain by a thin lens of clean brown loam, corresponding to a turf- or soil-layer that had developed after initial stabilisation of the fills. The ditch was finally levelled in a single episode by a backfill deposit of compact mixed chalk rubble and silty loam, probably comprising material from an associated up- cast bank. The weathered nature of the outer edge of

BECKHAMPTON AVENUE AND A ‘NEW’ NEOLITHIC ENCLOSURE NEAR AVEBURY

Longstones Field

a

Silbury Hill -. Sanctuary “tag

So West Kennet enclosures

NS 2 15 N\

8 Tr. V4

aN Ridge and furrow

8 y

Longstones

Figure 1. Longstones Field, Beckhampton: Plan of the excavations

4 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

the ditch, and the occasional presence of lenses of slump material sitting atop the turf-line on the inner face, suggest that the up-cast bank was internal.

Finds from the ditch fill were remarkably few. A small amount of animal bone came from the soil at the base of the tertiary fill, as did a substantial fragment of a single Grooved Ware vessel (from trench 12). Most likely placed deposits, in each instance these may have been sitting in shallow re-cuts. A number of fresh pieces of worked flint were also present, especially in trench 11. The only finds from the primary fills comprised a small spread of bone from the ditch terminal, and a sherd of pottery from trench 11, provisionally identified by Ros Cleal as earlier Neolithic.

Although only limited areas of the interior were exposed in the trenches, no features were observed, nor were pieces of worked flint recognised in the ploughsoil over this area.

The Beckhampton Avenue

‘The avenue was investigated by means of a single open area (trench 10), centred over the four possible stone

pits located on the 1999 geophysical survey, and subsequently extended to the south-west in order to investigate two further anomalies faintly visible on the 1989 survey (Figure 1). A number of features were revealed, prominent amongst which were remnants of medieval ridge-and-furrow cultivation running north-south across the area. Of greater significance were six discrete features belonging to the Avenue, arranged in two parallel lines running south-west to north-east, and directly aligned on the remaining Longstones. Consisting of medieval stone burials and post-medieval stone destruction pits, the longitudinal intervals between each ranged between 22—30m, and the transverse intervals 14-17m. Of the six pits, three contained buried sarsen stones (F.22, 25 and 26), one had evidently contained a stone which had been subsequently removed (F.23), and in the remaining two were layers of stone destruction debris (F.21 and 24). The original stone sockets were found immediately adjacent to each of the buried stones, and another in the base of destruction pit F.24. More intensive searching (precluded due to time constraints) may well have located those for the two remaining stone settings.

Figure 2. Longstones Field, Beckhampton: Stone burial pit F.26. Note the original stone socket to the left of the pit

BECKHAMPTON AVENUE AND A ‘NEW’ NEOLITHIC ENCLOSURE NEAR AVEBURY 5

Figure 3. Longstones Field, Beckhampton: Stone destruction pit F.21 partially excavated, showing the spread of sarsen flakes and charcoal

The alignment of the Avenue is such that it runs through the eastern entrance of the enclosure, though off-set to the south-east in such a manner that a substantial gap (c. 20m) is left between the stones and the northern ditch terminal. Correspondingly, on the south-east side the stones appear to line up with the southern ditch terminal; and it may be that at this point the Avenue was reduced to a single line of standing stones.

Stone burials

Whilst the soil fills of F.21, 23 and 24 made them immediately visible following stripping, establishing the position and extent of the burial pits F.22, 25 and 26 involved very careful investigation. The tops of buried stones just projected from the chalk rubble backfill of F.22 and 25, but the extent of F.26 only became apparent after a rain storm (which served to distinguish clearly the much cleaner rubble fill from the surrounding weathered natural). All the burial pits were regularly cut, their shape and size closely matching that of the stones. The deepest burial pit was that of F.26 (1.42m in depth), and here the side adjacent to where the stone had stood was ‘battered back’, presumably to avoid destabilising it whilst the burial pit was dug (Figure 2).

Following the toppling of the stones, the pits were rapidly filled with chalk rubble, often highly compacted in the top of the fills. The only finds from the pit fills consisted of small sarsen blocks (perhaps disturbed packing stones), a quantity of worked flint, and a chopped large- mammal rib from F.26.

The stones, like those of Avebury and the West Kennet Avenue, were unmodified blocks of sarsen. Their shape and size varied, ranging from 2.34 — 3.00m across. That in F.26 was particularly unusual. Its upper face was bisected by a series of large cracks and folds in the rock, and there were numerous natural depressions and perforations (Figure 2). On the southern end three large perforations ran through the full thickness of the stone, one being filled with soil containing a curious assemblage of worked flint and a split large-mammal longbone. On the exposed upper surface of the sarsen in F.25 were a number of features of anthropogenic origin, including a sub- rectangular wedge-hole on its northern end, and a set of three stone axe polishing marks on the southern. The low position of the polishing marks in relation to the presumed original base of the stone (its southern end) demonstrates its utilisation for axe working took place prior to its erection.

6 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Stone destruction pits

Two of the pits, F.21 and 24, clearly relate to post- medieval stone destruction events. In contrast to the burial pits, these were shallow and irregular features; oval, up to 5.40m across and 0.35m deep, with very uneven bases. The basal fills comprised extensive spreads of flaked and burnt sarsen within a matrix of burnt straw and charcoal (Figure 3). These destruction deposits were sealed by layers of ploughsoil containing small sherds of medieval and early post-medieval pottery. Dating for the destruction is perhaps provided by fragments of later 17th century clay pipe from the burning deposits.

Stone sockets

Four original stone sockets were discovered (adjacent to F.22, 24, 25 and 26); the destruction event represented by F.21 perhaps removing the socket in that instance, and the incomplete excavation of F.23 precluding more detailed search and investigation here. These were irregular in shape, roughly oval or sub-rectangular, and up to 0.50m deep. The primary fills comprised some compact chalk packing, incorporating sarsen blocks in two instances. Sets of post-holes were discovered in the bases of three sockets, related either to the setting-up of the stones, or forming pre-stone settings. All were filled with loose chalk rubble, and none showed evidence of post- pipes, indicating deliberate withdrawal of the posts rather than in situ decay.

DISCUSSION

The results of the 1999 excavations proved more successful than anticipated. A ‘new’ Neolithic enclosure has been added to the growing corpus of prehistoric monuments in the region, and the antiquarian observations of William Stukeley have been vindicated through the ‘re-discovery’ of the Beckhampton Avenue.

The enclosure lies not far from the southern slope of Windmill Hill and is overlooked by its more famous neighbour. It is tempting to infer some relationship, though both enclosures are of rather different character, in terms of form and the known range of associated activities. Unlike the ditches of Windmill Hill, with their evidence of repeated depositions of animal bone, pottery, flint and other materials (Whittle et al. 1999), little appears to have entered the ditch of the Beckhampton enclosure, at least during its primary phase. It might even be regarded as

anomalously ‘clean’. The same might be said for the interior — note the seeming absence of worked flint in the ploughsoil — though this remains to be properly tested. On these grounds there is certainly no reason to assume occupation. The monument might even have been deliberately avoided or ‘abandoned’ once constructed. Perhaps its location adjacent to two earlier long mounds is a clue, the enclosure serving a special purpose for mortuary rituals or other special practices separated from the routines of living. A range of interpretive possibilities presents themselves.

Until radiocarbon determinations are available, the date of the enclosure can only be inferred from limited artefactual remains, its morphology and relationship to the Avenue. The position of a substantial fragment of a Grooved Ware vessel from not far above the primary fills (even if in a shallow re- cut) would imply construction in the 3rd rather than 4th millennium cal BC. Morphologically though, the enclosure shares many features with earlier enclosures, such as the suggestion of an internal bank, and the semi-segmented nature of the ditch. Though not of the same geometric regularity, it is in this respect similar to the first phase of Stonehenge (Cleal et al. 1995). Other parallels might be provided by continuously-ditched earlier Neolithic enclosures in Sussex, such as Bury Hill, Houghton (Bedwin 1981). A search for more than general analogies may, however, prove futile. Within its local context the enclosure is unique. It also presents unusual features, such as the exceptionally wide entrance through which the Beckhampton Avenue later ran. The sequence of fills, with a clear episode of backfilling and levelling whilst the secondary silts were forming, implies a short life for the monument, measurable maybe in tens rather than hundreds of years. It remains to be determined whether such planned ‘destruction’ relates to the construction of the Avenue and Longstones ‘Cove’, or whether the use-life of the monument was prescribed from the outset.

Whilst the form of the enclosure finds little ready analogy within the immediate region, the stone settings have a direct parallel in those of the West Kennet Avenue leading from the southern entrance of the Avebury monument. The longitudinal and transverse intervals between the stone settings in each monument are identical, as is the range of stone size (Smith 1965, 206). Details of the sockets, and perhaps by extension the processes of erection, can also be matched. The post-holes cut into the bases of the stone sockets fit very closely Smith’s description of so-called ‘anti-friction stakes’ encountered by Keiller during the course of excavations along the West

BECKHAMPTON AVENUE AND A ‘NEW’ NEOLITHIC ENCLOSURE NEAR AVEBURY ‘fl

Kennet Avenue (Smith 1965, 219). There is, therefore, a temptation to regard both Avenues as part of a unitary episode of construction, perhaps also tied in with the stone settings inside the Avebury henge. Another possibility is for more protracted, perhaps episodic, construction. Changes in alignment along the West Kennet Avenue could indicate that this was built in a piecemeal manner, its final course not being determined from the beginning, but developing through repeated addition and re-working (Burl 1993, 45-7). Dating of the stone settings both at West Kennet and Beckhampton remains to be resolved (no dateable material came from the stone sockets during the recent excavations). Only a broad span, most likely somewhere in the 3rd millennium cal BC, can be offered (Pitts and Whittle 1992).

Though the precise chronology of the Avenue’s construction still remains to be determined, we can be more confident about the dates of its destruction. The method of stone burial is identical to that observed at Avebury, with stones being toppled into carefully cut burial pits matching the size and shape of the sarsens with precision, and ‘the whole levelled so that no trace remained of the operation’ (Smith 1965, 177). Jope considers that stone buriai in Avebury occurred over a short period during the early 14th century, the motive being provided by ecclesiastical disapproval of superstitious practices associated with the stones (Jope 1999, 67). Perhaps it is significant that the deepest of the burial pits at Beckhampton (F.26) contained the most unusual and ‘featured’ sarsen — the stone most likely to attract folkloric practices?

The second episode of destruction was that recounted by Stukeley (1743). Driven by economic expediency, and undertaken by local farmers such as Griffin and Richard Fowler, this involved breaking up the stones through controlled fire-setting and the use of sledge hammers. Dating is provided by fragments of late 17th century clay pipe. Although large quantities of sarsen chips remained in the backfill of destruction pits F.21 and 24, in each instance this must represent but a small fraction of the original stone. Most of the debris comprised flakes produced through the trimming and shaping of larger blocks, these having been taken away for use as building stone.

Having established that the Beckhampton Avenue exists, that Stukeley’s observations on its course from the western of Avebury to Longstones Field are most likely accurate, and that it met the same fate in the hands of medieval zealots and post-medieval farmers as befell Avebury and the West Kennet Avenue, many questions still present themselves. Does the Avenue

terminate at the Longstones, or continue further to the south-west as envisaged by Stukeley? If so, could there be another Sanctuary-style structure at its termination? What is the precise chronological relationship between the Avenue and the enclosure? Following Stukeley, is the most westerly of the Longstones (Adam) the sole remnant of a ‘cove’, and are the scale and position of this feature related to its location at the point where the Avenue runs over the line of the enclosure? Could the Avenue be of more than one phase of construction? To an extent, several of these questions can be answered by further excavation and careful geophysical prospection, which are planned for the future.

Perhaps the most striking result of the fieldwork is that it emphasises once again the sheer scale and magnitude of monumental construction during the later Neolithic of the region. The new enclosure and Beckhampton Avenue can be added to a list which already includes Avebury itself, the West Kennet Avenue, the Sanctuary, Silbury Hill, the West Kennet palisaded enclosures, and perhaps several as yet uninvestigated small stone circles in the region. All these monuments may have come into being over a time-span of just a few hundred years, though the exact ‘periodicity’ or rhythm of monument construction needs to be established. Whittle considers that the motivation and necessary mobilisation of effort for such constructions could have come as much through the force of shared religious belief — participation in what seemed respectful and proper — than coercion (Whittle 1997, 165-6). Indeed, the construction of the Avenues might be read as a statement of unity, drawing disparate fragments of an earlier landscape into a symbolic whole. Thus, the Beckhampton enclosure, or at least explicit memory of it, was physically drawn into the Avebury complex through the construction of the Avenue. Given the remarkable cleanliness of the enclosure and its location alongside the South Street and Beckhampton (Longstones) long barrows, this may have been a monument closely associated with dealings with the dead and ancestors (whatever form they took: Whittle 1998). As one interpretation, by the 3rd millennium BC this particular place was set apart from the routines of daily practice, already embodying associations of deep-time and mythical beginnings (note the evidence for early occupation and other activities under South Street: Ashbee et al. 1979). Though we would not subscribe to the structural rigidity of their scheme, Parker Pearson and Ramilisonina’s (1998) interpretation of the West Kennet Avenue as a pathway for ‘the ancestors’ might equally apply to that at Beckhampton.

8 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Acknowledgements

First and foremost, we would like to acknowledge the friendly co-operation and support given by the owner of Longstones Field, Robin Butler. Necessarily, a project such as this is the work of many hands, and for practical support, input and advice we would particularly like to thank Ros Cleal (Alexander Keiller Museum), Andrew David (Ancient Monuments Laboratory), Dave Field (English Heritage), Rosina Mount (for vital camp-site support), and by no means least Gill and Robin Swanton (North Farm). Glyn Goodrick provided invaluable assistance with the excavation supervision. The enthusiastic workforce was composed of students from Leicester, Newport and Southampton, along with local and more distant volunteers. The many visitors to the site offered constructive advice and comment. Alasdair Whittle and Julian Thomas are thanked for their support of the project. The work was generously funded by the AHRB, the Society of Antiquaries of London, the Prehistoric Society, and the Universities of Leicester, Newport and Southampton.

References

ASHBEE, P., SMITH, I.F. and EVANS, J.G., 1979, “The excavation of three long barrows near Avebury, Wiltshire’, Proceedings of the Prehistoric Society 45, 207-300

BARKER, C.T., 1984, ‘The Long Mounds of the Avebury Region’, WAM 79, 7-38

BEDWIN, O., 1981, ‘Excavations at the Neolithic Enclosure on Bury Hill, Houghton, W. Sussex 1979’, Proceedings of the Prehistoric Society 47, 69-86

BURL, A., 1993, From Carnac to Callanish: the prehistoric stone rows and avenues of Britain, Ireland and Brittany. New Haven: Yale University Press

CLEAL, R.M.J., WALKER, K.E. and MONTAGUE, R., 1995, Stonehenge in its Landscape: Twentieth-century Excavations. London: English Heritage

CUNNINGTON, M.E., 1914, ‘The Re-erection of Two

Fallen Stones and Discovery of an Interment with Drinking Cup, at Avebury’, WAM 38, 1-11

DAVID, A., MARTIN, L. and PAYNE, A., 1999, Beckhampton, nr. Avebury, Wilts. Draft Report on Geophysical Survey, May 1999. Unpublished report, Ancient Monuments Laboratory, English Heritage

EVANS, J.G., LIMBREY, S., MATE, I. and MOUNT, R., 1993, ‘An environmental history of the upper Kennet valley, Wiltshire, for the last 10,000 years’, Proceedings of the Prehistoric Society 59, 139-195

HOLGATE, R., 1987, ‘Neolithic settlement patterns at Avebury, Wiltshire’, Antiquity 61, 259-63

JOPE, E.M., 1999, “The Saxon and Medieval Pottery from Alexander Keiller’s Excavations at Avebury’, WAM 92, 60-91

PARKER PEARSON, M. and RAMILISONINA, 1998, ‘Stonehenge for the ancestors: the stones pass on the message’, Antiquity 72, 308-3206.

POLLARD, J. and GILLINGS, M., 1998, ‘Romancing the Stones: towards a virtual and elemental Avebury’, Archaeological Dialogues 5:2, 143-164

RCHME 1998, An Enclosure at Beckhampton, Wilts: Aerial Photographic Transcription and Analysis. Unpublished report

SMITH, I.F., 1965, Windmill Hill and Avebury: Excavations by Alexander Keiller 1925-1939. Oxford: Clarendon Press

STUKELEY, W., 1743, Abury: a Temple of the British Druids. London

UCKO, P.J., HUNTER, M., CLARK, A.J., and DAVID, A., 1991, Avebury Reconsidered: From the 1660s to the 1990s. London: Unwin Hyman

WHITTLE, A., 1993, ‘The Neolithic of the Avebury area: sequence, environment, settlement and monuments’, Oxford Journal of Archaeology 12, 29-53

WHITTLE, A., 1997, Sacred Mound, Holy Rings: Silbury Hill and the West Kennet palisade enclosures: a Later Neolithic complex in north Wiltshire. Oxford: Oxbow Books.

WHITTLE, A., 1998, ‘People and the diverse past: two comments on ‘Stonehenge for the ancestors”, Antiquity 72, 852-4

WHITTLE, A., POLLARD, J. and GRIGSON, C., 1999, The Harmony of Symbols: the Windmill Hill causewayed enclosure, Wiltshire. Oxford: Oxbow Books

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 9-19

‘A Very Pretty Seat’: Erlestoke Park, 1780—1999

by Isabel Ide

This paper offers a short history of Erlestoke Park, its owners and tenants, including the builder Joshua Smith, M.P,, the next occupier, George Watson Taylor, M.P., a long-term tenant, John Cam Hobhouse, M.P., later Lord Broughton de Gyftord, the return of the Watson Taylors, and the final dispersal of the estate after the First World War, and the fire of 1950.

ris os sated wis 2) Zot Beate Bact aee

Ee

Fig. 1. Erlestoke Park from the east: engraving published by John Britton in 1825

‘A very pretty seat,’ William Cobbett remarked when riding past Erlestoke Park in 1826, noting the black swans on the lakes and the clematis-covered cottages in the village.! The present-day traveller is more likely to observe the ornamental black and gold gates and the notice board bearing the legend ‘Her Majesty’s Prison Erlestoke Park’. Upon the site there had been an Elizabethan manor house on the sheltered meadowland of which no trace remains. Peter Delmé

Rhotteridge Farm, Lower Woodrow, Forest, Melksham, Wilts SN12 7RB

sold this earlier house to John Smith, a merchant of Lambeth, in 1780° and he had it demolished immediately in order to make way for a new house. Shortly after completing the transaction John Smith died and it was his eldest son, Joshua Smith (1732-— 1819), who was responsible for building the new mansion. The architect whom he chose to build Erlestoke Park was George Steuart (1730-1806), about whom not a lot is known. He is said to have

10 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

started his working life as a housepainter, but his transformation to an architect is shrouded in mystery. It has been said that his country house architecture is characterised by an elegant restraint verging on bleakness.’ ‘Stoke Park’, as the house was originally named, had the thinly modelled elevations, chaste neo-classical motifs, and compact planning that were typical of George Steuart’s designs.

The new house was considered sufficiently large for polite society, but not extravagantly grand. On entering the house the visitor would have found himself in a hall 40 feet wide by 32 feet deep surrounded by Grecian columns. On the right hand side was the main drawing room and the dining room. Through the hall was a large library. On the other side of the entrance hall was the breakfast room, the stairs and a dressing room. Above the public rooms were eleven bedrooms and one water closet.

In his design Steuart had omitted any external doors to the central part of the house. Behind the steps and imposing portico of the principal front were three tall sash windows, so the only entry was through an inconspicuous door in the west wing. In contrast inside the house there were plenty of doors; the library alone had four real doors and two false ones. The main block was three stories high over the basement and cellars; the service wings were two stories high.

While the house was being built the surrounding countryside was being converted into a handsome park by the landscape designer George Eames and completed in 1786.*There is a description of the park in its full glory in John Britton’s Beauties of Wiltshire (1800):

The sides and summit of the escarpment edge of the plain have been thickly planted with wood, which as it advances in growth will give the seat an additional beauty. The Park abounds with many fine large elm trees and is enriched with a sheet of water, after forming seven different cascades in its progress it is collected into a lake of considerable dimensions. This spot abounds with a choice collection of botanical plants.

Not only did Joshua Smith spend lavishly on the house and its environment; he also improved the village of Erlestoke, as Britton recounted: ‘The poor villagers’ humble cottages were formerly devoid of comfort, the houses being situated in a narrow valley, subject to the inundations of every trifling flood. This has been remedied by the proprietor and a comfortable habitation has been provided for the peasant and his family with a sufficiency of garden ground to supply them with vegetables. I feel considerable pleasure in relating these instances of benevolent condescension to the wants and distresses

of the poor; but the pleasure would be heightened into rapture, if any encomiums of mine could shame the penurious or the inconsiderate to similar actions.’

Joshua Smith first stood for Parliament at Penryn in 1784 as a supporter of Pitt’s administration. Having failed to get elected, however, he turned his attention closer to his new home and in 1788 stood for the parliamentary borough of Devizes.° At this time one of the town’s two representatives was Henry Addington, later Viscount Sidmouth. Remarkably, Addington represented Devizes unopposed for 30 years from 1783, a feat which Smith emulated from 1788. This can be partly explained by Smith’s generosity. The electors of Devizes benefited from gifts from Smith of £500 in 1791 and of £1,000 in 1803 for improvements to the town. He could not be described as an over-active Member of Parliament, and there is no record of his ever having actually spoken in the House.

In 1766 Joshua Smith had married Sarah Gilbert, the daughter of a judge who was a member of the Antigua legislative council. Smith can be described as an acolyte of Addington, whose policies were pro- slavery, anti-emancipation for Catholics, and very much against any hint of Parliamentary reform. After his 80th year Smith’s health deteriorated; he took leave of absence from the House of Commons in 1816 and 1817, and finally retired in 1818 at the age of 86. By this time three of his daughters had married M.P.s, respectively Lord Compton, William Chute and Charles Smith.’

After Smith’s death in 1819 his daughters sold Erlestoke Park to George Watson Taylor, a theatrical, profligate, larger than life character. The fourth son of George Watson of Sauls’ River, Jamaica, George Watson (as he then was) was educated in England, at Lincoln’s Inn and St Mary’s Hall, Oxford. He originally made his name as a playwright. His play England Preserved was performed at the Theatre Royal, Covent Garden at George III’s request in February 1795, being much applauded for its anti- French sentiments!* In addition he was an author of poetry and political pamphlets. At the age of 24 he was appointed private secretary to the lst Marquess of Camden when he was Lord Lieutenant of Ireland during the years from 1795 to 1798, so Watson was a witness to the atrocities that took place in Ireland in 1798.° Subsequently he held office again in Ireland, probably as assistant to Lord Castlereagh while he was Chief Secretary for Ireland (1798-1801).

When Castlereagh was moved to the India Board in 1802 Watson continued to be his private secretary, but later returned to the service of the Marquess of

“AVERY PRETTY SEAT’: ERLESTOKE PARK, 1780-1999 sal

DLrgenwec S Subrtabed ty tyeomn Aarhivivabver

Fig. 2. Erlestoke Park: first floor (above) and ground floor (below) plans, 1799 (Wiltshire Buildings Record, file B1664)

12 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Camden and, through Camden’s influence, was appointed in 1806 to a deputy tellership and a Commissioner of Excise at a combined salary of £2,200 per annum.'°

When he was 38 George Watson started courting Anna Susanna Taylor, the daughter of Sir John Taylor of Lyssons in Jamaica. ‘Although it was acknowledged that he bears a most excellent character and is much esteemed by all his friends and relations, his lack of a private income stood in the way of his happiness.’ Eventually in 1810 the match was allowed to proceed on the understanding that a favourable settlement was made on her and any children of the marriage.'!

Anna Taylor was the niece of Simon Taylor (1740- 1813), a bachelor, the owner of six sugar estates and three cattle ranches, who also acted as a plantation attorney for absentee proprietors. He was commonly supposed to be the richest man in Jamaica, and was determined that his nephew and heir, Anna’s brother Simon Richard Taylor, would inherit his fortune and thereby become the richest commoner in England.'” The years between 1790 and 1799 brought unprecedented prosperity to plantation owners, as the result of the St Dominican slave revolt and the Napoleonic War. Sugar prices doubled and more prolific types of sugar cane were imported from Tahiti and Bourbon. From 1800 profits declined from both sugar and coffee, as production costs rose and customs duties were increased.

At the time of Simon Taylor’s death in April 1813 he owned the estates of Lyssons, Holland, Llanrhumney, and Haughton Court. He had additional property of £740,000 bringing in an annual income of £47,000.'? His heir, Sir Simon Richard Brisset Taylor, sadly died only two years after inheriting all this wealth, which then passed to his sister Anna Susanna Watson. The fortunate couple took the additional name of Taylor, and as Lady Charlotte Bury commented: ‘What a wonderful change of fortune for these two persons, from having an income of two to three thousand a year with tastes far beyond such limits, to almost boundless and unequalled riches! It is said they are full of projects of splendour and enjoyment.’!*

‘The newly enriched couple embarked on the joyful process of house-hunting with a deep purse. One of the houses under contemplation was Houghton Hall in Norfolk, which had been built for Sir Robert Walpole in 1730. In 1819, however, the Watson Taylors decided to buy Erlestoke Park and the surrounding estates from the executors of Joshua Smith for £200,000." Since 1816 Watson Taylor had been M.P. for Newport on the Isle of Wight. Between 1818 and

1820 he had represented Seaford in Sussex and then from 1820 until 1826 East Looe in Cornwall.'® His reasons for entering Parliament included a desire to participate in any debates that concerned the West Indies, and particularly Jamaica. By 1808 public opinion had slowly begun to move against slavery and the import of slaves into British colonies had been banned. These moves had a deleterious effect on the profitability of the sugar plantations, the source of Watson Taylors’ income. George Watson Taylor was immediately elected to the standing committee of the West Indian Planters and Merchants on his arrival in the House of Commons.

When a vacancy occurred at Devizes in 1826 Watson Taylor put himself forward, but not all the electors were in favour of returning a member whose income was dependent on slave labour. A local newspaper commented that, ‘MrWatson Taylor of Earl Stoke has offered himself to the notice of the electors of Devizes, lately represented by Mr Estcourt, and will most likely to be returned, unless the Quackery of negro emancipation should interfere, Mr Taylor being the possessor of between two and three thousand slaves.’!’ The ‘Quackery’ was insufficient to prevent Watson Taylor being returned.

Although he attempted to respect his wife’s desire that her slaves should be treated with considerations of pure humanity, benevolence, justice and liberality, and indeed boasted in 1824 that he had spent over £140,000 in attempting to ameliorate the condition of his slaves, Watson Taylor was obdurate in his antagonism to abolition, and objected, ‘to the way that itinerant adventurers had collected signatures for anti-slavery petitions by inflaming the passions of the people’.'* Quite suddenly, however, the profits to be made from sugar plantations fell due to preferential tariffs being abolished, and the replacement of West Indian sugar with sugar extracted from home-grown sugar cane.'”

Despite being commoners the Watson Taylors consorted with the highest in the land. In addition to Erlestoke Park they had a magnificent mansion in Cavendish Square, which was ‘superbly illuminated’ on the acquittal of the tragi-comic Queen Caroline in the celebrated divorce case brought against her by her husband George IV prior to his coronation.”° Mr and Mrs Watson Taylor were rarae aves in being on friendly terms with both the Duke of Clarence and the Duchess of Kent and her younger daughter Victoria, The duchess strongly disapproved of her brothers-in-law, and would allow the young heiress to the throne very little contact with them.

Already by 1823 George Watson Taylor’s extravagance had led him into financial trouble, and

‘AVERY PRETTY SEAT’: ERLESTOKE PARK, 1780-1999

he was forced to sell some of his finest books and paintings. The sales of the books alone covered nine days, the total amount raised being over £30,000. Despite this setback they continued with their generous hospitality, as shown by the following account from the Devizes and Wiltshire Gazette of 16th August 1827:

FETE CHAMPETRE AT ERLESTOKE PARK Mr and Mrs Watson Taylor, with that munificence by which they are characterised, on Tuesday last gave one of the most splendid fetes ever witnessed in this part of the country.

The invitations were principally confined to Devizes and its neighbourhood, but we observed many of the most respectable families from different parts of the county. The number on the ground we should suppose could not be less than seven hundred. The beautiful scenery of Erlestoke Park is well calculated to set off to advantage such assemblage of fashion and beauty and display, with full effect, the accompanying variety of costume. The weather in the morning was lowering and inauspicious, but toward mid-day it brightened, and continued free from rain until eight in the evening. The gates of the park were open at two o’clock, and within a short time afterward, the company began to arrive. The carriages drew up at the front door, and after passing through a spacious entrance hall (the butler announcing the names as they entered) the company were received in the Library in the most polite and affable manner by both Mrs and Mr Watson Taylor. They then passed into a beautiful flower- garden, and after promenading here for some time, proceeded to the extensive pleasure grounds. The excellent arrangement of the walks in these grounds (extending over 600 acres) and the order and care in which they are kept, excite the admiration of all who visit them. In different parts were stationed bands of music, playing some of the most popular airs, which greatly added to the enchanting pleasures of the day.

About three quarters of a mile from the house, and as an ample lawn, gently rising above the water which winds its course through the pleasure grounds into the Park, was a temporary erection, seventy feet square, and of proportionate altitude. This erection was neatly thatched, and the pillars supporting it tastefully decorated with laurels and evergreens; within, and on the turf four long tables, capable of containing 500 persons were laid; and from the variety and fanciful arrangements of the viands, they had quite a picturesque appearance. At a short distance, a room between 60 and 70 ft in length, with an excellent flooring, was erected for dancing, supported by columns (rendered exceedingly graceful by wreaths of flowers and evergreens) forming a beautiful arcade with a piazza on either side. In front of these rooms, on the lawn, was the principal promenade before dinner, and it is impossible to imagine a more gay and imposing scene. No one could view, without delight and rapture, the

13

groups of lovely women, glowing with animation and gracefully and splendidly attired, parading to and fro upon the verdant lawn: they vindicated their just claim to the character ascribed to them of ‘giving the country its charms’, as in less ostentatious situations they merit that of ‘imparting to home its delights’. It was indeed a scene of enchantment. A few parties perambulated the various walks, where the music now and then break in, in full choir upon the air, came with added sweetness because unseen.

Soon after four o’clock, the company crowded to the dining room, where there was an ample supply of the best and most substantial viands, of the choicest wines, and of all the delicacies of the season. Soups, fowls, lamb, venison, lobsters, tongues, hams, Etc, Etc, (the soups and fowls hot). Confectionery in great variety — under the direction of Mr Kemp of South Audley Street, whose arrangements evinced great taste and judgement; fruits rich and abundant; Sparkling Champagne, Burgundy, Madeira, Port, Sherry Etc, Etc.

The room for dancing was, in the meantime, lighted with variegated lamps, formed in festoons; and at about half past five o’clock quadrille parties were arranged, and quadrilles danced with grace and softened animation, to the tones of an efficient quadrille band from Bath. Other parties separated to a distant part of the lawn, where the more rural country dance was kept up with great spirit; but the greater part of the company indulged in the pleasures of the promenade.

Throughout the evening, ices, jellies, with a variety of other sweet things, and confectionery, lemonades, Roman punch, wines, tea, etc, etc were distributed in abundance.

Variegated lamps forming two large stars and various festoons in different parts illuminated the walk leading from the dancing room to the gate at the entrance of the village of Stoke, at which place the carriages were brought up; but it was between nine and ten o’clock before the great bulk of the company thought of separating; the hours flew with the wings of birds of paradise; the pace of time gave no echo to the sense; and never was the sentiment of the poet more entirely realised —

‘Noiseless falls the foot of Time Which only treads on flowers’

We do not think the day will ever be forgotten by those who were present; it will be ranked among the most happy of their lives. The extreme affability and kindness of Mr & Mrs Watson Taylor, have excited an impression that will never be effaced. All were alike happy — all delighted.

Silk hats, ornamented with flowers or feathers, were generally worn by the ladies, & most of the dresses very handsome * * * *

It affords us sincere pleasure to state, that throughout this gay and happy day not the slightest accident occurred. Mr Watson Taylor, with his accustomed politeness and attention, sent into Devizes on the following morning, to ascertain the safe returns of his friends.

14 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Fig. 3. Erlestoke Park, south front: engraving published by John Britton in 1825

Between 40 and 50 pairs of horses were ordered from the Bear Inn alone, on the occasion; and the excellent appearance of those horses, together with the good arrangement of the carriages, and the personal attention of Mr E. Parsons, reflects on him considerable credit.

Within a couple of months Mr and Mrs Watson Taylor were again entertaining the local gentry. This time the report in the Devizes and Wiltshire Gazette gave greater details of the furnishing of Stoke Park:

SPLENDID DINNER On Friday last a most Splendid Dinner Entertainment was given by Mr & Mrs Watson Taylor, at Erle Stoke Park. We have taken some pains to collect the particulars, and we believe the following to be correct: — The invitations included the resident members of our Corporation; the Members for the County; the neighbouring Magistrates and other gentlemen of distinction. The company assembled in the magnificent and well furnished library —aroom of large dimensions, and which, in addition to a valuable collection of books, is ornamented with many paintings and busts of poets and scientific, literary, or remarkable persons:— Pope, Dryden, Lord Byron, Sir Walter Scott, Sir Joseph Banks, Dr Johnson, Warren Hastings, Sir Joshua Reynolds, Baretti, Etc. It is also adorned with splendid mirrors, of great size, and which are so placed as to reflect and apparently multiply, the

various objects of Taste, Literature or vertu with which the rooms abounds.

Here the party was met by the amiable Host and Hostess, (attended by their fine young family) and received with that kind hearted and well bred courtesy, by which Mr and Mrs Watson Taylor are so eminently distinguished. After a short interval, the dining room (40 feet by 38, we believe) was thrown open, and the company were at once struck by the magnificent scene before them. A plateau of massy silver gilt, (about 30 feet long, it is said) blazing with lights, and surrounded by classical and admirably executed Tripods or Candelabra, and groups of figures, nearly filled the entire centre of a long table — The Sideboards also, resplendent with superb plate and numerous lamps and branches rendered the coup d’oeil as brilliant as it was beautiful. The figures on the Plateau bespoke at once the taste and the opulence of the proprietor; whether as related to the designs, the intrinsic value (we do not vouch for the exact amount, but we have heard £18,000) — or the skill of the artists employed in its execution, — The chief subject was ‘The Graces unrobed by the Loves’, and we are informed that nothing could exceed the voluptuous, yet chaste beauties of the one, except, perhaps, the light, airy, and playful archness and ardent interest of the others.

The Sideboards were covered, one with gilt, and the others with silver plate, uniting beauty and utility. Some of the gilt salvers were of great size, and covered with admirably executed designs in alto or basso relievo. The

“AVERY PRETTY SEAT’: ERLESTOKE PARK, 1780-1999

dinner service for the first course was entirely of highly embossed silver, and for the succeeding one and the dessert beautifully painted china. The viands consisted of everything rare or excellent, which the season or the various elements afforded. Turtle — different sorts of Fish —Venison and Game — besides the usual combinations of the Cuisine Francaise. The wines were of exquisite flavours, and seemed, from variety and number, to have laid France, the Rhine, Spain, Portugal and Madeira, all under liberal contribution. The liqueurs, also Maraschino and Curacao, were excellent of their kinds. The dessert was not inferior to the other parts of the repast — the pines, in particular, were both unusually numerous and fine; ices too were in abundance.

Nothing could exceed the animated and general attention of the amiable hostess, who with Mr Watson Taylor, left an impression on the numerous guests, which will not soon be effaced. Instead of haughty condescension which sometimes marks and disgraces similar entertainments, where the ranks of the parties invited are not and cannot be equal — the humblest person invited felt that he had fully participated in the attentions of the day; while those of a higher class received every notice to which they were justly entitled, either by birth, rank, or official station. Of the corporation, or those connected with it, including the worthy mayor (the Rev Mr Bayntun) there were twenty present; the other guests were, our two county members (Mr Bennet and Sir J. D. Astley) T. G. B. Estcourt esq. M.P. for the University of Oxford, Sir Edw. Poore, Col. Baker M.P., Mr T. H. Phipps and son, Mr Warriner, the Rev Mr Edmonstone, and Dr Segrim, who with the host and hostess, and their two elder sons, formed a company of 34 persons. We cannot here omit noticing that the conduct of Mr & Mrs Watson’s two sons, was distinguished by good sense, candour, and manners greatly beyond their years, and did such honour to those who have the superintendence of their education.

The party broke up at about 10 o’clock, (after partaking of coffee) and returned to their respective residences, loud in the praises of what they had witnessed and grateful for hospitalities, which are not likely soon be equalled — still less surpassed.?!

George Watson Taylor appears briefly on the pages of history in the following anecdote. When the Duke of Clarence unexpectedly inherited the throne on the death of George IV he was uninterested in royal etiquette, having spent his life in the Royal Navy or in comparative seclusion surrounded by a large family of illegitimate children. In July 1830, five days after his brother’s funeral, William IV reviewed his Guards _ dressed in uncomfortable uniform, so he decided to change into civilian clothes and strolled down St James Street. Here he met his old friend Mr Watson Taylor and arm-in-arm they were swept along by an enthusiastic crowd; the jolly monarch was even kissed on the cheek by a street walker. Outside White’s Club

15

the crowd brought the two friends to a halt, so members of the club rushed out and escorted the pair to the safety of St James Palace. The King was completely unconcerned and thanking his escort he remarked, ‘Oh never mind all this, when I have walked about a few times they will get used to it, and will take no notice!’*?

The Taylors’ hospitality was crowned on Saturday 23rd October 1830 when they were honoured by a visit from the widowed Duchess of Kent and her daughter Princess Victoria, then aged 11.** The royal visitors changed horses at the Bear Hotel Devizes, where Princess Victoria was shown a drawing by Sir Thomas Lawrence which she later purchased. They arrived at Erlestoke Park at 5 o’clock where they were met by Thomas Moore the poet, and Mr Fisher the Duchess’s chaplain. The whole evening was spent in singing, the Duchess and her daughter singing several duets. The next morning they all went to Church. Later there was a large dinner party, the guests including Lord and Lady Sidmouth, the members for the County, the members for Devizes and the Mayor of Devizes. On Monday morning there was more singing and, after lunch, the guests departed in order to view Stonehenge. This visit from the future Queen Victoria was the apogee of the Watson Taylors’ social success. Nemesis, however, was close at hand.

George Watson Taylor was totally incapable of adapting his life style to his declining income. Shortly after the royal visit he was given leave of absence from Parliament on account of the disturbed state of his neighbourhood, but Taylor ensured his personal popularity by increasing wages, reducing rents and ending the preservation of game.”

The Watson Taylors had tried to ignore the diminution of their income, but the end of slavery and the exhaustion of the soil in Jamaica by overcropping had lowered their income by at least 70 per cent. There had been sales in 1821 and the one already mentioned in 1823 that had raised over £30,000, but the problems continued over the next few years, and finally in 1832 the crash came, bringing down George Watson Taylor and many of his dependants.

The local newspaper commented on his sad dilemma:

Notwithstanding that Mr Watson Taylor was surrounded by a degree of splendour, which it has been well said, might have excited the envy of royalty itself, his mind was scarcely for a moment at ease — he appeared to have an insatiable thirst for something he did not possess. He could not for a moment have thought of the money he was expending.”

16 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

This time everything had to go. The pictures by Rubens, Murillo, Corregio, Parmegiano, Guido, Hobbima, Poussin and Zoffany were sold. So, too, was all the furniture, porcelain and plate, the library of over 4,000 books and no fewer than 5,000 rare exotic plants.*°

By autumn 1832 Watson Taylor was residing in Holland and he left the House of Commons at the dissolution that December. He was never formally declared bankrupt but in 1839 he was still in debt to the tune of over £60,000. He died in Edinburgh in June 1841. His wife outlived him. Fortunately, the Erlestoke and Jamaican estates were entailed and she remained in control of them until her death in 1853.°’

For five years the great house lay empty, until in 1837 it was let, the new tenant being John Cam Hobhouse. A radical politician and reformer, he was a prominent supporter of the Parliamentary Reform Bill of 1832. Now, however, he is best remembered for his great friendship with Lord Byron. Hobhouse was born in Bristol in 1786. His father was a politician and his mother a dissenter. He was sent at an early age to a Unitarian School and then to Westminster School.** From there he went to Trinity College, Cambridge, where he was a contemporary of Byron. Hobhouse and Byron travelled together to Portugal, Spain, Albania, Greece and finally Constantinople. Subsequently Byron dedicated the fourth canto of Childe Harolde to Hobhouse, and Hobbouse was Byron’s best man at the poet’s misconceived marriage to Anne Milbanke.’? Hobhouse was present in Paris when Louis XVIII entered his capital in May 1814, and published an account of the Hundred Days which was notably anti-Bourbon and sympathetic to Napoleon. This led to his first brush with the law when the French translation was seized and the printer and translator were fined and imprisoned.* In 1819 Hobbouse made his first attempt to become a Member of Parliament, standing for Westminster as a Whig. He was defeated by George Lamb, Lord Melbourne’s brother, by 4,465 votes to 3,861.?!

Meanwhile Hobhouse continued to publish anonymous political pamphlets. Unfortunately, one of these, entitled A Trifling Mistake, was held to be in breach of privilege by the House of Commons” and on 14th December 1819 he was committed to Newgate prison. In the pamphlet Hobhouse had asked, ‘What prevents the people from walking down to the House and pulling the members by the ears, locking their doors, and flinging the key into the ‘Thames’, to which he answered, ‘their true practical protectors are to be found at the Horse Guards and the Knightsbridge barracks’. On appeal on 5 February

1820 the Court of the King’s Bench refused to interfere with the Speaker’s warrant and so Hobhouse remained in prison until the dissolution of Parliament on 29 February. After his release Hobhouse stood again and this time he defeated George Lamb by 446 votes.*? On Byron’s death in 1824 Hobhouse, who was one of the poet’s executors, arranged his funeral and persuaded Thomas Moore to destroy Byron’s memoirs in order to keep private Byron’s liaison with his half-sister. **

Hobbouse was very conscious of his diminutive stature and comparatively humble origins. After some abortive efforts at courtship he finally achieved a happy marriage with Lady Julia Hay, the youngest daughter of the 7th Marquis of lweeddale. Sadly Lady Julia was consumptive and the marriage lasted just seven years, leaving Hobhouse a widower with three small daughters.” Thus he decided that in addition to his house in London he would also need a home in the country for his three little girls. He had been acquainted with Wiltshire through his friendship with the Lansdownes at Bowood, and the Methuens at Corsham. Indeed, it was probably Lady Methuen who suggested that Hobhouse should rent Erlestoke Park.*® The little girls were tended by a French governess and took an active interest in village life. At Christmas time 1840, Sir John, as he had become, distributed clothes to nearly one hundred children from Erlestoke and the surrounding parishes while his daughters, Julia, Charlotte and Sophia, waited on the village children serving them with cakes and ale.’

As relaxation from his parliamentary duties Hobbouse’s_ greatest pleasure was entertaining his friends. He moved in the highest circles, including among his close friends King Leopold of the Belgians. When Prince Albert and Queen Victoria desired to raise the cultural standards of their dinner party conversation they consulted Sir John as to which scientists and authors would educate and amuse the royal couple.** Among Hobhouse’s close friends was the Prime Minister, Lord Melbourne, and he was a member of the erudite Holland House set. Visitors to Erlestoke during these years included Thackeray, Thomas Love Peacock and Disraeli with his rich, elderly, silly wife. The normally polite Hobhouse was quite acidic about Mary Anne Disraeli, noting in his diary that, ‘her £3,000 or £4,000 a year are dearly purchased’.*’

In January 1841 Erlestoke Park caught fire and the house would have been destroyed had Hobhouse’s nephew not been woken by the smell of burning. Fortunately, this time the fire was extinguished before much damage had been done.”’ In 1849 his beloved

“AVERY PRETTY SEAT’: ERLESTOKE PARK, 1780-1999

CATALOGUE

OF THE

MAGNIRIGAND ASSAMBLAGH OF PROPERTY

ExrlestokeNWiansio:

NEAR DEVIZES, IN WITS,

ACCUMULATE!

> WHTHAN THIS

FAR-FAMED ABODE OF TASTE ASD Vira,

During the last Twenty Years, at an enormous expense, the whule selected by

GEORGE WATSON TAYLOR, Eso. ™.r.

It is only necessary to observe, that within this classic Residence will be found as extensive a Collection

of objects of superior clegance and taste as that which adorned 1] HE ABBEY OF FONTHILL;

WHICH WILL BE SOLD BY AUCTION, BY

ON-PHE PREMISES,

On MONDAY, the 9th Day of JULY, 1632,

And Twenty sneceeding Days, at Twelve o'Clock, (Sundays excepted),

THE SPLENDID

FURNWEIT

Throughout the Mansion is adorned and fitted up in the most superb style of eclesance ;

THE DRAWING ROOM SUITES comprise very beautiful satin and India silk Preneh curtains, costly carved

and gilt sofas, ottomans, fantenil and route chairs, cheval screens, solid rose-wood sofas and chairs, nd rich

Axminster carpets. THE PURNITURE OF THE DINING ROOM is no less complete. To the LIBRARY

is a range of elegant maltugany bovkcases, winged and single ditto, THE BOUDOSR is furnished with rich Indian silk hangings, &e.

THE BED CHAMBERS

Are fitted up in a very superior manner, The principal Kooms with splendid solid satin wood.oak and mahog+

any bedsteads, with rich silk tabaret and cotton hangings, (altogether upwards of forty,) with bedding com.

plete; winged and single wardrobes of the choicest gatin-wood and imahogany, wilh every other requisite for the Sleeping Apartments.

THE COLLECTION OF VALUABLE AND

CELEBRATED PICTURES,

ERMEBLTS MANY CHIEF DQLUVRES BY

GUIDO PARMEGIANO —ALBANO N. POUSSIN MURILLO = RUDENS HOBBIMA POTTER RUYSDALE CORREGLO GREUZE te. ke,

Gallery of Portraits of Distinguished Characters,

By Sir Joshua Reynolds, Sir Thomas Lawrence, Hagarth, Gainsborough, Debson, Zoifuny, Sir P, Lely, Philips, Mopuer, é&c. aud

THE BOURBON & BUONAPARTE FAMILIES

By Lefevre, Mignaud, Duplessis and Mirevelt. (TURK OVER.)

Fig. 4. Erlestoke sale catalogue, 1832 (WRO 1335/1)

18 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

eldest daughter Julia caught cholera whilst on holiday in Guernsey; she died a few days later at Erlestoke at the age of eighteen. Affairs of State prevented Hobbouse mourning in private, but the Cabinet showed him much kindness, particularly Prime Minister Melbourne."!

Here is not the place to go into close details of Hobhouse’s political career, but he did much to ensure the success of the 1832 Parliamentary Reform Bill. While Secretary at War he was responsible for restricting flogging as a punishment. Also he is supposed to have invented the phrase ‘His Majesty’s Opposition’.*” He was created Baron Broughton de Gyfford in 1859. His later years were consoled by the companionship of his remaining daughters, their husbands and, in time, his grandchildren. His diaries, Recollections of a Long Life, were privately printed in 1865.’ By then, however, another change had taken place, and by about 1860 the Watson Taylors’ financial situation enabled them to return to Erlestoke, so Lord Broughton removed himself firstly to Corsham Court and from there to Tedworth House."

Simon Watson Taylor, like his father, was elected a member of the standing committee of the Planters and Merchants in 1832, when he was only twenty- one. Later he stood for Devizes as a Liberal. Despite supporting the total abolition of income tax, however, he does not appear to have pleased the electors and was only in Parliament from 1857 to 1859.° He married Lady Charlotte Hay, the daughter of the 8th Marquis of Tweeddale, and it was she who was responsible for the demolition of the decayed medieval church and its resurrection by George Street in 1880 in the perpendicular style with a porch tower. In the interior some of the Norman bases of the pillars were reused. The rebuilding cost £6,000 and Lady Charlotte dedicated it to the memory of her father. There is a local rumour that something in the rebuilding displeased her, and that her ghost haunts the mansion.*° Simon Watson Taylor lived to be over ninety years old, surviving into the twentieth century. His obituary notice in this society’s journal reminded readers that for nearly fifty years while the self-elected Corporation of Devizes returned two members of Parliament, one of these had been the owner of Erlestoke Park , and that this process only ended with the passing of the Reform Bill of 1832.”

After Simon’s death in 1902, George Watson Taylor inherited a property that stretched from Brattton to Urchfont, and north to the outskirts of Devizes. These estates remained in the ownership of the Watson Taylor family until, like so many other estates, they were affected by the financial devastation

of the First World War. As a result, the estate was put on the market. It was reported in the Wiltshire Gazette of 25 September 1919 that the estate had been bought by a timber merchant and that after the timber had been felled the estate was to be divided and resold. The new owners were Messrs Green & Co, and the new tenant was a Mr Potter of Croydon. The Wiltshire Gazette understood that he proposed to use it as a ‘Spiritual Healing Home’. Mr Potter was a Christian Spiritualist and Sir Oliver Lodge was a frequent visitor. Potter styled himself as the Rev. J. W. Potter, and although he was short of financial resources he had grandiose ideas. One of these was to turn the Park into the Wiltshire equivalent of Kew Gardens, and to further this plan he circulated articles in order to try and raise the £8,000 necessary for this project. He thought that the garden would provide work for the unemployed, of which there was a multitude in England in 1935, and that Erlestoke would provide an ideal centre for either restful or invigorating holidays.** When Mr Potter ran Erlestoke as a guest house, there was an impression of damp and disorder, mauve distemper and a bath sitting disconsolately in the middle of a passage in one of the wings of the house.’” But the commemoration gardens remained an unfulfilled dream for the Rev Mr Potter, who died unexpectedly while undertaking a lecture tour in America in 1939.”

On the outbreak of hostilities in that year Erlestoke Park was taken over for a Senior Officers Training School. As late as 1940, however, it was still possible to trace the outlines of the original gardens along the paths besides the lakes, the protective iron fences having decayed. After the liberation of Europe in 1944 officers from many of the Allied Countries, including China, Jordan, Iran and Holland undertook courses at Erlestoke. As a Senior Officers Training School after the end of the War, the house remained fundamentally unchanged until the end of June 1950. Then a major fire, originating perhaps in a faulty chimney flue, caused considerable damage to the first floor and completely gutted the second. Fire fighting appliances were brought from as far as Bulford, and the fire was the largest that the Wiltshire brigade tackled in 1950. Water was pumped from the big lake near to Erlestoke church, and it was used by three major pumps to feed the firefighters at the face of the building, where the men of five brigades were directing the jets of water into the heart of the flames. Shortly after the arrival of the first firemen the whole of the roof crashed in. Fortunately there were no casualties and by 11.30 p.m. the fire was under control.’! The Senior Officers School continued in the main wings of the house until

“AVERY PRETTY SEAT’: ERLESTOKE PARK, 1780-1999

1961 when, with the addition of many new buildings in the park, it became a detention centre for young offenders. At the same time the War Office insisted on demolishing the bridges across the road which linked the separate parts of Joshua Smith’s original gardens.

Despite protests from the village the demolition went ahead, the excuse being that the War Office was unable to afford the upkeep of the bridges. Erlestoke was officially opened as a detention centre for young offenders in 1962 and since then it has been a penal institution, originally for young offenders and, more recently, as a category C adult prison.” In 1993 Alison Gomme was in charge of the prison and the television drama “The Governor’ by Linda La Plante was loosely based on Erlestoke prison. Today the old house is almost unrecognisable, surrounded as it is by new red brick buildings. The prison library is in one of the few rooms in the main house to have survived the fire of 1950. It is slightly ironic that the remnants of a classical mansion erected by aWhig magnate to display his wealth and success are now occupied by the failures of our present society.

Acknowledgements

The author would like to express her gratitude to Valerie Cromwell of the History of Parliament Trust for permission to reproduce excerpts from the unpublished account of George Watson Taylor, and for the assistance of Mrs Margaret Parrot and Ms Gill Izzard of the Wiltshire Buildings Record, Mrs Pamela Colman and Mrs Lorna Haycock of Wiltshire Archaeological & Natural History Society’s Library, the librarian of Erlestoke Prison and staff of the Trowbridge Reference and Local Studies Library. Copyright of any material that has been extracted from the History of Parliament is herewith acknowledged.

Notes

1 W. Cobbett, Rural rides, 1885 ed., p. 410

2 VCH Wilts., vol. 7, 1953, p. 85

3 J. Lever and M. Richardson, Great drawings from the collection of the RIBA, c. 1983 (RIBA)

4_N. Pevsner, Wiltshire, 2nd ed., 1975 (The Buildings of England), p. 241

5 J. Britton, The Beauties of Wiltshire, vol. 2, 1801, p. 203

6 R. G. Thorne (ed.), House of Commons, 1790-1820, 1986, vol. 5, p. 197 (The History of Parliament)

7 Ibid. p. 198

8 The Times, 23rd Feb. 1795

9 Centre for Kentish Studies U 840 C90/3

19

10 Univ. London, Institute of Commonwealth Studies, Simon Taylor MSS VI/A/8

11 Ibid.

12 M. Nugent, Lady Nugent’s Journal. . ., rev. ed., 1966 (Institute of Jamaica)

13 R.B. Sheridan, ‘Simon Taylor, sugar tycoon of Jamaica’, Agricultural History, vol. 45, 1971, p. 290

14 Lady Charlotte Bury (i.e. C.S. M. Campbell), Diary of a lady in waiting, 2 vols., 1908, pp. 181-2

15 VCH Wilts., vol. 7, 1953, p. 84

16 Uniy. London, Institute of Commonwealth Studies, Simon Taylor MSS VIHT/A/3

17 WANHS Library, cuttings book 2, p. 180

18 The Times, 11 Feb. 1824

19 S.J. and E. F. Hurwitz, Jamaica: a historical portrait, 1971, p. 30

20 Devizes & Wiltshire Gazette, 16 Nov. 1820

21 Devizes & Wiltshire Gazette, 16 Aug. 1827; 4 Oct. 1827

22 R. Fulford, Royal dukes: the father and uncles of Queen Victoria, 1933, p. 139

23 Devizes & Wiltshire Gazette, 17 June 1897, diamond jubilee edition

24 Ibid. 9 Dec. 1830

25 Ibid. 21 and 28 June 1832

26 Ibid.; sale catalogue in WRO 1335/1

27 Salisbury & Winchester Journal, 1 Oct. 1832; The Times, 25 Feb. 1839

28 Dictionary of National Biography, vol. 9, p. 941

29 T. Moore, Life of Lord Byron, with his letters and journals, 1860

30 Gentleman’s Magazine, vol. 89 (2), 1819, p. 450

31 Dictionary of National Biography, vol. 9, p. 941

32 Parliamentary Debates, vol. 12, pp. 995-6, 998-1004, 1009-26

33, Dictionary of National Biography, vol. 9, p. 942

34 §. Smiles, A publisher and his friends: memoirs and correspondence of John Murray, 2 vols., 1891

35 M. Joyce, My friend H: John Cam Hobhouse. . ., 1948, p. 271

36 Ibid. p. 280

37 Ibid. p. 295

38 S. Weintraub, Albert: uncrowned king, 1997, p. 108

39 Joyce, op. cit., p. 317

40 Ibid. p. 295

41 Ibid. p. 320

42. Dictionary of National Biography, vol. 9, p. 943

43 Ibid. p. 944

44 Joyce, op. cit., p. 347

45 M. Stenton (ed.), Who’s Who of British Members of Parliament: a biographical dictionary, vol. 1, 1976, p. 373

46 Devizes & Wiltshire Gazette, 3 Aug. 1961

47 WAM, vol. 33, p. 69

48 brochure in Wiltshire Buildings Record

49 author’s memories

50 R. Wightman, Rural rides, 1957, p. 133

51 Devizes & Wiltshire Gazette, 29 June 1950

52 information from the Home Office

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Xe Laverstock pottery kilns

Trench 2-7 ' : eae ' \ \

: |

= 2 ee Trench 1 aso

‘'s Arms ,

Ld

Yj ML. & api oe / Vi We fs Li Wi, a / Yi, ae, Ye Why wy Vs WILE Usetisaa

Fig. 1. Site Location

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 20-62

Excavations at Ivy Street and Brown Street,

Salisbury, 1994

by Mick Rawlings’

with contributions by Michael J. Allen', John Chandler’, P. Hinton’,

S. Hamilton-Dyer*, Emma Loader’, Lorraine Mepham’, and Sarah F. Wyles'

Archaeological excavation 1n advance of development revealed evidence of occupation relating to the original establishment and settlement of the city of New Sarum in the early 13th century AD. Parts of several tenements were examined, with the differences in land use in each tenement being clearly defined. A medieval building aligned along the street frontage had at least three rooms and an extension to the rear that contained a cess pit. The building continued to be occupied 1n the later medieval period and examination of the contents of the cess pit of this period revealed much evidence of the

sources of dietary material.

INTRODUCTION

An archaeological excavation was undertaken at the junction of Brown Street and Ivy Street, Salisbury (centred on SU 146 298; Figure 1), on land which was to be redeveloped for residential purposes. The excavation was commissioned and funded by Salisbury District Council, and was carried out by Wessex Archaeology in July and August 1994. It was hoped that evidence of the 13th century AD establishment and early development of this part of the city would be recovered.

ARCHAEOLOGICAL AND HISTORICAL BACKGROUND

The city of New Sarum (Salisbury) was laid out on a regular grid pattern commencing in or around AD 1219 (RCHME 1980), resulting in a series of rectangular blocks known as ‘chequers’. The excavation site lies on each side of the south-east corner of Antelope Chequer, named after the

Antelope Inn which was located in the central part of the chequer. The site is bounded by Ivy Street to the south and Brown Street to the east. Ivy Street is a continuation of New Street and originally bore that name. This street was probably one of the first to be laid out during the development of the new town, and thus it is presumed that the frontages along New Street would have been amongst the first to be occupied.

PROJECT BACKGROUND

The excavation represented the major component of a staged programme of archaeological investigation undertaken in advance of the construction of a new residential development at the site. A major aim of this programme was to locate and record structural remains and deposits of medieval date, especially those relating to the establishment and early development of the city. Although material of later medieval, post-medieval and modern date was also anticipated on the site, it was hoped that information on these periods could be provided by documentary research and thus

1. Wessex Archaeology, Portway House, Old Sarum Park, Salisbury SP4 6EB 2. Salisbury & South Wilts Museum, 65 The Close, Salisbury SP12EN 3. Hillview, Higher Totnell, Leigh, Sherborne, Dorset DT9 6HZ 4. 5 Suffolk Avenue, Shirley, Southampton SO15 5EF

22 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

excavation resources could be concentrated on the earlier levels.

In order to test the archaeological potential of the site, a series of small test-pits was excavated in areas which were to be car parking spaces or courtyards within the redevelopment block. These test pits established that deposits of medieval date were sealed by at least one metre of later material, mostly of 18th century or later date. Prior to the current redevelopment, the land had been used as a car park, with a tarmac surface laid directly over recent demolition deposits.

EXCAVATION METHODS

‘Two trenches were opened at the site; each measured c. 160m° and was positioned directly within the proposed locations of the buildings that comprised the new residential development (Figure 1). In both cases the tarmac surface and the underlying layers of hardcore, demolition debris and modern concrete floors were removed using a 360° tracked excavator fitted with a toothed bucket. The underlying material was then removed in horizontal spits using a toothless bucket until medieval deposits or natural gravels were reached. All further excavation was by hand.

RESULTS

TRENCH 1 (Figure 2)

This was located adjacent to the Ivy Street frontage. During machine excavation no obvious floor surfaces were observed, and only one length of wall was recorded; this was quickly investigated but was found to be of post-medieval date. Following the completion of the machine excavation, all of the trench sections were cleaned and recorded, and all features seen cutting into the basal gravels were investigated. The overall depth of excavation varied between 1.3 and 1.75m below current ground levels and for reasons of health and safety, a 1m wide step was created at a depth of 0.9 - 1.0m below the current surface level.

The lower 0.3-0.4m of the trench fill sequence comprised dark loamy soils which contained pottery of 13th and 14th century date. Along the southern section, close to the Ivy Street frontage, two slot trenches, each 1m wide, were excavated by hand through the lower step. Both of these slot trenches found that along this edge of the trench the lower 0.3 - 0.4m of the sequence was made up of a series of thin spreads of compact deposits.

These were predominantly clays and gravels, along with at least two layers of crushed and puddled chalk and occasional lenses of darker, more organic material. Although some of these deposits produced no artefacts at all, the pottery that was recovered was consistently of later 12th - early 14th century date.

A number of the excavated features in Trench 1 were found to be of medieval date. A large subcircular pit (288) was cut into the gravels to a depth of 1.05m and was shown to have just penetrated the current groundwater level. Bulk soil samples taken from two of the lower fills (281, 287) were discovered to be waterlogged and detailed analysis has indicated that these deposits contain material typical of that found in cess pits. Pottery of 13th - early 14th century date was found in these lower fills, along with a fragment of Purbeck Marble which appears to be from an architectural piece.

A second pit (234) 0.7m deep was cut wholly into the upper fills of pit 288. Although this later feature was also subcircular in plan, it was lined internally with a number of roughly-hewn chalk blocks to form a square setting measuring 1.2m? which was recorded to a height of two courses or 0.6m. These blocks were set against a framework of horizontal timbers which were well-preserved in situ. Again the fills were found to be waterlogged and to contain pottery of 13th - early 14th century date, but the subsequent environmental analysis of these soils indicated that they did not contain materials indicating the presence of cess.

To the north of the two pits described above was a group of three shallow, irregular, intercutting pits. The earliest and deepest of these (291) was 0.5m deep and filled with a silty deposit which contained six sherds of pottery of 13th - early 14th century date. The upper part of the pit had been truncated by a second pit (290) which again contained medieval pottery. The latest pit in this group (233), however, contained sherds of later medieval and post-medieval date as well as some medieval pottery. It is possible that these pits were originally for the purpose of gravel extraction or some other function rather than refuse disposal.

A further shallow pit (323), this one having distinctive squared corners, was located partially within the eastern baulk section. Excavation showed this to be 0.6m deep and it was clearly cut through the loamy medieval soils which formed the basal part of the sequence in this part of the trench. The sides and base of the pit were concave and the fills (321, 322) were dark and humic and again appeared to

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

sainjeay [je

€

I youaly, °c Shy

Bas

\

(- | | | & 81S [PULequr Jo 693 = | 8 i. a : :

-~—--~.. — wr Yuen jo e6py See eee ie ' ua yo e6P3 s

aS |BULEIUI JO BBpy ~~

JEHOW eee

sajnpou july = Ze aunyeay 10)e7

syooiq Mey = AL. | a S490|q HEY 7

saunyee} ! payepun pue jersipaw—jsog Pra sainpeay |eAdipay\| oy

, youasy | Youddy

-—_——__ ° aa eae wed

24 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

contain cess or similar materials. Only one sherd of 13th — 14th century pottery was recovered, but several fragments of roof tile, including two glazed pieces, are also thought to be of medieval date.

In the north-eastern corner of the trench was a large rounded pit (192); this was also cut through the medieval loamy soils which made up most of the lower step in this area. The full size of this feature was not ascertained, but examination of the main trench section revealed it to be at least 0.7m deep. The main fill (194) contained three sherds of 13th — 14th century pottery along with a large number of cattle horncores, probably representing the waste products from industrial activity such as tanning.

Adjacent to the northern baulk of the trench were two more pits. These were both rounded in plan, although the full outline of the earliest one (284) could not be defined as it had been truncated by a later feature. Pit 284 was c. 0.75m deep and excavation of part of the fill sequence resulted in the recovery of sherds of 13th —14th century pottery and fragment of roof tile. A single sherd of post-medieval pottery was also found but is considered to be intrusive.

The southern edge of pit 284 was cut by a much more shallow example; pit 165. This was only 0.2m deep; the single fill (166) contained a substantial amount of medieval roof tile fragments and two sherds of 13th century pottery.

Several of the features investigated in Trench 1 were found to be of post-medieval date in addition to pit 233 mentioned above. Two of these were conjoined and comprised square pits lined with chalk blocks, both located adjacent to the northern baulk (301, 302). These were preserved to a depth of c. 1m and the central party wall included two courses of brick in addition to six of ashlar chalk blocks. Both this wall and the west wall of pit 301 were mortared, suggesting that these formed a single unit, possibly a successor to pit 302. No dateable artefacts were recovered from pit 302, although several roof tiles were recorded in the section. Pit 301, however, contained pottery of 18th — 19th century date along with pieces of glass of a similar date.

A further pit (157), this one very square or rectangular in plan, was located partially within the western baulk. It was 0.75m deep and was excavated 0.4m into the basal gravels. Pottery of late 17th — 18th century date was found in several of the fills of this feature, along with other materials (glass, clay pipe, roof tiles) of similar date. The layers at the very base of the feature were waterlogged and some pieces of wood were preserved in these deposits. Another

chalk-lined square or rectangular pit (147) was located to the south but was not dated, although it certainly cut through the medieval soils which formed the lower part of the sequence.

At the southern end of the main eastern baulk was a circular chalk-lined well c. 1m in diameter (176). Although this was not investigated and thus no artefacts recovered from within it, the level from which the feature was initially excavated indicates a post-medieval date, probably 17th - 18th century.

Projecting from the Ivy Street frontage at the southern end of the trench was a section of wall footing recorded for a total length of approximately 4m (306). This was 0.5m wide and made up of a single course of chalk blocks set in a mortar matrix, with occasional large flint nodules. Close examination of the material below the wall and also of the main southern baulk revealed that the wall footing was definitely of post-medieval date. Although there were associated surfaces on either side of the footing, none of these were within buildings (until very late in the sequence) and thus the footing probably represents a property boundary wall. A number of later walls or footings were built directly on top of this one, indicating the continuity of this boundary. The most recent of these was made up of five courses of red brick and was directly below the modern tarmac surface of the car park.

Summary

No evidence of medieval buildings was found during the investigation of Trench 1. Instead, the frontage along Ivy Street appears to have been left as open ground, with a series of compact dump deposits which may represent attempts to raise the ground level, possibly in order to avoid the high water table. In the backlands a few pits were excavated, at least two of which were cess pits, but over much of the area soils were able to develop. This area could have been a garden or orchard.

In the post-medieval period a boundary wall was established running from the frontage into the backlands, but no major buildings were constructed. Again a few pits were excavated in the backlands, along with a well. Only two sherds of later medieval pottery were recovered from this trench, although the paucity of material of this date is a common phenomenon within the city and does not necessarily indicate a lack of activity.

TRENCH 2 (Figure 3 and 4) This trench was excavated to a depth of 1.1m-1.3m below the current surface level, but it was not

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

Trench 2 Medieval and later medieval features : men J) e. Ecoe!

ke Modern Petrol tank

ia

Mortar

——}) Chalk blocks

°°, Flint nodules

854 Chalk surface

ap Greensand blocks Gravel surface

Later feature

Post-medieval features 2 Edge of trench pe eared SS - 898 Modern ' i i ' \ Peirol tank rs] pe 907

Paving

_ >

832

840 a 836 i as : 838

pone , {uwSCS Ferruginous sandstone ¢3 ¢?” Greenstone | ( ae Ss ee

oye: Flint nodules a », Tile 0 5 10m t I i I | — |

Fig. 3. Trench 2, all features

26 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

necessary to create any internal steps in the baulk sections. The north-west corner of the trench was not fully investigated due to the presence of a large subsurface petrol tank sealed within a concrete bund. This was disused and had been filled with gravel. At the eastern end of the trench, adjacent to the Brown Street frontage, the remains of a medieval structure were uncovered and excavated. The northern boundary of this building was actually a tenement plot wall which continued through into the backlands.

Although a number of small postholes were found below the chalk floors of the earliest building, not all of the floors were removed and so it is not clear if these postholes represented some form of structure preceding the one which was excavated. A detailed analysis of the pottery from this sequence could not distinguish any difference between the material found below the floors and that from within or above them; all of the sherds recovered were of 13th - early 14th century date. Also below the chalk floors, but sealing the postholes, were thin interleaving spreads of clayey gravel and more organic material which probably represent levelling or make- up material and are thus part of the initial phase of construction.

The medieval building at the east end of the trench was aligned along the Brown Street frontage and was at least 7.5m long and 3.5m wide. Estimation of the original position of the medieval frontage below the current pavement area suggests a true width of c. 5.5m for the building. At the northern end of the structure an extension to the rear resulted in a potential original width of c. 9.5m. This extension was found to be an integral part of the building from the initial phase of construction rather than a later addition, although examination of the floor deposits suggested that it may have been a separate room.

The external walls of the building were founded on dwarf wall footings (857, 979, 985), each of which comprised a band of gravel and mortar c. 0.3m wide with occasional flint nodules set into this mixture. There was some evidence for a northern wall directly abutting the main tenement wall (984), but this was fairly ephemeral and certainly the west wall (985) of the building directly abutted the tenement wall and appeared to be bonded with it at foundation level. Within the backlands, the tenement wall footing showed evidence for several stages of alteration and rebuilding along its recorded length. In this frontage area, however, the wall footing was of consistent build type and thickness, and no alterations were noted.

Where investigated, the wall footings were found to be placed directly on the basal gravels and had not been placed into any cut or trench through the natural deposits. Large crudely-worked blocks of greensand were placed at some wall junctions, presumably at the base of larger structural members within the main timber frame of the building. Within the narrower part of the building the ground area was divided up into smaller rooms by two narrow partition walls represented by beam slots. The most southerly of these (854) appeared to have two small postholes along its length, suggesting the use of vertical timber supports. The other beam slot (964) formed the southern side of a wider area of unfloored ground, suggesting that the partition wall had something built against it prior to any floor being laid in this area. The building thus had at least two rooms along the frontage; a northern one measuring c. 4.5 by 5.5m and a southern one measuring c. 2.5 by 5.5m. The edge of a third room was also uncovered at the southern edge of the trench. The rear extension measured 3.7 by 2.1m.

The floor surfaces were made up of crushed chalk and were generally between 40 and 80mm in thickness. These deposits were patchy and showed signs of repair and replacement. Pottery recovered from the surface of these floors was of 13th - early 14th century date. Although described here as floors, it is equally plausible that the actual floors were made of timbered planking suspended above these surfaces and thus the chalk would have been a sealant of underlying materials rather than a floor in its own right.

Located within the rear extension was a rounded pit (966) which was 1.5m deep. Most of the fills of this pit had been removed by a chalk-lined example of later medieval date (590). Pit 966 measured c. 1.1m in diameter and the waterlogged basal fills (777, 965) contained sherds of 13th - early 14th century pottery. Analysis of one of the fills showed the presence of excreta in addition to the remains of a variety of plant species. This pit did not appear to be lined, and it was almost certainly an integral feature within the building, although it clearly was cut partially through the edge of the southern wall of the rear extension.

A second medieval pit (932) was recorded outside the building but adjacent to the rear extension. This example was sub-circular in plan, measuring c. 1.8m in diameter but only 0.45m deep. A large quantity of 13th - early 14th century pottery was recovered from the fills of this pit, along with roof tiles and animal bones. A few sherds of post-

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

medieval pottery found in the upper fill layer are assumed to be intrusive.

To the north of the tenement wall there was no evidence of buildings located adjacent to the street frontage. Instead, a series of thin medieval soils lay directly over the basal gravels and no features were recorded in this area. Cutting through these soils was a square pit (926) lined with blocks of chalk. Although the outer dimensions of the feature resulted in a pit measuring 1.6m square, the lined part of the pit was only 1.15m square internally. The pit was excavated to a depth of 0.8m but the base was not reached. Each of the ashlar chalk blocks measured c. 300 x 150 x 200mm, with the latter measurement representing the height of each course. In some places ceramic tiles had been used as a levelling material between courses.

The upper 0.65m of the pit was filled with a homogenous deposit (925) which contained pottery of both 13th — 14th century and 15th —1 6th century date, suggesting that this deposit was of later medieval date. The lower fill (958) was much darker and more organic, and analysis of the plant remains from this deposit indicated that this feature was indeed a cess pit.

A similar feature was located in the rear extension of the building to the south of the tenement wall, cutting through and removing most of the medieval cess pit here (966). This later pit (590) was also lined with chalk blocks and had internal dimensions of 1.2 x 1.1m. Excavation proved the full depth to be 0.9m below the surface of the basal gravels, and at the base of the pit was a floor made of two layers of planks, the uppermost aligned east/west and the lower one north/south. Overlying the timber floor was a thin layer of chalky material followed by a very dark and organic deposit (594). Once again, analysis of this material indicated that much of it was composed of cess. This deposit was sealed by a further thin layer of chalky material, above which were two fills (692, 591) which appeared to represent refuse disposal. Pottery of 15th — 16th century date was recovered from the upper fill (591), indicating a later medieval date for the feature. A few pieces of medieval tile were found in the layer of cess but are probably residual. Two copper alloy pins were also found in these upper fills, one of which has been

identified as a later medieval type. The clear difference between the areas either side of the tenement wall was maintained in the backlands. To the south of the wall and to the rear of the building, a number of pits dated to the post- medieval period were recorded. One of these (770)

Qik

cut through the south wall of the rear extension of the medieval building. This feature was sub-square in plan, measuring 1.1m across and 0.75m deep. The lower part of a wooden barrel had been placed in the base of the pit, rising to a height of 0.55m and then provided with a ‘rim’ of ceramic roof tile fragments and flint nodules. The wooden staves of the barrel were very degraded and could not be retained for analysis, although an L-shaped iron fitting was recovered and could be part of this construction. Although exclusively medieval pottery was recovered from this pit, this material is considered to be residual as the pit clearly cut through an adjacent pit (587) which was definitely of post-medieval date.

Pit 587 was circular in plan, measuring 0.85m in diameter. It was only 0.4m deep and had vertical sides and a flat base. The upper fill appeared to contain lenses of cess along with a considerable quantity of pottery of 17th — 18th century date. Another pit (585) immediately to the south-west was only partially excavated and was found to be irregular in form. This pit was also of post-medieval date (1 6th —17th century) and appeared to have been used for refuse disposal.

In the frontage area, both within and without the medieval building, was an east/west alignment of six postholes (Figure 3; 830 - 840). These cut through the floors and the western wall of the earlier building, and although one of them (838) contained a sherd of 13th century pottery and another (836) a sherd of 15th century date, this group is thought to be of post-medieval date. The postholes varied in recorded depth from 0.06 to 0.24m and may not be exactly contemporary, but probably represent a linear structural feature such as a fence.

Also located within the earlier building was a circular chalk-lined well (865). This lay in the area between the main part of the building and the rear extension, suggesting that the building was no longer in use when the well was first constructed. This feature was excavated to a depth of 0.7m below the surface of the basal gravels, and the homogeneous fill contained much pottery of 17th — 18th century date.

Further into the backlands area, however, beyond the pits described above, the trench was not extensively excavated. Most of this area appeared to be occupied by large and amorphous pits of 19th — 20th century date (e.g. 559, 867) cutting through post-medieval dump deposits. The main feature investigated in this area was a surface (869) made up of limestone flags surrounded by red bricks,

28 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Trench 2

7

Modern petrol tank

maat

LY

=) JDO

{i

Be fe SSSeasnSses3

Se ySsc

Loo

ICC jcc

oe

Brick Chalk blocks

Fig. 4. Trench 2, Paving 869

located in the north-west corner of the trench. No walls were recorded in association with this surface, but the chalk rubble layer which formed the bedding for the surface contained a single sherd of 17th century pottery which provides a terminus post quem for the feature. This surface is likely to have been an exterior yard rather than a floor within a building. ‘To the north of the tenement wall the differences between the tenements were again clear. In the small part of this backlands area which was available for excavation, a number of wall footings (882, 907, 978) were aligned perpendicular to the tenement wall. There were no clear surfaces associated with these footings but they were cut through medieval soils similar to those recorded at the frontage in this area. They were also obviously later than the tenement wall, which showed signs of having been rebuilt and realigned in this area. These wall footings probably represent small outbuildings of post-medieval date. The remains of a small pitched-tile hearth or flue (898) were also recorded in this area and were probably a feature within one of these outbuildings.

Summary ‘Trench 2 was found to contain parts of two separate tenements. In the northern one, soils which formed

Flint nodules

Ferruginous sandstone

in the medieval period were cut by a later medieval chalk-lined cess pit. Within the backlands area, small outbuildings of post-medieval date were constructed up against the tenement wall, and one of these contained a small pitched-tile hearth or flue.

South of the tenement wall the recorded deposits were very different. Along the frontage was a medieval building with internal partition walls and a rear extension within which was a cess pit. This building utilised the tenement wall as the main north wall, and otherwise comprised dwarf walls or footings on which the timber superstructure would have been built. The internal cess pit was replaced in the later medieval period by a chalk-lined version similar to the one recorded to the north of the tenement wall.

In the post-medieval period a number of pits were excavated to the rear of the building, and an alignment of postholes cut through the floors and the west wall. A well was also constructed within the earlier building. In the backlands a series of modern pits had disturbed the post-medieval and medieval deposits, but a flagstone floor of 17th — 18th century date represented an exterior yard surface.

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

THE FINDS

THE METALWORK by Emma Loader

The assemblage of metalwork comprises 161 objects; 147 of iron, 10 of copper alloy and four of lead. Of this total, 39 objects came from medieval contexts. All iron and copper alloy objects have been X- radiographed. Most of the objects which have been dated to the post-medieval period are not discussed here; details may be found in the archive.

IRON OBJECTS

The 36 iron objects from medieval deposits comprise one knife, 30 nails and five objects which were too corroded for any positive identification.

A whittle tang knife was recovered from the make- up of one of the wall footings (857) for the building adjacent to the Brown Street frontage in Trench 2. This is acommon medieval type, comparable to knives found, for example, at Clarendon Palace (Goodall 1988, fig. 74, 4). The X-radiograph of this object suggests that it is constructed from iron with a steel cutting edge. The purpose of combining the two metals was to produce a knife with a sharper blade but which also did not break easily. Iron is more ductile than steel, but it cannot be sharpened to produce a good blade, whilst heat-treated steel produces a good cutting edge (Wilthew 1987).

Tylecote (1981) identifies four methods of combining steel and iron to produce blades, and all four methods were commonly used in the production of knives at this time. The method of combining the two metals in this instance is comparable with his type B whereby the steel cutting edge was welded onto the iron strip. This method enables the blade to be continually sharpened until eventually the steel wears away and the knife is discarded. The indication that the steel has not been worn away suggested accidental loss rather than intentional discard, as the object was still functional.

Thirty nails were recovered from a number of medieval features. Classification has been made by visual examination of the objects and the X- radiographs. Three types were identified; round-

headed (seven examples), flat-headed (four examples) and T-headed (one example); all types have square shanks. Other nails were unidentifiable. All types were probably used in woodworking, and no horseshoe nails or large masonry nails were observed.

29

Of the five unidentified objects, two are lumps of iron, possibly large nail heads. Another object is possibly an L-shaped fitting, which was recovered from the base of post-medieval pit 770 in Trench 2, in which the remains of a wooden barrel were identified. Another L-shaped piece of unknown function was recovered from the fill of a beam slot for one of the partition walls within the medieval building adjacent to Brown Street in Trench 2. The fifth object consists of five fragments of iron and is completely unidentifiable. It was recovered from a medieval soil layer immediately to the rear of the same building.

COPPER ALLOY OBJECTS

The two objects of copper alloy recovered from the earlier deposits comprise one pin and one pin/needle, both found in the fill of pit 590 in Trench 2 and associated with pottery of 15th— 16th century date. The pin is very fragmented, with a head formed by wrapping wire once or twice around the shaft and shaping this to create a globular head. This type is comparable to the small type 2 pins found at Colchester (Crummy 1988) and is likely to be of later medieval date. The pin/needle has a flattened head with three small perforations. The end of this object is hooked, probably unintentionally, and its function is uncertain.

LEAD OBJECT

A flat triangular piece of lead was recovered from a soil layer to the rear of the medieval building along the Brown Street frontage in Trench 2. It has two cut marks at the wider end and is probably a waste fragment from a larger object.

THE POTTERY by Lorraine Mepham

The complete pottery assemblage recovered during the excavations at Ivy Street/Brown Street amounts to 1671 sherds (24,712g), all of medieval to post- medieval date, of which 1069 sherds (13,775g) from stratified medieval contexts have been analysed in detail and are discussed here.

This is an assemblage of relatively modest size, but as it represents the first stratified medieval pottery assemblage of any size from Salisbury to be published, analysis has been undertaken in some detail in order to establish a type series for Salisbury which can be used as the basis for future analyses.

30 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Reference will be made to other, as yet unpublished, assemblages from Salisbury (Mepham and Underwood n.d.) in order to support chronological and other conclusions drawn here.

METHODS

Analysis of the medieval pottery has followed the standard Wessex Archaeology recording system (Morris 1992), involving the identification of fabric types on the basis of the range, size and frequency of macroscopic inclusions, and the grouping of these types according to the dominant inclusion type or known source. In this instance the fabric series 1s based around the Laverstock-type fabrics coded within Wessex Archaeology’s ‘established wares’ series (Group E); other fabrics fall into two fabric groups: sandy fabrics (Group Q) and limestone- tempered fabrics (Group C). Fabric totals are presented in Table 1.

Vessel forms have been defined using rims and other diagnostic sherds, and follow the recommended nomenclature for medieval vessel forms (MPRG 1998). Details of surface treatment, decoration, manufacture and evidence of use have

Table 1: Pottery fabric totals (medieval contexts)

Fabric No. Weight % of type sherds (g) group

% of total

LAVERSTOCK TYPE COARSEWARES

E422a 9 140 1.5

E422b 222 4627 49.7

E422c 432 4539 48.8

Sub-total 663 9306 67.5

LAVERSTOCK TYPE FINEWARES

E420a 25 323, 7.6

E420b 74 684 16.2

E42la 130 1263 29.9

E421b 68 1301 30.8

E421c 83 655 15-5

Sub-total 380 4226 30.7

OTHER FABRICS

“Tudor Green’ 2 6 2.5

C400 1 2 0.8

F400 1 8 33

Q400 4 76 31.3

Q401 18 bowl 62.1

Sub-total 26 243 1.8

Overall total 1069 13775

also been recorded; detailed pottery records by context are held in archive.

Pottery from post-medieval contexts has not been subjected to the same level of analysis, but has been scanned for medieval types not represented amongst the stratified assemblage. Post-medieval pottery types are briefly summarised at the end of this report.

In the fabric descriptions below, the terms used to describe the density of inclusions follow Terry and Chilingar (1955), and are defined as follows: rare (1-3%); sparse (3-10%); moderate (10-20%); common (20-30%). The fabrics identified are discussed below within groups according to known or putative source.

LAVERSTOCK-TYPE WARES

The overwhelming majority of the medieval assemblage comprises coarseware and fineware sherds which are comparable to products of the Laverstock kilns located just outside the city (Musty et al. 1969). The coarsewares are sufficiently visually similar throughout the medieval assemblage as to represent variations of a single fabric (E422); this has been subdivided here into three fabrics on the basis of the size of the quartz inclusions. Two basic fineware fabrics have been defined (E420, E421); again, both have been subdivided on the basis of inclusion size. These subdivisions are somewhat arbitrary, but have been made in order to determine whether variations in coarseness may be explained by chronological factors, or whether they are influenced rather by vessel form.

The putative source for these fabrics is, of course, the Laverstock kilns. The excavated kilns, however, have a restricted estimated life span of 1230-75 (ibid. 93), although there 1s indirect documentary evidence for the operation of kilns in the Laverstock/ Clarendon Park area during the period 1318-23 (bid. 83, footnote). References to pottery production near Milford Bridge in 1270 probably apply also to Laverstock (Robinson 1988, 170).

It is apparent that Laverstock-type fabrics continued to be produced after the known date of the excavated kilns, and there is no evidence either from this assemblage or from other medieval assemblages excavated in the city that other pottery types made significant inroads at any point into the Laverstock monopoly. Tantalising hints at pottery production within the city itself were revealed during construction of the inner ring road in 1972/3 - a pit filled with possible potting clay and containing a 13th century baluster jug was observed at London Road/ Rampart Road (WAM 68, 137), and a group of

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

‘wasters’ recovered elsewhere along the route (not mentioned in the WAM notes, but cited by Spoerry (1990, 3); this may be the basis for Robinson’s reference (1988, 170) to a kiln at Guilder Lane, but there is no other mention of kiln structure(s)). Two ‘wasters’ are also known from West Grimstead and Ashley Hill near Petersfinger (1bid. 170).

Whilst it is highly likely that a production centre existed in or close to the city throughout the medieval period, either at Laverstock or nearby, it should be pointed out that visually similar coarseware and fineware fabrics are common throughout south-east Wiltshire and east Dorset. Recent chemical analysis (Spoerry 1990) has failed to differentiate between coarseware pottery samples from Laverstock and south Dorset, while a documentary search has demonstrated the likelihood of the existence of further medieval production centres exploiting the clays of the Reading Beds and London Clay which outcrop in a band from south-east Wiltshire to Purbeck, particularly in the area of the post-medieval Verwood industry (Spoerry 1988).

Coarsewares

E422a Hard, moderately coarse matrix; common, fairly well sorted, subangular/ subrounded quartz, sometimes iron-stained, <l1mm; rare iron oxides. Handmade; firing varies from completely oxidised to completely unoxidised; ‘pimply’ surface finish.

E422b As E422a but with quartz <0.5mm; slightly ‘pimply’ surface finish.

E422c As E422a but with quartz <0.25 mm.

The range of coarseware vessel forms is limited, and consists largely of jars, generally round-based although some have a barely discernible basal angle. The term ‘jar’ is used here deliberately, in line with recommended nomenclature (MPRG 1998), and in preference to ‘cooking pot’, although from the presence of sooting on many of these vessels a cooking function may be implied. Four main jar rim forms may be defined:

Type 1: Everted, long, slightly thickened, rounded rim; everted at approximately right angles to the body; the _ orientation of the rim varies from flared to almost horizontal, to give either a rounded or more upright (convex) body profile (Figure 5, 1). Equivalent to Laverstock rim type I (Musty et al. 1969, fig. 7).

Type 2: Everted long, slightly thickened, rounded rim as

31

Type 1, but with a slight lid-seating (Figure 5, 2). Equivalent to Laverstock rim type II (ibid.).

Type 3: Everted, long, thickened and slightly moulded rim, sometimes with slight lid-seating, and with a more noticeable neck zone than Types 1 and 2 (Figure 5, 3). Broadly equivalent to Laverstock rim type II (ibid.).

‘Type 4: Everted, short simple rim, squared, sometimes with groove along the top to give an almost bifid profile (Figure 5, 4).

Jars with rim type 1 represent the survival of a form which is common in the Salisbury area at least from the late 11th century, and they were found in some numbers in 12th century pits at Laverstock. However, their occurrence in such numbers at Ivy Street/Brown Street and their association with glazed jugs in the Laverstock kiln groups amply demonstrates that they continued in use well into the 13th century. The other three rim types, all attested at Laverstock, do not appear before the 13th century.

Jars are frequently scratchmarked on both exterior and interior surfaces below the rim. This scratchmarking ranges from deeply incised marking which is restricted to exterior surfaces to a finer marking which occurs on both exterior and interior surfaces. Generally speaking, the coarser the fabric variant, the deeper the scratchmarking, the most pronounced marking nearly always (but not exclusively) occurring on vessels in fabric E422a. There is no certain evidence of glaze on any of the jars, and only one rim is embellished in any way; a type 3 rim with finger impressions around the inside of the rim (Figure 5, 3). All vessels are handmade, although the rims are invariably wheel-thrown or at least wheel-finished.

Variants of the basic jar form include two handled jars. The first was probably double-handled, and has a convex profile, flattened everted rim and vertical loop strap handle(s) (Figure 5, 5); a closely comparable example came from a 12th century pit at Laverstock, recut probably in the 13th century, (Musty et al. 1969, fig. 9, 31). The second is a jar with a type 4 rim, slightly concave to produce a slight lid-seating, with a vertical strap handle springing from the rim. The vessel is partially glazed internally. Again this may have been double-handled, and could have been a cauldron, examples of which were also found at Laverstock in 13th century contexts (ibid. fig. 11, 48).

Other vessel forms are uncommon. There are two

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Fig, 5.

c > | eS

7

Medieval Pottery

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

dishes, both in the finer variant E422c. One is partially glazed internally, and has stabbed decoration along the top of the rim (Figure 5, 6); other internally glazed sherds may derive from similar open vessels. The illustrated example is sooted on the exterior, indicating a probable use in cooking. Two thickened rims from open vessels with slightly convex profiles (Figure 5, 7) are likely to represent bowls - comparable forms are known from Laverstock, some of them handled (Musty et al. 1969, fig. 12, 55) - although the possibility that they could be curfews should not be entirely ruled out. In addition, one acute base angle from an inturned or ‘West Country’ dish, in fabric E422c, was recognised (ibid. fig. 11, 41-4).

Finewares

E420a Hard, fine matrix; common, well sorted, subrounded/subangular quartz, not iron stained, <0.5mm; rare iron oxides. Handmade or wheel-thrown; oxidised (very pale-firing: buff to cream).

E420b As E420a but with quartz <0.25mm.

E42la Hard, fine matrix; moderate to common, well sorted, subrounded quartz, iron-stained, <0.5mm; rare iron oxides. Handmade or wheel-thrown; oxidised (pale-firing: buff to pale salmon pink).

E421b As E421a but with quartz <0.25mm.

E421c As E421b but quartz not iron-stained.

A similarly restricted range of vessel forms is found in the fineware fabrics; these are represented almost entirely by jugs of various forms and decoration. These jugs have a long upright or slightly everted neck, occasionally collared, and a flattened rim of triangular profile; handles are of strap or rod form (Figure 6, 8-12). Most vessels are at least partially glazed, on the upper part of the body, and glaze is either colourless (appearing yellow) or an even to mottled apple green. No complete profiles were reconstructable, which has hampered close dating since the chronological progression from baggy, rounded body profiles to more slender, baluster forms during the lifetime of the kilns has been _ demonstrated at Laverstock (Musty et al. 1969, 112). One specific, more unusual form, however, could be identified: a strut jug with incised struts and rod handle, sherds of which were recovered from various contexts (Figure 6, 10). Strut jugs were found in three of the Laverstock kilns which fall in

33

the middle of the chronological sequence (ibid. fig. 21, 166-8).

Some chronological clues may also be gained from the decorative treatments. The earliest groups of jugs from Laverstock have a range of linear and curvilinear combing, applied foliage motifs, and stamping straight onto the body or on applied clay elements; one example from within the medieval building on the Brown Street frontage has ring-and- dot stamps around the bridge spout and on clay pads around the rim (Figure 6, 9). This is closely comparable to a jug from the earliest kiln group at Laverstock (ibid., fig. 13, 68), and there are other examples of similar stamping as well as a few comb- decorated sherds.

Later jugs display a much greater range, although often less well executed, of applied and stamped motifs. At Ivy Street/Brown Street slipped decoration is well in evidence: pellets, pads, scales and strips, which themselves may be stamped, impressed or incised. The slip is either brushed or trailed on in a thin solution, or applied in a thicker, more plastic form; the slip colour is generally red, contrasting with the pale body colour, although purple-brown (manganese) colouring is also used (e.g. Figure 6, 8). The fragmentary nature of the assemblage means that decorative schemes are rarely reconstructable; the most complete example displays a vibrant design of applied scales and stamped pellets (Figure 6, 12).

Other fineware vessel forms are limited to a small number of shallow saucer-shaped forms, internally glazed, occasionally with pulled or pinched lips (Figure 6, 13); these are probably the bowls from saucer lamps (see Musty et al. 1969, fig. 24, 185).

NON-LOCAL FABRICS

Four other fabrics were identified, all coarsewares, one limestone-tempered, one flint-tempered and two sandy. They are described as follows:

C400 Soft, moderately coarse matrix, slightly micaceous; common, poorly sorted, subangular limestone fragments <lmm; rare iron oxides. Handmade; unoxidised.

F400 Hard, moderately coarse matrix; sparse, poorly sorted, subangular flint <2mm; rare subrounded quartz <0.5mm; rare iron oxides. Handmade; oxidised (pale- firing).

Q400 Hard, moderately coarse, micaceous matrix; sparse, poorly sorted, subangular/subrounded quartz <1mm; sparse, poorly sorted, subangular limestone <]mm; rare iron oxides. Handmade; oxidised with unoxidised core.

34 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Fig. 6. Medieval Pottery

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

Q401 Hard, moderately coarse, micaceous matrix; moderate, fairly well sorted, subangular/subrounded quartz <0.25mm; rare subangular limestone fragments <0.5mm; rare iron oxides. Handmade; oxidised (pale-firing) with unoxidised core.

All four fabrics occur in very small quantities, constituting between them less than 2% of the total medieval assemblage by weight (see Table 1). Two rim sherds are present: one probable jar rim, everted and externally thickened, with a thin colourless external glaze, in fabric F400; and a triangular sectioned rim from a long-necked jug or pitcher in fabric Q400. Other sherds are plain, unglazed and undiagnostic. The micaceous wares C400, Q400 and Q401 find general parallels amongst assemblages from sites in north-west Wiltshire, such as Knook shrunken settlement, Market Lavington and Warminster (Mepham 1993; Mepham forthcoming; Smith 1997); such wares have a potential source in the putative production centre at Crockerton near Warminster. Micaceous fabrics are recorded from various 12th century contexts at Laverstock, Gomeldon and Old Sarum (Musty et al. 1969; Musty and Rahtz 1964; Musty and Algar 1986), but outside Salisbury have a longer life-span, at least into the 13th century. Flint-tempered fabrics are also known in north Wiltshire and fabric F400 could have a similar source area and date range, although a source closer to Salisbury cannot be ruled out. This fabric does not have a micaceous matrix, and the quartz grains resemble those from the Laverstock type coarsewares described above.

LATER MEDIEVAL/POST-MEDIEVAL POTTERY

Later medieval/post-medieval pottery has been examined in less detail, and has been quantified by broad fabric type, on the basis of known wares, e.g. Verwood-type earthenware, or ware groups, e.g. stonewares (Table 2). The difficulties of isolating later medieval wares are explored further below, but for the purposes of discussion certain transitional later medieval/early post-medieval types are included here; these are visually quite distinct from the medieval wares described above.

“Tudor Green’ ware, which is not a single type, and is a somewhat misleading term, is now generally taken to refer to thin-walled vessels in a fine white fabric without visible inclusions, and was probably made at more than one centre within the Surrey/ Hampshire border industry from the late 14th

35 Table 2: Pottery from post-medieval contexts Ware No. sherds Weight (g) Residual medieval PIER 2219 “Tudor Green’ 1 2 Redwares 8 228 Slipwares 5 44 Verwood 193 3993 early Verwood 23 396 Staffs/Bristol types 4 Wi Stonewares Pa 1634 Tinglazed earthenware 21 102 Whitewares 7, 316 Creamware 12 544 Agate ware 1 4 White saltglaze 19 46 Porcelain 3 6 Total 333 7330

century into the 15th century (Pearce and Vince 1988).Three sherds were found at Ivy Street/Brown Street.

Verwood-type earthenwares, as produced in the Verwood area of east Dorset from at least the mid 17th century (Algar et al. 1979); documentary references to medieval pottery manufacture in this area, as noted above, are not as yet supported by archaeological evidence. The group from Ivy Street/ Brown Street includes probable examples of products from the earliest known kiln at Horton, distinctive by the marked iron flecking within the glaze (Copland-Griffiths 1989, plate 1), and also examples of the anomalously-named Wiltshire Brown ware, an 18th century type made at several of the kilns (Algar et al. 1979, 16). Verwood-type earthenwares are the most common post-medieval type found at Ivy Street/Brown Street, and occur mainly in a range of bowl forms; one example of a small handled bowl has an interesting graffito on the underside of the base, possibly the owner’s initials (Figure 6, 14).

A small group of sherds have been tentatively identified as earlier products of the Verwood kilns, of later medieval or early post-medieval date. These wares are visually similar to the Verwood type earthenwares, but are slightly coarser. They are more frequently unglazed but where glaze is present this often has a pitted appearance and is more unevenly applied than on the later wares. On other sites in Salisbury, these wares have been found mainly in thin-walled vessel forms, including jars, and are frequently associated with “Tudor Green’ and other

36 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

transitional later medieval/early post-medieval wares (Mepham and Underwood n.d.).

Coarse redwares, both glazed and unglazed. A number of potential sources are known, including Crockerton near Warminster in north Wiltshire and Wanstrow in east Somerset; there were probably similar centres in Hampshire. Redwares are relatively scarce in Salisbury, occurring mainly in early post- medieval contexts (Mepham and Underwood n.d.); from the 18th century the paler-firing Verwood-type earthenwares swamped the market almost to the exclusion of other coarsewares.

Slipwares, here consisting of redwares decorated with trailed slip. These are characteristic of the West Country production centres such as Wanstrow and Donyatt, which may be seen here as supplying a specific type not produced by the Verwood kilns.

Staffordshire or Bristol type slipwares and brown- glazed wares of 17th— 18th century date; cups/bowls and plates/dishes in a pale-firing fabric, with a clear lead glaze over trailed slip decoration, and one cylindrical mug in a similar fabric with a streaky brown (manganese) glaze. In this instance Bristol, being the nearer source, is the more likely.

Tinglazed earthenwares, here considered as a single group, with no attempt to distinguish English from imported wares, although the group from Ivy Street/ Brown Street is unlikely to contain any imports.

Stonewares, here grouped together. They include examples of German wares, such as Raeren, ?Frechen and Westerwald, as well as English wares, which here include the later white salt glaze.

Industrial wares of the mid 18th century and later, which here include small quantities of agate ware, creamware, pearlware, porcelain and later fine whitewares.

CERAMIC SEQUENCE/CHRONOLOGY

The construction of a ceramic sequence for Salisbury, either overall or for single sites, has previously been hampered by the relatively small quantities of medieval pottery recovered, and by the even smaller proportion recovered from undisturbed medieval contexts. From nine sites excavated up to 1990, fewer than 1,000 medieval sherds came from undisturbed contexts (Mepham and Underwood n.d.). There has also been a general dearth of

diagnostic material and a scarcity of either good stratified sequences or well- stratified groups (i.e. pit groups). This problem has been compounded by the lack of comparative material from kiln sites which post-date the late 13th century, and the life-span of vessel forms observed within the 13th century kiln groups is therefore unknown.

The assemblage from Ivy Street/Brown Street unfortunately does little to redress this imbalance. The proportion of stratified medieval pottery from this site is relatively high, yet within this group little or no definite ceramic sequence can be discerned.

The best stratigraphic evidence for the medieval occupation of the site comes from the building adjacent to the Brown Street frontage. From this sequence, however, only 187 sherds were recovered, the overwhelming majority (172 sherds) from the levelling/make-up layers below the chalk floors. This earliest group includes the full range of Laverstock- type coarseware and fineware fabrics, in jar and decorated jug forms which are well paralleled amongst the published 13th century kiln groups (Musty et al. 1969). The jugs carry decorative schemes of both ‘earlier’ (e.g. Figure 6, 9) and ‘later’ type (e.g. Figure 6, 10). Succeeding phases within the building produced insufficient pottery for the construction of any possible ceramic sequence.

Pottery from other medieval layers and features at Ivy Street/Brown Street would all likewise fit within a 13th century date range, which leaves an obvious hiatus in the ceramic sequence for the 14th and 15th centuries, even allowing for the continuation of the 13th century vessel forms described above into the early part of the 14th century, as is entirely possible.

This dearth of later medieval assemblages has been noted elsewhere in the city, although the apparent increase in popularity of the fineware fabrics E420 and E421 at the expense of the coarseware fabric E422 later in the stratigraphic sequence at certain sites has been tentatively taken as a real phenomenon, possibly linked to the increasing use of metal cooking vessels in the later medieval period (Mepham and Underwood n.d.). This is more difficult to demonstrate at Ivy Street/Brown Street, since the fineware fabrics are well-represented from the earliest contexts, and it may be that the differing coarseware/fineware ratios observed across the city have instead a functional explanation.

The difficulties of ascertaining the date range of Laverstock type wares has already been discussed, and the possibility that these wares continued in production and use far beyond the known life-span of the excavated kilns must be considered. However,

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

given the almost complete absence at Ivy Street/ Brown Street of any vessel forms in these wares which might be considered as later medieval, such as bunghole jars or pitchers, or bifid rim jars, the apparent hiatus would appear to be real. There is just one exception - a jar rim (type 4) in fabric E421b from pit 590 with a broad horizontal applied thumbed strip, a decorative trait paralleled, for example, in early 16th century forms from the Donyatt production centre (Coleman-Smith and Pearson 1988, fig. 53, 1/124).

This is not to say, however, that later medieval pottery is completely absent from the site. A handful of sherds of ‘Tudor Green’ ware, as well as sherds which have been tentatively identified as early Verwood wares, fall within a potential date range of late 14th to mid 16th century. Features which might date to this period include pit 590 and pit 926.

Post-medieval contexts are, unsurprisingly, dominated by the well-known Verwood type earthenwares of the late 17th century onwards. Other coarse earthenwares are conspicuous by their absence and those redwares which are present occur either as slipwares, a type not produced by the Verwood kilns, or as plainwares likely to pre-date the massive expansion of the Verwood market in the 18th century. Other wares - stonewares, tinglazed earthenware and later industrial wares - indicate the continuation of the sequence through the 18th century and into the modern period.

THE CERAMIC BUILDING MATERIALS by Emma Loader

The assemblage of ceramic building materials comprises a total of 572 pieces (32,463g) from features dated to the medieval, later medieval and post-medieval periods. They were quantified for each context by the following types:

flat roof tile, including peg tile ridge tile

finial

floor tile

brick

ME SS

No complete bricks were recovered, and this category is not discussed further here.

Bi,

ROOF TILES

Flat roof tiles formed the majority of the ceramic building material and made up 86% by weight of the total assemblage. The most commonly occurring diagnostic fragments were peg tiles with round holes, and, less frequently, square holes. The latter were restricted to post-medieval pit 233 in Trench 2. Of the total number of roof tiles, 27% (by number) were glazed, the colour of the glaze ranging from clear to mottled green or greenish-brown. Fabric analysis has not been carried out on the tiles, but it may be observed that the predominant fabric of the tile is poorly-wedged, sparsely sandy with prominent iron inclusions and is conspicuously pale-firing, ranging in colour from buff/pink to pale orange. A number of tiles appear to be burnt, indicating probable reuse in the make-up of hearths.

The main areas of manufacture of roof tiles for this region extends from Hampshire to Wiltshire, along the line of the Reading Beds (Hare 1991).The fabric of the tiles produced, and the methods of firing, produce widely-distributed tiles which are very similar in appearance. It is difficult therefore to accurately identify the source of the tiles. Production of roof tiles in Wiltshire was dominated by kilns at Alderbury for the first three quarters of the 14th century, and this centre may have been in operation by the 13th century, supplying Clarendon Palace (ibid. 89). It is probable that many of the tiles recovered from Ivy Street/Brown Street were manufactured at this centre, which is about 5km (3 miles) to the south-east of the city; most consuming centres would have been within 8-16km (5-10 miles) of the tile kilns (ibid. 97).

RIDGE TILES

A small number of ridge tiles were recovered, making up less than 1% by weight of the total assemblage. The majority were glazed with a green glaze although no other diagnostic features were present. One fragment shows the remains of a knife-cut triangular crest, a feature of ridge tiles which first occurred in the Oxfordshire region in the late 13th century; the fabric is conspicuously sandy and pale-firing, and matches the pottery fabrics of the Laverstock-type coarsewares described above. Evidence of ridge tile manufacture has been found at Laverstock (Musty et al. 1969).

FINIALS

The evidence for other roof fittings is rare, with only three possible fragments of finials being identified, two of which are green-glazed and

38 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

probably derive from the same object. The third fragment is in a sandy fabric comparable to the Laverstock-type coarseware pottery fabrics, and may derive from the same source.

FLOOR TILES

Floor tiles are less common and make up 10% by weight of the total assemblage. The majority are unglazed and undecorated, with only one decorated fragment. This fragment features a four- petal rosette which forms part of a larger design. Possible sources include the Clarendon and Naish Hill production centres which were both in operation during the 13th and 14th centuries, but the fragment is too small for positive identification. Floor tiles from some contexts appear to have been burnt, suggesting that these tiles were reused in a hearth.

THE GLASS by Emma Loader

The glass assemblage comprises a total of 45 pieces (1613g) and includes one almost complete bottle, as well as other bottle and window glass fragments. All of the glass is of post-medieval date.

The bottle glass dates from the mid 17th century or later. Identifiable bottles include one almost complete onion bottle of late 17th - early 18th century date (Hume 1961, type 4/5), at least one other onion bottle of unidentifiable type, and a cylindrical bottle with a high domed kick, probably late 18th — early 19th century (ibid. type 20/22). A small clear glass phial of early 18th century date, probably used for pharmaceutical purposes, was also recovered.

THE STONE by Emma Loader

Two fragments (2,265g) of worked stone were recovered during the excavations; one of these was found in the fill of a posthole sealed below the floor of the medieval building in the frontage area of Trench 2, the other was in the fill of pit 288 in Trench 1, thus both pieces and are clearly dated to the medieval period. Both of these fragments are of Purbeck Marble and have at least one worked surface; they are assumed to be architectural pieces, although of uncertain origin.

THE CLAY PIPE by Emma Loader

The assemblage of clay pipes from this site comprises 74 pieces (372g), 62 pieces of which are plain stems and 12 are diagnostic bowls. Five of these bowls have makers’ marks:

1. THO/HVNT incuse on stem of pipe within a shield. Spurred bowl, polished. Thomas Hunt is known as a Marlborough pipemaker. His work spans the transition of bowls with marked heels to those with spurs. This particular stamp dates to 1685 - 1750 (Atkinson 1965, fig. 2).

2. Three pipes have the stamp used by the Gauntlet workshop in Amesbury. Two stamps are represented: the ‘monkey’s paw’, dated to 1670-80, and the plain hand, a mid-17th century stamp (Atkinson 1970, 179). All three pipes appear to have been polished, and their forms appear consistent with those produced at Amesbury, although it is not impossible that they may be imitation pipes given the large number of Gauntlet imitations being produced during this period and later.

3. Large serif letters IM moulded in relief, on a spurred bowl. The stamp is that of John Morgan I, a Salisbury pipe maker, and is datable to between 1800-20 (Atkinson 1970, appendix E).

The remaining seven bowls range from forms with a small bowl and pedestal foot, dating to 1600-40, through to the later spurred pipes produced in the late 18th century.

THE WORKED BONE by Emma Loader

‘Two fragmented, undecorated combs were recovered from the site. One was a one-piece, double-sided comb with coarse teeth on one side and fine teeth on the other (Figure 7). This was found in the fill of the construction trench for the 17th — 18th century well (865) in Trench 2. Similar bone combs have been found at High Street C, Southampton, and are dated to 1630 - 1640 (Platt and Coleman-Smith 1975, fig. 249). The second comb was found in the fill of the post-medieval pit 233 in Trench 1, although there are also artefacts of medieval date from this feature. It is possibly made from antler or horn, and the teeth of the comb are aligned with the grain, resulting in a stronger comb (MacGregor 1985, 28). It is uncertain whether this comb was originally a one-piece, double-sided comb or a single-sided

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994 39

O 20mm {/-———~e!

Fig. 7. Bone Comb, Well 865

comb, since teeth are present on one side and on the opposite side cut marks are present, but the break appears to have been smoothed over and polished. The comb may have been reutilised following the loss of teeth on one side. There are no comparable antler combs from Southampton sites, although MacGregor (ibid. 81) notes that earlier medieval combs were often of horn or wood, these materials being preferable to bone, and it is possible that this comb is of medieval date.

THE LEATHER by Lorraine Mepham

Leather was recovered from two features, medieval _ cess pit 288 (Trench 1) and later medieval cess pit 590 (Trench 2).

Cess pit 288 produced a complete turnshoe sole with oval toe, together with part of the upper; the style of shoe is unknown. A second turnshoe sole, of similar form, came from cess pit 590, together with

five small waste pieces, probably representing debris from cobbling.

THE WOOD by Lorraine Mepham

Samples of worked wood were retained from waterlogged contexts within three features excavated in Trench 1 (medieval cess pits 288 and 323, and post-medieval pit 157).

The pieces from cess pit 288 (from lower fill 281 and from a secondary fill) comprise six small plank fragments, all approximately 20mm thick and ranging in length from 100 to 340mm, which may originally have formed some kind of lining to the pit. All the planks are tangentially split, and as far as can be ascertained, have been well shaped. No joints or traces of methods of attachment were observed.

One small fragment (65 x 35 x 35mm), possibly part of a squared stake, came from cess pit 323. One end appears to be sawn, but otherwise this piece is too small for detailed identification.

Two plank fragments, plus nine further small fragments which probably originally derived from planks, were retained from post-medieval pit 157. The two planks are both tangentially split. The larger of the two (455 x 35 x 10mm) is well-shaped, with one slightly convex surface, tapering slightly at both ends; this could be a barrel stave which, as for pit 770, may have formed part of the lining of the pit.

THE ENVIRONMENTAL MATERIALS

THE PLANT REMAINS by Pat Hinton

A total of thirteen bulk samples were taken from sealed and datable contexts. All were processed for charred plant remains using 10-15 litres of sample by standard flotation methods. The flot was retained on a 0.5mm mesh and the residues on a 1mm mesh. Following preliminary processing it was obvious that six of these samples were waterlogged. The treatment of these samples was therefore modified and after fractionation (>2.5mm - 0.5mm) they were stored in Industrial Methylated Spirit (IMS).

40

Table 3: Plant Remains

Key: (_) = identification uncertain

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

charred

mineralised _frs = fragments

*

totals estimated.

Feature Context Sample volume (litres)

234

288

590,

287 594

| 235 10

281 10

10 10

592 10

966 7717 10

Cereals

Triticum cf aestivum s.1. - grains - rachis

fragment

Hordeum vulgare L. - grains

- rachis frag Secale cereale L. Avena sp.

ent

Cerealia indet.

bread wheat

hulled barley

cereal grain fragments

Pulses

Vicia faba L. - seed fragments - hila Pisum sativum L. - seed fragments -hila

broad bean

Ficus carica L.

Rubus cf idaeus L. Rubus fruticosus agg.

Fragaria vesca L.

Prunus spinosa/ insiutia

Prunus domestica s.1.

Prunus avium (L.) L.

wild strawberry 5 4 3200* 13" sloe and bullace 108 + frs 10 + frs 28 + frs

wild cherry

Malus sylvestris/domesuca - seeds

crab or cultivated

&

GRE ‘ a z eo a

ii \o nw — {bt Sina =n

Rumex sp.

Brassica cf migra black mustard 6*(60*) Dye Coriandrum sativum L. coriander Laz Foeniculum vulgare (Miller) fennel 1 Anethum graveolens L. 20h ay ea a | 6” Apium graveolens L. (ane es ee es oe ae Cannabis sativa L. hemp 3 Linum usitaussimum L. Arable &/or ruderal Papaver rhoeas/dubium | —{| — 23% Fumaria sp. 2 Uruca dioica L. stinging nettle — |--}2+—— 22m Chenopodium album L. fat hen aw ae ee ee) 54% ole Fata ee common orache chickweed a Ee ial Agrostemma githago L. Boor 31 irs: frs Persicaria Japathifolia (L.) Gray 13 Polygonum aviculare s.1. knotgrass 3 Fallopia convolvulus (L.) A.Léve black bindweed 6 eas Rumex cf crispus curled dock 5 lie hos Rumex obtusifolius L. broad-leaved dock i Sa ee ae eee ewan an |

Malva cf moschata

Malva cf sylvestris

musk mallow common mallow

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

field pansy/violet

=

smooth tare

Euphorbia helioscopia L.

Vicia cf tetrasperma Vicia/Lathyrus sp.

vetch or vetchling

Scandix pecten-veneris L. shepherd’s needle

Aethusa cynapium L.

fool’s parsley

Bupleurum rotundifolium L.

1

| _Lamium cf purpureum | reddead-newle | | |_Galeopsis terrahitL. | commonhemp-newtle | | |_Lithospermum arvenseL. | field gromwell | | | Centaurea cyanusL. | comnflower_ | || | Sonchus cfoleraceous | smooth sow-thistle ||

Chrysanthemem segetum L.

Tmpleurospermum inodorum (L.)

Schulz-Bip

UW

thorow-wax

Bromus hordaceous/secalinus

Grassland Ranunculus acris/repens/bulbosus Vicia sauva L.

buttercup 1 common vetch 1)

cf Trifolium sp. clover aS etoae

Pasunaca/Heracleum sp. parsnip or hogweed Torilis japonica (Houtt.) DC upright hedge-parsle

Myosots sp.

forget-me-not 1

Prunella vulgaris L.

Plantago lanceolata L. ribwort plantain

| Euphrasiasp. eye bright Odontutes vernus/Euphrasia sp. red bartsia/eyebright

Euphrasia sp.

Cirsium cf eriophorum

self -heal

woolly thistle

Cirstum cf vulgare

spear thistle

Cirsium sp.

Hypochaeris radicata L. Leontodon autumnalis L. autumn hawkbit =a

cf Crepis sp.

thistle

hawk’s-beard

;

;

_ im

16*

70*(12) 28*

_ _

:

;

_

Lactuca serriola L.

prickly lettuce

Anisantha sterilis (L.) Nevski Poaceae

Scrub or wood margin plants

Salix sp. - buds

Sambucus nigra L.

Stachys sylvatica L. | hedge woundwort_ |_|

barren brome

grasses 2(4)

liillowsse eee tet 2 3

elder

cf Viburnam Jantana

wayfaring tree

Heathland plants

Preridium aquilinum L. - pinnules

Calluna vulgaris (L.) Hull - shoots - flowers - seeds

Erica cinerea L. - leaves - flowers

bracken

bell heather

Potenuilla erecta (L.) Rausch

tormentil

Damp/wet area plants

Sphagnum sp. - fragments

bog moss

Musci indet. - stem frag Caltha palustris L. Ranunculus flammula L. Lychnis flos-cucult L.

Apium nodiflorum (L.) Lagasca Bidens cernua/tripartita

Potamogeton cf polygonifolius

eee) | Conium maculatumE, “| hemlock | | aa

indeterminate mosses marsh marigold

lesser spearwort ragged robin

fool's water-cress nodding or trifid bur marigold

bog pondweed

41

42 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

common or slender spike-rush

Eleocharis palustris/uniglumis

Carex cf demissa

common sedge

Carex nigra (L.) Reichard

jistlesesntee fff

sedges

The ‘flots’ and residues were large, several >500ml, but in all cases the whole of the larger fractions were searched. The smaller fractions were scanned but in the case of large numbers of very small seeds only those from sub-samples were counted. Sorting and most identifications were made with a stereo microscope at x 7-40 magnification. Higher magnifications with transmitted light were used when studying some plant tissues. For identification it was often necessary to allow items, a few at a time, to partially dry so that surface features could be more easily seen. Extracted seeds (the term is used loosely to include fruits, caryopses, nutlets etc.) were then stored in alcohol and the remainder returned to IMS. Fruit stones were allowed to dry and later measured by eye-piece graticule, taking thickness to be the distance between the ventral and dorsal edges and breadth to be from side to side (Behre 1978). Nomenclature follows Stace (1991), using scientific names only at first mention.

In Table 3 all taxa are represented by seeds unless otherwise stated. Totals are unavoidably inaccurate but are used to provide some means of comparison between samples or contexts. Some seeds are accompanied by many fragments.

The remains are listed in terms of categories and most likely original habitats of the seeds but, as always, this cannot be applied too rigidly. Many plants in the arable, ruderal and grassland categories are not restricted to specific conditions and some wetland plants could have occurred occasionally as weeds in damper patches of fields.

PIT 234 —TRENCH 1

A sample from the main fill (235) consists mainly of compacted lumps of soil, bone, charcoal and small fragments of wood, mostly reddened; very few seeds were identified. There are numerous small, black, apparently charred, fungal sclerotia. These are mostly spherical, (c. 0.5mm - 2.0mm diameter), which when broken show a relatively thick wall and dense amorphous core. A few are lighter in colour and possibly have not been burned.

PIT 288 —TRENCH 1

The sample from fill layer 281 is very different and the contents are characteristic of cess pits. They consist largely of light brown concretions ranging from very

small crumbs up to 20 — 30mm, appearing as parts of a compacted mass. Most are friable but some are partly or completely mineralised. Many can be seen to incorporate small bones, stem and leaf fragments, and occasionally stones of Prunus species (sloes, plums) etc. and also voids left where fruit stones, seeds, and other fragments have become dislodged.

Apart from this compacted material there is a large volume of free plant remains, some of which are mineralised. The main constituents are fine shreds and occasionally larger fragments of charcoal, wood, twigs and monocotyledonous stems which have not been identified. Identified items include plum and sloe stones, seeds and endocarp (core) fragments of Malus sp. (wild or cultivated apple) and some Ficus (fig), Fragaria (strawberry) and Rubus sp. (blackberry/ raspberry) seeds. Other edible plants are represented by the seeds of Brassica nigra (black mustard), Apium graveolens (wild celery) and Anethum graveolens (dill) which may have used as spices or flavourings. Two detached hila are the only identified evidence of Vicia faba (broad/field bean) but small featureless fragments may be parts of legumes.

Cereals are represented only by a mineralised wheat grain, a few charred fragments and one seed of Bromus sp. (chess/rye brome). A mineralised seed of Linum usitatissimum (flax) may well have been consumed but possibly the plant was grown for fibre. There are small numbers of characteristic field weeds, of which Agrostemma githago (corn cockle) represented mostly by fragments, is the most significant. The only variants to these plants and weeds of supposedly domestic origin are a shoot of Calluna vulgaris (ling, heather) and a few fragments of moss, suggesting usage of heathland plant material.

Another sample from the basal fill (287) of the same feature, with quantities of flint and chalk rubble, has a similar content of charcoal and wood, but there are more mineralised concretions. Plant remains are similar to those in the later fill but there is a wider range of seeds of weeds and ruderals. Both Triticum (wheat) and Hordeum (barley) are present as charred grains, and the many fragments of corn cockle suggest that these were ingested with cereal products. Fragments of heather and of Pteridium aquilinum (bracken) are further indications of heathland, with more fragments of several species of mosses.

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

Both of the samples from pit 288 consist largely of faecal material and comprise obviously digested plant foods. Other items probably represent the disposal of kitchen and other household rubbish.

PIT 590 —TRENCH 2

A sample from the basal fill layer (594) is also composed of very characteristic cess pit components, i.e. compacted masses, some of which can be broken down and many others which are mineralised. These include very small shreds of stems or other vegetation, and as before incorporate fruit stones, smaller seeds and bone fragments. Cereals occur only as mineralised and are uncertainly identified but there is a large amount of comminuted vegetable material. Little attempt has been made to identify this but examination at x200 and reference to Dickson (1987) shows that there are many fragments of testa and integuments of cereals, i.e. ‘bran’. Other fragments are probably remains of other leafy foods.

Plum and sloe stones, whole, and in very many cases smaller fragments to only c. 2mm, seeds of apples and of Vitis vinifera (grapes), figs and the wild fruits such as strawberry and blackberry or raspberry form a large part of the assemblage. Beans and Pisum sativum (peas) appear only as detached hila and probable testa fragments, occasionally incorporated in a buff-coloured matrix which is almost certainly excrement. There are fewer seeds of field or grassland, but these include several very characteristic field weeds, in particular mineralised whole seeds and many fragments of corn cockle. No heather, bracken, or mosses were found in this sample which appears to consist almost entirely of food remains.

Another sample from a later fill (692) contains little plant material. There are two fragments only of charred probable cereal, a few mineralised seeds of fig and strawberry and fragments of sloe or plum stones. These appear to be chance inclusions of probable kitchen debris.

PIT 966 —TRENCH 2 The sample taken from the basal fill (777) is more informative since it contains not only obvious excreta but, with the food remnants, more seeds of crop weeds, grassland and ruderals. Charred wheat and barley, and one mineralised grain of Secale cereale (rye), ‘together with seeds of corn cockle, Centaurea cyanus (cornflower), Chrysanthemum segetum (corn marigold), and Anthemis cotula (stinking chamomile) are present in greater numbers than in the previously examined samples. Large fruit stones are absent from this sample but there are some seeds of smaller fruits

43

and possible spices. There is a larger presence of heathland plants, and plants not only of damp or wet grassland but, in the form of Potamogeton (pondweed) seeds, possibly open water.

PIT 926 —-TRENCH 2

A sample of the lower fill (958) included very many compacted lumps of faecal material and mineralised concretions. In some it is possible, as before, to detect traces of included seeds, insect pupae or voids left by their loss. Loose seeds found in the sample, with the exception of Cannabis sativa (hemp) and a few weed seeds (which may well have been inadvertently consumed), are almost entirely those of fruit or other edible plants, mostly preserved by waterlogging but many by mineralisation.

Fruit stones and seeds of figs, grapes and gathered soft fruits are abundant and seeds of flavourings slightly more numerous than in other samples. More unusual is aseed of Euphrasia sp. (eyebright), a complex series of plants usually of grass or damp heathland, but an infusion of this plant has been used medicinally Johnson 1862). Sub-fossil seeds of Euphrasia species and Odontites verna (red bartsia) are difficult to distinguish but in this case the mineralised seed is perfectly preserved and definite identification possible. Apart from one small carbonised shoot tip, heathers and bracken were not found in this sample but there are a few fragments of one moss species. However, some fragments of concretions incorporate lengths of what may be mineralised fragments of a moss.

DISCUSSION

Of the seven pit fills examined, two appear to be refuse deposits with only few chance inclusions of plant remains, but the other five are closely associated with the likely original use of the pits. The contents are typical of cess-pits of medieval date and, although there are many stem and straw fragments, the greater part consists of food residues which have passed through the human alimentary system. Remains of non-edible plants form a lesser part.

The larger fruit stones of plums, sloes and cherries and the small seeds of other fruits are the most noticeable components. Domesticated plums, distinguished by their greater size and relatively lower breadth are less in evidence than the smaller wild sloes. The sloe stones range in size from 7.0 x 5.3.x 5.8mm to 13.5 x 9.0 x 10.1mm and their range of measurements is similar to that from fruits collected recently in Wiltshire, Gloucestershire and Dorset. The modern hedgerow fruits included the very astringent small sloes through to larger and sweeter bullaces or

44 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

wild plums, reflecting the many cross-fertilisations of native and introduced species.

Despite their size, the stones may have been swallowed and there is unequivocal evidence of this from York (Kenward and Hall, 1995) where a sloe stone (max. diameter 9mm) was found embedded in human faeces. The degradation of fragments of Prunus stones in these samples, particularly that from the lower fill of pit 590, also suggests that these may have been digested. The apple seeds have not been distinguished as from wild or cultivated trees. Blackberries, raspberries and strawberries would have been gathered from the wild but the fully formed seeds show that the figs were imported. Grape seeds may perhaps have arrived in imported dried fruits but vines could have been in cultivation locally.

However, the large quantities of fragments indicate that a great part of the consumed food was in the form of leafy vegetables, salads and cereal products. The very small fragments of cereal bran have not been identified but charred and mineralised grains indicate wheat, barley and rye. Bread, or pottage, made from incompletely cleaned grains, could account for the many testa fragments of corn cockle and probably for many of the other seeds.

Weeds such as corn cockle (which has toxic effects if eaten regularly), Scandix pecten-veneris (shepherd’s needle), Bupleurum rotundifolium (thorow-wax), cornflower, Anthemis cotula (stinking chamomile) and Chrysanthemum segetum (corn marigold), which have now disappeared from cereal crops were once frequent contaminants.

Brassica seeds are notoriously difficult to identify and the separation of ‘vegetable’ types such as cabbage, turnip etc. from B. nigra (black mustard) is not often possible. Occasionally, however, the reticulation on the testa surface of some of the Brassica seeds is comparatively conspicuous and suggests the higher cell walls of black mustard. These are perhaps more likely to occur in cess-pits than the seeds of plants of which only leaves or roots are eaten, but the identification of many is uncertain. Other plants which might have been used as flavourings include Coriandrum sativum (coriander), Foeniculum vulgare (fennel), dill, and celery. The first three are not native plants but had been introduced earlier to England (Harvey 1981).Wild celery grows in wet soils although more frequently found in brackish conditions nearer the sea.

The seeds of flax may have been consumed when used in food preparation but it is probable that this plant, and also hemp, were grown for fibre. As these plants require steeping in water (often near the site of

cultivation) at a preliminary stage in processing, they are not uncommonly found in places where preservation of plant remains is by waterlogging. Urtica dioica (stinging nettle), apart from the young leaves used as a pot-herb, also provides fibres which have been used as a substitute for, and even considered superior to, flax (Johnson 1862).

Other plants may have had uses which are unfamiliar today, not only as food but perhaps for medicinal purposes. Hyoscyamus niger (henbane) provides a narcotic and Conium maculatum (hemlock) has also been used for this purpose, presumably with considerable care since this is a particularly poisonous plant. Even the leaves of sloes and wild plums, in addition to the use of the fruits, may have been used as an infusion; they were so used in the nineteenth century for adulterating tea. The bark too can be used as an astringent medicine and the flowers as a laxative. In fact, there is scarcely any plant which does not have some use, whether nutritional, medicinal, magical, or as a source of dye, fibre or fuel.

Bracken and heathers, prominent in samples from pit 966 and in lesser evidence in one from pit 590, were presumably gathered from heathland for use as bedding, flooring or fuel, but the seeds of wet grassland and water-side plants seem inappropriate, and even more so the pondweed seeds. Caltha palustris (marsh marigold), Lychnis flos-cucull (ragged robin), Eleocharis sp. (spike-rush) and the Carex (sedge) species suggest wet grassland but Apium nodiflorum (fool’s watercress) and Bidens species (bur marigolds) are more common at muddy stream sides. Pondweeds are aquatics, normally growing in either still or flowing waters and a possible explanation might be that they were introduced with water carried in from nearby watercourses, used for domestic or hygiene purposes, and discarded in the cess-pits. However, Potamogeton polygonifolius, identified from Jessen’s (1955) key, although typically growing in shallow water such as streams and ditches, frequently grows sub-terrestrially in flushes (Preston and Croft 1997). This plant is a calcifuge so its provenance may well have been the same as the bracken and heathers.

Mosses may have been introduced with the heath or wetland plants, perhaps unintentionally, possibly as wipes for toilet use. If the mineralised fragments in the sample from pit 926 are indeed moss this might account for their presence in this faecal deposit. Alternatively they, and other damp loving plants, may have been part of the immediately surrounding flora but this is perhaps unlikely in the case of Sphagnum sp. (bog moss)

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

found in pit 966. Time has not permitted further work on moss identification.

Stems of grasses, cereals and other plants occur in all cess-pit samples, and may indicate bedding, thatch or fodder. Seeds of grassland plants are more numerous in a couple of samples but are not sufficient to confirm the presence of hay.

There is some variation in the cess-pit contents in terms of species represented and numbers of seeds. The sample from pit 966, for instance, includes only a small number of fruit seeds but a greater range of seeds of non-edible plants, whereas in the sample from pit 926 the latter are scarcer and food plant remains are more abundant. These differences might be explained in terms of usage, if one household enjoyed a more varied diet than a neighbour, or conditions of preservation.

Similarly there is little by which the users might be distinguished or their social status inferred, unless the absence of less common fruits such as Mespilus (medlar) and Morus nigra (mulberry) could be taken to indicate a more lowly status. Both trees were valued in medieval gardens (Harvey 1981; Roach 1985), but the quoted inventories are usually those of aristocratic or monastic gardens. Alternatively it may be that evidence is lacking at this site, and it must be emphasised that many fragments, particularly those that are mineralised, are unidentified and even with more time may remain so.

Compared with the results from similar cess-pit deposits of this period, e.g. Worcester (Greig 1981), Chester (Greig 1988), Taunton (Greig 1990), Hastings (Hinton 1993), Canterbury (Hinton unpublished) and from documentary evidence (Green 1984), these conform in many respects. Cereal bran, arable weed seeds, fragments of beans and peas, plum, sloe and cherry stones and seeds of gathered soft fruits are almost universally present and often associated with other plants such as flax, hemp, heather, bracken and mosses, and occasionally wetland species.

THE MARINE SHELLS by Sarah F. Wyles and Michael J. Allen

A total of 156 shells were retrieved during the course of the excavations, and a further four were recovered from bulk soil samples. The shell, although fairly ubiquitous, was never recovered in large quantities from any one context. Only four deposits produced more than 10 shells and the largest assemblage was only 18. Generally all of the shells were in good condition and the majority were oyster (Ostrea

45

edulis). There was also one scallop (Pecten sp.), one whelk (Buccinum undatum) and occasional mussel fragments (cf. Mytilus edulis).

THE FAUNAL REMAINS by Sheila Hamilton-Dyer

Much of the faunal material derives from pits and is of medieval date. A smaller amount of post-medieval material is briefly noted. The remains were from both hand-—excavation and sieved samples.

METHODS

Identifications were made using the modern comparative collections of S. Hamilton-Dyer. The fragments have been recorded to species and anatomy where possible, undiagnostic fragments have been classified as horse/cattle-sized (LAR) and sheep/pig-sized (SAR). Some small fragments could not be ascribed to any group and are recorded only as mammalian (MAM). Measurements were taken using a vernier calliper and are in millimetres. In general these follow the methods of von den Driesch (1976) for mammals and birds, and Morales and Rosenlund (1979) for fish. Sheep withers heights are based on factors recommended by von den Driesch and Boessneck (1974). The archive gives full details of the individual bones and includes further information on butchery, ageing, sex measurements and so on not in text. Archive table Al lists all the species identified and their printout codes.

RESULTS

Over 2,000 bones were recovered for study, mainly from Trench 2. Sieving of bulk soil samples contributed almost half of this total, and almost all of the fish bones, which formed the bulk of the sieved bone. Much of the material is in excellent condition and includes several bones with an ivoried appearance. A few bones had been burnt and some had brown/green concretions adhering, typical of cess deposits. Gnawing is at a low level and mainly localised in just a few contexts.

Mammal species include the expected large domestic ungulates; cattle, sheep and pig, as well as some remains of dog, cat, hare, rabbit, and fallow deer, but no horse. Bird bones are frequent and mainly of fowl and goose. Many bones are of fish, mainly eel. Occasional bones of mice and amphibians were also encountered. A summary of the distribution of taxa is given in Table 4, details of the

46

Distribution Summary

ee ae ie pais

Species

Table 4

Hare and

3 é & S =

Sheep-size

Cattle-size

4

aaa]

Cess pit 234 Cess pit 288

Percent

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

wn] elanlo tls a“ va +N

N

= BSc Paes See an aS ae

2

3

Se ES ee SS SS]

7 ne Ee es ae 9

2

0.4

Sa EE Ee ee a isd [94 [72 | D

Cess pit 590

% cattle, % cattle, Pit 770 Percentage overall

% cattle,

4

[2 | went 86s

Undated

BSS SSS brie

ter medieval Post-medieval cera | 172

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

Details of sieved bone and fish distribution

SARA ca a ee | |

Table 5

Flatts

Bullhead

Cyprinid

Cyprinid

Whiting

Conger

Feature

Hand-collected fish Context ys 953 y 957 584 Porcent

elu eo |< - > = N = n n atx s

ne ' q T Herring

Sheep-size

Context 859 97 592 Percent

Medieval medieval

48 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

fish and sieved material is given in Table 5 and a complete listing of the number of fragments of the taxa from each context is given in archive Table A2.

MEDIEVAL AND LATER MEDIEVAL The total number of bones from contexts assigned to medieval and later medieval date is 1,347. Well over half of these were recovered from sieved samples and most of these were of fish, primarily from the later medieval cess pit 590 in Trench 2.

Apart from the fish, most of the bones are of the expected main domestic ungulates; cattle, sheep and pig and also fragments probably of these animals. Overall, sheep bones dominate but cattle occur in more contexts and probably represent a greater weight of meat. Of the 53 cattle bones, most are from prime meat areas with the remainder from the feet or the head, but including only one loose tooth. Six of the head bones are horncores chopped from the skull, presumably for hornworking.

In the unidentified large ungulate material, assumed to be all of cattle, a high proportion of the fragments are of ribs. Many of these had been chopped, as had several of the other bones. Most of the bone can be interpreted as kitchen or plate waste rather than primary slaughter and butchery waste. ‘There is some evidence for secondary products in the case of the horncores, indicating bone from a mixture of sources. Due to the small number of bones and the high degree of fragmentation, almost none of the bones was measurable.

Bones from immature animals include eight of very young calves (including examples from cess pit 234 and cess pit 590). These are mainly meat bones, some chopped, and represent veal remains rather than stock mortalities. Although calf bones can be found from any period, they are more often to be found in post-medieval contexts reflecting an increasing interest in veal. Further examples of bones from young calves were recovered from several post- medieval contexts.

The more numerous bones of sheep (78) also show a biased pattern in the anatomical distribution. In this case many of the bones are of feet. Head fragments are also frequent, although only two loose teeth were recovered, an indication of the good state of preservation. The metapodial bones of the foot, along with the distal tibia and distal humerus, are often well-represented as they fuse early and are more resistant to erosion, dog gnawing and other taphonomic factors. Here, however, there are just two fragments of humerus and four of tibiae and one of these is an unfused proximal end. In contrast

the total of foot bones is 36, 46% of all the sheep bones. The material is well preserved and this bias in favour of the feet appears to be genuine. Bones from the limbs, scapula and pelvis account for the rest of the assemblage. The proportion of rib fragments in the sheep/pig-sized material is very high; several of these are almost certainly of pig but a high proportion matched sheep. No bones were identified as goat, whilst a number were positively identified as sheep. These included horned and hornless animals.

Measurable bones are present including two complete ones from which withers height estimates can be made. These are a metatarsus from cess pit 234 which gives an height of 0.567m and a large radius in cess pit 590 which gives an height of 0.631m. The smaller is typical of the animals reported from medieval Southampton (Bourdillon 1980). The largest (and latest) is substantial for later medieval material, though some from Saxon Southampton exceed this measurement and it is not large by modern standards. An increase in the size of post-medieval sheep has been reported by several researchers (e.g. O’Connor 1995), but the few previous examples of later medieval and early post-medieval date from Salisbury have been of quite small animals, between 0.49 and 0.60m (Coy 1986).

The high number of foot bones may imply waste from processing skins, but there is also some kitchen and plate waste represented in the assemblage. The foot bones are not concentrated in any one feature.

The pig bones are more evenly spread across the body and no anatomical concentrations were noticeable. This is a very small sample of only 37 bones and also the pig carcase, including the head and feet, has more meat value than cattle and sheep.

Other mammal remains are very rare; there is one bone of hare (cess pit 590) and two of a mouse (probably house mouse) in the same feature. No dog bones were found in dated medieval contexts but some bones were gnawed.

Most of the bird bones are of domestic fowl. These occur sporadically throughout the assemblage but not in any concentrations. Some bones can be attributed to hens in lay (Driver 1982). Other birds represented are goose (7), woodcock (1), kite (see pit 192 below) and finch (cess pit 590). The goose bones match greylag/domestic; some are large. At this date they are all likely to be of domestic birds. Woodcock is often one of the most common wild birds in assemblages and makes good eating; this bone is a femur.

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

Pit 192 in Trench 1 contained 14 bones; three of bird, six cattle and a few scraps of other bone. The interest in this small pit lies in the bird bones. These comprise a femur, radius and ulna of a kite. There are two species of kite in Europe; red kite (Milvus milvus) and black kite (Milvus migrans). The red kite is rare but still present in Britain whereas the former distribution of the black kite is not known but has certainly been absent in recent records. The bones here are small and therefore this latter species cannot be ruled out. Kites hunt small mammals but are also carrion feeders and were once common in towns (Reid-Henry and Harrison 1988). They may also have been considered a threat to young poultry, in the same manner as buzzard and crow, and destroyed as vermin.

The eleven bones of a small passerine recovered from bulk soil samples taken from cess pit 590 could be from any one of a number of related songbirds but they most closely match the greenfinch. One of the largest of the common finches, this resident bird could have been a pet or eaten as food.

As indicated above, the 787 fish bones were mainly recovered by sieving of bulk soil samples, particularly from the later medieval cess pit 590 in Trench 2. Only 14 of these bones were recovered by hand-collection, all bones or fragments of large fish. Details of the material from the processed samples, and of the overall fish species distribution is given in Table 5.

Bones of conger, cod and flatfish were identified in the hand-collected material; other fragments were undiagnostic. The four bones of conger found in a medieval soil layer in Trench 2 were vertebrae and head bones of a large fish, probably about 2m long. A fifth bone recovered from the underlying layer was another large vertebra which had been chopped both axially and across, implying that the fish had been split in half and further divided into sections. One of the reasons for splitting such a fish is for preservation. Divided congers were found in the stores of the Mary Rose (Hamilton-Dyer 1995).

Almost all of the fish bones recovered from bulk soil samples were from cess pit 590, but there were also a few bones of herring, eel and unidentified fragments from a soil layer. The species identified from cess pit 590 include ray, eel, herring, cyprinids, ~ stickleback, and bullhead (Table 5). Most of these, well over 400 bones, were of common eel. These remains were mainly vertebrae but also included head bones and cleithra. Herring bones were also common, in this case vertebrae and prootic bones. None of the other head bones was identified, but

49

since most herring bones are very thin and fragile and may not have survived, some of the unidentified fragments may be of herring.

The nine ray remains are of teeth and denticles, including some positively identified as male thornback, the most frequently caught of the ‘skate’ family. The 81 cyprinid bones included 29 inferior pharyngeals from at least 15 fish. These are diagnostic elements and all were identified as dace; the other cyprinid bones are probably of the same species. Dace can achieve a length of 0.25m and a weight of 0.6kg but these specimens are from much smaller fish, no more than a few centimetres long. The stickleback and bullhead are also small fish. All of the fish bones from this feature are small and a number had been crushed. Such damage occurs when fish are eaten by humans (Jones 1984; 1986) and, along with the botanical evidence, confirms the supposition that this pit was used for cess disposal. Very few other bones were found in this feature, but the sieved remains included five of the very small peripheral toes of pig, one of which was partly digested, and many small scraps of mammal bone associated with mineralised cess accretions.

DISCUSSION

This assemblage, though not large, is important as it comes from the centre of the medieval town. Although some faunal analysis has been previously undertaken on material recovered from Salisbury, there was little material from sealed and well-dated medieval material and reports have not yet been published.

The initial impression of the assemblage from the excavations at Ivy Street/Brown Street is of well- preserved material broadly similar to that from other medieval urban deposits in the area.

The mammal bones are mainly of sheep, cattle and pig. Individual contexts vary but overall more sheep bones were recovered than cattle. This may not reflect the true picture of meat usage however, as cattle have a much greater meat weight and their larger bones may have been partly disposed of elsewhere. Pig bones are relatively common; at Gigant Street, although the sample sizes are small, pig bones appear to show a decrease over time (Coy 1986). Pig bones were also low in numbers at Andover (Hamilton-Dyer n.d.), and in the post- medieval material at Romsey (Bourdillon n.d.). Part of the problem of interpreting small assemblages is the bias of context types, and their position in the settlement. Large semi-industrial waste dumps are different in character from the material in rubbish

50 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

and cess pits in the backyards of houses, and even pits on the same property can be variable, for example at Bull Hall, Southampton (Driver n.d.).

Horse and goat are absent from this small assemblage. There are no dog bones either but there is indirect evidence of dog with gnawing on some bones. Wild mammals are represented only by one bone of hare and two of a mouse. Birds include the expected fowl and goose together with a few fragments of other species including woodcock, greenfinch and a kite of uncertain species.

The marine fish, thornback, conger, herring, cod and plaice, almost certainly came through the port of Southampton, as indicated by documentary records (Coy 1996), and these species are frequently found in excavations in the town. Some may well have been preserved in salt, pickled or both (particularly cod and herring) and possibly smoked in the case of conger. The Brokage Books of Southampton record many barrels of herrings and sometimes carts of congers (Stevens and Olding 1985). Herring, conger and codfish could have been brought from some distance, as there are records of boats from Plymouth and the Channel Islands. Other marine fish are likely to have been locally caught in the Solent.

The eel is anadromous but the remains here are all of small eels, neither elvers nor silver eels, and were almost certainly caught locally. The dace, bullhead and stickleback were probably incidental catches but utilised anyway. Nets and wicker traps are the most likely method of collection. If caught together the source is likely to have been a clean stream or shallow river with a stoney bed, rather than a pond. The botanical evidence indicates the accessibility of plants from similar habitats, though not from the same pit as the fish. There is no shortage of suitable streams and rivers round Salisbury.

The remains of these small fish probably represent cess deposits, as quite a number of the vertebrae are characteristically crushed sideways (Jones 1984; 1986). Isaak Walton (1633) describes a recipe for ‘minnow tansies’, a kind of herbed fish omelette, and these small fish here could have been treated in this way or fried as ‘whitebait’. The bones are small and could easily have been eaten and passed through the digestive system. A similar mixture of eel, herring, ray and small freshwater fish was found in a cess pit in Andover (Hamilton-Dyer n.d.), and in the castle garderobe at Southampton (Hamilton- Dyer 1986).

Butchery marks were frequently observed, in part

because the material is very well preserved. Almost all of the butchery was carried out with a heavy bladed cleaver or axe; knife marks were very rare. Precise para-median chopping on sheep and pig vertebrae indicate carcase division of the type described by Bourdillon (1980; n.d.) and others for medieval sites in the area. There is evidence that geese too were divided in half, and jointing appears to be carried out by chopping. Some of the cattle bones had been reduced to small fragments, probably for stews, and all types of ribs were frequently chopped neatly into small sections.

Few metrical data are available but the sizes of the animals are similar to those from other sites both in Salisbury (Coy 1986) and elsewhere (Bourdillon 1980; Hamilton-Dyer n.d.).

Excepting cess pit 590, which has its totals boosted by the small fish bones, the features are not rich in terms of bone density. Cess pit 288 offered 26 bones from several fills, and rubbish pit 932 only seven (the largest number came from the post- medieval pit 585, which contributed 125 from a single fill). Most features were half sectioned but even small parts of some medieval pits at the Lower High Street site in Southampton produced many hundreds of bones (Hamilton-Dyer in prep.). Whether this paucity is standard for Salisbury or is peculiar to the site is not clear. At other Salisbury sites the bone assemblages do not appear to be large, but a meaningful comparison requires extraction of detailed information from the archives.

Apart from cess pit 590, none of the other features which had been used for cess disposal, as indicated by the botanical remains, contained this concentration of fish and other bone which almost certainly originated from the cess itself. Those pits intended for cess disposal do not appear to have been much used for general bone waste; most bone in these features is from the secondary backfill.

The material overall does not appear to originate from high status households. Wild resources, other than fish, are poorly represented and several of the sheep bones are of possible craft waste. Some of the cattle bones are also of low value parts of the carcase, yet they include veal bones and overall the remains give the impression of mainly domestic waste. The marine fish bones show the probable ultimate destination of some of the fish leaving Southampton, while the freshwater species indicate that local resources were also being utilised. Urban refuse disposal in the medieval period is often variable and selective, and only limited aspects of the faunal economy can be suggested from a small assemblage

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994 51

such as this. Nevertheless, it has provided a rare opportunity to examine material from the heart of the town.

POST-MEDIEVAL AND UNDATED Cattle, sheep and pig are the main species identified in the post-medieval assemblage of 469 bones. Bones of other species include all of the dog and cat remains as well as a few bones of fallow deer, rabbit, fowl, goose, cod and haddock. Cattle form a slightly higher proportion of the bones than in the earlier material. Several bones were again from veal calves; this is consistent with a post-medieval date at other urban sites while here they are also frequent in the medieval assemblage (though it should be remembered that both samples are relatively small).

Several of the sheep bones from well 865 are large; a complete metacarpus gives an estimated withers height of 0.694m. This is larger than the size expected for medieval material. The 18th century saw the development of ‘improved’ sheep and the origins of recognisable breeds such as the Southdown and the Leicester Longwool. This bone is relatively slim and could indicate a slow maturing type like the Leicester (O’Connor 1995). However, a radius from pit 585 gives an height of 0.559m which would be considered of average size for medieval material. There is also a very small metacarpus from the same feature. This bone gives an estimated withers height of only 0.482m, comparable with the very smallest values from medieval Southampton (Bourdillon 1980) and smaller than any previously recorded from Salisbury from medieval and post-medieval material (Coy 1986).

Well 865 (Trench 2) also contained sawn cattle ribs; this type of butchery is usually associated with assemblages of a late, even modern, date. Pit 157 (Trench 1) contained a group of sheep metatarsal distal ends, the bones having been broken or chopped off mid-shaft. These may represent tanning waste.

The only other notable bone assemblage is from the pitched-tile hearth (898) in Trench 2. All but one of the 15 fragments are calcined, eight are elements of a sheep forefoot. Given that this is a hearth and that burnt bones were almost absent from the rest of the assemblage, these are likely to be contemporary with use of the hearth.

The small number of bones from undated contexts is mainly composed of sheep and sheep- sized bones. Other species include cattle, pig, fallow deer, rabbit, fowl, goose, conger, ling and red seabream. These general clearance contexts contained bones of small sheep comparable with the

medieval material; a complete metacarpus gives an estimated withers height of 0.551m, and a distal tibia measurement of only 21mm is also very small.

DOCUMENTARY REPORT by John Chandler

INTRODUCTION

This report was undertaken in 1997 as part of the programme of analysis following the excavations of 1994. The aim of the documentary research has been to use maps, printed and manuscript sources in order to comment on the land use history of the site. Research has been conducted in the Wiltshire and Swindon Record Office, Trowbridge (hereafter WRO), the Salisbury Cathedral Archives (hereafter SCA), Salisbury Local Studies Library, and Trowbridge Reference and Local Studies Library. Unfortunately, because of the indisposition of the Salisbury Cathedral Librarian, Suzanne Eward, it has not been possible to examine all relevant sources in SCA.

RESEARCH STRATEGY

The corner site occupied by the building now known as the Queens Arms Inn was the property of the Dean and Chapter of Salisbury Cathedral from the early 15th until the late 19th century. There has been an inn on the site since the 16th century or earlier, known first as the Maiden Head, then the Blue Lion, and by the 17th century under its present name. It has thus provided a landmark for deeds, lists and surveys, and has been used as a reliable point of reference in this research.

‘Brown Street’ was thus named by the late 13th century, but the name ‘Ivy Street’ is not recorded before c. 1600; it was earlier regarded as part of New Street, of which it is the eastern continuation. The chequer in which the site lies is now known as Antelope, but has also been called White Bear and, around 1600, Blue Lion Chequer. The chequer lay in the parish of St Thomas, and in Martins ward. This report discusses the corner tenement and the excavation site together during three periods; medieval, post-medieval and modern. A brief summary is appended.

Sources for the medieval and later history of properties in Salisbury survive in profusion, notably among the records of Salisbury City Council (WRO G23), the city parishes (St Thomas’s is WRO 1900), and the muniments of the Dean and Chapter (SCA). In the WRO are also relevant collections deposited

52 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

by museums, solicitors, and individuals. Many deeds have not, however, been catalogued in sufficient detail to locate easily the properties to which they refer, and so research of this nature is somewhat frustrating, since it is very likely that relevant material has been overlooked.

THE MEDIEVAL PERIOD

The standard size of Salisbury burgage tenements, as laid down in the bishop’s charter of 1225, was 7 x 3 perches (RCHME 1980, xxxii), which is the equivalent of 115.5ft x 49.5ft, or approximately 35 x 15 metres. Antelope Chequer, as is clear from Naish’s 1716 map (RCHME 1980, pl. 16), was laid out with the long axis of tenements running east-west, and a straight boundary line bisecting the chequer north- south. If, therefore, the standard size was adhered to in the area of the site, it and the corner plot would have impinged on the three southernmost tenements fronting Brown Street, and the rear of the two southernmost tenements fronting Catherine Street. However, because of their prestige and commercial possibilities, it is clear that from an early date corner tenements and adjacent land were subdivided and rearranged in various ways. A notable example of this, which has been studied in detail, was Balle’s Place at the corner of Winchester Street and Rollestone Street (Bonney 1964; RCHME 1980, 135-7).

The earliest reference to the corner tenement discovered by this research is the will of Richard de Berewyke, proved in 1361 (WRO G23/1/212, f.22v). The relevant passage may be translated thus: ‘Also I leave and bequeath to William de Berewike to sell after the death of my wife Edith, all that [corner] tenement of mine in which I dwell, which is situated in the foresaid city [New Sarum] in New Street and Brown Street, with its gardens, curtilage, and all things pertaining, together with one plot of garden ground in St Martin’s Street annexed to the said tenement, which I hold by concession of Agnes my former wife...’. It is not clear from the wording whether the tenement as well as the separate garden came to Richard via his first wife. The St Martin’s Street garden cannot have been literally adjoining his tenement. A copy of part of the same will, transcribed on the recto of the same folio, adds the word ‘angular’ (corner) as indicated above.

In fact the property was sold less than a year later, in 1362, by Richard’s executor, Stephen de Botelesham, to Thomas de Hungerford (SCA press 1, boxes 13-15, Sal 3/63). The property is described as a corner tenement in New Street and Brown Street, with garden and curtilage; and a site in St

Martin’s Street. Thomas was a citizen and merchant of Salisbury, and served as mayor, a kinsman of (or even the same as) Thomas Hungerford of Heytesbury, founder of the Hungerford dynasty (see Kirby 1994, xvi). Thomas either let the property or sold it, by the 1390s, to Alice Rusteshale (see below), who appears to have been related to its next owner, John Caundel, a ‘mattins clerk’ in the cathedral.

In his will, proved in 1400 (WRO G23/1/213, f29), John Caundel bequeathed to his servant Margaret Deneman, for the term of her life, his tenement in which he lived in New Street. After her death, it appears to have been Caundel’s intention (although the wording is not altogether clear) that the house should be sold and the proceeds used to endow a chantry or obit for prayers for his soul and those of John Dewel and Alice Rusteshale, who was once Dewel’s wife (this suggesting that Caundel himself was related to Rusteshale). In the event the property was granted to the Dean and Chapter by Margaret Deneman’s executor after her death in 1410 (WRO G23/1/213, f.92).

This deed is important, because it gives the earliest firm indication discovered during this research of the ownership of the plots on either side of the corner tenement. The deed’s description of the premises may be translated thus: ‘All that corner tenement with everything pertaining to it in which the foresaid testator, John Caundel, lived, and before him Alice Rusteshale, situated in New Sarum in New Street and Brown Street, between the tenements of William Warmwell on both sides’. William Warmwell was a prominent citizen who in c. 1399 was living in Market ward, and who was one of the highest- assessed contributors to a tax levied in that year (Chandler 1983, 263, no. 39). He died in 1412, and his will (WRO G23/1/214, f.88) refers to a number of city properties, but none that can be identified with the site. He may have sold it before his death, although a search of enrolled deeds in the Domesday Books (WRO G23/1/213-14) did not discover any evidence for this. Alternatively he may have retained it, but regarded it as too insignificant to be separately itemised in his will. In any event, it is important that the corner tenement was already by this date surrounded by land in single ownership (as it continued to be for several centuries), suggesting that the original arrangement of tenements had broken down by c. 1400 and been replaced by something more complicated.

There is a little evidence (Helen Bonney, pers comm.) that the c. 1399 ward lists reflect the topographical sequence of householders, and if this

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

is the case we may surmise that the names adjacent to John Caundel’s in the list were his neighbours. The sequence, with their tax assessments, was as follows (WRO G23/1/236, transcribed in Chandler 1983, 268, nos. 653-9): John Route (6d); Cristina Handle (8d); William Busshel (4d); John Caundel clerk (40d); the wife of William Pobeman (4d); Thomas Denham (4d); Richard Pympol (4d). If the list is sequential we may deduce that the value of the building on the corner tenement, or of its owner, was a great deal more than any of the neighbours.

Cristina Handle appears to have been involved in the cloth trade, since an aulnage return of 1396/7 reveals that the alnager sealed 40.5 of her cloths, making her the 52nd highest (out of 292) Salisbury cloth-producer in that year (derived from list in Chandler 1983, 260-2). It should be noted that the east range of the present Queens Arms Inn is regarded as being of 14th century origin (RCHME 1980, 111, no. 225), and so dates from the time of Caundel and his predecessors.

The next positive evidence of the corner site occurs in a bishop’s rental of the whole city, the Liber Niger, dating from 1455 (Nevill 1911). Listed among the chapter property is: a tenement formerly of Alice Rusteshale in New Street and Brown Street, assessed at 5.5d (ibid., 69). All the other relevant properties are doubtless listed, but it is impossible to locate them precisely. An indication that, despite Salisbury’s wealth and rising population during the early 15th century, houses in this area of the city were being demolished to make way for tentering racks and gardens, comes in a deed of 1433 (WRO 164/1/11) which relates to, ‘my toft with racks adjoining the toft, which lies between tofts with racks and gardens in Gigant Street and Brown Street... and a toft with adjoining garden in New Street near Barnwell Cross’.

No evidence for the site during the later 15th or early 16th centuries has been found.

THE POST-MEDIEVAL PERIOD Throughout this period the corner tenement belonged to the Dean and Chapter, who leased it to individuals. Sometimes the lessee sublet it to a tenant. The earliest book of chapter leases begins in 1533 (WRO CC Bishoprick 460) and extends to 1561, but includes no relevant lease. The next book (WRO CC Chapter 207), spanning 1563-1608, has a lease of 1570 (p. 104) to Robert Stephens of New Sarum, ‘weafer’, of: ‘All that theyre corner tenemente little courte and stable roomes commonly called the Mayden hedd with all and singuler their appurtenances nowe in the tenure and

53

occupation of George Vynynge bocher sett lyinge and beinge in the ... streate there called New Streate and Browne strete between the lands sometyme belonginge to Thomas Chaffyn the eldre late deceassid nowe in the severall tenures and occupations of the said George Vinynge and Nicholas Dycar currier on the weste and northe parties, and the Queenes highe wayes on the east and south parties.’

Thereafter the renewals of the lease may be traced through the chapter lease books (SCA 41-59) up to the beginning of the 19th century. The dates of the leases, with lessees’ names and occupations, and document references, are as follows: 1637, George Mustin, innholder (SCA 42, f.36); 1699, Thomas Jatt, wheelwright, formerly Peter Dove of Pyt House (SCA 47, 376); 1715, Thomas and Mary Martin, of the Close (SCA 49, f.151); 1729 and 1743, as 1715 (SCA 50, f.113; SCA 51, f.107); 1757, John Wilkes, innholder (SCA 52, f.126; alsoWRO 1075/001/90); 1771, Elizabeth Wilkes (SCA 53, f.95v); John Mills, cordwainer (SCA 56, f.7v); 1806, Jasper Fawconer, of Charlton in Downton (SCA 57, f.97); ?1821, James Sutton and Thomas Budden; 1835, John and Mary Grimes (SCA 59, f.92). By 1637 (and indeed somewhat earlier, see below) the name Maiden Head had been replaced by Blue Lion, and by the 18th century the name changed again, to the Queens Arms. From 1699 the descriptions include the phrase ‘lately rebuilt’ in parentheses.

The descriptions of the premises and the abuttals included in these leases are of limited value, because nearly all details are merely copied from the previous lease. The information to be gleaned from the abuttals is discussed below. In addition, however, there are two more detailed descriptions of the premises, dating from 1644 and 1649. The will of Elizabeth Mustion (WRO Cons Sarum wills 1644) includes a room-by-room inventory, and the rooms are listed as: hall, parlour, little buttery within the parlour, room next to parlour, room next to wash house, wash house, great chamber, chamber over parlour and drawing chamber within it, chamber next the street called road chamber, two chambers over the hall and taphouse, taphouse, backsides. The 1649 Parliamentary survey of church property in Salisbury (SCA 11, f.5) describes the premises as: a hall, a parlour, two tapp houses, a little buttery, three drinking rooms, a little parlour, five chamberes, two other over roomes, two stables containing two bayes of building with a little courtyard of 20 feet square.

More important for present purposes than the information about the Queens Arms is the evidence which these leases and other documents give about

54 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

the adjoining properties to north and west. The 1570 lease quoted above refers to Thomas Chaffyn the elder, deceased, as the former owner of the surrounding land, and George Vinynge and Nicholas Dycar as the current tenants. Vinynge, a butcher, also occupied the corner tenement, and Dycar was a currier. The owners and tenants given as abuttals at later dates (so far as they are not merely repetitions of names from earlier documents) are as follows: 1637, Thomas Chafin the elder, gent (owner), George Mustin and Margarett Lawrence, widow (occupiers); 1649, the inn called the Antelopp to the north, the land of Mistress C. Chafyn, widow, or her assigns, to the west; 1715, Thomas Chafin Marks (owner), Richard Petty, wheelwright, and Thomas Durdall (present or former occupiers).

Thus the owners of the Ivy Street portion of the site, west of the corner tenement, from before 1570 until after 1715, were members of the Chafyn family. The 1623 heralds’ visitation of Wiltshire includes a pedigree of this family (Squibb 1954, 36-7) which suggests that the Salisbury branch was one of several claiming as a common ancestor Thomas Chafin of Warminster, who must have flourished in the early 15th century. They were well-established in Salisbury by 1545, when Thomas Chaffyn senior and Thomas Chaffyn were two of the most highly rated citizens for a tax (Ramsay 1954, 38-9). Both died c. 1558/9 (Squibb op. cit.), but the younger’s son and heir, Thomas Chafin, appears to have lived until 1619. At his death he was a major landlord in Salisbury, owning no fewer than 46 properties, as well as 22 in Warminster, and others elsewhere (Fry and Fry 1893, 196).

He was succeeded by his son, also Thomas, who was born c. 1582, and died in 1646 (Matthews and Matthews 1906). Indexes of wills proved in the Prerogative Court of Canterbury have been checked to 1700, and reveal that Charles Chafyn, gentleman, Dorothy Chafyn, widow, and William Chafin, gentleman, all of Salisbury Close, died in 1657, 1659, and 1665 respectively (Blagg 1936; Morrison 1935); and that Edmund Markes, apothecary of Salisbury, died in 1675 (Ainsworth 1942). No relevant wills of the Chafyn or Marks families have been found via WRO indexes.

The property which they owned west of the Queens Arms Inn was presumably the house of c.1500 now known as No. 7 Ivy Street (RCHME 1980, 111, no. 226). Its lessee at the time of her death in 1644 was Elizabeth Mustion of the Queens Arms, since her probate inventory (WRO Cons. Sarum wills, 1644) includes, ‘one other chattell lease of a

tenement and garden with the appurtenances adjoining to the foresaid tenement [Queens Arms] held from Thomas Chafyn gent for certain years yet to come’. It had perhaps been leased by the occupant of the Queens Arms for many years, since George Vyninge of the Queens Arms appears to have held the lease in 1570 (see above). Stephen Grist, who inherited the Queens Arms from Elizabeth Mustion, also occupied this tenement immediately after her death, according to a land tax assessment (WRO G23/1/174).

The Brown Street property abutting the Queens Arms on the north also belonged to the Chafyn family in 1570 and 1637 (see above), whose tenant in 1570 was presumably the currier, Nicholas Dycar. In 1637 the tenant was Margaret Lawrence, widow, who was the licensee of the important inn, the Antelope, which lay to the north-west in the centre of the chequer (Gordon n.d., 33). In 1649 the Antelope is named as the northern abuttal, and this area seems to have continued to have formed part of the Antelope well into the 18th century. A lease of 1768 (WRO 952/2) includes the Antelope Inn together with, ‘all that piece or parcel of ground on the east side of the inn and the backhouse adjoining lying in Brown Street, formerly converted into two tenements but now used as stables, and also three other stables in Brown Street part of and adjoining the Antelope Inn’. By the mid 18th century the Antelope was becoming an important coaching and post-chaise inn (Gordon, n.d., 33-6, 451), which would explain the conversion of tenements to stables. Former tenants of the Antelope, according to the 1768 lease, were William Little, Thomas Shuter, George Petty, and Widow Stephens.

With so many names of occupiers, both of the Queens Arms and adjacent properties, it should be possible to tie the evidence from leases with that from topographically arranged tax lists. There survives in the St Thomas parish records a notable series of Easter offerings books, listing from c. 1573 to c. 1602 adult parishioners including wives and servants, and arranged by street or chequer. Some are too fragile to be produced, but books from 1574, 1584, 1600, c. 1600, 1602, and two undated 17th century books have been examined (WRO 1900/44; 45; 51; 52; 54- 6). Unfortunately only two, of c. 1600 and undated (WRO 1900/52; 55), appear to have any names which correspond with those in the leases; both list George Mustian at the head of the section for Blue Lion chequer, and his neighbours in each case are Robert Hatt, Thomas Soper, and Bridgett Gibbes. However,

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

it is not clear whether the sequence of names is running north along Brown Street or west along Ivy Street.

More promising is a list of taxpayers for a royal aid in 1667 (Nevill 1910, 427). This includes a sequence: ‘Land of Mr Chaffin, tenant Widow Durdall; land of Mr Chaffin, tenants John Taylor and John Coombes; land of Mr Doves [i.e. the Queens Arms], tenant John Gumbleton; land of Mr Chaffin, tenants widow Gower, widow Williams, John Gumbleton a stable’.

A generation later, beginning in 1703, there is a series of land tax assessments for Martins ward, and these have been checked to 1717 (WRO G23/1/182). Through this period Thomas Jatt (or Iatt) was lessee of the Queens Arms until his death in 1709 (WRO Sub-Dean Wills, 1709), then his widow to 1716, and thereafter Mr Martin. On one side of him Thomas Durdall was the lessee throughout, and appears to have lived in the property himself with two sub- tenants. On the other side the land belonged to Mr Marks until 1715, whose tenant was Richard Petty; in 1716 Thomas Wigmore acquired the land, but Petty remained lessee. It is not clear which was the northern and which the western abuttal, but the reference to a George Petty in the 1768 lease (WRO 952/2: see above) as a former occupant of the Antelope militates in favour of Durdall occupying No. 7 Ivy Street, and Richard Petty, wheelwright, the Brown Street property.

THE MODERN PERIOD

As might be expected, sources of information about the site become more prolific and varied from the 18th century onwards, and include for the first time large-scale plans. For the recent period sources have been examined selectively.

Throughout the period, and to the present day, the Queens Arms appears to have existed as an inn. The list of lessees taken from Dean and Chapter leases up to 1835 is given above. The lessees were not necessarily the occupants; a list of Queens Arms innkeepers compiled from directories and other sources (Gordon n.d., 7) offers the following: 1743, George Rattew; 1809, James Ainsworth; 1822, Aaron Vousden; 1836 and 1842, Ann Vousden; 1851, George Moore; 1867, John Hibberd. It is also

- possible that the Queens Arms and the Antelope were being run together, since a 1773 ratebook (WRO 1900/200) brackets together John Wilkes and Mrs Best with a single assessment. It is known that Martha Best was landlady of the Antelope until her death in 1798 (Gordon n.d., 33), and a 1773

5)

newspaper advertisement includes the instruction, “Enquire of Mr Wilks at the Queens Arms Inn’ (ibid., 6). In 1779 Martha Best paid a similar assessment alone (WRO 1900/200), suggesting that she was then responsible for both establishments.

During the 19th century both the Queens Arms and No. 7 Ivy Street were held, from 1849, by a 40- year lease granted to Ann Vousden (WRO 1075/001/ 90), the landlady since at least 1836. But during the 1870s the premises were occupied (and the lease probably purchased) by Messrs. Weeks and Son, the Weyhill Brewery. In 1877 the Ecclesiastical Commissioners, as successors to the Dean and Chapter, finally sold the property, consisting of the Queens Arms itself together with No. 7 Ivy Street and a cottage fronting Brown Street which was attached to the northern range of the inn. It was described (WRO 1075/001/90) as a public house, brewhouse and tenement lately occupied by Messrs. Weeks and Son. The abuttals were a stable yard to the west (Ivy Street) and stables and stores to the north (Brown Street), all occupied by Messrs. Weeks and Son.

Sale catalogues of 1879 (WRO 374/130/110) and 1888 (WRO 1075/001/90) included the same premises, and the latter described No. 7 Ivy Street as a tenement, late the Weyhill Brewery Office. In 1891, according to the census returns, Harriet Norton, a 78 year old widow, lived alone as innkeeper at the Queens Arms, and No. 7 Ivy Street was occupied by a carpenter, John Foot, and his family. The cottage forming part of the complex north of the inn was occupied by a dressmaker, Martha Maton.

The sequence of occupation has been traced through Salisbury directories up to 1962. From before 1925 until after 1947 the licensee was A. Mitchell and then Mrs F. Mitchell (presumably his widow). E.L. Scammell, and from 1935 Mrs Scammell occupied No. 7 Ivy Street until after 1950. By 1953 No. 7 Ivy Street was part of the Queens Arms, but from 1959 it was used as stores for the neighbouring Farway Garage (see below).

Turning now to the land west and north of the corner tenement, the best starting point is the sequence of maps, which begins in 1716. Naish’s map of that year (Figure 8) shows the Brown Street frontage entirely built up, with (north of the Queens Arms) three rectangular backlands running back to a straight rear boundary which is not, however, parallel with the street. On the Ivy Street frontage there is a considerable gap in the building line, around the mid-point, leading to an open area

56 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

) Vee Tif Loe Queen's Arms‘A \

/ af

/ Queen's Arms

Fig. 8. Cartographic regression analysis

behind the Queens Arms. An enclosed courtyard within the inn complex is also shown. This map was revised in 1751, but the only difference in this area appears to be the addition of a watercourse along the centre of Ivy Street (perhaps accidentally omitted from the first edition).

The earliest really detailed plan of the area is found on a series of maps of Salisbury drawn in 1854 in connection with the Board of Health’s enquiry

into the sanitary condition of the city (WRO G23/ 701/1PC: see Figure 8). This distinguishes by colour between dwellings, other buildings and pavements, and open areas. By comparing this with the Ordnance Survey plan at 1:500 scale, surveyed in 1879 and published in 1880 (sheet Wiltshire 66.15.4: see Figure 8), it is possible to distinguish between other buildings and pavements on the Board of Health plan. Other useful 19th century cartographic sources are plans on deeds (in WRO 1075/001/90), and two undated (but c. 1870) plans of the Queens Arms (WRO CC Maps 45/11 and 45/18).

Trench 1 would appear to have impinged on the rear of three cottages, subsequently numbered Nos. 1, 3, and 5 Ivy Street, together with their rear courtyards and outbuildings. The course of a prominent north-south boundary wall, presumably part of the centreline of Antelope Chequer as originally planned, runs the length of the trench, and seems to have been picked up at its southern end during the excavation. To the west of this line was open land in 1716/51, but had been largely covered by non-dwelling buildings by 1854.

Little has been discovered about these cottages before the 19th century. Earlier tax lists and ratebooks give the impression that lowly-rated domestic properties extended both west and north of the Queens Arms from at least the 16th century onwards. The 1716/1751 maps suggest that there was then no passageway west of No. 7 Ivy Street, but continuous building along the frontage. If so, it may perhaps be assumed that the block of cottages which existed until the 1960s (Nos. 1-5 Ivy Street), was part of the Chafin property traced above. The undated Church Commissioners plans of c. 1870 (WRO 45/11 and 45/18) both describe the land west of No. 7 Ivy Street as Mr Weeks’s stables, but by 1880 a plan attached to a deed (WRO 1075/001/90) attributes this area, and the stables to the north, to Mr Thomas Crutcher. He in 1891 (census return) was the occupant of No. 5 Ivy Street, and described himself as a horse dealer. The cottages appear to have continued as private dwellings until the 1960s.

Trench 2, which explored tenements north of the Queens Arms fronting Brown Street, appears to have impinged on the stables and courtyard which belonged to the Queens Arms and/or the Antelope. The lease of 1768 (WRO 952/2) described above seems to refer to this area as the Antelope stables, but a newspaper advertisement in 1772 (Gordon n.d., 6) mentions houses in Brown Street between the Queens Arms stables and Mr Brooks’ house. If, as suggested, the two inns were being operated

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

in tandem at this period, the stables presumably served for both. The 1854 map implies that the buildings in this area were still stabling, or at least in non-domestic use, and this is confirmed by the abuttal recited on an 1877 deed (WRO 1075/001/ 90). In 1880 they were used by the horse dealer, Thomas Crutcher.

During the 20th century these premises were used variously as a mineral water factory, smith’s shop and builder’s workshop (1925), a tinsmith and motor engineer (1927), a cabinet maker (1931), and a motor cycle dealer (1935). By 1959 part of the site was occupied by Farway Garage, motor car agents and dealers, who disappeared c. 1965, by when Colletts Garage had established itself along the Brown Street frontage to the north.

SUMMARY

The excavation site lies within one of Salisbury’s 13th century ‘chequers’ surrounding its south-eastern corner. It may be assumed to have impinged on parts of five original tenements, and the north-south rear boundary dividing them. The corner tenement occupied by an extant medieval building, known since the 17th century as the Queen Arms Inn, can be traced intermittently in documents since 1361, in particular through the leasebooks of the Dean and Chapter of Salisbury, to whom it belonged from 1410 until 1877. Although in the 14th century it was a high-status private house, it appears to have become by the 16th century an inn, which it remains.

By 1570 the adjoining cottage west of the inn fronting Ivy Street (now No. 7 Ivy Street) was in the same occupancy as the inn, and later it seems to have been acquired by the Dean and Chapter. Land and buildings further west and to the north, with which the excavation was concerned, never belonged to the corner tenement, but were in the hands of wealthy Salisbury landowners, including William Warmwell in c. 1400, and the Chafyn family, c. 1550 -c.1715.

Although names of occupants in the general area abound in medieval and later documents, it is rarely possible to identify their premises. No documentary evidence has been found of particular industrial buildings or high-status dwellings on the site; rather, both street frontages appear to have been lined by “cottages. The western half of the site continued to be occupied by modest dwellings and their backyards until the 20th century. An important inn, the Antelope, lay to the north-west of the site, and by the 18th century its stables extended to Brown Street. The northern half of the site became stables, a

Dil.

courtyard, and associated outbuildings, associated with both the Antelope and the Queens Arms Inn, and taken over by a horse dealer in the late 19th century. In the 20th century this area served various small industrial functions, including prior to its clearance a motor garage.

DISCUSSION by Mick Rawlings

The excavations at Ivy Street/Brown Street have produced a substantial amount of evidence concerning the nature of settlement in this part of Salisbury, particularly during the period immediately after the establishment of the new city in the first half of the 13th century. The importance of the work lies not only in the information recovered concerning the arrangement of activities within the tenements or burgages and in the details of construction of the buildings along frontages, but also in the wealth of environmental data retrieved from waterlogged deposits in the lower parts of some pits.

In the decade preceding these excavations, a number of pieces of archaeological fieldwork had been undertaken within the limits of the medieval city (Figure 9). These were mainly excavations carried out by the Trust for Wessex Archaeology in association with other parties and were aimed at fulfilling the aims of a Project Research Design which had itself resulted from earlier concerns over the potential loss of archaeological deposits as a result of proposed development (cf. Borthwick and Chandler 1983). All of this work remains unpublished, except in the form of a number of very preliminary summary reports (e.g. WAM 1988, 178). However, a document presenting the outline results of the programme was produced (Hawkes n.d.) and the work at Ivy Street/Brown Street can be assessed against the information presented in this document.

The pattern of excavated medieval structures in the eastern chequers was previously shown to be mainly single-roomed buildings with rear yards containing a well, and subsequent developments comprising extensions at the back. The dimensions of the rooms were in line with those recorded here in Trench 2, but in this instance the building was clearly aligned along the frontage and had at least three rooms plus an extension at the rear.

The walls of this building, however, did conform to the type previously recorded, representing flint-

58 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

defences

Known line of city

Blue Boar

Market Place

Sd

BROWN STREET Rolfe’s /IVY STREET | EXCAVATIONS

ei Approximate area

; WA excavations 1984-1990 ® Location

Watching brief

Fig. 9. Archaeological Fieldwork in Salisbury, 1984-1990

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994

and-mortar dwarf walls which would have supported timber-framed buildings. The external walls were free-standing and of variable width, with greensand blocks often (but not exclusively) incorporated at corners and wall junctions. The internal partition walls were represented by narrow beam slots, whereas the previous investigations had found that this build-type was exceptional and that internal walls were more usually based on dwarf walls which were similar to, but narrower than, the external walls.

Floors were previously recorded as being mostly composed of layers of compacted chalk up to 70mm thick separated by levelling layers of sand, soil or occasionally of clay. At Ivy Street/Brown Street this pattern was repeated, although the levelling material was most often clay or gravel, or a mixture of both. There was no indication that the floors had been tiled, although the possibility remains that the actual floor surface was made up of timber planking suspended above the compacted chalk. However, the presence within the floors of numerous episodes of patching, repair and replacement makes this unlikely.

The backlands at Ivy Street/Brown Street showed limited evidence of pit-digging in the medieval period, with the exception of cess pits. There are several potential explanations for the almost complete absence of refuse pits, including the possibility of alternative refuse disposal systems such as collection for removal to extra-mural areas or direct disposal into the street water-courses. On the other hand, it may just have been the relatively high water table which provided the deterrent. This absence of refuse pits of medieval date was noted during the previous campaign, although the investigation of a site on the projected line of the northern city rampart revealed the presence of more than 20 pits, thus lending further credence to the suggestion that extra-mural disposal was the more favoured option (Hawkes n.d., 11).

The cess pits appear to have started out in the medieval period as rounded, unlined excavations into the basal river gravels, but by the 15th century these were replaced by well-built square or rectangular pits lined with ashlar chalk blocks. It is the pits of this type investigated at Ivy Street/Brown Street which have provided the great wealth of environmental data described in this report. This represents the first time in Salisbury that such material has been recovered and examined in detail, and it has thrown up a number of points and questions around which future research aims and objectives can be set.

Although the evidence recovered does not suggest that the occupants of the Ivy Street/Brown

59

Street site were of especially high social status, access to a rich and varied diet can be assumed. This may not have been a consistent phenomenon, but those cess pits which were examined revealed that the available foods included both local orchard and hedge fruits (apples, plums, sloes, strawberries, raspberries, blackberries) along with imported ones such as figs and possibly grapes. A number of types of birds were available for consumption (chickens, geese, woodcock and even finches) and these were bred or caught locally. Sheep, cattle and pigs were all found as kitchen or plate waste, and it is likely that the shellfish (oysters, scallops, whelks and mussels) were also brought to the site as comestibles.

Perhaps the most informative faunal remains are the fish bones. This assemblage shows the importance of fish in the diet, both of locally caught freshwater species (eel, dace, stickleback, bullhead) and of other types imported from the coast, probably Southampton. This latter group included conger, cod, herring, flatfish and rays, mostly thornback. In the post-medieval period this list was expanded to include ling and red sea-bream. The cod and herring could have been salted and/or pickled, and the conger was split for the purpose of transport and possibly for salting. The freshwater fish were almost certainly from the local chalk streams, and the predominance of eel within the assemblage indicates that this was available in some considerable quantity.

The stability of tenement layout was emphasised in the sites examined in the 1984-90 campaign, with both building plans and property boundaries being maintained right through to the 19th century, even during the wholesale replacement of timber-framed buildings with brickwork structures. This element of continuity can be seen clearly in Trench 2 at Ivy Street/Brown Street, where the tenement wall formed the boundary at the frontage until the 19th century and probably in the backlands until the same period. The lack of any structure on the frontage immediately to the north of the tenement wall suggests the presence here of an access through to the backlands. This is a common occurrence in Salisbury, where the grid street layout precludes the establishment and use of back lanes (RCHM 1980, xlii).

The documentary survey which has been carried out as part of this report serves to emphasise the richness of the available material and the rewards of being able to link archaeological excavation and historical research. This union deserves to be a crucial aspect of any work undertaken in the medieval core of Salisbury. In the case of the excavations at Ivy Street/Brown Street, the availability of such a rich

60 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

source of documentary material was offset against the necessary expedience with which the excavation was undertaken, i.e. the loss of information caused by the machine-excavation of the upper levels of the site was to a great extent compensated by that which was recovered by the documentary survey.

The archive

The archive has been deposited at Salisbury and South Wiltshire Museum, Salisbury, under the project code W7924.

Acknowledgements

The project, including the publication of this report, was commissioned and funded by Salisbury District Council, initially through the Relph Ross Partnership. Wessex Archaeology would like to thank the officers of the District Council and the staff of the Relph Ross Partnership for their assistance throughout the course of the project. The collaborative assistance of the staff of the County Archaeological Service, in particular Helena Cave- Penney, is also appreciated.

The excavation was managed by Kit Watson, and was directed in the field by Mick Rawlings with the assistance of Dave Murdie and Rod Brook. Much of the initial post-excavation work was undertaken by Nicholas A. Wells, with the finds analyses co- ordinated by Lorraine Mepham and the environmental analyses by Michael J. Allen. The illustrations were prepared by Elizabeth James.

Bibliography

AINSWORTH, J., 1942, Index of wills proved in the Prerogative Court of Canterbury, vol. 9, 1671-5, British Record Society, Index Library, vol. 67

ALGAR, D., LIGHT, A., and TREHANE, P., 1979, The Verwood and District Potteries: a Dorset Industry. Ringwood

ATKINSON, D., 1965, ‘Clay Tobacco Pipes and Pipemakers of Marlborough’, WAM 60, 85-95

ATKINSON, D., 1970, ‘Clay tobacco pipes and pipemakers of Salisbury, Wiltshire’, WAM 65, 177-89

BEHRE, K-E., 1978, ‘Formenkreise von Prunus domestica L. von der Wikingerzeit bis in die frihe Neuzeit nach Fruchtsteinen aus Haithabu und Alt-Schleswig’, Berichte der Deutsche Botanikscher Gesellschaft 91, 161-79

BLAGG,T.M., 1936, Index of wills proved in the Prerogative Court of Canterbury, vol. 8, 1657-60. British Record Society, Index Library, vol. 61

BONNEY, H.M., 1964, ‘“Balle’s Place” Salisbury: a 14" century merchant’s house’, WAM 59, 155-67

BORTHWICK, A., and CHANDLER, J., 1983, Our Chequered Past: An Archaeological Appraisal of Salisbury. Wiltshire County Council

BOURDILLON, J., 1980, ‘Town life and animal husbandry in the Southampton area as suggested by the excavated bones’, Proceedings of the Hampshire Field Club and Archaeology Society 36, 181-91

BOURDILLON J., n.d. Animal Bone Assemblages from Medieval and Post-medieval Romsey. unpublished report for Test Valley Archaeological Trust

CHANDLER, J., 1983, Endless Street: a history of Salisbury and its people. Salisbury

COLEMAN-SMITH, R., and PEARSON, T., 1988, Excavations in the Donyatt Potteries. Chichester

COPLAND-GRIFFITHS, P., 1989, ‘Excavation near a 17" century kiln at Horton, Dorset’, Proceedings of the Dorset Natural History and Archaeology Society 111, 71-85

COY, J.P., 1986, Animal Bones from Archaeological Evaluations at W139, 39 Brown Street, and W129, Gigant Street car park, Salisbury. Ancient Monuments Laboratory Report

COY, J.P., 1996, ‘Medieval records versus excavation results - examples from Southern England’, Archaeofauna 5, 55-63

CRUMMY, N., 1988, The Post-Roman Small Finds from Excavations in Colchester 1971-85. Colchester Archaeology Report 5

DICKSON, C., 1987, ‘The identification of cereals from ancient bran fragments’, Circaea 4, 2, 95-102

DRIESCH, A. von den, 1976, ‘A guide to the measurement of animal bones from archaeological sites’, Peabody Museum Bulletin 1, Harvard

DRIESCH, A.von den, and BOESSNECK, J., 1974, Kritische Anmerkungen zur Widerristhohenberechnung aus Langenmassen vor- und frtthgeschichtlicher Tierknochen. Saugetierkundliche Mitteilungen 22, Munich, 325-48

DRIVER, J.C., 1982, ‘Medullary bone as an indicator of sex in bird remains from archaeological sites’, in B. Wilson, C. Grigson and S. Payne (eds.), Ageing and Sexing Animal Bones from Archaeological Sites. 251- 4. Oxford: British Archaeological Report 109 (British Series)

DRIVER, J.C., n.d. Animal bones from Bull Hall, Westgate. unpublished report for Southampton City Museum Service

FRY, E.A., and FRY, G.S. (eds.), 1893, Abstracts of Wiltshire Inquisitiones post mortem...1n the reign of Charles the First. British Record Society, Index Library, vol. 23

GOODALL, L., 1988, ‘Iron objects’ in James and Robinson, 1988, 208-23

GORDON, R.G., n.d. The inns of Salisbury. [manuscript ; copies in Salisbury and Trowbridge libraries]

GREEN, F-J., 1984, ‘The archaeological and documentary evidence for plants from the medieval period in England’

EXCAVATIONS AT IVY STREET AND BROWN STREET, SALISBURY, 1994 61

in W. van Zeist and W.A. Casparie (eds.), Plants and Ancient Man. Rotterdam, 99-114

GREIG, J., 1981, “The Investigation of a Medieval Barrel- latrine from Worcester’, Journal of Archaeological Science 8, 265-82

GREIG, J., 1988, ‘Plant Remains’ in S. Ward, Excavations at Chester, 12 Watergate Street 1985; Roman headquarters building to medieval row, 59-69. Chester, Chester City Council

GREIG, J., 1990, ‘Plant and parasite remains from 16th century pits 22, 41, 56’ in I. Burrow (ed.), ‘Excavations at 5-8 Fore Street, Taunton, 1979’, Proceedings of the Somerset Archaeology and Natural History Society 132, 139-49

HAMILTON-DYER, S., 1986, ‘A note on the sieved material from the garderobe layers A296-8’ in J. Oxley, Excavations at Southampton Castle. 37-8. Southampton City Museum

HAMILTON-DYER, S., 1995, ‘Fish in Tudor naval diet - with reference to the Mary Rose’, Archaeofauna 4, 27-32

HAMILTON-DYER, S., n.d. Animal bone from excavations in Andover at Chantry Street and Winchester Street. unpublished report for Test Valley Archaeological Trust

HARE, J.N., 1991, “The growth of the roof tile industry in later medieval Wessex’, Medieval Archaeology 35, 86- 103

HARVEY, J.,; 1981, Medieval Gardens, London

HAWKES, J., n.d. Excavations in Salisbury 1984 — 1990. Wessex Archaeology Unpublished Report

HINTON, P., 1993, ‘Plant Remains’ in D. Rudling, L. Barber, and D. Martin, ‘Excavations at the Phoenix Brewery site, Hastings 1968’, Sussex Archaeological Collections 131, 101-2

HUME, I.N., 1961, ‘The glass wine bottle in Colonial Virginia’, Journal of Glass Studies 3, 91-117

JAMES, T.B. and ANDERSON, A.M., 1988, Clarendon Palace. Report of the Research Committee of the Society of Antiquaries of London, 45

JESSEN, K., 1955, ‘Key to Subfossil Potamogeton’, Botanisk Tidsskrift, 52, 1-7

JOHNSON, C.P., 1869, The Useful Plants of Great Britain: a treatise upon the principal native vegetables capable of application as food, medicine or in the arts and manufactures. London

JONES, A.K.G., 1984, ‘Some effects of the mammalian digestive system on fish bones’ in N. Desse-Berset (ed.), 2nd fish osteoarchaeology meeting. 61-5. Valbonne Cedex: Centre National de la Rechurche Scientifique, Notes et Monographies Techniques 16

JONES, A.K.G., 1986, ‘Fish bone survival in the digestive systems of the pig, dog and man: some experiments’ in D.C. Brinkhuizen. and A.T. Clason, Fish and Archaeology. 53-61. Oxford: British Archaeological Report 294 (International Series)

Kelly’s Directory of Salisbury and District, various issues, 1925-62

KENWARD, H.K., and HALL, A.R., 1995, Biological Evidence from 16-22 Coppergate. The Archaeology of

York, vol. 24: The Past Environment of York. York: Council for British Archaeology

KIRBY, J.L., (ed.), 1994, The Hungerford Cartulary. Trowbridge: Wiltshire Record Society, vol. 49

MACGREGOR, A., 1985, Bone, Antler, Ivory and Horn: The technology of Skeletal Materials since the Roman Period. London

MATTHEWS, J., and MATTHEWS, G.F., 1906, Year book of probates from 1630. London

MEPHAM, L.N., 1993, ‘The Pottery’ in D. Coe and R. Newman, ‘Archaeological investigations at the shrunken village of Knook’, WAM 86, 81-5

MEPHAM , L.N., forthcoming ‘The Pottery’ in P. Williams and R. Newman, Excavations at Grove Farm, Market Lavington, Wiltshire, 1986-1990, The Development of a Roman, Saxon and Medieval Settlement. Wessex Archaeology Report, Salisbury

MEPHAM, L.N., and UNDERWOOD, C., n.d., ‘The Pottery’ in J. Hawkes, Excavations in Salisbury 1984- 1990. unpublished report, Wessex Archaeology, Salisbury

MORALES, A., and ROSENLUND, K., 1979, Fish Bone Measurements. Steenstrupia, Copenhagen

MORRIS, E.L., 1992, The Analysis of Pottery. Wessex Archaeology Guideline 4, Salisbury

MORRISON, J.H., 1935, Prerogative Court of Canterbury wills, sentences, and probate acts, 1661-1670 (inclusive). London

MPRG., 1998, A Guide to the Classification of Medieval Ceramic Forms. Medieval Pottery Research Group Occasional Paper No. 1

MUSTY, J., and ALGAR, D., 1986, ‘Pottery’ in J. Musty and D. Algar, ‘Excavations at the deserted medieval village of Gomeldon, near Salisbury’, WAM 80, 154-66

MUSTY, J., and RAHTZ, P.A., 1964, ‘Medieval pottery’ in J. Musty and P.A. Rahtz, “The suburbs of Old Sarum’, WAM 59, 145-51

MUSTY, J., ALGAR, D.J., and EWENCE, P.F., 1969, “The medieval pottery kilns at Laverstock, near Salisbury, Wiltshire’, Archaeologia 102, 83-150

NAISH, W., 1716, The City of Salisbury with the adjacent Close, Church and River accurately surveyed [also 1751 edition]

NEVILL, E.R., 1910, ‘Salisbury: a royal aid and supply for 1667’, WAM 36, 413-34

NEVILL, E.R., 1911, ‘Salisbury in 1455’, WAM 37, 66-91

O’CONNOR, T.P., 1995, ‘Size increase in post-medieval English sheep: the osteological evidence’, Archaeofauna 4, 81-91

ORDNANCE SURVEY, 1:500 scale series, sheet Wiltshire 66.15.4, surveyed 1879, published 1880

PEARCE, J., and VINCE, A., 1988, A Dated Type-series of London Medieval Pottery. Part 4: Surrey Whitewares. London and Middlesex Archaeology Society

PLATT, C., and -COLEMAN-SMITH, R., 1975; Excavations in Medieval Southampton 1953 - 1969. Vol. 2: The Finds, Leicester University Press

PRESTON, C.D., and CROFT, J.M., 1997, Aquatic Plants in Britain and Ireland. Essex

62 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

RAMSAY, G.D. (ed.), 1954, Two sixteenth century taxation lists, 1545 and 1576. Wiltshire Record Society, vol. 10

REID-HENRY, D., and HARRISON, C., 1988, The history of the birds of Britain. London

ROACH, F.A., 1985, Cultivated Fruits of Britain. Their Origin and History. Oxford,

ROBINSON, A.M., 1988, ‘Pottery and roof tiles’ in T.B. James and A.M. Robinson, 1988, 169-93

RCHME, 1980, Ancient and Historical Monuments in the City of Salisbury, Volume I. Royal Commission on Historical Monuments (England)

SMITH, R.W., 1997, Excavations at Emwell Street, Warminster: the Early Economy and Environment of a Wiltshire Market Town. Wessex Archaeology, Salisbury

SPOERRY, P.S., 1988, ‘Documentary and other evidence for medieval and post-medieval ceramic production in Dorset’, Proceedings of the Dorset Natural History and Archaeology Society 110, 29-35

SPOERRY, P.S., 1990, “Ceramic production in medieval Dorset and the surrounding region’, Medieval Ceramics 14, 3-17

SQUIBB, G.D. (ed.), 1954, Wiltshire visitation pedigrees, 1623. Harleian Society, vol. 105/6

STACE, C., 1991, New Flora of the British Isles. Cambridge

STEVENS, K.F., and OLDING, T-E., 1985, The Brokage Books of Southampton 1477-8 and 1527-8. Southampton Records Series, 38

TERRY, R., and CHILINGAR, D., 1955, ‘Summary of “Concerning some additional aids in studying sedimentary formations’ by M.S. Shvetsov,’ Journal of Sedimentary Petrology 25.3, 229-34

TYLECOTE, R.F., 1981, “Che medieval smith and his methods’ in D.W. Crossley (ed.), Medieval Industry. Council for British Archaeology Research Report 20, 42-50

WALTON, L, 1633, The Compleat Angler. (reprinted 1939, Penguin), London

WILTHEW, P., 1987, ‘Metallographic examination of medieval knives and shears’ in J. Cowgill, M. de Neergaard and N. Griffiths, Knives and Scabbards. Medieval Finds from Excavations in London, vol. 1, HMSO

MANUSCRIPT SOURCES

Census Returns, Salisbury St Thomas, 1891

S[alisbury] C[athedral] A[rchives] Press 1, boxes 13-15, Sal 3/63: Deed of 1362 [catalogue only seen]

SCA 11: Parliamentary survey of Salisbury church property, 1649

SCA 41-59: Chapter lease books, 1608- 19th century

WRO 952/2: Lease of Antelope Inn, etc, 1768

WRO 164/1/11: Deed, Moris to Cove, 1433

WRO 1075/001/90: Deeds re Queens Arms Inn, 1757-1888

WRO 1900/44, 45, 51, 52, 54-6: St Thomas parish easter offerings books, 1574, 1584, 1600, c.1600, 1602, and undated 17th-century

WRO CC Bishoprick 460: Bishop’s and Chapter leases, etc. 1533-61

WRO CC Chapter 207: Dean and Chapter leases 1563- 1608

WRO G23/1/174, 176: Martins ward land tax assessments, 1648/9, 1651

WRO G23/1/182: Martins ward land tax assessments, 1703- 52, 1797

WRO G23/1/212: Salisbury Domesday Book, vol.1

WRO G23/1/213: Salisbury Domesday Book, vol.2

WRO G23/1/214 Salisbury Domesday Book, vol.3

WRO G23/701/1PC: Board of Health Drainage Map of Salisbury, 1854

WRO Cons Sarum wills: Elizabeth Mustion, 1644

WRO Sub-Dean wills: Thomas Iatt, 1709

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 63-70

The Great Bustard in Wiltshire: Flight into Extinction?

by James Thomas

Once a proud resident of Wiltshire, the Great Bustard, the largest land bird in England, was systematically eradicated by various forces and agencies. Its form and treatment between 1500 and 1850 are here subjected to

analysis and comment.

At first sight there would appear to be little or no connection between an Indian state, a noted English naturalist, and coats of arms for two English counties. And yet, as if by coincidence, nothing could be further from the truth. In an article published early in 1999 headed ‘Great bustard of Rajasthan is close to extinction’, the Daily Telegraph’s correspondent in New Delhi, Jan Stock, recounted how this large, cumbersome bird, frequently over four feet tall and weighing up to 40 lbs., was being slaughtered indiscriminately by poachers for its meat and preyed upon by wealthy hunters for sport. Taking a long while to be airborne, it is, in consequence, the journalist noted, an easy prey for hunters. Stock cited the national outcry in 1980 when a Middle Eastern prince planned to visit Rajasthan to hunt this seemingly inoffensive bird, an outcry which forced the potentate to abandon his trip. Stock also cited a spokesman for the Bombay Natural History Society as stating “People don’t realise how near to extinction this bird is”.! Substitute Salisbury Plain for the remote areas of Rajasthan’s Thar desert, where the Great Bustard can be found today, and there is a repetition of what took place in Wiltshire and other parts of England in the nineteenth century. Greed, farming and the desires of sportsmen spelt doom to the bird. To understand the bird’s significance in the context of Wiltshire, however, it is necessary to consider three questions. What were the bird’s prime characteristics? How did ‘it fare in early modem England? What happened to it in the eighteenth and nineteenth centuries? When answers are provided to these questions it should become abundantly clear why it was so fitting that in

1937 Wiltshire should have been granted a crest described in heraldic terms as: ‘On a wreath of the colours, A Bustard, wings elevated and addorsed, proper’; and why, three decades later, a badge should have been granted: ‘On a rounded barry of eight argent and vert a Bustard, wings addorsed, proper’.’

Outs tarda, as the Great Bustard is known officially to ornithologists, belonged to the Order of Rasores (now Stuthioniformes) and to the family Struthionidae. It was the largest British land bird, and Wiltshire ‘was probably its last breeding ground in England’. Though habitually shy, it was nevertheless capable of being aggressive, as in 1856 when a wounded specimen bit a boy’s fingers. The eggs, larger than those of a swan, are usually two in number and are olive brown in colour with darker spots. Its courting habits were pronounced, if not to say extreme. The gular pouch, running down the thorax in front of the windpipe, is accentuated in males at this time. The tail is turned back revealing the white under-tail coverts, while the primaries are crossed over the back so as to hold down the inverted tail. The head is sunk on the back, plume feathers project upwards on each side, resulting in a white fluffy mass totally unlike the normal bird. Thus arrayed, the male struts before the female, producing a guttural ‘hoc, hoc, hoc’ noise, which the hen appears to take no notice of.’ In mid-January 1853 William Yarrell, Vice President of the Linnean Society’, read a paper to the Society entitled ‘On the Habits and

School of Social and Historical Studies, University of Portsmouth, Mildam, Burnaby Road, Portsmouth PO1 3AS

64

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

The Great Bustard, drawn by Joseph Reed and engraved by T. J. Smyth, published in WAM, 1856

THE GREAT BUSTARD IN WILTSHIRE: FLIGHT INTO EXTINCTION? 65

Structure of the Great Bustard. (Otis tarda of Linnaeus)’, noting early in his presentation ‘the great scarcity, or rather, the now rare occurrence of the bird in this country, affording but few opportunities for observations’. One of his correspondents noted his experiences in Spain: ‘I may add that the bustards when flushed generally fly two miles or more, sometimes at least a hundred yards high. They never try to turn;... They fly with a regular flap of the wings, and much faster than they appear to go’.’ Straight-billed with the point of the upper mandible curved, the bustard had long, strong, muscular legs and was three-toed. Its head and neck were a bluish grey, while its back and upper breast were buff orange; all of its underparts were white. The female was generally about a third of the male in terms of size. Prone to roving and being polygamous, the bustard was both wild and difficult to approach.°® In this respect it may have been seen as more of a challenge.

I

How did this seemingly ungainly bird fare in the early modern period? Contemporaries and historians are at one in their thinking on the subject. It was hunted, it was served up as food and, as an example of early Tudor benevolence and paternalism, it was protected by legislation. The Household book of L’Estranges of Hunstanton in Norfolk provides vivid evidence of the bustard being hunted, with the entry: “The xljst weke —Wedynsday: It[e]m, vii) malards, a bustard andj hernseme, kylled w[i]t[h] ye crosbowe’.’ An early seventeenth century manuscript account of Wiltshire life extolled the virtues of both the bustard and the hunter, the author observing: And last of all the Courser may have his belly full of rideing, as well as his doggs of running, if not at the Hares in the Hare warren about Wilton, yett at that Rare and Excellent Creature, the English Aestrich (as I call a Bustard), which the Grand-father of this present Earle of Pembroke used to catch with his Greyhounds.®

Writing in the second half of the seventeenth century John Aubrey waxed lyrical about the bird, pointing out that they were to be found on Salisbury Plain ‘especially about Stonehenge’ and ‘in the fields

-above Lavington’, though ‘they doe not often come to Chalke’. Later in his work, as if to emphasise the point, he noted “These plaines doe abound with hares, fallow deer, partridges, and bustards’.’ Britton, who produced an edition of Aubrey’s work, observed: ‘It was formerly very numerous on these plains, but the

murdering tube of the sportsmen, and pilfering hand of the shepherd, have nearly exterminated the whole race’. Gilpin perhaps came closer to the truth when he noted ‘as he is so noble a prize, the flesh so delicate, and the quantity of it so large, he is of course frequently the object of the fowler’s stratagem’.!° It should come as no surprise, therefore, to learn of the bustard being served up at table. In 1519 bustard was served to the Lords of the Star Chamber on three occasions. In one instance 4s. was defrayed for it; on the other two days it cost 2s. 10d.'! Such prices indicate, indirectly, that it was already considered something of a delicacy. The selling price of a bustard at Salisbury’s Poultry Cross Market in 1555 was 10s. a bird and such was its prolific nature on the Plain that that community’s Mayor used to include it in the menu for his inaugural feast, certainly up to 1800.”

Well before this time, however, the bustard had been protected by legislation, an early and intriguing example of concern for the environment and the wildlife it supported. By legislation enacted in 1535 the taking of the Great Bustard’s eggs was prohibited, a maximum penalty of 20d. being imposed for each egg removed.!’ As always with early Tudor legislation, however, it was one thing to pass it and another to make it work. By a proclamation of May 1544 prices ‘to sell all manner of wild foul and poultry wares’ were set, so that ‘the best crane, bustard or stork, not above the price of 4s’. Furthermore, it was ordered that ‘no foreign or foreigners sell or cause to be sold within the markets of Leaden Hall, Cheapside, and Newgate market, or any of them, or elsewhere within the... city of London and suburbs of the same .. / the mean bustard at 2s. and the best bustard at 2s. 8d.'* As a further means of protection, there is evidence to show that bustards were included in early menageries. During his visit to the Low Countries in 1641 John Evelyn (1620-1706) recounted with both relish and detail many of the sights he had encountered. In Brussels early in October he noted, having visited the riding school and gardens, with their impressive backdrop of fountains and music:

There is likewise a faire Aviary; and in the Court next it are kept divers sorts of Animals, rare and exotic fowle; as Eagles, Cranes, Storkes, Bustars, Pheasants of Severall Kinds, and a Duck having 4 Wings and c: In another division of the same Close, Connys of an almost perfect yellow Colour: There was no Court now in the Palace, the Infanta Cardinal, who was the Governor of Flanders being dead but newly, and every body in deepe Mourning, which made us quitt the Towne sooner than happily we should else have don.!”

60 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

II

It was the eighteenth and nineteenth centuries, however, that were to spell doom to this magnificent bird. The naturalist Gilbert White (1720-93), who saw bustards as large, somewhat ungainly creatures, also pointed to their passing in his highly detailed journals. Thus for 13 February 1770 he noted: ‘Saw bustards on Salisbury plain: they resemble fallow-deer at a distance. Partridges pair. Wild-geese in the winter do damage to the green wheat on Salisbury plain.’

Three years later, however, he noted that bustards were being bred in the Sussex parish of Findon. In November 1787, by contrast, he was able to record a conversation with a carter on Mr Treadgold’s North downs farm between Andover and Winchester. While the farm was ‘much annoyed with Norway rats’, the carter noted changes in local life. He ‘also told us that about 12 years ago he had seen a flock of 18 bustards at one time on that farm, and once since only two. This is the only habitation to be met with on these downs between Whorwel and Winchester. !° White’s correspondent Thomas Pennant felt them to be more common in Wiltshire at this time, observing that “in autumn these are generally found in large turnip fields near the downs, and in flocks of fifty or more.!’

While specimens were offered for sale, such as the “Three fine BIRDS’ put up by William Hussey at Tilshead near Market Lavington early in November 1788,'* change was under way. Towards the close of the century the bird’s eggs were systematically gathered for incubation and hatching under hens. Such was their increasing rarity value that 10s. 6d. per egg was often paid while the young, when not fully grown, frequently commanded 10-12 guineas a pair.'” Here could be seen the beginning of the end, for the greater the obtainable price the rarer the bird was likely to become, and thus a vicious downward spiral would begin to take hold. But the egg hunter and specimen collector caused only part of the problem, for the poor bustard appeared to be fighting an unwinnable war on three fronts. While the egg hunter and his associates constituted one threat danger came, too, from the advance of enclosure and of the sportsman with his ‘murdering tube’. The bird was being driven from Wiltshire and, it should be noted, from elsewhere as well, by the simple but lethal process of eradication. The late eighteenth century rise in wheat prices, succession of bad harvests and war-time restrictions on grain imports meant that the area under the plough had to be extended and the type of cultivation intensified. As a result, the bustard

was driven from a large area that had once constituted its natural habitat. But it was not just a matter of the appearance of the egg collector and the relentless advance of enclosure. The rise of the sportsman also did more than a little to end the bustard’s days in Wiltshire.

Game and its acquisition was an important dimension of landed society in the late eighteenth and early nineteenth centuries. As sport rose in social and economic significance so the game laws grew harsher and the poacher became ever more devious. On many a Wiltshire estate the gamekeeper, who had to be licenced, and the shepherd were viewed as the top hands. Landowners could become irate when supplies of game were not forthcoming. Salisbury solicitor John Hodding, Receiver of the Duke of Queensberry’s estates in Wiltshire, most of which were centred on Amesbury, from 1788 onwards, was moved to write to his employer as follows in mid-September 1807:

I have received Mr Douglas’s Letter, with a Copy of a Letter from Mr Dubois to the Gamekeeper at Bentley; and the Gamekeeper has been with me on the Subject of it — he seems much hurt that your Grace should think he sells or gives away Game, which he declares he never did.

The Season for killing game at Bentley is not yet arrived, there being no partridges or any Land there to which they resort —- And no Game but Pheasants, Hares and Woodcocks, are to be found there, and the time for shooting them, does not arrive, until the first of October; immediately after which he will supply Your Grace with all the Game he can possibly procure.”°

Significantly the steward made no mention of the availability of the bustard on the ducal estate, suggesting that it had perhaps disappeared by then. By the same token, the pre-printed Game Book completed by members of the Penruddocke family between 1823 and 1848, recording shooting activity on their lands at Compton Chamberlayne and Baverstock, both to the west of Salisbury, make no mention of the bustard. Moorhens, snipe, rabbits, pheasant, rooks and other birds fell victim to their guns, but of bustard there was no mention.”! The evidence would appear to point to the bird’s disappearance from Wiltshire by this time. Indeed, a combination of written and early oral history would also appear to confirm that conclusion.

Over time bustard numbers, encounters and sightings in Wiltshire were to decline considerably. In 1818 William Chafin recalled a bustard hunt on the downs near Winterslow Hut in the late 1760s. There were about 25 birds in the flock and although he took a pot shot at one, he failed to bag it. With a certain amount of prescience he observed ‘I believe such a

THE GREAT BUSTARD IN WILTSHIRE: FLIGHT INTO EXTINCTION? 67

number of bustards will never again be seen together in England’.*? In c.1785 or 1786 came record of another sighting, this time of several near Chitterne Bam, while the authority, who employed the initials ‘J.S.’, noted that he had heard local farmers talking of reasons for the bird’s relative scarcity, which they attributed ‘to the heath, ..., being broken up and converted to tillage, and to the corn being weeded in the spring, whereby the birds were disturbed and prevented making their nests’. Within a year the same person saw a pair flying ‘over our heads and within gun shot, and I could distinctly see the colour of their plumage’. In c. 1792 a young bird was taken between Devizes and Salisbury and given to Mrs Steedman who kept the Red Lion in that city. She tamed it and within three months ‘it could eat off the table in the bar’. She was offered 10 guineas for the bird but declined it, only to subsequently lose it when a pointer gained entry to the parlour and slaughtered it.*? Four year later Dew, an observant and credible sportsman, saw seven or eight together on the Downs near Winterbourne Stoke but, added the author, G. Maton, ‘IT have not met with any one since who has actually seen the bustard in Wiltshire subsequently to that year’.*?

John Britton took a particular interest in the birds, recording seeing a brace in the summer of 1800, feeding on a pasture track near Tilshead: ‘I felt much gratified in beholding and examining these rare and majestic birds; and having amused myself by looking at them for some time through a telescope, I approached within eighty yards, when they sprang immediately from the ground.”

Even when away from his beloved Wiltshire, Britton was kept apprised of relevant information by his fellow county devotee and scholar William Cunnington. Thus late in July 1802 Cunnington wrote to him as follows:

Dear Sir

I was duly favored with your Book and Letter and inclosed I return you a Guinea and half for the former also all the Drawings we have left as part were returned before I purposed sending you an account of the Bustard but am disappointed in not receiving the information promised me from Tilshead — but in a fortnight I shall see the person who kept the Bird and I will then write you — I don’t see what I can write so as to be of any service to you — Camps Barrows Religious Circles and c. are so numerous

that single details would be of no service to you.*°

Britton’s knowledge upon this increasingly rare inhabitant of the county was considerable and he supplied Yarrell the noted ornithologist with much valuable material for his paper entitled, “On the habits

and structure of the Great Bustard’. Thus he recounted the experience of the rider en route from Tinhead to Tilshead very early on a June morning in 1801. A large bird, ‘which afterwards proved to be a bustard’, landed in front of the horse and ‘indicated a disposition to attack, ...’. After an hour-long struggle the rider secured the bird, offering it to his host, J. Bartley of Tilshead. Kept in a staked cage, it soon became tame and lived on a diet of sparrows, charlock flowers, rape leaves and the odd mouse as a nutritional supplement. Some two weeks after this particular incident Farmer Grant of Tilshead, while returning from Warminster Market, was attacked in a similar fashion by, it was thought, the same bird’s mate.*’ Other reports occur of encounters with the bustard in early nineteenth-century Wiltshire, but they all predate Waterloo in 1815. In 1801, for example, Reverend Wyndham’s grandfather recorded in his game book seeing a hen bustard in flight whilst riding to Upavon. In 1803 and 1804 John Waters, renting Normanton Farm on the Salisbury side of Amesbury, killed what he maintained was the ‘last of the bustards seen about at that date’. On this count, however, he was mistaken. From Lake and from Eastcott, from West Lavington, from Broad Hinton to Langley came sightings and reports of bustards being brought down. In 1802 Montagu observed that the bustard was to be found only upon the large extensive plains and that Wiltshire was virtually its last home ‘where they had become very scarce within these few years’. Increasingly, authorities referred to the bird’s scarcity, with the ornithologist Graves explaining in 1821: the enclosing and cultivating those extensive downs and heaths in various parts of Great Britain, on which formerly this noble species was seen in large flocks, threatens within a few years to extirp ate the bustard from this country; instead of being met with in flocks of forty or fifty birds, it is a circumstance of rare occurrence that a single individual is now seen.”*

Just four years later Selby maintained that the bustard was ‘extinct upon our extensive downs, of which it once formed the appropriate ornament’.”” Thereafter it was a matter of sightings and the occasional shooting or capture of visiting specimens, some of which may well have flown off course. (Examples are given in Appendix B.)

Although the bustard had disappeared from its native habitat of Wiltshire, it was not forgotten. Firstly, it lingered on in local folk memory, enabling Canon Bennett, Shrewton’s incumbent, to write in 1861 “The oldest inhabitants remember the Bustards existing in flocks but “very shy” on the downs east of Shrewton’. There were very occasional sightings of the bird, such

68 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

as that captured in the neighbourhood of Hungerford early in January 1856. One was seen at Berwick St. James in 1870, while during the following year at least seven found their way to Salisbury Plain, with sightings at Berwick St. James, Maddington, Market Lavington, and Shrewton. Accounts survive, with varying degrees of accuracy, of what fate befell them, while there are attempts at explaining why they should have arrived back in their native habitat. One theory was that they were refugees from the Franco-Prussian War, being frightened away by the roar of incessant gunfire.*” Whatever the reasons, there were a number of sightings and shootings in Wiltshire early in the 1 870s.

One female specimen was shot at Maddington Manor Farm in January 1871 by a keeper named Stephen Smith, who was in the employ of Mr. E. Lywood. The bustard was shot at a range of over 100 yards and was downed with a ‘marble’. The bird, accompanied by two other specimens, suffered a broken wing before crashing to the ground, one of its two erstwhile companions wheeling round in a seemingly frantic search for it. Measuring 31 inches in the body and with a 62 inch wingspan, the bird was presented by Mr. Lywood to the Salisbury and South Wiltshire Museum. Seemingly this specimen was one of a flock of eight, at least two of which had been shot and stuffed in neighbouring Somerset. They fetched up in the hands of Reverend Murray Matthews, Vicar of Bishops Lydeard near Taunton and Mr. Cecil Smith, squire of the same parish, ‘both of whom have large and very perfect collections of our British birds’.*!

This particular incident led to three developments. Firstly, Henry Blackmore, formerly of Salisbury, and A.P. Morres, who subsequently wrote a very useful article about the bustard, consumed a hearty lunch of bustard flesh. Secondly, Mr. King of Warminster, preserved and stuffed the bird for posterity and the benefit of natural historians. Morres wrote subsequently: “There are now 2 pairs of these grand birds in our Salisbury and South Wiltshire Museum, one pair coming from Yorkshire, killed in 1825, and the pair from our own plain — 1871.

The third development was the sending of a letter to the local journal, entitled ‘A Plea for the Bustards’, which read as follows:

Sir - may I request your valuable assistance by inserting a plea for the lives of the beautiful pair of Bustards which are still walking over the lands of this and the parishes adjoining, as doubtless they would breed, the close time being so very near, and thus pleasantly add to the unexpected novelty of their re-appearance in our generation.

In the event, however, the plea was wasted for, as Morres noted, “They soon, however, disappeared’.*?

And yet, although they had disappeared, there was little or no chance of them being forgotten in the county because of place and allied name evidence. Near Martin was Bustard Farm, put up for auction on 6 October 1796, with live and dead stock, four wagons, a cart, ploughs, six ‘good horses’ and ‘208 sound sheep’.*! By the side of the old road across Salisbury Plain is the area known as the Bustard. Immediately to the north of Rollestone Bake Farm was the original Rollestone Camp, while opposite the Bustard and occupying both sides of the road was Bustard Camp. Indeed, one of the results of the army’s arrival on the Plain was the decision to revive the Bustard Inn which stands on the edge of the ranges less than a mile to the north of the present-day Rollestone Camp.” In Trowbridge today can be found the Bustard Club, a social venue for county council employees. Though gone, the bustard 1s not forgotten.

IV

During the last thirty years there have been efforts made to return the bird to its native Wiltshire habitat. A Great Bustard Trust was established and attempts made from 1970 onwards to breed the bird on land at Porton Down, using birds and eggs imported from Germany and Russia. As the bird lays only two eggs per annum, however, numbers would be slow to recover once stocks have been allowed to fall, while the necessary habitat was no longer found to exist in Wiltshire. In the end the Trust was forced in January 1998 to admit defeat and disband itself.*° While there was a breeding programme at Whipsnade Wildlife Park in 1997 and one extant specimen called ‘Keto’, it was felt to be unlikely that ‘the magnificent great bustard will be seen roaming the Wiltshire countryside again’.*’ When, therefore, visitors to Devizes Museum stop and look at the two bustard specimens on display there, they should perhaps spare a thought or two for them. Enclosure, the seed drill, the horse hoe, predators of various sorts whether human or in the form of Vulpes Vulpes, and the wilful sportsman together made sure that England’s largest land bird was driven and eradicated from Wiltshire. And the real tragedy is that the bustard, as the great Chinese ornithologist Cheng T’so-Hsin explained, was ‘the bird most beneficial to agriculture’. Perhaps the final remarks should be the observations of that evocative observer of rural life W.H. Hudson (1862-1922), who commented in the early part of this century:

THE GREAT BUSTARD IN WILTSHIRE: FLIGHT INTO EXTINCTION? 69

Wiltshire, like other places in England, has long been deprived of its most interesting birds — the species that were best worth preserving. Its great bustard, once our greatest bird — even greater than the golden and sea eagles and the ‘giant crane’ with its ‘trumpet sound’ once heard in the land — is now but a memory. Or a place name: Bustard Inn, no longer an inn, is well known to the many thousands who now go to the mimic wars on Salisbury Plain; and there is a Trappist monastery in a village on the southernmost border of the county, which was once called, and is still known to old men as, ‘Bustard Farm’. All that Caleb Bawcombe knew of this grandest bird is what his father had told him; and Isaac knew of it only from hearsay, although it was still met with in South Wilts when he was a young man.*®

References

1 Daily Telegraph, 11 February 1999.

2 D. Buckeridge, Heraldry in Wiltshire, Vol.1(1995), n.p.

3 C. Straton, ‘The Great Bustard’ in The Festival Book of Salisbury 1864-]914 (Salisbury, 1914), pp. 11-13. A member of W.A.N.H.S., Straton died on 22 February 1918, aged 75: WAM 40 (1917-19), pp. 199-201.

4 Carl Linnaeus (1707-78), was a Swedish natural historian who contributed much to the concept and application of classification. He was the first to enunciate the principles for defining genera and species.

5 Yarrell’s paper was reported in The Morning Chronicle. William Yarrell (1784-1856) was a noted zoologist and bookseller based in London. A Fellow of the Linnaean Society in 1825, he served as its Treasurer between 1849 and 1856 and was an original member of the Zoological Society in 1826. His published work included History of British Fishes (1836) and History of British Birds (1843): D.N.B.

6 Rev. A.C. Smith, “The Great Bustard’, WA.M. 2 (1856), pp. 129-30. Smith was incumbent of Yatesbury, a few miles to the east of Calne.

7 H. Fraser Fortescue, ‘A Lost British Bird: the Great Bustard’, The Badminton Magazine (July 1903), p. 92.

8 H.C. Brentnall, ‘A Longford Manuscript’, WA.M. 52 (1947-8), p. 20.

9 J. Aubrey, The Natural History of Wiltshire (ed.) J. Britton (Wiltshire Topographical Society, 1847), pp. 64, 108.

10 J. Britton, The Beauties of Wiltshire, vol. 2 (1801), p.114.

11 A.L. Simon, A Concise Encyclopaedia of Gastronomy (1983 edn.), p. 515.

12 C. McKeown, ‘Gone Forever? Wiltshire’s Lost Birds’, Wiltshire Life, September 1997, p. 25; R. Whitlock, A Victorian Village (1990), p. 182; A Tryon, ‘Return of the Great Bustard’, Country Life, 26 July 1973, p. 244.

13 H. Fraser Fortescue, p. 92.

14 H.E.S. Fisher and A. R. J. Jurica (ed.) Documents in English Economic History: England from 1000 to 1760 (1984), pp. 484-5.

15 E.S. de Beer (ed), The Diary of John Evelyn (5 vols.,

1955), vol. 2, p. 72. Evelyn was in error about the Cardinal Infanta’s death as Ferdinand died 31 October 1641.

16 W. Johnson (ed), Journals of Gilbert White (1982), pp. 20, 69, 280. Rattus norvegicus or Rattus decumanus, the brown rat, had probably reached England in the late 1720s aboard ships trading with Russia and had spread rapidly: Johnson, Gilbert White (1982 edn.), p. 76.

17 Quoted Rev. A. C. Smith, “The Great Bustard’, WA.M. 3 (1856), p. 132. Thomas Pennant (1726-98), traveller and naturalist, produced British Zoology (1766) and History of Quadrupeds (1781): D.N.B.

18 Salisbury and Winchester Journal, 3 November 1788.

19 H. Fraser Fortescue, pp. 95-6.

20 John Hodding to the Duke of Queensberry, 15

September 1807: WRO 377/1. Bentley Woods, located

in West Dean parish, contained about 672 acres,

producing in tmber and underwood felled and sold, approximately £450 per annum for the Duke.

WRO 332/282, Game Book 1823-48, np.

R. Whitlock, A Victorian Village (1990), p. 182.

J. S., ‘The Bustard’, WA.M. 1(1854), p. 212.

4 G. Maton, “The Natural History of a Part of the County of Wiltshire’ (1843), p.5. Copy contained inW.A.N.H.S. Library, Wiltshire Tracts no. 2.

25 J. Britton, The Beauties of Wiltshire, vol. 2, (1801), p.115.

26 William Cunnington to John Britton, 21 July 1802: W.A.N.H.S. Library, MS. 2600, Cunnington MSS., Box 326, n.f. At this time Britton (1771-1857), the noted antiquary and topographer, was residing at no. 18, Wilderness Row, Goswell Street, London.

27 Rev. A. C. Smith, ‘The Great Bustard’, WAM. 3 (1856), pp. 134-5. Mr. Bartley subsequently sold his bustard to Lord Temple for 30 guineas. Temple was already in possession of one specimen.

28 A.P. Morres, ‘On the Occurrence of some of the Rarer Species of Birds in the Neighbourhood of Salisbury’, WAM. 20 (1882), p. 179. At least one downed specimen at this time was sent to the Duke of Queensberry. No acknowledgement of the gift was received. Morres was incumbent of Britford, to the south-east of Salisbury.

29 Quoted Rev. A. C. Smith, “The Great Bustard’, WAM, 3 (1856), p. 138. John Prideaux Selby (1788-1867) naturalist, was High Sheriff for Northumberland in 1823. Between 1825 and 1834 he published I//ustrations of British Ornithology. In 1842 he published British Forest Trees: D.N.B.

30 WAM 46 (1932-4), notes, p. 392; 25 (1891), p. 360; B. McGill, Village under the Plain: The Story of Market Lavington (1995), p. 121.

31 Devizes Advertiser, 2 February 1871; A.P. Morres, ‘On the Occurrence of some of the Rarer Species of Birds in the Neighbourhood of Salisbury’, WAM 20 (1882), p. 181.

32 Morres, p. 182. For a bustard recipe — Bustardu Double Beurre — see A. L. Simon, p. 515. Taxidermy, private and museum acquisitions deserve closer study in the

be

NS Ww bw bo Q

70 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

county. A Great Bustard shot in Allington Mead on the banks of the Avon in early February 1891 was sent to Foot the Bath taxidermist for preservation: WAM 25 (1891), p. 361. A male Great Bustard, captured near Hungerford in January 1856, was among the articles exhibited in a temporary museum at the town hall in Warminster in August 1856. The town’s Literary and Scientific Institution also exhibited a specimen. Articles contributed to the Loan Museum in 1875 by S. A. Jeffreys of Melksham included two specimens of bustard. In 1868 William Cunnington purchased a specimen from the Warminster Institution for £15 and disposed of it to the W.A.N.H.S. Council ‘on the condition that in the event of the Society being broken up, the specimen was to be offered, at the same price, to me, or to members of my family’. By comparison £18 6s. 7d. was spent in the same year on ‘Stationery, postage, Carriage, Advertising and c.’: WAM, 3 (1857), pp. 267, 268; 15 (1875), p. 138; 12 (1870), interleaved vol., note by Cunnington.

33 Quoted A. P. Morres, p.181.

34 Salisbury and Winchester Journal, 3 October 1796.

35 N.D.G. James, Plain Soldiering: A History of the Armed Forces on Salisbury Plain (1987), p. 124.

36 S. O’Neill, ‘Time and space run out for the Great Bustard’, Daily Telegraph, 20 January 1998.

37 C. McKeown, ‘Gone Forever? Wiltshire’s Lost Birds’, Wiltshire Life, September 1997 pp. 24-5. Other sources dealing with the re-establishment programme include a pamphlet produced by the Great Bustard Trust and A. Tryon, ‘Return of the Great Bustard’, Country Life, 26 July 1973, p. 244.

38 W. H. Hudson, A Shepherd’s Life (1946 edn.), p. 84.

Appendix A Ornithology and Wiltshire Inns, 1800 — 1850

Set out below are details of the Wiltshire towns containing inns with bird names. While they show a certain amount of continuity, they may also provide food for thought for other researchers. Of particular note is that 14 of the 23 establishments involved the swan, important for its regal pose and its role as a symbol of innocence. Of these, four were the White Swan suggesting, perhaps, a greater degree of innocence. The sole Black Swan, located in Devizes, may have had connections with empire. The black swan (Chenopis atrata), native to Australia, was introduced to New Zealand. Aboard the French vessel Naturaliste, held at Gosport in May 1803, were two black swans, a pair of emus and many potted plants: William Cole to Sir Joseph Banks, 31 May 1803:W.R. Dawson (ed.), The Banks Letters (1958), p. 223.

BRADFORD Swan 1822 - 1838

CHIPPENHAM

Cock 1822

Swan 1838 - 1848 DEVIZES

Black Swan 1809 - 1844

Pelican 1830 - 1848

White Swan 1838 - 1848 HIGHWORTH

Swan 1830 HINDON

Swan 1830 MALMESBURY

Swan 1830 MERE

Swan 1830 SALISBURY

Bird in Hand 1822

Chough 1822

Falcon 1838

Pheasant 1822

Spread Eagle 1822

Swan 1830

Three Swans 1830

White Swan 1822 TROWBRIDGE

Swan 1830 - 1848

White Swan 1822 WARMINSTER

Cock 1822

Swan 1822

White Swan 1838 - 1848

Sources: Holden’s Triennial Directory (1809); Pigot’s Directory (1822), (1830); Robson’s Commercial Directory of London and the Western Counties, vol. 2 (1838); Hunt and Co’s Directory (1848)

Appendix B Sightings of the Bustard, 1877 — 1998

1877 Salisbury Plain 2

5 Dec 1879 Woodham Ferrers, Essex 1 8 Dec 1879 St. Clement, Jersey 2

Dec 1879 Romney Marsh, Kent 1

1880 Great Chard, near Ashford, Kent 1

10 Jan 1880 Cranborne Downs, Dorset 1

Jan 1880 Wye, Kent 1

6 Feb 1880 West Wickham, Cambridgeshire 1 1958 — 98 Twenty sightings 20

Nov 1998 Poole, Dorset 1

Sources: A. P. Morres, ‘On the Occurrence of some of the Rarer Species of Birds in the Neighbourhood of Salisbury’, WA.M, 20 (1882), p. 182; Daily Telegraph, 20 Jan 1998; Sunday Telegraph, 8 Nov 1998.

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 71-81

Beaker Pits at Crescent Copse, near Shrewton, Wiltshire, and the Effects of Arboreal Fungi on

Archaeological Remains

by Michael Heaton! and Rosamund M J Cleal’ with contributions by Peter Higgins,’ Peter Bellamy,* Sheila Hamilton-Dyer’ and John Wilson®

Excavation of six pits of ‘Beaker’ date during the summer of 1997 revealed, in addition to prehistoric materials, evidence of extensive, and possibly recent, fungal activity, resulting in an almost total absence of identifiable organic remains despite the presence of large quantities of charcoal dust. Because of the small scale of the work, the resultant report does not attempt to draw wide-ranging conclusions about the topographic or temporal distribution of Beaker pits, but presents descriptions of the principal assemblages together with some observa- tions on the action of fungal mycelia on buried organic remains. Published studies on soil micro-organisms suggests that these may have spread from the adjacent conifer plantation, either as the natural result of afforesta-

tion on otherwise dormant soil fungi, or as a result of fungi introduced with the pine trees.

INTRODUCTION

Crescent Copse is situated at the head of one of the tributary dry-valleys of the Till catchment, 3.5km WSW of the village of Shrewton, on the chalk massif of the Salisbury Plain Training Area - hereafter ‘SPTA’ (Figure 1). The adjacent section of Track 21G, one of a network of un-surfaced routes that criss-cross the SPTA, had, by winter of 1996, become sufficiently impassable to warrant upgrading. In line with the guidance of PPG16, the Defence Estates Organisation (Lands) South-West (hereafter DEO(L)SW) considered the likely archaeological impact warranted evaluation and, if necessary, mitigation. Accordingly the impact of the proposed works was evaluated by Wessex Archaeology (Wessex Archaeology 1996), leading to targeted excavation of selected features by Michael Heaton the following year. The entire project, from evaluation to publication, has been carried out under the aegis of Gifford and Partners.

The excavation concentrated on two adjacent 50m and 125m stretches of the track. Topsoil stripping revealed three groups of potential archaeological features; two of which, at the northern end of the site and its centre, proved not to be anthropogenic, but included a cluster of six small pits (numbers 125-7, 131, 150 and 159) at the northern end in which sherds of prehistoric pottery were visible. One of the pits had been masked by a thick layer of vehicle-compacted chalk. There were no extensive soil or colluvial deposits.

METHODOLOGY

The upper profiles of all six pits (not considered to be pit fills per se) were excavated rapidly without sampling, and artefacts within them were recorded as bulk finds only. All other artefacts from the lower layers - the pit fills proper - were recorded to individual 3D

1.12 Victoria Road, Warminster, Wiltshire. BA12 SHE 2. Alexander Keiller Museum, Avebury, Wiltshire 3.Southern Archaeological Services Ltd.,

Unit 7 Kingsbury House, Kingsbury Road, Southampton, SO14 OJT 4.51 Fordington High Street, Dorchester, Dorset, DT 1 1LB 5.5 Suffolk Avenue, Shirley, Southampton SO15 5EF 6.5 Stuart Place, East Twerton, Bath.

&, si Sea] ,

Fig. 1. The Site and its situation

‘

/

-’ <__Rellestone Camp

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

y , iY Nor \ ie N

< / Robin Hood"s "Bail

\ Crescent Copse_ pig LJ ers

1

SU 45 Carkhill 1

7 re a

a

\. SU 10270

edges of strippe

BEAKER PITS AT CRESCENT COPSE NEAR SHREWTON

positions for which numeric references (2000>) were assigned. These positions were recorded on levelled 1:10 single context plans of each layer. Where groups of more than one object were superimposed or situated immediately adjacent to each other (generally <5mm apart), they were recorded as decimal subdivisions of one object position; e.g. 2012.1, 2012.2, etc. All the soil from these layers was passed through a lcm sieve on-site to remove chalk and flint lumps, and the residues retained for flotation separation of their palaeo-environmental component.

RESULTS

The six pits had been excavated into bedrock chalk immediately adjacent to each other within an area of approximately 5 x 5m (Figure 1). All were slightly oval, approximately 0.70m in diameter, with depths varying between 0.30m and 0.70m, and had slightly rounded bases. They shared elements of a common stratigraphic sequence, represented by Pit 127 (Figure 2):

A. Compacted topsoil containing fragments of coal, and therefore of recent derivation, which had been pushed down into the tops of the pits to a depth of between 0.10-0.15m.

B. Shingly ‘A/C’ horizon of patinated flint pieces, largely natural but containing some worked pieces. This layer is present at the base of the ‘natural’ soil sequence across much of the chalk downs and is considered to be a natural sorting product of pegogenesis. It filled, level, a deep depression in the surface of the underlying layer.

C. Pit fills. Three recognised: an upper fill of yellowish brown silty loam, forming an inverted cone up the sides of all of the pits; a middle fill of almost black silty loam permeated by extensive fungal mycelium, in most of the pits; and a Jower fill of very powdery, ashy, grey silty dust lying across the base of one or two of the pits, also penetrated by a network of fungal

1

mycelium. Within these, however, no finer horizon boundaries were discernible, despite incipient layering/grouping being evident in the distribution of stones.

>z

\ Flint

A Burnt flint

@® Pottery

— Bone Sandstone

Fig. 2. Pit 127. Horizontal and vertical distribution of artefacts

0 10 20 30 40

I dis 1 0 A a MB cts

Within each pit, the bulk of the artefacts and animal bone were contained within the dark ashy layer. There appears to be no other significant spatial patterning, the majority of the material being principally around the edges of each pit, sometimes with a weighting to one side, and without any apparent patterns in the vertical distribution. There were no groups of artefacts or significant single artefacts lying on the base of any of the pits. This distribution appears similar to that noted at Dean Bottom, considered there to represent incidental incorporation within the pit fills. Total artefact and animal bone quantities appear to reflect pit volume; the deepest (125 and 127) producing the largest amounts of worked flint and pottery (Table 1).

Table 1. Finds totals per feature. Quantities given as number/weight (g). Burnt worked flint is counted as worked flint.

FEATURE 125 126 MATERIAL

Worked flint 45/975 26/225 Burnt flint 21/1340 5/192 Pottery 9/68 4/35 Animal bone 9/38 1/1

Stone 1/6 0/0

127 131 150 159 47/725 14/189 9/124 4/216 4/250 8/318 4/246 5/302 11/171 4/20 1/4 0/0 7/59 0/0 0/0 0/0 2/24 2/24 0/0 0/0

50cm

74 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Artefacts

POTTERY (Figure 3) by Rosamund M. J. Cleal

The assemblage comprises 28 sherds (weighing 291g), representing at least seven vessels. With the exception of one sherd from F127 (2030), which is perhaps intrusive, all are assignable to the Beaker tradition of the later 3rd millennium or earlier 2nd millennium cal BC. Counts, weights and ‘P’ numbers are given in Table 2, by context. ‘P’ numbers indicate identifiable separate vessels, for each of which only a few sherds have survived; not all are illustrated here.

The recovered pottery has been examined by using a x10 hand lens. Fabrics were defined on the basis of the non-plastic inclusions (1.e. the non-clay materials present in the clay either as naturally occurring inclusions or added by the potter). Frequency of inclusions was estimated by eye by surface area (using comparative charts) and should be regarded as approximate.

Description

P1. Seven sherds of a single vessel decorated with square- tooth-comb impressions and other impressions. The fabric is soft, with a hackly fracture, containing approximately 10- 15% small to moderate bone (<4mm, most <2mm), sparse sand (mostly fine) and rare flint (<6mm) (Fabric code BS:1). The bone mainly comprises small white fragments and appears to have been well crushed into small pieces, including fragments with rounded edges as well as angular pieces. A minority of fragments are bluish-grey, at least in part. Exterior and interior sherd surfaces: medium red- brown; core dark grey. The condition is fair, with some abrasion; where the original surfaces survive they are smoothed. The decoration has been applied with two different implements, one a comb with roughly square teeth, the other perhaps a worn comb (the condition of the sherd precludes a certain identification).

Six sherds were from the middle and upper fills of feature 125, and one from feature 127. Because of the soft, friable nature of this fabric it is unlikely that the sherds had been exposed for a long period before being deposited in the feature.

P2. Single body sherd in a soft fabric with a smooth fracture and slightly sandy feel, containing moderate fine sand and rare coarse sand. Some fine (<Imm) grog (crushed potsherd) is present but it is difficult to distinguish from the matrix; there are rare fine calcareous fragments (too small for certain identification), possibly chalk (Fabric GS:1). Exterior and interior sherd surfaces: pale orange brown; core: dark grey. ‘The edges and surfaces are in fair condition. The decoration is incised. From lower fill of pit 126.

P3. Single sherd in a soft fabric with a smooth fracture and feel. Fine grog (<1mm) is present, probably sparse, but it is difficult to distinguish from the matrix; there is also sparse to moderate fine sand (fabric code GS:2). Exterior surface: pale brown; interior surface pale grey brown; core: grey. The sherd is in fair condition and the decoration is of shallow closely-spaced grooves. From middle fill of pit 125.

P4. (Not illustrated) Single rim sherd in fabric GS:2. It is probably an everted rim, but the rim angle is uncertain. The decoration is of horizontal lines of small square-tooth- comb impressions. Exterior orange; interior: grey brown; core: bi-coloured as for surface colours; the sherd is in fair condition. From upper fill of pit 125.

P5. Two large body sherds, conjoining along an ancient break, and one sherd not joining, of a large, thick-walled vessel decorated with large applied pellets or bosses of clay; four other plain sherds are almost certainly of this vessel. The fabric is soft, but brittle, with a hackly fracture and rough texture; 1t contains moderate (c. 10-15%) angular flint (<4mm, most <3mm), sparse to moderate grog (<10%; although it is difficult to distinguish from the matrix and could be present in greater frequencies than this), and sparse sand, mostly fine (fabric code FGS:1). Exterior surface: orange brown; interior: dark brown; core bi-coloured, exterior third orange, interior two-thirds black. The orientation of the sherds is not certain. The sherds are in fair condition. From the fill of pit 127.

Po. (Not illustrated). Three plain body sherds almost certainly of one vessel in a soft fabric with a hackly fracture containing moderate to common grog (c. 20-25%, <3mm, most <lmm, but difficult to distinguish from the matrix), rare fine calcareous fragments (?chalk; too small for certain identification) and some sand (fabric code GS:3). Exterior: orange-red; interior pale brown; core: black. The sherds are in fair condition. One sherd was in the A/C horizon at the top of pit 127, one in its fill and one from the fill of pit 131.

P7. (Not illustrated) One decorated body sherd and one fragment probably of the same vessel, in a soft fabric with a hackly fracture and slightly sandy feel, containing sparse, heat-crackled flint fragments (<3mm, most <Imm) and sparse to moderate coarse sand. Exterior: orange-brown; core: dark grey; interior: orange-brown. The decoration is a filled horizontal band demarcated on at least one edge by an incised line. The filling is of oblique incised lines, possibly similar to Clarke’s Motif no 10 (Basic European Motif Group 1, Clarke 1970, 425), but the sherd is too fragmentary to be certain of the motif. From the fill of pit 126.

Other sherds Six other sherds were not assignable to a particular vessel, although all were probably derived from them.

BEAKER PITS AT CRESCENT COPSE NEAR SHREWTON

Table 2. Catalogue of pottery

Feature No. Object No. Context No. Qty Weight 125 2049 (136) 1 4g 2053 (137) 1 5g 2056 (137) 1 4g 2059 (137) 1 4g 2060 (137) 1 2g 2074 (138) 1 7g 2088 (138) 3 42g Totals 9 68g 126 2097 (else) 1 lg 2109 (158) 1 4g 2110 (158) 1 25g 2110 (158) 1 2g Totals: 4 35g 127 2003 (129) 1 10g 2011 (130) 4 Og 2019 (130) 1 7g 2022 (130) 1 54g 2026 (130) 1 20g 2030 (130) 1 66g 2030 (130) 1 2g 2033 (130) 1 3g Totals iat 171g 131 2034 (132) 1 2¢ 2035 (134) 1 12¢ 2035 (134) 1 4g 2043 (134) 1 2g. Totals: 4 20g Grand total: 28 291g

Vessel Observations pointed rim with single incised line; not assignable to a vessel; fabric GS:2 is black on surfaces and along ancient broken edges, possibly from burning in antiquity

Pal

Pal

Pil

P4

P3

Pl

P7? Pi P2

conjoining along ancient break with sherd from 2022

possibly intrusive

crumbs, with some flint

Po plain body sherd, flint tempered sandy

A single small rim from pit 125 in fabric GS:2 could belong to P3 on the grounds of fabric, but this seems unlikely on the grounds of the decoration (see Table 2). It also appears to have been burnt as a sherd. It is possible that this represents another vessel, not otherwise represented.

Discussion

This is a small assemblage but is important in two respects: firstly because it is one of a relatively small number of non-funerary Beaker assemblages in Wiltshire; and secondly because it is one of only a handful of known occurrences of bone used as an added non- plastic material in pottery production at this period.

Wiltshire (and the Wessex region in general) is well-known as an area rich in funerary Beakers; it is much less rich, however, in Beakers from other contexts and in particular in what have traditionally been referred to as ‘domestic’ Beakers (e.g. as by Gibson 1982). Isolated finds do occur, but larger assemblages are rare. This is not universally true of Beaker finds in Britain as in some areas, such as East Anglia, there are both plentiful burial finds and many non-funerary assemblages.

The classification of Beakers such as those in the features here is not easy, as the vessels are each represented by only a few sherds, and the classification even of whole Beakers is still a matter for debate. In

76 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Fig. 3. Pottery vessels P1, P2, P3 and P5 (with inset)

terms of date it is only possible to state with confidence that they are likely to belong to the period 2600-1800 BC (Kinnes et al. 1991) and probably to within c. 2400-1800 BC). Humphrey Case, however, in a recent reassessment of Beaker typology and dating (Case 1993) has revisited his earlier scheme (Case 1977) and proposed a series of Beaker groups, A - E, with a chronology slightly more refined than this. In addition, he has made a particular examination of Wiltshire Beakers (Case 1995a) in which he classifies the majority as falling within his groups B and D. Group B generally is dominated by Case’s style 3 Beakers (formerly ‘Late Style’) and D largely by those of style 2 (formerly the ‘Middle Style’). Group D he suggests as emerging around the middle of the 3rd millennium BC, surviving until perhaps the second quarter of the 2nd millennium, with Group B emerging at about the same time, in northern Britain, spreading to the south during the fourth quarter of the 3rd millennium BC (Case 1993, 1, 4). (This is of necessity a highly simplified and abbreviated account, to set the Crescent Copse assemblage in context.)

The form of none of the Crescent Copse vessels is certain. The whole vessel illustrated (Figure 3), is intended to suggest the general type of vessel rather than to be a reconstruction of P5, which is too fragmentary for even the profile to be at all certain

although it is perhaps likely to have been slightly slacker than that illustrated. Coarse, thick-walled vessels with heavy plastic rustication are a feature of ‘domestic’ beaker sites in East Anglia and elsewhere, and such vessels are assigned by Case to his Group B, particularly in East Anglia (Case 1993, fig. 13). Case notes the occurrence of three Group B non- grave groups in south Wiltshire, noting particularly that from Butterfield Down, near Amesbury (8km south-east of the present site) (Case 1995, 4). The use of such distinct bosses as on P5 is unusual, particularly in such a rusticated manner; it is more usual to find them as a single row, such as at Stonehenge (Cleal 1995, P17, 363) and indeed no close parallels for the decorative scheme on P5 are known to the writer. Such decoration is, however, more nearly represented among Dutch Beakers, as illustrated by Gibson (1982, figs N MISC 1, 3 and 4).

Of the other vessels little can be said, given the broad range of the motif types represented. Those of P1 and P2 are ubiquitous, falling into Clarke’s motif group 1, the Basic European group (Clarke 1970, 425) but the horizontal grooves of P3 are more unusual and suggest a northern contribution. This type of decoration (Clarke’s Motif 21 in his late Northern British motif group, op cit 426) is usually found on

BEAKER PITS AT CRESCENT COPSE NEAR SHREWTON

the neck, which is often fairly straight (as illustrated in Case 1995a, fig. 2: 6), but in this case it is impossible to determine the location of the sherd on its vessel. This hint of northern-ness, however, is in keeping with a classification of the assemblage as Case’s group B. This being so, the suggested date for it, according to Cases’s chronology, is probably no earlier than the last quarter of the 3rd millennium BC (Case 1993, 254, 257).

At Butterfield Down, one of the geographically closest sites with Group B Beakers, cited by Case, an exceptionally large Group B vessel with incised decoration was found in a small pit in which a finger- pinched vessel and a comb-decorated vessel were also represented. The two latter were tempered with bone (Cleal 1996), giving another point of similarity with the Crescent Copse assemblage. It seems that bone must only rarely have been used as a tempering material or it would have been more widely recognised by now, but new cases occasionally come to light. When the two Butterfield Down vessels were recognised as having bone, only two other occurrences in Wiltshire were known to this writer, through the publication of a list of such occurrences by Smith and Darvill (1990, 152). Of the seven cases listed by them the two Wiltshire occurrences were a sherd presumably of earlier Neolithic bowl from Robin Hood’s Ball (Thomas 1964, table 1; type of pottery not specified but the rest of assemblage is bowl) and a sherd from the henge ditch at Avebury (Gray 1935, 138, note 1). The latter is published as Peterborough Ware (Gray 1935, 138, fig. 7, 163) but appears from the illustration to be perhaps Beaker or another Bronze Age type (the sherd has not been examined by this writer, but Smith notes that there is almost certainly no Peterborough Ware among the sherds identified by Gray as such, suggesting instead that they include Beaker, Collared Urn or Biconical Urn and sherds in a Bronze Age type fabric, Smith 1965, 229). In addition to these, and to the Butterfield Down and Crescent Copse Beakers, there is now a Grooved Ware vessel from Marlborough (personal information and writer’s own observation; publication in preparation by Cotswold Archaeological Trust), and a Beaker from a probable post-hole on Overton Hill, Avebury parish (unpublished, Site SY72, Alexander Keiller Museum, Avebury; identification by the writer).

The Butterfield Down assemblage, as well as including bone as a temper, is also exceptional due to the unusually large size of the incised Beaker, which has an estimated volume of about eight litres. This is at the extreme end of the range of Beaker volume,

tate

which is generally under two litres (Case 1995b, 56). At Crescent Copse there is also clearly a very large Beaker, represented by the sherds of P5. As a sherd of P1 occurred in the same feature (F127) as the sherds of P5 and the two are, therefore, likely to have formed part of the same assemblage, it can be suggested that Crescent Copse is demonstrating a similar pattern of deposition to Butterfield Down in that at both sites there are very large vessels probably accompanying smaller, finer ones, including crushed bone in at least one of the vessels. Whether this represents a pattern repeated over many years and generations, or is the result of a restricted group of people moving around a fairly limited area over a few years or generations is impossible to say. More light may be shed on this by future finds and by radiocarbon dating of material associated with such finds, or dating material within the vessel walls themselves as that becomes increasingly common.

WORKED FLINT by Peter Bellamy

The flint assemblage comprises 154 pieces weighing a total of 2603g, plus a further 20 unstratified pieces recovered during the evaluation. The artefacts are all made from ‘chalk flint’. The nodules have many thermal surfaces and many of the pieces exhibit faults and fractures, suggesting that the raw material was obtained from the surface, rather than quarried flint. All the pieces are white or white/grey patinated, some with calcareous concretions. Generally they are in a fresh condition, indicating that most of the pieces were buried not long after manufacture.

The flint from all the pits is similar in character and probably derives from a single industry. The material from the topsoil and ‘A/C’ horizons is also similar. The vast majority of the pieces are waste primary and secondary core-trimming flakes, most of which have fairly thick plain butts and were probably removed with a hard hammer. A very small number of small blade-like flakes are also present. There are three cores, together with four pieces of naturally thermally-fractured flint with one or two flakes removed (the result, either of testing of the raw material, or of shattering at an early stage of preparation). Two of the cores are small, with several platforms exhibiting traces of small flake removals. All the cores were exhausted. The only implements present are six scrapers (from pits 126, 127, 131 and 150). Three of the scrapers (2028.1, 2047, 153) have regular semi-abrupt retouch on fairly thick flakes; two

78 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

(2102, 2012.2) have fine regular retouch on small thin flakes; and one (2099) had coarse, almost denticulate retouch on a small broken thermal blank. There is little chronologically diagnostic material present in this assemblage, but it does exhibit some characteristics of a late Neolithic/Beaker industry

Palaeoenvironmental materials

PLANT, INSECT AND MOLLUSC REMAINS by Peter Higgins

The retained pit fills were disaggregated in dilute hydrogen peroxide, wet sieved and floated over a 250 micron mesh, dried, scanned under low magnification and sorted. The assemblages are summarised numerically in Tables 3-4. With the exception of mollusc shells and animal bone fragments (see below) the results of flotation separation were disappointing. Plant remains, with the exception of hazelnut fragments, were virtually non-existent, whilst the poor condition of all arthropod remains precluded identification to species. Terrestrial molluscs were present in larger numbers, though many were damaged, accounting for a relatively high number of ‘indeterminate’ identifications. Many of the shells were from immature individuals. Of the species identified (see Table 4), Helix sp. is probably Helix apersa, though this is not a secure identification; this species was introduced in the Ist century AD, is often found near human habitation, and hibernates by burrowing into the soil. It is thus probable that human occupation of the area was not confined to the Neolithic/Bronze Age and that contamination of these deposits has occurred. Discus rotundatis, Pupiula muscorum and Vellonia costata occur together in open, watered grassland. Nesovitrea

Table 4. Catalogue of molluscs. (NB: Helix sp. present only as small fragments, therefore recorded only as present (P))

FEATURE 125 126 127 131 159

TOTAL 368 146 553, 155 362 SPECIES % Indeterminate 18.5 26.03 6.87 12.9 16.48 Discus rotundatus 31 29.54 40.5 51.61 24.59 Pupilla muscorum 9 16.44 33.81 29.03 30.39 Velloniacostata 12 0 0 0 0 Nesovitrea

hammonis 21 Lida 9.2 0 0 Oxychilus alliarius 6 14.38 5.97 6.45 14.09 Columella

edentula 0.6 26.03 7.23 12.9 20.8 Punctum

pygmaeum 0 0 0 0 4.14 Helix sp. Pp P 12 P 1e

hammonis and Punctum pygmaeum are sometimes found in association with them, although it seems more likely, given the presence of Oxychilus alliarius and Columella edentula, that these four species are from open wooded habitat. Both these habitats could be present at the same time in the area, or they could represent a succession. However, the presence of all these species together in most deposits suggests a degree of pre-deposition mixing.

ANIMAL BONE by Sheila Hamilton-Dyer

The small assemblage of animal bone is summarised in Table 5. The charred and calcined fragments were fragile but comparatively well preserved; the other bones were all much eroded. None of the fragments was sufficiently well preserved for the observation of gnawing or butchery marks, and most were too small

‘Table 3. Catalogue of all materials recovered during flotation separation

FEATURE 127 131 125 126 159

MATERIAL

Animal bone 5/<lg I/<lg 6/2g I/<lg 0 Mollusc 553 155, 368 146 362 Arthropod 6/<lg 0O 0 0 0 Hazelnut shell 0 I/<lg I/<lg 0 0 Pottery Z2<1ig: 0 I/<lg 0O 1/<lg

Burnt flint 18/lg 5/6g 21/3g 19/lg 5/4g

Table 5. Catalogue of animal bone (combined manual recovery and sieving results)

FEATURE 125 126 127 131 TOTAL 5 2) 12 1 SPECIES

Unidentified 1 1 4 1 Pig 1 2

Pig-sized 1 3

Cattle 1 Cattle-sized 3 2

BEAKER PITS AT CRESCENT COPSE NEAR SHREWTON

for precise identification. Heavily charred bone from Pit 125 included parts of one or more cattle; that from Pit 127 comprised several bones comparable with pig, from the forelimb and the heel, and part of a cattle humerus. Uncharred bone fragments were too eroded and fragmentary to be certain of identification, but matched pig better than the other possibilities, such as sheep or dog.

DISCUSSION

Salisbury Plain encompasses one of Europe’s most extensive surviving tracts of prehistoric, Romano- British and historic landscape. The SPTA borders the UNESCO Avebury and Stonehenge World Heritage Site, and Crescent Copse is intervisible with Stonehenge itself. In addition to the numerous long barrows, round barrows, linear earthworks, deserted settlements, colluvial accumulations and field systems that characterise the archaeology of the Plain, Crescent Copse is directly overlooked by the Neolithic causewayed enclosure of ‘Robin Hood’s Ball’ situated approximately 400m to the north (Figure 1).

Pits are ubiquitous amongst the archaeological features of almost all periods of Britain’s past, but those of small surface area (<0.5m?’) for which utilitarian functions are not readily apparent are typical of the Neolithic and early Bronze Age. Unlike ditches, settlement earthworks and other extensive archaeological deposits which are detectable by aerial photography, geophysical survey or by linear evaluation trenches, these small earlier prehistoric pits are invariably revealed as chance finds during the investigation of larger archaeological sites. As result, their geographic distribution is poorly understood.

Nonetheless, within a 20 x 20km area centred on Crescent Copse there are approximately 14 published sites of pits dated by ‘Beaker’ pottery, of which most comprise 2-3 individual pits, and most are situated at the heads of tributary dry valleys. They are characterised by surface diameters of less than 1m, slightly rounded profiles invariably less than 0.30m deep, containing at least one layer of dark, ashy loam with a wide variety of artefacts and palaeo- environmental materials. Although individual cases have produced considerable quantities of pottery such as at Butterfield Down near Amesbury (Rawlings and Fitzpatrick 1996) or large flint and palaeo- environmental assemblages such as at Dean Bottom

19

near Ogbourne St Andrew (Gingell 1992), others have proved to be almost culturally sterile. They remain one of the more enigmatic of Britain’s archaeological ‘type’ features, but are generally recognised to be of non-utilitarian function, especially because of the otherwise almost entirely funerary distribution of Beaker pottery, particularly within the vicinity of Stonehenge (Cleal, in Richards 1990).

The pits revealed here, therefore, are of a not uncommon form. They are undistinguished in the range, quantity, and, with the singular exception of the small pottery assemblage, quality of materials recovered from them. Their principal significance lies in their pottery assemblage, their topographic situation isolated from funerary monuments at the head of a watercourse, and the apparent post- depositional de-pauperisation of their organic content.

PREHISTORIC CONTEXT OFTHE SITE by Rosamund M. J. Cleal

The condition of both the flint and pottery from the pits suggest that they were used and deposited within a fairly short period of time. Neither artefact type shows the sort of wear which would result from trample or even only from exposure to weather on a ground surface.

Because the organic component of the pits has apparently been impoverished by post-depositional processes, it is difficult to reconstruct much of the surrounding environment or of the economic activities in which the users of these materials would have engaged, but the site is important nonetheless because it joins a relatively small number of non-funerary sites with Beakers in the region. If isolated finds of a few sherds are disregarded, the list of Beaker non-funerary finds in Wiltshire is a short one. If low density surface scatters such as are known from the Stonehenge area and parts of the Marlborough Downs are also excluded the list barely reaches double figures. This may be in part a reflection of lack of excavation in those areas which do have low density scatters and of the destruction of material in the ploughsoil, but despite the considerable amount of archaeological monitoring of ground disturbing works over large areas of the county, the number of finds remains low. It seems likely then that this represents a real state of affairs; that is, that occupation did not occur intensively in limited areas but was generally transitory and did not routinely include the digging and filling of pits.

80 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Only a few larger sites are known, those in southern and central Wiltshire comprising Easton Down (Stone 1930/32), 17 km east of Crescent Copse, Downton (Rahtz 1962), 25 km to the south and Robin Hood’s Ball, less than 1 km to the north of the site. In each of these cases there are circumstances which may have made the location particularly attractive for settlement: in the case of Downton, just off the chalk, it was probably the proximity to the resources of a river; at Easton Down it was presumably the availability of flint, while Robin Hood’s Ball seems to fit a pattern seen elsewhere, of causewayed enclosures being later a focus for settlements of more than usual duration or size - compared to the majority of sites in this region (as at Windmill Hill, in the north of the county: Smith 1965). The site just outside Robin Hood’s Ball is as yet unpublished but is noted as Group D by Case (1995a, 4).

It may be that the proximity of Robin Hood’s Ball and the Beaker period occupation outside it had a direct influence on the nature of the Crescent Copse occupation. As noted, the occurrence of a group of pits must be regarded as unusual and, as discussed above, there are features of the ceramic assemblage which are certainly rare. The struck flint, however, shows no special features, nor are there any obvious reasons to suggest that the deposit was a carefully selected one. It is difficult to conclude other than that this was a fairly short occupation which for some reason was marked, perhaps on completion, by burial of some of the occupation debris. In that alone it is still a welcome addition to the small corpus of settlement evidence from this still enigmatic period of Wessex prehistory, for which such evidence remains so perplexingly rare.

FUNGAL ACTIVITY IN ARCHAEOLOGICAL DEPOSITS

Neither the poor palaeoenvironmental assemblages nor the presence of fungal mycelia aroused much interest until the actions of arboreal fungi on adjacent organic materials were explained during a lecture attended by the excavator as part of his continuing studies in building conservation. Prior to that, the excavator had assumed, perhaps glibly, that fungi attacked organic remains immediately after insertion into, or removal from, the burial environment; the possibility that fungi could be introduced to a burial environment had never been considered. Informal consultation with mycologists, specifically Alan

Rayner (President of the British Mycological Society), who specialises in the association of trees and fungi, suggested that the results of the excavation warranted consideration from a different viewpoint. Whilst it is recognised that more detailed examination of the fungi on-site would have been desirable, two observations are pertinent to the conclusions of this report.

Firstly, organic materials would be anticipated in this calcareous burial environment. Protected from the overlying argillaceous soils, the bone content of chalkland archaeological features invariably survives in excellent condition as do mollusc shells and carbonised plant remains - it is a characteristic of chalkland archaeology. Furthermore, these features had once contained organic materials; fine charcoal dust was evident throughout, and flotation separation recovered much plant and animal material identifiable to type, but not to species, all of it in very poor condition. Whilst charcoal is ubiquitous in archaeological deposits, it invariably survives as identifiable fragments, not just dust. Pre-depositional attrition may have been responsible, but the condition of neither the pottery nor the flint support this, and the incorporation of some un-eroded material would be expected; the fact that none was, suggests that an explanation has to be sought for the apparent destruction of this material in situ.

Secondly, the fungal mycelia were active; they were white and ‘fluffy’, and were extending westwards from the direction of the adjacent pine plantation into the pits, not out from them. Only the pit fills proper were permeated by them, not the overlying A/C horizon or the compacted topsoil above, or the fills of any of the other features investigated. There can be little doubt, therefore, that the fungi were a recent arrival and had selectively exploited these deposits.

Fungi are an active component of all ecosystems, essential for the propagation and decay of all life (Dommergues 1977, Rayner 1995). Whilst each ecosystem has its own characteristic fungal population, the spores of thousands of other species are also ubiquitous in air, soil and water as dormant organisms that can be activated on encounter with suitable hosts, including plants (Gwynne-Vaughan and Barnes 1951).Trees - particularly pine trees - are mutually dependent upon a wide range of fungi, the mycelia of which can extend beneath the ground over hectares in search of organic carbon, water and minerals (Rayner 1998). Indeed, some species of woodland fungi in Northern America constitute the largest single living organisms on the planet. Alterations in hygrography, plant succession or land use, can create major imbalances in the fungal

BEAKER PITS AT CRESCENT COPSE NEAR SHREWTON

population, leading to rapid changes in cycles of growth and decay that may have been stable for millennia. Coniferous afforestation schemes are particularly pernicious in this respect (Dommergues 1977) because breakdown of the leaf litter, unpalatable to other life forms, requires and favours a wider range of fungi than does that of deciduous forest, many of which supplement their diet by sub-surface mycelial ‘foraging’ (Rayner 1995). The introduction of coniferous trees to a chalkland landscape will, therefore, also introduce a range of otherwise alien fungi to a relatively nutrient-poor environment, that will then seek out additional sources of nutrients in the vicinity. This much, at least, is scientifically accepted fact.

The mycelia observed in these pits bore all the characteristics of woodland species and were clearly radiating from the direction of Crescent Copse. What cannot be proven - post-excavation - is whether they actually emanated from the pine trees and whether they alone were responsible for destruction of the organic remains in these pits, though the excavator believes this is the case. Given the serious implications, for archaeological deposit survival, of these assumptions being correct, and in the absence of contradictory experimental data, the excavator believes that this conclusion should be accepted until proved wrong by experiment. The ramifications for the environmental management of Salisbury Plain - and all archaeologically sensitive areas - are serious; afforestation of hitherto stable un-wooded landscapes is likely to encourage the outward spread of archaeologically destructive fungi, well beyond the immediate area of afforestation.

Note

1. The identification of the bone in both the Butterfield Beakers and the Crescent Copse Beaker was in the first instance by the author, but was subsequently confirmed by Jacqueline McKinley, a human bone specialist. In neither case was it possible to establish whether the bone was human or animal because of its fragmentary nature.

References

CASE, H.J., 1977, ‘The Beaker culture in Britain and Ireland’, in R.J. Mercer (ed.), Beakers in Britain and Europe, 71-101. Oxford: British Archaeological Reports

CASE, H.J., 1993, ‘Beakers: Deconstruction and After’, Proceedings of the Prehistoric Society 59, 241-68

81

CASE, H.J., 1995a, ‘Some Wiltshire Beakers and their contexts’, WAM 88, 2-17

CASE, H.J., 1995b, ‘Beakers: loosening a stereotype’, in I. Kinnes and G. Varndell (eds) ‘Unbaked urns of rudely shape’. Essays on British and Irish pottery for Ian Longworth, 55-67. Oxford: Oxbow Books

CLARKE, D.L., 1970, Beaker pottery of Great Britain and Ireland. Cambridge: Cambridge University Press

CLEAL, R.M.J., 1995, ‘Earlier prehistoric pottery’, in R.M_J. Cleal, K.E.Walker and R. Montague, Stonehenge in its landscape. Twentieth-century excavations, 349- 367. London: English Heritage

CLEAL, R.M.J., 1996, ‘The pottery from pit 2’, in M. Rawlings and A. P. Fitzpatrick, ‘Prehistoric Sites and a Romano-British Settlement at Butterfield Down, Amesbury’, WAM 89, 1-43

DOMMERGUES,.Y.R., 1977, ‘The Plant-Micro-organism system’, in Dommergues and Krupa (eds), Interactions between non-pathogenic soil micro-organisms and plants. Amsterdam: Elsevier

GIBSON, A.M., 1982, Beaker Domestic Sites. Oxford: British Archaeological Reports 107

GRAY, H. St G., 1935, “The Avebury Excavations, 1908- 1922’, Archaeologia 84, 99-162

GWYNNE-VAUGHAN, H.C.I., and BARNES, B., 1951, The Structure and Development of The Fungi. Cambridge

KINNES, I., GIBSON, A., AMBERS, J., BOWMAN, S., LEESE, M. and BOAST, R., 1991, ‘Radiocarbon dating and British Beakers: the British Museum programme’, Scottish Archaeological Review 8, 35 - 68

RAHTZ, P.A., 1962, ‘Neolithic and Beaker sites at Downton, near Salisbury, Wiltshire’, WAM 58, 116-41

RAYNER, A.D.M., 1995, ‘Fungi, a vital component of Ecosystem Function in Woodland’, in Allsop, Colwell and Hawksworth (eds), Microbial Diversity and Ecosystem Function. Oxford: CAB International

RAYNER, A.D.M., 1998, ‘Fountains of the forest- the interconnectedness between trees and fungi’, Mycological Research 102 (12), 1441-1449

SMITH, LF., 1965, Windmill Hill and Avebury. Excavations by Alexander Keiller 1925-1939. Oxford: Clarendon Press

SMITH, I.F. and DARVILL, T., 1990, ‘The prehistoric pottery’, in A. Saville, Hazleton North. The excavation ofa Neolithic long cairn of the Cotswold-Severn group, 141 - 152. London: English Heritage Archaeological Report No. 13

STONE, J.E.S., 1930/32, ‘A settlement site of the Beaker period on Easton Down, Winterslow, S. Wilts.’, WAM 45, 366-72

THOMAS, N., 1964, ‘The Neolithic causewayed camp at Robin Hood’s Ball, Shrewton’, WAM 59, 1-27

WEBSTER, J., 1970, Introduction to Fungi Cambridge

WESSEX ARCHAEOLOGY, 1996, Track 216 Crescent Copse SPTA; archaeological evaluation. Unpublished client report

82 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE Ce) Manors held by Thomas Tropenell

Corsham according to his IPM & Church

Castle

y% Great © re] Chalfield Manor House

m Ilrowbridge

e Great Cheverell

West Codford

0 E © Maiden Bradley : ee

@ Chicklade Little © Hindon © Durnford

SALISBURY ©

East Harnham

miles 5) (ee ——— eee ———— aed

Figure 1. Principal interests of Thomas Tropenell as given in the text

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 82-89

A ‘Perillous, Covetous Man’: the career of Thomas Tropenell Esq. (c. 1405-88), a Wiltshire lawyer, Parliamentary burgess and builder of Great Chalfield!

by J. T: Driver

During the course of a very long life Thomas Tropenell spent much of his energy establishing his rights to various manors, especially that of Great Chalfield, three miles north of Trowbridge, where his fine manor-house still stands. His legal training stood him in good stead when It came to estate litigation, but an equally important factor was his long connection with the leading family of Hungerford, staunch backers of the Lancastrian government. Tropenell took his part in the usual tasks 1n local government. He was elected twice to the Commons, first in 1429 and again in 1449, and served as a Justice of the Peace.

Thomas Tropenell of Great Chalfield, Neston, Salisbury and elsewhere, was born c.1405, the son of Henry Tropenell and his wife, Edith, the daughter of Walter Roche. Through the marriage of his forebear, Walter Tropenell, to Katherine Percy he could trace his ancestry back directly to the Percy family who had held Great Chalfield in the thirteenth century.’ Tropenell married twice. By 1431 he had wed Agnes, widow of Thomas Burton of Burton in Gloucestershire, and Atworth in Wiltshire. By 1456 he had married Margaret, widow of John Erley, who had been elected to the Commons for Ludgershall in 1450. She was the daughter of William Ludlow of Hill Deverill, a royal official under the Lancastrians and parliamentary burgess for Ludgershall in 1432, 1433, 1435, 1453 and possibly in 1455, and for Salisbury in 1439.’

During his life Thomas Tropenell was to build up a modest estate as a Wiltshire squire, take some part in local administration as befitted his status as a gentleman and lawyer, and, perhaps most important of all, to build the manor-house at Great Chalfield. Initially, however, his public and professional career should be examined, and his contacts with contemporary notables discussed.

Tropenell must have been in his mid-twenties when elected as second burgess for Bedwin to the parliament which met at Westminster on 22nd

September 1429.* This would seem to mark the beginning of his career. For the next few years he appears to have taken no other part in local affairs, but his presence among the Wiltshire electors in December 1436 is an indication of his rising status.’ Indeed, under the statute passed in the parliament of 1429-30 the county franchise was limited to the 40s. freeholder. Tropenell attended the elections at Bath in January 1437 and ten years later again took part in the shire elections.° Furthermore, his name was put down as one of the manucaptors (i.e. sureties for appearance in parliament) for John Fortescue, one of the parliamentary knights for Wiltshire elected in December 1436.’

By this stage in his career Tropenell was already associating with several south-western gentry, some of whom, as himself, were both lawyers and clients or servants of the important Hungerford family. An early example of such involvement came in March 1438 when he, Henry Long, Richard Chokke, John Fortescue and John Whittokesmede, all lawyers, were feoffees or trustees for Sir Walter Hungerford.* A few years later Tropenell was admitted to Lincoln’s Inn in Michaelmas term 1445, being granted ‘special admission’ to that Society in 1470.” Meantime he had been elected to the second seat for Bath in the parliament which met at Westminster on 12th February 1449.'° In September 1450 he was a

46 Crossley Crescent, Hoole, Chester CH2 3EZ

84 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

member of a commission, headed by Sir Robert Hungerford and Henry Long, to summon the king’s lieges in Wiltshire; and in the following year he was among those ordered to enquire into two cases of robbery in Salisbury and Swindon.'! For some eight years or so Tropenell seems to have been absent from official commissions. The 1450s were a time of growing political tension between the supporters of the increasingly ineffectual Lancastrian king, Henry VI, and their opponents led by Richard, Duke of York. Among those who followed the latter were the Herberts. It is of significance, therefore, that Tropenell, as one having close attachments to the Lancastrian Hungerfords, should have been included in a body to arrest the Yorkist squire Thomas Herbert in Somerset in June 1459.""

In the same month, he and others were appointed to ascertain what lands Sir Robert, Lord Hungerford had held.'? Once again there is a lack of evidence about Tropenell’s public activities, except for service as bailiff of Salisbury for the year 1462-3, and as alderman of the city in 1461—2 and 1479-80.'" He was probably occupied with his own matters regarding property, and with building Great Chalfield. Another possible reason for his lack of involvement in more public affairs could have been political, in that the 1460s were essentially years of Yorkist ascendancy, when people such as Tropenell with ‘Lancastrian’ affiliations were excluded from positions of local authority. It is particularly interesting that, despite Tropenell’s Lancastrian connections, he was able to enlist the support of Edward, Earl of March (later Edward IV) in recovering Hindon and Chicklade manors in 1460.'" What should also be noted, however, is that he sued for a number of official pardons during these years: in July 1459, for breaking one of Henry VI’s statutes; and in 1464, 1468 and 1471.'° On the last occasion he was, in fact, described as ‘of London, gentleman’.

However, in October 1470 his name was included with Henry Long and John Whittokesmede in an investigation into felonies, murders and homicides committed in Wiltshire;!’ and on 18 November he was made a Justice of the Peace in the county, though he was not nominated to the quorum. In the event Tropenell was only to be a member of the bench for a few months — until 20 June 1471.'* The most likely explanation for such a short tenure of office was that his appointment had been made under the Readeption parliament of Henry VI, and that he had been removed by the restored Edward IV soon after his final victory at Tewkesbury on 4 May 1471. However, his appointment to the Wiltshire bench by the Readeption

government, eight days before the meeting of parliament on 26 November 1470 could possibly give some support to the claim that he sat in that parliament. After his nomination in August 1473 as one of the commissioners to look into the neglect of the sheriff of Wiltshire to return royal farms to the Exchequer, Tropenell does not seem to have figured on any more official government bodies.'”

Central to his career and the building-up of his estate were his connections with the Hungerford family, which lasted throughout much of his life. As demonstrated already, his links with the Hungerfords were established by 1438.That he was clearly a trusted servant is indicated by a reference in the account of John Mervyn, receiver-general for 1447-48, that on 23 November 1447 Tropenell had paid to Lord Moleyns (i.e. Robert Lord Hungerford) £16 13s. 4d. by hand.”° By 1448 he had been given the important post of receiver-general for Sir Walter, first Lord Hungerford, and after the latter’s death in the following year he retained the office under Sir Robert, second Lord Hungerford, until 1458.7! In 1449 he was one of anumber of Hungerford feoffees, including Richard Chokke, John Fortescue and John Whittokesmede, in the manors of Imber and Winterbourne Homington [Homanton in Maddington] in Wiltshire and of Folke in Dorset.” Six years later, in February 1455, he was one of the witnesses along with Sir Edmund Hungerford, a younger brother of Sir Robert, to a charter concerning Little Chalfield manor.’ It was, however, with Sir Robert, second Lord Hungerford, his wife, Margaret Lady Botreaux, and their son, Robert Lord Hungerford and Moleyns, that his links were closest. Although the significance should not be stressed too much, it is important to note that Tropenell seems to have had a central role in the Hungerford ‘circle’ for many years from the 1440s, and especially from the 1450s when the family was hard-pressed to meet the heavy ransom of Robert, third Lord Hungerford, who had been captured in 1453." A clear mark that Tropenell stood high in Hungerford favour was that Robert ‘the elder’ bequeathed him a silver cup engraved with the family arms and that Lady Margaret made him her steward, executor and feoffee.”

The connection enjoyed by Tropenell with the Hungerfords gave him wide contacts with local notables, gentry and lawyers. On 31 March 1460, for instance, his name was listed among a large panel of feoffees for Sir Robert Hungerford ‘the younger’ in the manors of Chippenham, Warminster, Imber and elsewhere; and in another document, dated the same day, he was given as a witness with John

A‘PERILLOUS, COVETOUS MAN’: THOMAS TROPENELL (c. 1405-88) 85

Mompesson and others to a grant of manors to Sir Robert and his mother, Lady Margaret.”° When, in September and October 1463, Lady Margaret established trusts for her lands in Cornwall and Wiltshire, Tropenell was among an impressive list of feoffees, some of whom were of more than local standing, in particular George Neville, bishop of Exeter, Chancellor of England, Richard Beauchamp, Bishop of Salisbury, and the important knight Roger Tocotes, who had married the widow of Bishop Beauchamp’s brother. A further member of this panel of trustees was Gregory (or George) Westby, quondam receiver-general for Lady Margaret Hungerford.*’

In October 1467 Tropenell and Tocotes were named as principal witnesses to a grant of Little Cheverell by Lady Margaret to John Mompesson; in June 1470 he and the substantial squire and royal servant, Avery (or Alfred) Cornburgh, witnessed a charter of Walter Hungerford, son of Robert, third Lord Hungerford. In January 1471 he was listed as a Hungerford feoffee with the south-western justice Sir Richard Chokke.** His links with the Hungerfords continued until the end of his life for, in April 1486, Tropenell was mentioned as one of the patrons of Mildenhall in Wiltshire, along with Mompesson and Mervyn, all feoffees of the late Robert, Lord Hungerford and Homet.”’

Such an intimate and lengthy involvement with the staunchly Lancastrian Hungerford family would suggest a similar alignment of Thomas Tropenell, too. Yet there is some evidence to suggest that, politically speaking, he managed to tread a careful path. Indeed he was described as a squire, ‘which had the lyvereys of Kyng Harry the vj and of Kyng Edward the iiij"’.*° Furthermore, as ‘Thomas Tropenell of Neston, gentleman’ , he had been included in a general pardon issued under Edward IV in 1464, and was to receive another pardon under Richard III.*!

It is, however, not possible to prove how politically committed Tropenell was during this difficult and disturbed period. If his intimate attachments lay with the Hungerfords, they could well have been essentially personal and professional. Certainly his involvement with the family could have proved politically embarrassing in the 1460s, when Robert, Lord Hungerford and Moleyns, and his son Sir Thomas both suffered execution for rebellion against Edward IV. What does seem possible is that Thomas Tropenell was astute enough to concentrate upon what were his own essential interests — that of taking a modest part in the accepted round of local office-holding, as was appropriate to his status as a member of the

county squirearchy, including occasional service in parliament, and of building up his estates in Wiltshire. The former aspect of his career has already been considered. It is time now to focus on the latter, beginning with his struggle to establish his claim to the manor of Great Chalfield.

In the thirteenth century, possibly even in the late twelfth century, this had belonged to the Percy family. Tropenell’s claim was based upon the marriage of Walter Tropenell, his direct ancestor, to Katherine, daughter of Sir William Percy.*? However, at some point between 1437 and 1443 a dispute arose when William Rous argued that, at Tropenell’s request, he (Rous) had enfeoffed him (Tropenell) and his associates Richard Chokke and Henry Long in the manor of Great Chalfield.*? Rous later claimed that Tropenell had not fulfilled the trust’s condition and proceeded to law. However, when the case came up Rous failed to appear and so Tropenell was left in possession. Nevertheless, other legal actions were to follow before Tropenell finally secured possession. In 1444—45 another protagonist appeared in the person of Thomas Beverley, who unsuccessfully sued Tropenell and his co-feoffees, Long and Chokke, for the estate.

Two years later, in 1447, Tropenell and his colleagues conveyed Great Chalfield back to Rous, who then settled it upon his second wife, Isabel, and their issue. In spite of that arrangement Rous soon afterwards leased Great Chalfield to Tropenell, who claimed to be the tenant when Rous died in 1452. Isabel, however, contested the matter and actually entered the manor, only to be ejected by Tropenell. In 1454 she surrendered her claim for 100s. and in 1459 Tropenell finally settled with her by a composition of £53. Unfortunately, he still had others with whom he had to contend before he achieved complete possession. In 1454 Thomas Beverley renewed a claim, but then released his rights to Tropenell; and in 1459 he tried once more to acquire the manor, but Tropenell produced documents to retain his right. Moreover, at the same time, Tropenell faced another challenge to Great Chalfield from an aunt of William Rous, Joan Beaushyn, who, in the event, relinquished her interest. At last, in 1467 Tropenell finally secured the estate, but only after yet another lawsuit brought against him by Beverley. The latter, in fact, won his case, but then sold his rights to Tropenell.** His persistence in fighting for the estate could partly have owed something to his desire to become a man of substance, but most likely to his determination to secure a property which he believed was rightly his through inheritance from Walter

86 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Tropenell and ultimately from the Percy family. Although it is not possible, at this stage, to establish the costs of his lawsuit in his pursuit of his claim to Great Chalfield, they must have been quite substantial. The cost of the final composition with Beverley was no small amount in fifteenth-century terms. Thomas Tropenell’s determination to secure his rights to Great Chalfield with the accompanying office of the constableship of Trowbridge is revealed in the comment that, “Tho. Tropenell, in 1) the last yeres of the said Kyng Harry the sixte, oft tymes sued to the counseil of the seid duchie of Lancastre, for the seid offices [i.e. the constableship of Trowbridge], shewyng his evidence and title to theym thereof, and prayed to be amytted therto.’”

In addition to Great Chalfield, Tropenell acquired other manors and properties. One was Fisherton Anger near Salisbury, which consisted of two small estates, obtained in 1457 and 1465.*° Another was Little Durnford which he apparently purchased of John Wodehull esquire and others in 1474—75.*” At the time of Tropenell’s death the manor was valued at 100s.** His tenacity in pursuing his claims suggests that he was a determined hard man of business, a characteristic borne out by his acquisition of Hindon manor. Tropenell had apparently acquired it in 1452, when Robert Hardell mortgaged it to him. Four years later he occupied Hindon, together with lands in Milton, East Knoyle, and'Tollard Royal, for which he paid £40 to Robert Hardell and his wife.*” Included in the transaction were valuable assets such as 19 messuages or dwellings, a dovecote, a bull and 700 sheep, all of which suggests that Tropenell could have been farming on quite a large scale. Indeed, though firm evidence is lacking, he could have been involved in the wool-trade. At his death Hindon was said to have been held of the Bishop of Winchester.*" A few years earlier, in 1451, Tropenell bought the Bell Inn at Chippenham, for which he paid 22 marks (£14 13s. 4d.) to William Jerveys and his wife, Isabelle, and 10 marks (£6 13s. 4d.) to Hugh London (or Bushyn).*! Tropenell also obtained the manors of Maiden Bradley and Chicklade which, with Hindon, were valued at his death at twenty marks a year. In addition to these estates Tropenell held the manor of East Harnham, three messuages and land in West Codford, worth 30s., and a messuage and land in East Codford, worth 10s.”

It should be noted that the Hungerford family were involved in Tropenell’s acquisition of Codford, together with the advowson of Great Cheverell. A confirmation and release of Codford and the advowson in 1469 referred to an earlier grant by

Robert, second Lord Hungerford, on 9th December 1447, to Thomas Tropenell and his wife, Agnes. A further deed of May 1465 recorded a grant of parcels of land, rents and two cottages in West Codford [Codford St Peter] by Margaret, Lady Hungerford and Botreaux, to Thomas Tropenell ‘esquire’, for the annual payment of a red rose at the Nativity of St John the Baptist (24th June).™

By c.1465—67 Tropenell had succeeded in building up an estate of several manors and properties in Wiltshire which must have brought a steady income. Without the survival of a rental, however, it is not possible to estimate the size of his income and expenditure. Nevertheless, although his total holdings were not extensive they must have provided much of the resources needed to re-build his imposing manor house at Great Chalfield and the Tropenell chapel in the church of All Saints. It would be extremely interesting to know the cost of the building projects at Great Chalfield. Unfortunately, however, no building contracts or accounts have so far come to light, though some have survived for properties elsewhere in the fifteenth century. The architectural features of Great Chalfield and the chapel have already been described by various authorities.” Here it is appropriate only to draw attention to some of the principal features, taking first the chapel, which dates from c. 1480. It has a panelled wagon roof and a stone screen which includes five shields describing heraldically the principal families connected with the manor. These include the arms of Tropenell impaling Percy; Tropenell (Gules, a fesse engrailed argent powdered ermine, between three griffons’ [1.e. griffins’] heads of the same erased); and Tropenell impaling Ludlow — the last referring to his second wife, Margaret, daughter of William Ludlow.’° Indeed, similar heraldic devices relating to the Tropenell and Ludlow families can be seen in various parts of the house, as well as on the tomb of Thomas and Margaret at Corsham.”’ In the house itself Tropenell arms are displayed on the bosses in the bays at the east end of the hall and were once on wooden bosses there. Still to be seen on the original timbers are his painted motto Le jong tyra belement (‘The yoke pulls well’) and his badge of a double ox yoke. By such visual means did Thomas Tropenell esquire announce his arrival into the ranks of the gentry.

However, more important perhaps, is the high quality and beauty of the architecture itself at Great Chalfield. In particular there are the two fine oriel windows on the north front, the well proportioned hall, whose ceiling has moulded beams, and the very arresting three spy-windows in the form of masks of

A‘PERILLOUS, COVETOUS MAN’: THOMAS TROPENELL (c. 1405-88) 87

a bishop, a devil and another face. Perhaps even more significant still is that many of the windows have uncusped lights, which were rather ahead of their time. Again this underlines the unusual quality of a building erected by aman whose estate was by no means large and whose social rank did not rise above that of ‘esquire’. This prompts the question as to how he secured the services of masons and other craftsmen of such high calibre. Perhaps his connections with the Hungerfords, whose seat was not far away at Farleigh Hungerford, enabled him to obtain highly- skilled men to build and adorn his splendid new house at Great Chalfield? Although Tropenell enjoyed a close relationship with the Hungerfords, the part they played in his grand project at Great Chalfield must essentially have been confined to providing useful contacts rather than financial support, since they themselves were going through a time of financial embarrassment in the late fifteenth-century.”” Again, Tropenell could well have taken some ideas from the manor house at nearby South Wraxall, which belonged to the Long family. That there were such skilled craftsmen locally is given added support by the magnificent south chapel of St Nicholas’ church, Bromham, which was built some four or five years after Tropenell’s death by his former associates, Sir Richard Beauchamp and Sir Roger Tocotes.

In August 1486, possibly conscious of his advancing years, Tropenell made a grant of an annual rent from his lands in Chicklade to his son, Christopher, and his heirs.*° Just over a year later, on 28 October 1487, his name appeared as one of the patrons, as a co-feoffee of the late Robert, Lord Hungerford and Homet, in the institution of Henry Mompesson to the living of Mildenhall in Wiltshire.”! Some eight days later, on 5 November, he drew up his will, probate of which was eventually granted on 26 February 1488. A fortnight or so after making his will the writ diem clausit extremum was issued on 22 November which ordered an enquiry as to his lands.” The resulting inquisition, held at Wilton on 3 May 1488, listed his principal holdings as the manors of Great Chalfield, Maiden Bradley, Chicklade, Hindon, Little Durnford and East Harnham, with properties and lands elsewhere. To ensure that the provisions of his will were carried out Tropenell enfeoffed Sir Richard Beauchamp and a judge, Sir John Catesby. His heir was stated to be his son, Christopher, aged twenty-five or more. The jurors further stated that Tropenell had died on the last day of January last (i.e.1488).** In his will he asked to be buried in the south chapel of Corsham church in the tomb made for himself and his second wife, Margaret. Among

the several monetary bequests were 6s.8d.each to Salisbury Cathedral and Corsham church; 26s.8d. to the rector of Edington; and 13s.4d. to the Prior and community of Bath Abbey. More revealing, perhaps, as an indication of his personal wealth, taste and religious devotion, was his gift to Robert Cheverell of two silver candlebra and a missal. Unsurprisingly, Tropenell did not forget his long-lasting ties with the Hungerfords in his provision for a priest to pray for the souls of Robert, late Lord Hungerford, his wife Lady Margaret, and their grandson, Sir Thomas.” His charity even extended to a grant of thirty sheep to the church of Codford St Mary.

Thomas Tropenell esquire and lawyer appears to have spent most of his life in the south-west, especially in Wiltshire. In fact he does not seem to have cut much of a public figure on a wider stage — though he was certainly elected twice to the Commons in his earlier years and held most of the usual offices appropriate to that of a country gentleman, except that of sheriff. The real significance and importance of his career is threefold: first his long association with the Lancastrian Hungerfords; second his building of the fine manor house at Great Chalfield; and third the unusually full and rare collection of deeds relating to his properties, which he began to gather together in 1464, and which were published by the Wiltshire Archaeological and Natural History Society in 1908 as The Tropenell Cartulary. This collection of deeds, many of which go back to the early fourteenth century, reflect clearly the character and determination of this Wiltshire lawyer to assert and retain his rights. Indeed, they seem to lend substance to the pejorative comment made of him that he was a ‘perillous covetous man’.

Perhaps most fascinating of all is the mural painting in the dining room at Great Chalfield which could well be that of Thomas Tropenell himself. The subject depicted is of a man of mature age, substance and authority. He is shown with five fingers on each hand. Should this portrait indeed be that of Tropenell, then it could be the earliest known painting of a member of the gentry and a parliamentary burgess.”°

Notes

1 I should like to thank Professor C.T. Allmand and Dr J. H. Thomas for reading an earlier draft of this article and for their constructive suggestions, and my wife for preparing the typescript.

2 J. Silvester Davies (ed.), Trop(enell)Cart(ulary), 1908 (WANHS, Devizes), vol. 1, pp. x-x1; facing p. 272 (geneal. tables); 272-88; vol. 2, p. 163. Tropenell was described

88 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

as ‘gentleman and merchant’ [of Salisbury] in 1452, Ibid. vol. 2, p. 28; ‘of Neston in the County of Wiltshire, gentleman’, Select Cases on the Law Merchant, Selden Society (1929), vol. 2, 20); and as armiger [esquire] in 1458, Trop.Cart. vol. 1, p. 369.

3 Ibid. vol. 1, pp. 122-23; vol. 2, pp. 163-64; J. C. Wedgwood and Anne D.Holt, History of Parliament: Biographies of the Members of the Commons House, 1439-1509, 1936, H.M.S.O., p. 875. For the career of William Ludlow, see Ibid:, p:561.

4 PRO €219/13/6, pt.1, no. 91. Official Return of Members of Parliament (The Blue Book), Pt.l, 1213-1702, 1878, p. 317. Tropenell took second place to Richard Bridges of Salisbury, who was probably a lawyer. No date was given for the borough election.

5 PRO €C219/15/I, pt. 3, no.105. The elections were held at Wilton for the parliament which met at Westminster on 21st Jan.1437.

6 PRO C219/15/), pt. 3, no.105; C219/15/5, no. 14.

7 PRO C219/15/), pt. 3, no.105.

8 Calendars of) P(atent) R(olls), 1436-42, p.152; J. L. Karby (ed.), Abstracts of fines relating to Wiltshire, 1377- 1509, 1986, (Wiltshire Record Society, vol. 41), no. 511.

9 W. B. Baildon (ed.), Admissions Book of Lincoln’s Inn, 1420-1893, 1896, vol. 1, pp. 10, 18. ‘Special admission’ generally implied one or more of three privileges: exemption from keeping vacations, exemption from holding certain offices, and the right to sit at the Masters’ (i.e. Benchers’) Commons. In a letter written by Sir Miles Stapleton to his ‘Right welbeloved and trusty freende’, Thomas Tropenell, he refers to the receipt of 100s. via Tropenell’s clerk, Robert, which could suggest that, as a lawyer, Tropenell had a small staff (Trop. Cart. vol. 2, p. 93). The money was for three closes in Codford which ‘Tropenell wanted to have.

10 PRO C219/15/6, no. 88.

11 C.PR., 1446-52, pp. 434, 443.

12 Ibid. 1452-61, p.518.

13 Ibid. p. 496. Robert, second Lord Hungerford, died early in the summer of 1459. For his career, see G.E.Cokayne et al (ed.), The Complete Peerage, 2nd edn., vol. 6, pp. 617-18.

14 WRO G23/1/2, ff. 58v.,55v.,135v.

15 Trop.Cart. vol. 2, pp. 63-65.

16 C.PR, 1452-61, p. 509. Wedgwood and Holt, (op. cit., pp. 875-6) suggested that the description ‘of London’ used of Tropenell in 1471 could have pointed to his election to the Readeption Parliament of 1470-71, but more substantive proof would seem needed to assert that he was an M.P. in this body. However ‘of London’ could more plausibly suggest that’Tropenell had a house there.

17 “CPR: 1467-77, p-247,

18 Ibid. p. 635. It should be noted that his associates, Henry Long and John Whittokesmede, were made ‘of the quorum’: PRO C66/491, mm. 25d, 26d.

19 C.PR., 1467-77, p. 406.

20 PRO SC6/1119/12.

21 J. L. Kirby, “The Hungerford Family in the Later Middle Ages’, unpublished London M.A. thesis (1939), p.112. 1

am indebted to Mr Kirby for allowing me to cite from his work. ‘Two accounts of Tropenell’s immediate predecessor, John Mervyn, have survived for the financial years 1444— 45 and 1447-48 (PRO SC6/1119/ 11, 12). The next surviving roll (SC6/1119/ 13) is without a heading. Since ‘Tropenell took over from Mervyn in 1448, this last roll must have belonged to the period of Tropenell’s time as receiver-general.

22 C(alendar of) C(lose) R(olls), 1447-54, pp. 147, 148. 23 C.C.R., 1454-61, p. 55. 24 For Robert, third Lord Hungerford, see Complete

Peerage, vol. 6, p. 618. After his release from French captivity he continued to support the Lancastrian cause, for which he was to be attainted in 1461 and executed after the battle of Hexham in 1464.

25 Wedgwood and Holt, op. cit., p. 875; F. W. Weaver (ed.),

Somerset Medieval Wills, 1901 (Somerset Record Society, vol. 16), pt. 1, p. 189; N. H. Nicolas (ed.), Testamenta Vetusta, 1826, vol. 1, pp. 311-20.

26 C.C.R., 1454-61, pp. 439, 441. Mompesson sat for

Wilton in the parliament of 1453-54. He, too, was a lawyer with Hungerford connections. He was appointed, as was Tropenell, to the Wiltshire bench on 17th Dec.1470 by the Readeption government, but removed when the Yorkists returned to power: Wedgwood and Holt, op. cit. pp. 599-600

27 C.C.R., 1461-68, p. 271; J. L. Kirby (ed.), The

Hungerford Cartulary: a calendar of the Earl of Radnor’s Cartulary of the Hungerford Family, 1994 (Wiltshire Record Society, vol. 49), no. 666. Sir Roger Tocotes of Bromham married Elizabeth, widow of Sir William Beauchamp, Lord St Amand. Tocotes was returned for Wiltshire in 1467 and 1472-75. Towards the end of his life he and his stepson founded the splendid chantry at Bromham: Wedgwood and Holt, op. cit., pp. 858-9; J. T. Driver, “The Parliament of 1472-75, with particular reference to the personnel of the Commons’, unpublished Liverpool Ph.D. thesis (1982), vol. 2, pt. 3, pp. 949-59. For Westby as receiver-general, see Kirby “The Hungerford Family’, p. 114.’Two accounts survive from his period as receiver-general, one for 4-5 Edward IV (1464-65), and the other for 14-15 Edward IV (1474-75): PRO SC6/1119/ 14, 16.

28 C.C.R., 1468-76, pp. 65, 142, 171. For Cornburgh,

who served the governments of Henry VI, Edward IV, Richard III, and Henry VI, see Wedgwood and Holt, op. cit., pp. 223-24.

29 Kirby, Hungerford Cartulary, no. 690; D. P.Wright (ed.),

The Register of Thomas Langton, Bishop of Salisbury, 1485-93, 1985 (Canterbury and York Society, vol. 74), no. 48.

30 Trop.Cart. vol. 2, p. 163. 31 C.PR., 1461-67, p. 339; British Library, Harl. MS. 433,

f. 83b , see now R. Horrox and P. W. Hammond (ed.), BL.Harleian Manuscript 433, 1979 (Richard III Society), vol. 1, p. 230. The entry reads: “Thomas Tropenelle of Chaldefeld in the Countie of Wiltshire Squier hathe a generale pardone’.

32 Trop. Cart., vol. 1, introduction, p. x1.

A‘PERILLOUS, COVETOUS MAN’: THOMAS TROPENELL (c. 1405-88) 89

33 For this outline of Tropenell’s efforts to recover Great Chalfield, the account in V.C.H. Wilts., vol. 7, 1953, p. 61 has been used. See also the evidence in Trop, Cart., vol. 1, pp. 272-96.

34 Ibid. vol.1, pp. 396-98.

35 Ibid. vol. 1, p. 295.

36 V.C.H.Wilts., vol. 6, 1962, p. 186; Trop.Cart., vol. 1, pp. 152, 153, 166, 170.

37 C.C.R., 1468-76, pp. 348, 347, 63-64; V.C.H. Wilts., vol. 15, 1995, p. 84; Trop. Cart., vol. 2, pp. 286-90, 298- 99; R. Colt Hoare et al, The History of Modern Wiltshire: Hundreds of Everley, Ambresbury and Underditch, 1826, pp. 126-28.

38 Cal(endar of) I(nquisitions) P(ost) M(ortem): Henry VIL, vol. 1, 1898, no.351.

39 Kirby (ed.), Wiltshire Fines, no. 628.

40 V.C.H. Wilts., vol. 11, 1980, p. 99, n.13; Trop. Cart., vol. 2, pp. 17-18, 27-28; Kirby (ed.), Wiltshire Fines, no. 628. In an obligation, taken by Hardell before the mayor of Salisbury in Oct.1452, to secure Tropenell’s rights to lands and properties in Chicklade and Hindon, the latter was described as ‘gentleman and merchant’ . More interesting, however, is that in the fine printed in Trop. Cart. (vol. 2, p. 17) and that printed in Wiltshire Fines (no.628) are two parts of what was probably originally a tripartite indenture, the third part of which would have been given to Hardell and his wife, and which has now been lost. It is quite rare for more than one part of a tripartite indenture to have survived.

41 Trop .Cart., vol. 1, pp. 89, 91-92.

42 Cal.I.P.M., Henry VII, vol. 1, no. 351.

43 Ibid.

44 C.C.R., 1468-76, pp. 63, 64; ibid. 1485-1500, p. 47.

45 J. Silvester Davies, ‘The Manor and Church of Great Chalfield’, Transactions of the Bristol and Gloucestershire Archaeological Society, vol. 23, 1900, pp.193-261; H. Avray

Tipping, ‘Great Chalfield Manor’, Country Life 15th Aug. 1914, pp. 230-7; 29th Aug. 1914, pp. 294-301; N. Pevsner, Wiltshire. 2nd ed., 1975 (The Buildings of England), pp. 257-9; Great Chalfield Manor, Wiltshire, 1966 (National Trust).

46 Trop. Cart., vol. 1, p. xii. The advowson of Great Cheverell remained in the hands of the Tropenells until 1553. The church was much remodelled between 1476 and 1553: Trop. Cart., vol. 2, p. 269; V.C.H. Wilts., vol. 10, 1975, pp. 49, 51.

47 Trop.Cart., vol. 1, p. xvi; Great Chalfield Manor (National Trust), pp. 7, 10.

48 Pevsner (op. cit., pp. 257-58) points out that such a feature belongs to the period of Henry VIII rather than that of Henry VII. The stone for the house came from the quarries at Hazelbury in which Tropenell had acquired an interest in 1465: V.C.H. Wilts., vol. 4, 1959, p. 247.

49 K. B. McFarlane, The Nobility of Later Medieval England, 1973 (Oxford), pp. 29-30.

50 C.C.R., 1485-1500, p. 58.

51 Wright (ed.), Register of Thomas Langton, |xxiv, no. 133%

52 PRO Prob. 11/8 (P.C.C., 7 Milles).

53 PRO C/142/3/103; Cal.IL.P.M., Henry VII, vol. 1, no. 351; Calendar of Fine Rolls, 1485-1509, p. 70.

54 This statement seems to conflict with the issuing of the writ diem clausit extremum in November 1487, since that would presume that Tropenell had already died. One possible explanation could be that the jurors or a clerk had stated or written ‘January’ instead of ‘November’ at the inquisition. If so, then Tropenell’s death would have taken place late in 1487 and not early in 1488.

55 Sir Thomas had been executed at Bemerton in 1469: Complete Peerage, vol. 6, p. 621.

56 Great Chalfield Manor (National Trust), p.11.

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 90-104

Excavations along the Littleton Drew to Chippenham

Gas Pipeline by Clifford Bateman

with contributions by Fiona Roe, Jane Timby and Tracey Stickler

In 1997 Cotswold Archaeological Trust undertook a continuous watching brief and associated excavations along the route of the Littleton Drew to Chippenham gas pipeline. This report details the findings of two excavations undertaken during the programme of works, one a multi-phased site with evidence of early to middle Iron Age and Romano-British activity, the other a hitherto unknown area of Romano-British activity. A gazetteer of isolated features and findspots of all periods identified during the watching brief is also included.

Cotswold Archaeological Trust was commissioned by Transco to undertake an archaeological watching brief along the route of the Littleton Drew to Chippenham gas pipeline (ST 8335 7950 to ST 9140 7535), a distance of approximately 12km, through a landscape of proven archaeological interest. The pipeline crossed the predominantly limestone geology of the Cotswold dipslope, which largely comprises middle Jurassic Great Oolite and Cornbrash. Middle Jurassic Forest Marble Clays and upper Jurassic Kellaways Clay, associated with the North Wiltshire Clay Vale, are more prevalent at the south-east extent of the scheme. ‘Topographically, the pipeline route consisted of gently undulating ground along the north-west section, becoming more uniform 3.5km towards the south- eastern end of the route. At its north-western extent the pipeline lies at approximately 130m OD; the south-eastern end lies at approximately 90m OD.

ARCHAEOLOGICAL AND HISTORICAL BACKGROUND

Within the Chippenham environs, evidence has been recorded indicating occupation and activity from the earliest prehistoric period up to the present day (Figure 1). Prehistoric activity is attested by the lithic scatters (Tucker 1985) and funerary monuments identified within the immediate hinterland of the pipeline. However, evidence of contemporary settlement within this area remains limited. Iron Age

and Romano-British activity is concentrated to the west of the pipeline upon the Cotswold dipslope. Evidence for Iron Age settlement is dominated by Bury Camp hillfort (Wiltshire SAM 130), although isolated findspots of Iron Age pottery have also been recorded, most commonly associated with Romano- British settlement. The spatial distribution of Romano-British settlement activity within the general area is focussed upon the Fosse Way (PRN 300), the major Roman road linking Exeter to Lincoln. A roadside settlement at Nettleton Shrub (Wiltshire SAM 311, PRN 302) containing twenty six buildings, a temple and cemetery was excavated between 1938 and 1967 (Wedlake 1982). A villa complex and associated cemetery were excavated at Truckle Hill, North Wraxall, in 1859-60 (Jackson 1862). No sub- Roman or Anglo-Saxon activity is recorded within the immediate vicinity of the pipeline, although the excavation of a Saxon sunken-featured building to the south-west of Chippenham (NGR ST 898 727) suggests occupation in the area (Anon 1991).

The pipeline route itself passed through three previously identified archaeological sites. At ST 8397 7937 it crossed a section of the Fosse Way (Area A). At Down Farm, Heywood, (ST 8880 7653) the route crossed an area where an assemblage of early prehistoric flints (PRN 052) had previously been found (Area B).The lithic material includes flint cores, scrapers, blades and flakes from the Mesolithic period. At Lodge Farm, Heywood (Area C), the pipeline intersected an area of cropmark features previously identified from aerial photographic evidence (PRN

Cotswold Archaeological Trust, Headquarters Building, Unit 9, Kemble Business Park, Cirencester, Glos GL7 6BQ

91

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE

= ST90000 ST88000 ST86000 ST84000 ST82000

Solis [BIISOTOIeYIIe UMOUY O] UOTeIAI UI UONeIOT :aurjedid ureyueddiyy 01 MoIg uoIIpIINT *[ o1ns Ly

spuly ysiiuig-OURWOY © spuly a5y vos gy

quawaljlas

@ smoueg d01sIYaId4 © ysnug-ouewoy

juawaj|ijas aby uol| w WUI}} WOS!IYaIg H

= p lf aN i / OJ [[IH wy Uol] / ( <a Be weg Aing wee)! pee [A a, O00PL1S ) Se a

i

“ uoiseppig

“jjeyeappip

xy BI[IA URWIOY ‘ 000921S —W [exelA, YON

a

Aa|Bue7 uozbuly p.

g eaVy og {jeuAey UoHeA —o2e:

ee ii yo. 4 Realy ine 7 quays wojapieNn

JeeyoiA) 1S UO WwUly

i a | ane

i JUSWST}OS UPLUOY

43 4 sey ti aquuoy) ajiseg saddy

000821S

ce .

“uaa uoja|18N uoyBuines : ae We eh

Se fork

pI ABMIOLOW uoung

ouatuejaq yBia7 *

o00081S

92 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

625). The cropmarks, centred on ST 8982 7656, include ring ditches, linear features, a D-shaped enclosure and a possible Romano-British building. Archaeological fieldwalking and excavation immediately south of the cropmarks complex recovered Romano-British pottery and tile (PRN 301 and 304).

Other archaeological sites within close proximity to the pipeline route include a probable long barrow at Green Barrow Farm (PRN 600) (ST 8545 7782), which is no longer extant (having been levelled in 1852), and an undated semi-circular enclosure (PRN 606) north of Park Farm at ST 8650 7810.

METHODOLOGIES

The archaeological recording was undertaken by a staged approach. The initial phase comprised the machine stripping of topsoil within three previously identified archaeological areas in advance of the main phase of construction, namely:

Area A: 100m either side of the Fosse Way (PRN 300)

Area B: 400m centred on the area of the flint scatter (PRN 052)

Area C: 500m through the cropmark complex (PRN 625)

A site meeting was held with representatives from Wiltshire County Council Archaeological Service and Transco, to determine the extent and nature of any further works. It was agreed that archaeological excavation should continue by hand within Area C, but the limited archaeological evidence from Areas A and B did not justify full-scale excavation.

A watching brief was to be maintained during all intrusive groundwork along the pipeline corridor, with further contingency for excavation 1n the event that significant archaeological deposits were encountered. One such site, Area E centred on ST 8790 7698, was identified and subsequently excavated.

RESULTS OF THE ARCHAEO- LOGICAL RECORDING

Excavations at Area C

The site was located on gently undulating agricultural land, ranging from 104.45m OD at the western extent

to 99.61m OD at its eastern limit. The underlying geology consisted of middle Jurassic Cornbrash. An area totalling 510m in length by 1.8m in width was machine stripped to the top of the natural substrate, with archaeological excavation continuing by hand thereafter (Figure 2).

Early to middle Iron Age

Ditches [536] and [542] correlate closely with linear cropmarks, which may now be interpreted as of early to middle Iron Age date. Aerial photographic evidence suggests that ditch [536] may represent an antenna ditch associated with the D-shaped enclosure 65m south of the pipeline.

Romano-British Ditch [503] continues the alignment of a north-east to south-west orientated cropmark, which may now be provisionally dated to the lst — 2nd century. Its close proximity to ditch [534], aligned perpendicular to the cropmark, may suggest that the two ditches formed the north-eastern corner of a Romano-British agricultural enclosure or paddock. Such an interpretation would suggest that the linear alignment of undated shallow pits [549], [551] and possibly [553], adjacent to ditch [534] are inside the postulated enclosure, and are broadly contemporary in date. Ditch [507] and stone-lined culvert [512] correlate closely with the alignment of previously identified linear cropmarks, and represent the western limit of activity identified from both the aerial photographic evidence and the excavation. Artefactual material retrieved from culvert [512] might indicate that the cropmarks are 2nd century in origin, although its close proximity to a similarly aligned redundant post- medieval field boundary may suggest the pottery is residual. The quality of dressed limestone utilised within the construction of the culvert, in conjunction with the deliberate sealing of the structure with non- local blue-green clay (514), suggests it was designed to carry a water supply rather than function as a land drain. The alignment of the culvert suggests it is associated with an amorphous depression, measuring approximately 20m in length by 15m in width, visible 15m north of the pipeline. The depression may represent an infilled water source subsequently channelled towards the area of contemporary settlement. Such a hypothesis would suggest the area of occupation lies to the south-west of the study area. Ditch [562] correlates with the projected alignment of a cropmark, which may now be interpreted as a 2nd — 3rd century agricultural boundary. The two phases of recuts within this ditch

93

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE

° 3° ° D re wn

‘posodulliodns soinjeay paviosse pue surjodid yum 9D variy jo uondriosued o1ydeisojoyd [eniay °Z a1ngh]

4 Ie

uolssaideg

ST89800

gos

juow

SS

—{—>

seo 9

uljadid

$T89700

sainieaj jeaibojoaeuaiy

sauepunog pjalj sayeq

sysewidoi9

OO0S9LLS

94

suggest longevity in its function as a boundary. Interpretation of pits [510], [520] and [530] is limited, although the moderate quantity of daub retrieved from the features reinforces the hypothesis of occupation within the immediate vicinity.

Undated

Ditches [528], [544], and [560] remain undated. Although they correlate closely with known cropmark alignments, the paucity of datable artefactual material retrieved from the features prohibits their interpretation.

Finds

POTTERY by Jane Timby

‘Twenty-two body/base-sherds (137g) of Iron Age date were recovered. Although the pottery was in relatively fresh condition, the lack of diagnostic sherds frustrates

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

close dating of the material which is all fossil shell and limestone-tempered. Ditch [542] produced exclusively coarse shell-tempered sherds (fabric H1) normally typical of the earlier Iron Age. Ditch [536] produced limestone and shell-tempered sherds (fabrics L3, L4), probably of early or middle Iron Age date.

Sherds of Roman pottery were recovered from seven individual features with an additional 30 sherds of unstratified material. Few of the features produced chronologically viable groups, with only two pits [520] and [530] yielding in excess of ten sherds. Most of the material would appear to belong to the 2nd century AD suggesting a hiatus in activity in the later Iron Age/early Roman periods. Pit [520] is probably slightly later, extending activity into the 3rd century AD. Sherds of Oxfordshire colour-coated ware from the unstratified material would confirm use of the area in the second half of the 3rd century but there is limited evidence for the 4th century. The majority of the wares are local products from the Wiltshire industries including jars from the Savernake Forest kilns. Imports are limited to two sherds of Dorset black burnished ware and two small scraps of samian.

Table 1. Pottery quantities (sherd count and weight) from Areas C and E

AREA C FABRIC No % WT ( % No % WT (g) | % IRON AGE HI 9 a Bee ae 13 3 2.75 E 2.40 en Rese L4 10 9.00 4 6.15 ROMAN: native wares eI ee ee ae aan ee ee MALVLI 2 : Semis eee GROG 6 1.60 44 1.05 ROMAN: local wares SAV 13 12.00 23 28 16 4.40 1115 26.45 WMBBW 14 12.80 fe 3 fe 23 6.40 89 Dall WILOX 1 0.95 23 3.00 eerie SWOX 4 3.50 12 1.60 10.80 10.00 ROMAN: traded wares SAM 3.50 7 0.95 ll 3.00 134 3.20 BBI 2 1.80 10 1.35 35.60 | 1493 35.40 Oxce 9 8.25 87 11.60 1.40 15 0.35 SVW 3 2.75 44 5.90 3 0.85 30 0.71 MICGW 4 3.50 25 3.33 16 4.40 248 5.90 ROMAN: source unknown Rl 1 0.95 2 R2 i 6.40 20 R3 1 0.95 9 R4 l 0.95 10 R5 3 2.75 7 R6 3 2.75 24 R7 Misc. reduced 12 11.00 64 Misc. oxidised 6 5.50 13 : ‘i Total 110 100.3. | 749 100 362 99.99 [| 4213. [100 |

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE 95

Table 2. Pottery fabrics and date

Description and forms

IRON AGE Handmade. Brown paste with a sparse to moderate density of | Early Iron Age Piso see eee medium-fine fossil shell and limestone ROMAN: Native Wares

Ital Handmade brown or orange-brown ware with a high density | Ist to 2nd century of very fine discrete ooliths of limestone with occasional larger fragments

L Brown medium-fine sandy ware with sparse fragments of fossil shell/limestone

GROG Handmade grog-tempered wares. Form: jar ee eo. Sloe hae, Seer

MAN: Local Wiltshire Wares

SAVGT Savernake ware (Annable 1962). Form: Large storage jars

WMBBW Wheel-made black burnished ware (cf Rigby 1982, 153, Cirencester TF 5). Form: Necked jars

a

Q

5 2 ; to

; 5 <a 5 © =} 3 Qa a fe} 5 fa) eat: Q 3 O a Le) (a) o a pt} rej ov : esl : es) oO ~ a: OQ a tea 3 ca) = 2 Qa oO 5 oe g Kd

a 2)

WILGW

WILRE WILOX SWOX

South-west orange sandy ware. Gritty feel. Forms: Small flagons, white-slipped mortaria, necked everted rim jars/bowls

ROMAN: Traded Wares

SAMCG Central Gaulish samian. Forms present, Dragendorff 30, 38, | Mainly 2nd century types 31, Curle 35

DORBB1

Bidwell 1991). Forms: Jars, straight-sided dishes, flat-rimmed bowls and grooved rim bowls

OXCC Oxfordshire colour-coated ware (Young 1977). Forms: | Later 3rd to 4th century tablewares and mortaria

NECC New Forest colour-coated ware (Fulford 1975). Form: beaker | Early 4th century.

MICGW Grey or brown micaceous sandy ware. Forms: copies of DORBBI forms, jars, straight-sided dish

ROMAN: Source unknown

Thin walled, hard, black medium sandy wheelmade ware. Ses Forms: jars, beaker

Fine, sandy ware with blue grey surfaces and a red-brown inner core. Wheelmade. Forms: Handled jug with vertical burnishing on the neck

ve)

ALE

R3 Black sandy, finely micaceous ware with a brown core Finer, black sandy ware with a slightly gritty feel. ?Originally

with a white slip

R5 A fine or medium grey sandy ware with occasional buff clay

pellets R A dense medium sandy ware with a speckled grey surface ROO/R15 Miscellaneous reduced wares

000/015 Miscellaneous oxidised wares

Ww Q

96 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Table 3. Animal bones from Area C

Early to Middle Iron Age

No frags | Species Identified

Ditch 536 26

Romano-British i 19 | Ditch 524 [18 | 14 cattle, 2 sheep

1 dog tooth 1 fox tooth

Pit 530

1 Red deer, 1 sheep, 1 cattle

1 adult horse; 4 cattle, 1 sheep, 1 Small

1 red deer tooth, 1 sheep, 6 LAR, 2 SAR

Comments 1 cattle butchered, 1 Small green fractured; root damage, condition poor

Poor condition, possibly exposed

1 butchered, 1 green fractured

Root damage, 2 gnawed

Adult cattle metatarsal with unusually large proximal articulation foramen, withers height 1236.4mm

bt a oe i nC! ho a ee)

ANIMAL BONE by Tracey Stickler

For material from the excavations at Areas C (Table 3) and E (Table 4), where possible, specimens were identified to species, or to higher order taxa of: LAR (large artiodactyle); SAR (small artiodactyle); or Small. The osteological differences between sheep and goat were identified after Boessneck (1969), weathering stages were assigned after Lyman (1994), and animal ages ascertained from epiphyseal fusion and tooth eruption after Silver (1969), and from tooth wear after Grant (1982). Measurements were taken as outlined by Von den Driesch (1976), and withers height calculations as presented by Matolsci (1970). Details can be found in the archive report.

In Area C the material associated with the Iron Age ditches is of very poor condition. This is probably due to extensive root action. The condition of the Romano-British material is better in general than that of the Iron Age, with less evidence of root damage. The high level of fragmentation is probably associated with the extraction of bone marrow, indicated by green fracturing and heating to melt the fats. Marrow extraction also continues to be strongly suggested by the post-medieval material. The presence of a naturally cast antler may indicate a local environment supporting established populations of deer.

DISCUSSION

‘The excavation at Area C revealed the presence of early to middle Iron Age and Romano-British activity. The strong correlation between the excavated features and the aerial photographic transcription has enabled

the provisional dating of a number of components of the previously identified cropmark complex.

The early to middle Iron Age activity identified is restricted to two ditches at the eastern extent of the excavations. Detailed interpretation of the features is limited, even allowing for their correlation with known cropmarks by the excavation, although ditch [536] may be interpreted as an antenna originating from the D-shaped enclosure.

Romano-British activity is more dispersed in nature, and largely comprises ditches of field enclosures and paddocks. Artefactual evidence retrieved from these deposits suggests 2nd — 3rd century activity, although the pottery retrieved from ditch [503] may hint at an earlier, lst century presence. Although no evidence of contemporary settlement was identified during the excavation, the identification of Roman tile during previous fieldwalking throughout the study area suggests occupation within the immediate environs. The significance of stone-lined culvert [512] remains undetermined due to its close proximity to both an undated cropmark and relict post-medieval field boundary.

Excavations at Area E

Area E, centred on ST 8790 7698, was initially revealed during the watching brief (Figure 3). It quickly became apparent that the significant archaeological deposits could not be recorded in tandem with the construction of the pipeline and contingency archaeological recording was implemented. This comprised the recording and planning of all features exposed within the wayleave,

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE

ST87900

Ww Zz Oo Kk ep) WwW = =

Grove Farm

ST87300

ST76700

Figure 3. Location plan showing Areas E and F

97

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

98

WOG

t90S ZS0S

aoueqinisip UWJapol

Mojaq panuljuod

OOLS

ued :q boy ‘fF ons]

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE 99

and the excavation of features affected by the cutting of the pipe trench. An area totalling 220 x 6m was machine stripped to the top of the natural substrate, with archaeological excavation continuing by hand (Figure 4).

The site is located on moderately sloping agricultural land, ranging from 120.5m OD to 117m OD. The underlying geology consists of middle Jurassic Cornbrash and Forest Marble clay. The excavation results can be assigned broadly to two main phases of activity, with evidence for the deliberate infilling of features in the late 3rd - early 4th century AD.

Phase 1 (1st to 2nd century AD)

The earliest activity was concentrated upon the gentle slope demarcating the Cornbrash from the Forest Marble clays. Ditch [5013] and gully [5011] may be interpreted as boundary ditches forming the north-western corner of a lst century enclosure or paddock, the 1m wide gap between the respective termini of the ditches representing an entranceway to the enclosure. It remains undetermined whether pit [5015], containing the articulated skeletons of two sheep/goats, was contemporary with this enclosure. Ditches [5009] and [5064] were more substantial in construction. Both were orientated approximately east to west, cutting across the natural slope of the land.

Interpretation of postholes [5004], [5006], and [5008] is problematic. They appeared to form a curving alignment, suggestive of a building. A similar north-western return of postholes was not identified during the excavation, although such may have been truncated by 2nd — 3rd century ditch [5018]. Given such an interpretation, the structure would have been centrally positioned between ditches [5009] and [5064], suggesting they formed a boundary around the building. Based solely upon three postholes, and given the high level of modern truncation within the general vicinity, such a hypothesis may be seen as speculative and interpretation as the corner of a fenced boundary could be of equal validity.

Phase 2 (2nd to 3rd century AD)

At the junction of the Cornbrash and Forest Marble clays were two wells. The close proximity of these to each other suggests they were not in contemporary use, but were situated to exploit the change in the local geology and utilise the same water resource. No evidence for the date of construction of well [5087] was retrieved, although artefactual evidence from the deliberate infilling of the shaft suggests it was

redundant by the late 2nd — 3rd century. The construction technique utilised to form the shaft of well [5066] suggests it had originally contained a lining, presumably of wood or wicker, with redeposited clay (5068) between the lining and the original well cut. Artefactual evidence from this deposit suggests it was constructed in the 2nd or 3rd century.

The relationship between ditch [5027] and well [5087] remains speculative, but it is possible that [5027] and associated recut [5048] acted as a drainage outlet from the well, feeding the excess water into drainage ditch [5033]. Gully [5021] is broadly contemporary with ditch [5027], although its function remains undetermined. The construction of stone-lined ramp [5059] on the south-western side of ditch [5027] was undoubtedly to allow access to the ditch, in all probability for livestock. Furthermore, the irregularity of the southern side of the ditch may also have resulted from the encroachment of livestock poaching the ground along the edge of the ditch. The function of pits [5057] and [5061] identified to the north of ditch (5027], immediately opposite the stone-lined ramp, remains undetermined.

Ditches [5041] and [5043] delineate the south- eastern extent of activity identified during the excavation, and may form an agricultural enclosure/ paddock, utilising drainage ditch [5033] as its western boundary. Ditches [5018] and [5092] were revealed on the higher ground cutting phase 1 ditch [5064], and may represent further agricultural field systems. Pit [5100] was revealed adjacent to ditch [5092], and is broadly contemporary.

Phase 3 (late 3rd to early 4th century AD)

Deposit (5053)/(5072), consisting of small fragments of limestone rubble within a clay matrix, sealed the north-western extent of ditch [5027] and wells [5066] and [5087]. The deposit is likely to represent an attempt to consolidate an area of increasingly wet/ marshy ground for continued use. Artefactual material retrieved from the deposit suggests that these features had become redundant by the early 4th century.

Unphased

At the northern limit of the excavation, ditch [5073] was flanked on its southern side by an elongated sub- oval pit [5077] and by two postholes [5075] and [5096]. This arrangement was mirrored on its northern side by pit [5081] and posthole [5079]. The lack of artefactual material retrieved from these features prohibits their dating.

100 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

O 100mm

O 50mm

Figure 5. Early prehistoric pebble-hammer stone and Romano-British copper alloy brooch

Finds

POTTERY by Jane Timby

An assemblage of 363 sherds (4214 g) was recovered from Area E (see Table 1). The pottery suggests activity from the ist century through to the early- mid 4th century. Particularly large assemblages were recovered from ditches [5027] and [5033], and from well [5087], which account for 58% of the assemblage. The majority of the wares are of local origin supplemented in the later Roman period by products of the large regional industries. Foreign imports are limited to a few samian tablewares.

Amongst the earlier Roman features investigated was ditch [5064], which produced later Ist century material. Further small groups of potential 1st century wares were recovered from features [5013] and [5018]. Second century wares are more prolific and include several pieces of samian, further Savernake ware, wheel- made black burnished ware, along with greywares. Features with typical 2nd century material include [5004], [5011], [5015], [5018], [5025], and [5048].

Pottery spanning the later 2nd — 3rd centuries was recovered from wells [5066] and [5087], in particular fills (5068) and (5089) respectively, and from ditches [5027] and [5092]. Wares dating to the second half of the 3rd century into the 4th century were associated with the upper fill of well [5087] and deposit (5053/ 5072):

WORKED STONE by F. Roe

Part of a pebble-hammer stone was retrieved from Romano-British ditch [5033] (Figure 5). It was

broken across the hour-glass shaped shafthole, the surviving end being somewhat battered. It appears to have had a secondary use as a whetstone. The original pebble would have been more rounded in shape, but both sides are now flattened by wear, and there also appears to have been some use on the top and bottom surface.

The stone is a grey, slightly micaceous sandstone, possibly originating in the Cretaceous Lower Greensand. It may have been collected as a pebble from local gravels belonging to the River Avon, or else from local Pleistocene Drift.

This type of shafthole implement is early prehistoric in date, but is difficult to date more precisely than to a generalised Mesolithic to Bronze Age date range (Woodcock et al 1988, 30; Roe 1979, 36). Pebble hammers also occur quite frequently as apparently residual finds in Iron Age or Roman contexts, but the re-use as a whetstone is unusual, and seems likely to be contemporary with the finds of 3rd century-pottery from the ditch. It is difficult to know what the purpose of these drilled pebbles may have been, but use as weights for bow drills or small, all purpose hammers are possibilities. In Wiltshire, other examples of pebble hammers include finds from Neolithic sites at Durrington Walls and Windmill Hill, and an Iron Age site at Fifield Bavant.

ANIMAL BONE by Tracey Stickler

That activity occurred in the Ist century in Area E is about all that can be deduced from the material. The 2nd century is similarly represented with the exception of pit [5015], which produced articulated skeletons of sheep/goat. Unfortunately, taphanomic interpretation is not possible and the conditions of

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE 101

deposition cannot be ascertained, but there is no evidence for a period of exposure or scavenger activity, suggesting intentional burial rather than refuse pit discard. The consistent presence of Red Deer teeth is likely to indicate an established local population, which was utilised in the 2nd — 3rd century at least. The 2nd — 4th centuries also saw active marrow extraction.

DISCUSSION

The rapid response excavation revealed a hitherto unknown area of dense Romano-British activity. Analysis of the artefactual evidence suggests that activity spans the Romano-British period, with a noticeable concentration of activity within the 2nd — 3rd centuries. Detailed interpretation of the activity is restricted by the limited confines imposed on the excavation, but the general character of the deposits pertains to an agrarian landscape. The initial Ist — 2nd century activity is restricted to the higher

Table 4. Animal bones from Area E

First century Context

Ditch 5064 Second century PH 5004

PH 5008

Ditch 5009

Pit 5015

Ditch 5092 Second to third centu Ditch 5018

Second to early fourth

Ditch 5027

Ditch 5027

LAR

Well 5087 Layer 5091

Red deer

Species Identified Comments 1 Small Root damage and chop marks

Heated

1 Small Fractured when green and heated 4 cattle, 2 dog Wermien Sts? Anis ee ce ee

166 1 Sheep/goat skeleton, 1 sheep skeleton 1 Mature, ?male sheep/goat;1 immature sheep Ditch 5025 1 Juvenile sheep, 1 Small, 1 horse

ee ee | Practived wher green and/heated

19 11 Red deer, 2 SAR 1 Red deer green fractured and butchered; 1 unidentified carbonised

2 Red deer, 8 LAR, 2 SAR

1 Horse, 2 SAR, 2 LAR

5 Cattle, 1 LAR, 1 Small, 1 horse, 2 sheep, 1

Ditch 5033 16 9 Cattle Cattle humerus either butchered or worn from suspension after most of meat removed

ground, with expansion of activity into the heavier, low-lying Forest Marble clays during the later 2nd and 3rd centuries.

No evidence of the contemporary settlement associated with the features was revealed. Such a settlement is likely to be situated on the higher (presumably drier) ground associated with the Cornbrash. As no further Romano-British features were revealed to the north-west of Area E, it may be suggested that the settlement lies either to the west of the study area, in the general vicinity of Grove Farm, or to the north-east of the pipeline.

A Gazetteer of Archaeological Sites Noted Along the Littleton Drew to Chippenham Gas Pipeline

The following gazetteer provides details of the archaeological sites identified during the watching brief maintained throughout all intrusive groundworks along the pipeline route (Figure 1).

Small butchered; horse green fractured; 4 unidentified burned

Butchered scapula 4 LAR green fractured, SAR and 8 unidentified heated

SAR heated and butchered; LAR heated and green fractured

2 Small green fractured, heated and

punctured; 5 unidentified green fractured and heated

102 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

AREA A. ST 8397 7937

Two linear ditches were recorded in section, on either side of, and parallel to, the existing road between The Gibb and Grittleton. Ditch [103], 3.5m wide and 0.4m deep, was 1.75m east of the modern road. Ditch [105], 3.2m wide and 0.75m deep, was 7m west of ditch [103], and was partially sealed by the modern road. Although no artefactual material was retrieved from either of the ditches, and no evidence of Roman road surfaces encountered, they may be interpreted as the Roman roadside ditches of the Fosse Way.

AREA B. ST 8880 7653 Advance topsoil stripping centred on the previously identified Mesolithic flint scatter (PRN 052) identified three features. A sub-square pit [201], 2.5m wide and 0.65m deep, was revealed at ST 8862 7660. Six undated worked flint flakes were retrieved from the pit as well as 55 animal bones, which included four cattle teeth of mid-range wear. All fragments were unaltered and showed no signs of weathering or modification. Feature [208] was revealed at ST 8893 7643, measuring at least 0.85m long, 0.4m wide, and excavated to a depth of 0.2m. No artefactual material was retrieved from this. Ditch [206], at ST 8878 7653, was orientated north-north-west to south-south-east, and measured 1.1m wide and 0.5m deep. One sherd of post-medieval pottery was retrieved from its fill. A Romano-British copper alloy disc brooch dated to the mid-1st - late 2nd century (Hattatt, 1982, 137) was retrieved from the ploughsoil at ST 8894 7644 (Figure 5). The brooch is decorated with a central raised ring, the centre of which contains a small recessed boss. The outer edge of the brooch is moulded with a further area of raised moulding midway between the edge and central ring. Small traces of gold leaf are apparent between the outer edge and the central raised ring. The hinged pin was missing. The lack of contemporary features within the immediate vicinity suggests the brooch represents a stray find.

AREA D. ST 8492 7868

A sub-oval pit [403], 1.1 x 0.6m and 0.1m deep, produced one sherd of 2nd — 3rd century AD pottery and a worked flint flake. Pit [405] was revealed 2m north-east of [403]. Measuring 1.2 x 0.9m, and 0.07m deep, it contained no finds. A small flint assemblage, consisting of six flakes, one broken flake, two broken blade flakes and two burnt worked pieces was recovered from the ploughsoil within the general area.

This material is broadly late prehistoric in character, although two flints may be Mesolithic.

AREA FE. ST 8771 7716

Ditch [601] was orientated north-south and measured 1.7m wide and 0.45m deep. Three sherds of Iron Age pottery (now lost) were retrieved.

AREA G. ST 8676 7762

A ‘bowl shaped sub-circular pit [703], 0.5m in diameter and 0.15m deep, contained a significant concentration of charcoal and slag suggestive of in- situ smithing activity. No datable artefactual material was retrieved.

OVERALL CONCLUSIONS

The programme of archaeological recording undertaken during the construction of the gas pipeline has revealed seven sites of interest. Two topographical zones were crossed by the pipeline: the majority of the route lay on the limestone geology of the Cotswolds, while the final section from Kington Langley to Chippenham cut through the Kellaway clays of the North Wiltshire Clay Vale. Observations made on the pipeline have helped elucidate the pattern of past settlement at the southernmost limit of the Cotswolds.

Archaeological deposits of varying significance were encountered along the lengths of the pipeline that crossed the known archaeological Areas A, B and C. The watching brief succeeded in identifying significant and hitherto unknown Romano-British deposits at Area E, as well as a number of isolated features that ranged in date from prehistoric to post- medieval. Although the limited nature of the groundworks and the nature of the topsoil stripping reduced the ability to interpret the archaeological features encountered during the watching brief, the methodologies employed were sufficient to identify archaeological deposits and to retrieve artefacts from within the topsoil/ploughsoil.

The earliest activity detected is represented by six flint artefacts from Area B, the Mesolithic flint scatter previously identified by Tucker (1985, table 1, Allington 2; Anon 1985, 254 (5)). Unfortunately none of the newly discovered flint was diagnostic, and there can be no certainty that the few cut features found in this area have any association with the overlying surface scatter. A second flint scatter was found at

EXCAVATIONS ALONG THE LITTLETON DREW TO CHIPPENHAM GAS PIPELINE 103

Area D, which contained two pieces suggestive of Mesolithic technology. Little further comment can be made on these two sites, save to note that they are typical of the scatters recorded by Tucker (1985) in the Chippenham region, and which occur across the southern Cotswolds generally (Saville 1984). They presumably indicate the sites of temporary camps utilising woodland and riverine resources. A number of finds of Neolithic-Bronze Age flintwork have been found in the general vicinity of the pipeline (Figure 1), although their distribution is more a reflection of the location of individual survey programmes than of past settlement patterns.

A pebble-hammer stone was found in a Roman ditch at Area E. Such artefacts have a generalised Mesolithic to Bronze Age date range. While noteworthy in its own right, its subsequent re-use (in the Roman period?) as a whetstone means that it need not have originally been deposited within the vicinity of Area E.

The earliest period for which we have structural remains is the early to middle Iron Age activity identified at the eastern extent of Area C. Pottery recovered from the pipeline suggests that part of the cropmark complex previously identified from aerial photographs is of this date. Although the D-shaped enclosure to the south of the pipeline was not investigated, its possible association with ditch [536] suggests that it may now be interpreted as an early to middle Iron Age enclosed settlement. The D-shaped enclosure (internal area approximately 500m’) appears to be associated not only with linear (?field) boundary [536] but also with a sinuous ‘antenna’ ditch. This ditch was sectioned in the pipe trench as [528]. It contained 30 fragments of animal bone but no dating evidence. If an early-middle Iron Age date is accepted for the enclosure it can be classified alongside the ditched farmstead enclosures of the Wessex chalklands (Cunliffe 1984). A number of the other ditches might be part of a contemporary field system. Area C lies 8km north-east of Bury Wood Camp, a hillfort which has produced evidence of intensive occupation in the middle Iron Age (4th — 2nd century BC) (King 1967). Other Iron Age finds from the general vicinity (Figure 1) comprise small collections of pottery recovered from various sites examined in advance of the Chippenham bypass and ring main pipeline, and a stray find of a gold stater of Corio from near Yatton Keynell.

The study of the Romano-British settlement pattern in this part of North Wiltshire has traditionally focused upon the rich villas, and Branigan (1977, 24- 31) has stressed the importance of proximity to both

roads and the town of Bath in their siting. It is noticeable that the villas to the north-west of Bath (such as North Wraxall) are concentrated to the west of the deeply incised valley of the By Brook which must have hindered east-west communications. Further south, villas tend to concentrate close to the Bath-Mildenhall road, especially around the small town of Sandy Lane ( Verlucio). The area to the north- west of Chippenham, although away from the main concentration of villas, was not devoid of settlement in this period as the results from Areas C and E testify. The majority of the small pottery assemblage in this period from Area C seemingly dates to the 2nd or 3rd century, and there is no evidence, or requirement, to seek continuity from the previous early-middle Iron Age occupation. The material recovered in the pipeline excavation may be added to the 20 sherds of pottery retrieved during fieldwalking by Chippenham College Archaeology Group in 1984, and the further pottery and tile found in a small excavation by the same group. The tile suggests the presence of a building in the Roman architectural tradition somewhere on the site. The pottery from Area E dates to the mid-1st to mid- 4th century, and includes a number of imports from outside the region. Oxford and New Forest colour- coated wares, and Central Gaulish samian, indicate that the site had access to systems of regional trade, and the inhabitants were clearly above subsistence level. Nevertheless, there is currently no evidence to suggest that either site need be classified as a villa (and certainly none to make them fall within the definition of a villa adopted by RCHME (1976, Xxxvili) for the Gloucestershire Cotswolds). It is noteworthy that amongst the small animal bone assemblages from both sites there is evidence that the local environments supported established populations of deer. Due to the small size of the assemblages, interpretation of the significance of deer within the local economy must be viewed as speculative. It remains undetermined whether the bone is representative of the ad hoc subsistence killing of deer, or whether deer were managed either for recreational hunting, or as a resource to provide food and skin on a commercial basis.

Previously unknown Romano-British field systems in the area around Chippenham have come to light during development works over the last few years (Anon 1993; Bateman and Enright 2000). Associated settlements (although perhaps not rich villas) must await discovery. The results of the pipeline investigations help to fill out a picture of a managed agricultural landscape in this part of North Wiltshire in the Romano-British period.

104 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Acknowledgements

The archaeological recording along the Littleton Drew to Chippenham pipeline was funded by Transco. The author would like to thank Paul Swinbourne, Mike Jefferies and Andy Leach (Transco) and Duncan Coe (Wiltshire County Council Archaeological Service) for their assistance during the course of this project. The project was managed by Neil Holbrook. The excavations at Areas C and E were directed by Clifford Bateman and the watching brief supervised by Mark Brett and Simon Roper-Presdee. The illustrations were drawn by Richard Morton. The project archive has been deposited at Devizes Museum.

Bibliography

ANNABLE, F. K., 1962, ‘A Romano-British pottery in Savernake Forest, kilns 1-2’, WAM, 58

ANON 1985, ‘Wiltshire Archaeological Register for 1983’, WAM, 79, 254-9

ANON 1991, ‘Excavation and Fieldwork in Wiltshire 1989: Chippenham Without’, WAM, 84, 143

ANON 1993, ‘Excavation and Fieldwork in Wiltshire 1991: Chippenham’, WAM, 86, 159

BATEMAN, C. M., and ENRIGHT, D., 2000, ‘Excavations of Bronze Age and Romano-British Sites Along the Chippenham Western Bypass A4 to A350 Link’, WAM, 93, 233-8 [this volume]

BOESSNECK, J., 1969, ‘Osteological differences between sheep (Ovis aries Linne) and goat (Capra hircus Linne)’, in D. Brothwell and E. S. Higgs (eds), Science in Archaeology, 331-58. London

BRANIGAN, K., 1977, The Roman Villa in South-West England

CUNLIFFE, B., 1984, ‘Iron Age Wessex: Continuity and Change’, in B. Cunliffe and D. Miles (eds), Aspects of the Iron Age in Southern Britain 12-45. Oxford: University Committee for Archaeology

FULFORD, M. G., 1975, New Forest Roman Pottery. Oxford: British Archaeological Reports

GILLAM, J P., 1976, ‘Coarse fumed ware in North Britain and beyond’, Glasgow Archaeological Journal 4, 57-80

GRANT, A., 1982, “The use of tooth wear as a guide to the age of domestic ungulates’, in B. Wilson, C. Grigson, and S. Payne (eds), Ageing and Sexing Animal Bones

from Archaeological sites, 91-108. Oxford: British Archaeological Reports

HATTATTER., 1982, Ancient and Romano-British Brooches. London

HOLBROOK, N., and BIDWELL, P.T., 1991, Roman Finds from Exeter. Exeter: Exeter Archaeol Report 4

JACKSON, J.E., 1862, ‘On a Roman Villa discovered at North Wraxhall’, WAM, 7, 59-80

KING, D.G., 1967, ‘Bury Wood Camp, Excavations in the Area of the South-West Opening’, WAM, 62, 1- 15

LYMAN, R.L., 1994, Vertebrate Taphonomy. Cambridge: Cambridge University Press

MATOLSCI, J., 1970, ‘Historische Erforschung der KorpergroBe der Rindes auf Grund von ungarischem Knochenmaterial’, Zeitschrift f. Tierzuchtung. und Zuchtungsbiologie 87, 89-137

RCHME 1976, Ancient and Historical Mounuments in the County of Gloucester; Volume I Iron Age and Romano- British Monuments in the Gloucestershire Cotswolds. RCHME

RIGBY, V., 1982, “The pottery’, in J. S. Wacher and A. D. McWhirr, Early Roman Occupation at Cirencester, 153- 200. Cirencester Excavations 1

ROE, F., 1979, ‘Typology of stone implements with shaftholes’, in T. H. Clough and W. A. Cummins (eds), Stone Axe Studies, 23-48. London: Council for British Archaeology, Research Report 23

SAVILLE, A., 1984, ‘Palaeolithic and Mesolithic evidence from Gloucestershire’, in A. Saville (ed.), Archaeology in Gloucestershire, 59-79

SILVER, I.A., 1969, ‘The ageing of domestic animals’, in D. Brothwell and E. S. Higgs (eds), Science in Archaeology, 283-302

TUCKER, J.H., 1985, ‘Upper Palaeolithic and Mesolithic Sites in the Chippenham Area’, WAM, 79, 226-228

VON DEN DRIESCH, A., 1976, A Guide to the Measurement of Animal Bones from Archaeological Sites. Harvard

WEDLAKE, W.J., 1982, The Excavation of the Shrine of Apollo at Nettleton, Wiltshire, 1956-1971

WOODCOCK, A.G., KELLY, D.B., and WOOLLEY, A.R., 1988, ‘The petrological identification of stone implements from south-east England’ in T.-H. Clough and W.A Cummins (eds), Stone Axe Studies Volume 2, 21-33. London: Council for British Archaeological, Research Reports 67

YOUNG, C. J., 1977, Oxfordshire Roman Pottery. Oxford: British Archaeological Reports

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 105-19

Early Castles in the Medieval Landscape of Wiltshire

by Oliver H Creighton

The evidence for early castles in Wiltshire 1s reviewed in order to establish a list for the county more reliable than those hitherto available. Further understanding of these enigmatic sites can only be achieved through integrated analysis of topographical, documentary and, where possible, archaeological evidence. The above ground remains of these sites are usually complex and multi-phase field monuments that dety analysis through simple classificatory schemes. A holistic approach to early castles emphasises their important contribution to the medieval landscape, as evident in their interrelationships with settlement topography, patterns of ecclesiastical provision and teneurial

geography.

INTRODUCTION

Previous studies of Wiltshire castles have tended to focus upon a small number of the more major sites: primarily baronial and royal castles which are both fully documented and associated with upstanding masonry remains. This paper aims to go some way towards redressing the balance in favour of more minor, predominantly rural, early castle sites (i.e. those with likely occupation during the period c. 1066- 1216). The majority of these sites, which have suffered a dearth of academic scrutiny, have little or no contemporary documentation and survive as earthworks that are often ill-defined and heavily eroded. The interpretation of earthworks is thus a vital tool in the study of early castles and will form an important component of this study. It is also important, however, to frame these sites within the context of contemporary medieval landscapes. The early castles of Wiltshire, as elsewhere in lowland England, were suspended within the web of medieval landscape at a variety of levels. As military sites with strategic or tactical roles, mottes and ringworks were often sited to dominate key resources and routes of communication, yet as manorial centres, many were related closely to networks of estates and functioned as integral components within medieval settlement patterns. In addition, as centres of élite consumption and symbols of seigneurial power, early castles were powerful icons of Norman lordship.

WILTSHIRE CASTLE STUDIES

Summary accounts of Wiltshire castles have appeared within national listings of castle sites (Renn 1968; King 1983). Likewise, other nationally based lists relating to certain classes of castle earthwork, or castles of a certain period, have included the relevant evidence from the county (Brown 1959; King and Alcock 1969). However, synthetic studies of the castles of Wiltshire in their own right are somewhat lacking. An inventory purporting to list all Norman castles within the county was published in the Wiltshire Archaeological and Natural History Magazine early this century (Downman and Goddard 1919), drawing upon a remarkable series of earthwork plans commissioned by the Society and executed by Rev. Downman in the period 1901-09." These illustrations form part of a series of seventy Wiltshire earthwork plans that, along with the accompanying annotations and summary, provide a unique insight into the condition of several early castle sites prior to destruction or mutilation. However, the limited scope of Downman and Goddard’s list ensures that a more comprehensive and updated statement is essential. This study, based on an unpublished thesis (Creighton 1994), presents the evidence for early castles in Wiltshire with the benefit of recent advances in castle studies, not least a major reassessment of the importance of earth and timber fortification (Higham and Barker 1992). A preliminary section of

School of Geography and Archaeology, University of Exeter, Laver Building, North Park Road, Exeter EX4 4QE

106 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Ashton Keynes Oaksey &

Ds ae

: } vee AAT, a

Great Somertord &

Vig BEES OG ase Hinton

Castle i. La Aldbourne Chase Chippenham [~~ @ Membury

jee St Andrew

Lyneham

a Marlborough Sibury Hill

Hartham Park

a C on

ae er Trowbridge l Littleton Pannell ; Ludgershall ® Bishopstrow Farm =) A . =< ‘ yi ) { / Sherrington Stapleford , Stourton Wilton @ ( Old Sarum West Dean re ica ae / . » Downton ae A Motte @ Ringwork i ° ee [eee wee Genes See w Vanished/Possible Early Castle N

Fig. 1. Location of sites mentioned in the text

EARLY CASTLES IN THE MEDIEVAL LANDSCAPE OF WILTSHIRE 107

this paper synthesises the results of archaeological research relating to urban castles in the county. The remaining sections deal thematically with the evidence for lesser Norman castles, drawing particular attention to questions of earthwork interpretation, pre-castle occupation and the landscape context of these enigmatic sites. Figure 1 presents a location map of the sites discussed.

The field archaeology of early castle earthworks is well established, not least owing to the contribution of three seminal papers that have established the existence of two distinctive forms of Norman fortification among others: the motte and the ringwork (Renn 1959; King and Alcock 1969; King 1972). Detailed studies of castle earthworks at a local or regional scale demonstrate, however, the existence of a catena of intermediate earthwork forms between the motte and the ringwork, as opposed to a strict dichotomy between the two (Higham 1982, 109-110). In essence, any clear-cut distinction between mottes and ringworks is ultimately misleading, as it neglects the sheer variety of early castle forms and their reference to existing landscape features, in addition to the fact that many earthworks are the amalgam of several phases of development. Accordingly, this study uses the terms ‘motte’ and ‘ringwork’ as convenient labels as opposed to watertight and inflexible categories.

Major Castle Sites: Excavation and Fieldwork (Figures 2-4)

Only two early Wiltshire castles have been excavated systematically in the modern era. Excavations at Trowbridge (ST 856579), conducted on an intensive rescue basis, have added substantially to our understanding of the castle’s defences, if not its internal structures. Significant questions also remain concerning the chronological development of the site: although Humphrey de Bohun’s castle was besieged in 1139, the excavations did not establish whether or not the defences were established substantially before this date, nor demonstrate whether the putative motte ~ and two baileys were the product of a unitary phase of construction or a process of longer-term development (Graham and Davies 1993, 57-77, 147- 48). On the basis of morphological evidence alone, however, a likely scenario appears to be the secondary addition of motte to a pre-existing sub-rectangular ringwork (cf. Goltho, Lincs.: Beresford 1987). Remarkable evidence was also revealed of a Saxo- Norman manorial settlement that was cleared to make

way for the castle, and a church and cemetery that were incorporated within its perimeter (Graham and Davies 1993, 146).

At Ludgershall (SU 264512), a programme of research excavation between 1964-71 revealed a complex sequence of continually rebuilt timber and stone buildings within the double ringwork (Addyman 1969; 1973), allowing remarkable correlation with an extensive corpus of royal documentation (Colvin et al. 1963, 729-31; Stevenson 1992, 70-72). The excavated sequence provides a sobering reminder of the manner in which the final phase of a castle earthwork can potentially shroud multiple phases of development. For instance, the first timber-revetted earthwork defences were preceded by a cluster of modest stone and timber structures associated with stake fences that may, in common with Trowbridge Castle, indicate an antecedent phase of manorial occupation (Addyman 1973, 8, 11). The final publication of the excavations includes a valuable analysis of the castle within the context of its immediate environs and wider regional context that emphasises how the castle’s surroundings - earthworks, parks and borough - were manipulated in order to create an ornamental setting commensurate with the castle’s status (Everson et al. forthcoming).

The great foreland ringwork at Devizes (SU 002613) is associated with two concentric D-shaped baileys, the outer constituting a town ward. Evidence for the castle is summarised adequately elsewhere, as is the site of Old Sarum (SU 138327), where the largest ringwork in the country is set within the defences of an iron age hillfort that similarly enclosed a castle-dependent borough (Cunningham 1945-47; Rahtz and Musty 1957; RCHM 1980, 1-15, 173-74). More obscure are the earthworks on the site known as the Moot, Downton (SU 181214). Here, the castle earthworks have been modified radically from the eighteenth century to create a formal designed landscape in the grounds of the mansion known as Moot House (Clark 1875, 305-09; Squarey 1906, 3). Although remodelled as a remarkable earthwork theatre with associated fishpond, terraces and a hexagonal Temple of Mercury, the earliest identifiable earthworks can be rationalised as a large ringwork with a single bailey to the east. Whilst conceivable that the ringwork is only partial, forming a crescent backing onto the River Avon, the manner in which the north-west corner of the ringwork bank shows signs of deviating to the south suggests, however, that the west side of a formerly oval earthwork has been removed during quarrying. Raised in 1138 by Henry

108 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

de Blois, Bishop of Winchester, and almost certainly slighted in 1155, the fortification was apparently short- lived (Schortt 1947, 166; Biddle 1969, 30-31).

The ringwork lies immediately east of the Old Court site, separated from it by a post-medieval mill leat. Here, substantial masonry remains thought to be indicative of the Bishop’s Palace have been noted, while a large Saxon gravel pit nearby in Castle Meadow was almost certainly used to provide make- up for the site (Rahtz 1964; Musty 1966). These observations raise the distinct possibility that the castle was conceived as an appendage to an extant archiepiscopal palace site, and thus represented the temporary transfer of functions to a fortified nucleus as opposed to the creation of a new seat of lordship. The identification of Castle Meadow as the site of the siege castle erected in opposition to Downton in 1148 (Squarey 1906, 30-31) is thus almost certainly erroneous. Whilst feasible that the siege castle is lost, it can probably be identified as the ringwork and double bailey erected within an Iron Age hillfort at Godshill, Hants., c. 5.5km to the south-west at SU 166162 (Renn 1968, 173).

The castle at Marlborough (SU 184687) was a royal foundation and is documented extensively from 1110 when Henry I stayed there (Stevenson 1992, 72-73). Interpretation of the present field monument is problematic due to extensive landscaping from the seventeenth century, which included the cutting of a spiral walkway on the motte and the insertion of a summer house and pond on its summit (VCH Wilts. XII 1983, 169-70). Opinion is divided concerning the origin of the castle mound. A recent review of the evidence does not rule out the possibility that the great conical earthwork, c. 14m in height, is a de novo Norman creation (Whittle 1997, 169-70). The combination of the size of the earthwork, the origins of Marlborough’s place name (‘barrow of Maerla’) and the recovery of red deer antler fragments from the mound early this century, argues strongly, however, that the nucleus of the motte is formed by a prehistoric barrow (Haslam 1976, 41-42; Stevenson 1992, 70). In addition, sufficient vestiges remain to suggest the former existence of a single quadrilateral bailey to the south-east of the motte, provided with moated defences fed by a branch of the River Kennet, as depicted on the Tithe Map.’

Evidence for a number of other urban castles is marginal. Anarchy-period fortifications are documented at Calne, Cricklade and Wilton, although their sites are unknown and remain the subject of speculation based on topographical and place-name evidence (Thomson 1958-61, 71-72; Haslam 1976,

13, 18, 68-69). The evidence for an undocumented motte immediately west of the Market Place in Chippenham (ST 922732) is fragmentary yet suggestive. During urban development early in the nineteenth century a large earthen mound was described in conjunction with masonry remains that included a Romanesque doorway (Ibid., 16). Whilst tradition dictates that a Saxon palace lay in the vicinity, this evidence is not inconsistent with a motte. Further interpretation is undermined, however, by a lack of corroborative documentation.

Rural Mottes and Ringworks (Figures 3 and 4)

Two Wiltshire ringworks are distinguished by their isolated positions within the landscape as well as their small size and lack of a bailey or outworks: Lewisham Castle, Aldbourne (SU 244739), and Cam’s Hill, Malmesbury (ST 941858). Lewisham Castle survives as a powerful univallate earthwork that commands much of Aldbourne Chase from a locally prominent ridge-top position. The circular interior of the work has been consolidated with a compact layer of flint nodules, ensuring that it is raised slightly above the surrounding terrain. Whilst unexcavated, medieval activity at the site is confirmed by the recovery of thirteenth-century pottery and iron arrowheads (Carrington 1855, 127-29; Brentnall 1945-47, 472- 73).The Cam’s Hill earthwork is of similar, if slightly more compact, form and occupies the point of a ridge, c. 1.5km south of the town, overlooking an important crossing of the Avon. There is little reason to doubt the thesis that this work is one of the three erected to besiege Malmesbury castle in 1144, although the linear earthworks at the base of the ridge, between Cam/’s Hill and the Avon, appear to be field boundaries/flood defences as opposed to a siege-line (see King 1983, 499). It is possible that Lewisham Castle could have similar origins as a minor fortification erected out of short-term military necessity, as in April 1217 the mercenaries of Louis the Dauphin, following expulsion from Marlborough Castle, were ensured safe passage to ‘their own place’ following the despoliation of Savernake Forest (Brentnall 1945-47, 472-73; see also Stevenson 1992, 74). Whilst credible that the castle was thus a temporary raiding base, it cannot be discounted that it originated as, or continued to function as, a hunting lodge, situated admirably for the administration of the surrounding chase and the accommodation of hunting parties.

EARLY CASTLES INTHE MEDIEVAL LANDSCAPE OF WILTSHIRE

109

Trowbridge Town Plan Devizes Town Plan

0 200m

Tt, / ~ doned0Gn ‘“ s Calta, 5 \ \\4 Ges es = WK eee i \Y NS 1 pe —= ~ Sy ; = 1 \Y S — _ SY aS foo = = ‘ ~~ i v= z) L == 6 — Ean) yy = 4 . cae = Ts ae *, = pose on Ve =) SY ep = = — = SI ae Ae Se Aye SS ~ + ae “Y mn 3 < . Trowbridge ee: Array .\h Devizes * t; ay, Hy oe “> 44 Ky > s hy SA i

~ : Haw ee es Ye tod bd Uh sepa rtts

Ts

Suadlaty, STUPET py PR eVdVTEP trees

aN

aeesdaseray

. wwe?! PVE Pes

are

3

ety

200m

Fig. 2. Wiltshire ringworks with multiple baileys

110

Malmesbury

veer per hy ernng,, Piotr ol ees tae ways. ‘er

Downton

book Jah Bo, $e ZA.)

BRU eth ite

eee OY YS Te

Fig. 3.

West Dean

Wiltshire ringworks

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Wi WNrettaus’, - ea Vunusitangon ee ts, ass as ry IS ie Soe Casolen aegyy tr, ene Aas Pre tess <> 3a pier ee 4 Ota ate”) nt. “4 . va ean as ry ose ame =: mS os . -:: - “NA ee, ale ~ Pre Reo) ts. -%, » tar Nie Ae) ~ vA Canin, oY - bi a Tes Leet ae Uiyaay ae Nansen aA Caer 5 Aldbourne Chasc Od “erie, S a NG . . Paar we Sas, a ’ aa Ry Whe . Sak Jegte is San Ce ht Beis Fs : Pa A ; a . ~~ = ’ - Be, ~ = > 2 = °, =: Membury « ry} se tay) : Ashton Keynes ae 7 CEA a sas . = R ey ae as ) ¥ OPP i = type gaunt ty? - ane , we oa iE gag rceayet % =3 if te ifs a : 53 = =i “ . S ms atrttiry, z << ‘ NS é ee bane an ¥ SS + x AN) : ei Sit, t> ‘ . yo? \y sv Ae rrr iene Sey WW y asad atte a8 Tory ss ee ae Sues tee Teed he SAN MIN ks ee yt ag A eS Lepr ee, Nise Sree Cee A PLAN U ada tal aaa! igs um ~ or = ati Rees BR le ~ oN ey oN Be SS a iers eee ey a oie ey one ae *, a7 6 Ls oN Sign, aa a ~*~, os Pan “4 PI fon ‘4, ee 47 a eS ar vis a pe try Pers 7. ar . ? ax Scie ees: =< 5 re ‘ 7 ‘se we = . = ~ vor a Reis es % = - ~ a ~> 7 FN a = ae Si ae = See ~ Y mee ~ = . ~ a Cy Gr «a ig ry = ye aypeyy era es . 2 ayy ‘ a i. Pay serttad pa way cos Ludgershall ’ vig - a Ok pes rea yys a Thyyyat®

0 200m

EARLY CASTLES INTHE MEDIEVAL LANDSCAPE OF WILTSHIRE 1g

The ringwork at Stapleford (SU 069379) is horseshoe-shaped in plan, due to its setting adjacent to the River Till, which deviates to skirt the west of the earthwork and feed its moated defences. Within the interior of the ringwork a series of indistinct scarps doubtless represent the vestiges of internal masonry structures, although they form no coherent plan (Offer and Hoare 1825, 22).The large right-angled earthwork enclosure appending to the north and west of the ringwork, enclosing an area of floodplain in excess of 2.5 hectares, appears too weak and extensive to represent a bailey enclosure with military or defensive functions. In terms of construction, this enclosure appears a secondary addition to the powerfully- defended ringwork. For instance, the western bank of the rectangular fishpond set within the east side of the enclosure continues to run along the northern counterscarp bank of the ringwork, implying that it is either contemporary with, or later than, the ringwork. The earthworks at Stapleford thus illustrate that a well- sited, lowland earthwork castle with room for expansion could continue to operate in a manorial capacity long after military needs had declined, hence the addition of the fishpond and suite of paddocks.

The earthwork known as Hall’s Close, Ashton Keynes (SU 049945) comprises an irregular embanked platform suggestive of a ringwork, adjoined by a single sub-rectangular bailey. Although the earthworks presently appear rather weak, the opening of a trial trench in 1959 revealed a substantial wall that surmounted the ringwork, and demonstrated the surrounding spring-fed ditch to be revetted with timber and lined with puddled clay (Knocker 1958- 60, 241-42; Wilson and Hurst 1960, 156). The material assemblage from the excavation included floor tiles, and both glazed and unglazed pottery of early-twelfth through to thirteenth-century date (Wilts. Arch. and Nat. Hist. Soc. 1974, 187; 1975- 76, 137).To the south of the ringwork can be identified the vestiges of linear earthworks that are suggestive of a series of former adjoining closes, thus adding to the impression that the site represents an intermediate stage between a ringwork and bailey and a moated - manor. None of this evidence contradicts the hypothesis that the Hall’s Close earthwork (also known locally as ‘The Battlefield’) is the site of the castle captured by King Stephen in 1139, said to be at South Cerney (Potter 1955, 62; Renn 1968, 314). Indeed, the form of the earthworks and excavated evidence are entirely consistent with a short-term fortification that was subsequently adapted as a manorial residence of the de Keynes household (Knocker 1958-60, 242).

An earthwork at West Dean (SU 257275), whilst undoubtedly originating as a medieval fortified site, has been modified radically as an ornamental feature within a formal garden setting. The present field monument comprises a circular entrenchment of c. 60m diameter with an interior raised little more than c. 2.5m. Given the large diameter of the earthwork, which argues against it originating as a motte, the most likely scenario is that the rampart of a ringwork has been flattened in the post- medieval period to form the raised bowling green recorded early in the nineteenth century (Master 1855, 242). The artificial reduction of a former rampart also seems likely given the dimensions of the surrounding ditch (c. 1.2m deep yet over 12m wide), which appears to have been substantially filled in. Immediately to the west, the flanks of the gentle ridge that the earthwork surmounts have been modified through the creation of a series of successive garden terraces. In part, these earthworks are almost certainly on the site of a bailey enclosure that formerly enclosed the parish church of St Mary’s (RCHM 1987, 119-210), thus demonstrating the juxtaposition of a seat of secular power with a private ecclesiastical foundation.

Norwood Castle, Oaksey (ST 985944), represents what has often been accepted as the ‘classic’ form of a motte and bailey, yet on a tiny scale (Wilts. Arch. and Nat. Hist. Soc. 1952, 227). The motte, little more than c. 1.5m in elevation, is adjoined to the north- west by a single bean-shaped bailey. This site is not to be confused with the earthworks south of the churchyard in Oaksey village (ST 992934) that indicate the former position of the defended manor of the Duke of Lancaster described by Aubrey in 1670 as ruinous (Jackson 1867, 298-99).

The motte and bailey at Castle Orchard, Stourton (ST 769319) is adapted from a natural promontory. A circular motte, supporting the base of a small keep, is isolated from an oval bailey by a rock-cut ditch; a subsidiary enclosure on lower ground to the north is defined by a linear rampart and ditch and a natural scarp running along the west bank of the upper Stour (VCH Som. II 1911, 517-18). Twice in the late nineteenth century, areas of the earthwork were sampled archaeologically during investigations into the origins of the ‘Pen Pits’ (see Lane-Fox 1879; Pitt- Rivers 1884; Winwood 1884). The excavations confirmed that a number of the pits were the by- product of quarrying for mill-stone, yet also sectioned the rampart at the extreme west end of the bailey, which was demonstrated to overlie such a pit (Lane- Fox 1879, 11), and showed the motte ditch to contain greensand rubble and to have silted to a depth of c. 4ft (c. 1.2m) (Winwood 1884, 150-51).

ge THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

The Castle Orchard motte and bailey must, however, be understood as one of a group of three closely-spaced early castles, two of which lie within Somerset. A complex earthwork castle at Cockroad Wood, Charlton Musgrove (ST 746323), and another known as Balland’s Castle, Penselwood (ST 753311), are, in association with the Castle Orchard site, disposed around Penselwood village in a triangle with sides of c. 1.7km, 1.2km and 2.1km. King (1983, xxix- xxx) has defined three potential reasons for the existence of multiple castles in a restricted area: the successive occupation of sites; the raising of a siege castle against another; or the sites having separate administrative histories.’To this model may be added the scenario of two or more sites existing contemporaneously within a unified strategy (Lewis 1989, 167). In the absence of any supporting documentation, or any archaeological dating of the sites, their interrelationship remains a matter for speculation. Topographically, however, the manner in which the sites are deployed, almost in mutual support, would favour the hypothesis that, at some stage, they formed elements within a coherent and unified programme of castle building. In seeking a possible historical context for these fortifications, it may be significant that a cluster of Domesday manors in the Penselwood-Bruton area of eastern Somerset show a marked reduction in value during the period 1066-86, presumably as a result of Norman subjugation in the wake of the 1069 rising against Montacute (Welldon Finn 1971, 289-90). Although the origins of these early castles may well owe to these events, they appear not to have been short-term fortifications: Balland’s Castle may have been associated with a small deserted medieval settlement and church; the Cockroad earthwork indicates the addition of a motte and bailey to a primary ringwork; and the Castle Orchard site was fortified with a buttressed rectangular keep (VCH Som. IT 1911, 513- 17; Aston 1982, 125).

Aspects of Pre-Castle Occupation (Figures 2-4)

The notion that a castle mound lies between Bishopstrow Farm and Bishopstrow House at ST 898443 (Cunnington 1949, 155; VCH Wilts. I(i) 1957, 160) is undoubtedly mistaken, and apparently based on confusion between the tradition of a castle in the village and the mis-identification of a mutilated round barrow. The identification of a complex of earthworks immediately east of Bishopstrow Farm (ST 901440) as a motte and bailey seems, however, correct.’

Although the earthworks, lying in an area known locally as ‘motte field’, are now almost totally mutilated by agriculture and partly obscured by buildings, aerial photographs reveal an enditched mound of c. 17m diameter at the centre of two successively larger enclosures. Limited excavation in 1981 has given further credence to the notion that the earthworks may represent, in part, a medieval fortified site, as a small trial pit in the north-west sector of the ‘motte’ ditch recovered two sherds of glazed pottery of probable twelfth-century date from the upper fill.t The wealth of residual iron age pottery scattered throughout the sub-soil in this test pit, combined with the excavation of a pit cluster of similar date immediately south of the mound? suggests, however, that the motte may be a short-term fortification sited within the extant earthworks of an iron age domestic complex (Scott and McOmish 1989, 103).

The suggestion that Si/bury Hill (SU 100685) is a Norman motte, erected de novo, has occasionally been made (e.g. Downman and Goddard 1919, 352). Whilst this is patently mistaken, excavation has provided tantalising evidence that the mound was modified as a defensible feature at an indeterminate date. The evidence centres on two enigmatic terraces at the summit of the mound: an upper terrace surrounding the summit entirely, and a lower, discontinuous, feature. It is unclear whether either or both terraces were topographical features in the Neolithic period or originated as the result of subsequent modification. The terraces were, however, vertically revetted with timber secured with iron nails at some stage; both were associated with Saxo- Norman wares, and a single coin of Ethelred IIT (AD 1009-16) was recovered in close association (Atkinson 1970, 314; 1978, 170; see also Whittle 1997, 22). The excavator was keen to interpret this evidence as a defensive feature raised against a Viking threat, yet the motte-like aspect of the defences cannot be denied. Indeed, subject to detailed re-interpretation of the ceramic evidence, there is no compelling reason to discount the possibility of Norman re-occupation of Silbury Hill as an expedient fortification or sentinel post overlooking the Roman road.

Excavation of the elliptical motte known as “The Mount’, Great Somerford (ST 963831), whilst limited and poorly recorded, has afforded remarkable insight into antecedent occupation on the site. In 1811, and again in 1910, the mound was opened, revealing substantial masonry remains at its core; these comprised a length of walling pierced by a doorway, and two windows of Romanesque form (Goddard 1930, 88-89).

EARLY CASTLES IN THE MEDIEVAL LANDSCAPE OF WILTSHIRE

\We Ve. ,

S '- NM

> Ay Sa \e ny aA 7 Vie

Great Somerford Ogbourne St Andrew — Hartham Park Littleton Pannell

Ze Lyneh : Stourto : ea yneham s E = ao Pe z ee Ss a . rai “4 vaca aN x witteaart yyy, rn AN) + ae oe ay Ml Lhe WE : (oe, oy s \ HWM, SSW oe Aura ws ~ Sy 7) I “s =) all . os Hasty , > = =: re ay sy" SS : rif aS S a : pyrtoyst a 7} 4 2) Aye :

ESI ini S4 7I\\\

“Fang gqaee’

aa AE eG SO NL AeA . Sous Q 7+ Ni Wit on SUL SSC

Bishopstrow Farm

4 ve " ny

vis ~ Pyenet®

“ye

an We =

NS

<

, Bream aan a ‘ , n . COD eT es ce

eye a i: Y4, SS eee ee A Son

. a reecrrened ne << & S Zi Z, Z Bs

ee

Broad Hinton

200m

Fig. 4. Wiltshire mottes

ls}

114 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

This evidence could be interpreted in one of a number of ways. Given the dimensions of the motte, the possibility that the earthwork is the product of rubble derived from the masonry structure consolidating and grassing over (cf. Middleton Stoney, Oxon.: Rahtz and Rowley 1984, 61), seems unlikely. Equally, the earthwork is of sufficient magnitude not to have constituted an earthwork abutment to a small keep (cf. Ascot Doilly, Oxon.: Jope and Threlfall 1959). Whilst feasible that the masonry remains indicate a manorial precursor to the castle site after the manner of Sulgrave, Northants. (Higham and Barker 1989, 50-51), the available evidence favours the notion that the motte has been imposed over a church site in order to overlook a ford over the Avon that lies immediately to the north. A church at Great Somerford is documented from the late twelfth century (VCH Wilts. XIV 1991, 202), and the present parish church of St Peter and St Paul, lying adjacent to the motte, in all probability represents a re- foundation. Although the date of the motte’s construction remains a matter for speculation, given the feature’s topographical position, it is entirely possible that it is one of the three siege works erected rapidly against Malmesbury Castle by Robert, Earl of Gloucester in 1144 in response to raids by the garrison (Potter 1955, 113). A number of other castle sites imposed upon and displacing parish churches appear to have origins as Anarchy-period fortifications, as at Eaton Socon, Beds. (Lethbridge and Tebbutt 1952), while the castle at Malmesbury itself encroached upon the abbey cemetery (Haslam 1976, 35).

Although the castle earthwork at Castle Combe (ST 839779) is listed occasionally as a motte and bailey (e.g. Palmes 1967, 24), in reality the site comprises an irregular ringwork containing the base of a small square keep, associated with a minimum of five bailey enclosures. The exceptional size of the site can likely be attributed to the re-use of prehistoric defences (VCH Wilts. IG) 1957, 264). Although this cannot be confirmed, the format of the perimeter defences is certainly consistent with a late prehistoric promontory fort that has been reconditioned and remodelled in the Norman period through the insertion of the ringwork, and sub-divided through the addition of a series of transverse earthworks. Recent survey of the castle has shown the outer banks of these internal earthworks to be stone-revetted and confirmed the existence of building foundations within the two inner baileys (Wilts. Arch. and Nat. Hist. Soc. 1992, 158; 1993, 159). The large outer enclosure appears to have remained free of structures and may have been dedicated to manorial functions

from an early date; the earthworks of two pillow mounds can also be recognised here. Although, somewhat puzzlingly, the castle has no conventional documentation, there is little doubt that it served as the caput of Castle Combe barony, most probably when the fee was consolidated by the de Dunstanvilles during the reign of Henry I, although an initial Anarchy-period fortification on the site cannot be discounted (see Scrope 1855, 134-36; Sanders 1960, 28).

Castles and Domesday (Figure 4)

No castles are mentioned in the Wiltshire Domesday: they were sources of expenditure as opposed to taxable assets and thus beyond the concerns of the commissioners. Domesday is, however, a much under- used resource for understanding many early castles in their appropriate context as seats of Norman administration in the immediate post-Conquest period (Pounds 1990, 10).The interpretation of land- holding patterns at Domesday can be problematic due to the difficulties of equating Domesday manors with present parishes/townships, and where a single settlement within the present landscape is subdivided manorially in 1086. Nonetheless, the Wiltshire Domesday suggests a circumstantial link between the pattern of land-holding in 1086 and the respective locations of early castles at Sherrington (ST 960392) and Bicknoll Castle, Broad Hinton (SU 108793). The motte at Sherrington lay centrally within a compact block of estates in the hands of the Norman magnate Osbern Giffard, comprising nine manors in Heytesbury, Branch and Dole hundreds.° The fact that Giffard held little land outside Wiltshire (aside minor holdings in Berkshire and Dorset), combined with an exceptionally low level of subinfeudation (of Osbern’s twelve Wiltshire manors only two were in the hands of sub-tenants, and neither lay within the compact estate), increases the confidence with which the motte at Sherrington can be identified as the caput of a small eleventh-century honour. The motte, artificially elevated c. 6m and with an embanked summit c. 28m in diameter, is of considerable size relative to other rural Wiltshire mottes, yet evidence of a related bailey is ephemeral. An archaeological evaluation commissioned during drainage operations has sectioned a substantial ditch, c. 25ft (7.6m) across, c. 110m north-west of the motte, which seems likely to represent the northern arm of the bailey defences (Wilts. Arch. and Nat. Hist. Soc. 1973, 137-38). In addition, the Sherrington enclosure map’ depicts that the pattern of roads to the south of the motte form a

EARLY CASTLES INTHE MEDIEVAL LANDSCAPE OF WILTSHIRE 115

conspicuous D-shape that embraces the parish church of St Cosmos and St Damian. This feature seems likely to fossilise the perimeter of a second bailey, or feasibly indicate the limits of a large single enclosure whose moated defences were doubtless fed by a diversion of the Wylye. Both the proximity of the manorial mill to the motte (a mill at Sherrington is recorded from Domesday), and the likely origins of the parish church as a castle chapel (a priest at Sherrington is recorded from 1130), indicate the status of the castle as a tool of Norman lordship and an expression of seigneurial power (VCH Wilts. XV 1995, 237-40).

The relationship between Gilbert of Breteuil’s manors at Bicknoll and his other estates within the shire indicates a similar pattern of teneurial geography. The core of Gilbert’s Wiltshire holdings comprised five apparently contiguous manors at Bicknoll, [Broad] Hinton and Clyffe [Pypard], all of which were held in chief.* Significantly, of Gilbert’s four other outlying Wiltshire manors (Beckhampton, Chisbury, Moredon and Stanmore), all but one were subinfeuded. Bicknoll Castle thus lay at the centre of the most concentrated and valuable block of Gilbert’s manors, which lay mostly, although not exclusively, within the shire. The physical development of Bicknoll Castle is related strongly to its topographical setting on a narrow chalk spur projecting north from the watershed between the Kennet and the Bristol Avon. A motte, c. 3m in elevation and much depleted by chalk quarrying, occupies the apex of the tongue of land, isolated from a trapezoidal bailey by a crescentic ditch. The bailey is defined by a curving embankment and ditch constructed transversely across the neck of the steep-sided promontory, and the merest vestiges survive of a second outwork that formerly defined an outer bailey (Goddard 1913-14, 213). Despite its present isolated status in the landscape, the castle formerly lay within the separate township of Bicknoll, where taxpayers and a chapel are recorded from the thirteenth century, and may have been associated with a small village or hamlet (VCH Wilts. XII 1983, 108, 114-15). The remains of a deserted settlement are visible as earthworks on the slopes immediately north - of the castle in Bicknoll Dip,’ whilst comparable earthworks, in association with medieval pottery, have been identified within the bailey enclosures.'°

The compact nature of these small fees is especially striking when viewed within the context of the large number of Saxon landholders from whose confiscated estates they were formed: Osbern’s twelve manors were in the hands of eight antecessors and Gilbert’s nine manors were previously held by seven separate Saxon landholders. These statistics create

the impression of élite strongholds imposed at the core of newly created blocks of estates. Further work elsewhere within England is revealing interesting patterns of regional variation in this respect; in certain areas Norman castles tended to establish new centres of seigneurial authority, yet elsewhere the trend was towards the perpetuation of late Saxon administrative arrangements (Higham 1999-2000, 7).

Yet by no means all compact 1086 estates in Wiltshire contained castles. The powerful magnate Ralph of Mortimer, for instance, held six apparently contiguous manors in the Grittleton-Hullavington area,'! yet his estates did not contain a castle. Here, the absence of a castle may be explained by the fact that he held considerable estates in eleven other shires. Conversely, not all early castles were related spatially to compact 1086 estates; Stapleford, for example, was the only Wiltshire manor held by Svein, one of the King’s thanes,'? whilst Stourton was one of two manors held by Walscin of Douai, subinfeuded to Ralf.'? The distinct possibility exists that such castles were constructed under the orders of sub-tenants as opposed to being temporary fortifications, especially as their size and complexity argues against the thesis that they are purely Anarchy-period foundations.

Possible Early Castle Sites (Figures 3 and 4)

A further issue in the field archaeology of early castles is the differentiation of isolated mottes bereft of bailey enclosures from earthworks of similar form yet different origin, such as barrows, windmill mounds and prospect mounds. Particular difficulties emerge, however, as the expedient nature of castle building ensured that earlier features could be adapted as the sites for castles, whilst the location of mottes and ringworks near settlements meant that many were often modified continuously after their military functions had expired.

Earthworks at Hartham Park (ST 858724) and Littleton Pannell (ST 999540) exemplify the problem of distinguishing genuine medieval mottes from post- medieval garden features. Both features are relatively tall (c. 3.5m and 4.5m high respectively), of conical appearance with very small flat tops, and have been identified as mottes or possible mottes (Rahtz et al. 1969, 17; King 1983, 502). However, the setting of these features within the immediate vicinity of post- medieval mansions, combined with the lack of a bailey in both instances, raises the distinct possibility that either or both may have originated as gazebos or viewing platforms erected in conjunction with schemes of formal garden creation.

116 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

The earthwork known as Clack Mount, Bradenstoke (ST 997793), may be a further example of a garden feature listed erroneously as a motte. The site has been described as a motte with an angular ward (King 1983, 499), yet the present field monument does not have the appearance of a fortified site. The mount, little more than c. 1.5m in height and c. 12m in diameter, is not circumvallated and lies entirely within a double-banked trapezoidal enclosure, with signs of a second enclosure to the south. The earthworks, previously described as the ‘site of a pleasaunce and a fishpond’ (Cong. Brit. Arch. Assoc. 1881, 146-47), are thus more consistent with a formal garden feature erected in the environs of Bradenstoke Abbey than a fortified site. However, the local place- name ‘Barrow End’ and derivation of the name ‘Lousy Clack’ (derived from the Old English hlaew: burial mound), raises the possibility that the mount is itself a modified barrow (VCH Wilts [IX 1970, 91; Lewis 1995, 190).

The identification of a low earthen mound within the churchyard of St Andrew’s, Ogbourne St. Andrew (SU 189723), as a motte, albeit a weak or mutilated example (King 1983, 500), seems equally specious. Excavation of the mound in the nineteenth century revealed a series of central cremations with intrusive Saxon and medieval inhumations (Cunnington 1885, 345-48). In addition, documentary evidence suggests that the feature served as a windmill mound in the post-medieval period (VCH Wilts. XII 1983, 147).A close spatial relationship between motte and parish church is a recurrent feature within the lowland zone of medieval Britain (Creighton 1997, 30-31), whilst the coincidence of barrow and church is not unknown (Morris 1989, 40-41, 255-58). Here, however, the insufficient magnitude of the mound (it is elevated little more than c. 1.5m) and lack of a bailey, combined with clear evidence of burials within the mound, confirm its origins as a round barrow and recommend strongly against the thesis that it was adapted as a motte.

The identification of the enigmatic earthwork south of Hillocks Wood, Lyneham (SU 026804) as a motte and bailey (VCH Wilts. IG) 1957, 267; [IX 1970, 91) can be rejected outright. Only vestiges of the earthwork survive, comprising two linear depressions meeting at right angles at the crest of a gentle north- facing ridge; these are almost certainly hollow-ways converging upon former farm buildings at the hill top, where a brick barn survives." To this list of extremely doubtful castle sites we may add the sub-rectangular univallate earthwork immediately south of the Iron Age hillfort at Membury (SU 305745). Excavated

c. 1941, a structure interpreted as a twelfth-century rectangular keep was revealed, overlain by a thirteenth-century house with a chapel (O’Niel 1948, 33). Although the earthen defences, now entirely denuded by arable cultivation, were supplemented with a single round tower, the setting of the site on a plateau with no natural defence is more consistent with a fortified manorial site as opposed to a twelfth- century castle.

CONCLUSIONS

This review of the evidence for Norman castle sites in Wiltshire has highlighted some of the problems involved in their study and indicated some potential avenues for future research.

It has been demonstrated that castle earthworks can be complex and multi-phase field monuments. The earthworks of an early castle site can potentially shroud antecedent phases of domestic, military or ecclesiastical occupation. Equally, however, a motte or ringwork could itself be adapted and remodelled in post-military phases. This study has also identified a number of earthworks that, despite traditional identification as early castle sites, may have alternative origins as post-medieval garden features. The previous mis-identification of such earthworks as mottes may well be a reflection of the period-based biases of archaeological fieldworkers, both past and present. Indeed, it is only relatively recently that modern archaeological survey has emphasised the ubiquity of formal garden earthworks (Everson and Williamson 1998, 139). Nonetheless, considerable difficulties exist in differentiating isolated mottes from post-medieval prospect mounds, as both classes of field monument tend to occupy similar topographical positions and are often found in the vicinity of medieval/post- medieval manor houses and halls.

Other regional surveys have demonstrated the shortcomings of a rigid classificatory approach to castle earthworks (Baker 1982, 38-39; Welfare et al. 1999, 60), and this study is not an exception. The reasons for differences in the form of castle earthworks are complex and interrelated and owe as much to post- abandonment sequences as to the original intentions of the castle builders. This study accepts tentatively, however, that the “human variable” may be the key determinant factor that explains the respective distribution of mottes and ringworks (King and Alcock 1966, 103). Whilst the decision to raise a motte as opposed to a ringwork, or vice versa, is ultimately a matter of seigneurial preference, the evidence from

EARLY CASTLES IN THE MEDIEVAL LANDSCAPE OF WILTSHIRE 117

Wiltshire highlights three important factors that also merit consideration. First, as a form of construction, the ringwork is more appropriate for the enclosure of extant structures or buildings, either in a time of crisis, or as a deliberate domination of an extant manorial centre as an act of usurpation and conquest. Second, the aggregation of ringworks and mottes in respective groups (e.g. the cluster of ringworks in the south- east of the county: Figure 1) may indicate the diffusion of a concept of fortification through competitive seigneurial emulation or unity of patronage. Third, the manner in which ringworks may have been economic fortifications in terms of time and labour points towards their employment as rapid and expedient forms of fortification (e.g. siege castles such as Cam/’s Hill, Malmesbury), especially where local geological conditions were less favourable for motte construction, as demonstrated in a recent study of early castles in Glamorgan (RCAHMW 1991, 34-36). It remains essential, however, to acknowledge that the classification of castle earthworks in this manner should be seen as an inherently limited tool of analysis that can only provide a platform for further detailed study. Indeed, one may question whether terms such as ‘ringwork’, ‘motte’ or ‘ringmotte’ would have had any real meaning in medieval terms; to contemporary observers, the visually striking aspect of these sites would surely have been their timber superstructures as opposed to the ground plans of their associated earthworks.

This study has also stressed the importance of integrating early castle sites within contemporary medieval landscapes. Unambiguously, the early castle was also an icon of seigneurial power. In an “imitative age” (Lewis et al. 1997, 231), when lordship was reinforced by mechanisms of patronage and display, even the most humble of earth and timber castles was a symbol of conspicuous seigneurial consumption as much as a military strongpoint. The particular functions of an early castle could vary, from garrison block (Cam’s Hill, Malmesbury) to quasi-palatial residence (Old Sarum); what is consistent is that the castle always represented the administrative, economic and coercive apparatus of land management and/or territorial control (see Creighton 1998). In particular, mottes or ringworks that were physically associated with parish churches or mills (e.g. Old Somerford; Sherrington; West Dean) emphasise that early castles were often integral components within the machinery of Norman lordship; these sites were hubs of manorial administration and ecclesiastical provision as well as centres of military power.

Acknowledgements

The author wishes to thank Dr N Christie (School of Archaeological Studies, University of Leicester) for his supervision of the MA and PhD theses from which this research is derived, and Dr Robert Higham (Department of Archaeology, University of Exeter) for his comments on an earlier draft of this paper. In addition, the staff at the Wiltshire and Swindon Record Office, and the Wiltshire County Sites and Monuments Record, Trowbridge, are thanked fer their assistance.

Notes

1 The Downman Collection, Wiltshire Archaeological and

Natural History Society Library, Devizes Museum Wilts. and Swindon Record Office: Marlborough Tithe

Award and Map, 1843

English Heritage Scheduled Ancient Monument No.

10211

Fieldwork Notebook 2 (29.1.1981): Wilts. SMR No.

ST94SW464

Wilts. SMR Nos. ST94SW201/202

Domesday i, 72d Wilts. and Swindon Record Office: EA/41

Domesday i, 71b-71c

Wilts. SMR No. SU17NW454

10 OS Record Card No. SUI7NW2

11 Domesday i, 72b-72c 12 Domesday 1, 74b 13 Domesday 1, 72a 14 OS Record Card No. SU 08 SW 5

Q bo

aS

woomonNnta uv

Bibliography

ADDYMAN, P.V., 1969, ‘Excavations at Ludgershall Castle, Wiltshire’, Chateau Gaillard IV, 9-12

ADDYMAN, P.V., 1973, ‘Excavations at Ludgershall Castle, Wiltshire, England (1964-1972)’, Chateau GaillardVI, 7-13

ASTON, M., 1982, ‘The medieval pattern 1000-1500AD’, in M. Aston and I. Burrow (eds), The Archaeology of Somerset. 123-33.Taunton: Somerset County Council

ATKINSON, R.J.C., 1970, ‘Silbury Hill 1969-70’, Antiquity 44, 313-314

ATKINSON, R.J.C., 1978, ‘Silbury Hill’, in R. Sutcliffe (ed.), Chronicle: Essays from Ten Years of Television Archaeology. London: BBC, 159-173

BAKER, D., 1982, ‘Mottes, moats and ringworks in Bedfordshire: Beauchamp Wadmore revisited’, Chateau Gaillard 9-10, 35-54

BERESFORD, G., 1987, Goltho: The Development of an Early Medieval Manor c. 850-1150. London: HMSO

118 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

BIDDLE, M., 1969, ‘Wolvesey: the domus quasi palatium of Henry de Blois in Winchester’, Chateau Gaillard 3, 28-36

BRENTNALL, H.C., 1945-47, ‘Lewisham Castle [Note]’, WAM 51, 472-473

BROWN, A.R., 1959, ‘List of English Castles, 1154-1216’, English Historical Review 74, 249-280

CARRINGTON, F.A., 1855, ‘Intrenchments at Aldbourne’, WAM 11, 126-129

CLARK, G.T., 1875, “The earthworks of the Wiltshire Avon: Old Sarum and Downton’, WAM 32, 290-309

COLVIN, H.M., BROWN, R.A. and TAYLOR, A.J. (eds), 1963. The History of the King’s Works, vol. 2. London: HMSO

CONGRESS OF THE BRITISH ARCHAEOLOGICAL ASSOCIATION, 1881, ‘Excursion on Thursday, August 19th’, WAM 19, 132-160

CREIGHTON, O.H., 1994, The Wiltshire Castle in the Landscape. Unpublished MA Thesis: University of Leicester

CREIGHTON, 0O.H., 1997, ‘Early Leicestershire castles: archaeology and landscape history’, Transactions of the Leicestershire Archaeological and Historical Society 71, 21-36

CREIGHTON, O.H., 1998, Castles and Landscapes: an Archaeological Survey of Yorkshire and the East Midlands. Unpublished PhD Thesis: University of Leicester

CUNNINGHAM, R.H., 1945-47, ‘Devizes Castle: a suggested reconstruction’, WAM 51, 496-499

CUNNINGTON, F.G:S., 1885, ‘Barrow at Ogbourne St Andrew, Wilts’, WAM 22, 345-348

CUNNINGTON, M.E., 1949, An Introduction to the Archaeology of Wiltshire: From the Earliest Times to the Pagan Saxons. Devizes: C.H.Woodward (4th Edition)

DOWNMAN, E.A. and GODDARD, E.H., 1919, ‘Plans of Wiltshire Earthworks’, WAM 40, 352

EVERSON, P. and WILLIAMSON,T., 1998, ‘Gardens and designed landscapes’, in. P. Everson and T. Williamson (eds), The Archaeology of Landscape. 139-165. Manchester: Manchester University Press,

EVERSON, P., BROWN, G. and STOCKER, D., forthcoming. ‘Chapter 5, The Castle Earthworks and Landscape Context’, in P. Ellis (ed.), Excavations at Ludgershall Castle. Devizes: WANHS Monograph no.2

GODDARD, E. H., 1913-14, ‘List of prehistoric, Roman and pagan Saxon antiquities’, WAM 38, 153-378

GODDARD, E.H., 1930, ‘The mount at Great Somerford’, WAM 45, 88-89

GRAHAM, A.H. and DAVIES, S.M., 1993, Excavations in Trowbridge, Wiltshire, 1977 and 1986-1988. Salisbury: Wessex Archaeology Report no. 2

HASLAM, J., 1976, Wiltshire towns: the archaeological potential. Devizes: Wiltshire Archaeological and Natural History Society

HIGHAM, R.A., 1982, ‘Early castles in Devon, 1068-1201’, Chateau Gaillard IX-X, 101-116

HIGHAM, R.A.,1999-2000, ‘Castle studies at the Leeds International Medieval Congress’, Castle Studies Group Newsletter 13, 7

HIGHAM, R.A. and BARKER, P.A., 1992, Timber Castles. London: Stackpole

JACKSON, J.E., 1867, ‘Ancient chapels and churches in the county of Wiltshire’, WAM 10, 253-322

JOPE, E.M. and THRELFALL, R.I., 1959, ‘The twelfth- century castle at Ascot Doilly, Oxfordshire: its history and excavation’, Antiquaries Journal 39, 127-131

KING, D.J.C., 1972, “The field archaeology of mottes in England and Wales: eine kurze ubersicht’, Chateau Gaillard 5, 101-112

KING, D.J.C., 1983, Castellantum Anglicanum, 2 vols. Millwood: Kraus

KING, D.J.C. and ALCOCK, L., 1966, ‘Ringworks of England and Wales’, Chateau Gaillard 3, 90-127

KNOCKER, G.M., 1958-60, ‘Hall’s Close, Ashton Keynes’, WAM 57, 241-242

LANE-FOX, A.H., 1879, ‘Report of Pen Pits Exploration Committee’, Proceedings of the Somersetshire Archaeological and Natural History Society 25, 7-17

LETHBRIDGE, T.C. and TEBBUTT, C.T., 1952, ‘Excavations on the castle site known as “The Hillings” at Eaton Socon, Bedfordshire’, Proceedings of the Cambridgeshire Antiquarian Society 4, 48-61

LEWIS, C., 1989, ‘Paired mottes in East Chelborough, Dorset’, in M. Bowden, D. Mackay and P. Topping (eds), From Cornwall to Caithness: Some Aspects of British Field Archaeology, Papers Presented to Norman V. Quinnell, 159-173. Oxford: BAR no. 209

LEWIS, C., 1995, ‘Pattern and process in the medieval settlement of Wiltshire’, in M. Aston and C. Lewis (eds), The Medieval Landscape of Wessex. 171-193. Oxford: Oxbow

LEWIS, C., MITCHELL-FOX, P. and DYER, C., 1997, Village, Hamlet and Field: Changing Medieval Settlements in Central England. Manchester: Manchester University Press

MASTER, G.S., 1855, ‘Collections for a history of West Dean’, WAM 22, 239-317

MORRIS, R., 1989, Churches in the Landscape. London: Dent

MUSTY, J., 1966, ‘Old Court, Downton, and the Moot earthworks’, WAM 61, 98-99

OFFER, J. and HOARE, R.C., 1825, The History of Modern Wiltshire. Volume I: Hundred of Branch and Dole. London: Nichols and Nichols

O’NIEL, B.T., 1948, ‘War and Archaeology in Britain’, Antiquaries Journal 28, 20-44

PALMES, J. W., 1967, ‘Castle Combe’, Archaeological Review 2, 24

PITT-RIVERS, A.H.L.F., 1884, Report on Excavations at the Pen Pits, near Penselwood. London

POTTER, E.R. (ed.), 1955, Gesta Stephani, the Deeds of Stephen. London: Nelson

POUNDS, N.J.G., 1990, The Medieval Castle in England and Wales: a Social and Political History. Cambridge:

EARLY CASTLES IN THE MEDIEVAL LANDSCAPE OF WILTSHIRE 119

Cambridge University Press

RAHTZ, P.A., 1964, ‘Saxon and medieval features at Downton, Salisbury’, WAM 59, 124-129

RAHTZ, P.A. (ed.), 1969, ‘Norman Castles’, in Medieval Sites in the Mendip, Cotswold, Wye Valley and Bristol Region. 14-17. Bristol Archaeological Research Group, Field Guide no. 3

RAHTZ, P. and MUSTY, J.W., 1957, ‘Excavation at Old Sarum’, WAM 57, 359-364

RAHTZ, P. and ROWLEY, T., 1984, Middleton Stoney: Excavation and Survey in a North Oxfordshire Parish, 1970-82. Oxford: Oxford University Press

RENN, D., 1959, ‘Mottes: a classification’, Antiquity 33, 106-112

RENN, D., 1968, Norman Castles in Britain. London: Baker

RCAHMW 1991, An Inventory of the Ancient Monuments in Glamorgan, vol. 3, Part 1a. London: HMSO

RCHM 1980, Ancient and Historical Monuments in the City of Salisbury, vol. 1. London: HMSO

RCHM 1987, Churches of South-East Wiltshire. London: HMSO

SANDERS, L_J., 1960, English Baronies, a Study of their Origin and Descent 1086-1327. Oxford: Clarendon Press

SCOTT, D. and McOMISH, D., 1989, ‘Non-Hillfort Settlement and its Implications’, in M. Bowden et al. (eds), From Cornwall to Caithness: Some Aspects of British Field Archaeology, Papers Presented to Norman V. Quinnell. Oxford: BAR No. 209, 99-111

SCROPE, G.P., 1855, “The Manor and the Ancient Barony of Castle Combe’, WAM 2, 133-158

SHORTT, H., 1947, “The Moot, Downton’, Archaeological Journal 104, 166

SQUAREY, E.P., 1906, The Moot and its Traditions. Salisbury: Bennet

STEVENSON, J.H., 1992, ‘The castles of Marlborough and Ludgershall in the Middle Ages’, WAM 85, 70-79

THOMSON, T-.R., 1958-61, Materials for a History of Cricklade. Oxford: Oxford University Press

VCH 1906-92, Victoria History of the County of Somerset, Vols 1-6. Oxford: Oxford University Press

VCH 1957-95, Victoria History of the County of Wiltshire, Vols 1-15. Oxford: Oxford University Press

WELFARE, H., BOWDEN, M and BLOOD, K., 1999, ‘Fieldwork and the castles of the Anglo-Scottish borders’, in P. Pattison, D. Field and S. Ainsworth (eds), Patterns of the Past: Essays in Landscape Archaeology for Christopher Taylor. 53-60. Oxford: Oxbow

WELLDON FINN, R., 1971, The Norman Conquest and its Effects on the Economy: 1066-86. London: Longman

WHITTLE, A., 1997, Sacred Mound, Holy Rings. Silbury Hill and the West Kennet Palisade Enclosures: A Later Neolithic Complex in North Wiltshire. Oxford: Oxbow

WILSON, D.M, and HURST, D.G., 1960, ‘Medieval Britain in 1959’, Medieval Archaeology 4, 134-165

WANHS 1952: Notes. WAM 54, 226-228

WANHS 1973:Wiltshire Archaeological Register for 1972. WAM 68, 126-139

WANHS 1974: Wiltshire Archaeological Register for 1973. WAM 69, 183-187

WANHS 1975-76: Wiltshire Archaeological Register for 1974-75. WAM 70-71, 132-138

WANHS 1992: Excavation and Fieldwork in Wiltshire, 1990. WAM 85, 1992, 156-162

WANHS 1993: Excavation and Fieldwork in Wiltshire, 1991. WAM 86, 1992, 158-164

WINWOOD, H.H. 1884. ‘The result of further excavations at Pen Pits’, Proceedings of the Somersetshire Archaeological and Natural History Society 30, 149-153

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 120-30

Looking for Dr Ingen Housz

The evidence for the site and nature of the burial, in Calne, of the famous Dutch physician and scientist of the eighteenth century by Norman and Elaine Beale

Dr Jan Ingen Housz, the famous physician and scientist whose experiments, in 1779, revealed the mysteries of photosynthesis, spent his final days in Wiltshire. He died during the early hours of 7 September 1799, at Bowood House. His funeral was organised by his long-standing friend and final host, the First Marquis of Lansdowne (who was 1n residence at Bowood House when the doctor died) and took place on 9 September 1799. That Dr. Ingen Housz was buried 1n Calne 1s confirmed by the parish register. It has long been assumed that he was buried in Calne St. Mary’s churchyard and that the site of the grave and its markings have been lost. Evidence Is presented here that his remains were deposited, within a lead coffin, 1n a vault under the church. The burial vault must have been already in existence and was owned by the Marquis of Lansdowne. We suggest that the relevant vault was that which ‘belonged’ to Castle House, Calne and discuss the evidence. The ‘Castle House vault’ was almost certainly under the chancel, probably to the north side but the vaults, crypts, and graves under the chancel of St. Mary’s Church, Calne appear to have been filled in during the mid-Victorian re-ordering of the building. Complete re-flooring of the Chancel destroyed all the ledger stones and vault entrance traps. It 1s suggested, however, that the coffins beneath the chancel were left in situ during such backfilling and that the mortal remains of Dr. Ingen Housz still lie under the chancel of St. Mary’s Church, Calne. This conclusion could only be verified by a formal archaeological exploration if the opportunity ever arose.

INTRODUCTION BRIEF BIOGRAPHY OF DR. INGEN HOUSZ Where notable people were buried is sometimes unknown. There is, for instance, no means of locating, Jan Ingen Housz was born on 8th December 1730 exactly, the site of the grave of Mozart. On the other in the small town of Breda, south Holland. The hand there are many, many sepulchral monuments, second son of a leather merchant and apothecary, often imposing and exuding self-importance, that he showed an outstanding talent for the classical mark the graves of people about whom nothing is languages at school. He trained to be a doctor at the known. Then, for some people, we have lost both their University of Louvain, graduating MD in July 1753.! personal heritage and their grave. Such a case is Jan His unusually varied postgraduate training in Ingen Housz, an important eighteenth century figure Leyden, Paris and Edinburgh was fostered and who deserves to be better-known and remembered. encouraged by Sir John Pringle, the famous Scottish Evidence is assembled in this paper, most of it physician’ who happened to be a friend of the Ingen previously unpublished, for the nature and location Housz family. of the final resting place of Ingen Housz. The Ingen Housz established a medical practice in conclusions will contradict the popular notion that Breda while keeping up his interest in chemistry and he was buried in the churchyard of Calne and that physics, especially in static electricity. After his father the markings of the grave have been lost, like those of died in 1764 he joined Pringle in London where his so many souls over the centuries. host and mentor introduced him to many of the

3, Main Road, Cherhill, Calne, Wilts SN11 8UX

LOOKING FOR DR INGEN HOUSZ

leading literary, scientific and political figures of the day. Important friendships developed — with Doctors William and John Hunter, with Joseph Priestley,’ and especially with Benjamin Franklin and the second Lord Shelburne. It was from Dr William Watson that Ingen Housz learned the technique of inoculation against smallpox,’ proving to be an outstandingly safe exponent. He was asked to go to Vienna to inoculate those Habsburgs who had not succumbed to the 1767 smallpox epidemic. The Empress, Maria Theresa, was so grateful that the devastation of her family was curtailed that she gave the young doctor a substantial pension for life and made him one of her personal physicians.

Financially secure, Ingen Housz was able to spend much of the rest of his life performing scientific experiments, and in travelling around Europe visiting other scientists. He was elected a Fellow of the Royal Society in 1769? and was later to serve on its Council. He married the sister of Nikolaus Jacquin, Professor of Botany at Vienna, but the marriage was late in both their lives and not emotionally successful.°

In England in 1779 Ingen Housz ensconced himself in a country house near London and began a marathon of over 500 experiments by which he unravelled the processes involved in what is now known as photosynthesis. He was the first to demonstrate that it is the green parts of plants that absorb carbon dioxide in sunlight and produce, at the same time, our oxygen; and that the carbon element becomes, eventually, more plant matter and ultimately our food.’ In later years he also proved that different metals conduct heat at different rates, developed the oxygen mask for clinical use, and invented the cover slip for microscope slides. When the French Revolution broke out in 1789 he was in Paris. Joint experiments with Lavoisier had to be forgone. As doctor to the Queen’s brother,* Ingen Housz had to flee. He travelled north to safety in England, but also into exile.

THE DEATH OF INGEN HOUSZ

- Ingen Housz never returned to Vienna. Although he recovered from the gall stones and kidney stones that had been troubling him on arrival in England in late 1789° he lost his appetite for travel and the spread of republicanism and successive wars on the continent put him, officially a Royal Courtier, in personal danger in several countries.'° His long-standing friendship with Lord Shelburne! was re-kindled and Ingen Housz spent long and happy periods, during the 1790s, at Bowood House as a guest.'? The Bowood ‘laboratory’

Figure 1. Ingen Housz in 1769 (Cunego, Rome, by courtesy of the Trustees of the Bowood Collection)

developed by Priestley in the 1770s and in which oxygen had been discovered, saw a new lease of life.

Ingen Housz was at Bowood in early Spring 1799, anxious that he could have a new and potentially fatal illness. He travelled to Bath to consult his friend, the physician Dr. William Falconer’? who must have confirmed Ingen Housz’ worst fears for, on returning to Bowood, the Dutch physician announced that he had no intention of becoming a nursing burden to the household and proposed to return to London. The Marquis persuaded him to stay ‘...at Bowood, surrounded by people he knew and where he was sure to find friendly care and sincere concern ...’' rather than suffer a lonely demise in rented rooms in London. So Ingen Housz and his manservant of 30 years — Dominique Tede!’ — spent what was to be their last summer together, at Bowood.

By June 1799 Ingen Housz knew that he was dying. He wrote a last letter to Vienna'® in which he expressed his desire that his wife could be with him to help to nurse him, although they had now been apart for over 11 years. Despite deteriorating physically and having a very troublesome dryness of the mouth and throat!’

122 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Ingen Housz still socialised and ate with the Marquis, his family and guests. As popular as ever with the ‘Bowood company’, Ingen Housz continued" to demonstrate his experiments.!’ Less happily, he also discussed with Lord Lansdowne, seemingly quite frankly and in some detail, his wishes for his funeral, future arrangements for Dominique, and the settlement of his other affairs.*°

On Wednesday 4 September 1799 he dined at table for the last time.*! During the next day he became too weak to leave his room on the first floor of Bowood House and on the 6thhe lapsed into a coma. He died, without regaining consciousness, at 3.30 a.m. on 7 September 1799.**

THE FUNERAL OF INGEN HOUSZ

The Marquis of Lansdowne organised the funeral of his friend ‘...quietly ... as he had wished ...’”’ for Monday 9 September 1799. His Lordship wrote to Ingen Housz’ widow, Agatha (née Jacquin), and his nephew, Josef Jacquin, both at Vienna. He also dictated a letter to the nephews at Breda,” enclosing it in a further letter that he sent to Ingen Housz’ bankers in London — Drummonds of Charing Cross.” The bank relayed the sad news, by means of letter and enclosures, to the family at Breda, informing them also of the will in their possession and of its attached instructions. It is from this, the ‘Drummond’ letter,”° that the details of the funeral emerge.

The burial register for Calne*’ records that Ingen Housz was buried on 9 September 1799.** The entry is corroborated by a certificate of burial’? signed by Thomas Greenwood, Vicar of the parish, and by Samuel Viveash and William Savory, Guild Stewards of the Borough of Calne.*° This was issued on 14 November 1799, its signing witnessed by Henry Maundrell, a local attorney at law, who took it to London and swore an affadavit to its veracity in the presence of Counsel.*!

The ‘Drummond’ letter’’ reveals that the mourners present at the funeral were few but mostly important. They were: Lord Henry Petty representing his father the Marquis of Lansdowne who was ‘...not well enough to go himself...’; The Reverend Mr. Dumont; the Guild Stewards of Calne; ‘...the medical people who attended him...’ and his manservant Dominique Tede. But the letter also contains information on the site and nature of the burial of Ingen Housz for it reads ‘...that he was interred in the Church of Calne ...’. [authors’ italics].

32

Figure 2. The First Marquis of Lansdowne (1737-1805). (E. Bartolozzi after Gainsborough, by courtesy of the Trustees of the Bowood Collection).

EVIDENCE THAT INGEN HOUSZWAS LAID IN A VAULT IN ST. MARY’S CHURCH, CALNE

Besides the explicit statement in the ‘Drummond’ letter, there is other evidence, both factual and circumstantial, that Ingen Housz was placed in a vault®? underneath St. Mary’s Church, Calne. The pointers are as follows. Firstly, in the will made by Ingen Housz in London in 1796, is the following phrase: ‘...I desire to be Buried in the Church of the Parish in which I will die...’. Ingen Housz obviously felt that he was an important enough person to merit burial within the walls of any church rather than in the churchyard even if he was not a resident parishioner and contemporary burial practices in Vienna, where he had lived for some years, would have reinforced this view.*! Secondly, the 1799 burials list for Calne shows, specifically, for September, the entry: Dr Jno Ingenhousz. The prefix ‘Dr’ is an important piece of evidence — only notable members of communities were customarily recorded with such titles by the clergy of the late eighteenth century.

LOOKING FOR DR INGEN HOUSZ

Concomitant with this entry format signifying high social rank would be the permission for burial inside the church. As Litten states, ‘... The use of a prefix in a burial register is of importance. In the majority of instances those afforded a prefix were considered as of gentle birth and merited burial/deposit within the building. The normal rule-of-thumb is “prefix equals preferred burial”...’*°. Thirdly, there is the fact that the bank ledgers recording Ingen Housz’ account at Drummonds have survived and those for 1799 show an entry, on 24th September, for funeral expenses, of £25 2s. 8d. This figure is corroborated, apart from a minor discrepancy of 4d., by an entry, on the same day, in the Day Book of Mr. Cross, Bowood Agent of the First Marquis, of ‘...Bills for Dr. Ingenhouz’s funeral £25. 2s. 4d...’.*° This is asum of money which is significantly larger than the known costs of an average funeral in the late-eighteenth and early- nineteenth centuries. Litten, records*’ that as late as 1838, a finished elm coffin with inscription plate, handles, lining and pillow cost only 17 shillings. On the other hand: ‘...an Elm shell, covered with 4lb lead, lined, ruffled, and pillow...for the grandest vault.:.would have been sold to the client for about £25 ...’° It may be concluded therefore, that Ingen Housz, normally a very thrifty man but yet more so in the last decade of his life when his pension was not regularly paid and he was heavily taxed, spent such a large sum of money only as necessary. There is no evidence of £25 being spent on an extravagent wake and we know from his will it is evident that he wanted his “...ffunerals ...to be simple and of little expense...’ Therefore Ingen Housz was almost certainly buried in a lead coffin — at an expense that could only be justified if his destiny was deposition in a vault rather than interment in a burial ground.’ Fourthly, the Drummond letter provides clinching evidence that Ingen Housz’ coffin was placed in a vault inside St. Mary’s Calne, since it continues: ‘...that he was interred in the Church of Calne in a Vault of his Lordship’s...’. {authors’ italics]. Vault burial in England in the eighteenth century and the location of a vault owned by the Marquis of Lansdowne under St. Mary’s - Church, Calne are obviously the next discussion points. The exact wording of the letter — ‘... a Vault of his Lordship’s ...’is a highly pertinent clue to locating it.

VAULT BURIAL WITHIN ENGLISH CHURCHES, 1650 — 1850

The peak time for constructing family vaults for burial under churches in Britain was from about 1650 until legal strictures ended the practice soon after 1850. In

123

fact the earliest recorded burial within St. Mary’s Church, Calne is 1598.*° During the next two and a half centuries few churches were not undermined to meet the wishes of ‘important’ parishioners who desired burial places ‘worthy’ of their status. Recent demolition and archaeological investigations of some inner city churches, where the parishes were very densely populated, have provided dramatic views of the subterranean overcrowding of such buildings.

But even churches in smaller, market towns such as Calne contained, by 1800, many vaulted chambers under their floors as well as intramural graves as so many church wall monuments evince with the words

. In a vault near this place...’. In some churches the wardens kept good records of the positions, dimensions, ownership and occupancy of vaults and intramural graves and in some instances these have survived.?! From the monumental inscriptions surviving in Calne church as at 1898*" we know that there must have been at least 77 vaults and graves within the building and its porches but there seems to be no surviving map locating them.

In fact, ownership of burial vaults became linked, like that of the pews above, to the substantial houses of a parish. Just as families of consequence bought and sold estates with the advowson and the named box pews where they would worship, so they sometimes bought rights to burial in particular vaults underneath the church.”

The quid pro quo of vault deposition was, as discussed above, that the corpse would be encased in a ‘triple’ coffin. Various designs for the inner lead shells were in vogue in different localities and at different times. By the second quarter of the eighteenth century the typical coffin, shell and case provided for vault deposition comprised an inner coffin of wood encased within a lead shell. This in turn was usually placed into an outer wooden case, often upholstered with velvet, and on to which was affixed the metal coffin furniture. Outside the metropolis, and certainly in places like Calne, the local plumber would usually be engaged to provide and fashion the sheet lead and to seal the plumeous layer hermetically with lead solder. Such additional and bespoke preparations for burial were time-consuming, expensive and made the coffins very heavy. Access to vaults was usually via a stone slab, possibly an inscribed ledger stone. Manoeuvring a four to five hundredweight coffin into a dark confined space was a challenge for the undertakers; subterranean reordering and relocation within the vault were not uncommon practices in order to accommodate the latest occupant. Triple coffins can, though, survive intact for a very long time and some

124 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

have been found to be in excellent condition as much as 200 years later. Although they may become dislodged, depositum plates (lead, brass, stamped iron, tin, or pewter) recording identification marks and dates (usually) of the occupant are often perfectly legible. Alternatively, identification and other details chased into outer lead casings themselves can also be well preserved.

IDENTITY OF THE VAULT, IN ST. MARY’S CHURCH, CALNE, OWNED BY THE MARQUIS OF LANSDOWNE.

Surprising though it may appear, the location of the vault or vaults owned by the First Marquis of Lansdowne within St. Mary’s, Calne is not known. The lapse of only two days between the death of Ingen Housz and his funeral must mean that there was a

pre-existing vault. But hunting for the ‘Bowood’ or ‘Lansdowne’ vault is fallacious; there never was one. At least, there is no evidence that any of the Shelburne/ Lansdowne family have ever been buried at St. Mary’s, Calne. The widow of John, the first Earl, built a family mausoleum on high ground above Bowood House where she and her husband are entombed." Other family members were either laid there; in the dynastic vault under the Church at High Wycombe, where the family also lived; at Bremhill; or, later, at Christ Church, Derry Hill, consecrated in 1840. Lateral thinking implies that the burial place of Ingen Housz must have been a vault in St. Mary’s Church that happened to be, in 1799, in the ownership of his friend, the First Marquis of Lansdowne, perhaps simply by chance. What we should be looking for is *...a Vault of his Lordship’s...’ and not ‘his Lordship’s vault’ i.e. vault of the Lansdowne family.

Figure 3. Interior of St. Mary’s Church, Calne, 1846, seen from the Chancel. (Courtesy of Mr Ray Downham MBE.)

LOOKING FOR DR INGEN HOUSZ

The purchase of Bowood Park by John, the First Earl of Shelburne was completed on 1 January 1754.” The previous owners — the Bridgeman family — had been assigned a mortgage on the property in February 1722. There is no evidence that the Bridgemans ever acquired property in Calne itself but they must have worshipped at St. Mary’s Church for, on 21 January 1754, the following entry appears in the Churchwardens’ Minutes: ‘...At a vestry held in the Parish Church of Calne it is unanimously agreed ...that the Seats and Gallery...which formerly belonged to Sr. Orlando Bridgeman Bart. dec’ed be forthwith granted to the Right Hon’ble John Earle of Shelbourn and his Successors for ever Owners of Bowood House... in consideration of twenty Guineas...’.*° There are, however, no references to an advowson related to Bowood ownership, no references to a burial vault, and no Bridgeman memorials in the church.*””

After the first Earl of Shelburne died in 1761, Bowood was inherited by his son, William, the second Earl. He continued the improvements and extensions of Bowood House and Park, retiring from politics temporarily between 1763 and 1765" in order to have the time to supervise the works (and similar activities at his High Wycombe and London properties). During the next forty years he also acquired substantial tracts of property in Calne and it is possible to show that he thereby acquired a superfluous burial vault in St. Mary’s Church. It is equally possible to identify this vault and its likely location within the building.

In the will of Richard Stokes, ‘Gentleman’ of Calne, written in 1723 is the following instruction to executors: ‘...my body to be buryed ...in a private and decent manner and to be laid as close to my Late Dear Wife as possible it can be in the vault in Calne Church which belongs to my Dwelling house...’. Moreover, in 1713, the churchwardens of St. Mary’s had issued a demand for a fee to be paid to them, by Mr. Stokes for ‘...opening of the vault...’. Stokes had, apparently ‘...lately had occasion to lay a child init...’. But the demand for payment begins with a useful and pertinent description of the vault itself: ‘... To the - house in which he lives, belongs a vault which was made by Mary Norborne widdow decs. (upon the death of her husband Walter Norborne Esq.) between 50 and 60 years ago, ... in which vault there now lies the body of the said Walter Norborne, and one of his daughters...’.?”

In other words it would appear that Richard Stokes acquired a burial vault that had been constructed by Mary Norborne to receive the remains of her husband. The demand for payment by the Churchwardens

1125,

clearly implies that the vault was considered to be part of the domestic property then owned, in Calne, by Richard Stokes and that he was in the habit of placing deceased relatives within.

On 20th April 1650 Walter Norborne had agreed and signed, with the widowed daughter of a Temys Jordan of Calne, a ‘...Feoffment of 2 messuages , one called the Castle in Castle Street, Calne ...’.°° Walter Norborne died in 1659 and it is not clear what then happened to his properties but on 23 and 24 June, 1708 Richard Stokes bought, from the Viscountess of Hereford, a‘... messuage called the Castle, in Castle Street, Calne (Part of the same property as in a Feoffment of 1650, April 20th)...’.?! It is clear, therefore, that the ownership of Castle House (as it is now known) passed to Richard Stokes, who was actually the grandson of Mary Norborne, and that this must be the *...house in which he lives...’ and to which ‘...belongs a vault...’ as referred to by the St. Mary’s Churchwardens in their demand for payment.”

Richard Stokes, shortly before his death, disinherited his elder son for disobeying his bar on marrying a cousin. The son joined a regiment of the East India Company’s army” and left England for the orient. Judith Stokes, the widowed second wife, appears also to have left Calne and leased Castle House to John Bull, Gentleman,” an important Calne citizen serving as Guild Steward in 1735.” During the 1750s and 1760s he acted as Steward to the Duckett family” and helped Thomas Duckett sell, in 1764, the manor of Calne and Calstone to the Second Lord Shelburne for £27,000.’ The Bull family, presumably now owners of the property, remained at Castle House until 1792 when it was bought by the Fripp family.** Then, for 1798, ‘The Calne Churchwardens and Vestry Book, 1795-1824’ clearly shows that Castle House had been bought by the First Marquis of Lansdowne and that Sir Geo. Colnbrook was occupying the house, stables and gardens as his Lordship’s tenant. But the same Vestry Book also shows that on 15 March 1799 the Marquis of Lansdowne was both the owner of Castle House and also the ‘...occupier...’. There was now no tenant and since the Marquis himself lived at Bowood House it may be assumed that, except for servants keeping the property aired and secure, the house was empty. This remained the situation until a tenant was listed in 1806.°° What is particularly important is that the First Marquis of Lansdowne had acquired Castle House and that it was in his ownership in 1799 when Ingen Housz died. He must also have acquired what might be called the ‘Castle House vault’ in St. Mary’s

126 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Church. This had been built by Mary Norborne in about 1660, and contained only, as far as we know, the remains of Walter Norborne and his daughter; Eleanor Stokes, a Stokes child, and Richard Stokes, the Bulls and Fripps having their own vaults/graves in the nave of St. Mary’s.°!

Thus when Ingen Housz was buried in 1799 the Castle House vault in St. Mary’s Church, Calne appears to have been redundant for many decades. It is suggested that this vault was the one in which the remains of Dr. Ingen Housz were deposited on 9 September 1799. Admittedly the evidence is circumstantial but the contention can be supported by negative evidence. All of the 23 other houses and cottages in Calne owned by Lord Lansdowne in 1799 were occupied by named tenants. Castle House appears to have been, in 1799, the only Calne property both endowed with a burial vault and for which burial rights were then at the disposal of Lord Lansdowne himself. Responding to the wishes of his deceased friend by placing his coffin in this vault did not mean, for the Marquis, that he must embarrass the sepulchral privacy of his own or any other family and, presumably, there must have been remaining space within the chamber for at least one more coffin.

POSSIBLE LOCATION OF THE BURIAL VAULT CONTAINING INGEN HOUSZ WITHIN ST. MARY’S CHURCH, CALNE

The next question must be the location of the vault. Here there is no decisive evidence. The obvious expected marker — a ledger stone or memorial tablet to Ingen Housz on the church floor or wall respectively and near to the putative vault — was either erected and later removed or was never realised. There are several reasons to suspect the latter. Although the Marquis of Lansdowne wrote, on 9 September 1799, in his letter to the family in Holland that ‘*... I reserve the right to erect a simple monument on his grave...’ it is more than possible that circumstances led to its never being commissioned. Firstly, Lord Lansdowne was ill. Unable to attend the Ingen Housz funeral himself, he had, within a few weeks, rented a house in Bath, ‘taking the waters’ in an attempt to restore his vigour.’’ Unfortunately his health declined further and in a letter to his ‘sisters’ [sic] at Bowood (Caroline Fox and Elizabeth Vernon) in May 1800 he made clear reference to having lost the function of a hand.°’ Clearly his illness had prevented him

being in London for the winter ‘season’ of 1799— 1800 and he was therefore unable to meet the two Ingen Housz nephews from Breda who came over to the capital in late November 1799 to prove their uncle’s will. Lansdowne did write to them, however, inviting them to use, if necessary, the services of his personal solicitor, Mr. Smith of Drapers’ Hall, but made no mention of any memorial. The nephews returned directly to Holland, having had difficulty obtaining passports in any case.°! Perhaps it is not surprising that Lansdowne, confronted by illness and by growing financial problems,” lost the impetus to erect a memorial to his friend. In any case he was, by this time in his life, not a supporter of sepulchral ostentation.

In a letter to the editor of The Gentleman’s Magazine, published in May 1791°° Lord Lansdowne had decried the adulteration of many English churches by excessive and unnecessary monuments to the dead. His ‘Enlightened’ perspective in this respect was likely to have been discussed with Ingen Housz to whom it was certainly nothing new. The physician’s close friendship with his patient, Emperor Joseph II, would have exposed him, fifteen years earlier, to similar beliefs. In Emperor Joseph’s view ‘...it was wrong to glorify a dead man’s corpse, the empty husk of the spirit...’.°’ But consensual or not, it is certainly the case that Lansdowne now held intellectual reservations about sepulchral monuments that would have presaged the practical difficulties when it came to establishing a memorial to his friend.

Direct evidence for the location of the Castle House vault may be missing but there are three clues in the guise of other surviving monumental inscriptions. Firstly, the memorial tablet for Eleanor Stokes, the first wife of Richard Stokes, who died of smallpox in October 1705 is a very large monument inscribed in Latin. It is high on the east wall® of the north chapel of the Chancel (see ‘S’ on the pre-1864 floor plan of St. Mary’s Church, Figure 4) and is currently obscured by the organ.°’ The Latin text is headed ‘HI S Q’ -hic 1acet sepultus quidam — here lies buried a certain. . .. However, this is not the only inscribed reference, within St. Mary’s Church, to the Castle House vault. There is, secondly, a memorial tablet to Walter Norborne high on the west wall” of the entrance to the North Transept (see ‘N’ on Figure 4). This is also in Latin, an epitaph written by his friend Dr. Pierce, then President of Magdalen College, Oxford.’! It makes no specific reference to its geographical association to the, then, Norborne vault and is some distance from the Stokes monument at ‘S’. Perhaps Castle House vault was located

LOOKING FOR DR INGEN HOUSZ

127

Figure 4. Sketch plan of east end of St Mary’s Church, based on pre-1864 floor plan, showing possible locations of the

Castle House vault

somewhere along the axis ‘S-N’ (see Figure 4). A third piece of evidence introduces other possibilities, although still pointing to the Chancel. On the east wall of the south chapel of the Chancel are two modest tablets referring to William Powell Bendry and his wife, Mary (see point ‘B’ on Figure 4). The Bendrys were tenants of the Third Marquis of Lansdowne, Henry Petty, at Castle House between 1815 and their respective deaths in 1816 and 1835. The tablet to William Bendry ‘... of Castle House in this Parish Esq.re ...’ as he is described in his widow’s tablet, is in Latin and begins ...Juxta conduntur reliquiae ... - “Nearby were placed the remains of ...’. But the tablet to Mary Bendry, his widow, is in English and states, - unequivocally, that she was buried ‘...in a vault adjoining ...’. The word ‘adjoining’ is difficult to interpret per se but must surely mean somewhere in the overall area of the Chancel other than immediately below the tablets.

In essence, then, the internal evidence for the possible location of the Castle House vault points to the Chancel and the best of that evidence, assuming the oldest and most imposing tablets to be located, still, in their original positions, to the north side of

the Chancel. It is suggested, therefore, that the remains of Dr. Ingen Housz were deposited, on 9 September 1799, in the Castle House vault which was most likely situated between points ‘S’ and ‘N’ in the north chapel (aisle) of the Chancel of St. Mary’s Church, Calne. These tentative conclusions for the final chapter of the story of Dr. Ingen Housz may be permissible but there is still no information, from any era, on the structure of any such vault, its dimensions, design, capacity, cost, state of repair, or construction materials.

HOW THE VAULT CONTAINING DR. INGEN HOUSZ WAS LOST AND HOW IT MIGHT BE FOUND

The Chancel in St. Mary’s church, Calne has undergone considerable changes. In 1863 William Slater, the famous London ecclesiastical architect, was commissioned to draw up specifications for a major refurbishment of the building that would incorporate changes dictated by Victorian ideas of a ‘modern’ church. The resulting faculty directed that, after removal of the box pews and galleries, the floor of the

128 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

church was to be lifted, wooden and stone parts alike, including the ‘...(uneven) stone pavement of the passages and of the chancel...’ [authors’ italics]. Then the ‘Digger’ was told to ‘...clear out the soil from the whole of the internal area of the Church to an average depth of at least one foot below the level of the underside of posts of seating...’ Later the contractor was told to make a new ‘...Chancel pavement ...of Mintons 4/2 red buff and black encaustic tiles ... in patterns ...’.’”” Any pointers to the sites of vaults or their entrances on the floor of the chancel will obviously have been destroyed. That even some might have survived seems naive especially when a later faculty in 1934” gave permission for removal of the Chancel floor tiles and for their replacement by new stone flags.

There is, however, one helpful clue in the succession of faculties that have sanctioned the numerous changes, installations and adjustments to the eastern part of St. Mary’s church during the last two centuries. The document permitting installation of the present organ was authorised in 1908. The musical instrument itself is acknowledged as unique and valuable. It is also considerable in size and occupies virtually the whole of the north chapel of the Chancel, the very part of the church on which the search for the burial place of Ingen Housz has come to focus. Bulk and weight were obviously major considerations when structural problems were discussed by the organ installers, for the authors of the faculty felt it necessary to state, categorically, ‘...that remains and deposits need not be disturbed...’. In other words, they knew that they were to be working above vaults or graves and that these were still occupied. Moreover, in the locked archives still held at St. Mary’s church, there is a copy of an Act of Parliament of 1852” explicitly banning any further burials in the churchyard. This was twelve years before any of the major structural changes instigated by the Slater proposals and it is difficult to conceive that builders would have attempted to lift old, fragile, and very heavy lead coffins from vaults that they were obliged to breach if re-interment involved (as it would have done after 1857) a journey to the new burial grounds at Holy Trinity, Quemerford. It seems more likely that contractors, being only human, would have filled in such vaults, leaving the occupants ‘in situ’.

Since the pointers to the whereabouts of the Castle House vault on the walls of St. Mary’s are somewhat ambiguous and since clues such as worn or chipped stones on the floor of the building that might have admitted to steps to vaults have been obliterated, deeper, subterranean explorations under the Church

would be needed to obtain any further information in the hunt for Dr. Ingen Housz. A destructive assault on the building would obviously be illicit. It would also be impractical other than as a tangential exercise to essential refurbishments. These are, though, the kind of activities which have sometimes given archaeologists opportunistic access to ‘lost’ vaults in other churches. It is just possible that such exploration might reveal the existence of a subterranean cavity, probably scalped and back-filled in 1864, that was once the vaulted burial chamber commissioned, by his widow, to receive the remains of Walter Norborne in 1660 and which, 140 years later was utilised as the resting place of a man who wished to be *...buried in the church of the Parish in which I shall die...’, a certain Dr. Jan Ingen Housz.

Acknowledgements

We owe thanks to very many people, especially the following: the Marquis of Lansdowne, Bowood; Dr. Kate Fielden, Bowood; the Reverend Robert Kenway, Vicar of Calne and Blacklands, Barbara Gleed, senior Churchwarden, the parish of Calne and Blacklands; the Venerable John Smith, Archdeacon of Wiltshire (now retired); Julian Litten FSA; Professor Don Emblen; Philip Winterbottom, Archives of the Royal Bank of Scotland; Eric Boore; Robin Holley; Peter van der Pas; Jonathon Forsyth; Duncan Coe. The staff at the following institutions have all been extremely helpful ~The British Library (Manuscripts);The Royal Society Library; Wiltshire and Swindon Record Office; Devizes Museum; Gemeentearchief (Municipal Archive), Breda; Vienna Town Hall; the Library of the American Philosophical Society, Philadelphia. Julian Litten, Eric Boore, and Kate Fielden all helped us with the script and we express particular gratitude to them.

Bibliography and Notes

1. P. van der Pas, Dictionary of Scientific Biography. (Charles Scribner’s Sons, New York 1981); H. Reed, ‘Jan Ingenhousz, plant physiologist. With a history of the discovery of photosynthesis’. Chronica Botanica, 11 (1949) pp. 285 — 396; J.Wiesner, Jan Ingen-Housz. Sein Leben und sein Wirken als Naturforscher und Artz. (Verlagsbuchhandlung Carl Konegen, Wien 1905); N and E Beale, Who was Ingen Housz, anyway? (Calne Town Council, Calne 1999).

2. ‘Trained in Leyden under Boerhaave, Pringle (1708 — 1782) rose from army doctor to President of the Royal

LOOKING FOR DR INGEN HOUSZ

Society and physician to George II. He wrote a seminal book on diseases of the army and coined the words ‘septic’ and ‘antiseptic’.

3. Priestley (1733 — 1810) had also been introduced to Lord

Shelburne, probably by Benjamin Franklin, and in 1773 took up the position of librarian at Bowood House, Calne. He established a laboratory at the house and it was here, on 1 August 1774, that he discovered oxygen.

Live smallpox serum was scratched into the skin of the recipient in the hope of provoking a mild attack of the disease to ensure later protection. Jenner later used the same clinical technique but with (much safer) cowpox serum.

5. Ingen Housz was elected Fellow of the Royal Society on

25 May 1769 having been proposed on 15 February and balloted at 11 subsequent meetings. His sponsors were W. Watson, W. Huck, M. Blair, G. Baker, Sir John Pringle, W. Watson junior, W. Heberden, B. Franklin, Gowin Knight, James Parsons and M. Maty. Royal Society: Certificates of Election III, 69.

6. Reference is made, in correspondence after Ingen Housz

~)

had died, to the ‘coolness’ of the marriage: Marquess of Lansdowne to Caroline Fox and Elizabeth Vernon, 9 January 1800: Shelburne MSS. 6; ff 95 — 6.

. J. Ingen-Housz, Experiments upon vegetables, discovering

their great power of purifying the common air in the sunshine, and of injuring it in the shade and at night. (Elmsly and Payne, London 1779).

. Marie Antoinette (1755 — 1793) was the 15th child of

Empress Maria Theresa and younger sister of Emperor Joseph II.

9. W. Falconer, Letter from Ingen Housz to the author in:

10.

Wil

We

13:

An account of the efficacy of the Aqua Mephitical Alkalina. (Cadell, London 1792). A letter from Ingen Housz to the British Inland Revenue, 1798 or 1799, explaining why he was stranded in England, that he was not earning fees as a physician but was being taxed on his pension from Vienna is contained in Breda Gemeentearchief IV, 16A — 13. Lord Shelburne had been made First Marquis of Lansdowne by George III after serving as Prime Minister between July 1782 and April 1783, a brief but active spell in office during which the American War of Independence was concluded. Widowed for the second time in 1789 and retired from politics, Lansdowne spent long sojourns at Bowood among a revolving circle of family members, friends, and distinguished visitors. Ingen Housz was one of the most regular guests and was even left ‘in charge’ at times - Caroline Fox, London to Ingen Housz at Bowood, 2 November 1795: Breda Gemeentearchief IV, 16A — 13; Ingen Housz to First Marquess of Lansdowne, 29 July 1797: BL. Add. MS. 51821, ff 61 — 62. Dr. William Falconer MD (1744 — 1824), was, at this time, the physician in charge of the Bath General Hospital in Upper Borough Walls (now the Royal National Hospital for Rheumatic Diseases) and had an extensive private practice in Bath, a mecca for the infirm. Ingen Housz had known him for some years.

14.

SY

16.

Lie

18. Ws

j=)

bt be —

wo N bw Be WO Lo

26. 2s . At that time the only legal burials permitted were in

Bie

32; 33.

129

First Marquess of Lansdowne to the oldest nephew of Ingen Housz in Breda, 9 September 1799: Breda Gemeentearchief IV, 16A — 13.

Caroline Fox to her brother, Third Lord Holland, 16 September 1799: BL. Add. MS. 51735, ff. 140 — 1. Published in Fremden-Blatt (Foreigners’ Newspaper) Issue 164, pp 15 - 16, Vienna, 16 June 1905.

Ingen Housz to Dr. William Falconer, 24 August 1799: Breda Gemeentearchief IV — 16A — 13.

Breda Gemeentearchief IV — 16A — 13. See note 14. Ingen Housz appears to have developed a repertoire of simple experiments that required little apparatus and with which he could entertain in company. One of these was the burning of a length of iron wire in pure oxygen, a phenomenon that gives a spectacular light.

. Breda Gemeentearchief IV — 16A — 13, see note 14. . Caroline Fox to Third Lady Holland, 8 September 1799

BL. Add. MS. 51744 ff. 115.

= Ibid: . Breda Gemeentearchief, IV, 16A — 13, see note 14. . From handwriting comparisons it is apparent that the

letters, though dictated by the Marquis, were actually written by the Reverend Etienne Dumont. Dumont (1759 — 1829) was secretary and close friend to the Marquis: Jefferson P. Selth, Firm Heart and Capacious Mind: the Life and Friends of Etienne Dumont. (University Press of America 1997), passim; Breda Gemeentearchief IV, 16A — 13. Ingen Housz had always remained close to his older brother, Ludovic (1729 — 1788), who carried on the family business in Breda and who had a large family. After Ludovic died following a carriage accident, Ingen Housz showed much care and concern for his Breda nephews and nieces.

. Riksarchief in Noord-Brabant: catalogue Family van

Lanschot, 1262. A copy of the letter of condolence that Ludovic Ingen Housz’ widow sent to her bereaved sister- in-law in Vienna relates the exact postal arrangements made.

Breda Gemeentearchief IV, 16A - 13. see 18

WRO 2083/20 Calne Burial Register, 1792 — 1812.

Anglican churches and churchyards and since St. Mary’s was then the only church in Calne identity and location are unambiguous.

. Breda Gemeentearchief IV, 16A — 13. 30.

Guild Stewards were, effectively, joint Mayors, serving for a year. Calne elected Mayors only after the passing of the Municipal Corporations Act of 1835.

The procedure required at that time, we assume, in order that a will held in a country other than where the death took place could be proved. Ingen Housz had a substaatial will in Vienna: Landes- & Stadts-archiv, Vienna, 1799, 2318.

Breda Gemeentearchief IV, 16A — 13.

For the purpose of definition ‘...a burial vault is a subterranean chamber of stone or brick capable of housing a minimum of two coffins, side by side, and with an internal height of not less than 1.74 metres ... a

130

34.

3D) 30.

40.

41.

42. 43. 44,

45. 46. 47. 48.

49. 50:

i, 52. 53):

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

vault need not necessarily have access steps and neither was there any compunction for its presence ... to be identified...’. J. Litten, The English way of death. The common funeral since 1450. (Robert Hale, London 1991) p. 207.

Breda Gemeentearchief IV — 16B — 18b, The will, in England, of Ingen Housz; V. Braunberens, Mozart in Vienna. (English Translation) (Andre Deutsch, London 1990), p. 414.

Personal communication, J. Litten.

The ledgers of Drummonds Bank, now the Royal Bank of Scotland, for 1799: Archives of Bowood House: General Daybook Accounts 1797 — 99, manuscripts Lansdown to Mr. Broad (agent).

. J. Litten in Grave Concerns. Death and burial in England

1700 — 1850. (CBA Research Report 113. Council for British Archaeology, York, 1998) p.9.

. Litten, Joc. cit. . “...lead coffins ... are almost invariably associated with

vaults...”: W. Rodwell, Church Archaeology. (Batsford, for English Heritage, London, 1998), p. 166. A. Schomberg. “The monumental inscriptions of Calne church, Co. Wilts’. The Genealogist, New Series, Volume 14, (1898), pp 37 — 44, 90 — 94. This is the case for Trowbridge church for which there is a detailed floor (and under-floor) plan, as at 1816, recording the location of some 100 sites of deposition. WRO 206/37

Schomberg, loc. cit.

Rodwell, p. 158.

The memorials to the Lansdowne family are mostly in the north transept/tower area of St Mary’s Church, Calne, but none relates to actual burials. They are almost certainly in this part of the church only because the family pews were here. Earl of Kerry, ‘King’s Bowood Park (no. 2). WAM, vol. 41, (1920 -1922), pp. 502 — 521

Ibid.

WRO 212B, 1220; WRO 2176/2.

Schomberg, loc. cit.

Lord Edmond Fitzmaurice, Life of William, Earl of Shelburne. (Macmillan, London, 1875), vol.1, pp. 307-8 WRO 212B, 1237.

WRO 212 B, 1008. A feoffment was a conveyance evidenced by a fee.

WRO 212B, 1142.

WRO 212B,1237.

A. and H. Stokes, Stokes Records. Notes regarding the ancestry and lives of ... (privately printed, New York 1910).

54.

55

63.

64.

65.

60.

67.

68.

69.

70.

TA U2: US. 74.

A. and H. Stokes, Joc. cit. . A. Marsh, A history of the borough and town of Calne. (Heath, Calne 1903).

G. Duckett, Historical and genealogical memoirs of the family of Duket from the Norman conquest to the present time. (J. Russell Smith, London 1874), p. 66. . Duckett was in debt to precisely this amount.

. WRO 2167/2. . WRO 2167/3. . By this time Castle House was owned by Lord Henry

Petty, second son of the first Marquis.

. Schomberg, loc. cit. . Lord Lansdowne, Bowood to the Ingen Housz nephews in London 28 November 1799: Gemeentearchief Breda

IV, 16A — 13 Letter of 5 May 1800: Shelburne MSS.,2. It is unclear whether the loss was attributable to gout or a stroke.

Ingen Housz nephews at Breda to their cousin, Josef Jacquin, Vienna, 5 October 1799: Gemeentearchief IV 16A, 13.

Earl of Kerry. ‘King’s Bowood Park [No.3]’. WAM, vol. 42, (1923), pp.18 — 38.

Gentleman’s Magazine, vol. 61, part 2, pp.395 — 396. 23 May 1791.

E. Crankshaw, Maria Theresa (Constable, London 1983) p. 310; The Marquis certainly practised what he ‘preached later in his life. Although he, too, was buried in a vault (under the Chancel of All Saint’s Church, High Wycombe with other members of his family) he forbade the erection of any monument. The only notification of his burial place is a modern window marking his association with Benjamin Franklin and the settlement of the American War of Independence.

The east wall was constructed in the 13" century: Harold Brakespeare FSA, ‘Notes on the architecture of Calne Church’ in A. Marsh, A history of the Borough and Town of Calne. (Heath, Calne 1903), pp. 150 - 167.

The organ was installed in 1908: WRO 2586/50, Faculty

of 12 December 1907.

The wall was reconstructed by 1650 after the fall of the central spire/tower in 1638: A. Marsh, A history of the Borough and Town of Calne. (Heath, Calne 1903), p. 153.

ibid. p. 188.

WRO D1/61/15/14, Faculty of 1863.

WRO 2586/50.

‘An Order in Council closing Calne Parish Churchyard, 11 August 1854’.

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 131-180

Neolithic activity and occupation outside Windmill Hill causewayed enclosure, Wiltshire: survey and

excavation 1992-93

by Alasdair Whittle,' Jessica J. Davies,” Ian Dennis,' Andrew S. Fairbairn’ and Michael A. Hamilton,’

with a contribution by Joshua Pollard *

As part of a regional programme of research into the Neolithic of the area, which included excavations at the Windmill Hill causewayed enclosure in 1988, fieldwalking, test pitting, geophysical survey and limited excavation of selected areas and features were carried out on the southern slope of the hill below the enclosure in 1992-93. Rich collections of worked flint from this location were formed from the end of the last century. The activity of Kendall and Keiller from 1904 to 1929 was especially important. Surface lithic survey and test pitting showed some variation in lithic densities, which were however surprisingly low; it 1s likely that nearly a century of flint collecting has almost picked out the lithic scatter. There is probably much more Later Neolithic material in the Kendall-Keiller collection than Earlier Neolithic. Geophysical survey indicated few subsoil features. Excavation in selected areas showed one small concentration of Earlier Neolithic pits, and one small concentration of Later Neolithic pits. The pits were mostly unweathered and backfilled, and contained deposits of artefacts and animal bone. Both sets contained the remains of wild and domesticated resources, though the earlier pits had more cereal remains, perhaps the result of deliberate burning. The relationship of the earlier pits to the enclosure Is not known, though the activity in them bears resemblances to that seen in the enclosure ditches. How the activity of both phases relates to wider patterns in the area remains unclear, though the evidence recovered does not support a model of fully sedentary existence in either phase. Both earlier and later pits contain the remains of domestic activity, including flintworking and the processing of meat and plant foods. The later pits and the bulk of the lithic scatter may also mark the continuation of a tradition of special visits to a special place.

the physical environment; the further hope was to gain new insights into the circumstances in which

INTRODUCTION

by Alasdair Whittle

The research context and aims

From 1987 to 1993 a field research project was carried out to investigate the Neolithic of the Avebury area. Its primary aims were to refine our knowledge of the Neolithic sequence in the area by obtaining more radiocarbon dates, and to extend our understanding of spatial and temporal variation and development of

monuments and other sites were created, used and abandoned (Whittle 1993). Sites investigated were Easton Down long barrow (Whittle et al. 1993), Millbarrow chambered tomb (Whittle 1994), Windmill Hill causewayed enclosure (Whittle et al. 1999;Whittle and Pollard 1998), and the West Kennet palisade enclosures (Whittle 1997a). The earlier excavations at Silbury Hill were also published in full (Whittle 1997a), as well as those from slightly further afield at Wayland’s Smithy (Whittle 1991);

1. School of History and Archaeology, PO Box 909, Cardiff University, Cardiff CFl 3XU_ 2. School of Archaeological Sciences, University of Bradford, Bradford BD7 1DP_ 3. Institute of Archaeology, University College London, 31-34 Gordon Square, LondonWC1H OPY 4, Department of Humanities and Science, University of Wales College, Newport, Caerleon Campus, PO Box 179, Newport NP18 3YG

132 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

radiocarbon dates were also obtained from Avebury (Pitts and Whittle 1992), and the Sanctuary was re- assessed (Pollard 1992).

While enormously productive of new data and a fruitful source of new interpretations, the project had largely concentrated on the monuments themselves. Though that did not exclude direct evidence for activity and occupation, for example from pre- monument contexts at Windmill Hill and Easton Down, it was certainly a skewed approach (Edmonds 1999), which has yet to be corrected in the Avebury area. We know of lithic scatters (Holgate 1987; 1988a) and small foci of occupation, variously preserved in the area (Summarised in Thomas 1991;Whittle 1993; cf. Evans and Smith 1983). There has still not been systematic survey of the kind carried out in the Stonehenge area (Richards 1990), or to more limited extents in Cranborne Chase (Barrett et al. 1991), or in the eastern area of the Dorset Ridgeway (Woodward 1991).

In an attempt to redress the balance of research, and to assess the potential for future such investigations, the area of the large, well known lithic collection on the southern slope of Windmill Hill was selected for more detailed investigation. It was recognised that it is very likely that this was not a typical situation, since the collection was large (I. Smith 1965, 18, note 1) and close to the causewayed enclosure, but the link with research already carried out at the enclosure was compelling within the terms of the project as a whole.

The setting

The physical setting of Windmill is well known (I. Smith 1965; Whittle et al. 1999) and needs little rehearsal (fig. 1). An outlier of Middle Chalk, with Lower Chalk around its base, the hill stands out in the local landscape. Investigation so far has produced very little sign of Mesolithic activity on the hill, and the place seems to have become significant in a landscape in which looking out and looking at were important. Nonetheless, there are significant differences in aspect. The causewayed enclosure, from the detail of its layout on the hill (especially of inner and outer circuits), can be seen as looking largely north and north-west (Bradley 1998, 122), while the lithic scatter is on the long southern slope, with a direct view to and from a much wider sweep of landscape. It is of course possible that the southern slope was selected for some quite different reason, for example sunshine or warmth.

Previous research: Kendall and Keiller

The story of the investigation of the causewayed enclosure from the 1920s onward is well known (Crawford 1953; 1955; 1. Smith 1965; Malone 1989; Whittle et al. 1999; Barber et al. 1999). Its first investigator, the Rev. H.G.O. Kendall, rector of Winterbourne Bassett (soon to be helped by O.G:S. Crawford, who in due course guided Alexander Keiller to the site), was by then already responsible in large measure for the formation of a very large flint collection from the slopes around Windmill Hill.

In his book on the area published in 1885, the Rev. A.C. Smith had judged of the Windmill Hill enclosure that ‘in all probability this was a British camp’, speculatively of Neolithic age (A. Smith 1885). Ordnance Survey maps record worked flints as having been found on Windmill Hill since 1888. The only north Wiltshire material illustrated by John Evans in his Ancient Stone Implements (1872) was from barrow excavations, but it is clear (Martyn Barber, pers. comm.) that collecting had gone on since the latter part of the nineteenth century. In the 1930s, Cunnington and Goddard (1934, 6) noted that the largest lithic collection from a single site held by Devizes Museum came from Windmill Hill, ‘many the gift of the late Wm Browne, of Avebury’, and Keiller (1934, 138) referred toWindmill Hill as having been ‘famous for decades...as a paradise of the surface flint hunters’. In discussion following Kendall’s 1914 paper, a Mr Dale refers to a visit he made in 1882 to ‘Mr John Brown at Avebury, who first discovered the site on Windmill Hill and had a good collection, but its fate was unknown’ (Kendall 1914). Perhaps either the Brown(e)s were related or were the same man. Kendall also quotes a Dr Blackamore who ‘remarks that, in a collection formed many years ago, ‘fabricators’ were specially numerous’ (Kendall 1914, 74).

The subsequent credit for the early exploration of the site must go to Kendall. He came to north Wiltshire in 1904 after several other posts, and stayed there till 1924. His notebooks hint at a lively, energetic and involved man, with a weak heart and prone to indigestion and overwork on behalf of his parishioners. He was already an enthusiastic student of flints and a believer in eoliths; his first paper, “Eoliths and pseudo- eoliths’ was published in Manin 1905. Earlier papers were concerned with Grime’s Graves and with ‘Palaeolithic’ and ‘Eolithic’ industries from Hackpen Hill on the Marlborough Downs and Knowle Farm

133

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE

UOMESISIAUI JO BAIL DY JO UL[Y ‘[ AINSI]

OSrl oor ose: ooel osol 0001 osé °

Jayowoipeiy ---- sajjow OOS

Ayyiqudaosns osjeaubeyy ——

seaiy payeaeoxy

P/aly YON AINGSAJEA

plaid sqqid

WNINHD J3]NO

THIH THWAONIM

e {HH 1HWpUlAy

134 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

in Savernake Forest near Marlborough. Kendall was elected FSA in 1913, and was a president of the Prehistoric Society of East Anglia, to which he lectured during the war. Already by 1910 he had a substantial collection of flints because a group from the Wiltshire Archaeological Society visited the rectory in that year to see it; Palaeolithic and Eolithic specimens were displayed in the study, and Neolithic finds in the drawing room. The Wiltshire Archaeological Magazine also records the first donations by Kendall of axe fragments from Windmill Hill in 1910 (vol. 36, 358 and 508).

Notes and papers referring to the site appeared from 1912 onwards. Searches were carried out ‘many times a year’ (Kendall 1916, 232). In Kendall’s day, the upper part of the hill would have been parcelled up into a number of fields, most of them under cultivation. ‘An immense number of chipped flints have been found there [on Windmill Hill]...including leaf-shaped implements, celts of a certain kind, and so on. Arrowheads of every variety have been found in large numbers, and immense quantities of scrapers’ (Kendall 1916, 230). Among the arrowheads, chisel- ended types were prominent (Kendall 1916, 234; 1922, 522). Cruder as well as well finished tools were also found, but Kendall argued that the assemblage must be treated as of essentially one date, Neolithic or later. ‘One cannot search such a site as Windmill Hill, as it has been the writer’s good fortune to do, many times a year, for a period of twelve years, without being convinced that the chipped flints on the hill...all belong to the same industry, or two industries’ (Kendall 1916, 232).

In 1924 Kendall moved to a living near Salisbury, and sold much of his Windmill Hill collection to Alexander Keiller, in two lots. In 1924, ‘perhaps over a thousand’, ‘thinned out of my cabinets’ (Kendall letters in Avebury Museum) were sold, and in 1925 some 1,819 specimens changed hands. There was a great variety of implements, including 428 scrapers, 554 ‘knives, scrapers and rechipped celts’ and 163 arrowheads (‘a good many are chisel-ended, and many others are broken, chiefly in ancient time’). Some struck flakes were included. In his letter to Keiller of 16 November 1925, Kendall wrote that ‘I have assigned to you a little under half of the arrowheads, and favouring myself somewhat, tho not altogether, in quality. On the other hand, I have only burdened you with a smallish quantity of flakes (not but what they have their importance). I have retained only a small representative lot of scrapers; and have also given you the majority of knives, cores, fabricators etc’. The remark about flakes should imply that he had collected

many more. In a slightly earlier letter (6 October) Keiller had written, “Yes I should like cores included. I am becoming increasingly keen on cores’. Keiller appears to have paid Kendall £30 for the 1924 sale and £25 for the 1925 purchase.

Kendall fell ill in 1927 with a nervous breakdown. During a fieldtrip to Windmill Hill in 1979, I heard R.J.C. Atkinson recount the tale that in his last days Kendall suffered delusions that he was a flint, but the rather cruel story may be apocryphal. Kendall certainly suffered more illness and died in April 1928, aged 62. Keiller bought the rest of the collection from his widow, paying her this time £100.

Keiller continued to build up his collection for the area as a whole, by purchase from the local farm workers. Receipt books record acquisitions from 1925-1929, when he was at work on the enclosure. Keiller was not always satisfied. Of the collection from Yatesbury Field of A.G. Rogers, he recorded (6 May 1927): ‘I have never had a rottener lot than this submitted to me, here or anywhere else. It is not worth my while to give the time required to examine such stuff. In future nothing will be paid for Yatesbury Field stuff unless of real merit’. One wonders whether this kind of attitude could have distorted the provenances of other flints offered to Keiller. Nothing was paid for worked flakes, but there were set prices for other categories. Scrapers dominate the finds. Provenance by field was recorded for 1928-29, when most of the material came from Windmill Hill and North Field, but other details are frustratingly absent. An annotated map in Avebury Museum notes many cores from what is now the top north-east part of North Field.

Some collections were also made by A.D. Passmore (Holgate 1988a, 286), although much of his collecting was on Hackpen Hill, and by others. Holgate (1988a, table 4, 242) has listed a total (from the Avebury, Devizes, Swindon and Ashmolean museums) of 13,000 worked flints from the Windmill Hill surface site, including nearly 1300 flakes, over 700 cores, over 7000 scrapers and about 380 leaf arrowheads and nearly 1000 Later Neolithic arrowheads. Other specimens are known to exist in the Knowles collection, now in Ireland, and the local tradition of flinting by farm workers has continued to the present day. It is likely therefore that the original scatter was indeed prolific, with apparently at least 10,000 implements of all kinds. The original density of flakes is uncertain, though cores were definitely present. The proportion of the total collection which was formed by Keiller is not clear, though it may have been at least half, and it could be that Keiller’s purchasing policy led to more flints being ascribed to

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 135

Figure 2. Distribution of worked flint found in the field walking survey in North Field

136

AH-Arrowhead AZ-Adze B-Blade/bladelet J-Rejuvination flake N-Notched flake P-Piercer

K-Knife

Se-Serrate

C-Core F-Fabricator H-Hammerstone

R-Retouched flake

S-Scraper

U-Utilised flake

Ax-Axe

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Figure 3. Distribution of artefacts found in the field walking survey in North Field

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 137

or claimed from Windmill Hill than was actually the case. In any event, this was clearly a major assemblage, one of the largest recorded in southern England (Holgate 1988a).

FIELD SURVEY by Ian Dennis

The field on the south slope of Windmill Hill known as North Field was selected for a systematic field- walking survey conducted during the autumn of 1992 (Figs 1-2). The field-walking collection pattern used in the survey was aligned on the national grid. This system was compatible with other field collection surveys, used in other projects within the Avebury location (Holgate 1987) and neighbouring regions (Holgate 1988a; Richards 1990; Schofield 1991).

A grid was laid over the field and divided into 19 transects 50m apart aligned with grid north; these were further divided into metre-wide 50m long sections forming the basic collection unit. The transects were then walked from south to north and all material along each 50m strip was bagged as a discrete collection unit. There were in total 284 collection units covering the 19 transects. The numbers of units varied with the total length of each north-south transect.

The material collected may represent approximately 5% of the total available material on the ground at any one time. Distribution maps showing overall density (Fig. 2) and diagnostic flints were plotted (Fig. 3). Only data from whole collection units were used to maintain statistical coherence.

Artefact visibility during the field survey was very good. The field had been ploughed and exposed to weathering for at least a month, effectively washing the surface soil off the flint, allowing the white patinated flint to show against the dark soil background, considerably facilitating its retrieval.

TEST PIT SURVEY by Ian Dennis

- A test pit survey was conducted together with excavations during the summer of 1993 further to investigate the nature and extent of the flint scatter. A grid was constructed using 19 transects aligned with grid north, spaced at 50m intervals. Each transect was further divided into 50m sections and 1 by 1m test pits were dug at these points. The test pit grid was positioned so as to bisect the transects of the field- walking grid. Soil from all the test pits was sieved using

lcm meshes. Results from the test pits were plotted on to distribution maps showing overall worked flint density (Fig. 4) and artefact location (Fig. 5).

A total of 156 pits were excavated across the north part of North Field, along with 16 test pits dug in Gibbs Field, 8 around Horslip longbarrow and 12 down the centre transect towards the southern end of North Field. These 12 test pits were dug to investigate whether colluvial deposits are masking surface data at the base of the hill. Sieving was considerably more difficult owing to the larger proportion of clay within the matrix, making this particularly dense and compact. Colluvial build-up is apparent, but results indicate that this does not appear to mask significant finds.

Only two test pits produced any subsoil features. Test pit 700-650 contained a single stakehole, 5cm in diameter at the top and 10.5cm deep, while Test pit 600-750 contained a single posthole, 28cm in diameter and 20cm in depth. However, neither of these features can be dated and are presumed to be modern.

GEOPHYSICAL SURVEY by Michael A. Hamilton

Geophysical surveys were initiated and supervised by the author, with the further contribution of Amanda Banham. The main technique employed was a fluxgate gradiometer survey. Four main areas were surveyed (Fig. 1). In addition, the centre-north part of North Field was also subject to a magnetic susceptibility survey. Seven and half 100 by 100m grids were undertaken (7.5ha); three 100 by 100m grids were also surveyed in the north-west of North Field (3ha). A limited amount of resistivity was conducted. This involved seven 20 by 20m grids to the centre-north of North Field (0.28ha) and one grid in the north- west of North Field (0.04ha).

Survey organisation

The gradiometer survey was based on 20 by 20m grids, named after eight-figure national grid co- ordinates.

Equipment

A Geoscan FM36 fluxgate gradiometer was used with a sensitivity of 0.1nT. Readings were taken with a transverse interval of one metre and a sampling interval of 0.5m, making 800 readings for every 20

138 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Figure 4. Distribution of worked flint found in the test pit survey in North Field

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 139

by 20m grid. A Bartington magnetic susceptibility meter was used. 100 by 100 m grids were surveyed. Three or four readings were taken every 5m and a single average reading recorded by hand. Each grid consisted of 400 recorded readings and these were then entered manually into the Geoplot 1.2 programme. A Geoscan RM4 resistivity meter and DL10 datalogger were used. These were recorded in 20 by 20m grids with a sampling and transverse interval of 1m, making 400 readings per grid. The data were initially processed using Geoscan Geoplot 1.2 and 2 programmes but were completed using Geoplot 2 and 3.

Results

Centre-north of North Field (Fig. 6)

Linear features

Ll. On both the gradiometer and magnetic susceptibility printouts was a strong feature, running NNW to SSE, which continued the modern fence that forms the eastern side of the salient of preserved grassland which extends into North Field. This feature

AH-Arrowhead C-Core

AZ-Adze F-Fabricator

B-Blade/biadelet

H-Hammerstone P-Piercer

extends for 200m on the magnetic susceptibility survey, and probably reflects a vanished field boundary. The feature on the magnetic susceptibility was very broad, perhaps reflecting relatively recent destruction.

L2. A similar feature, but more north-south in orientation, is recorded some 185m to the east of the modern fence mentioned above. Only the magnetic susceptibility covered this area. This feature runs for 300m. In orientation it conforms to the modern eastern boundary of North Field and again could be a vanished field boundary.

L3. Between A and B was a less distinct feature on the same orientation as A.

L4.A NW-SE linear feature which does not appear on the magnetic susceptibility survey. This feature may terminate on L5.

L5. An east-west feature, absent from the magnetic susceptibility survey.

Lo. A north-south feature, absent from the magnetic susceptibility survey.

L7.A feature on the magnetic susceptibility survey.

L8.A feature on the magnetic susceptibility survey, parallel with L7.

J-Rejuvination flake | R-Retouched flake

N-Notched flake S-Scraper

U-Utilised flake

Figure 5. Distribution of artefacts found in the test pit survey in North Field

140 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Limit of gradiometer survey

r ny

SoS

NaS Ae

aS

Figure 6. Outlines of possible features found by gradiometer survey 1n the north part of North Field (for details see text)

L9. A vague curving feature on the magnetic susceptibility survey, with no real support on the gradiometer survey.

Other features

I. Avebury G46a (Grinsell 1957, 154) shows up very clearly on the gradiometer survey, as does a possible grave/disturbance at its centre. There are also signs that either the barrow was incorporated into a later enclosure to the north, or it was built to the south of an existing feature. The magnetic susceptibility shows a much wider (at least 70m across) irregular area of higher readings, suggesting either the spread of material from the mound, or perhaps that the barrow was the centre for later activity, or sited on earlier material. There is no rea! confirmation for this feature in the testpit survey.

II. Avebury G48b (Grinsell 1957, 155). Again this is clear on the gradiometer survey. The high readings in the magnetic susceptibility survey are limited to the area within the ditch.

II. An area of high readings to the NW of the surveyed area. This shows up especially well in the magnetic susceptibility survey and corresponds to Area D which produced a series of Earlier Neolithic pits (see below). The feature is not as well defined in the gradiometer survey, but there is a very subtle area of higher readings (+ 0.5 nT); yet it was from the gradiometer surveying that this feature was identified. This does raise the question whether similar subtle variations in the gradiometer and magnetic susceptibility surveys could have a similar interpretation. At least one similar anomaly was tested (excavation Area E) at the top of North Field and appeared to be little more than an effect of ploughing. However, this feature was less prominent in the magnetic susceptibility survey.

IV. The magnetic survey suggests a vague large circular feature, c. 80m diameter. It is difficult to interpret. It is possible that this is the barrow identified by Grinsell (G48a) from a St Joseph aerial photograph (Grinsell 1957, 155). This feature could not be located on oblique aerial photographs held by the RCHME in Swindon.

V. A vague suggestion of a rectilinear enclosure in the magnetic susceptibility survey.

VI. An area of very low readings 1n the magnetic susceptibility survey.

There were also numerous iron spikes. Many were test-excavated and produced only modern iron objects.

North-west of North Field

This showed a series of linear features orientated NNW-SSE and spaced 25m apart, possibly ridge and furrow, or strip division of the field. There is one other rather irregular linear feature which runs SW to NW. The magnetic susceptibility survey suggested more activity over a 90m area around SU086711. This is supported by the gradiometer survey which has more readings + InT in that area; however, there is no coherent pattern.

The area excavated as Area M (see below) is not particularly striking on the gradiometer or magnetic susceptibility survey. Attention was drawn to this area by the discovery of a stone axe fragment in a routine test pit, which coincided with an area of higher flint density recorded by the field survey. The test pit was then extended, and a resistivity survey and a closer-

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 141

spaced gradiometer survey revealed features subsequently identified as Later Neolithic pits (212 and 202).

North of Gibbs Field Only vague suggestions of features. Centre of Horslip Field

No features. Discussion

Overall the results were disappointing. Two pit clusters were located but only with great difficulty. It is impossible reliably to extrapolate from those tested by excavation to suggest others. More trial excavation might provide the data for a predictive methodology, though this is not certain. The only clear features were a series of ring ditches and a number of linear features, presumably Later Bronze Age and later. With the exception of the ring ditches most features in the surveyed areas produced very little response, and this is not dissimilar to the pattern found around the enclosure (David 1999). None of the magnetic susceptibility anomalies seem to correspond to any high densities of flint from the testpit survey. It is notable that the Earlier Neolithic pits in Area D corresponded to a small magnetic susceptibility anomaly whilst not showing up in the testpit lithic survey.

SAMPLE EXCAVATION by Alasdair Whittle

Several promising magnetic anomalies were investigated, in 18 different locations, in the hope that they would prove to be subsoil features (Fig. 1). With the exception of Area D, all proved to be iron in the topsoil. The circumstances in which the features in

_Area M were found have already been described above.

Area D (Fig. 1; Fig. 7)

An area 4 by 4m was progressively opened, to reveal a small concentration of intercutting pits of Earlier Neolithic date. The earliest feature was a chalk-filled pit, 324, sterile and probably backfilled directly after being dug. This was cut by other pits, 308 and 323,

and 325 was close by. These three contained humic basal fills, with Earlier Neolithic sherds, bone, antler, sarsen and flint.

324. Oval in plan; 2.4m maximum diameter, up to 47cm deep, with steep to near-vertical sides and a more or less flat base. Its edges were rather disturbed, by animal burrowing, and by cutting of 308 and 323. Its fill was largely fresh chalk pieces, with some grey- brown soil in the upper part; the chalk fragments were up to 15-20cm long. There were no finds. This had clearly been deliberately backfilled very soon after being dug.

308. Irregularly sub-circular; when excavated it gave the appearance of two intercutting features, but no relationship or difference in fill could be seen. The main part of the feature comprised an oval pit 1.6 by 1m, whose lower sides were steep, upper sides shallower and very weathered, and base slightly rounded. On the west side a semi-circular bay c.60-80cm in diameter might be another feature; this was also steep-sided and its base was at the same level as the main part. The thin basal fill (312) consisted of large fresh chalk pieces. Directly above this was a thicker layer of dark grey- brown soil, with some small chalk pieces (310 and 311). This contained Earlier Neolithic sherds, animal bone and antler, as well as a large sarsen quern. Above this were two layers of dark brown soil with small chalk fragments (309 and 307). 312 and 310/311 appear to be deliberate deposits, along with possibly deliberately placed finds; 307 looks to be ploughsoil, and 309 could be similar or a slower secondary weathering product.

323. Oval-circular; c.90cm diameter, 45cm deep; steep sides and dished base. Its more or less uniform fill (22) was a dark brown soil with plenty of small chalk pieces, becoming slightly greyer towards the base and in central parts less chalky. This contained Earlier Neolithic sherds, flints and sarsen, but not in evident groupings or placings. The pit appears to have been backfilled in one go.

325. Circular; c.90cm diameter; steep sides at top, with under-cutting lower; slightly dished base. Its lower fill was a very dark grey-brown soil, with small pieces of chalk, and some flecks of charcoal towards the base. This contained animal bone, Earlier Neolithic sherds and flints, and two sarsen rubbing stones placed against the south wall. The upper fill (326) was a dark brown soil with some small chalk pieces. This pit also appears to have been backfilled in one go.

Features 315 and 328, to the north of 324, are part of a complex of animal burrows.

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

142

308

Figure 7. Area D: plan and sections of pits

143

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE

50cm

Figure 8. Area M: plan and sections of pits

144 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Area M (Fig. 8)

Three adjacent areas, over c.15 by 10m in total, were progressively opened, to reveal three dispersed features, two certainly of Later Neolithic date. There were a deep pit (202); a substantial pit with internal chalk walling and abundant animal bones (212); and a shallow pit or posthole (207).

202. Circular; 85cm diameter, 75cm deep; steep sides, undercut towards the base; slightly dished base. The basal fill was a dark grey-brown soil with a little wood ash and chalk silt. This contained animal bone and pieces of sarsen. The main fill was dark grey- brown soil with large angular chalk pieces (206, 210, 211). The upper part of this was a little less chalky. Animal bone was found in this fill (214) (Fig. 9), and there were a few Grooved Ware sherds. The upper fill consisted of a thin very dark grey-brown soil across the whole feature, possibly a remnant soil formation (204). Above this was dark brown soil with small rounded chalk pieces (203), presumably a plough deposit.

The possibility of a central postpipe was considered during excavation but discounted. This appears to be another pit deliberately backfilled before it had time to weather naturally.

212. Circular; maximum diameters 1.9 by 1.7m, and 94cm deep; steep sides, occasionally undercut, more sloping lower down, to form dished base. The basal fill was compact chalk rubble (225) and a mixture of chalk blocks and loose grey soil (224). Above this was found a penannular ring of compacted chalk (217), seeming to form an internal wall a few centimetres out from the chalk edges of the pit. This could be seen from an early stage in the excavation of the pit (Fig. 8). Between it and the true walls of the pit was compacted chalky silt, with occasional chalk blocks and patches of charcoal-rich soil (220 and 221). There were animal bones and other finds in the lower fill. In the main part of the pit, the fill was a dark brown soil with some angular chalk pieces (222, with 219 above). This contained abundant deposits (218) of animal bone (Fig. 9), and also Grooved Ware sherds. The upper fill was also dark brown soil with abundant chalk inclusions (216, 215, 213).

The pit appears to have been largely unweathered, and its fill was artificial. Animal bone and other finds were deliberately deposited with the main fill, but whether just dumped or otherwise more carefully placed is hard to say.

207. Irregularly circular; 1m diameter, maximum 35cm deep; sloping, irregular sides and small base.

The lower fill (209) consisted of light grey silt with angular pieces of chalk, with above this brown soil and chalk (208). There were no finds.

FINDS

Pottery by Michael A. Hamilton

The extensive test pitting and excavations conducted in 1993 did not produce much ceramic material. Even Romano-British material was less abundant than that obtained from the upper layers of many of the enclosure ditches. This may be explained by the long history of ploughing of the hill.

All the pottery was examined by a x20 magnifier, though only the fabric descriptions for the Neolithic material are included in the text.

Earlier Neolithic (Table 1)

Eight Earlier Neolithic fabrics were identified:

EN 1: Inclusions of: c.5% small iron oxides and some fine sand; grog (amount uncertain) up to 3 mm diameter; and long thin organic material. Soapy feel.

EN 2: Inclusions of: c.15% small sand: c.5% flint (up 3 mm in length, but mostly around 1 mm). Gritty feel.

EN 3: Inclusions of: c.3% small iron oxide and some fine sand; there are some larger sand particles (just visible with the naked eye); c.3% fossil shell; and probably c.10% grog.

EN 4: Inclusions of: c.20% shell; and some sand.

EN 5: Inclusions of: c.5% sand and iron oxide; and occasional tiny fragments of flint. Very well fired pottery with possible burnished exterior.

EN 6: Inclusions of: ?c.5% sand and iron oxide; and c.8% shell, including pieces greater than 3 mm in length, but mostly fine. Very parched appearance to interior surface. Exterior has a series of pop marks (c.4 mm in diameter), possibly as result of defects during firing, though they could be deliberate.

EN 7: Inclusions of: 10% flint (up to 4 mm).

EN 8: Inclusions of: 10% sand; and 10% flint (mostly below 1 mm).

In addition there was one possible Earlier Neolithic fabric:

UN 1: Inclusions of ¢.5% sand, occasional large white flint (up to 4mm), and probably much grog (up to 6 mm).

Largely on the basis of fabric it is possible to identify a minimum of eight vessels amongst the 22 sherds. Four of the vessels are identified by single rim

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 145

Pit 212 Bone deposit 218

ie) 50cm 1m feel [ass a a)

Pit 202 Bone deposit 214

Figure 9. Bone deposits in Area M, pits 212 and 202

146 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Table 1. Pottery from Areas D and M

Area D EN Prehistoric? LBA RB Post-Medieval Brick/tile Pit 308 307 - ? 1 6 307/309 - - - 1 309/310 1 310 4

Pit 323 322 2471 3'+ 71

Pit 325 326 - - 3 2 330 14 =

Area M GW Prehistoric? RB Medieval Post-Medieval Notes

Pit 202

203 - - 4 - - slag

206 -

223 4

to

Pit 212 213 215 - - ?1 216 2

216/217 - - - 1 217 1

to rw

sherds. In terms of fabric, they do not significantly differ from the enclosure assemblage, though grog inclusions seem better represented. In terms of form, probably all the rims are rolled (I. Smith 1965, 45, form B), though perhaps with a little expansion in two badly preserved examples.

Only one of the rims is well preserved (Fig. 10, 2) and this is also the only decorated sherd. It has radial lines on the rim top and oblique shallow grooving on the interior. Decorated simple rims are less common (Zienkiewicz 1999). This decorative scheme seems more common with closed or neutral vessels (I. Smith 1965, figs 24-26). In terms of chronology there are clearly decorated sherds from the primary use of the enclosures ditches (especially in the outer ditch) as well as sealed below the outer bank (Zienkiewicz 1999). However, decorated pottery seems to become more common in the higher levels of the inner ditches, which might suggest that the pits reported here are more likely to be contemporary with the outer ditch than the inner ditches. Alternatively, decorated pottery might have been largely restricted initially to the outer ditch and beyond, including the pit complex in Area D.

Most of the Earlier Neolithic pottery is remarkably fresh in appearance, suggesting no prolonged exposure before burial. Very few sherds conjoin and eight vessels are represented by only 22 sherds, suggesting that this collection is only a

fragment of a much larger assemblage. There is no clear patterning to suggest deliberate selection of specific sherds.

Later Neolithic (Table 1)

Three Later Neolithic fabrics were identified:

LN 1: Inclusions of c.15% shell, varying in size from 5mm in size down to barely visible fragments. Very fragile and crumbly.

LN 2: Inclusions of up to c.5% tiny sand and iron oxide. Probably ¢.5% grog (around 1 mm). Occasional shell.

LN 3: Inclusions of ¢.5% tiny sand and iron oxide. Probably c.5% shell. Occasional grog. Very soft and largely dissolves on gentle washing.

As is typical for Grooved Ware from north Wiltshire, this material is very poorly fired and fragile, actually more fragile than the dried soil adhering to many sherds. The assemblage here lacks the better made material which forms a minority of Grooved Ware assemblages (Hamilton 1997, 115). Fabric LN 1 could represent three vessels or more as there is considerable variation in sherd thickness and decoration. Most of the sherds appear to be plain. A small number of sherds have grooves and pushed up cordons but are otherwise plain. Only sherd Fig. 10, 5 has more decoration. This kind of decoration has considerable local parallels (Burderop Down, Cleal 1991, fig. 49.14; Windmill Hill, Hamilton and Whittle 1999, fig. 4.5.34; West Kennet Avenue, I. Smith 1965, fig. 79, 366, and other sites). I have argued elsewhere (Hamilton and Whittle 1999) that this material, with the emphasis on the horizontal, may reflect an earlier phase of Grooved Ware development, pre-dating an emphasis on the vertical and on panel decoration.

Fabrics LN 2 and LN 3 came from the bases of two small cups. Only one out of five sherds was not from the base-angle. This seems a high proportion and deliberate selection of sherds may have occurred. Small bowls are a regular feature of Grooved Ware pits, usually in the form of Woodlands-style Grooved Ware (Wainwright and Longworth 1971, 238), though these are generally larger than the cups reported here.

Sherds of fabric LN 1 occurred in pit 202 and 212. It is not possible to say if sherds from the same vessel occurred in both pits. This material is so fragile that it is unlikely to have been exposed for very long. Parts of the same fragile vessel from pit 212 were distributed in different layers (221 and 216), perhaps suggesting that these reflect tip layers in backfilling.

Grooved Ware seems more common from cuttings across the enclosure to the west and south and less

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE

1

Figure 10. Pottery (see text for detail)

common to the north-east (Hamilton 1999). This notion is supported by the existence of the pits described here.

Later Bronze Age/prehistoric

Fabric UN 1 may belong to a LBA vessel (Fig. 10, 10). There were other possible LBA or prehistoric sherds, largely identified on the resemblance to fabrics found in the enclosure (Hamilton 1999). This material occurred in the upper fill of the Area D pits suggesting LBA activity on the hilltop, as indeed do some of the cuttings across the enclosure (Hamilton 1999).

Romano-British

Small assemblages were recovered from both the test pits and excavations. Three bead-rims of Iron Age -appearance were recovered, though similar material was identified by Annable (1965, 173) as coming from local Savernake Romano-British kilns. Nothing in the assemblage was diagnostically Later Roman or imported from outside Wiltshire. There was a concentration of Romano-British material from the north end of Gibbs Field. In North Field the pottery was concentrated around 0950/1050 and extended in a SSE line towards 1150/0750, with very little material elsewhere.

Sem

Medieval

Only six possible sherds of medieval pottery were recovered, including a well preserved rim and decorated body sherd.

Post-medieval

The small post-medieval assemblage, and to a lesser extent the tile/brick, showed a bias for the middle part of North Field, between 0750/0700 and 1200/0850, in marked contrast to the Romano-British finds. It is possible that part of the tile/brick is Romano-British.

Illustrated sherds (Fig. 10)

1. 308/310 3003. Earlier Neolithic rim. Fabric 2EN 3. Dark throughout.

2. 308/310 3056. Earlier Neolithic rim. Fabric EN 2. Dark throughout. Oblique grooves on interior. Radical grooves on rim-top. Slightly burnished appearance to exterior.

3. 323/322 3333. Earlier Neolithic rim. Fabric EN 8. Dark throughout with a possible burnished exterior.

4. 325/330 3349. Earlier Neolithic rim. Fabric EN 1. Badly damaged during excavation and showing signs of animal damage. Brown exterior and dark core and interior. Thin organic impressions on interior surface.

148

‘Table 2. Details of struck flints from the field and test pit surveys

Flakes Chips = Scrapers Axe Blades Cores Awss frag Field Survey 710 60 42 2 17 28 2 Test Pit Survey 541 238 13 - 4 6 cy Total 1251 298 55 2 21 Sa) 5

5. 212/221 2250. Decorated Grooved Ware sherd. Fabric LN 1. Mixture of dark grey, brown and orange in colour. Decorated with two grooves either side of a pressed up cordon. It has a stroke across this, probably fingernail. At right angles to the cordon is a line of short vertical strokes.

6. 212/221 2263. Grooved Ware base-angle. Fabric LN 2. Buff exterior, buff/brown interior and grey core. Small vessel of 5cm diameter. This is not the same vessel as 2522. Possible seed impression on base.

12 ,212/221 2522: (joins 251 1\ and! 2410). Probable Grooved Ware base-angle sherd, represented by 50% of circumference (but only 2522 illustrated). Fabric LN 2. Dark patches (possible later localised heating) but otherwise orange/brown throughout. Most of the sherds are not evidently decorated, and originally over a quarter of the circumference was probably left plain. 2522 is decorated with three intermittent horizontal grooves. Several depressions (including one on 2522) may be firing defects or fingernail impressions. This is a tiny cup of 6cm diameter.

8. 212/221/222 2423. Grooved Ware rim sherd recovered during wet sieving of environmental samples. Fabric LN 1.Yellow/orange interior and dark exterior and core. A tiny impression below the rim could be the end of a finger-nail impression.

9. 212/221/224 2505. Decorated Grooved Ware sherd. Fabric LN 1. Dark brown throughout. One of three sherds under this finds number which was decorated, probably originally all one sherd. Decorated with irregular grooves, combined with pushed up cordons. Across the cordon is a possible fingernail impression. The sherds are so uneven it is not clear if the lines are horizontal, vertical or diagonal. The illustrated sherd may also have the edge of a boss or other raised feature, though the irregular shape of the sherds makes confirmation difficult.

10. 322 3102 Rim. Fabric UN 1. Brown exterior and interior and dark core. Signs of organic material burnt off the surface. The form of this rim could be Earlier Neolithic, but the flint is rather larger and coarser than normally used. This seems to be the same vessel as sherd 3107, which has a rounded shoulder. This vessel may be akin to the LBA sherds from

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Arrow- Saw Notched Re- Untlised Rejyuv. Knife Hammer Bumt = Fabnc Other heads touched flake stone ator

7 1 2 12 8 13 4 1 181 1

2 1 Il 6 1 334 1

9 1 3 20 14 14 4 2 515 1 1

Burderop Down (e.g. Gingell 1991, fig. 75.1) which feature everted rims and rounded shoulders.

Flint by Ian Dennis

Local raw materials

The Middle Chalk has no flint nodules. Nodular flint from elsewhere in the area appears to be the main raw material for lithic production on the hill. All the lithic artefacts recovered were patinated white or light blue to white.

Surface and test pit survey lithic assemblage analysis

Classifications have been adopted from previous studies carried out within the Avebury area and elsewhere (Holgate 1988a; Clark 1934; 1960). The results from the field survey and the test pits can be seen in Table 2. These figures do not include the results from the excavated areas, or test pit extensions.

Some further analysis was made of the implements, the most reliable indicator of the chronology of the scatter (Table 2). Because of the mixed nature of the assemblage, no systematic analysis was made of the core reduction process, but details of cores are noted.

Cores.

The total number of cores from both surveys is 44. The cores from these samples have been analysed according to the system adopted in the Hurst Fen report (Clark 1960, 216). They consist of: two Al single platform cores flaked all round, five A2 single platform cores with flakes removed part of the way round, three B2 cores with double platforms at oblique angles to each other, eight B3 cores with double platforms at right angles, 20 C cores with three or more platforms, one D keeled core with flakes struck from two directions and one E keeled core. Four burnt cores were unclassifiable.

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 149

The Windmill Hill industry single platform A cores from Keiller’s excavations (I. Smith 1965, 87) account for 40% of the total, but this type of core accounts for only 16% of the total from both surveys. B cores with two platforms account for a further 31% of Keiller’s total, compared with 25% from the two surveys. The most dominant core type present from the field surveys is the C class, with 46% of the total, compared to less than 9% from Keiller’s excavations. From the cores found by Keiller in the primary levels the A type appears to be of an Earlier Neolithic date, while the C cores from secondary levels would appear to be of a Later Neolithic origin. From this it may be implied that there was later occupation across the southern slope of Windmill Hill.

Rejuvenation flakes. Fifteen rejuvenation flakes were recovered during the field surveys.

Scrapers.

There are 55 scrapers from both surveys, of which 43 are flake scrapers. These have been classified using the system from Hurst Fen (Clark 1960, 217) and Windmill Hill (I. Smith 1965, 95). Of the flake scrapers recovered there are: five Al long end scrapers, 14 A2 short end scrapers, seven C disc scrapers, one D1 long side scraper, six D2 short side scrapers, five E scrapers with the bulbar end removed and one unclassifiable. These may be of Earlier Neolithic date because of parallels from sealed excavated contexts. There are 13 small scale-flaked scrapers, probably Later Neolithic to Earlier Bronze Age in date (Clark 1933, 271). These have been classified as six Au short end scrapers, six Aiii round end scrapers, and one B side scraper (Fig 11, 1-8).

Axes.

Two axe fragments were found in the field survey. One is of a flaked unpolished flint axe broken in half with the butt end missing (Fig. 11, 9). It has straight sides with a broad gently curved edge, which has a clear junction with the sides and a medium to thick profile. Adkins and Jackson (1978, 28) classify this form as type H and this example is very similar to their illustration no. 122 from the London area. The second axe consists of a small polished blade fragment of unidentified stone (Fig. 11, 10).

Awils and piercers.

Five awls or piercers were recovered, including both short points with minimal retouch as well as long carefully worked points.

Arrowheads.

The field surveys produced six arrowheads of various types (Fig. 11, 11-16). There were two leaf-shaped points, one class A with bifacial retouch and shallow pressure flaking over the whole of both faces, and one class B with retouch confined to its extreme edges (Clark 1960, 220). Four transverse and petit tranchet derivative arrowheads (Clark 1934) were recovered, which included two class Cl, 1 class C2, and one class H arrowhead. The class C examples have also been classified as chisel arrowheads by Green (1980). The distribution of arrowheads is again in contrast to the 1920s excavation (I. Smith 1965), which produced far more leaf-shaped arrowheads than the later transverse type.

Serrated flakes.

These flakes have been provided with minute denticulations along part or the whole of one or both edges. They have been identified with Earlier Neolithic technologies at Windmill Hill (. Smith 1965, 91). The field surveys only produced one example of this artefact type.

Notched flakes.

Four notched flakes were recovered by the field survey. All have a single notch. Smith (1965, 239) suggests that these artefacts were used as small hollow scrapers, and that they may be of Later Neolithic date due to the number recovered from the West Kennet Avenue.

Utilised flakes.

The surveys recovered 14 examples. Two types of utilisation can be distinguished: class A in which the edges and occasionally the distal ends of the flakes have been blunted by the removal of fine regular spalls at a steep angle, and class B, in which the long edges exhibit irregular chipping as a result of use for cutting or sawing resistant substances (Clark 1934, 121). Out of the 14 examples recovered all but one belonged to the class A category. This is consistent with the results from Keiller’s excavations.

Miscellaneous retouched flakes. Twenty-three pieces were recovered by the field survey.

Knives.

Four knives were located by the surveys. Three are blunted-back knives displaying characteristic steep retouch on the opposite edge to the blade. The other piece, a plano-convex knife (Fig. 11, 17), shows small scale-flaking confined to one side, and is similar to

150

Figure 11. Worked flint (see text for details)

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 151

the Beaker knives from the excavations by Keiller (I. Smith 1965, 109).

Hammerstones. Four hammerstones were found in the field survey.

Fabricators.

Two fabricators were recovered by the field surveys. Both are square in section, showing signs of slight wear and retouch at the proximal end.

Discussion

Overall there was generally a good correspondence between the results of the surface survey and the test pit survey in the distribution density of the lithic debitage. 5% of the test pits had more than 15 worked pieces of flint in total. Cores and implements were moderately represented in their distribution across the survey, with 4% of the test pit assemblage consisting of implements compared with 13% from the surface survey. The large difference in the number of chips between the two collections can be attributed to the use of 1 cm sieving during the test pit excavations.

It was possible to see variations in the density and limits of the scatter to the east, south and west (Fig. 4). Some high concentrations of flint appear to be focused in close proximity to Bronze Age round barrows (Fig. 2). The main area of the scatter seems to be some 300 m by 300m, centred in North Field to the south-east of the causewayed enclosure; the lower density in Gibbs Field to the west may represent the outer limits of the scatter. Within the main area, there were zones of slightly higher density; none was observed immediately adjacent to the enclosure. Variation in flint density may indicate a series of knapping episodes over various areas of Windmill Hill rather than one continuous accumulation of flint working. Core and tool types may suggest that much of the assemblage is of Later Neolithic and Earlier Bronze Age origin. Although Bronze Age material can be shown to concentrate around the barrows, the other periods represented have no discrete locales and are

uniformly spread throughout the scatter, with the density of finds increasing towards the enclosure.

A preliminary model is of activity during the Earlier Neolithic concentrating on or near the causewayed enclosure. Sporadic activity followed on the hill, increasing in the Later Neolithic. This activity may have concentrated in the northern part of the field, but away from the enclosure. This was followed by further activity during the Earlier Bronze Age, with

flint working concentrated around the round barrows built during this period. Such activity may have been intensive for short periods, but does not necessarily signify occupation over a large area at any one time or for a prolonged period of time. It may suggest repeated visits to the hill rather than permanent settlement.

Even at the low densities surviving, over a main area of 300 by 300m there could be some 100,000 or more flints in the plough soil. The original figure is likely to have been higher. In 1928 and 1929 Keiller acquired some 3900 and 2000 worked flints from the hill. If approximately 5% of an assemblage is visible on the surface at any one time and taking the 1928 figure, an original total of around 80,000 implements is indicated. There is of course no certainty that what was sold to Keiller was collected in one year, nor that it was all collected from Windmill Hill. We also have no information on how much of the hill was searched, nor how much was in cultivation in one season. These uncertainties may tend to support an originally higher total.

The Neolithic pits

The pits were tentatively located by geophysical survey and plough soil excavations were conducted within the vicinity to locate them precisely. All flints were recorded in 1 by 1m collection units. Sixty-four 1 by 1m squares were excavated in area D and 61 in area M.

Area D: ploughsoil

511 pieces of worked flint were recovered from Area D which was located above the Earlier Neolithic pits. These included narrow unretouched blades, miscellaneous retouched flakes and flakes showing signs of utilisation. There were also six cores (one C with the rest unclassifiable due to post-depositional damage) and a rejuvenation flake. Six scrapers were recovered, one A2, one B2, three small scale-flaked of type Aii and one unclassifiable. The surface excavations also produced one flaked stone axe, whose source has not yet been determined by petrology.

Area D: features (Table 3)

Pit 308.

There were 95 pieces of worked flint, with 17 implements. Only 18 pieces, including three implements, came from the primary layer (310), and the rest came from the secondary and tertiary layers

152 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Table 3. Worked flint from Area D features

Flakes Chips Blades Cores Arrow- heads

Pit 308

306 8 - 1 - -

307 33 3) 2 4 2 309 4 6 1 - -

310 6 - - - -

316 2 - 1 - - Total Be} Di 4 2

Pit 323

B22 29 2 1 z) -

326 25 5 2 - - Total 54 7 3g 3 -

Pit 325

330 20 5 1 1 -

(306, 307, 309). 310 produced rejuvenation flakes and a knife. The knife was made from a secondary core preparation flake. The proximal end forms the back of the knife, while the distal end is retouched as the cutting edge. The blank for this tool was produced by hard- hammer technique, and the bulb of percussion shows signs of trimming and reduction (Fig. 11, 18). The implements from the upper fills included narrow blades, a utilised flake, cores (one B3, one C and two unclassifiable) and rejuvenation flakes. 307 also contained another smaller knife and two arrowheads: a class D PTD, and a barbed and tanged similar to F177 in Smith (1965, 109) (Fig. 11, 19). The barbed and tanged arrowhead was stratigraphically higher than the PTD, possibly suggesting that the pit was only partially backfilled then left to silt up over a period of time.

Pir323.

There were 57 worked flint pieces, which included nine implements. These included a narrow blade, cores (A2, B2 and C), rejuvenation flakes and a pointed flake with some signs of serration along both edges. The main finds from the pit were two burnt knives, one broken and the other complete with a conjoining thermal fractured flake (Fig. 12, 1). The broken knife shows minimal reworking on the ventral surface, while the dorsal surface shows abrupt retouch along the back of the knife and the remaining distal end has invasive retouch. The other knife has minimal retouch on the ventral surface, while the working across the dorsal surface could have produced a double-edged implement. There were 15 other pieces of fire-cracked flint.

Saw Utilised Rejuv. Knife Burnt Other

flake z iT = = Le - 1 1 6 - 1 : 3 - - 2 1 9 - = 1 = = = 1 D) 2 18 - - 2 2 15 3 1 : 3 5 é I 2 2 20 3) - 2, - 24 1

Pit 324. There were two flakes from the uppermost fill (327).

Pit 325.

There were 93 pieces of flint, including blades, a utilised flake, a serrated flake , an A2 core and rejuvenation flakes. There were also 29 burnt pieces. The serrated flake, worn and of hard-hammer technique, is from the top of the upper fill (326).

Posthole 315. There are two flakes, one blade and a rejuvenation flake, all from the upper context 316.

Area M: ploughsoil

A total of 442 pieces of worked flint were recovered from the ploughsoil in Area M. There were a narrow unretouched blade, three miscellaneous retouched flakes and three utilised flakes. There were three cores (A2, B3 and C ) and two rejuvenation flakes. Four scrapers were also recovered (2 A2, 1 D2, and 1 unclassifiable), along with four awls.

Area M: features (Table 4)

Pit 202.

There were 201 worked flints, including narrow blades, retouched and utilised flakes, three cores (one B3 and two C), a rejuvenation flake and a hammerstone. There were also 15 burnt pieces. The majority of the waste flakes were hard-hammer struck.

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 53)

Pit212.

There were 404 worked flints. Included in this total are blades, utilised flakes, serrated (Fig. 12, 2-4) and retouched flakes. There were seven cores (three A2, two C and two unidentifiable), rejuvenation flakes and a hammerstone. There was also a flaked flint axe and an adze (Fig. 12, 5), both broken and with signs of heavy burning. Three knives (two blunted-back knives and a sickle knife similar to F57 in I. Smith 1965, 97) (Fig. 12, 6-7) and seven scrapers (three A2, two C, one D2 and one unidentifiable) were also present. Most flakes show signs of hard-hammer technique.

Discussion

The flint from all primary pit contexts was in mint condition though with a light grey patina. Material from the upper pit fills was generally worn with a heavy white patina. Patches of calcium carbonate concretion from groundwater leaching through the soils were

deposited on flint from all contexts; this was particularly heavy on those pieces from primary layers. Similar effects were also noticed on artefacts recovered from Robin Hood’s Ball near Stonehenge (Harding 1990).

The presence of cores, rejuvenation flakes and small pieces of debitage suggests that flint knapping was carried out over the site as a whole. All pits except pit 324 contain evidence of core reduction, although very few conjoins could be recognised. Some conjoins were found within discrete contexts, but none between contexts within individual pits. However, find no. 2384, a broken flint flake from the middle fill 219 in pit 212 conjoins with find no. 2781, a broken retouched flake from the lowest fill 223 in pit 202, and together these may form a scraper or a small knife.

Pit 323 shows signs of ash in the lower fill (322), and it has been suggested that this was due to deliberate burning of cereals as a part of ceremonial consumption, and placed with other symbols and

Figure 12. Worked flint(see text for details)

154

Table 4. Worked flint from Area M features

Flakes Chips Scrapers Axe Blades Cores frag

Pit 202. 203 20 15 - - 3 204 15 3 - - 1 - 204/206 1 - - - - 1 206 49 27 - - 3 1 210 1 1 211 3 es 223 22 3 1 1 233 2 - Total 113 49 8 3 PrzgiZ 213 39 22 - - 2 215 17 10 4 1 216 44 26 - 1 5 1 216/217 1 - 217 14 d 1 1 219 17 4 1 2 220 6 5 - - 221 49 22 2 3 2 222 20 6 - 3 224 7 4 - - Total 214 103 7 1 14 7

Arrow heads

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Saw Re- Utilised Rejuv. Knife Hammer- Burnt Adze touched flake stone 1 1 1 - 7 3 2 1 5 3 1 1 1 2 7 4 1 1 1 15 4 1 8 1 - 1 1 2 - 3} 2 1 1 1 2 1 2 1 5}

to ey ee

& i) S Gy = nN oS nN

residues of consumption, such as quern stones and animal bones. The burnt flint implements deposited in the pit, as tools associated with harvesting of grain and butchery of livestock, could themselves be further symbols of such consumption events.

Acknowledgements

Thanks are due to David Gilbert and Howard Mason for invaluable help with both finds and text, and to Lisa for support.

Worked sarsen by Joshua Pollard

Area D pits

Three substantial pieces of worked sarsen were recovered from the Earlier Neolithic pits: a quern fragment from pit 308 (context 310) and two intact rubbing stones from pit 325 (context 330) (Fig. 13). The quern fragment comprises a roughly flaked block with one, slightly concave, surface exhibiting extensive pecking and use-related smoothing. Clearly a substantial fragment of saddle-quern, the implement is similar to Earlier Neolithic examples from the enclosure (I. Smith 1965, 121-23;Whittle er a/. 1999, chapter 15). The two rubbing stones were found placed together in pit 325. Both appear to have been produced on large sarsen flakes. Minimal trimming through secondary flaking is evident on the dorsal

sides of both, whilst the convex ventral (working) surfaces show extensive pecking and abrasion. The smaller of the two (3309) could have been worked easily with one hand, whereas the larger (3308) would have required both hands to operate. Comparable examples are again known from the enclosure.

Area M pits

From pit 212 (context 221), there is an intact hammerstone, nearly spherical and with extensive signs of use, and a fragmentary example.

There were also quantities of unworked sarsen. Much of the material from the Grooved Ware pits in Area M is highly fragmentary and burnt, perhaps being used as hearthstone.

Test pits

A single hammerstone on a small sarsen lump came from 1400/750.

THE ANIMAL BONES by Jessica J. Davies

In Area D, three of the four Earlier Neolithic pits contained animal bone. The assemblage derived from 308, 323 and 325, whilst the larger pit, 324, yielded no animal bone. In Area M both of the Later Neolithic

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 155

Figure 13. Quern (centre) and rubbing stones from the Earlier Neolithic pits

pits excavated contained bone in their fills. All the animal bone material was retrieved by hand during excavation.

Methodology

The bones from each context were separated into identifiable and non-identifiable specimens. The identifiable specimens were those which possessed diagnostic characteristics allowing them to be identified to body part and species. Some bones could not be assigned to a single species, and grouped categories of similar species were also used, such as cow/horse. Non-identifiable specimens were those fragments that were too badly preserved, or lacking diagnostic characteristics to be able to ascertain body part and/or species. These were not recorded or quantified. Identifications were made with reference to comparative modern and archaeological examples at the Department of Archaeology and Prehistory, University of Sheffield (details held in archive).

No bones of goat were identified, but where possible sheep were distinguished from the sheep/goat grouping. This was accomplished with the use of Boessneck (1969) and Prummel and Frisch (1986), as well as of comparative skeletal elements.

Wear patterns of mandibular teeth were recorded: for pig, after Grant (1982) and Halstead (1992), and for cattle Grant (1982). Few teeth that could be recorded in this way were present, with no sheep/goat teeth being found. Thus this data will provide an idea of the age of some of the animals present, but will not be used to produce mortality profiles. Fusion stages were recorded following Silver (1969), and will also be used to look at the age of the main domesticates. Pig canines and/or their sockets were used for sexing mandibles. Although anatomical features can be used to determine the sex of other species, the appropriate characteristics were not present on any of the diagnostic zones.

All dental, cranial and post cranial fragments identified were recorded as minimum number of anatomical units (MNAU), where proximal and distal halves serve as separate anatomical zones. The effectiveness and reliability of other methods of quantification have often been debated. Various authors such as Grayson (1978), Klein and Cruz- Uribe (1984), Lyman (1992), O’Connor (1985), Payne (1972) and Ringrose (1993) have scrutinised methods such as MNI, NISP, bone and meat weights, and so on. With regard to this assemblage, bone and meat weights would be unrepresentative,

156 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

being affected by the small number of fragments. NISP has too many inherent problems (described by Ringrose 1993, 125-26) to be of use here. MNI is favoured by many archaeozoologists. However, at the present time this would not be a suitable method to use in quantifying the remains from each pit. It is possible that the deposition of the bones of animals was not exclusive to one pit, with body parts being divided between them. Thus in calculating the MNI for each pit, numbers of animals present could have been over-estimated. MNAU was considered the best method to quantify the assemblages from the Earlier and Later Neolithic pits. Over-representation is avoided with this method, as well as any possibility of the proportion of any one species being exaggerated.

State of fusion and the side of the bone were recorded. Fragmentation was recorded indicating the prevalence of old and new breakage. Gnawing, burning and butchery provide additional information about the use and treatment of the bones before deposition. Binford (1981) was used as a standard for the description and position of different types of butchery marks.

All measurements taken follow von den Dreisch (1976). Few bones were complete enough for measurements to be taken. Much of the bone was weathered and thus not in a particularly good state of preservation, but measurements were taken of as many bones as possible. The measurements displayed in

bold type in Table 8 were analogous with those taken and recorded by Grigson (1965) on material from Windmill Hill causewayed enclosure.

Species utilised will be considered and, whilst there may not be enough suitable remains to reconstruct the subsistence strategies employed, the deposition of the assemblage will be discussed. It already seems evident that, during the Later Neolithic period, more wild species were being utilised than before. This seems to be a common trend, seen also in the analysis of the plant remains (below).

Results

Earlier Neolithic

Few bones were present within the contexts from pits 308, 323 and 325 (Tables 5-6). As the assemblages from these pits are so small they are considered collectively. Domestic species predominated, particularly cattle, with red deer being represented by a single antler fragment in 308. The body part representation suggests predominance of particularly lower limb bones in 323, head and more extreme limb bones in 325, with odd fragments in 308 (Table 5 and Fig. 14). This may indicate that the deposited bone was predominantly the product of the secondary stage of carcass reduction. There seems to have been no deliberate or ritualistic selection of particular body

Table 5. Assemblages from the Earlier Neolithic pits

Pit 308 Cattle

Pit 323

Red deer Cattle

DMt - -

Ph 1 - - - Antler /H - 1 - Occip C - - - MdH - - - Md - - - MdT 1 - 1 MxT - - -

Pit 325 Pig Sheep/ Cattle Pig Sheep / Goat/Roe Goat l . “ é 3 2 s 1 2 1 2 1 S l = E cs 1 = Z i 1 z = hs 1 1 ZS

Cattle Hors¢

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE NB}/

Table 6. Total assemblage from the Earlier Neolithic pits

Cattle Pig Sheep/ Cattle/ Sheep/Goat Red deer Goat Horse Roe deer Total number 13 3 2 1 1 1 of fragments Total% 61.9 14.3 9.52 4.76 4.76 4.76

parts or bones. The bones may have been deliberately placed or disposed waste fragments. Eight of the ten bones recorded from pit 325 were from the left-hand side of the animals; the other two were of indeterminate side.

All the bones present were older than new born (Table 9). One cattle proximal tibia suggested an age at death of less than 30-42 months. The fusion data for pigs and sheep/goat were indeterminate. No tooth wear data could be recorded for any of the species. One possible female pig was noted from 325.

No butchery marks or gnawing were evident on the Earlier Neolithic material. Much of the bone that

could be identified was weathered. This may have masked or destroyed any cut or gnaw marks, particularly the latter. Whilst weathering may have occurred in the deposits, the fact that not all the bone was weathered suggests that there was either a variation in the burial environment or that some bones may not have been buried immediately and exposed to the elements before deposition. Burning was recorded on four bones from 325; one of cattle, three of pig. This may have been the result of cooking or disposal of the bone. Both old and new breaks were recorded. No whole long bones were seen, indicating that bones may have been broken as part of their disposal, or to remove marrow and fat deposits. New breaks can be attributed to excavation and storage of the assemblage.

Only one measurement could be taken. The distal breadth of the metacarpal (Table 5) is compatible with the range previously recorded for domestic cattle from the Windmill Hill enclosure (Grigson 1965).

ea ae ae naa etn se pease Se Same cn eecl et cee eeacss-ca sete csees° — —= — — — —= == = = > = = < = — = = = = = = = = —| = = = 1 tex 5 : os Sy 50) oy RS xs Be NY Xo S is) ix? NY Bo NS +, ae ry ty Ss oc) S [Xo Sf Ss ae S eet S et N Se) (xX? SN <3 KS) N +e tat ae S cx> J ve NY eee Lx cH S KH 2) KS SN LA 0 Co tS (Ye oO

DR PMc DMc

PF

i Red deer £9 Cattle / Horse A Sheep / Goat / Roe |

E] Sheep / Goat | { B Pig | | i Cattle = ti ag ae ae ue ae ae aa ae on as os oe =n on Lis ae as ae ae oe on as Lil oe ae ue as =S S$.2 2s © $2 * § 2 Be Ot g = fs 2 3 = 3 2 A te eS) 5 <£o4 << Sanice 5 os he = Saws" Cmca 3 2a 8 oO a fo) 4 2 5 Oo

Body Part

Figure 14. Body part representation for the whole Earlier Neolithic assemblage (pits 308, 323 and 325)

158

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Table 7. The animal bone assemblage from the Later Neolithic pits

Pit 202

Cattle Pig Sheep Sheep Dog Cattle Sheep /Goat

/Goat /Red deer /Roe

2 =] wre

PF

DF 3 = E PT 2 : 3 DI 2 : 3 PM 3

DM

U = PMp : \ DMp e 1 ¢ a ~ A Phil

1 Phil 1 - - Ph UI 3 1 1 Nav Cub 1 Carpal 1 Tarsal 1 Patella - 1 Ses 1 - At 1 1 Ax Cervical vert Rub - - 1 Antler/H B = 1 Scaph 2 Oceip C 1 MdH - - Md - 2 - - = Mx rl = - 2 1 Md T 6 2! - - 1 Mx T 2 6 = : 1 Column total 62 18 3 Column %

wr oo x

Later Neolithic

Pit 202.

The assemblage from this pit is dominated by domesticated species, principally cattle, followed by pig (Table 7). Even when sheep/goat and sheep categories are combined they are still less prevalent than pig remains. This is the only pit in which the remains of dog were encountered.

Most body parts of domestic cattle are represented, suggesting the presence of fore and hind limbs, ankle, wrist and foot bones, as well as areas of the skull around the teeth and vertebrae closely associated with the head (Fig. 15). This may indicate that cattle carcasses were being reduced at, or very close to the site of the pits. Main limb bones, which bear the most meat, were also present, suggesting the material deposited was not purely waste from butchery of carcasses. Bones of the head and other areas, such as ribs and vertebrae, were limited in number, suggesting they may have been deposited elsewhere or not utilised. The presence of loose teeth, for this and other species, indicates that there must have been a considerable loss of bone at some point in the history of the material. This bone loss may have occurred prior to deposition, through butchery or the extraction of marrow in the mandible, or during deposition, particularly if the bone had been churned

Pit 212

Cattle

Cate ‘Aurocl

Catue

Red deer

Pig Sheep Wild Goat pig

Sheep Goat/Roe - > - : 2

1 2 1 1 = : 1 1 2 : : = 1

bee ere tre

1 15 2 2 2 30 21 4 3 4 3

26.0 18.3 3.5 2.6 BIS) 2.6 3.5

va ete At Oe eo) a eee Cea Dod) det teen atthe cto Th aiegh Ay Hieeirenbaie 1 Q

and mixed in redeposition from middens. Alternatively bone loss may have occurred as a result of the material becoming friable whilst buried, with subsequent excavation adding to the loss. Similar loss of bone was reported from Cherhill (Grigson 1983, 71).

For domestic pig, only two counts of a main limb bone were recorded, which may have been from a single animal. The body part representation for this species indicates the presence of bones predominantly from the extremities; these may be associated with butchery waste of animals that may have been used for ceremonial feasting elsewhere (Fig. 16). However, they could alternatively represent parts of animals consumed close to the point of deposition.

Fig. 17 shows the body part representation for sheep/goat. The primary meat-bearing limb bones were present, with fewer bones from the extremities. This suggests that this material was the possible refuse from consumption.

The bones from other species were limited in number and little can be interpreted from these (Table 7). It is possible that these bones may be butchery waste, or equally the result of accidental deposition. These are also plausible suggestions for the presence of the bones of dog.

Over half of the post cranial cattle bones were from new born animals (Table 9). About one quarter of the combined sheep/goat and sheep bones were

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 159

=

MNAU

0 :

ee ee ee Seg) psa aoe eet 14 DOR UE PE SE CS Sages eet ae ee ea ae a eet 3 23 8 = = =

2, 5 < ° 5 =) Body Part

Figure 15. Domestic cattle body part representation (pit 202)

MNAU

a t as a 2 Oo - Shey Ce Caer RL San Epes) eS ise re Ee ee DORR ES che ie eds melee IS 4 Se Sa ie BiG ee beaeoo oe eens gosads: SE ey oe ~ a A = 8 ors) 3 a8 8 tS s= a a oO Ss fo) cs = < 38 Body Part

Figure 16. Domestic pig body part representation (pit 202)

160

neonatal. The post cranial bones of the other species were all older. The fusion and tooth wear data add further information to this. The tooth wear patterns for cattle indicate that at least one individual died between 18 and 30 months, whilst another was over 36 months. For pig, one individual was less than 6 months old at death, and another individual was possibly only weeks old. This was indicated by the presence of a deciduous fourth premolar, which had not fully erupted. The eruption time for this tooth in pigs is between 1 and 7 weeks. This individual may well have been a neonate at death. The fusion data indicate the pattern set out in Table 11.

Whilst more bones were from the left-hand side, an emphasis on the deposition of bones from one side only does not seem to have prevailed in this pit. No data regarding the sex of individuals were present.

Only one bone displayed a cut mark. This was on a cattle proximal radius from context 206. One bone fragment showed signs of gnawing, but again the weathered nature of much of the bone material would have made gnawing and cut marks difficult to distinguish if present. These marks may have been eroded during weathering also. Two identifiable bone fragments were burnt, suggesting that methods of

MNAU

Ber 22 Ars vhs a. a

Figure 17. Sheep/goat body part representation (pit 202)

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

disposal or cooking that may result in burning were not common practice.

Whilst only one long bone was recorded as being whole, the majority of small bones remained intact. More old breaks were seen on the long bones, suggesting breakage for marrow/fat removal, or as part of the deposition process. New breakage is again attributable to excavation and storage and more small bones had new breaks than old.

As Table 8 shows, for each element the majority of measurements taken were from one bone. Thus ranges of measurements could not be produced, nor statistical methods used to examine the data. In general all the bones fit within the ranges given by Grigson (1965).The size of the pig atlas was compared with measurements from West Kennet (Edwards and Horne 1997, 123). The measurements of the specimen from pit 202 were smaller (4.1mm (height (H)) and 2.1mm (greatest breadth of the facies articularis cranialis (BFCR)).

Pit 212.

Again domesticates dominate the assemblage from this pit. Cattle were predominant, though less prevalent than in 202, with pig and sheep/goat

Lote Peto eer ar eH ameter la prin oP tp ey deste mc Ved 1d aie Big 2 G8 eee 9S ye ee Fe fel ee see > > 56 & 5 Bice ena z e g &

Body Part

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 161

Table 8. Measurements from Earlier and Later Neolithic bones. Measurements in bold type are analogous with those taken from Windmill Hill, reported in Grigson 1965.

EARLIER NEOLITHIC: Pit 325: GATTLE Metacarpal BP BD GL SD = 513: - = LATER NEOLITHIC: Pit 202: CATTLE Metatarsal BP BD GL SD = 528 - & Ulna DPA SDO LO - 619 Calcanea - - GL GB = = 1290 430 = - 1380 440 Phalanx I BP BD GL SD 280 253 629 - Phalanx III LD MBS DLS - 529 207 706 - 518 208 - - Atlas LAD H GL BFCR = - = 1033 SHEEP / GOAT Humerus BP BD GL Bis = 271 - 266 Radius BP BD GL SD 286 270 1550 150 Tibia BP BD GL SD - 244 - 139 PIG Atlas LAD H GL BFCR 170 410 - 532 Pit 212: CATTLE Metatarsal BP BD GL SD = = - 210 Phalanx I BP BD GL SD 286 276 579 250 Phalanx II BP BD GL SD 289 - 414 221 * = 2 04 B.p. 407 57 3 B.p. 405 345 570 308 Axis BFCD H LCDE BFCR 437 - - 887 Horn core - GB GL - - 574 1700 -

* Possible Bos prinigenius specimens.

162 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Table 9. Proportions of Neonatal and Older Bones (Later Neolithic pit 202)

Domestic Domestic Sheep/ Cattle/ Sheep/Goav Row Total Cattle Pig Goat Red deer Roe

New-born 27 - 4 : - 31 50.9% 25.0% 38.8%

Older 26 th 12 3) I 49 49.1% 100.0% 75.0% 100.0% 100.0% 61.3%

Column Total 53 fi 16 3 1 80 66.3% 8.8% 20.0% 3.8% 1.3% 100.0%

following. These were proportionately greater than in 202. More wild species were identified from this pit than from any other contexts. Thus the diversity of species present is also greater (Table 7).

For domestic cattle, the main limb bones present were those that bear the most meat. Some bones of the feet, neck and head were also identified. Elements such as cervical vertebrae, rib, horn core and mandible, which were not present in pit 202, were identified. As only one of each of these elements was present, deposition of these parts into one pit rather than the other cannot be claimed. Parts of the head are limited, although the overall skeleton is less well represented than in pit 202. This material may indicate the refuse from butchery for consumption (Fig. 18).

Domestic pig body parts were present in relatively low numbers, apart from mandibular teeth (Fig. 19). Upper limb bones only were indicated, with some foot elements. This material indicates a limited usage of pig meat, with possible deposition of waste from the butchery of carcasses, at least parts of which could have been consumed and deposited elsewhere.

The material from pit 212 shows a predominance of upper limb bones for sheep/goat (Fig. 20). Pelvis bones and a metatarsal, as well as rib bones, were also present and the remains may be indicative of the refuse associated with consumption of meat in the vicinity of the pits. The predominance of forelimb bones, however, suggests that the deposits may have been structured in their content and deposition.

Fig. 21 shows the presence of the body parts of other species. Many of the elements identified suggest that wild species may also have contributed to the meat available for consumption. Humeri of wild pig and aurochs suggest the use of prime meat parts, although no butchery marks were evident. The

presence of red deer antler may indicate that this was collected for use as a raw material. It was not possible to tell if the antler had been shed or removed from a hunted animal.

A much smaller number of the bones from this pit were neonatal (Table 10). These were from cattle, pig and sheep/goat only, with sheep/goat having the highest proportion at 28.6%. The tooth data suggest a range of ages for both cattle and pig. For cattle only one tooth could be placed in an exact age stage, F (young adult, 1.e. >36 months). Another tooth was in wear at a point between stages D and E (18-36 months). Other teeth were recorded as being at stages B+ , C+, D+ and E+. This places many of the teeth at various points between 1-8 months and senile age. Although there is a lack of specificity of ages, the tooth wear data suggest that cattle were being killed at a variety of ages from as young as 1-8 months, up to > 36 months and possibly older. For pig the ages recorded were very young (i.e. 1-7 weeks), < 6 months, 6-12 months and + 24 months. Thus individuals were being killed generally within the first year of life and around 24 months. The fusion data add little information. For sheep/goat the fusion data were of indeterminate age stage. For cattle and pig, the information is set out in Table 11.

Only one possibly female pig mandible was recorded. No other remains with sexually diagnostic features were present.

Cut marks were noted on three bone fragments. One, a cattle/red deer metatarsal, had been worked into a point. A cattle/red deer mandibular hinge displayed cut marks that may have resulted from dislocation of this joint or removal of meat from the head. A cattle first phalanx displayed cut marks that may have been allied to the dismembering or removal of the foot. No gnawing marks were noted and only a fragment of red deer antler was burnt. Again much of the material was weathered.

Few whole long bones were seen, with the vast majority of breakage being old. About a third of small bones had old breaks, but only two had new breaks, indicating that marrow may have been utilised.

Table 10. Proportions of Neonatal and Older Bones (Later Neolithic pit 212) Domestic Domestic Sheep/ Boar Aurochs Cattle/ Cattle/ Sheep/ Cow/ Red deer Row Total Cattle Pig Goat Horse Red deer Goat/Roe Horse/ Aurochs New-born 3 1 6 - - - - 10 12.0% 6.7% 28.6% 12.5% Older 22 14 15 2 4 2 4 2 2 70 88.0% 93.3% 71.4% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 87.5% Column 25 15 21 2 4 2 4 2 2 80 Total 31.3% 18.8% 26.3% 2.5% 3.8% 5.0% 2.5% 5.0% 2.5% 2.5% 100.0%

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 163

wn

MNAU

Rub ff

DoD & 2 0 rete Tied: edd SP eS Gx SON eh a 6 EC ae ; & z is 2. a § 5 Ss g y oO Body Part

Figure 18. Domestic cattle body part representation (pit 212)

MNAU

Ses

At Ax Rib

Antler/H Scaph Occip C MdH

Nav Cub Carpal Tarsal Patella

Cervical vert

Body Part

Figure 19. Domestic pig body part representation (pit 212)

164 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

w

25

MNAU

0.5 4

a See £82 0< 27S 8 ae Gomis BS E29 BE BEE e bE 5 J ee ee ee be ee Se Z é < GQ Body Part Figure 20. Sheep/goat body part representation (pit 212) 6 Ged deer | ‘Si Cattle / Horse / Aurochs | i eer slow de Mad oe eda es ere ia elena si ed eS oe Me Ae one 2G eM CEN oy 2 Me I oe oe let, A OR | 2 Sheep / Goat / Roe | ‘Cattle / Red deer | Ei Cattle / Horse | 4 = BREE ESSE EERO eer es me Ce See SE er i ee eae ne eI ts 5 se r hs N A | N Hy | Wild pig fl >| N i 3 7 oe S a Nc Re eee OR OR cee A i> 2 Mp SO onan fee PD aiesee Pa EEL ec at cle DPE Reon, I oe Ree ao RE kN 8 ee ee ee Fee range nea cee Spee yan sean ou: eee l om mjtoy ine, cleere wigieielsiayerdalac wn 2 = == == == Kt al on =l= ee 4 » ee ; =A= EH 3 7 Yaa olin ta teat ent ee = moe Se0¢ 522 8a Sf 8 = BSEEOT PSHE Rae eee er Sg Se Sea Gaede eae z 3 < °o 8 Body Part

Figure 21. Body part representation for other species identified (pit 212)

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 165

Table 11. Fusion data from Later Neolithic pit 202

No. of bones Age at Death Cattle 1 <6-10m 1 <13-16m 2, >6-10m 2 >18-28m 2 >30-42m Pig 1 <6-10m <13-16m 1 >18-28m Sheep/ 3 <30-42m Goat 3 >6-10m >18-28m >30-42m

Table 12. Fusion data for cattle and pig from Later Neolithic pit 212

No. of bones Age at Death Cattle 1 <30-42m 3 >13-16m 1 >18-28m Pig 1 <6-10m 3 <13-16m 2 >6-10m

Measurements were compared with those in Grigson (1965). Again the majority of measurements were within the ranges from the Windmill Hill enclosure. Two possible aurochs second phalanges were identified during this comparison (Table 8). The measurements for proximal breadth are greater than those given by Grigson (1965, 159). Whilst size should not be used without care to distinguish wild from domestic cattle (Grigson 1969, 288), the much greater size suggests that these phalanges are more likely to be from wild cattle. The two phalanges were enantiomorphic, suggesting they were from one individual. The breadth of the trochlea for both ‘aurochs humeri was greater than those of domestic cattle at Windmill Hill. The breadth of the wild pig trochlea was also greater than comparative specimens in Grigson (1965). This specimen was also much larger than the measurements presented by Payne and Bull (1988) for wild pig and the wild pig identified from the Neolithic pit at Puddlehill (Grigson 1976, 16). The lateral metapodial from a wild pig was larger than the comparative modern boar specimen

examined. Measurements of this element are not considered by Payne and Bull (1988) and thus other factors such as age, sexual dimorphism and individual variation may have affected the size of this metapodial.

Discussion

As so few of the domestic cattle bones were measurable, both from the Earlier and Later Neolithic, little suggestion can be made regarding the process of domestication of this species. What is evident is that by the Later Neolithic, at the latest, there was a distinct size difference between aurochs and domestic cattle. Overall, domestic cattle and pig seem distinct from their wild counterparts, by the Later Neolithic, evident in the differences in size.

The tooth wear information and fusion data are somewhat scant and seem to provide little indication for the seasonality of occupation or use of the pits at the site. There is a total lack of any ageing data from the Earlier Neolithic pits. The presence of very young pig teeth, in pits 202 and 212, indicates these animals were being killed soon after birth. To infer seasonal activity must depend upon the acceptance that domestic pig in the Later Neolithic were producing young at a similar time of year to their wild counterparts (i.e. in spring, around April). However, pigs can be highly productive, producing more than one litter in a year. Therefore it would be rather erroneous to try to use the presence of neonatal remains to indicate seasonal activity. The presence of bones of older animals, suggested by tooth wear and fusion, generally of mixed ages for all three main domesticates, further suggests that consideration of seasonality may not be necessary. Grigson warns against the use of such data to infer seasonal occupation at similar sites, suggesting that there may be a great variation in tooth wear due to variable eruption and birth times (1966, 85). The spread of ages indicates that activity, whilst not necessarily permanent occupation, may have occurred in manifold episodes during the year. Exploitation of domesticates was not confined to distinct age groups. The data available cannot be refined to give a more accurate conclusion upon this point.

Further, it may be inappropriate to suggest seasonal activity if the bones were contributing to midden accumulations prior to deposition in the pits. Whilst Legge notes that many bones from chalkland burial environments display considerable surface erosion (1991, 54) it is possible that middening occurred at this site. Therefore the whole process of use and deposition of the archaeozoological remains

166 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

considered here may not reflect a single season’s deposition or ‘seasonal’ activities at all.

The paucity of tooth wear data and sexable bones reduces what can be inferred about husbandry practices. For the Earlier Neolithic, it is possible to say little more than which animals were being kept or utilised and deposited in the pits. It is possible that domestic cattle, pigs and sheep/goat formed an important display within social exchanges occurring in the vicinity of Windmill Hill. These domestic species may have functioned as a symbolic resource, although the small amount of bone retrieved sheds little light on their social function. It is likely that at least some of the animals may have been used for their meat and marrow before deposition. By way of comparison, the Earlier Neolithic bones from Coneybury are suggested to represent primary butchery of domestic cattle. Equally, upper limb bones were infrequently encountered, with dumps representing possibly a single butchery event. The lack of upper limb bones in the Earlier Neolithic pits to the south of Windmill Hill may indicate a similar situation, with butchery to a secondary stage, occurring around these. The jointing of bones for food is also interpreted at Durrington Walls (Stone et al. 1954). The predominance of cattle bones fits with the accepted view of the predominance of cattle in the area in the Earlier Neolithic (Thomas 1991, 163). The presence of red deer antler may suggest that some wild resources were still utilised and important at this time. The division between upper and lower limb bones was not so distinct in the Later Neolithic assemblage. It is suggested that the bones from the Later Neolithic pits at Down Farm had a primarily social function, with deliberate selection and deposition of skull and jaw bones, as well as of large pieces of bone.

The presence of bones of neonatal sheep/goat, pig and cattle suggests that animals may have been born at, or very close to, the site. Bones of a very young calf and pig were also recovered from the ditches of Windmill Hill (jope 1965, 145). As male cattle are generally absent from enclosures, it has been suggested that culling at a young age may be responsible for this (Thomas 1991, 24). It was not possible to tell the sex of the sheep/goat or cattle to determine if these were the young males, culled as surplus to requirements in a dairy economy. The lack of tooth wear data means that a mortality profile could not be produced to test this proposal. The presence of juvenile bones has been used to infer dairying at Fussell’s Lodge long barrow (Grigson 1966, 85-86). It can be seen that animals were being utilised throughout their investment phase and into

adulthood. Pigs serve little purpose other than to provide meat and fat. Molluscan and soil evidence suggests that the area around Windmill Hill was not extensively cleared in the Earlier Neolithic (Whittle 1996, 272; Whittle et al. 1999) providing suitable pannage for wild and domestic pigs.

Edmonds suggests that enclosure sites probably acted as congregational places for dispersed populations, with gathering events being closely related to the husbandry of livestock (1993, 105). Although the pits were in a location removed from the enclosure, the area around may have been influenced during events at the enclosure. At least some of the remains present within the Earlier Neolithic pits indicate that animals were probably being utilised, at least in part, for their meat. Whilst domestic pig predominate in the assemblages of many ceremonial sites (Thomas 1991, 22), their presence in the Later Neolithic pits may represent the use of some pig meat outside an enclosure context, as well as deposition of waste elements from primary, and possibly secondary, butchery of carcasses.

Domestic cattle predominate in the assemblages from both Later Neolithic pits, followed by pig and sheep/goat. At Down Farm, pig are the most common species ((Legge 1991, 65). The predominance and proportions of the main domesticates in the pits at Windmill Hill are comparable to the remains from the Windmill Hill enclosure ditches (Jope 1965, 144; Whittle 1996, 235). Horse could not be distinguished, although the identification of specimens to cattle/ horse, may suggest that the species was present. The presence of this species has been considered to be practically certain during the Later Neolithic (Legge 1981, 80). The remains of wild animals, other than red deer antler, were present within pit 212 only. Remains of similar species were found within the Windmill Hill enclosure, but it was thought that hunted animals contributed little to the diet of those at the site (Jope 1965, 144-45). Admittedly, the number of remains of wild species is low, but their presence indicates that they may have been a more important utilised resource than in the Earlier Neolithic. Large hunted animals may have been butchered where they were killed, with only a few bones left attached to the meat and therefore being transported to the site.

The use of animals may indicate something of their social significance. Throughout time, food has been used to express social values. The consumption of meat communicates a shared set of meanings, protocols and behaviour; communal eating and drinking form bonds of friendship and obligation

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 167

(Fiddes 1991, 34). Cut marks were few, although this may be a factor of the weathering of bones. The presence of bones from meat-rich parts of the bodies of many of the species present suggests the use of meat for food. Fragmentation of limb bones prior to deposition suggests that the extraction of marrow was probably practised in the Earlier and Later Neolithic, and that many animals present were being used in a rational food procurement strategy. Butchery of carcasses of domesticates may have been taking place close to the pits, with deposition of primary and secondary waste parts, typified by the presence of body parts such as foot and head elements. Particularly in the Earlier Neolithic, consumption and deposition of some carcasses may have taken place elsewhere. Although a comprehensive conversion to sedentary agriculture is unlikely to have occurred during the Earlier Neolithic, domesticates would have played a significant role in social relations (Edmonds 1993, 101). The lack of wild species in the Earlier Neolithic pits may relate to social display, with utilisation of domesticates only.

Evidence from other earlier and later sites of this period seems to suggest that the material from these pits fits with the general occurrences of this period. The majority of sites studied are in some way ceremonial or ‘ritual’ sites. As already noted, at many sites, such as Cherhill (Grigson 1983), domestic cattle dominate the assemblages. Whilst there is variation in the proportions of domesticates between sites such as Abingdon (Case 1965), Horslip long barrow, Beckhampton Road long barrow and South Street long barrow (Ashbee et al. 1979), as well as Coneybury (Maltby 1990), this may well relate to preferences and specific practices at each site rather than the abundance of species specifically.

Perhaps the most closely comparable sites, in terms of context type, are Puddlehill pit 6, a Later Neolithic storage pit (Grigson 1976) and Down Farm (Legge 1991). Wild and domestic cattle and pig, plus sheep/goat and red deer were present at both sites. Although present in different proportions, the Later Neolithic assemblage from the pits to the south of Windmill Hill enclosure concurs with this trend. Wild ‘Species were seen at all three sites in small quantities, and are likely to reflect the local abundance and preference for specific species at the individual sites.

It appears that similarities in animal bone assemblages from the compared Earlier and Later Neolithic sites are context-specific. There are similarities with assemblages from henge sites, in terms of the domesticates present. However, at henge and enclosure sites domesticates are present in much

larger quantities, with few wild species, but a more visibly structured or ceremonial pattern of deposition (for example Jope 1965; Edwards and Horne 1997). The ways in which carcasses were disarticulated and jointed for meat appears to be more variable at henge sites, with deliberate wasting of whole limbs. This kind of conspicuous wasting does not seem to have occurred in the pits to the south of Windmill Hill enclosure. However, it is impossible to determine if the bone assemblage from non-domestic sites, such as the Windmill Hill enclosure, reflect the local abundance of species as the context of consumption and deposition is unlikely to reflect a rational food use strategy.

As discussed previously, the deposition of animal bones in the pits may have formed a specific practice. Deliberate selection of left-side bones is suggested for pit 325. At West Kennet, the overwhelming majority of bones were from the right side of the species identified (Edwards and Horne 1997, 125). The left side represents different things in different cultures, including death and feminine gender, and is often equated with impure aspects (Hertz 1960, 99-102). If this was an element of the belief system of the Earlier Neolithic, the deposited bone in the pits may have been associated with profane or impure activities or social elements. This explanation may sound extreme, but perhaps what the predominance of left-sided bones indicates is that activities around the pits were, socially, visibly different to those occurring inside the enclosure. The left side also predominates in the Later Neolithic assemblage, but not to such a marked degree.

Pits such as these have been considered to be initially created for other purposes, with subsequent deposition of ‘rubbish’ (Holgate 1988b, 106). The appearance of the Earlier Neolithic pits suggests that only one, 308, was open for any length of time. The Later Neolithic pits were little eroded. This suggests that the majority of pits were either filled soon after opening or, if used for another purpose previously, were not exposed to the weather. Whilst the erosion of the surface of bones, seen at other chalkland sites also, may be a factor of taphonomic processes in the burial environment, it is not possible to state that middening did not take place. Deposition of the bones themselves seems to have been deliberate, particularly in the Later Neolithic. Plans of the deposits confirm that, unlike some of the bones from the ditches of Windmill Hill enclosure (Whittle 1996, 274; Whittle et al. 1999), the material had been completely disarticulated prior to deposition. This kind of spatial separation of bones and body parts was also seen in

168 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

the Coneybury pit (Maltby 1990, 61). Edmonds suggests that such structured deposits at enclosures may reflect episodes of consumption (1993, 112). It seems likely that the deposition of animal bones in this deliberate manner would reflect a similar practice. The presence of worked bone and antler may reinforce the different nature of the pit deposits to those from enclosures. With debris from several activities being deposited together, a more everyday domestic constitution to the process of deposition may be indicated. However, apart from the proportions of species present, the Earlier Neolithic assemblage compares well with that from Coneybury, and the Later Neolithic with that from Down Farm, in terms of context and deposition.

CHARRED SEEDS, FRUITS AND TUBERS by Andrew S. Fairbairn

Sampling during the 1988 excavation of the adjacent causewayed enclosure produced a small quantity of charred remains of cereals, seeds, fruits and tubers (Fairbairn 1999). These provided evidence that the debris of both wild and crop plants had been incorporated into the enclosure ditch segments after use as food, beverages and medicines during the acts of consumption that accompanied the exchanges and ceremonies occurring within the social arena of the enclosure. Charred cereals were also present in the soils buried during the construction of the outer bank and in an Earlier Neolithic grave fill. Further investigation of Neolithic contexts on the hill provided an opportunity to extend botanical investigations beyond the bounds of the enclosure in a different set of features. Four Neolithic pits were sampled.

Sampled contexts

Thirteen bulk samples were collected from the Earlier and Later Neolithic pits in Areas D and M respectively. Sample volume varied from 4-10 litres, most samples being 6-7 litres in size (Tables 13-14). All the samples were collected from the lower and basal pit fills from archaeologically sealed contexts without any obvious contamination from post-Neolithic activity.

In Area D three samples were collected from the lower ashy fill (322) of pit 323, with one from a similar basal fill (330) of pit 325. In Area M a single sample was collected from the lower fill (223) of pit 202, with

samples from the basal fill (224), lower fill (221) and fill (219) of pit 212.

Field and laboratory methods

The samples were processed on a ‘Siraf’ type flotation tank, using a 250um mesh sieve to collect the floating fraction (flot) and a 500um mesh sieve to collect the heavy residue. The volume of the dried flots was measured and each was sorted in its entirety using a low-powered dissecting microscope. All of the charred non-wood charcoal plant remains were picked out from the flots. Identification was completed with the aid of the comparative seed and fruit collections of the University College London Institute of Archaeology, with the parenchymatous remains identified after fracturing using a scanning electron microscope (SEM). Help in identification was given by Gordon Hillman and Jon Hather. The identified remains are recorded in Tables 13-14 for each sample, the nomenclature for cereals following van Zeist 1984, and that for the wild taxa following Stace 1991.

Results

Abundant plant remains were recovered from the pits, with both wild and domestic remains being present including cereals, wild seeds, fruits, nuts and the remains of vegetable tubers. The flot sizes were variable and most contained a significant quantity of mollusc shell. Therefore, the actual quantity of charred plant remains was often much less than the total flot volume recorded in Tables 13-14. Preservation of the remains varied within individual samples, although it was mostly very poor. Cereal remains were vesicular and often fragmented, with wild plant seeds and fruits often badly damaged and lacking features crucial for identification. Poor preservation also characterised the plant remain assemblages recovered from the 1988 excavations. Small fragments of parenchymatous tissue were also preserved, although many were glassy and lacked any discernible structure, making identification impossible.

Plant remains from Area D (Earlier Neolithic: Table 13)

The flots from Area D contained a high percentage volume of mollusc shells. Relatively little wood charcoal was preserved and much of the volume of the charred plant material in the flots consisted of cereal grain fragments. Cereal grains were preserved

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 169

Table 13. Charred plant remains from the Earlier Neolithic pits. (Numbers refer to whole specimens and numbers in brackets refer

to fragments unless stated.)

Sample 1 Context Pit 323 Pit Fill basal Sample Volume Sule; Flot Volume 52ml Sorted flot volume 2ml Taxon Component

Domestic taxa

Triticum cf dicoccum grain

Triticum cf dicoccum spikelet forks (glumes)

Triticum grain

dicoccum/monococcum

Triticum sp. grain 1

Triticum sp. spikelet forks (glumes) 1

Hordeum vulgare var. grain

nudum

Hordeum vulgare grain 4 (2)

cf. Hordeum vulgare grain

Cereal indet grain BL C718)

Wild taxa

Corylus avellana nutshell (all fragments) 1

322-3119

2) 3 4 322-3223 322-3329 330-3344 323 323 325 basal basal basal 4L. 6:5 1: elles 20ml 45ml 82ml 2ml 5ml 12ml

1

2 (3)

2

i 22 (4)

2 3 (1) 13 (71)

1 6 We) 6 (10) 2 6(43) 11 (420) 36 (c.700) 4 38 149

in abundance in samples from Area D, although they were usually highly vesicular, distorted and had often lost much of their outer surface..This made identification difficult even to genus level in most cases and only occasionally could cereal grain identifications be made to species or sub-species level.

Several typical emmer wheat (Triticum dicoccum) grains were distinguished in sample 4 from Pit 325. This sample also included two grains which may have derived from either emmer (Triticum dicoccum) or einkorn (Triticum monococcum). A single well preserved naked barley grain (Hordeum vulgare var. nudum) was distinguished in sample 4 from pit 325, recognisable by its rounded form and wrinkled surface. The good preservation of this grain in comparison to the others was noticeable and may indicate a different preservational history. Many grains were only identifiable to general barley or wheat categories, and both barley and wheat were recorded in each pit. Many grains were highly compressed and distorted. This distortion and the vesicular appearance suggest that they may have been charred at high temperatures in a confined space, such charring effects having been observed in laboratory experiments (Fairbairn 1991). All of the identified cereal taxa are well known in the British Neolithic (Moffett et al. 1989). They were also identified by Helbaek in his work on the causewayed

enclosure and in the 1988 excavations with the exception of the naked barley.

Cereal chaff was preserved in the form of wheat spikelet forks and glume bases in both pits. The narrow spikelet forks with wide disarticulation scars typical of emmer were identified in samples from pit 325. However, most of the specimens were highly fragmented and beyond identification.

Several fruit and nut remains were preserved in the pits, hazelnut shell fragments being recorded in all the samples. In most cases few fragments were recorded, amounting to less than the equivalent of one whole nutshell per sample. Sample 4 (pit 325) contained the largest number and volume of shell fragments and the widest range of soft fruit remains, including the stone of a hawthorn fruit (haw) probably from Crataegus monogyna. Several fragments of the stone of sloe (Prunus spinosa) were identifiable by the characteristic sculpted surface and were freshly broken, suggesting that they were originally part of one specimen which had been broken during archaeological recovery. Fragments of apple pip, probably from the wild crab-apple (Malus sylvestris) were identified in sample 4 and sample 1, this being the only soft fruit remain from pit 323.

The few wild plant/weed seeds identified from Area D were all from sample 4 with the exception of one fruit of a bedstraw species (Galium). This was

170

identified by its characteristic shape, although the loss of the diagnostic outer surface meant that it was not distinguishable at the species level. The assemblage in sample 4 was small but relatively diverse. Three groups of taxa from the goosefoot family were identified including one specimen of Chenopodium murale, the fragment having the characteristic cell pattern and marginal keel of the species (Bergerren 1981). The typical trigonous, angled fruits of a dock species (Rumex sp.) were identified as well as a single fruit of greater plantain (Plantago major). Two other fruit types were only identifiable at the genus and family level, one from the mint family (Lamiaceae), and the second a fruit of one of the meadowgrasses (Poa). Several probable seeds remained unidentified from samples 1, 2 and 4.

Several small fragments of vegetative parenchyma were identified in samples from both pits (samples 3 and 4). Unlike that in the samples from Area M the specimens contained small (10-15 uum wide), densely packed cells and the specimens lacked vascular bundles. Taxonomic identification was impossible, although the cell size and structure suggest that the fragments are derived from charred endosperm or cotyledon tissue from a large seed or fruit (J. Hather pers. comm.), possibly from the broken sloe also identified in the sample.

A variety of other remains was recognisable in the remaining assemblages from both pits. The only identified specimen was the culm fragment of a grass species, identified on the basis of its anatomical

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

characteristics observable on the SEM. Other culm fragments remained unidentified from pits 323 and 325 (samples 3 and 4), with the damaged remains of buds present in samples 1 and 4. Several small woody structures, possibly stems (c.2mm wide), were recorded in samples 1 and 4, all being damaged and unidentifiable. A possible fruit pedicel was identified in sample 4, of the type seen in the Rosaceae, possibly deriving from either sloe or haw. The final class of material has been labelled ‘vesicular material’. It was common in samples 1, 2 and 4, the specimens consisting of fused, glassy, dense material, probably of plant origin but with no clear features visible.

Plant remains from Area M (Later Neolithic: Table 14)

The flots from Area M contained a higher proportion of charred plant material than those from Area D, much of which was wood charcoal. Unlike those from Area D they contained few cereal remains, all of which derived from the lower fill of pit 212. These possible cereal remains were distorted, indistinct, unidentifiable and included one whole grain and several fragments in sample 7 and one fragment in sample 9. Hazelnut shell fragments were present in all the samples, although again usually only in small amounts equivalent to less than one whole nut with the exception of sample 11. Other small wild seeds and fruits were present in samples from both pits, although unidentifiable. A small woody axis was

Table 14. Charred plant remains from the Later Neolithic pits. (Numbers refer to whole specimens and numbers in brackets refer to fragments

unless stated.)

Sample 5 6 Context 219-2176 221-2443 Pit 212 212

Pit fill fill lower fill Sample Volume OL OL

Flot Volume 22ml 24ml

Sorted flot volume <Iml

Taxon Component Domestic taxa Cereal indet grain Wild taxa

Corylus avellana nutshell (all fragments) 11 8 indet. (?monocotyledon) stem indet aerenchyma P indet strorage parenchyma indet parenchyma P

Indet woody axes

s)

Indet seed/fruit 2

Indet vesicular lumps P

221-

212 lowe OL. 73m Iml

86

a 8 9 10 12 13 11 2534 221/2- 221/2- 221/4- 224-2538 .. 224-2539 © 223-2 2414 2423 2508 212 212 212 212 212 202 sr fill lower fill lower fill lower fill — basal basal basal 7h, 1OL CRE 8L Tele 6oL ] 24ml 34m 19ml 25ml 24ml 47ml <Iml Iml <Iml 3ml Iml 2ml 1 frag 32 33 114 6 32 115

P Pp P Pp P P

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 171

recorded, as in the Area D samples and was again unidentifiable. The vesicular material described above was present in the three layers of pit 212.

A large number of fragments of vegetative parenchyma were recorded in the samples from Area M, although few were identifiable due to poor preservation of anatomical structure. Glassy material with distinctive outer surface (epidermis?) was present in both pits but was unidentifiable. Ten of the best preserved specimens from this area were selected for SEM investigation. Aerenchyma was identified in five samples from each of the excavated pits and all sampled layers with the exception of the upper fill of 212 (Figs 22-23).This special form of parenchyma is the main storage tissue in the rhizomes of aquatic plants and has a characteristic structure of air spaces separated by membranes consisting of large parenchyma cells (Fig. 22). The species was indistinguishable from the specimens, although most aquatic aerenchymatous plants are monocotyledons. The structure of the outer layers of cells was recorded in one specimen (Fig. 23), showing the periderm and inner aerenchyma with layers of non-aerenchymatous cells between. This type of aerenchyma and the tissue arrangement is seen most commonly in families such as the Alismataceae (although seen occasionally also

TE gs oon

fi aaaant

Figure 22. Close-up of aerenchyma from the lower fill of Later Neolithic pit 212 (sample 8)

Figure 23. Section of parenchymatous tissue from the lower fill of Later Neolithic pit 212 (sample 6). Visible in this specimen are the epidermis, sub-epidermal zone of small dense cells and the larger air spaces of the aerenchyma

in some members of the sedge family, the Cyperaceae). Jon Hather has suggested that the most likely source would be one of the water plantains (Alisma spp.) or arrowhead (Sagittaria sagittifolia), both of which produce edible tubers known to have been collected and used as foods (Hather 1993). Three other types of parenchymatous tissue were recorded. The first, recovered from sample 8 (pit 212), consisted of large-celled storage parenchyma, possibly from a root . Another specimen from pit 202 is shown in Fig. 24. This consisted of dense parenchyma cells containing a large number of vascular bundles, possibly coming from stem of monocotyledon. Several specimens from pit 212 (sample 9) had similar form and appeared to derive from a single plant structure. The outer surface had a distinctive undulating appearance with collapsed areas between peaks (Fig. 25). Internally the tissues were indistinct, although in one large fragment a smooth pit was preserved, possibly the imprint or testa of a seed. The seed imprint suggests that the seed took up a small total percentage volume of the whole. The reconstructed size of the specimen was about 4-5 mm, the shape

172 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Figure 24. Dense tissues from the lower fill of Later Neolithic pit 202 (sample 11). Possibly a monocotyledon stem

spherical to ovoid. It is possible that the specimen is the remains of a small soft fruit, of the same size as elder (Sambucus nigra).

Interpretations

Theoretical considerations.

Plants are routinely used and exchanged in non- industrial societies throughout the world for many purposes including, but not exclusively limited to, use as food (Moore 1986; Campbell 1994) in systems of value very different from those which operate in the contemporary western world (Appadurai 1986; Thomas 1991). Use and exchange of any material goods by individuals and communities are contingent on the beliefs and values of the social groups of which they are part. If the local Neolithic can be characterised by at least partly mobile settlement by an indigenous population only gradually adopting Neolithic material culture (Whittle 1993) and in a state of social and conceptual flux (Thomas 1991), with values rooted in the gatherer-hunter societies of the region, then exchanges and consumption would have been determined by these values during the

SoS BR ee OR Re ew

1, 68mm

1 20Ky.

Figure 25. Outer surface of a possible charred soft fruit from the lower fill of Later Neolithic pit 212 (sample 9) showing characteristic collapsed pattern

everyday events which physically sustained life, as well as in less prosaic activities which may have punctuated, defined and driven the life of the wider community.

The preservation of charred plant remains requires the exposure of plants to heat. It is usually assumed that most charred plant remains, with the exception of those burnt in catastrophic conflagrations, are charred by either accidental exposure to fire or the deliberate burning of waste during production, processing or consumption related activities (Moffett et al. 1989, 245), and then dumped into pits or other contexts whose primary use was for some other purpose. Activities include the parching of cereals prior to pounding (Hillman 1981), roasting nuts, roasting grain, cooking food and burning the debris of food preparation (Legge 1989). For the purposes of this study, however, the focus shifts to the acts of charring and deposition themselves and questions such as how and why such activities were carried out, by whom and how they reflect and acted upon the groups which carried them out in the Neolithic social milieu. Several modes of charring can be suggested here for the archaeological past distinguished by the relationship of the process of charring to human agency:

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 173

a. Deliberate burning. The use of plant fuels such as wood and peat are an obvious case. Other possibilities include: the burning of grain and other stored foods to remove pests; the burning of the by-products of the production and consumption of plant products; plants burnt as incense; and the deliberate burning of plants for ceremonial and ritual purposes.

b. Accidental burning, of plants during use for another purpose, as in spills of grain during processing for food, or accidental charring during cereal parching (Hillman 1981).

c. Incidental burning of plants independent of any direct usage, for example as a result of the burning of soil seed-banks below hearths, or the charring of plants blown on to open fires.

In the absence of any evidence for in-situ burning within the Windmill Hill pits, it must be assumed that the charred remains were incorporated into the pits after transformation by fire elsewhere. Again several! modes of entry can be suggested for the inclusion of plant remains in the pit fills:

i. Deliberate selection. Charred plant remains were deliberately selected for inclusion in the deposit as charred cereals, fruits and nuts.

ii. Ash incorporation. Plant remains were incorporated into the deposits through the deliberate inclusion of the residues of burning, although the specific plant remains were not picked out or selected.

ili. Mixed incorporation. Charred plant remains were incorporated into the deposits through human activity which included ash and charcoal as part of a more diverse group of remains (i.e. incorporation of a midden into a pit).

iv. Incidental. Incorporation of plant remains into a feature is totally independent of human action due to natural processes of erosion and deposition acting on charred material dumped elsewhere.

Earlier Neolithic pits of Area D.

The assemblages from Area D consisted mainly of charred cereal grains preserved with a little charcoal in an ashy soil excavated in association with bones and artefacts, including sarsen grinding stones presumably used for processing plant products. The pit sides were generally unweathered and fresh suggesting that they were infilled soon after excavation. All four cereal assemblages from pits 323 and 325 contained large numbers of charred grains, wheat spikelet forks and glume bases. Wheat and barley were

identified in all four samples, with emmer wheat and naked barley also present in the sample from pit 325. These crops are unlikely to have been gathered, processed, stored or cooked together because glume wheats, hulled and naked barley require different husbandry and processing techniques (Hillman 1981) and have different cooking properties. The lack of culm nodes, awn fragments and weed flora suggests that the cereals were charred in a final processed form, in the spikelet in the case of the glume wheats and as processed whole grains in the case of barley.

Large concentrations of grain and chaff were recovered from both pits and in the case of pit 323 similar assemblages were recorded in all three samples from the ashy soil layer. Although the presence of the cereals could be explained by the burning of the cleanings from stored grain thrown on to the fire prior to the preparation of food, such an explanation would seem unlikely. The prime grain represented in the samples would be only a small fraction of the whole quantity that initially entered the fire, most having been burnt, and if food preparation was the final aim of the processes during which plants were charred it would seem curious to throw such a large quantity of that food on to the fire. Ultimately, the repeated high concentrations of cereal remains would be difficult to account for without deliberate burning unless an unlikely series of burning events produced, entirely by coincidence, identical charred remain traces. The lack of any obvious signs of insect infestation or sprouting of the cereals suggests that this burning was neither the result of the destruction of spoiled grain nor the remains of grain malted for beer. If the thesis that processed prime produce was the source of the remains the most likely site of burning would have been a fire. The lack of charcoal could indicate that the cereals were burnt separately.

The non-domestic plant remains would also be most conveniently accounted for by the deliberate burning of waste from food preparation. Peelings, stones, shells and other waste would easily be disposed of in this way and the harder plant parts such as the nutshell and the haw stones would have required the sustained heat of a fire to ensure that they were completely charred. As with the domestic produce this need for sustained heat runs counter to the presence of only tiny quantities of charcoal in the samples from Area D, more of which would have been expected it the remains were generated in hearths. The other wild plant species may have been burnt as tinder in the case of the non-domestic grass remains (culm and seeds) and possibly incidentally through charring of wild seeds in the soil when fires were set

174 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

(e.g. the seeds of great plantain (Plantago major) and goosefoot (Chenopodium sp.)).

Despite the lack of charcoal, the association of wild and domestic species in the ashy soil suggests that they were burnt in association, the cereals being deliberately burnt as a form of ceremonial consumption, and the wild plants either deliberately burnt or derived from burnt waste. The plant remains ended up in the pit after the ash from the fires in which the plants burnt was placed in the pit along with other residues of consumption, and other symbols of that consumption such as the grinding stones and animal bones. The emphasis of the assemblages is upon domestic products and wild woodland resources. Individual plant remains were not selected from the ash left in a fireplace, but were incorporated as part of a deliberate ash deposit. Direct transference of ash from a fire seems the most likely route when the archaeological context and artefact associations are considered. It is also possible that the remains were transferred from an ash heap or midden after accumulation over an unknown time. This explanation seems less likely given the character of the pit fills, which seem to be more characteristic of single events.

Later Neolithic pits of Area M. The assemblages from the pits in Area M contained only tiny plant remain fragments in large quantities of charcoal suggesting that the plant remains were generated in a hearth and placed in the pits as part of an ash deposit. The fresh pit sides again suggested that the pits were back-filled rapidly after initial excavation. Hearth debris may have been transferred directly into the pits with other residues of the activities associated with the fires or after collection in a midden. Selection and disposal of ash and associated debris were deliberate, although it would also necessarily involve unconscious selection of the individual components of the plant remain assemblages if all of the ash and hearth debris was not collected. The plant remains from Area M were less numerous than those from Area D but were present throughout the sampled pit fills and contained few domestic remains. It is unclear which activities produced the remains but they could have been generated through the burning of waste generated by activities related to food consumption. The presence of large quantities of charcoal may indicate that the fires in which the plant remains were burnt were not tended for that purpose as in Area D.

The only cereals in these pits were recovered from the lower fill of pit 212. These few fragments were

badly preserved and may be derived from the burning of the residues of food preparation or may be residual, being present in the pit as the result of erosion from the topsoil ‘charcoal bank’ identified in the excavation of the causewayed enclosure (Fairbairn 1999). In contrast to the Area D pits, the emphasis in Area M is on wild plants from both wetland and woodland areas. The presence of the wetland plant resources is of some interest as despite the low level of taxonomic identification, the most likely source of the remains is from a range of water plants which grow in slow- moving or still waters and which are permanently inundated. The nearest source is at least 1 km from the site, indicating definite importation of plants to the site and probable storage. This is different from the other wild plant taxa, all of which may have grown close to the pits in the woodland/scrub habitat of Neolithic Windmill Hill. Such finds are significant as they back up the observations from other sites (Murphy 1988; Moffett et al. 1989) that wild resources from a range of habitats were collected well into the Neolithic.

Context and meaning.

In the Earlier Neolithic world, domestic plants may have carried weight as novel resources outside traditional forms of knowledge. Consumption in feasts accompanying the ceremonies at the site may have endowed status and drawn upon the symbolism of the domestic and tame to claim high rank and position (Thomas 1993). Deliberate destruction of cereals may have served similar purposes. As with the traditional wild foods, consumption may have also been an explicit or implicit expression of control or access to land and resources that grew or were planted there. Wild plants may also have carried the influence of tradition, their use acting as a means of invoking ancestral claims to position and land access, as in the presence of the wetland plant tubers in the Later Neolithic pits. The ash deposits themselves may have acted, with the other incorporated artefacts, as metonyms for the completion of these ceremonies and exchanges, while also providing a means of fixing these acts in the location of the hill and so being the material basis for the creation and later invocation of tradition (cf. Edmonds 1993).

Temporal and spatial patterns of plant use on Windmill Hill.

Over the history of Neolithic use of the hill trends are apparent in the representation of plants in the excavated features. Assemblages from the Earlier Neolithic pits were burnt in acts of deliberate

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 175

destruction. Other pre-enclosure contexts included charred plant remains, and plants were a general feature of the Earlier Neolithic use of the hill, whether the interpretations presented here are accepted or not. Plant remains from the Earlier Neolithic enclosure deposits were generated mainly as the result of accidental burning during the consumption of plant foods and included cereals and wild foods. In some contexts deliberate burning generated the remains as the plants were used as a metaphor for the body and its ills. The Later Neolithic plant use in the enclosure is characterised by similar deposits to that of the earlier phase of use, yet that beyond the enclosure sees only the residues of wild plants as definite utilised resources.

Discussion.

The temptation is to dwell only on those assemblages which are well preserved in easily understood contexts, applying quantitative techniques to discern slight changes in the patterning of archaeobotanical assemblages that are then given meaning only in relation to subsistence practice. The interpretation of less well preserved assemblages is not attempted, or written off due to ‘taphonomic bias’. The experience with the botanical remains from Windmill Hill is that even with poorly preserved remains, a consideration of issues beyond the bounds of subsistence within a deeper consideration of the social context of charring and deposition allows a wider set of interpretations to be generated which provide a more relevant contribution to the understanding of the archaeological features. The botanical investigations at Windmill Hill, including those of the pits and causewayed enclosure, have provided scope for such an interpretative approach to be attempted and new perspectives on the Neolithic use of plants to emerge.

DISCUSSION

by Alasdair Whittle, Jessica Davies,

Ian Dennis, Andrew Fairbairn and _ Michael Hamilton

Chronology

So far, both features and the components of the lithic scatter can only be assigned generalised dates, in the Earlier and Later Neolithic. Despite the overall aims of the project, radiocarbon dates have not so far been

obtained from the pits. It was felt that isolated dates, for example on bone samples selected before full analysis of the whole assemblage including assessment of their weathering, would be of little value, and it will be far more useful in the future to obtain both dates on varying samples (plant remains and fresh animal bone, for example) from the pits and further dates on samples from the ditches of the causewayed enclosure (Whittle et al. 1999). The Later Neolithic pits presumably date to after the primary use of the enclosure, but the position of the Earlier Neolithic pits is so far unclear; even with radiocarbon dates it may remain so, given the imprecision of the method. In a sense this does not matter, as the features and finds are sufficently interesting in their own right and we should not be obsessed by the enclosure, but it is important to attempt relative dating, since pits earlier than the enclosure would contribute to understanding of the early context, while pits contemporary with the enclosure would give a relatively rare insight into what went on around such an enclosure.

The near-absence of decorated pottery from the earlier pits could suggest that they belong, relative to the enclosure, earlier than or early in its existence, but the number of sherds is very small, and the inference is unreliable. In the absence of radiocarbon dates, it can be stressed that both sets of pits appear to have been fairly short-lived. Most are unweathered and there are frequent deliberate fills. There are conjoining flints in 202 and 212. These simple observations may in fact be at least as valuable as radiocarbon determinations, since they immediately give insight into the social context of activity (cf. Pollard 1999) outside the enclosure.

The Earlier Neolithic setting

The final significance of the earlier pits has yet to be determined. At one level, they can be seen as part of the series of sporadic, perhaps transitory or short- lived, occupations that characterise the area in the Earlier Neolithic (Whittle et al. 1999 and references). The current belief in the relative mobility of Earlier Neolithic settlement (e.g. Thomas 1991;Whittle 1996) has rightly been challenged (Evans 1999, 24), and one alternative is to envisage a spectrum of kinds of occupations of varying duration and character (Whittle 1997b; Pollard 1999). The specific evidence here is ambiguous. There were no reliable ageing data from the animal bones; the assemblage could be seen as butchery waste of carcasses exploited elsewhere, implying movement of some kind. The plant remains

176 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

imply the existence somewhere nearby of plots or gardens at the least, but it was argued above that the cereal remains may have been deliberately burnt, to enhance their novel significance: acts of deliberate spoilage and ceremonial consumption. This could imply that the scale and duration of cultivation were not intense, but equally allows the possibility of at least short-term sedentism. The quern and rubbing stones could also be viewed as principally symbolic or routine. People stayed long enough to work flint right through the reduction process, and sherds selected for deposition could suggest separate middening; context 310 from pit 308 and 330 from 325, which produce most of the Earlier Neolithic pottery, represent sizeable proportions of those pits but produced relatively little flint, perhaps suggesting a separation of the middening of flint and pottery (Hamilton 1995, 224). This might again imply some history to the use of the locale (in the terms of Evans 1999). The earlier pits themselves were generally unweathered and backfilled, implying short-lived events rather than long occupation, and there were no signs of post-framed or other structures, but the areas opened were small and the chalk subsoil has presumably been weathered to some degree. The earlier component of the lithic scatter may suggest only sporadic occupations, but periodic visits to the same small places cannot be excluded (contra Pollard 1999, 82).

At another level, the significance of the earlier pits resides in the perspective they give on the enclosure. It is easy to be over-swayed by what went on in the enclosure as a whole and to see the enclosure as a single phenomenon. It is argued in the enclosure report (Whittle et al. 1999) that despite the large- scale of the layout, many of the depositions in the ditches may have been small-scale and themselves episodic. Though they can be characterised by a great range of combinations of material and of degrees of intentionality, deliberately placed deposits were recurrent (Whittle er al. 1999, chapters 3, 4, 11 and 17). The selection for inclusion in the pits outside the enclosure of animal bones, flints, sarsen artefacts and well burnt cereal remains is strongly reminiscent in general of the activity seen in the enclosure ditches. The earlier pits may therefore serve importantly to underline the frequency of this kind of activity, whether few or many people were involved in either case. The pits also involved digging into the ground, and in the case of 324, deliberate backfilling of chalk. Clearly this is not on the monumental or cohesive scale of the enclosure, but the enclosure perhaps only had meaning in a world in which these kinds of activity

were widespread or at least recurrent in other contexts (cf. Barrett 1994). In this sense, and for the time being, it may matter less that we do not know whether the pits pre-dated or were contemporary with the enclosure.

The Later Neolithic setting

At a time when there was only limited use of the enclosure ditches (Whittle et a/. 1999), occupation of the southern slope of Windmill Hill seems to have intensified. In this, the locale is similar to several others in the area (Thomas 1991; Holgate 1987; Whittle 1993 and references). The later component of the lithic scatter was probably significantly greater than the earlier, as discussed above. It has been argued on various grounds that the Later Neolithic saw a trend to increasing or established sedentism (Thomas 1991; Bradley 1987). The evidence from specific locales may put this in question, as discussed elsewhere with reference to Silbury Hill and the West Kennet palisade enclosures (Whittle 1997a). The evidence from the southern slope of Windmill Hill is once more ambiguous. There is no evidence for post-framed or other above-ground structures, but the same qualifications as above must apply. Pits were unweathered and backfilled; conjoining flints reinforce the impression of short life. Fills, charcoal (not yet identified) and burnt sarsen indicate fires or hearths. The whole flint reduction process is represented. Sherds were selected for deposition, which may again imply middening of some kind. The plant remains seem to emphasise wild species more than earlier. This may be compatible with continued mobility or short-term sedentism, or a reduced scale of cereal cultivation, or a different season or spatial location of cereal processing. The much less abundant plant remains from the later pits seem, however, to have been burnt during their consumption as food, beverages or medicines, which could imply routine circumstance. The ageing data from the animal bones show the presence of animals of varying ages, including young ones; as discussed above, no specific or exclusive season of occupation can be claimed, and the year-round presence of people with their animals somewhere within reach is not to be excluded. To a greater degree than in the earlier pits, there were meat- bearing bones in some abundance and there was primary carcass reduction nearby. Wild animals continued to be exploited in the area.

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 177

It is probably a mistake to seek a single answer from this kind of evidence, and better to envisage shifts along the spectrum of mobility and short-term sedentism. The Later Neolithic trend in the area, if indeed there is a single trend (cf. Evans 1999), may be to reduced mobility and the more frequent establishment of short-term sedentism (cf. Whittle 1996; 1997a). It is important also not to neglect the possibly enduring significance of the locality (in the terms of Evans 1999), since the place represented by the hill and its tradition may have lived on in continuing and increasing visits, to be enhanced now by the view down from an older history on to the other spectacular and new additions to the Later Neolithic landscape of the area; both Avebury and Silbury Hill are directly visible from the southern slope of Windmill Hill. Some of the deposition in pits 212 and 202 may hark back to earlier ways of doing things, but it would be premature to assign the pits to special times and the lithic scatter to the routine, or vice versa. Nonetheless the later pits and some at least of the their contents might have been associated with the exterior of the enclosure, in line with the interpretation of the enclosure ditch circuits as realms of value, drawing on the symbolism of domesticity at its core and the wild at its periphery (Whittle 1996; Whittle et al. 1999). Equally, through time, such distinctions may have become blurred and the total history of the place may have become its dominant drawing power.

The search for useful occupation

The investigations reported here were hard work, and subsoil features were not located till a relatively late stage. It was easy to feel frustrated during the process of survey and excavation, because there remains an expectation that if only we try hard enough or look cleverly enough in the right places, a more abundant and better preserved Neolithic settlement record will somehow emerge. Circumstances will vary, but the kind of evidence reported here may be as good as we are likely to get. The total range of evidence recovered, despite its fragmentary nature and the difficulties of locating it, is in fact impressive: for moving, dwelling, “using plants and animals, and in general inhabiting a history. The Neolithic archaeological record of the area need not be seen as wholly dominated by monuments, even if it will require continued hard work, as well as refined search methodologies, to find the relevant occupation evidence. The harder tasks, however, may be to re-think both our expectations and our interpretations.

Acknowledgements (AW).

Grateful thanks are due to: Robin Butler and the National Trust for access; The British Academy and The School of History and Archaeology, Cardiff University, for financial support; Joshua Pollard for assisting direction, and Ian Dennis for supervision, and for the published drawings; Michael Hamilton for geophysical survey; Julian Richards and the late Tony Clark for advice; Stanley Jenkins for guidance in the Alexander Keiller Museum, Avebury; Martyn Barber and Dale Serjeantson for information; Robin and Gill Swanton for every help; and the Cardiff University students and other volunteeers for surviving hot weather, wet weather and sieving.

Bibliography

ADKINS, R. and JACKSON, R., 1978, Neolithic stone and flint axes from the River Thames. London: British Museum

ANNABLE, F.K., 1965, ‘Roman Period’, in I.F. Smith, Windmill Hill and Avebury, 172-74. Oxford: Clarendon Press

APPADURAI, A., 1986, ‘Introduction: commodities and the politics of value’, in A. Appadurai (ed.), The social life of things: commodities in cultural perspective, 3- 63. Cambridge: Cambridge University Press

BARBER, M., DYER, C. and OSWALD, A., 1999, The Neolithic causewayed enclosures of England. London: RCHME

BARRETT, J.C., 1994, Fragments from antiquity. Oxford: Blackwell

BARRETT, J. C., BRADLEY, R. and GREEN, M., 1991, Landscape, monuments and society: the prehistory of Cranborne Chase. Cambridge: Cambridge University Press

BENDER, B., 1978, ‘Gatherer-hunter to farmer: a social perspective’, World Archaeology, 10, 204-22

BERGERREN, G., 1981, Atlas of seeds and small fruits of north-western European plant species, with morphological descriptions. Part 3, Salicaceae- Cruciferae. Stockholm: Swedish Museum of Natural History

BINFORD, L., 1981, Bones: ancient men and modern myths. New York: Academic Press

BOESSNECK, J., 1969, ‘Osteological differences between sheep (Ovis aries Linné) and goat (Capra hircus Linne)’, in D. Brothwell and E.S. Higgs (eds), Science in archaeology (second edition), 331-58. London: Thames and Hudson

BRADLEY, R., 1987, ‘Flint technology and the character of Neolithic settlement’, in A.G. Brown and M.R. Edmonds (eds), Lithic analysis and later British prehistory, 181-86. Oxford: British Archaeological Reports

178 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

BRADLEY, R., 1998, The significance of monuments: on the shaping of human experience 1n Neolithic and Bronze Age Europe. London: Routledge

CAMPBELL, A. T., 1995, Getting to know Waiwa1i: an Amazonian ethnography. London: Routledge

CASE, H., 1956, ‘The Neolithic causewayed camp at Abingdon, Berkshire’, Antiquaries Journal 36, 11-30

CLARK, J.G.D., 1933, ‘Report on an Early Bronze Age site in the southeastern fens’, Antiquaries Journal 13, 266-96

CLARK, J.G.D., 1934, ‘Derivative forms of the Petit Tranchet in Britain’, Archaeological Journal 91, 32-58

CLARK J.G.D., 1960, ‘Excavation at the Neolithic site at Hurst Fen, Mildenhall, Suffolk (1954, 1957,1958)’, Proceedings of the Prehistoric Society 26, 202-45

CLEAL, R., 1992, ‘The Neolithic and Beaker Pottery’, in C. Gingell, The Marlborough Downs: a Later Bronze Age landscape and its origins, 61-71. Devizes: Wiltshire Archaeological and Natural History Society

CRAWFORD, O.G.S., 1953, Archaeology in the field. London: Dent

CRAWFORD, O.G.S., 1955, Said and done: the autobiography of an archaeologist. London: Weidenfeld and Nicolson

CUNNINGTON, M.E. and GODDARD, E.H., 1934, Catalogue of antiquities in the Museum...at Devizes, Part IT. Devizes: Wiltshire Archaeological and Natural History Society

DAVID, A., 1999, ‘Geophysical survey’, in A. Whittle, J. Pollard and C. Grigson, The harmony of symbols: the Windmill Hill causewayed enclosure, Wiltshire, 17-23. Oxford: Oxbow

EDMONDS, M., 1993, ‘Interpreting causewayed enclosures in the past and present’, in C. Tilley, Interpretative archaeology 99-142. Oxford: Berg

EDMONDS, M., 1999, Ancestral geographies of the Neolithic: landscapes, monuments and memory. London: Routledge

EDWARDS, A and HORNE, M.., 1997, ‘Animal bone’, in A.Whittle, Sacred mound, holy Rings. Silbury Hill and the West Kennet palisade enclosures: a Later Neolithic complex in north Wiltshire, 117-29. Oxford: Oxbow

ENTWISTLE, R. and GRANT, A., 1989, ‘The evidence for cereal cultivation and animal husbandry in the southern British Neolithic and Bronze Age’, in A. Milles, D. Williams and D. Gardner (eds), The beginnings of agriculture, 203-15. Oxford: British Archaeological Reports

EVANS, C., 1988, ‘Acts of enclosure: a consideration of concentrically-organised causewayed enclosures’, in J. C. Barrett and I. A. Kinnes (eds), The archaeology of context in the Neolithic and Bronze Age: recent trends, 85-96. Sheffield: Department of Archaeology and Prehistory, University of Sheffield

EVANS, J., 1872, Ancient stone implements, weapons, and ornaments of Great Britain. London: Longmans, Green, Reader and Dyer

EVANS, J.G., 1999, Land and archaeology: histories of

human environment in the British Isles. Stroud: Tempus

EVANS, J. G. and SMITH, I. F., 1983, ‘Excavations at Cherhill, North Wiltshire, 1967’, Proceedings of the Prehistoric Society 49, 43-117

FAIRBAIRN,A.S., 1991, The role of electron spin resonance spectroscopy 1n the identification of preservational bias in archaeological cereal assemblages. Unpublished BSc dissertation, Department of Human Environment, Institute of Archaeology, University College, London

FAIRBAIRN, A., 1999, “The charred plant remains’, in A. Whittle, J. Pollard and C. Grigson, The harmony of symbols: the Windmill Hill causewayed enclosure, Wiltshire, 139-56. Oxford: Oxbow

FIDDES, N., 1991, Meat: a natural symbol. London: Routledge

FIELD, N. H., MATTHEWS, C. L. and SMITH, I. F., 1964, ‘New Neolithic sites in Dorset and Bedfordshire with a note on the distribution of Neolithic storage-pits in Britain’, Proceedings of the Prehistoric Society 30, 352-81

GINGELL, C., 1992, The Marlborough Downs: a Later Bronze Age landscape and its origins. Devizes: Wiltshire Archaeological and Natural History Society

GRANT, A., 1982, ‘The use of tooth wear as a guide to the age of domestic ungulates — a brief explanation’, in B. Wilson, C. Grigson and S. Payne (eds), Ageing and sexing animal bones from archaeological sites, 91-108. Oxford: British Archaeological Reports

GRAYSON, D. K., 1978, ‘Minimum numbers and sample size in vertebrate faunal analysis’, American Antiquity 43, 53-65

GREEN, H.S., 1980, The flint arrowheads of the British Isles. Oxford: British Archaeological Reports

GRIGSON, C., 1965, ‘Faunal remains: measurements of bones, horncores, antlers and teeth’, in I.F. Smith, Windmill Hill and Avebury, 145-67. Oxford: Clarendon Press

GRIGSON, C., 1966, “The animal remains from Fussell’s Lodge long barrow (including a possible ox-hide burial, with discussion on the presence of the horse in Neolithic Britain)’, in P. Ashbee, “The Fussell’s Lodge long barrow excavations 1957’, Archaeologia 100, 63-73

GRIGSON, C., 1969, ‘The uses and limitations of differences in absolute size in the distinctions between the bones of aurochs (Bos primigenius) and domestic cattle (Bos taurus)’, in P. Ucko and G.W. Dimbleby (eds), The domestication and exploitation of plants and animals, 277-94. London: Duckworth

GRIGSON, C., 1976, ‘The animal bones from Neolithic Pit 6’, in C. L. Matthews, Occupation sites on a Chiltern ridge: excavations at Puddlehill and sites near Dunstable, Bedfordshire. Part 1: Neolithic, Bronze Age and Early Iron Age, 11-18. Oxford: British Archaeological Reports

GRIGSON, C., 1983, ‘Mesolithic and Neolithic animal bones’, in J. G. Evans and I. F. Smith, ‘Excavations at Cherhill, North Wiltshire, 1967’, Proceedings of the Prehistoric Society 49, 64-72

GRINSELL, L.V., 1957, ‘Archaeological gazetteer’, in R.B.

NEOLITHIC ACTIVITY AND OCCUPATION OUTSIDE WINDMILL HILL CAUSEWAYED ENCLOSURE 179

Pugh and E. Crittall (eds), Victoria County History of Wiltshire 1(2), 21-279. Oxford: Oxford University Press

HALSTEAD, P., 1992, ‘Dimini and the ‘DMP’: faunal remains and animal exploitation in Late Neolithic Thessaly’, The Annual of the British School at Athens 87, 29-59

HAMILTON, M.A., 1995, The Latest Neolithic and Earlier Bronze Age pottery of southern Britain. Unpublished PhD thesis, Cardiff University

HAMILTON, M.A., 1997, ‘Pottery’, in A. Whittle, Sacred mound, holy rings. Silbury Hill and the West Kennet palisade enclosures: a Later Neolithic complex in north Wiltshire, 93-117. Oxford: Oxbow

HAMILTON, M., 1999, ‘Pottery, Part Two: Late Neolithic and Bronze Age’, in A. Whittle, J. Pollard and C. Grigson, The harmony of symbols: the Windmill Hill causewayed enclosure, Wiltshire, 292-317. Oxford: Oxbow

HAMILTON, M.A. and WHITTLE, A., 1999, ‘Grooved Ware of the Avebury area: styles, contexts and meanings’, in R. Cleal and A MacSween (eds), Grooved Ware in Britain and Ireland, 36-47. Oxford: Oxbow

HARDING, P., 1990, ‘The comparative analysis of four stratified flint assemblages and a knapping cluster’, in J.Richards, The Stonehenge environs project, 213-25. London: English Heritage

HATHER, J.G., 1993, An archaeobotanical guide to root and tuber identification. Volume 1: Europe and the Near East. Oxford: Oxbow

HERTZ, R., 1960, Death and the right hand. Aberdeen: Cohen and West

HIGGS, E.S., (ed.) 1972, Papers in economic prehistory. Cambridge: Cambridge University Press

HILLMAN, G. C., 1981, ‘Reconstructing crop husbandry practices from charred remains of crops’, in R. Mercer (ed.), Farming practice in British prehistory. 123-62. Edinburgh: Edinburgh University Press

HOLGATE, R., 1987, ‘Neolithic settlement patterns at Avebury, Wiltshire’, Antiquity 61, 259-63

HOLGATE, R., 1988a, Neolithic settlement of the Thames basin. Oxford: British Archaeological Reports

HOLGATE, R., 1988b, ‘A review of Neolithic domestic activity in southern Britain’, in J. C. Barrett and I. A. Kinnes (eds), The archaeology of context in the Neolithic and Bronze Age: recent trends, 104-12. Sheffield: Department of Archaeology and Prehistory, University of Sheffield

HUBBARD, R. N. L. B. and CLAPHAM, A., 1992, ‘Quantifying macroscopic plant remains’, Review of Palaeobotany and Palynology 73, 117-32

JACOMEYT, S., 1987, Prahistorische Getreidefunde. Basel: Botanisches Institut der Universitat, Abteilung Pflanzensystematik und Geobotanik

JONES, M.., 1980, ‘Carbonised cereals from Grooved Ware contexts’, Proceedings of the Prehistoric Society 46, 61- 63

JONES, M., 1984, ‘The plant remains’, in B. Cunliffe, Danebury: an Iron Age hillfort in Hampshire. Volume 2, the excavations 1969-1978: the finds, 483-95.

London: Council for British Archaeology

JOPE, M., 1965, ‘Faunal remains: frequencies and ages of species’, in I.F. Smith, Windmill Hill and Avebury, 142-45. Oxford: Clarendon Press

KEILLER,A., 1934, ‘Excavation atWindmill Hill’, Proceedings of the First International Congress of Prehistoric and Protohistoric Sciences, 135-38. London: International Congress of Prehistoric and Protohistoric Sciences

KENDALL, H.G.O., 1914, ‘Flint implements from the surface near Avebury: their classification and dates’, Proceedings of the Society of Antiquaries of London (second series) 26, 73-85

KENDALL, H.G.O., 1916, ‘Windmill Hill, Avebury, and Grime’s Graves’, Proceedings of the Prehistoric Society of East Anglia 2, 230-39

KENDALL, H.G.O., 1922, ‘Scraper-core industries in North Wiltshire’, Proceedings of the Prehistoric Society of East Anglia 3, 515-41

KLEIN, R. G. and CRUZ-URIBE, K., 1984, The analysis of animal bones from archaeological sites. Chicago: University of Chicago Press

LEGGE, A. J., 1981, “The agricultural economy’, in R. J. Mercer, Grimes Graves, Norfolk: excavations 1971-72: volume 1, 79-103. London: HMSO

LEGGE, A. J., 1989, ‘Milking the evidence: a reply to Entwistle and Grant’, in A. Milles, D. Williams and D. Gardner (eds), The beginnings of agriculture, 217-42. Oxford: British Archaeological Reports

LEGGE, A. J., 1991, “The animal remains from six sites at Down Farm, Woodcutts’, in J. Barrett, R. Bradley and M. Hall (eds), Papers on the prehistoric archaeology of Cranborne Chase, 54-100. Oxford: Oxbow

LYMAN, R. L., 1992, ‘Anatomical considerations of utility curves in zooarchaeology’, Journal of Archaeological Science 19, 7-22

MALONE, C., 1989, English Heritage book of Avebury. London: Batsford/English Heritage

MALTBY, M., 1990, ‘Animal bones’, in J. Richards, The Stonehenge environs project, 57-61. London: English Heritage

MOFFETT, L., ROBINSON, M. and STRAKER, V., 1989, ‘Cereals, fruits and nuts: charred plant remains from Neolithic sites in England and Wales and the Neolithic economy’, in A. Milles, D. Williams and N. Gardner (eds), The beginnings of agriculture, 243-61. Oxford: British Archaeological Reports

MOORE, H., 1986, Space, text and gender; an anthropological study of the Marakwet of Kenya. Cambridge: Cambridge University Press

MURPHY, P., 1988, ‘Carbonised Neolithic plant remains from The Stumble, an intertidal site in the Blackwater Estuary, Essex, England’, Circaea 6, 21-38

O’CONNOR, T. P., 1985, ‘On quantifying vertebrates - some sceptical observations’, Circaea 3, 27-30

PAYNE, S., 1972, ‘On the interpretation of bone samples from archaeological sites’, in E.S. Higgs (ed.), Papers in economic prehistory, 65-81. Cambridge: Cambridge University Press

180 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

PAYNE, S. and BULL, G., 1988, “Components of variation in measurements of pig bones and teeth, and the use of measurements to distinguish wild from domestic pig remains’, Archaeozoologica 2, 27-66

PITTS, M. and WHITTLE, A., 1992, ‘The development and date of Avebury’, Proceedings of the Prehistoric Society 58, 203-12

POLLARD, J., 1992, ‘The Sanctuary, Overton Hill, Wiltshire: a re-examination’, Proceedings of the Prehistoric Society 58, 213-26

POLLARD, J., 1999, “These places have their moments’: thoughts on settlement practices in the British Neolithic’, in J. Brick and M. Goodman (eds), Making places in the prehistoric world: themes in settlement archaeology, 76-93. London: UCL Press

PRUMMEL,W. and FRISCH, H-J., 1986, ‘A guide for the distinction of species, sex and body size in bones of sheep and goat’, Journal of Archaeological Science 13, 567-77

RICHARDS, J., 1990, The Stonehenge environs project. London: English Heritage

RINGROSE, T. J., 1993, ‘Bone counts and statistics: a critique’, Journal of Archaeological Science 20, 121-57

SCHOFIELD, A.J., 1991, ‘Lithic distributions in the upper Meon Valley’, Proceedings of the Prehistoric Society 57, 159-79

SILVER, I. A., 1969, “The ageing of domestic animals’, in D. Brothwell and E.S. Higgs (eds), Science in archaeology (second edition), 250-68. London: Thames and Hudson

SMITH, A.C., 1885, Guide to the British and Roman antiquities of the North Wiltshire downs in a hundred square miles round Abury (second edition). Devizes: Wiltshire Archaeological and Natural History Society

SMITH, I.F., 1965, Windmill Hill and Avebury: excavations by Alexander Keiller 1925-1939. Oxford: Clarendon Press

SHANKS, M. and TILLEY, C., 1987, Reconstructing archaeology: theory and practice. London: Routledge

STACE, C., 1991, New flora of the British Isles. Cambridge: Cambridge University Press

STONE, J. F. S., PIGGOTT, S: and BOOTH, A. St. J., 1954, ‘Durrington Walls, Wiltshire: recent excavations at a ceremonial site of the early second millennium B.C’, Antiquaries Journal 34, 155-77

THOMAS, J., 1991, Rethinking the Neolithic. Cambridge: Cambridge University Press

THOMAS, J., 1993, ‘Discourse, totalisation and the Neolithic’, in C. Tilley, Interpretative archaeology, 357- 94. Oxford: Berg

VAN ZEIST,W., 1984, ‘Lists of names of wild and cultivated cereals’, Bulletin on Sumerian Agriculture 1, 8-15 VON DEN DREISCH, A., 1976, A guide to the measurement of animal bones from archaeological sites. Peabody Museum Bulletin 1, Harvard University

WAINWRIGHT, G. J. and LONGWORTH, I.H., 1971, Durrington Walls: excavations 1966-1968. London: Society of Antiquaries

WHITTLE, A., 1991, ‘Wayland’s Smithy, Oxfordshire: excavations at the Neolithic tomb in 1962-3 by R.J.C Atkinson and S. Piggott’, Proceedings of the Prehistoric Society 57 (2), 61-101

WHITTLE, A., 1993, ‘The Neolithic of the Avebury area: sequence, environment, settkement and monuments’, Oxford Journal of Archaeology 12, 29-53

WHITTLE, A., 1994, ‘Excavations at Millbarrow chambered tomb, Winterbourne Monkton, north Wiltshire’, Wiltshire Archaeological and Natural History Society 87, 1-53

WHITTLE, A., 1996, Europe in the Neolithic: the creation of new worlds. Cambridge: Cambridge University Press

WHITTLE, A., 1997a, Sacred mound, holy rings. Silbury Hill and the West Kennet palisade enclosures: a Later Neolithic complex in north Wiltshire. Oxford: Oxbow

WHITTLE, A., 1997b, ‘Moving on and moving around: Neolithic settlement mobility’, in P. Topping (ed.), Neolithic landscapes, 15-22. Oxford: Oxbow

WHITTLE, A., ROUSE, A. and EVANS, J., 1993, ‘A Neolithic downland monument in its environment: excavations at the Easton Down long barrow, Bishops Cannings, north Wiltshire’, Proceedings of the Prehistoric Society 59, 197-239

WHITTLE, A. and POLLARD, J., 1998, ‘Windmill Hill causewayed enclosure: the harmony of symbols’, in M. Edmonds and C. Richards (eds), Understanding the Neolithic of north-western Europe, 231-47. Glasgow: Cruithne Press

WHITTLE, A., POLLARD, J. and GRIGSON, C., 1999, The harmony of symbols: the Windmill Hill causewayed enclosure, Wiltshire. Oxford: Oxbow

WOODWARD, P., 1991, The South Dorset Ridgeway: survey and excavations 1977-84. Dorchester: Dorset Natural History and Archaeological Society

ZIENKIEWICZ, L., 1999, ‘Pottery, Part One: Early Neolithic including Ebbsfleet’, in A. Whittle, J. Pollard and C. Grigson, The harmony of symbols: the Windmill Hill causewayed enclosure, Wiltshire, 258-92. Oxford: Oxbow

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 181-204

Excavations in Wilton, 1995-6: St John’s Hospital

and South Street

by Phil Andrews, Lorraine Mepham and Rachael Seager Smith with contributions by Michael J. Allen, Sheila Hamilton-Dyer and Pat

Hinton; illustrations by S. E. James

This report describes the results from two recent excavations in Wilton where comparatively little archaeological work has been undertaken despite the known importance of the town between the 9th and 13th centuries. The larger site, at St John’s Hospital, revealed a section through the late Saxon defences that sealed deposits of Romano-British date. The other site in South Street demonstrated occupation from the 9th/10th century onwards and produced notable assemblages of pottery and 1ron-working slag.

PROJECT BACKGROUND

Wessex Archaeology was commissioned to undertake two excavations in Wilton in 1995 and 1996, at the New Doctor’s Surgery, South Street and St John’s Hospital respectively. Both sites lie in an area of high archaeological potential, St John’s Hospital across the line of the late Saxon defences in the north-west of the town, and South Street within the medieval and possibly also the late Saxon settlement (Figure 1). The excavations followed earlier evaluations at both sites, and the work at St John’s Hospital also included a watching brief during development. The majority of this report is concerned with St John’s Hospital, with asummary report of the South Street excavation, although the finds, particularly the pottery, from the latter site are dealt with in more detail.

HISTORICAL AND ARCHAEOLOGICAL BACKGROUND

The historical and archaeological background for Wilton presented here is largely derived from James’s and Haslam’s work (James 1962; Haslam 1976) which provide the most comprehensive studies of the town. However, there remains much further work which could be done on the documentary sources and very

little archaeological excavation has yet been undertaken.

The slightly raised gravel island between two rivers on which Wilton is situated would have provided an attractive settlement site for early inhabitants of the Nadder and Wylye valleys. The existence of prehistoric and Roman activity in the immediate area of the town has been postulated although, prior to the excavations reported on here, no archaeological evidence for this had been forthcoming. Haslam has suggested that there may have been an early river crossing utilising the gravel island, but implies that a significant Roman settlement is unlikely as the main road between Dorchester and Old Sarum crossed the Nadder approximately two kilometres east of Wilton (Haslam 1976, 67).

A likely starting point for the Saxon settlement of Wilton is in the 6th or 7th century, although there remains a lack of evidence for its existence prior to the 9th century, by which time it had become the capital of the shire. The first documentary reference to Wilton (Uilton) dates to AD 838, in the form of a concordat between the King of Wessex and the Archbishop of Canterbury written at Kingston upon Thames and confirmed at Wilton. This implies a special legal status for Wilton, which at this time was probably one of the more established royal seats in Wessex, where the king is likely to have had a palace.

Wessex Archaeology, Portway House, Old Sarum Park, Salisbury SP4 6EB

J WILTO ,

2km

N.

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

SU

DITCHAMPTON

St John's i Hospital

Fig 1. Site location plan

St Andrew's

Church

My /, if

Kingsbury // Square x Lyf, Sey f f

Hospital

EDs yy, site of Jl Pte pgs Mill OW vy je SERED

New 7 Doctor's Surgery

site of Y

site of Abbey

[| Archaeological excavation

YZ, Probable extent of Saxon occupation y 4% WUT dh: (after Haslam 1976)

EXCAVATIONS IN WILTON, 1995-6

In Wilton, place-name evidence for such a building is possibly provided by the modern street name Kingsbury Square, located to the south-east of the Market Place (see Figure 1). Kingsbury may derive from Kings Burh, identifying the area as associated with a royal burh or defended settlement. Wilton was also an occasional seat of the West Saxon bishop, and a nunnery was founded in the town during the 9th century that was subsequently to become a major monastic establishment. However, by the time of Alfred’s reign in the later 9th century the importance of Wilton as a royal seat was on the wane, and Winchester had become the principal town as well as being the ecclesiastical focus of the Kingdom.

During the 9th and 10th centuries Wilton was an important defended centre, and the last battle of one of the Danish campaigns was fought there in 871. The Burghal Hidage, an early 10th century document recording defended sites established by King Alfred, lists Wilton (with 1400 hides) as one of three key sites in Wiltshire defending the boundary of Wessex, and the town may have been replanned and had its defences remodelled at this time. Throughout this period Wilton was the predominant mint for the shire although, after 1003, when the town was sacked by the Danes, the moneyers appear to have left it temporarily and moved to Old Sarum. However, Wilton was not completely abandoned as a mint until 1250.

The Domesday survey records that Wilton paid a rent of £50, though the number of burgesses was not recorded, and this indicates its continued importance amongst the towns of Wiltshire, with eight parish churches and an additional four in the suburbs listed. The granting of a charter to the Guild Merchant in the early 12th century, conferring the right to free passage and tolls, was a major economic benefit to the town. Documentary sources indicate that Wilton continued to expand throughout the 11th, 12th and 13th centuries, and the presence of a large Jewish community suggests a significant urban population involved in commercial and industrial activity.

Wilton suffered a decline in the late 13th and 14th

_centuries caused by the growth of New Sarum (Salisbury), 5km to the east, and the construction of the Harnham Bridge in 1244 which allowed direct access from Salisbury westwards without the need to pass through Wilton. This decline was probably exacerbated by depopulation suffered through the Black Death, and the fortunes of the town did not revive until the 18th century with the growth of the carpet industry.

183

The lack of recent redevelopment within the historic core of the town has meant that there have been few opportunities for archaeological intervention, and those that have taken place have been of very limited scale. During the early 1970s two evaluation trenches were excavated under the direction of David Hill across the line of the Saxon defences immediately south of St John’s Hospital (WAM 1971, 191; WAM 1972, 175-6).The ditch and bank, which was subsequently investigated in 1996, was found in these trenches, and the results of this earlier work are discussed further below. An evaluation trench nearby, to the south-west of the church of St Mary and St Nicholas (at SU 0944 3120), failed to find any ditch and bank, although structural remains, provisionally identified as medieval, were revealed. Unfortunately, these remains could not be further investigated due to the high water table (WAM 1972, 176). In 1963, examinations carried out in Kingsbury Square during construction work recorded structural remains, including a medieval house of probable 13th or 14th century date, but failed to identify any deposits earlier than 12th century (WAM 1964, 189). However, Haslam comments on the discovery of an iron- smelting site of probable Saxon date during the construction of a new health centre to the east of the market place in 1975 (Haslam 1976, 68-9).

ST JOHN’S HOSPITAL

(Figures 2-4) INTRODUCTION

Wessex Archaeology was commissioned to carry out a programme of archaeological work on land to the south of St John’s Hospital, West Street (SU 0938 3139), following the submission of a planning application to construct 22 new almshouses on the site. The programme of work comprised an evaluation undertaken in November 1993, followed by excavation in June and July 1996, and a watching brief in October 1996 during development (Figure 2). All of these stages of work were undertaken according to briefs issued by the County Archaeological Officer for Wiltshire, and the results are presented and discussed together below.

The area proposed for development covered c. 3000m? on the north-west side of Wilton, and lies approximately midway between the River Wylye to the south-west and the River Nadder to the north- east. The north and west limits were closely defined

184 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

St John's Square

St John's Hospital

Excavation trench Test-pit 1970s trench

Watching brief area

Fig 2. St John’s Hospital: Extent of archaeological work and Saxon defences

EXCAVATIONS IN WILTON, 1995-6

by an existing ditch/culvert which connects the two rivers and is the traditional border between Wilton and the northern suburb of Little Ditchampton (formerly in the parish of Burcombe). The southern limit was formed by Netherwells (or Netherwalls) Lane, and the eastern limit by the rear boundaries of existing almshouses fronting onto West Street. At the time of the archaeological excavation, the northern half of the development area was covered by allotments and the southern part by waste ground and several mature trees. The ground surface sloped very gently down to the north, from a height of c. 54.70m OD to c. 54.10m OD adjacent to the culvert. The drift geology of the area consists of a spur of valley gravel, overlying solid geology comprising Cretaceous Upper Chalk (Ordnance Survey 1976).

St John’s Hospital, immediately to the north of the proposed development area, was founded in the 12th century to provide accommodation for old and infirm poor people, and in this capacity has continued as an almshouse to the present day. The 1844 tithe map for Wilton and the 1859 Wilton inclosure award do not show the hospital, as it lay within the parish of Burcombe at that time, but they do indicate at least three buildings fronting on to the north side of Netherwells Lane.

During the early 1970s two separate evaluation trenches were excavated within the proposed development area (WAM 1971, 191; WAM 1972, 175- 6), although their precise locations remained uncertain until the recent work (Figure 2). These collectively indicated a 7m wide south-west to north- east aligned ditch at least 1m deep, with a possibly associated bank of river gravel 0.7m high to the south- east. A possible timber revetment consisting of posts and wattle was identified on the south side of the ditch. None of these features was dated, but the bank had been cut by a foundation trench for a stone wall which contained undiagnostic medieval pottery. It was suggested that the bank and ditch was that referred to in an Anglo-Saxon charter of 1045 relating to Ditchampton (Grundy 1919, 76, 290), and also part of the defensive earthwork recorded in the Burghal Hidage.

~METHODS

The evaluation (Wessex Archaeology 1993) comprised four 2m square, hand-dug test-pits, which were sited so as to avoid allotments in use at the time (Figure 2, TPs 1-4). The subsequent excavation (Wessex Archaeology 1996a) comprised a single

185

trench measuring 25m by 5m, representing approximately 22% of the proposed building footprint. This trench was located within the building footprint and positioned so as to lie across the line of the defences deduced from the test-pits and the earlier, 1970s evaluation trenches (see Figure 2). The topsoil and various post-medieval and modern levelling deposits, mainly confined to the western half of the trench, were removed by machine, as were the majority of the bank deposits in the eastern half of the trench, and the features revealed were then excavated by hand. A watching brief was undertaken during the initial stages of development (Wessex Archaeology 1996b) and this allowed the recording and limited investigation of a large area of the site, which was subject to soil stripping and groundworks. This generally confirmed observations made during the excavation, and enabled some clarification and modification of the interpretation of the defensive sequence.

RESULTS Romano-British deposits (Figure 3a)

A sequence of deposits up to 0.4m thick was present in the eastern third of the excavation trench, sealed and preserved beneath later (Saxon) bank material (Figure 4). Several deposits produced small quantities of Roman pottery, but in a fairly abraded condition and perhaps only layer 125 can be ascribed with reasonable certainty to the Romano-British period. Layer 125, the earliest deposit, directly overlay natural gravel and is interpreted as a buried soil. It comprised a homogeneous brown silty clay loam up to 0.15m thick, and produced 29 sherds of Roman pottery including several dateable to the Ist — 2nd century, the remainder being undiagnostic. Partly overlying this buried soil and a possible post-hole, 130, was a gravel surface, 127, which produced four sherds of undiagnostic Roman pottery and a Romano-British penannular iron brooch, and may have continued further to the west as surface 75. A shallow, irregular feature, 93, cut gravel surface 75, but its relationship with a relatively thick layer of rammed chalk, 116, was unclear. This chalk surface was up to 0.2m thick and overlay buried soil 125, protecting it from erosion or mixing with later material in this area. A thin silty layer, 126, over chalk surface 116 produced two sherds of 3rd — 4th century pottery, the only late Roman pottery identified from the site. Finally, an apparently linear spread of gravel, 92, running south-east to

186 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

A Romano-British

Medieval feature LJ Post medieval/modern feature

ieee Compacted chalk/chalk & mortar spread

Fig 3. St John’s Hospital: Phase plans

north-west, sealed the majority of the putative Romano-British deposits, but this spread produced no dating material.

Ground conditions during the watching brief did not enable the full extent of any of these layers to be determined.

The Saxon defences (Figures 3b and 4)

The principal remains revealed by the archaeological work comprise a ditch, bank and associated features. At least two phases of construction were identified in the 1996 excavation and watching brief, with

additional information provided by the evaluation trenches of the early 1970s. The earliest phase has been assigned, despite a paucity of dating evidence, to the burh defences of the late Saxon period, with subsequent phases attributable to either the late Saxon or medieval periods.

Ditch 24/42

A large ditch, 24/42, 6m wide, crossed the centre of the site, running east-north-east to west-south-west. A single section excavated across this in 1996 showed it to be only c. 1m deep, but to have been recut on at least one occasion. Both phases of ditch had fairly

187

EXCAVATIONS IN WILTON, 1995-6

soougJap UOXeS JO SUOIIIAS PajgN.AISUOIIT PUB pap1OIOY -jeudsoH s,uyof1g “fF FL

(uoxes aye94) Eeseud [SY (Ainjueo YIOLsé — UOXeS a]eq) Zz eseud KAY (Aunjueo ues — uoxes a}e7) | eSeUd ESQ

(ynd—91)

fepesied seq Pel

SESSA

-seouajap SSO1Oe UOI}OaS peseud

peaids sej1OW 9 WeYd/HJ/EYO pajoedwoDg Fe]

jeuajyew yueg RX

jeune NZ jeamen 22 8Y Id

Sel

OL

SAQA AS PE.’ youad) uoenjeag

y WSS = LL6L

WSS ,

zl Wy [ ; tL SS, i]

eb/ve YONG

188 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

gently sloping sides and flat to very gently rounded bottoms. The earlier ditch, 42, contained primary fill 34 which was a black, slightly clayey silt. The subsequent fills, 43 and 27, were both dark grey clays, with layer 27 containing more gravel inclusions. None of these ditch fills produced any dateable finds.

Ditch 42 was recut slightly to the north as ditch 24, which was approximately 5m wide. The primary fill, 32, was a dark grey clayey silt, overlain by a greyish brown loamy clay, 23, containing occasional fragments of unworked stone. This in turn was sealed by a thick deposit of re-deposited gravel, 22. Fills 32 and 42 contained a few fragments of animal bone, but no dateable finds. A shallow gully, 8, and a post-hole, 36, lay on the north-west side of ditch 42, but both are undated and their relationship unclear.

Post-hole alignment 134

Approximately 4m to the south-east of ditch 24/42 was a line of seven post-holes, 134 (comprising post- holes 109, 112, 114, 118, 120, 122 and 124), each between 0.18m and 0.28m in diameter (average 0.23m) and between 0.08m and 0.28m deep (average 0.17m). These were spaced at regular 0.7m intervals, cut gravel surface 92 (of possible Romano-British date), and the majority were cut by the foundation trench for wall 15. All had steep sides, rounded bottoms and were filled with homogeneous grey silts containing variable amounts of flint gravel and occasional chalk flecking. One post-hole, 109, contained larger flint nodules possibly used as post- packing, but none produced any finds. A further post- hole, 133, lay at 90° to this line and had a more pointed profile. Post-hole alignment 134 continued to the south-west beyond the limits of the trench, but apparently did not continue further to the north-east (but see below). Unfortunately, ground conditions during the watching brief did not permit the identification of further examples either to the south- west or north-east. However, a post-hole on this alignment was recorded immediately to the north- east in the 1970 evaluation trench (David Algar pers. comm.). It is suggested below that post-hole alignment 134 marked a timber palisade or timber- faced revetment to an earthen bank.

Test-pit 2 revealed part of a large, narrow slot or ditch, 0.45m wide and c. 1m deep with vertical sides and a flat bottom. This ran at 90° to ditch 24/42 (although the relationship between the two features could not be ascertained), was stratigraphically early, but produced no finds. The function of this slot is unclear, but it may have been associated with post- hole alignment 134.

Wall 15

Immediately to the north-west and cutting several of the post-holes in alignment 134 was a broad, shallow foundation trench, 89, which ran parallel to ditch 24/ 42. This was 1.2 - 1.5m wide, up to 0.3m deep, had steep sides and a flat bottom, and contained the remains of a wall, 15, and rubble foundation 38. Both wall and foundation had been extensively robbed, but sufficient was recorded, particularly during the watching brief and the evaluation (in test-pit 4), to indicate that they continued on broadly the same alignment for at least 30m across the development area. Rubble foundation 38 comprised flint nodules in a matrix of red, sandy clay, with wall 15 built on top of this along the south-east side. Wall 15, where it survived, was approximately 0.7m wide and up to 0.8m high. The base of wall 15 comprised mainly flint nodules with some fragments of chalk and Chilmark stone, perhaps forming a levelling deposit. The wall did not survive above this level in the excavated area, but the watching brief revealed that on top of this, in places, were roughly-squared blocks of Chilmark stone (maximum observed dimensions 0.3 — 0.4m) with some remnants of a pale greyish brown mortar surviving in between. Insufficient material of wall 15 survived or was exposed to enable more precise details of its construction (e.g. coursing) to be determined. To the north-west of foundation trench 89, a thin, discontinuous spread, 91, of greyish brown mortar mixed with frequent small chalk fragments and occasional flints extended as far as the edge of ditch 24/42. It is suggested below that wall 15 was a revetment to an earthen bank, repiacing an earlier timber-faced revetment or palisade.

Bank

Several deposits have been interpreted as forming bank material associated with either or both phases of ditch 24/42, post-hole alignment 134 and wall 15. Most of these deposits were recorded in section only and so the precise sequence and their extent remains uncertain, with the schematic section presented in Figure 4 representing a combination of the evidence from the excavation and watching brief. This indicates two, or possibly three phases of bank, some 10m wide in total and surviving to a maximum height of approximately 1.5m.

The earliest bank material appears to have been 110, a layer of light greyish brown to dark yellowish brown clayey silt containing an increasing amount of gravel towards the east (observations made during the watching brief indicate that this layer extended beneath wall 15, but not as far to the north-west as

EXCAVATIONS IN WILTON, 1995-6

ditch 24/42; no relationship could be determined between layer 110 and post-hole alignment 134). The only finds recovered from layer 110 were some animal bone and a small quantity of iron-working slag.

Layer 110 was sealed by layer 29, a light greyish brown clayey silt up to 0.4m thick (observations made during the watching brief suggest that much of this layer comprised redeposited natural sand and gravel) which produced four sherds of late Saxon pottery, three residual Roman sherds and a single, possibly intrusive medieval sherd. Layer 29 was restricted to the south-east of wall 15, with layer 12 to the north- west possibly contemporary. Layer 12 sealed chalk and mortar spread 91 (see above) and comprised a greyish to orange brown sandy clay containing flint gravel and larger nodules. Finds from this layer which, like 29, had suffered from some later disturbance caused by treeholes, comprised animal bone, a fragment of lava quern, nine sherds of Saxo-Norman pottery and one medieval sherd.

The final deposit assigned to the sequence of bank material was layer 98, a layer of redeposited natural sand and gravel up to 0.3m thick which partly capped layer 29, but produced no finds.

Medieval and later features (Figure 3c)

Only two medieval features were identified, ditch 4 and pit 48. Ditch 4, exposed in the corner of the excavation trench, ran parallel and approximately 3m to the north-west of ditch 24/42. The profile of this ditch was not ascertained, but it was at least 2m wide and in excess of 1m deep, with the south-east side sloping at approximately 45°. The watching brief failed to distinguish any division on the surface between the dark greyish brown fills of ditch 4 and the more recent fills of the ditch/culvert which extended around the north and west sides of the development area. Ditch 4 did not appear in test-pits 1 and 3, both of which exposed spreads of dark greyish brown silty soils between 0.13 and 0.65m thick overlying natural gravel. These spreads produced pottery of 12th — 15th century date, and were sealed by deposits containing 18th century material. Only a small segment of ditch _4 was excavated, but this produced 17 sherds of medieval pottery, one residual Saxo-Norman sherd and a small quantity of animal bone. Pit 48 (see Figure 4) was a small, sub-oval pit up to 0.45m deep that was partly exposed along the southern edge of the excavation trench. It cut wall foundation 38, contained some burnt clay in its upper fill, and produced two sherds of medieval pottery.

189

A number of post-medieval/modern features were recorded, but are not described in detail here. The majority of these are shown in Figure 3c and comprise several shallow scoops, some probably tree holes; one possible structural feature, 58, filled with layers of compacted mortar; and a relatively substantial V- shaped ditch, 18/45, up to 1.5m wide and 1.4m deep, which ran north-west to south-east across the site and was filled with a homogeneous dark grey silt.

SOUTH STREET (NEW DOCTOR’S SURGERY)

by Rachael Seager Smith (Figure 5)

INTRODUCTION

In September 1995, Wessex Archaeology was commissioned to carry out an excavation in South Street (at SU 09535 31010) in advance of the construction of a new doctor’s surgery. The development area lay on the south-east side of South Street, at a height of approximately 53m OD, less than 100m to the north-east of Bull Bridge and close to the site of St Michael’s Church (see Figure 1). The excavation followed an earlier evaluation of the site which had identified evidence for structures and deposits of medieval and post-medieval date extending back from the South Street frontage (Wessex Archaeology 1995).

The development site covered c. 1060m2, and the principal aim of the excavation was to record a sample area of the archaeological remains before their possible destruction during development. Much of the site was designated to be a car-park, but an area of c. 240m? was to be directly affected by the foundations of the proposed new building. Therefore, the excavation trench, covering approximately 80m? (33% of the area to be affected by the foundations), was located largely within the footprint of the proposed building.

At the time of the excavation the site was covered by overgrown gardens, the layout of which had remained unchanged since at least 1869 (OS 25” map, 1st series). Topsoil and subsoil (up to 1m thick) was removed by machine and all subsequent excavation undertaken by hand. Although not all of the deposits were fully excavated, the entire sequence was investigated. The underlying geology comprised alluvial clay deposits over Valley Gravels.

190 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

RESULTS attributable to before the 9th century and, although

the stratigraphy appears to represent several phases The excavation revealed a sequence of late Saxon, of medieval activity, the pottery indicates a medieval and post-medieval deposits, the majority of comparatively tight date range, extending little beyond which are of 9th - 13th century date. None is definitely the early years of the 14th century. The principal post-

[| Excavation trench

@—mm Evaluation trench

Footprint of new doctor's surgery

261

Z, STA W210 Mi, AS A DX, z t Chalk cob wall 110 B29WAM SSN AY fa oe | yy’ A

wt 34255 a

WIR NAS NSO YAN AN ESANSN IO

Hearth Wall 255 360

Robbed wail 219

ARS 9th —12th century =| 17th—18th century 12th-13th century [___] 19th-20th century 16th century MM Mortar mixing pit

Fig 5. South Street: Site location and phase plan

EXCAVATIONS IN WILTON, 1995-6

medieval remains are probably of 16th century date, but there were no associated artefacts, and the majority of the later features cannot be closely dated.

A number of features have been assigned to the 9th/10th — 12th century on the basis of ceramic dating or stratigraphic relationships, and at least three of these are of probable pre-Conquest date. These earlier, late Saxon features comprise pits 263, 285 and 288, and the later, Saxo-Norman features comprise small pits or scoops 316, 322, 339 and 354, hearth 360, and possible structural features represented by post- holes 284, 351 and 366. Several of these features had been cut by later, medieval ditches and an insufficient area was exposed to discern any structural arrangements or feature groupings

A considerable amount of iron-working (smelting) slag was present (a total of 126.64kg was collected from a much larger quantity), especially in the south- eastern corner of the site, suggesting that it lay in close proximity to an iron-working complex probably involved in bloomery smelting, with the possibility of some smithing. A small quantity of pottery found with the slag suggests that it is likely to have been deposited in the 12th century or earlier, though it probably post- dates the group of-features assigned to the 10th — 12th century.

Ditch 269/278 was probably the earliest medieval feature assigned to the 12th — 13th century, running east-west across the site and possibly marking a property division. Ditch 269/278 had been recut, with the earliest phase possibly dating to the 12th century and the later phase to the 13th century. Ditch 269/ 278 was subsequently replaced by ditch 299/320 to the north, which had itself been recut and is broadly dated to the 13th century. Several pits assigned to the 13th century were either cut by ditch 299/320 (pit 329), cut ditch 269/278 (pits 274, 318 and 371), or had no relationship with these ditches (pit 312). Towards the base of the sequence the features and deposits were waterlogged, preserving organic materials such as wood, textiles and leather, and one of the pits, 329, had a wicker-work lining.

A substantial flint-and-mortar wall footing, 255, was built over the slag deposits in the southern part

- of the site and several, probably related, construction features, including two wood-lined mortar mixing pits (243 and 342), were also identified. Dating evidence was sparse; a small quantity of 13th century pottery was present, but stratigraphically these features represent a later phase of activity and it is probable that the pottery was re-deposited. A stone-built, clay- lined culvert, 261, dug into the top of ditch 229/320, may also have been associated with this phase of

19]

construction. Its presence suggests a continuing need for drainage on the site, a possibility hinted at by the presence of medieval ditches 269/278 and 229/320, both of which had been recut.

Four small pits or post-holes (252, 293, 327 and 330) may all have belonged to a later group of features of 17th — 18th century date, possibly forming part of a single structure. The ceramic assemblage from these features consists of residual 13th century material, as well as some post-medieval sherds from the upper fills only. Pit 213 is broadly contemporary with this group, and a right-angled chalk cob wall footing, 110, exposed in the evaluation trenching (but not further investigated) is most likely to have been of the same, post-medieval date. However, there is a possibility that this wall may have dated to as early as 16th century or as late as 19th century. Wall 110 was almost 1m wide and probably marked the rear of a building fronting onto South Street approximately 10m to the west.

In the 19th century, a trench (225) was dug, possibly robbing out an earlier wall foundation, and various other robbing episodes were identified, including that of the eastern part of the substantial flint-and-mortar wall footing 255 (robber trench 219). Post-medieval/modern features also included a small, rectangular, chalk-lined structure (206), probably a cess pit, demolished and backfilled during the 19th century, though possibly constructed as early as the 17th or 18th century.

FINDS by Lorraine Mepham

The emphasis in this section is on finds from medieval or earlier contexts, together with finds presumed to be of this date but found redeposited in later contexts. Of the post-medieval finds only the pottery, and certain other objects of intrinsic interest, are considered in any detail here.

POTTERY

This report covers the pottery from both St John’s Hospital and South Street. The assemblage from St John’s Hospital amounts to 157 sherds (5385 g); this is a small collection of mixed date — material of prehistoric, Romano-British, late Saxon, Saxo- Norman, medieval and post-medieval date is present. General condition (apart from the more recent material) is poor, and many sherds have rolled edges

192 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

and abraded surfaces. The assemblage from South Street is larger (2081 sherds; 91,410 g), and is in markedly better condition; sherds are fresh and unabraded, and several partial profiles are present. This assemblage is more restricted 1n date range, and the primary interest here lies in the late Saxon/Saxo- Norman (mid 9th-12th century) component, although later medieval and post-medieval pottery 1s also present.

Methods

Analysis has focused on the late Saxon/Saxo-Norman and medieval pottery, and this has involved detailed analysis of fabric and form, following the standard Wessex Archaeology pottery recording system (Morris 1994). Fabrics have been defined and coded on the basis of dominant inclusion type, and comprise one calcareous fabric (Group C), two flint-tempered fabrics (Group F) and seven sandy fabrics (Group Q); there are also six fabrics of known type or source (Group E). Vessel forms have been defined using rims and other diagnostic sherds, and follow the recommended nomenclature for medieval vessel forms (MPRG 1998). Details of surface treatment, decoration, manufacture and evidence of use have also been recorded; detailed pottery records by context are held in archive. Prehistoric and Romano-British pottery, present in much smaller quantities, has not been analysed to this level of detail, but is broadly described in terms of the types present, as is the post-medieval pottery, all of which falls into well known local, regional or imported types. Fabric totals are given in Table 1.

Prehistoric and Romano-British pottery

With the exception of a single Romano-British sherd from South Street, all of the prehistoric and Romano- British pottery described here came from St John’s Hospital. All sherds are small and abraded (mean sherd weight 5.4 g).

Four sherds are of prehistoric date; these comprise two sandy sherds and two sherds with sparse organic temper. While the possibility that the organic- tempered sherds could be Saxon should be noted, particularly given the site context, the low frequency of the organic inclusions is more in keeping with a date range in the Early Iron Age, and the sandy sherds, although not diagnostic, could be of similar date. All four sherds were found as residual material in later (Romano-British and medieval) contexts at St John’s Hospital.

The remaining 44 sherds are of Romano-British date. These include one sherd of samian, one sherd of a North Gaulish roughcast colour-coated beaker, 26 sherds of coarse greywares and 15 sherds of coarse oxidised wares; the coarsewares are almost certainly from more than one source. Apart from the finewares, which are of Ist or 2nd century date, the only diagnostic sherd comes from a greyware dropped- flange bowl of 3rd/4th century type. Forty of these sherds came from contexts at St John’s Hospital which pre-date the late Saxon defences, although only buried soil layer 125 (29 sherds, including the samian and North Gaulish colour coat) has been confidently dated to the Romano-British period. Other sherds came from gravel surface 127 (four sherds), feature 93 (one sherd), silty layer 126 (two sherds, including the dropped-flange bowl), and layer 76 (four sherds).

Late Saxon and Saxo-Norman pottery

Six fabrics were defined as falling within a potential date range of mid 9th to 12th century:

E400 Cheddar fabric E (Peacock 1979); a hard- fired, wheelthrown fabric with varying frequency (rare to sparse) of quartz sand, limestone, often leached, degraded sandstone and patinated flint or chert; variable firing but generally oxidised surfaces.

C400 Limestone-tempered ware: hard, moderately coarse clay matrix, slightly micaceous; common, fairly well sorted, crushed limestone <2mm; rare subrounded quartz <lmm; rare iron oxides; handmade; generally unoxidised.

F400 Flint-gritted ware: hard, moderately coarse clay matrix; sparse, poorly sorted, subangular flint (patinated and unpatinated) <3 mm; sparse to moderate, subrounded quartz <1mm; rare limestone; handmade; variable firing.

F401 Flint-gritted ware: hard, moderately coarse clay matrix; sparse, poorly sorted, subangular flint (patinated and unpatinated) <2mm; rare iron oxides; handmade; generally unoxidised but variable.

Q400 Sandy greyware: very hard, moderately coarse matrix; abundant, well sorted, subrounded/ subangular quartz <0.5mm; wheelthrown; unoxidised mid to blue grey.

Q403 Possibly a less hard-fired, handmade version of E400; a similar range and frequency of inclusions (quartz, limestone, sandstone); a slightly soapy feel; variable firing but more frequently unoxidised.

EXCAVATIONS IN WILTON, 1995-6

Fig 6. Pottery (South Street)

Also potentially within, or at least with a late date within this date bracket, is the coarsest variant of the Laverstock-type wares described below (E422A). This fabric does occur here in association with the above fabrics, but is known to continue in use into the 13th century from evidence from Salisbury (Mepham 2000). The coarser of the two Crockerton-type fabrics (Q405, see below) probably has a similar date range; a comparable fabric was found in 10th-12th century and later contexts at Trowbridge (Mepham 1993, Q402). One of the six fabrics is of known type (Cheddar E), anda second fabric (Q403) is visually very similar; the two were frequently difficult to distinguish, but differ slightly in firing conditions, and the fact that while Cheddar E is always wheelthrown, fabric Q403 is handmade. Both fabrics occur in very similar jar forms, rounded with flared necks (at approximately _ 45° to body wall) and simple or slightly thickened rims; examples in Q403 are sometimes finger- impressed (Figure 6, 1, 3). One other unusual vessel form is present (Type 2): a lid in Cheddar E (Figure 6,6), which is as yet without direct parallel. The limestone-tempered and flint-tempered fabrics are all likely to be of at least fairly local manufacture. Only one diagnostic form is present, a

1193,

jar rim of a similar form to the examples in E400 and Q403 (Figure 6, 2).

The most interesting part of this group, however, comprises sherds in a hard, grey, wheelthrown fabric (Q400). This fabric occurs in jar forms with everted, thickened rims, with slight neck cordons or horizontal tooling and incised curvilinear decoration around the shoulder; two well-preserved partial profiles came from pit 263 (Figure 6, 4, 5). Several sherds have noticeable exterior burning or sooting. While this greyware fabric has not as yet been identified in other assemblages of this date in the county, it falls within the range of ‘Late Saxon Sandy Ware’ identified in small quantities at Winchester, where it is dated c. AD 850-pre 950 (Biddle and Collis 1978; Matthews forthcoming); vessel forms, decoration and the external burning/sooting can all be paralleled in the Winchester assemblages. At St John’s these sherds were found only as residual material; at South Street they occur in most instances in association with Saxo- Norman wares, but in isolation in pits 263, 285 and 288. The significance of these vessels, and of the rest of the late Saxon/Saxo-Norman group, is discussed further below.

194 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Table 1. South Street and St John’s Hospital: Pottery fabric totals

Fabric South Street St John’s Hospital TOTAL

[1] [2] [3] [1] [2] [3] [1] [2] [3] Prehistoric Sandy - - - 2 20 76.9 2 20 76.9 Sandy/organic - - - 2 6 23.1 2 6 23.1 Sub-total - - - 4 26 - 4 26 -

Romano-British

E100 - - - 1 16 7.0 1 16 6.7 E120 - - - 1 1 0.4 1 1 0.4 Q100 1 10 100 26 145 63.6 27 155 65.2 Q101 - - - 14 64 28.1 14 64 26.9 Q102 - - - 1 2 0.9 1 2 0.8 Sub-total 1 10 - 43 228 - 44 238 - L.Saxon/S-N

C400 20 242 4.2 - - - 20 242 4.1 E400 64 1354 ° 23:1 - - - 64 1354 22.8 F400 Jf 104 1.8 - - - 7 104 TEVA F401 Livi 138 2.4 - - - 17 138 23 Q400 102 2469 42.8 2 8 52 104 2477 =41.7 Q403 71 1483 25.7 10 145 94.8 81 1628 27.4 Sub-total 281 5770 - 12 153 - 293 5943 - Medieval

E420 6 66 tel 1 4 0.9 7 70 1.1 E421 29 458 7.9 5 68 14.7 34 526 8.3 E422A Le 1910 32.7 12: 159 34.3 129 2009 32.8 E422B 130 1960 33.9 15 84 18.1 149 2065 32.8 E422C 57 753 12.9 4 28 6.1 61 781 12.4 Q401 1 14 0.2 - - - 1 14 0.2 Q402 Dili 328 5.6 1 54 LO 28 382 6.1 Q404 1 6 0.1 - - - 1 6 0.1 Q405 10 224 3.8 3 32 6.9 13 256 4.0 Q406 2 10 0.2 1 34 7.3 3 44 0.7 “Tudor Green’ 1 1 - - - - 1 1 - Pearly Verwood 2 94 1.6 - - - 2 94 1.5 Sub-total 387 5845 - 42 463 - 429 6308 - Post-medieval

Verwood 994 54488 54 4497 1048 58985 Redwares 22 444 1 8 23 452 Slipware 7 220 - - 7 220

Staffs type 9 60 - - 9 60

Tinglaze 9 71 - - 9 71

Fine redware 3 38 - - 3 38 Creamware 63 1041 - - 63 1041 Pearlware 105 1135 - - 105 1135 Jackfield 4 70 - - 4 70 Stonewares 51 1411 - - 51 1411

White saltglaze 4 46 - - 4 46

Porcelain 2, 15 - - p) 15

Industrial 139 746 1 10 140 756 Sub-total 1412 59785 - 56 4515 - 1468 64300 - TOTAL 2081 71410 - 157 5385 - 2238 76795 -

Note: [1] = number of sherds; [2] = weight; [3] = % of period.

EXCAVATIONS IN WILTON, 1995-6

Medieval pottery

The later medieval assemblage is dominated by coarseware sandy fabrics comparable to products of the Laverstock kilns outside Salisbury, which were operating in the mid 13th century (Musty etal. 1969), although the overall date range for these types is likely to have been much wider. For assemblages within Salisbury, a type series has defined one basic coarseware type, subdivided on the basis of inclusion size, and two basic finewares, again subdivided (Mepham forthcoming a). The following types are present at Wilton:

E422 Laverstock type coarseware: tempered with common, subrounded quartz, frequently iron-stained; variants A (coarsest, quartz <1mm, ‘pimply’ surfaces), B (medium, quartz <0.5mm) and C (finest, quartz <0.25mm); all handmade; variable firing.

E420 Laverstock type fineware 1 (all variants), quartz <0.5mm, not iron-stained; firing buff to cream.

E421 Laverstock type fineware 2 (all variants), quartz <0.5mm, iron-stained; firing buff to pale salmon pink.

The coarsewares occur almost exclusively in jar forms, which again can all be paralleled within the Laverstock/Salisbury type series: rounded or slack- profiled jars, probably round-based, with a limited range of rim forms (Musty et al., 1969, fig. 7; Mepham 2000, types 1-4). Other vessel forms include one bowl or possible curfew and three possible pitcher forms in E422B, and an internally glazed dripping dish in E422C.

Five other fabrics were identified, all sandy:

Q401 Hard, moderately coarse, slightly micaceous matrix; moderate, well sorted, subrounded quartz <0.125mm; sparse iron oxides; wheelthrown; oxidised (pale-firing) with unoxidised core.

Q402 Hard, moderately coarse, micaceous matrix; moderate, poorly sorted, subrounded quartz <0.5mm; sparse limestone <0.25mm; rare subangular - flint; sparse iron oxides; handmade; variable firing.

Q404 Hard, moderately coarse matrix; common, well sorted, subrounded quartz <0.125 mm; rare iron oxides; handmade; oxidised with unoxidised core.

Q405 Hard, moderately coarse, micaceous matrix; sparse, poorly sorted, subrounded quartz <lmm; sparse degraded sandstone <3mm;; rare patinated flint <1mm;; sparse iron oxides; handmade; firing variable.

195

Q406 Hard, fine, slightly micaceous matrix; common, fairly well sorted, subrounded/subangular quartz <0.125mm; rare iron oxides; wheelthrown; unoxidised with oxidised (pale-firing) interior surface.

Two of these (Q402, Q405) are comparable to micaceous fabrics found widely across west Wiltshire and with at least one putative source at Crockerton outside Warminster (Smith 1997). Q405 is a coarser variant and could therefore be postulated to have an earlier start date than Q402. Here they are found only in jar forms, rounded with flared necks and thickened “T-shape’ rims.

Fabrics Q401, Q404 and Q406 are each represented by just one or two sherds, and are all of uncertain source. The single sherd of Q404 has applied rouletted strips and is likely to be of 13th century date. Sherds in Q401 and Q406 derive from glazed jugs; these finer fabrics are probably of late medieval date, perhaps later 14th or 15th century.

Post-medieval pottery

Predominant amongst the post-medieval assemblage are Verwood type earthenwares from east Dorset, and their pre-eminence here reflects Verwood’s virtual monopoly of the earthenware market from the 18th century onwards. Other coarse earthenware types are restricted to a handful of coarse redwares and slipwares, amongst which can be identified Crockerton type wares, and possible north Somerset products.

A few sherds of Staffordshire-type slipware and tinglazed earthenware are present. Eighteenth century wares are represented by white salt glaze, fine redware, creamware, Jackfield ware and pearlware; some of the stonewares could be of this date or later.

Late Saxon/Saxo-Norman and medieval ceramic sequence

The assemblages from St John’s Hospital and South Street combine to provide a good ceramic sequence from the late Saxon period through to at least the end of the 13th century, and then from the 18th century onwards.

Prior to this sequence, the occurrence of prehistoric and, in particular, Romano-British material, albeit in such small quantities, is nevertheless significant. Pottery of this date has not previously been found in Wilton, and this small group provides the first clear evidence for activity of this date in the town.

196 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Perhaps most interesting amongst the late Saxon/ Saxo-Norman assemblage is the small group of wheelthrown greywares (Q400), which are paralleled in well-stratified mid 9th to mid 10th century deposits in Winchester. As yet these wares have not been identified elsewhere in Wiltshire. This may be at least partly due to chronological factors in that sites of known 9th/10th century date have so far proved elusive. It may be noted that the greywares (and indeed any wares which could be definitively assigned to the period of the 8th to 10th centuries) were absent in assemblages from Trowbridge and Market Lavington, where early/middle Saxon (5th to 8th century) and Saxo-Norman (10th to 12th century) wares were identified (Mepham 1993; forthcoming), and none is known from late Saxon Cricklade (Jope 1972). Another factor, however, may be lack of recognition - these greywares, particularly undiagnostic body sherds, could easily be mistaken for Romano-British wares. Their identification at Wilton, therefore, is a particularly significant addition to our knowledge of the regional distribution of late Saxon ceramics, despite the fact that their precise source remains uncertain (Biddle and Collis 1978, 133; Matthews forthcoming).

For the Saxo-Norman period (10th to 12th century), the pottery of the region is as yet poorly understood in comparison to the later medieval period, and only one substantial and well stratified assemblage which spans part of this period has been published, from Trowbridge (Mepham 1993). The latter site provides parallels, from deposits associated with the Saxo-Norman manorial settlement (c. AD 950-1139), for the limestone-tempered jars, and the wheelthrown Cheddar type wares, found at Wilton (ibid., figs. 35, 37), and similar limestone-tempered jars found elsewhere in north Wiltshire are dated on stylistic grounds to this period (e.g. Currie 1993, fig. 8). The use at this time of calcareous fabrics for a limited range of simple vessel forms is characteristic of a widespread ceramic tradition across southern England; Cheddar yielded a range of limestone tempered fabrics alongside the wheelthrown wares (Rahtz 1979), and in Oxfordshire late Saxon shelly wares continue into the early medieval period (Mellor 1994).

In the case of both Wilton and Trowbridge, a dominant local industry is represented from an early date. At Trowbridge this was apparently in operation by the 10th century, probably in the Avon valley, and producing vessels in sandy and sandy/calcareous wares which were found alongside the wares described above (Mepham 1993, fabrics C401 and Q400). The sequence at Wilton demonstrates the early appearance

of Laverstock-type coarsewares (E422A) prior to the foundation of Salisbury and before the known date range of the excavated kilns; interestingly, evidence from Salisbury indicates the long survival of these archaic forms (Mepham 2000). The foundation of the city in the early 13th century presumably stimulated this local industry which quickly superseded the limestone-tempered wares to dominate the market. In Salisbury itself Laverstock type wares occur almost to the complete exclusion of other wares (Mepham and Underwood n.d.; Mepham 2000), although they do occur at Wilton, as is shown by the presence of the micaceous west Wiltshire (?Crockerton) types, albeit in very small quantities.

From the mid 14th century onwards there appears to be a similar hiatus in the sequence as at Salisbury, although whether this is a case of a real absence of pottery or a non-recognition of later medieval types is as yet uncertain. In this instance, there are only a handful of sherds which can with any certainty be attributed to the later medieval or early post-medieval period. These include a single sherd of “Tudor Green’ ware and two sherds of a possible early Verwood type ware (E642); both types occur only as residual sherds in later post-medieval contexts at South Street.

List of illustrated vessels (Figure 6)

1. Jar, fabric E400. South Street, PRN (Pottery Record Number) 185, context 250, clearance.

2. Jar, fabric C400. South Street, PRN 43, context 203, robbing event (wall 255).

3. Jar, fabric, finger-impressed rim, fabric Q403. South Street, PRN 219, context 250, clearance.

4. Jar, fabric Q400, incised curvilinear decoration around shoulder. South Street, PRN 252, Obj. No. 1001, context 264, pit 263.

5. Jar, fabric Q400, incised ?curvilinear decoration around shoulder. South Street, PRN 253, Obj. No. 1001, context 264, pit 263.

6. Lid, fabric E400. South Street, PRN 348, layer 335.

CERAMIC BUILDING MATERIAL

The ceramic building material recovered comprised a sample only of what was observed on both sites (273 pieces; 18,839 g). Of this total, only a small proportion (7 pieces, 274 g from South St; two pieces, 74 g from St John’s) came from well stratified medieval contexts (none was found in contexts dating earlier than the 13th century). This small group comprises peg tile fragments in coarse, poorly wedged fabrics,

EXCAVATIONS IN WILTON, 1995-6

some flint-tempered and some sandy; one of the flint- tempered fragments is glazed. Fragments in similar fabrics, including one other glazed peg tile and two glazed ridge tiles, occur in greater quantities in post- medieval contexts, particularly at South Street, where they are distinctive by the nature of the coarse fabrics and the fact that a significant proportion are pale- firing. These pale-firing fabrics have been found in quantity in 13th century and later contexts in Salisbury (Cleal n.d.; Loader 2000), and it is likely that a local source was supplying both centres; one such centre is known from documentary records at Alderbury south of Salisbury from the 14th century (Hare 1991), although this centre, or others, must have operated at an earlier date in the area.

Other ceramic materials comprise fragments of fired clay. Of the overall total recovered, most of which came from South Street, the majority derived from features of late Saxon/Saxo-Norman (1317 g) or medieval (1980 g) date. This consists of small, featureless and undiagnostic fragments, probably of structural origin, from hearths or wattle and daub structures. At South Street, moderate concentrations came from pit 274 and layer 306, and some correlation was noted with the concentration of iron-working slag in the south-east corner of the site, suggesting that at least some of this material may have derived from smelting furnaces or smithing hearths. However, no vitrified material was present nor any fragments with slag or other residues adhering. At St John’s Hospital, asmall group of fired clay (1495 g) from medieval pit 48 also represents hearth lining, but there is no evidence that this derived from anything other than a domestic hearth.

SLAG

A substantial quantity of iron-working slag was recovered from South Street. The majority of this came from deposits assigned to between the 10th — 12th and 12th — 13th centuries, with a smaller quantity probably residual in later contexts. Preliminary examination suggests that most of this is smelting slag from bloomery furnaces, but there “is also some evidence for bloom smithing, and plate hammerscale was noted associated with some of the material. This material appears to represent dumps of debris and no associated iron-working features were identified.

At St John’s, undiagnostic iron-working slag was found in small quantities in various contexts. A small piece of copper alloy waste came from medieval pit 48.

197

WORKED STONE

The stone recovered from South Street is all building material, and consists entirely of roof tiles in either slate or limestone, from both medieval (9 fragments) and post-medieval contexts (20 fragments). From St John’s, one fragment of lava quern, a continental import often found in Romano-British or middle to late Saxon contexts in this country, came from the second phase bank deposit (12).

WORKED BONE AND ANTLER

Seven worked bone (or antler) objects were identified, all from South Street. Also from this site, part of a crudely made iron tool, possibly a sickle, with an antler handle, was found, but was unstratified.

No objects were found in stratified late Saxon or Saxo-Norman contexts, although one object, a pin- beater, is likely to derive from activity of this date. This object (found in clearance levels) is cigar-shaped, with a point at either end (length 138mm), and is polished through use; this is a type well attested on Saxon sites, e.g. Southampton (Addyman and Hill 1969, fig. 29) and West Stow, Suffolk (West 1985, fig. 246, 15-17).

Two objects came from undisturbed medieval contexts: a skate or runner and a comb. The skate/ runner, found in pit 274, conforms to the types described by MacGregor (1985, 141-4 and fig. 76), which have a currency from the 8th to the 13th century in this country. It is made from a bos radius, trimmed so that both ends are ‘upswept’ although not perforated for strap attachments as some examples are. One surface has a heavy polish from use, through which can be seen numerous small striations. The comb (from medieval ditch 269/278), is of one-piece, double-sided form, and can be compared to examples found in Southampton in contexts dating between late 14th and early 17th century (Platt and Coleman-Smith 1975, fig. 249, 1944, 1946-7).

Three further objects are likely to be of medieval date: a spatula, a handle fragment and a decorative strip. The spatula, perforated (for suspension?) at the end of the handle and also at the base of the blade, which is broken, finds a possible medieval parailel at Northampton (Oakley 1979, WB60); this object came from post-medieval pit 213. The handle fragment comprises part of a side plate, decorated with incised diagonal hatching; this came from a post-medieval layer. The strip fragment (clearance levels) is 22 mm

198 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

wide and has incised ring-and-dot motifs; this is possibly a second handle or a decorative mount.

The remaining objects — another handle, and the antler-handled sickle - are likely to be of later medieval or post-medieval date.

METALWORK

A quantity of metalwork was recovered, mostly from South Street, but very few items came from stratified medieval or earlier contexts (six objects from St John’s and six objects from South Street). Objects from St John’s comprise a Romano-British penannular brooch from gravel surface 127, a knife blade from subsoil layer 115, a possible tool from buried soil 125 and at least three nails (?-bank material 98, and layers 126 and 127). From South Street came two iron objects from Saxo- Norman levels (a possible heckle tooth and an unidentified object), and four iron objects from medieval contexts (one horseshoe nail, two other nails, and one unidentified object). There are, however, some more interesting objects from post-medieval contexts at South Street, including three patten supports, spade and shovel blades, a cowbell, and a curry comb.

ORGANIC MATERIAL

Organic material, comprising fragments of wood and leather and other textile, was recovered from waterlogged deposits at South Street. The wood came from four pits, and comprises possible plank fragments from Saxo-Norman pit 339; fragments of wattles from medieval pit 329; a plank from post-medieval mortar mixing pit 342; and a small plank fragment from post- medieval pit 213.The leather comprises two small scraps, probably waste off-cuts, from medieval pit 329.

ENVIRONMENTAL EVIDENCE

ANIMAL BONES by Sheila Hamilton-Dyer

The 517 animal bone fragments recovered from St John’s Hospital amount to 370 separate bones. These were identified using the modern comparative collections of the writer. Undiagnostic fragments have been divided into cattle/horse-sized and sheep/pig- sized fragments. The few measurements are in millimetres and follow the methods of von den

Driesch (1976). Withers heights were calculated using factors recommended by von den Driesch and Boessneck (1974). The archive gives full details of each fragment.

Results

The condition of the material is variable but good on the whole; 65% of the bone could be identified to taxon. A total of ten taxa could be identified in the collection: horse, cattle, sheep/goat, pig, roe deer, dog, cat, hare, goose and domestic fowl. Sheep was positively identified, but no bones could be attributed to goat. A summary distribution of the taxa recovered from each phase is given in Table 2.

Bones from Romano-British contexts form the largest group, and the least well preserved. Most of the material was recovered from layers 125 and 126. Cattle and cattle-sized fragments comprise the bulk of the material and include several metapodia and other foot bones of immature beasts. Other fragments are of meat bones and the material appears to be derived from both slaughter and butchery. Although no dog bones were recovered, several fragments had been gnawed.

There are 82 bones from Saxon contexts and these include one bone each of roe deer, hare and goose in addition to the domestic ungulates. Again, gnawing indirectly indicates the presence of dog. Two complete horse bones offer estimated withers heights of 1.226m and 1.359m, animals of pony size typical of the period.

The 23 medieval bones include one each of dog, cat and fowl.

The remains of at least two dogs of different sizes constitute the majority of the 93 bones from post- medieval contexts. These 73 bones were all recovered from ditch 18.

Discussion

This is a very small assemblage and, therefore, more detailed analysis is inappropriate. Some general conclusions can, however, be made. The majority of the bones identified to taxon are of the expected main domestic ungulates, cattle, sheep and pig. Dog is indicated by gnawing as well as by remains. Birds are represented by a few bones of domestic fowl and goose. Roe deer and hare indicate some contribution to the Saxon diet by hunting.

The period assemblages are quite different in character from each other and probably represent a variety of activities and disposal practices at different times.

EXCAVATIONS IN WILTON, 1995-6

Table 2. St John’s Hospital: Animal bone totals

CHARRED PLANT REMAINS

by Pat Hinton

Samples of ten litres were processed by flotation using Wessex Archaeology’s standard methods with flots retained on 0.5mm mesh and residues on 1mm mesh.

199

Period horse cattle sheep/goat pig roe deer cattle-size sheep-size dog cat hare fowl goose Total RB 4 62 17 D - 62 - - - - - - 150 Late Saxon 3 1 13 4 1 39 9 - - 1 - 1 82 Medieval 2 3 4 2 - 5 4 1 1 . 1 - 23 Post-med 1 6 1 2 - 5 1 73 1 - 2 1 93 undated - 10 3) 2 - - 5 - - - - - 2D Total 10 92 40 15 1 111 19 74 2D, 1 3 2, 370 % 2H, 24.9 10.8 4.1 03 30 Sal 20 0.5 0.3 0.8 0.5 % 62.6 27.2 10.2 Table 3. St John’s Hospital: Charred plant remains Period Romano- Late Saxon Medieval British

Feature Buried soil Ditch 24/42 Pit 48 Context 125 34 46 Sample 3) 6 1 2 Sample vol. (litres) 10 10 10 10 Cultivated Triticum cf aestivum bread wheat 1 1 1 3 Tr. sf spelta spelt - A - - Triticum sp. unspecified wheat 2 1 3 1 Hordeum sp. hulled barley 1 1 1 2 Avena sp. oats - 1 - - Cerealia indet. — fragmented grains indeterminate cereals 0.5 ml <0.5 ml. <0.5 ml. c. 1 75iml: Arable or waste Chenopodium album L. fat hen - - - 1(1) cf Atriplex sp. orache - - - 1 Stellaria media/neglecta common/greater stitchwort = - - 2 - Lychnis flos-cucult L. ragged robin - - 1 - Polygonum cf aviculare knotgrass - - 1 - Centaurea sp. cornflower/knapweed - = - 1 Carex sp. sedge = & = 1 cf Arrhenatherum elatius (L.) false oat grass/ 1 - - - P.Beauv- onion couch stem node Poaceae indet. unspecified grass - - - 1 Wood margin/Waste = = Sambucus nigra L. elder - - = 3 ef Arctium sp. greater /lesser burdock - - - 1

_ Unidentified fragmentary seeds - 1 - 3

Flots and charred items extracted from the residues were examined by stereo microscope at x7-40 magnification.

Charred plant remains are sparse, poorly preserved and probably represent little more than the common residual spread of charred fragments. They

200 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

do, however, provide some illustration of utilised cereals, their weeds and nearby vegetation (Table 3).

The wheats from the Romano-British buried soil (context 125), probably Triticum spelta (spelt), and Triticum aestivum (bread wheat), together with Hordeum sp. (hulled barley) and Avena sp. (oats), are typical for the period. Other seeds are particularly scanty except for a probable fragment of Arrhenatherum elatius (false oat grass or onion couch), possibly present as a ‘weed’.

Late Saxon ditch 24/42, with small quantities of bread wheat and barley present, also included very few charred wild plant seeds, but the one Lychnis flos-cuculi (ragged robin) is of interest in indicating damp or marshy conditions. This sample also includes charred seeds (at various levels of degradation) of Ranunculus, Urtica, Stellaria, Cerastium spp. (buttercups, nettles, chickweeds), Hyoscyamus niger (henbane), Conium maculatum (hemlock), two Carex sp.(sedges), Mentha cf aquatica (water mint) and one Lemna sp. (duckweed). Most of these are common weeds, ruderals or grassland plants, but sedges grow in damp places, hemlock and water mint usually in or beside ditches or streams, and duckweed is a floating plant of still water. It is possible that the uncharred seeds are of comparatively recent date, but it is perhaps equally likely that periods of waterlogging may have slowed the decay of these seeds.

Medieval pit 48, with a cereal content similar to the earlier samples, includes more typical field weed seeds and also evidence, in Sambucus nigra (elder) and (Arctium sp. (burdock), of wood margins or scrub. Damp parts may again be indicated by the seed of Carex sp. (sedge).

THE BURIED SOIL by Michael J Allen

The Romano-British buried soil (context 125) was sampled in a single 210 mm long monolith tin. The sample tin incorporated the bA horizon but, unfortunately, did not cover the boundary between the top of the buried soil and the overlying deposits. The nature and structure of the soil is described below:

0—45mm Brown (10YR 5/3) silty loam, almost stonefree, 0.2% fine micropores, small granular peds in a massive structure, with charcoal fragments to 2mm, diffuse boundary.

Probably turf or upper trampled horizon.

45 — 180mm _ Dark brown (10YR 3/3), slightly darker silty loam (some fine sand) with occasional medium flints, weak medium blocky structure 0.1%

fine and medium micropores. bA horizon.

180-210mm _ Dark yellowish brown (10YR 4/4) firm silty clay to silty clay loam with occasional medium flints, no structure observable in sample, few fine macropores (1mm)

bB (top of B horizon)

The soil examined in the sample represents a buried grassland (or trampled horizon) of a brown earth soil, indicated by a probably worm-worked horizon and the crumb structure.

DISCUSSION

ROMANO-BRITISH

The excavations at St John’s Hospital have revealed the first clear evidence for Romano-British activity in Wilton, although the precise nature and focus of this remain uncertain. The small assemblage of pottery was generally abraded; the only features were two possible post-holes, and the spreads of gravel and the chalk surfaces, while considered most likely to be Romano-British, cannot be dated with certainty. However, it is thought unlikely that these spreads were associated with the construction of the later Saxon defences.

Analysis of buried soil 125 has indicated that it represents an organic grassland or a trampled horizon, and the small animal bone assemblage recovered is dominated by cattle, with evidence for both slaughter and butchery. Small quantities of cereals, including wheat and oats, were also present.

On the basis of this evidence it might be suggested that the remains indicate agricultural, principally pastoral activity, possibly associated with a nearby farmstead, spanning at least the 1st — 2nd centuries AD. The various gravel and chalk surfaces may represent yards or ‘hardstanding’ in what was a relatively low-lying damp area.

LATE SAXON/SAXO-NORMAN

It seems certain that ditch 24/42, the associated bank material, post-hole line 143 and wall 15 formed part of the late Saxon defences of Wilton. However, the

EXCAVATIONS IN WILTON, 1995-6

precise dating of this sequence is hampered by the almost complete lack of associated pottery. The sequence proposed here is, therefore, based largely on the stratigraphic evidence, documentary sources and comparison with late Saxon defences elsewhere. Two principal phases of construction were recognised from the 1996 excavations at St John’s Hospital (see Figure 4).

The earliest phase appears to have comprised a broad, rather shallow ditch, 24, and an associated bank fronted by a timber palisade or revetment represented by post-hole line 134. Between the ditch and bank was a berm approximately 4m wide. No dating evidence was found associated with any elements of this phase of defences, but ditch 24 was subsequently re-cut as ditch 42, and possibly contemporary with this was the replacement of the timber revetment by stone wall 15. Chalk and mortar spread 91 is interpreted as a construction level associated with this wall. Some material appears to have been ‘banked up’ against the front of wall 15 at this time (layer 12), though it is possible that this occurred during subsequent cleaning out or re-cutting of the ditch.

The investigations of the defences at St John’s Hospital in the early 1970s did not distinguish more than a single phase of construction, but this was probably due to the narrowness of the trenches and because work was hampered by a high water-table which prevented a complete section being excavated across the ditch and bank. However, waterlogged timber, possibly the remains of posts and wattle—-work, was recorded on the south side (WAM 1972, 175), perhaps representing the remains of a revetment along the inner edge of the ditch. Unfortunately, this could not be corroborated by the 1996 excavation as no waterlogged deposits were encountered (it seems probable that the water-table in the area has fallen during the past 25 years), and no post-holes were found within the ditch, perhaps a reflection of the narrow width of the excavated section. The 1970s evaluations also revealed the presence of a wall, the same as wall 15 recorded in 1996, and noted the presence of undiagnostic medieval pottery apparently found in association with it. A few sherds of medieval

_ pottery were found within the bank material (in layer 29, equivalent to layer 98; see Figure 4) in the 1996 excavation and, if not intrusive, it is possible that some of this may have been introduced during later refurbishment of the bank or robbing of the wall.

The construction of the earliest phase of ditch and bank at St John’s Hospital may have been associated with the defensive arrangements of the Alfredian burh of the later 9th century, or they may

201

have been in existence prior to this. As Haslam has noted, the last battle of a lengthy campaign by the Danes was fought at Wilton in 871, implying that the town was defended at this time (Haslam 1976, 67). However, the possibility cannot be discounted that an earlier, smaller defensive circuit may yet remain undiscovered in the town. The refurbishment of the defences and the construction of the stone wall at St John’s Hospital could have taken place in the later 9th or early 10th century, and a possible parallel for this sequence is provided by the northern defences of Oxford, another large de novo burh (Hassall and Hill 1970, 189). At Oxford, a timber faced rampart, separated from the ditch by a berm approximately 4m wide, was replaced by a stone revetment of regular ragstone blocks bedded in clayey mortar; these two phases of construction have been dated to the late 9th and early 10th century respectively (Blair 1994, 148, fig. 87).

The suggested extent of late Saxon occupation shown in Figure 1 (based on Haslam 1976, fig. 19) takes into account the Burghal Hidage assessment for Wilton of 1400 hides making it 8= in size of the 33 burhs listed in the early 10th century document (Hill 1969, table II). Whether the boundary to the settlement was defined on all sides by a ditch and bank seems unlikely, and the defences investigated at St John’s Hospital remain the only section yet identified, dug at a convenient point to cut off an area between the River Wylye and the River Nadder. As has been suggested earlier (WAM 1972, 176), it is probable that some of the profusion of watercourses which surround and pass through Wilton served defensive purposes as well as acting as boundaries. Indeed, the open, shallow profile of ditch 24/42 at St John’s Hospital may reflect the likelihood that it too was permanently or regularly flooded, a suggestion supported by the waterlogged conditions encountered in the 1970s evaluations and by the charred plant remains.

The paucity of the finds from the ditch and bank material at St John’s Hospital may indicate a low level of occupation in the immediate vicinity, although the excavation and watching brief could not demonstrate this with certainty. However, the excavation at the New Doctor’s surgery in South Street, just within the postulated southern limit of the late Saxon town, has produced fairly clear evidence for occupation during the 9th/10th — 12th centuries, and thus of probable pre-Conquest date. Some possible structural remains, including a hearth, and several pits were assigned to this period, but little more can be deduced about the nature of the settlement. Of particular interest,

202 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

however, is the group of grey, wheelthrown wares assigned a mid 9th — mid 10th century date. These have not as yet been identified elsewhere other than in Winchester and Kings Somborne (Charlotte Matthews pers comm.), where they occur in small quantities, and their presence in Wilton may be significant in terms of identifying their source and perhaps also in reflecting the date and status of occupation on the site in South Street.

Unfortunately, no other controlled excavations have yet been undertaken in Wilton which might provide information on the extent and nature of late Saxon occupation, and the few intermittent evaluations and watching briefs have produced no firm evidence for pre-Conquest settlement. Recent evaluations on the north side of Russell Street (OAU 1994), and just to the south within Kingsbury Square (Michael Heaton, pers. comm.) have revealed no Saxon features or finds. This might indicate a relatively low density of occupation within much of the postulated area of late Saxon Wilton, which might not be as extensive as indicated on Figure 1, particularly to the north-east of St Mary’s Church along North Street.

Clear evidence of a rectilinear street plan, indicated by the lines of North Street, South Street and West Street (the layout to the east is more confused) has been noted before (Biddle and Hill 1971, 81), and it has been suggested that this may reflect a ‘deliberate policy of urban foundation in response to a military situation’ in the Burghal Hidage places during the late 9th or early 10th century (Biddle and Hill 1971, 83). This seems a likely explanation for the street layout at Wilton, but remains to be demonstrated archaeologically, as does the full extent and nature of the associated late Saxon settlement.

A substantial deposit of iron-smelting slag probably post-dated the 10th — 12th century features at South Street, but is not likely to have been deposited later than the 12th century. Iron-smelting slag was also recorded in 1975 to the east of Kingsbury Square where it was attributed to the late Saxon period (Haslam 1976, 69), and it is conceivable that the material from South Street is of late Saxon date, and may represent ‘industrial activity’ towards the periphery of the town. A possible parallel might be drawn with Romsey, Hampshire, where large deposits of iron-smelting slag have been recorded in probable pre-Conquest deposits across the town, indicating the existence of a relatively substantial industry at an early date.

MEDIEVAL — POST-MEDIEVAL

Medieval Wilton had no town defences, and it is uncertain when the late Saxon defences ceased to be maintained and wall 15 robbed. The ditch and bank referred to in an Anglo-Saxon charter of 1045 relating to Ditchampton (Grundy 1919, 76, 290) could well be those at St John’s Hospital, the approximate line of which subsequently came to mark the parish boundary between Wilton and Burcombe. However, it seems that ditch 24/42 had probably been infilled by the 13th century as it produced no medieval pottery, whereas pottery of 13th — 14th century date was recovered from ditch 4 which perhaps replaced it a short distance to the north-west. Some pottery of this date was also recovered from the late Saxon bank, perhaps introduced when wall 15 was robbed or from a possible raising in height of the bank when ditch 4 was dug.

Ditch 4 may have had its origins in the 12th century, possibly dug when St John’s Hospital was established, and subsequently maintained as a drainage ditch or culvert serving the Hospital. No further details of this ditch or any structures relating to the Hospital were revealed during the excavation and watching brief, and the only other medieval feature was a single, small pit. This suggests that the area, set back from the street frontage, remained as largely undeveloped ground probably until the 18th or 19th century. The small quantity of charred plant remains from medieval pit 48 provide some support for this suggestion in that they included field weed seeds and seeds of wood margins or scrub.

The paucity of medieval features at St John’s Hospital contrasts with the density of features recorded at South Street where several ditches and pits, waterlogged organic material and a relatively large quantity of medieval pottery were present, attesting to domestic occupation on the street frontage in the 12th - 13th century. No structural remains were identified, but the ditches may have defined property boundaries as well as having been dug for drainage.

There was a marked fall-off in the evidence for settlement at South Street from the 14th century, and this in part is likely to reflect a general decline in Wilton’s fortunes which can be attributed to the growth in importance of Salisbury and the effects of the Black Death. The lack of later medieval features and finds has been noted on other recent, small-scale evaluation work at Russell Street (OAU 1994), Kingsbury Square (Michael Heaton, pers. comm.), and at King Street just to the north of Cross Bridge

EXCAVATIONS IN WILTON, 1995-6

in an area of medieval suburban development (AC Archaeology 1992). It is clear, however, that occupation did not cease, and at least one substantial structure was built on the site at South Street, perhaps during the mid 16th century. It is just conceivable that this structure, possibly a barn, was that built by one of the burgesses of Wilton near to the site of St Michael’s Church, which is referred to in the First Pembroke Survey commissioned in 1563, taking some ten years to complete (James 1962, 29). Other structural remains at South Street have been assigned to the 17th — 18th centuries, but the general character of occupation does appear to have altered, reflecting Wilton’s decline from a thriving urban centre to a relatively unimportant, small market town.

Both the recent excavations reported on here, and earlier work, have provided tantalising glimpses of the wealth of archaeological information that lies beneath Wilton. It is hoped that all future opportunities to undertake excavations will be grasped, particularly those which might shed light on what was happening in the town between the 8th and 13th centuries. It is sad to report that our knowledge of Wilton during this period has advanced very little in the quarter of a century that has elapsed since Haslam’s assessment (Haslam 1976).

Acknowledgements

The fieldwork and post-excavation work at St John’s Hospital were commissioned by The Sarum Partnership on behalf of the Trustees of the College of Matrons and the Trustees of St John’s Hospital. Mr John Overton and Mr J Ashenden of The Sarum Partnership are particularly thanked for their co- operation.

The evaluation was managed by Caron Newman and directed by Andrew Crockett, assisted by Damien Boden and Mel Gauden. The excavation was managed by Antony Firth and directed by Andrew Hutcheson assisted by Dominic Barker, Bob Davis, Joe Whelan and Richard May, with the watching brief undertaken by Ronald Ross assisted by Jane Liddle.

The fieldwork at the new Doctors’ Surgery, South

-Street, was commissioned by Dr A J Hall; and Dr Hall, Mr V Hall and Miss L R Oliver are thanked for their co-operation and assistance. The evaluation and excavation were managed by Roland Smith and Mick Rawlings respectively, and directed by Rachael Seager Smith and Andrew Hutcheson with the assistance of Joe Whelan, Simon Carlyle-Lancaster, Neil Walmsley, John Hart, Rosie Edmunds, Chris Hurn, Andrew Crockett and Dave Murdie.

203

The briefs for both projects were prepared by Helena Cave-Penny and Roy Canham of the Wiltshire County Council (Library and Museums Service), and their advice during monitoring of the projects is fully acknowledged.

The post-excavation programme was managed by Charlotte Matthews. Phil Andrews would like to record his thanks to the site directors who undertook the initial post-excavation work and discussed various aspects of the sites which have been taken account of in this report. The finds work has been managed by Lorraine Mepham, and the environmental work managed by Michael J Allen with environmental processing by Hayley Clark.

David Algar kindly made available some previously unpublished information relating to the 1970s evaluation trenches at St John’s Hospital, details of which have been included in this report. Finally, the author would like to thank Nick Griffiths, Honorary Curator of Wilton Museum, for providing him with the benefit of his local knowledge as well as a conducted tour of the town.

Bibliography

AC ARCHAEOLOGY, 1992, ‘Archaeological assessment: Landmark House, King Street, Wilton, Wiltshire’, unpublished client report, AC Archaeology

ADDYMAN, P.V. and HILL, D.H., 1969, ‘Saxon Southampton: a review of the evidence. Part II: Industry, trade and everyday life’, Proceedings of the Hampshire Field Club and Archaeological Society 26, 61-96

BIDDLE, M. and COLLIS, J., 1978, ‘A new type of 9th and 10th-century pottery from Winchester’, Medieval Archaeology 22, 133-5

BIDDLE, M. and HILL, D., 1971, ‘Late Saxon Planned Towns’, Antiquaries Journal 51, 71-85

BLAIR, J., 1994, Anglo-Saxon Oxfordshire. Gloucester: Alan Sutton

CLEAL, R.M.J., n.d. “The ceramic building material’, in J.W. Hawkes, ‘Excavations in Salisbury 1984-1990’, unpublished report, Wessex Archaeology

CURRIE, C.K., 1993, ‘Excavations at the Postern Mill site, Malmesbury, 1986-87, WAM 86, 58-74

DRIESCH A. VON DEN, 1976, A guide to the measurement of animal bones from archaeological sites. Harvard: Peabody Museum Bulletin 1

DRIESCH A. VON DEN and BOESSNECK J., 1974, Kritische Anmerkungen zur Widerristhohenberechnung aus Langenmafen vor- und frtihgeschichtlicher Tierknochen. Saugetierkundliche Mitteilungen 22, Munchen, 325-348.

GRUNDY, G.B., 1919, ‘The Saxon Land Charters of Wiltshire’, Archaeological Journal 76, 143-301

204 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

HARE, J.N., 1991, “The growth of the roof tile industry in later medieval Wessex’, Medieval Archaeology 35, 86- 103

HASLAM, J., 1976, Wiltshire Towns: The Archaeological Potential. Devizes: Wiltshire Archaeological and Natural History Society

HASSALL, J.M. and HILL, D., 1970, ‘Pont de l’Arche: Frankish Influence on the West Saxon Burh?’, Archaeological Journal 127, 188-195

HILL, D., 1969, “The Burghal Hidage: The Establishment of a Text’, Medieval Archaeology 13, 84-92

JAMES, M.K., 1962, “The Borough of Wilton’, in E. Crittall (ed.), A History of Wiltshire, VCH, Vol. VI, 1-34

JOPE, E.M., 1972, ‘Pre-conquest pottery’, in C.A. Ralegh Radford, ‘Excavations in Cricklade, 1948-1963’, WAM 67, 90-3

LOADER, E., 2000, ‘Ceramic building materials’, in M. Rawlings, ‘Excavations at Ivy Street and Brown Street, Salisbury, 1994’, WAM 93, 37-8 [this volume]

MACGREGOR, A., 1985, Bone, Antler, Ivory and Horn. London: Croom Helm

MATTHEWS, C.A., forthcoming, ‘The Late Saxon Pottery’, in K. Qualmann (ed.), Winchester Excavations since 1972, Vol 1: Western Suburb

MELLOR, M., 1994, ‘A synthesis of middle and late Saxon, Medieval and early Post-medieval pottery in the Oxford region’, Oxoniensia 59, 17-217

MEPHAM, L.N., 1993, ‘Pottery’, in A.H. Graham and S.M. Davies, Excavations in Trowbridge, Wiltshire 1977 and 1986-1988, Wessex Archaeological Report 2, 101-14

MEPHAM, L., 2000, ‘Pottery’, in M. Rawlings, “Excavations at Ivy Street and Brown Street, Salisbury, 1994’, WAM 93, 29-37

MEPHAM, L., forthcoming, ‘The pottery’, in P. Williams and R. Newman, Excavations at Grove Farm, Market Lavington, Wiltshire, 1986-1990, The Development of a Roman, Saxon and Medieval Settlement, Wessex Archaeological Report

MEPHAM, L.N. and UNDERWOOD, C, n.d., ‘The pottery’, in J.W. Hawkes, ‘Excavations in Salisbury 1984- 1990’, unpublished report, Wessex Archaeology

MORRIS, E.L., 1994, The Analysis of Pottery. Salisbury: Wessex Archaeology Guideline No. 4

MPRG 1998, A Guide to the Classification of Medieval

Pottery Forms, Medieval Pottery Research Group Occasional Paper 1

MUSTY, J., ALGAR, D.J. and EWENCE, P.F., 1969, ‘The medieval pottery kilns at Laverstock, near Salisbury, Wiltshire’, Archaeologia 102, 83-150

OAKLEY, G.E., 1979, “The worked bone’, in J.H. Williams, St Peter’s Street, Northampton. Excavations 1973-1976, Northampton Archaeological Monograph 2, 308-18

OAU 1994 ‘13A Russell Street, Wilton, Wiltshire: Archaeological evaluation’, unpublished client report, Oxford Archaeological Unit

ORDNANCE SURVEY 1976, Geological Survey of Great Britain (England and Wales) — Salisbury, Sheet 298

PEACOCK, D.P.S., 1979, ‘Petrography of fabrics A-H’, in Rahtz 1979, 310-4

PLATT, C. and COLEMAN-SMITH, R., 1975, Excavations in Medieval Southampton Volume 2: The Finds

RAHTZ, P., 1979, The Saxon and Medieval Palaces at Cheddar. Oxford: British Archaeological Reports 65

SMITH, R.W., 1997, Excavations at Emwell Street, Warminster: the Early Economy and Environment of a Wiltshire Market Town. Salisbury: Wessex Archaeology

WAM, 19064, ‘Excavation and Fieldwork in Wiltshire 1963’, WAM 59, 185-190

WAM, 1971, ‘Excavation and Fieldwork in Wiltshire 1970’, WAM 66, 188-191

WAM, 1972, ‘Wiltshire Archaeological Register for 1971’, WAM 67, 167-178

WESSEX ARCHAEOLOGY 1993, ‘St. John’s Hospital, Wilton, Wiltshire: Archaeological evaluation’, unpublished client report, Wessex Archaeology

WESSEX ARCHAEOLOGY 1995, ‘New Doctors’ Surgery, South Street, Wilton, Wiltshire: Archaeological evaluation’, unpublished client report, Wessex Archaeology

WESSEX ARCHAEOLOGY 1996a, ‘St. John’s Hospital, Wilton, Wiltshire: Archaeological excavation’, unpublished client report, Wessex Archaeology

WESSEX ARCHAEOLOGY 1996b, ‘St. John’s Hospital, Wilton, Wiltshire: Archaeological Watching Brief’, unpublished client report, Wessex Archaeology

WEST, S., 1985, West Stow, The Anglo-Saxon Village. Gressenhall: East Anglian Archaeology 24

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 205-217

Names on the Path to Remembrance: the building of Marlborough College Memorial Hall

by Brian Edwards

Marlborough College suffered a_ tragically high loss of Old Marlburians serving in the First World War, and decided to commemorate their sacrifice by building the present Memorial Hall. This paper examines the controversial decision to build the hall in the context of the public mood and opinion prevailing during and after the war, and assesses the shortcomings of the resulting structure as a memorial.

Have you forgotten yet?... For the world’s events have rumbled on since those gagged days, ... Siegfried Sassoon , ‘Aftermath’, 1919.!

The interior walls of Marlborough College’s Memorial Hall carry the names of the Old Marlburians who died during the Great War, but as another millennium starts and leaves behind a century shaped by this conflict, the names are now obscured by seating.’

That these names would be covered for even a single moment, let alone all but a few days of the year, is something that would cause tremendous resentment among the survivors of 1914-18,’ and mortify those who at the time of the 1945 armistice devised the term “The First World War’, ‘to prevent the millenian [sic] folk from forgetting’.* That nearly every town and village has a war memorial on which have been added the names of those killed on active service during the Second World War, is a fact that is increasingly overlooked by the vast majority of people on all but one or perhaps two days a year. Despite public demand having recently seen the return of ‘the silence’ being observed at 11.00 a.m. on the eleventh day of the eleventh month, the significance of this -time and date is lost on those of the post-war generations who have taken no active participation, other than observing that the dead of two world wars are commemorated each year on the Sunday nearest to the 11th November.’

Between 1919 and 1939, Armistice Day was the most important national day on the calendar. For

those growing up in the inter-war years, this was the day when the world stopped. ‘There was not the amount of road traffic then that there is today, but what traffic there was stopped at 11.00 a.m.’.° On Remembrance Sunday in the 1990s, the traffic passing the Marlborough College Memorial Hall on the Bath Road and nearly everywhere is likely to keep moving. Inside the Hall this is the only day when the memorial reverts to its primary function, as the Memorial Hall built in honour of the war dead has within the lifetime of their surviving comrades, been transformed from a ‘speech hall’ to what is principally a hall for music and theatre.

That the Memorial Hall is now being used for purposes other than originally intended could be taken as a sign of the times, reflecting today’s mood in comparison to that of yesterday. That however would be an unjust reflection on the present incumbents, as it is more realistic to recognise that their heritage in the form of the Hall was not in keeping with the overriding spirit of the time in which it was raised. When it was suggested and built immediately following the Great War, the choice of a functional memorial was in sharp contrast to the consensus of opinion. The more recent treaty of 1945 saw a very practical attitude to memorial as a nation rebuilding was inclined to ‘make do and mend’; but in 1918 the greater need was for a memorial idiom unencumbered by a requirement for utility.’ Discussion concerning the form that Great War memorials should take had begun on a broad scale as early as 1915, and it was stressed from the outset that uniformity was as important from the view of recognition as it was to

Mount Pleasant, The Cartway, Wedhampton, Devizes, Wilts SN10 3QD

206 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

protect public buildings, churches and the landscape from unsuitable additions or alterations.* An agreed form took shape at the end of the war, and despite a counter-case for memorials with a practical use, the broad consensus fell behind something simple of single and poetical importance but otherwise impractical in purpose.”

The appalling toll in the casualty reports of 1915 indicated that death rates would be highest among young officers, the ranks that consisted mostly of Oxbridge and old boy volunteers.!° The Marlborough College newspaper, The Marlburian, made depressing reading throughout the 1914—18 period; however the casualty rates were exceptionally appalling throughout 1915 even by Great War standards.'! The Marlburian of 11th February 1915 carried the most incredibly moving headings: ‘Killed In Action. .. Died Of Wounds... Missing Believed Killed... Wounded. . . Missing... Prisoners of War’. Each individual section carried an extensive lists of names. The number of former pupils of Marlborough College who had been killed in a little over a year had already rapidly climbed above 300 in total, and this would be a major factor in the resignation of the Master of Marlborough College. Wynne Wilson had been headmaster at Marlborough for only five years when in 1916, rather suddenly it seems, he decided to leave. A decisive factor could well have been the death of the war poet Charles Hamilton Sorley at Loos in October 1915, which must have been a particularly shattering blow to the Master.'? Wynne was especially friendly with Sorley, and they regularly exchanged letters. In fact Sorley’s last letter to the Master was dated 5th October 1915, only eight days before his death, perhaps arriving at Marlborough along with notification of his loss.'*? Following Sorley’s death, Wynne it appears simply could not take the ultimate cost of the fresh faces he had so recently taught, as page after page of The Times and The Marlburian bore increasingly greater numbers of past pupils.'*

The mounting casualties that finally saw the resignation of Wynne Williams perhaps also prompted the first discussions regarding a memorial. With the sheer number of Marlburian dead resulting from the heavy losses in the junior officer class — which consisted mostly of former public schoolboys — it is perhaps not surprising that someone associated with Marlborough College would be among the first to organise and start the process of considering a memorial. History has not revealed who that person was, but it can safely be assumed that he came from within the ranks of the Old Marlburians, and quite possibly from one of them serving at the Front.

The Old Marlburian association of old boys had perhaps already started considering memorials when their Committee received a letter from a group of Old Marlburian officers serving on the Western Front. The letter revealed that earlier that year, in January 1917, a dinner had been arranged immediately behind the front lines, which was attended by some 70 Marlburian officers. Those assembled collected around £ 170 and had sent it to the Old Marlburian Committee stating that, ‘it was their contribution to any fund to be formed, and they hoped it would be devoted to the education of the sons of the fallen’.'” The contents of this letter galvanised the Old Marlburians into action and a public meeting was arranged on 20th April 1917, specifically to launch the memorial project.

The meeting was held at the Surveyors’ Institute, Great George Street, Westminster, and around 150 Old Marlburians, parents and friends of College were present.'® Following a notable lack of any opening remarks whatsoever, the Chairman “The Right Hon. R. E. Prothero, M.V.O., MP., came straight to the point, which was that out of 3,000 Marlburians serving in the forces, a number in excess of 450 had already been killed and ‘many of those men were married and have left behind them young families’.'” The meeting concerned itself immediately with those left behind, rather than commemorating those who had served and died. The meeting was of course guided by the speakers from the platform, whose stance clearly reflected the exact line suggested by the letter from the Old Marlburians serving at the Front.

The letter from the serving Old Marlburians was read to the meeting, and coming as it had from the front line of the conflict, their message encouraged the meeting to pledge unanimously to support the children of Old Marlburian casualties in their education. Taken as it was before the total number of losses was known, and therefore numbers of children and the cost of any such provision could be calculated, this decision was perhaps an emotional reaction. This undertaking was no small assurance on the part of the Old Marlburian Committee, who as part of the project were offering their services to organise raising funds and running proceedings.

Initial funds for a memorial would undoubtedly be forthcoming, but the educational issue was a cause that would be difficult to gain and collect appeal funds for on an ongoing basis. The meeting was reminded by one of the few speakers, General Sir Edmund Barrow, that, ‘ Marlburians do not as a rule come from an affluent class, for we are mostly the sons of

NAMES ON THE PATH TO REMEMBRANCE: MARLBOROUGH COLLEGE MEMORIAL HALL 207

Figure 1. Memorial Hall. Drawing by E. M. Holman

professional men with limited incomes’.'* Barrow was subtly pointing out that capital funds for a memorial would be difficult enough to raise from among their class at a time when money would be short, and to make any additional commitment that could continue for up to two decades into the future might be an emotional reaction and well beyond their means. This latter prophesy proved accurate, because in reality they were considering providing for what could theoretically total a thousand or more children ranging from a 21-year-old then approaching his finals at Oxbridge to a babe in arms or even a foetus, and everyone in between.!° In view of the truth behind this statement the optimism in their ability to raise such funds was perhaps more emotionally than economically based. The meeting, almost as a -secondary matter, wholeheartedly supported a proposal that a cloister be built as a memorial attached to the College Chapel, and it was with this satisfactory agreement that the Old Marlburian Committee which had now also become the Marlborough College War Memorial Committee, adjourned.” Fully committed to the cause, the Old Marlburians set about raising funds and understandably approached the Marlborough College Council, the

school’s governing body, for assistance with regard to educating the dependants of the fallen. It appears however, that the College Council were not so emotionally committed to the cause as the Old Marlburians. The Council took until February 1919 to reply formally, which was nearly two years after the War Memorial Committee had approached them for assistance. The delay meant that dependants of school age in the meanwhile, whether at Marlborough College or elsewhere, were left in limbo during the Council’s lengthy deliberation. No consideration appears to have been given to these children by the Council, and the delay in formally replying suggests that, before committing themselves, the Council were waiting until the war was over and the likely number of dependants could be ascertained. At the end of 1918 it was adjudged that the total of Marlburian dead was in excess of 725, with a number of missing and seriously wounded which would contribute to the final figure. With this approximate total, a decision on the cost of the proposal regarding the dependants’ education could now be made, but this suggests the Council operated from a purely economic stance. The feeling that the College Council made a purely economically calculated decision with regard

—

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

208

— _

i

KIN |

Ks x

MIN

I

a

3

re

to Ta NS x

St =

COLL

|

|

is

|

OROL

B

LARI

\

Brae om Ae CS

ted Lo mang 180 +B |

Figure 2. Plan and Section of the Memorial Hall, published in The Marlburian, 20th December 1920

NAMES ON THE PATH TO REMEMBRANCE: MARLBOROUGH COLLEGE MEMORIAL HALL 209

to offering free places to dependants of the fallen is difficult to escape. With the responsibility they had to ensure the sound footing of the school, it is perhaps unfair to suggest that they were merely cold and calculating, but the collective evidence indicates a certain detachment from sombre obligation and sincere sentiment. By itself the decision to wait until the number of dependants was known can be considered sensible, and by stipulating in their offer to the War Memorial Committee that places would only be offered to the ‘sons’ of the fallen, and then only ‘on the grounds of need’, it could be considered that the Council were merely minimising any possible contribution on their part.?! Should there be any doubts as to the motive of the offer, however, the overriding condition placed on the offer of free places was that it was made subject to the College War Memorial Committee agreeing to erect a ‘suitable building or buildings’ for the school’s use. Free places at Marlborough in lieu of a new building was a good deal for the Council. The school’s popularity meant that demand for places far outstripped those available, and an assembly hall and other buildings were much needed. If not built by donation, the school would have to find the money themselves. Coming as they would in occasional flurries, the inclusion of a few free places at any one time would hardly be noticeable in comparison to raising the necessary capital for a hall, and it would save the College an enormous outlay at a time when the Council were being pressed to provide many new buildings as additions to the school.

There were undoubtedly those who thought that the money raised for any memorial should be put to a practical use in the College. The dining hall was apparently a disgrace, the assorted libraries throughout the school appeared in need of co- ordinating in one building, swimming baths were quite desired and a dedicated laboratory was another requirement. In addition there was nowhere suitable within the College, such as a hall, where the whole school could be assembled and addressed. All these requirements had come under consideration by the War Memorial Committee, but were rejected on the grounds of incompatibility with “a memorial generally, and a Great War memorial specifically. The inaugural memorial meeting had also recognised that the cost of such a venture could prove prohibitive, perhaps with the cost of war-time materials in mind, but particularly with regard to the limitations of their middle-class purses.** Despite this the odd protest was made in support of a hall, and in December 1917 The Marlburian started to publish letters of support for the proposed

cloister and counter letters calling for a hall, that continued to fuel a debate throughout 1918.

In February 1918 the President of the Marlburian Club, in answer to concerns of aesthetic nuances over an addition of a cloister to the College Chapel, revealed in The Marlburian the discovery of plans for such a cloister. These had been drawn up by the College architect, who had envisaged when he had designed the College Chapel that such an addition was desirable, and would be required when funds allowed. With this encouraging affirmation and the endorsement of the current College Architect, ©. E. Ponting, who was also as Diocesan Architect an experienced ecclesiastical designer, the cloister would have appeared to have seen off its detractors. The College Council, however, had seen an opportunity for saving the school the cost of a much needed hall and their offer of free places was in fact a counter- proposal which was in itself a fait accompli.”*

With many worried dependants waiting two years for the outcome, free places were finally offered to the Memorial Committee by the Council, but only in exchange for a building or buildings ‘to serve the school’. The ‘Cloistered Garth’ which was placed before the Council in November 1917 as the democratic and unanimous choice of the inaugural meeting, the expressed wishes of some of those still serving and the specific item under which donations had been sought and given — was entirely and obviously compromised.”? The Cloistered Garth was not a building that would serve the school as such so was therefore automatically ruled out if the Council’s offer was to be accepted. The situation was put before a second War Memorial public meeting, and the decision to build a hall was forced through amidst unhappy protests from those wishing the meeting to abide by the original decision to build a Cloistered Garth.”

A hall having been decided upon, the sensitivity in approaches to memorial design that was being observed by the prevailing populace, was completely ignored by the College. The Civic Arts Association had offered conferences and exhibitions on the design for war memorials which were widely promoted in the press, and the results of their competition to find simple and dignified designs were published in Country Life and other magazines.*° The Royal Academy was opened to official bodies for consultation on these matters, and they established a committee of leading artists, architects and sculptors, in order that the tasteless and unsuitable could be avoided.*’ This guidance was neglected by Old Marlburians in favour of the rather undignified self- promotion of a competition among themselves.**’ The

210 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

strict requirement of the competition being to produce a design for ‘a Hall in the Amphitheatre style’ suggested a groundbreaking exercise as this was a recent foreign importation in the architectural field.*” To stray from the indigenous and well established could be considered unusual for a memorial, the risk of creating offence being higher than if sticking to known areas. This makes it appear as if the competition was merely an opportunity for an Old Marlburian to make a name for himself.

In fact the competition was further limited not just to Old Marlburians, but to a certain section of them. Since all those serving would be unable to enter the competition, those who had returned home wounded would probably either not want to enter or be incapable of doing so, and those on leave would not have time; the likelihood is that in reality this competition was limited to those who had not personally experienced the awfulness of contemporary warfare.*” They would however, have been schooled, as all Marlburians were, in muscular Christianity and chivalry reinforced with heroic tales of valour.’! A reflection of one attitude current among this group was voiced by those who, after the full horrors of 1914-18 had been revealed, announced that they considered the Great War an ‘opportunity’.*” It was this attitude that ensured that the competition for a memorial design was exclusively limited to them, to produce a monument to war in the guise of an honourable shrine.*’

Although no protest regarding the design of the Hall appeared in the columns of The Marlburian or any local or national newspapers, the fact that a wave of propaganda appeared attempting to justify the size of the Hall and the detail it employed is rather intriguing. *' As it is apparent from the text that this is no self-congratulatory exercise, it is perhaps therefore a response to some unrecorded criticism.*? Certainly in comparison to the largely fine but plain brick buildings that the College mostly consists of, the Hall appears grandiose considering its primary purpose. Nikolaus Pevsner described it in these terms: ‘It is of brick with ample stone dressings. The front has two closed end bays and then, stretching between them, a long loggia of eight Adamish columns. The top is a parapet. The back of the building is semi-circular. It comes as near to the American Campus style of the same years as anything this side of the Atlantic’.*° ‘Terminology such as ‘ample dressings’, and ‘Adamish columns’ cannot be misinterpreted. Nor can the ‘semi- circular’ rear be regarded as less than flamboyant for a memorial. Compared to anything else in the United Kingdom, it is certainly not typical of the time in which it appeared.*’

Marlborough College was not entirely alone in constructing a practical building as a war memorial. Public Schools were unsurprisingly the main instigators of halls, libraries, sanatoria, pavilions and playing fields, most of which are simple affairs.** A minority of communities also favoured halls as a memorial to the dead of 1914-18, to perpetuate the camaraderie and levelling companionship of the war years, and to be used by the those who came back as well as the families of those who did not.*? However, it is rare to encounter among Great War memorials anything so splendid and generous in its architecture as the Marlborough College Memorial Hall.!° The Great War memorials found in churchyards, high streets and at cross-roads nationwide, are remarkable in their simplicity, as are those that pervade the Somme.*! Simplicity in design, if not singularity in purpose, was expected even by those calling for practical buildings as memorials.’ This is a sentiment to which the Hall clearly does not comply.

The Marlborough College Memorial Hall is an example of what Professor Adshead described in 1916 as, ‘arrogant architecture that was a war cry rather than a message of peace’. The dimension of overstated glory and pompous architecture employed is in outstanding contrast, not only to the simplicity of the Cenotaph which provides a sense of non- denominational timelessness and _ universal understanding, but also to the simply stated and dignified memorials to be found in every town and village." The Marlborough College Memorial Hall even stands out in comparison with the memorials employed by other public schools and wealthy institutions.’ The chapels at Charterhouse, Beaumont and Downside are all humble affairs, Edinburgh Academy and Campbell College, Belfast, too, are comparatively simple despite also employing old boys as designers. Some public and grammar schools such as Nottingham High School, and Fettes School, Edinburgh, did resort to sculptures of officers, but portrayed as they are in the act of falling or encouraging their men onward, these are used to front simple memorial buildings. Playing fields and pavilions were aptly chosen as memorials at some schools, and of those who chose libraries, museums, classrooms, wall friezes and memorial gates, simplicity was the common ingredient.’° Even Glasgow Academy which went to the extraordinary lengths of endowing the school as a memorial, has simple carved oak panels lining the school gallery to carry the names.

That simplicity in commemoration was the most widely agreed notion following the Great War is evident by the examples that so characterise the

NAMES ON THE PATH TO REMEMBRANCE: MARLBOROUGH COLLEGE MEMORIAL HALL 211

landscape of Remembrance. That it was not a principle strictly adhered to by Marlborough College suggests how out of touch with the outside world the powers driving the school memorial actually were. Able to view the whole subject of memorials more objectively was Cyril Norwood, who had been Headmaster of Bristol Grammar School before accepting the Mastership of Marlborough College. Only offered the post upon the refusal of Old Marlburian Cyril Alington, who went on instead to be Provost of Eton, Norwood was made to feel an outsider from the outset.’” On entering Marlborough College Norwood was booed and jeered by the many boys who resented the appointment of someone who was merely a former grammar school headmaster.** Yet Norwood was undoubtedly what Marlborough College needed, and was the only voice from within the school that spoke out against the Hall. Norwood revealed at his very first address to Marlburians that as a headmaster he would like to see additional buildings provided, but as ‘a memorial to the dead that would be unthinkable’.*? However, he made it clear that he had no wish to be directly involved in decisions regarding a memorial, because he felt that the choice should lie with the parents and Old Marlburian friends of the fallen.*® At the second and final public memorial meeting, Norwood continued to make clear his alienation from the project by stating that he was only acting as a, ‘sort of clearing house for opinions’. He did however also make clear that the call for a hall was led by those still at Marlborough, and therefore by implication those likely to benefit directly.2! Norwood obviously was not happy with how things had proceeded, but having made it clear from the outset that he was distancing himself from the project, he could not then be seen to interfere. Despite his original stance against a practical building as a memorial, Norwood attempted to make the best of it by hinting that his least objection would be against a ‘speech hall’. This at least would be dignified, but Norwood was being optimistic if he thought the building would be used purely for this purpose. The prospect of a design competition likely to produce architecture with unsuitable embellishment was clearly not, in his view, a fitting - way to proceed, and he tried subtly to deflect it by stating at the memorial meeting that, ‘you might throw it open to a competition amongst Marlburian architects; or you might decide on a different kind of hall altogether’. Despite Marlborough’s rapid progress under Norwood’s enlightened approaches to education, the College failed to listen to his concerns regarding the memorial.’? The Council and Old Marlburians carried on regardless, and Norwood left

Marlborough at the first opportunity, which arose, co-incidentally, shortly after the inauguration of the Memorial Hall.

Despite being out of step with the vast majority of grieving bodies, the unnecessary and inappropriate embellishment employed in the design of Marlborough College’s proposed memorial was not to be sacrificed when a financial crisis forced economies; instead it was a matter of reduction in size. Following the announcement of the winner in 1921, Old Marlburian William Newton, whose design was chosen by the President of the Royal Institute of British Architects, unveiled his plans for the Memorial Gardens that we see today accompanied by what was described in Hinde’s history of the school as, ‘a magnificent pilloried hall to seat 1,500’.?* No consideration to cost had been stipulated by the competition organisers and none was taken by either the competition’s judge or winner. Overlooked was the fact that the finances necessary to build such a building would have to be raised by donation in a period of serious economic circumstances and the threatening depression that inevitably followed war. The cost proved prohibitive, despite a year long wait for material prices to subside and for depression to eat into the labour costs. The overall cost fell by 25%, but even this massive cut was insufficient to bring the project within the range of the fund. Instead of examining what simplification could be applied to the design, the winning architect was asked to submit reduced plans. A rather pathetically pedantic debate ensued as to whether an average boy required 14 or 16 inches sitting space, an area that under-11s would find uncomfortable, let alone the seniors who would be expected to occupy these spaces for long periods. Despite the original claim of being able to seat 1,500, building began on the basis of a 1,150 seater hall, but in a realistic estimate excluding the additional 79 seats which obstruct the names of the fallen, the figure is closer to 600. The Memorial Hall could not seat the entire population of the school, even at the time it was built.

At the announcement of the competition, the school numbers had been rising consistently for over a decade. By the time the plans had to be re-drawn the number of seats required was well above 700, and at the time of the opening ceremony the newly opened hall fell far short of seating the entire school, which then totalled 722. The platform raised at the rear of the hall is a detail included for occasional standing overfills but which lends itself naturally to the placement of temporary seating. The placing of chairs on this ledge, therefore, is hardly surprising, since any

212 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

long ceremony or performance would require this additional seating. When in place however it is the backs of the chairs that hide more than half the names, and they are completely obstructed from view when the seats are occupied. With a little more foresight by the design team the tablets could have been placed higher up, or elsewhere, thereby ensuring that the names on the Roll of Honour would never be so ignominiously obscured. Quite clearly the Hall was never fit for its supposed purpose. Not only would it not house the school, but it fails in its primary function as a memorial to immortalise the names of the dead. This, of course, was widely regarded as of foremost importance, and it was particularly relevant at a school that had a greater number of war dead than any similar institution."

As a totem, individual Rolls of Honour inevitably reflect the vast numbers of those who enlisted in

Figure 3. Partly concealed memorial to C. H. Sorley

county regiments and Pals battalions, men from the same area who would have grown up together, went to school together, joined up together and died together. This cannot be said for former pupils of public schools. Sent away from home to school, they had no other such immediate connection when joining up, and as young officers had no associations similar to the Pals in army life or indeed in death. The memorials which sprouted in every village and town, to express the collective loss and binding together of local families in public commemoration, were not also the communal centre for grief for the families and school-friends of the public schools.” The names of public schoolboys could of course appear on a memorial in their home area, or perhaps at a club, but the only places where they could be commemorated with the majority of friends they had made in their short lives was at their school where they had formed most of their allegiances and associations.*° Appearing on a Roll of Honour therefore allowed individuals to retain their association in death as they had done in life. It also allowed organisations and institutions to illustrate their role in the sacrifice.

The Roll of Honour had developed from the Roll of Service, a record of the rush to join up in 1914— 15.°’ This Roll of Service was then used to encourage those whose names did not appear, so they would be shamed and respond by joining up. Absence from this list of names implied cowardice and encouraged ‘white feathering’, whereas individually and collectively the names on the list implied status. This provided any organisation that produced such a list the opportunity to exploit the situation and make a statement regarding what they were doing for the war effort. At the end of the war when this Roll no longer represented those who had gone to war, but those who were not coming back, the Roll of Honour on a factory wall no longer proclaimed that the firm was doing its bit, but had played its part.*® Inevitably this raises a question whether the original impetus behind the Roll of Honour for some organisations was to bring honour on themselves, as much as it was to honour their servicemen either living or dead.

The history of Marlborough College’s Memorial Hall makes it hard to come to any conclusion other than that it was built to serve and honour the school more than it was to honour their dead. In an article on memorials in the Cornhill Magazine in 1916, the Master of Magdalene College, Cambridge, A. C. Benson, urged those who sanctioned memorials to, ‘fight shy of elaborate design’, and to employ, ‘simplicity of statement’, with perhaps, ‘a touch of emblem but no more’. He further pleaded that

NAMES ON THE PATH TO REMEMBRANCE: MARLBOROUGH COLLEGE MEMORIAL HALL 23

Figure 4. Marlborough College Memorial Hall and Chapel

memorials should not be built in a style to ‘astonish tourists’ and ‘feed our vanity’. What the passer-by sees in the grounds of Marlborough College confirms Benson’s worst fears, and its history reveals his foresight. The Master of Magdalene had further warned about guarding against allegorical pantomime.*’ Those in control of Marlborough’s memorial had chosen to ignore his warning, and with hindsight might be seen to have been embarking on something of a pantomime of their own.

In April 1917, a cloister had been proposed as a memorial and funds raised from donations on this basis. By October 1919, the College Council had created an atmosphere that demanded the building of a hall -by offering to take over the education of sons of the Old Marlburian Great War dead, and amid protest a hall was agreed. The site was decided upon by the Council, and the design competition among old boys produced a plan published in December 1920. Amid rumours of unsuitability of the chosen site and concern over proposals for the foundations The Marlburian, on 24th May 1921, published a

“report on the proposed raft to take the structure. Then, as if building towards the climax of some ridiculous farce, the following day the Master reported to the College Council that the funds raised were far insufficient to cover the cost. Despite this lack of funds, the contractor started work two months later, and the designer was asked to submit reduced plans. It was at this point that Marshall Foch among other

celebrities refused without validated excuse to perform the opening ceremony. Then less than two months before the planned date of the ceremony the winning designer informed the Memorial Committee that the Hall was sinking.®°

A flawed design was compounded by alterations demanded by the overreaching nature of the project. Those responsible had little regard for the circumstances of the time, and clearly overestimated the willingness. of Old Marlburians and friends of the school to produce a budget to match the expectations of the design competition. Underpinning was required and the building was retrieved after further delay and not insubstantial cost. In 1916 the Royal Academy had issued a memorandum suggesting points of guidance which in terms of Marlborough College’s Memorial Hall can be revealed as a prophesy. It had stressed a need for professional design, and care in choice of site, advocating simplicity of design and attention to scale and proportion.°’ The Memorial Hall was badly planned and poorly initiated owing to the College’s reluctance from the beginning to consider involving outside agencies. Because of their failure either to seek advice or to heed the warnings regarding their over-ambitious plans, the War Memorial Committee and College Council of that same period presented successive generations with an unworthy and impractical memorial.

In contrast to the names of the fallen, which were consigned to obscurity lining the walls of the

214 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Memorial Hall, the iconography of war remains bathed in sunlight. From its commanding position at the eastern end of the College Chapel, the reredos still radiates the glorious and heroic portrayal of death in battle which encouraged a generation to war in 1914. Instead of toning down this imagery following two world wars, the salutation of battle-invoked death was embellished only five years after the second conflict by Sir Ninian Comper, who added highly imaginative and colourful wall paintings in 1951. Like the figures on the reredos, the paintings perpetuate the idolatry of war, by depicting young boys as angels clad in armour.

In 1933, Sassoon wrote of witnessing the ‘Prince of Darkness’ standing in front of the Cenotaph praying;

‘Make them forget , O Lord what this Memorial

Means; their discredited ideas revive;

Breed new belief that war is ...

Proof of the pride and power of being alive;

Lift up their hearts in large destructive lust;

And crown their heads with blind vindictive Peace.’

The Prince of Darkness to the Cenotaph

Bowed. As he walked away I heard him laugh.°?

In the case of the Marlborough College Memorial Hall, the Devil’s prayers were apparently answered from the outset.

By way of an epilogue, however, it must be noted that, because of the high running costs of the present hall, Marlborough College has plans for a new purpose-built building, which will take over all but the most solemn aspects of the Memorial Hall. It appears that the Hall will otherwise be unused, and so it is possible that the chairs hiding the names may be removed. Rather fittingly the Hall, empty and without practical purpose, will then at last perhaps become what it always should have been — a memorial to waste.

“The boast of heraldry, the pomp of power, And all that beauty, all that wealth e’er gave, Awaits alike the inevitable hour. The paths to glory lead but to the grave’ Thomas Gray’

Notes

1. ‘Aftermath’ verse in full:

Have you forgotten yet? ...

For the world’s events have rumbled on since those gagged days,

Like traffic checked awhile at the crossing of city ways:

And the haunted gap in your mind has filled with thoughts that flow,

Like clouds in the lit heaven of life; and you’re a man reprieved to go,

Taking your peaceful share of Time, with joy to spare.

But the past 1s just the same,- and War’s a bloody game...

Have you forgotten yet? ...

Look down, and swear by the slain of the War that you’ll never forget!”

Siegfried Sassoon, ‘Aftermath’, 1919.

NAMES ON THE PATH TO REMEMBRANCE: MARLBOROUGH COLLEGE MEMORIAL HALL 215

Figures 5 (left) and 6 (above). Wall paintin

2:

2s 1n Marlborough College Chapel, by Sir Ninian Comper

Pamela Colman, Marlborough In Old Photographs: A Second Selection, 1990, p. 133.

Jay Winter, Sites of Memory, Sites of Mourning: The Great War in European Cultural History, 1996. A.J.P.Taylor, English History 1914-1945, 1965, p. 2 footnote 1.

Charles Madge and Tom Harrison, Britain by Mass Observation, 1939, ‘Two Minute Storey’.

6. Oral testimony of Stan Philpot, Lockeridge, Marlborough,

Wilts. Born in 1911, Stan made a tape recording in the 1980s and a copy is in the possession of the author.

7. Geoff Dyer, The Missing of the Somme, 1994, p. 11. 8. Ibid. p. 11-12. 9. Ibid; Michael Heffernan, ‘For Ever England: The Western

10.

Front and the Politics of Remembrance in Britain’ in Ecumene 2: 3 July 1995, p. 299; Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, pp. 142-4. The Times, 2nd December 1918, p. 3: David Cannadine, ‘War and Death, Grief and Mourning in Modern Britain’, in Joachim Whaley (ed), Mirrors of Mortality: Studies in the Social History of Death, 1981, pp. 196-9; Jay Winter, ‘The Lost Generation of the First World War’, in Population Studies 31, 1977, p. 461 but also pp. 449- 4066. See also: Robert Wohl, The Generation of 1914, 1979; Peter Parker, The Old Lie: The Great War and the Public School Ethos, 1987.

1. Lyn Macdonald, 1915, 1993.

bo

13:

16.

M7

18. 19.

23.

. Jean Moorcroft Wilson, A Biography of Charles Hamilton

Sorley, 1985, pp. 201-2. See also Jean Moorcroft Wilson, The Collected Poems of Charles Hamilton Sorley, 1985; The Collected Letters of Charles Hamilton Sorley, 1990. Jean Moorcroft Wilson, The Collected Letters of Charles Hamilton Sorley, 1990, pp. 258-261.

Thomas Hinde, Paths of Progress : A History of Marlborough College, p. 133.

The Marlburian, 30 May 1917, p67; ‘First list of donations’ in The Marlburtan, 12 July 1917, p. 96. It is important to note that most parental fathers attending the meeting would not have been educated at Marlborough themselves, and therefore were not Old Marlburians. Despite the College having been in existence long enough for several generations of a family to have attended the school, the trend of a son following in the footsteps of his father was not at that time the rule it later became. Indeed it is not certain that Marlborough College and other public schools did not create this precedent by offering free places to the sons of the fallen. ‘Proceedings at Meeting of Marlburians with regard to Marlborough College War Memorial’, in The Marlburian, 30 May 1917, p. 64.

Ibid. pp. 64-70.

Marlborough College Council minutes 2nd February 1934, report that memorial funded students were now few in number and were expected to have left the school inside two years.

. The Marlburian, 30 May 1917, p. 70. . Barrow’s exact words to the inaugural meeting.

Proceedings at Meeting of Marlburians with regard to Marlborough College War memorial’, in The Marlburian, 30 May 1917, pp. 65-69.

The offer of free places was made on the understanding that a building woulda be built for the school’s use. The further consequences of accepting the Council’s offer meant that offers of free education for the dependants would extend only to “those in need”, and then only to the “sons” of Marlburians killed in the war. The letter from Old Marlburians at the front made no such stipulations, and the reports of the Memorial meetings

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

carry no such suggestion that it was intended that any child should be precluded on the basis of means or sex. That Marlborough College was a purely male establishment and was not a charity was obvious to everyone concerned, and it therefore could not offer places to dependant daughters or afford to give free places to those left well provided for. That the Council made these stipulations may therefore on the surface appear innocent, but with the overall evidence in mind it is not surprising that there may linger suspicions of an economic basis for their inclusion. The primary intentions of the Old Marlburians were perhaps only concerned with “sons” and “those in need” in an age which even more sanctioned sexism and patronising philanthropy than today. However, if that was the case the probability is that they would have been more specific in their vocabulary. Until the Council responded formally to the Old Marlburian request for places, the terminology used for those requiring educational assistance by the Old Marlburians, parents and friends who attended the memorial meetings was “dependants”. The term “sons” did not appear until the formal reply from the Council was published, but thereafter substituted for the word “dependants” which disappeared from the vocabulary associated with the Marlborough College Memorial. Nor was “need” or any similar meaning attached to any proposals, so these ideas although perhaps not conscious, fell with the adoption of the Hall. The economic demands placed on the Memorial by the Council put paid to any expectations on behalf of all dependants, for it was it was after all economy rather than memorial that was the driving force behind Council seeking the Hall. See ‘Proceedings at Meeting of Marlburians with regard to Marlborough College War memorial’, in The Marlburian, 30 May 1917, pp. 64-70.

College Council Minutes, November 1917, from the Marlborough College Archive. The Marlburian, 6 February 1918, pp. 5-6, Letters to the Editor; The Marlburian, 17 December 1918, p. 150, Letters to the Editor.

. The Marlburian, 6 November 1919. Report on meeting

27th October 1919. Three of only four speakers called after the Chairman, Sir Henry Wilson, and Seconder, The Very Rev. W.M. Furneaux, Dean of Winchester, had risen to propose the Hall, spoke in favour of maintaining the original decision to build a cloister. It was most forcibly pointed out by Sir E.C.K.Ollivant, that the money had been subscribed with this object in view. Mr ‘T.\W.Weeding whose three sons had been killed in France endorsed the sympathies of the previous speaker Lieut. S.A.P. Kitcat, and argued strongly for a cloister as it was “much more appropriate than a hall, besides being in the minds of a very large number of subscribers”. In addition he stated that he felt confident that the “boys that had fallen would like the Memorial to be associated with the Chapel”. It must have appeared that if speakers continued to rise to support the Cloister,

33.

34.

35.

30.

Silk 38.

39. 40.

the meeting would surely sympathise, the Chairman therefore intervened, suggesting the meeting was “as representative as they could get” and put the resolution that “the money be spent on a Memorial Hall”. Country Life, 5th February 1916, pp. 188-9; The Spectator, 5th February 1916 vol 116, pp. 183-4; The Architectural Review, 1916, vol 39, pp. xvii, 112; The Architectural Review, 1916, vol 40, pp. 35-40.

. Michael Heffernan, ‘For Ever England: The Western

Front and the Politics of Remembrance in Britain’ in Ecumene 2: 3 July 1995, p. 299; Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, pp. 142- 144.

. This particular competition was rather unlike the one held

by the Civic Arts Trust, which had imposed strict guidelines requiring simplicity in sympathetic design and ensured continuity of opinion by widely publicising illustrations of the results, all of which conformed to a distinct standard whereby the dead would gain proper recognition in a dignified way. It is apparent from his contribution to the Memorial Meeting (The Marlburian, 6th November 1919. Report on meeting 27th October 1919) that the Master of Marlborough College, Cyril Norwood, found the idea of this competition rather distasteful, and an undignified approach to the Memorial design.

. Nikolaus Pevsner, The Buildings of England: Wiltshire,

1963, pp. 341 (2nd edition 1975).

. Jay Winter, Sites of Memory, Sites of Mourning, 1995,

p. 204. See Brian Edwards, Cotton Wadding on the Path to War, 2000, Devizes Museum Library.

. J.R.Taylor, H.C.Brentnall and G.C. Turner, Revised

and Continued Edition of A.G. Bradley, A.C.Champney, and J.W.Baines, A History of Marlborough College, 1923, p. 325.

Gaynor Kavanagh, Museums and the First World War, 1994, esp chapter 12: ‘Memories, Memorials and Momentoes’. The Marlburian, 20th December 1920;The Marlburian, 23rd June 1921;The Marlburian, 19th June 1923; Country Life May 1921.

The absence of an element in society is not justified by a lack of written evidence in newspapers etc. Over- positive propaganda playing up a particular aspect does however reflect fears whether voiced or not. For arguments along the lines of this reverse logic see Steve Humphries and Pamela Gordon, Forbidden Britain: Our Secret Past 1900-1960, 1994.

Nikolaus Pevsner, The Buildings of England: Wiltshire, 1963, p. 341.

Ibid.

Colin McIntyre, Monuments of War, 1990, pp. 170- 200, esp 170-1, 173, 175-6, 184, 199-200.

Ibid. p. 166. Country Life, extract in The Marlburian 23rd June 1921, p. 71, sites Sheffield Memorial Hall as the only other example.

NAMES ON THE PATH TO REMEMBRANCE: MARLBOROUGH COLLEGE MEMORIAL HALL

41. 42. 43.

44. 45.

40.

47. 48.

49,

50. Sill D2.

53.

54.

Geoff Dyer, The Missing of the Somme, 1994, pp. 6-7. Ibid. p. 11.

Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, p. 143.

Colin McIntyre, Monuments of War, 1990, p..198. For ease of reference see Colin McIntyre, Monuments of War, 1990, esp pp. 170, 171, 175, 199, 200. See also: C.F.Kernot, British Public Schools War Memorials, 1927; Derek Boorman, At the Going Down of the Sun : British First World War Memorials, 1988.

Playing fields and pavilions: Christ College, Bradfield College, The City of London School, Hurstpoint College, Lorretto School; for details of schools see Paton’s List of Schools, 1898 -on. Note : there is also a playing field dedicated to an individual at Marlborough that was presented to the school by his mother. Other memorials: Aldenham School, Herts; Berkhampstead School, Herts; Felstead School, Essex; Harrow; Canterbury; Bedford Grammar : for details of schools see Paton’s List of Schools, 1898 -on.

Thomas Hinde, Paths of Progress, 1992, p. 134.

John Stallworthy, Louis MacNeice, 1995, p. 74; John Castello, Mask of Treachery, 1988, p. 72; Maurice Bowra, Memories 1989-1939, 1966, p. 165; Also see Bevis Hillier, Young Betjeman, 1988, p. 94.

The Marlburian, 6 November, 1919. p. 163. Norwood remained faithful to his own philosophy on the memorial, and avoided conflict with the College Council who had proved a problematic body to successive masters since the school first opened, and were a major influence in the curtailment of the careers of previous Masters. Having followed Willson into the post, Norwood was very aware of the effect that the Council could have on his own charted course, and would not wish to risk his steadily building reputation. He was ambitious and perhaps in the realisation that involvement with a school whose Council could turn a memorial into a farce was becoming a liability.

The Marlburian, 30 May 1917, p. 67.

The Marlburian, 6 November, 1919. p. 163.

Cyril Norwood, The English Tradition of Education, 1929.

Thomas Hinde, Paths of Progress, 1992, pp. 137-9. Note also : Newton was in fact an architect working for his father Ernest Newton RA, so it would at first be considered surprising that so flamboyant a building was proffered by him, but this was undoubtedly intended to be simple enough to satisfy modesty in a memorial but stylish enough to catch the judges eye. War Memorials aside, the American campus style was considered very innovative by architects at the time.

Marlborough College not only had a greater number of war dead than any similar institutions, but with 742 out of some 3000 recruits they lost a greater percentage of old boys serving than anyone else. It was regarded as of primary importance to include the names of the fallen on any memorial at Marlborough College. See

5D):

56.

Dik.

58.

Qi,

reports on meetings in The Marlburian, 30th May 1917;

The Marlburian, 6th November 1919. In particular the names were cited rather specifically by Sir John Butcher, K.C., M.P., at the meeting held 27th October 1919 (The Marlburian, 6th November 1919, p. 163), but the importance of “the names” was not exclusive to Marlborough College and was a general theme; see Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, p. 139; “But the names are what matter”. Colin McIntyre, Monuments of War, 1990, p. 171.

Geoff Dyer, The Missing of the Somme, 1994, pp. 121- 2; Joanna Bourke, Dismembering the Male, 1996, p. 227; Michael Heffernan, ‘For Ever England: The Western Front and the Politics of Remembrance in

Britain’ in Ecumene 2: 3 July 1995, pp. 294-6.

Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, p. 138; David Cannadine, ‘War and Death, Grief and Mourning in Modern Britain’, in Joachim Whaley (ed), Mirrors of Mortality: Studies in the Social History of Death, 1981, conclusion.

Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, p. 139.

Department stores such as Harrods erected a Roll of Honour. See Colin McIntyre, Monuments of War, 1990, p. 44.

59. A.C. Benson, ‘Lest We Forget: A World of War Memorials’,

60.

61.

63.

in Cornhill Magazine, September 1916, p. 299. We can only speculate as to the reason why a number of celebrities refused the “honour”. It is possible that some refusals could have been due to the manner in which a hall was implanted over the choice of a Cloister, since there certainly was strong feelings about this aspect. The refusals could also have been something to do with the chosen design of the hall, but this appears less likely. Bob Bushaway, ‘Name Upon Name: The Great War and Remembrance’, in Roy Porter (ed) Myths of the English, 1992, p. 144.

. Siegfried Sassoon, The Road to Ruin, 1933.

‘At the Cenotaph’ I saw the Prince of Darkness, with his Staff, Standing bare-headed by the Cenotaph; Unostentatious and respectful, there He stood, and offered up the following prayer.

‘Make them forget , O Lord what this Memorial Means; their discredited ideas revive;

Breed new belief that war is purgatorial Proof of the pride and power of being alive;

Men’s biologic urge to readjust The Map of Europe, Lord of Hosts, increase; Lift up their hearts in large destructive lust;

And crown their heads with blind vindictive Peace,’ The Prince of Darkness to the Cenotaph Bowed. As he walked away I heard him laugh. Thomas Gray (1716-1771), Elegy Written in a Country Churchyard.

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp.218—232

Investigation of a Roman villa site at Euridge Manor

Farm, Colerne by Larry Luckett'

with contributions by D F Mackreth, T S N Moorhead, Judith Roseaman

and Bryn Walters

This report details the results of amateur excavations undertaken during the 1950s on the site of a suspected Roman villa at Euridge Manor, Colerne. Necessarily constrained by the nature of those excavations and the length of time which has since elapsed, the report which follows 1s compiled from various interim accounts published in the Wiltshire Archaeological and Natural History Magazine, an unpublished letter, a description of the site as visited by Hugh Seymour and the writer, specialist reports on the coins and fibulae, and a description of such finds as have been placed in Devizes Museum. The results of a recent geophysical survey of the site are

also briefly mentioned.

The Site

Euridge Farm, in the parish of Colerne, and some 1.5 km from the village, is situated on a south-facing slope above a tributary of the By Brook. The name Euridge means ‘the ridge where Yew Trees grow’. It appears as Ewerigga in 1156, Iwerugge c.1250, and as Uridge on a map of 1826. Aubrey says ‘at Euridge....they (yew trees) also grow indifferently plentiful’ (Aubrey 1847, 55). The site is some 450 metres to the south of the farm buildings (Figure 1). According to F K Annable, in a letter to Dr St. Joseph, it covered an area of about 2.5 ha.

History of Research

Dr. Shaw Mellor visited the site in 1954 and reported (Shaw Mellor 1954) on a piece of terra sigillata that - had been repaired by riveting. This piece is in Devizes Museum (accession no.34/1967). In January the excavator — Mr H Morrison, a worker at the farm — visited the museum and informed Ken Annable, the Curator, that he had found Roman remains and building material. The latter, accompanied by Shaw

Mellor, visited the site and reported on what he found (Annable 1958). In this report he mentions, inter alia, a piece of Bath stone which had been sculpted. This was the subject of a report by Shaw Mellor (Shaw Mellor 1958). The slab, which is interpreted as depicting Hercules slaying the Hydra, is in the museum (11/58-337).

Unpublished is a letter from B Hartley, dated July 1963, reporting on a piece of red colour-coated ware, imitation samian, sent to him for observation. Hartley says that the sherd, which is stamped, is ‘one of the red colour-coated imitations of Samian, Form 31R, such as were commonly made in the Oxfordshire kilns in the later period. They frequently have potter’s marks, though they are normally, and perhaps always, meaningless ciphers. You will find some published by Thomas May, from Sandford, in Archaeologia 72, 233, and, I fancy... that some were also found at the Cowley pottery (Oxoniensa V1).’ Annable’s drawing of the sherd is reproduced as Fig 6.3. A note under the drawing says ‘Sherd in private possession. Stamped sherd found on surface at Euridge Farm, Colerne. Many more finds here, now in the possession of Mr Morrison, Euridge Farm cottages’.

1. 44, Manor Fields, Bratton, Westbury, Wilts BA13 4ST

220 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

The many more finds to which Annable (1958) refers have disappeared, with the exception of the repaired sherd, the sculpted stone mentioned above, some 177 coins, the fibulae, and the small finds described below. The large amount of pottery has gone. The finds, together with plans by Morrison, were deposited in Devizes Museum by Mrs Morrison (accession nos. 1983/73 and 1985/1-9). The site was visited in spring 1989 by Hugh Seymour and the writer. The field was under cultivation but the owners allowed us to walk over the ground. Scatters of small fragments of tile were found, but the reference line in Morrison’s plans — a fence — had gone. His plan showed it as starting at the south-west corner of Ashley Barn, but the farm manager thought it had run from the opposite corner making a difference of 10-12 metres. No pottery was found on this visit. Two further visits were made, one after harvest, the other after ploughing. On the first, in spite of torrential rain, several holes were dug. The last one uncovered part of a wall. The rain prevented further work, but it was possible tentatively to ascribe this wall to the building shown on Morrison’s plan as being opposite the 96 yard mark on his datum line. A further visit was made by Bryn Walters and the writer, after the winter rains showed more features than had been previously seen.

Recent fieldwork by Judith Roseaman

A geophysical survey of the site at Euridge Manor Farm was commissioned by WANHS Archaeology Field Group and carried out during the winter of 1997/98. The work was undertaken by Alister Bartlett of Bartlett-Clark Consultancy, with sponsorship and assistance from the landowner and members of the Archaeology Field Group.

The survey included resistivity, magnetometry and magnetic susceptibility. The full report is held with the rest of the archive at Devizes Museum, but the main results are summarised in Figure 2. There is a large symmetrically planned building to the north of the site, together with a group of other buildings arranged around a rectangular courtyard. Some of the walls noted on the 1950s sketch plan by Morrison have been located approximately and are marked. The buildings are set within and partly superimposed upon a complex system of rectilinear and curved enclosures, of probably more than one period.

It is intended that this geophysical survey should be the prelude to more thorough investigation of the site by the WANHS Archaeology Field Group.

The Finds

BROOCHES by D. F. Mackreth

Colchester Derivatives

1. The spring was held in the Polden Hill manner: an axis bar passed through both coils and pierced plates at the ends of the wings, the cord being held by a pierced crest on the head of the bow. Each wing has two sets of paired mouldings, separated from each other and the rest of the wing by flutes. The bow is relatively broad. The pierced crest has a notch beneath separating it from the beaded ridge down the centre of the bow. On either side of the upper part of the ridge is a step. The bow slopes away to bordering mouldings: a short one on either side of the head rising from the wings and dying back into the side of the bow, and a longer one beneath to the foot which has a seal projecting moulding.

The Polden Hill system belongs to the western parts of England, but this particular brooch does not belong to any definite group using this method of securing the spring. The elaborate form of ornament on the wings is more often found in the eastern parts of Britain and is more a mark of the Ist century than later. The expansion of the head of the bow by the addition of mouldings on each side is, however, a characteristic feature of a major family of brooches in the south-west. One almost exact parallel for the present specimen comes from Keynsham (to be published). The distinguishing feature of both is the added mouldings along sides of the lower part of the bow. These appear intermittently in other groups also belonging to the western parts of Britain (e.g. Hattatt 1985, 83, fig. 35, 377) and especially amongst a group which seems to have been made at Prestatyn (excavations, K Blockley, five examples to be published), but there is nothing to suggest that there is any real connection between these and the Colerne brooch. The indications are that this was an unusual ornament which might be applied almost at will. However, the ornament itself provided the only sign of date: the two main groups involved belong to the late 1st century and to the earlier 2nd with some probably running to c.150 AD as survivors in use.

2. The pin was hinged, its axis bar being housed in the tubular case formed by the wings. Each of these has two sets of three grooves right around, probably intended to look like two buried mouldings. On the wings and under the head of the bow are two pairs of grooves running back from the edge of the bow to

INVESTIGATION OF A ROMAN VILLA SITE AT EURIDGE MANOR FARM, COLERNE 221

Figure 3. Copper alloy and iron brooches, nos. 1-9. All at actual size

222 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

the end of the slot for the pin. The basic bow section is oval and tapers to a pointed foot. On each side of the head is a short bordering moulding rising from the wings and decorated along its outer edge with a series of cross-cuts. The central ridge on the upper bow is beaded and dies out in the middle of a lozenge lying across the bow and made up of pairs of grooves. The foot has three cross-cuts.

The additional mouldings at the top of the bow show that this brooch belongs to the south-western parts of Britain, but, apart from that, there is little to show that the brooch is part of a well defined group within the overall rather loosely defined family. That being the case there is no independent dating for the piece, the only signs lying in the ornament applied to it. The poorly defined wing decoration is, like that on brooch 1, more related to the eastern side of England than to the west. The absence of a proper foot-knob or a version of one, but the presence of simple decoration at the foot of the bow is more a lst century feature than later. On balance, these weak indications point to the later 1st century and are unlikely to be 2nd century.

3. The axis bar of the hinged-pin is housed in oval- sectioned wings. Each of these has a buried moulding at its end. On the head of the bow is the stub of a cast- on loop raised on a tab with ogee sides. The bow is divided into three sections. The top one 1s flat on the front and has two longitudinal recesses for enamel of two colours alternating to give a checked appearance. One colour is an opaque orange, the other is now an opaque grass-green. The rest of the bow has pairs of divided lenticular bosses lying across the bow, one under the enamelled zone, another halfway to the foot-knob and the third on the foot beneath a cross-moulding.

In contrast to Brooches 1 and 2, this one does belong to a distinctive family which is at home in the lower Severn Valley and generally in the south-west. Keeping only to the major variations, the bow may only have two decorative sections, with enamelling in the top part. The common form of enamelling is that on the present specimen, but this pattern may be replaced, usually with a median line of lozenges with infilling triangles on each side. The dating is weak to non-existent: Caerleon, with pottery dating to c.130- 180 (Wheeler and Wheeler 1928, 162. fig. 13, 13). The only other sign, in default of more published brooches from dated contexts, lies in one from Croft Ambrey which, itself, has the very wide time-span of c.75-160 (Stanford 1974, 144, fig.67,4). However, the overall form of that brooch is that of a typical ‘Dolphin’ of the kind found in the Severn Valley and spreading out sparsely over a good deal of Roman Britain. The

dating of that is the late 1st century into the 2nd and possibly surviving to about 150. The relationship between the two groups is very difficult to express, but it is more likely that the Croft Ambrey brooch should be placed with the main floruit of its family and that the present type is later, possibly entirely of the 2nd century, but hardly lasting, even as a survivor in use, as late as the last quarter.

Late La Téne

4. The spring is missing, but almost certainly had four coils and an internal chord. The bow has a thin rectangular section with parallel sides down to the top of the catch-plate where it tapers to a pointed foot. On the front of the bow are two groups of poorly executed grooves, one near the top and the other at the beginning of the tapering section.

5. Iron; The spring has four coils and an internal chord. The section of the bow is lenticular and only just wider than the spring to remove the brooch from the Drahtfibel derivative type, but not broad enough to classify it as a Nauheim derivative.

6. The pin and spring from a four-coiled-internal- chord brooch.

Nothing can be said about brooch 6 as only its spring and pin survive. Neither of the other two really declares any clear affiliation. Brooch 5 is too indefinite in form and brooch 4 lacks both the elongated triangular bow and type of decoration to be properly a descendant (e.g. Crummy 1983, 7-8, fig. 21) of the Nauheim itself. These two brooches seem to be a long way from the slack profiled relatively large brooches with framed catch-plates which belong to the middle of the Ist century BC. The decoration on brooch 4 does not belong to a proper group either. The best that can be said is that all three probably belong to the Ist century AD, and, while all could run to near the end of that century, the use of iron for brooch 5 may place it in the first half, if not actually before the Conquest, and the presence of ornament on 4 would seem at the present to be before the last third of the 1st century.

Unclassified 7. Iron. 8. Iron.

9. Iron.

INVESTIGATION OF A ROMAN VILLA SITE AT EURIDGE MANOR FARM, COLERNE

All three are of the same type and form: the axis bar of the hinged pin is housed in the rolled-under head of the bow. This is a tapering strip ending in a pointed foot. The catch-plate is short, the return at the end of the swept-back foot. No ornament is detectable.

The Strip Brooch is characteristic of the south- west although outliers are known. While its origins are mixed, deriving in part from the earlier stages of the development of the Aucissa and betraying an early acquaintenance with the Langton Down, the iron examples are not susceptible to close analysis. The use of the rolled-under head is typical of the whole family, irrespective of material and almost certainly comes from the earliest developments of the Alesia towards the standard Aucissa, which has a rolled-over head, and so is a trait of this particular school of brooch-makers rather than being an independent dating agent. However, the Strip Brooch was not free from influence from the Aucissa itself (e.g. Stead and Rigby 1986, 120, fig.48, 127; Leech 1986, 316, fig. 34, 1), but the numbers involved are so small that they may safely be discounted. The dating is not very good and this may be a product more of the way in which irom has been virtually ignored in site collections, thus reducing the numbers known by selection rather than being due to a genuine lack of specimens: Braughing, c.10 BC-AD 20 (Partridge 1981, 135, fig. 67, 11), AD 30-40? (ibid., 141, fig. 66, 6); Maiden Castle, before 43 (Wheeler 1943, 252, fig. 85, 35); Waddon Hill, Dorset, c.60-65 (Webster 1960, 97, fig. 7, 16); Puckeridge, Station Road, Claudius-70 (Partridge 1979, 10, fig. 6, 13); Camerton, 60-90 (Wedlake 1958, 216, fig. 50, 4); Nettleton, late 1st century — early 2nd (Wedlake 1982, 120, fig. 50, 2); Gadebridge, 150+ (Neal 1974, 123, fig. 54, 10).

The spread of dates runs, not surprisingly in view of the materials used, from Late pre-Roman Iron Age times through the Ist century. All examples after the 2nd century have been omitted: the British bow brooch ceases to be made before 200. There are at first sight, a surprising number of iron brooches from Late Roman contexts, but this is almost certainly due to the detail that copper alloy would be preferentially recovered if found adventitiously, because of its higher scrap metal value, while iron clogged with dirt would tend to be left. As iron is hardly ever used for any brooch dating after the demise of the Colchester, it would need powerful evidence to insist that it continued to be used for this type alone beyond 100 and it may be that all should be placed before c.75.

bo bo Qo

Trumpet Variety

10. The spring vas mounted on a loop behind the head of the bow, the piercing not being large enough for the rolled sheet metal tube typical of the Trumpet Type. The trumpet head is a little emaciated and there are traces of expansions on each side. In profile, the bow seems to be a thin bar running away from the bottom of the trumpet which, itself, is the end of an elongated element which joins the bow, then loops upwards to a boss and ending in a long tapering section finished off with a small blob. The whole is cast in one. The lower bow, with the catch-plate, is missing. The writer knows of one parallel for this brooch, from Camerton (Wedlake 1958, 224, fig. 51, 17), which shows that the foot ended in two cross- mouldings above a knob. The common form of this fairly rare type has a simple S-shaped profile with a minimal trumpet head, the elaborate form of the decoration has been reduced to a plate whose profile and a dimple betray its association with those like the present example (e.g. Kirk 1919, 11, fig. 3, 4). The origin of the type lies in the 1st century BC, in a type in which there is a small beak under a cross- moulding near the top of a high-arched bow (Hawkes 1940, 192). The next development produced the famous Birdlip Brooch (Smith 1909, 341-2, fig. 9) in which the beak is much more prominent, the trumpet head is reduced, and heavily decorated, and the foot broad. The beak has become much more curved and the next stage in development produced a closed loop (Reading Museum, Silchester Collection 03206). Further distortion (e.g. Hull 1967, 36, fig. 14, 19) introduces a change in the profile and a greater prominence for the loop. It is a matter of choice as to whether the common form mentioned above derived directly from this or whether that is a reduction of an extreme development encompassing the present example: there are nor really enough specimens known for this to be certain. None is satisfactorily dated and the best suggestion is based on the trumpet head and the spring-fitting arrangement which appear to owe something to the Trumpet type whose general date is latest Ist century to 150/175 AD.

11. The axis bar of the hinged-pin 1s housed in a semicircular projection along the bottom of the head- plate. This has an irregular top, possibly the result of having lost a cast-on loop rising from a semi-circular curve. The bow is plain and broad and has a slight swell down its front. The profile shows a recurve with a projecting foot under a small step. The sides of the

224 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Figure 4. Copper alloy brooches, nos. 10-12; other objects of copper alloy, nos. 13, 14, 16-18, 21, 22. All at actual size.

INVESTIGATION OF A ROMAN VILLA SITE AT EURIDGE MANOR FARM, COLERNE 225)

bow also show little in the way of shaping and the overall effect is that of a weak cabriole.

This is a member of a small and poorly understood group having a fairly restricted distribution mainly in Somerset. The dating evidence is meagre. One from Nettleton was dated by coins alone to the late 1st and early 2nd century (Wedlake 1982, 127, fig. 53, 49) and another from Ironmonger’s Piece, Marshfield, can best be assigned to the 2nd century (Blockley 1985, 145, fig. 45, 25). The form is not reminiscent of anything in the lst century and the 2nd century broadly may be the best range that can be suggested.

Penannular

12. The ring has a circular section. The brooch seems to have been forged from rolled sheet metal, the seam showing along the top surface of the ring. Each terminal is folded back along the top of the ring and appears to have two cross-grooves. The pin is straight with a simple wrap-round.

The form of the terminals is basic and the number of grooves may have no chronological significance; as they were made after the brooch vas formed, it must be a matter of chance whether the craftsman put one or two on, or whether there was space for three or more. The straight pin militates against a pre-Roman date and is entirely in accord with Penannulars of the Roman period made in the lowland areas of England: the humped-pin remained a characteristic of the upland areas of the west and north of Britain. Penannulars of this simple class are poorly dated and the bulk belong to the 1st and 2nd centuries AD. In the present instance, the detail that the object seems to have been made from rolled or folded sheet metal would tie in with a marked tradition of making bow brooches as well as Penannulars in pre-Conquest times. A general date- range, therefore, from the middle of the Ist century and possibly into the 2nd may be suggested.

OTHER COPPER-ALLOY OBJECTS

13. Aring made of one and half turns of square wire decorated with five sets of transverse decoration; each set is 8mm long. The ends of the wire are slightly flattened. See Crummy (1983, 47, 1758, fig. 50) fora ring of similar dimensions but different decoration.

14. Bracelet fragment. Square section having tranverse groove decoration (c.f. Crummy 1983, 40, 1676, fig. 44.1676). This appears to have been cast.

15. Part of a bracelet made of two strips wound round a central wire. Crummy illustrates a similar piece made of three circular strands from a late 1st century context (Crummy 1983, 38, 1628, fig. 41.1628). (Not illustrated).

16. Pin with a domed head c.f. Hawkes and Hull 1947, 333, pl.C. No.27)

17. A barrel shaped bead, octagonal in section. Made of a single piece of metal probably shaped over a core and then soldered. The core could have been of wood which was then drilled to take a cord; had the core been removed the bead would not have hung properly. These objects have been variously described as beads, toggles, ferrules, or collars. Allason-Jones (1984, 220-1, fig. 754) shows a ‘bronze barrel-shaped bead or collar’. Crummy (1983, nos 1383 and 1384) shows barrel-shaped beads albeit square in section. Close parallels for the Euridge specimen are described in Robertson (1975, 109-110, fig. 30.5 and 7; 116, 87, fig. 37.10-11).

18. Ligula. A very common toilet instrument. These have been described in many reports; typical ones are displayed in the British and Harlow museums (Brailsford, 1951, 12, 4; France and Gobel 1985, 83, 19; c.f. also Crummy 1983, 59, 1897;fig. 74).

19. Nail cleaner, with broken loop (c.f. Hawkes and Hull 1947, 334, 34, pl.C,34).

20. A boss 30mm in distance, less than 1mm thick, with the stub end of a shank. Similar items from Colchester and Harlow are described as studs (Crummy 1983, 116-117, fig. 20.3117; France and Gobel 1985, 91, 79a). (Not illustrated).

21. Ring, of probably 12th century date, the ends being joined under a plate pointed at both ends. This is decorated with recessed triangles at each end, each having a raised dot in the corners. They are separated from three recessed squares by two rows of four raised dots, with raised dots in each corner. For a similar ring from the Lark Hill hoard, c.1170, see Stratford, 1984, 293, No. 320c.

22. Abuckle, cast, having a central boss with tapering arms curving to meet a narrow cross-bar. There is a faint suggestion of raised decoration at the junctions of boss and cross-bar. 12th—13th century.

23. Part of a ring, cast hexagonal in section with 4mm sides, 48mm in length. (Not illustrated).

226 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Figure 5. Iron objects: nos. 26-36, at 1/2 size; no. 37 at 1/3 size

INVESTIGATION OF A ROMAN VILLA SITE AT EURIDGE MANOR FARM, COLERNE 2251

24. Abent piece of wire 90mm long, rectangular section.

25. Piece of strip, 95mm long, 4-8mm wide. (Not illustrated).

IRON OBJECTS

26. Socketed arrowhead. The socket was made by folding a flange into an approximately oval shape 14mm across. Brailsford (1951, 70, fig. 3.4) describes a similar object as a spear-head, as does Crummy (1983, 135-6, no. 4230).

27. Axe-head with a C-shaped socket: c.f. Wheeler (77, pl.34.2). The curious projection at the ‘front’ of the blade suggests some unusual function.

28. And 29. Miniature adze-hammers. Possibly votive

30. Sickle with a tang: c.f. Collingwood (1969, pl. XX.a) and Hawkes and Hull (1947, 343, 13, pl. CV).

31. Bar, terminating in a loop.

32. Two pieces of saw-edged blade which refit, making a length of 117mm. Possibly decorative.

33. And 34. Two pins (?) 35. A stud with hexagonal domed head.

36. Fire Shovel? The handle has a flat rectangular cross section which was twisted several times in the centre. The terminal, which probably ended in a loop and ring for suspension, is missing. The blade end is incomplete and in fragments: although concave now, it was originally flat. The shovel may be compared with fire shovels from Verulamium (Manning 1972, 164 and fig. 60, 6); Camulodunum (Crummy 1983, 112, fig. 115, 113, no. 2979) — dated to the second half of the 3rd century; Carrawburgh Mithraeum (Richmond and Gillam 1951, 84 and pl. XV 13); and Lakenheath (Manning 1985, A42).

37. Steelyard arm with remains of hooks: that at the end held the load to be weighed while the arm was suspended by one or other of the hooks. Examples in copper-alloy are published by Brailsford (1951, 78, 40 no.11) and Crummy (1983, 99, fig. 104.2508).

38. Knife with bolster, probably 16th century. (Not illustrated).

BONE OBJECTS

39. There are three pieces of worked bone: a broken pin, which resembles Crummy Type 2 with a Type 5 worked head (Crummy 1983.21.18.24,21); and two roughly shaped pieces, 45 and 55mm long, possibly rough-outs. (Not illustrated).

POTTERY

In spite a of the large amount of pottery and tile seen and mentioned by Annable (1958) there are only nine sherds in Devizes Museum, including the repaired sherd described by Shaw Mellor (1954), which come from the original excavations. In addition, 64 sherds, abraded and mainly small, were recovered on recent visits. Included are Savernake, Oxford, New Forest and BBI wares, also several samian sherds including a base, and a red colour-coated foot-rim. There are 17 rims and five base sherds. The piece of imitation samian referrred to in Hartley’s letter (Fig.6.3) was not among the sherds deposited in 1983. The eight pieces (D.M. 1983/73) are described below.

40. Part of samian base, type 31R. Stamped BELSAARVEI.

(BA . 9

Figure 6. Samian and imitation samian stamps

41. Samian base sherd stamped BANVILI. 42 and 43. Samian rim and wall sherd. 44. Part of base, hard buff ware, New Forest.

45. Base, hard buff ware black-coated, New Forest (cf. Sumner 1927, pl.11.1).

228 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

46. Rim sherd with handle. A coarse black fabric with buff surface, black-coated, with scratched decoration to one side of handle. Lines at slight angle to vertical. BBI

47. Half of a strainer bowl 170mm diameter, 60mm deep, with rounded perforated base. A dark buff coarse ware with many calcite grits and one prominent flint. The outer surface is blackened in part; the interior is blackened all over. Collingwood (1969, 273, 95d) shows a strainer bowl, but more elaborate, from Torksey, Lincs. Woodspring Museum has a strainer bowl, complete, in hard grey fabric. This comes from the Locking site. Miss Edith Allan, on seeing this bowl on a recent W.A.N.HLS. visit, said that she had seen a similar bowl being used, in a very old farmhouse in Morvan in the Burgundy region of France, for straining soft home-made cheese. Atkinson (1941, 17.4, no.8) shows a rounded bottomed strainer bowl, from the Oxfordshire kilns at Cowley, in a grey coated white fabric. Young shows round bottomed bowls in his report on Oxford Roman pottery, calling them colanders (Young 1977, 227.84, R.80; 1; R.80.2; 80.4; 228). These are dated mid 1st — 3rd centuries.

COINS by T:S.N. Moorhead

48. The coins reputed to have been found at Euridge number 177 with two associated objects (a lead and a bronze disc, the latter which might be a heavily worn Roman coin). However, the coins have been contaminated with at least 18 intruders which (on grounds of date and place or issue, and state of preservation) almost certainly did not come from the site. Another five coins could easily be intruders. Given that the finder of the coins also collected other pieces, it is possible that some others of the so-called Euridge coins were purchased or found elsewhere.

The main catalogue (held in archive) lists 157 Roman and two English coins. Of the 157 Roman coins, 131 pieces can be attributed to periods, although the identification of ten coins remains tentative, and three pieces listed as official issues might be irregular. It is these 131 coins which are used in the following analysis. Comparison is made with the coin record from the nearby Romano-British site at Broad Hinton! and other Romano-British rural sites, namely Chedworth (Glos.), Owlesbury (Hants), Trevelgue (Cornwall), East Anton (Hants.), Dorn (Glos.) and Sparsholt (Hants).?

The analysis is based on Table 1 and Figure 7 which show a breakdown of the coins by period for

Euridge and Broad Hinton and on Table 2 which displays the mints represented in Euridge coins.’

The numismatic evidence suggests that the site was occupied from the latter part of the 1st century AD or the 2nd century, until c.400 or later. In general terms, the distribution of coins displays a pattern which is quite compatible with other Romano-British rural sites. The small number of coins from the period AD 43-161 (6.1%) is not expected, nor is the absence of coins from the period 161-250 necessarily significant. It is quite probable that some of the coins struck prior to 161 were in fact lost in the period 161- 260. The distribution of coins prior to 260 is not dissimilar at Chedworth and Owlesbury, the latter also having a void in the period 161-260. It can be noted that Euridge is better represented than Broad Hinton in the period before 260 (6.1% to 2%), but this should be qualified by the fact that two of the Euridge coins might be intruders.

The upsurge of coins for Period 18 (260-73) (6.1%) reflects the changing nature of circulating currency in the Roman world and is a prevalent feature at most Romano-British sites (Chedworth 14.5%, Owlesbury 11.1%, East Anton 11.4%, Dorn 10.3%, and Broad Hinton 11.6%). However, Euridge does not have such a significant increase as other sites. The dearth of official coins for Periods 19 (1.2%) and 21 (0.4%) 1s paralleled at other sites, as is the presence of ‘Barbarous Radiates’ (2.9%) which are traditionally attributed to Period 19 (although some were undoubtedly struck in Period 18). To ascribe any specific socio-economic significance to Euridge on the grounds of the various fluctuations in coin frequency in the period 260-317 is therefore incorrect. It is possibly of significance that the coins for the period 260-317 represent 15.3% of the total at Euridge, but 22.7% of the total for Broad Hinton.

Period 22 (317-330) is only represented by three coins (2.3%), of which two might be intruders. Broad Hinton has 5.8% of its coins from this period, but other sites are poorly represented (Chedworth 1.6% and Dorn 2.7%), so this is not necessarily an exceptional feature.

In Period 23 (330-48) there is another major rise in coin loss (30 coinds; 22.9%) which is mirrored by numerous sites (Chedworth 22.3% Owlesbury 15.9%, Trevelgue 26.9%, East Anton, 35.4%, Dorn 20.3% and Broad Hinton 28.5%). This phenomenon can be explained by the increase in the number of official small denomination coins in circulation.

The slight fall-back in Period 24 (348-364) (11.4%) again follows the expected trend. The six irregular coins from this period probably reflect the

INVESTIGATION OF A ROMAN VILLA SITE AT EURIDGE MANOR FARM, COLERNE 229

Table 1. Analysis of the Euridge and Broad Hinton coins by period

Key: A — Probable Euridge site finds: Te II. II. B — Probable Euridge site finds: I. Il. C — Possible intruders (official)

Euridge Coins

Period AI Il Il BI II Cc Total

1 (43-54) 1 2 (54-68) 1

3 (68-81)

4 (81-96)

5 (96-117) 6 (117-138) 1 7 (138-161)

8 (161-180)

9 (180-192)

10 (192-217)

bo cs

bo

S Orel! HO Oo 16) a

11-17 (217-260) 18 (260-273)

19 (273-286) 20 (286-296)

21 (296-317) 22 (317-330) 23 (330-348) 24 (348-364) 25 (364-378) 37 26 (378-388) 1 26 or 27 (378-402) 5 27 (388-402) 5

bo

bo

Nb ee = oo bo bo — a eo So

Ole be co io) On

to

Totals 96 10 3 15

bo On

Irregular

131

increasing shortage of official low denomination pieces in the 350s and early 360s.

The coin record at Euridge peaks, as at many other Romano-British sites, in Period 25 (364-78 AD) (Euridge 32.1%), Chedworth 36.2%, Owlesbury 31.7%, Trevelgue 32.8%, Broad Hinton 33.3%). Again this can be explained by the large output of official low denomination bronze coins.

The dearth of coins for Period 26 (378-88 AD) ~ (0.8% or possibly more) is quite typical and reflects a shortage of low denomination coins, quite probably caused by Britain’s separation from the Central Empire during the usurpation of Magnus Maximus, 383-8 (Casey 1988, 47). However, the number of coins for Period 27 is noteworthy (5.3-9.1%). Some sites have few (Chedworth 0.3%) or no coins (Owlesbury) from this period suggesting decline or abandonment, but others like Euridge are better represented (Dorn 18.9%

Official issues Official issues, tentative identification Possibly irregular

Irregular, tentative identification

Broad Hinton Coins*

% Annual loss/ Total % Annual loss/

1000 coins 1000 coins

0.75 0.7 0 0.75 0.5 0 AS} 1.8 0

0

1 0.5 0.3 0.8 0.4 0 15 0.7 2 1.0 0.4

0

0

1 0.5 0.2

0 6.1 4.7 24 11.6 8.9 6.9 53 11 5S 4,1 1.5 15 8 3.9 3.9 0.8 0.4 4 1.9 0.9 A) 1.8 2 5.8 4.4 22.9 12277 59 28.5 57; 11.4 fie2 ip! a9) 33) 32:1 22.9 69 33'3 231 0.8 OL75 3.8 Gi2)a* 5 2.4 1,0** 53 3.8

207

and Anton 3.9%). Broad Hinton only has 2.4% for Periods 26 and 27 combined, which is considerably less than Euridge’s 9.9%. Does this reflect better fortunes at Euridge in the late 4th or early 5th century?

It seems quite likely that Euridge was still occupied c.400. Because supplies of official bronze coins ceased by 402, later low value coins are not found on Romano-British sites. However, this does not mean that earlier pieces did not continue to circulate for some years after 400. Therefore, Euridge might have continued in occupation beyond 400, a supposition that might be supported by the existence of worn coins from Pericd 27 (Casey 1988, 47-8). That Euridge is a quite normal rural site is supported further by the fact that there is a higher proportion of 4th century coins to 3rd century ones (Reece 1987,72).

An analysis of the mints represented does not provide any surprises either (see Table 2). For the

230 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

25 20 4 = - OEuridge eC | | | .

ro) | Broad Hinton oO | eee ; S45 oO - = ® Q. Ww ” 2 =, 10 =] i= Cc <

3)

) i a | a i

<O ae oe a or ees gS

OS Be oe oS BO o@ oS of GO

g CF RUE REQ ECE

QC Qe

Qo oe Qe

© Q Ee sO

Q

eF e& gf

Figure 7. Euridge and Broad Hinton: annual loss per 1000 coins. (Note that the figures are taken from table 1. All possible intruders and coins with tentative identifications are included. It is likely that the coins in Periods 26/7 belong to Period 27.)

period 43-286, the mints represented are as one would expect, notably Rome and various other mints of the Western Provinces (Lugdunum, the _ two undetermined mints of the Gallic Empire, Milan and ?Ticinum). For the period 330-78, the three Gallic mints of Trier, Lyons and Arles account for the bulk of the coinage with Trier predominant in Period 23 (45.4%, probably more) and Arles and Lyons taking the lion’s share in Period 25 (60.9% and 26.1% respectively). These proportions are quite compatible with Richard Reece’s findings after his analysis of coins from many Romano-British sites (Reece 1978).

In conclusion, it can be said that Euridge is a rural Romano-British site whose coin record conforms quite satisfactorily with the picture portrayed by other similar sites. It is interesting that, although they follow the same provincial trend, the coin records for Euridge and Broad Hinton do have some differences.

Notes

1. Broad Hinton is an unexcavated rural Romano- British site about 20 miles east of Colerne (GR c,1076). 256 Roman coins, two English coins and six miscellaneous metal objects were found with the aid of a metal-detector. The coins are now housed in Devizes Museum and are the subject of an unpublished report (Moorhead 1983).

2. These sites are the subject of discussion in Reece 1987, p.76ff. For all subsequent mentions of these sites, see p.77, Table 5.1.

3. In the analysis of the Euridge site finds, the following have been consulted: Reece 1987; Casey 1980, 1986, 1988.

INVESTIGATION OF A ROMAN VILLA SITE AT EURIDGE MANOR FARM, COLERNE 231

Discussion and Assessment of the Site by Bryn Walters

Since being brought to the attention of Devizes Museum in 1957, the site at Euridge Farm, Colerne has been considered as a possible Romano-British villa. The random series of stone foundations, sketch- planned by the late Mr H Morrison by themselves make little sense. More recent field-walking of the area (by LL and BW) has identified further areas of buildings with considerable quantities of fractured tegulae and imbrices among the surface debris, but no evidence of hypocaust tubulli, tesserae, or plaster fragments usually diagnostic of villas. Taking into account the considerable amount of pottery seen by Annable (1958), alas now lost, coupled with an interesting assemblage of iron, brooches and an unusually high coin list for a random series of exploratory holes, the site was likely to have been a well-built settlement probably with a minor industrial bias. Morrison’s building ‘A’ (approximately 25 x 7m) suggests a structure intended for industrial or agricultural use rather than residential. It is interesting, however, that a similar structure has now been identified on the same alignment a short distance to the south, both these buildings appearing to delimit the site on its south-east side. A further substantial

building is set back from the main area of debris on rising ground to the north and appears to define the edge of the settlement on that side. (This assessment of the site was undetaken prior to the recent geophysical survey.)

Acknowledgements

Thanks are due to Mr J G Robinson, present owner, and his manager, MrT Bunting, who permitted visits to the site and allowed trial holes to be dug; also to Don Mackreth (specialist report on the fibulae); Sam Moorhead (specialist report on the coins); Graham Webster; Alister Bartlett (geophysical survey); to Nick Griffiths for the drawings and advice, finally to Bryn Walters for the assessment of the site. The geophysical survey was paid for by generous donations from Mr Robinson, Hugh Seymour and the author.

Bibliography

ALLASON-JONES, L. and MIKET, R., 1984, The Catalogue of Small Finds from South Shields, Roman Fort. Society of Antiquaries of Newcastle upon Tyne, Monograph Series 2.

Table 2. Mints represented in the Euridge coins.

Note: Figures in parenthesis are only tentative attributions and are not included in the percentage calculations.

Period Britain Gaul

Une. Trier Lugd. Arles Mint I 1 2 1 3 4 5 6 7 (1?) 18 19 1 20 2 PAI 1 22 1 1 23 5(4) 45.4% 19.1% 3 27.3% 24 2(1) 50% (1) 2 50% 25) 6(3) 26.1% 14(8) 60.9% 26 27

Totals 2(1?) 9(5) 10(4) 19(8)

Italy Balkans Mint II Rome Eic: Milan Siscia Total 1 3 3 1 1 1(1?) 1(1?) 2(1) 3 1 6(1) 1? 2 2 1 2 218.2% 11(4) 4(2) 1(1) 4.3% 28.7% 23(12)

2(1) 11(1)(1?) 1? 1 2

232 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

ANNABLE, F.K., 1958, ‘A Roman Site at Colerne’, WAM Wi hs:

ATKINSON, R.J.C., 1941, ‘A Romano-British Potters’ Field at Cowley, Oxon.’ Oxoniensa 6, 9-21

AUBREY, J., 1847, The Natural History of Wiltshire. (Edited by J. Britton) London: Wiltshire Topographical Society.

BARTLETT,A.D.H., 1998. Euridge Manor Farm, Colerne, Wiltshire. Report on Archaeological Survey 1998. Unpublished report by Barltett-Clark Consultancy for WANHS

BLOCKLEY, K., 1985, Marshfield, Ironmonger’s Piece Excavations 1982-3: An Iron Age and Romano-British Settlement in the South Cotswolds. Oxford: British Archaeological Reports 141.

BRAILSFORD, J., 1951, Guide to Antiquities of Roman Britain. London: British Museum.

CASEY, P.J., 1980, Roman Coinage in Britain. Princes Risborough: Shire Publications.

CASEY, P.J., 1986, Understanding Ancient Coins. London

CASEY, P.J., 1988, Coins and the Archaeologist. 2nd Edition. London

COLLINGWOOD, R.C. and RICHMOND, I., 1969, The Archaeology of Roman Britain. (1930, re-issued 1969) London: Methuen

CRUMMY,N., 1983, Colchester Archaeological Report 2: The Roman small finds from excavations in Colchester, 1971-9. Colchester Archaeological Trust: Colchester

DUDLEY, D., 1967, ‘Excavations on Nor’ Nour in the Isles of Scilly, 1962-6’, Archaeological Journal 124, 1-64

FRANCE, N.E., and GOBEL, B.M., 1985, The Romano- British Temple at Harlow. West Essex Archaeological Group.

HATTAT, R., 1985, Iron Age and Roman Brooches: a second selection of brooches from the author’s collection. Oxford: Oxbow

HAWKES, C.F.C., 1940, ‘An Aylesford La Tene III Brooch from Arundel Park, and the Dating of the Type’, Antuquaries Journal 20, 492-495

HAWKES, G.F.C.;“°and HULL 3.- MiR; 1947, Camulodunum: !st Report on excavations at Colchester,1930-39. London: Society of Antiquaries Research Reports

HULL, M.R., 1967, ‘The Nor’Nour Brooches’, in D. Dudley, ‘Excavations on Nor’Nour in the Isles of Scilly, 1962-6’, Archaeological Journal 124, 28-64

KIRK, J.R., 1949, “Bronzes from Woodeaton’, Oxoniensa 14, 1-45

LEECH, R., 1986, ‘The excavation of a Romano-Celtic Temple and a later cemetery on Lamyatt Beacon, Somerset’, Britannia 17, 259-328

MANNING, W.H., 1976, Catalogue of Romano-British Ironwork in the Museum of Antiquities, Newcastle upon Tyne. Newcastle upon Tyne: The Department of Archaeology, University of Newcastle upon Tyne

MOORHEAD, T.S.N., 1983, The Broad Hinton Coins.

Unpublished manuscript in Devizes Museum

NEAL, D.S., 1974, The Excavation of the Roman Villa in Gadebridge Park, Hemel Hempstead, 1963-68. London: Society of Antiquaries Research Reports 51

PARTRIDGE, C., 1979, ‘Excavations at Puckeridge and Braughing, 1975-9’, Hertfordshire Archaeology 7, 28- 132

PARTRIDGE, C., 1981, Skeleton Green: A late Iron Age and Romano-British site. London: Britannia Monograph Series 2

REECE, R., 1978, ‘Bronze Coinage in Roman Britain and the Western Provinces’, in ‘Essays presented to Humphrey Sutherland’, Scripta Nummaria Romana. London

REECE, R., 1987, Coinage in Roman Britain. London: Seaby

RICHMOND, LA. and GILLAM, J., 1951, ‘Carrawburgh, a Roman Temple’, Archaeolgia Aeliana 29, 1-92

ROBERTSON, A.S., 1975, Birrens (Blatobulgium). Edinburgh: Dumfriesshire and Galloway Natural History and Antiquarian Society

SHAW MELLOR, A., 1954, ‘Roman Pottery Repairs’, WAM 55, 283-4

SHAW MELLOR, A., 1958, ‘Hercules Slays the Hydra at Euridge’, WAM 57, 240-1

SMITH, R.A., 1909, ‘On a late-Celtic mirror found at Desborough, Northamptonshire, and other mirrors of the period’, Archaeologia 61, 329-426

STANFORD, S.C., 1974, Croft Ambrey. Hereford

STEAD, I.M.and RIGBY, V., 1986, Baldock: The Excavation of a Roman and Pre-Roman Settlement, 1968-72. London: Britannia Monograph Series 7

STRATFORD, N., 1984, in English Romanesque Art, 1066 - 1200. London: Arts Council

SUMNER, H., 1927, Excavations in the New Forest. London: Chiswick Press

WEBSTER, G., 1960, “The discovery of a Roman Fort at Waddon Hill, Stoke Abbott, 1959’, Proceedings of the Dorset Natural History and Archaeological Society 82, 88-108

WEDLAKE, W.J., 1958, Excavations at Camerton, Somerset. Privately printed

WEDLAKE, W.J., 1982, The Excavation of the Shrine of Apollo at Nettleton, Wiltshire, 1956-71. London: Society of Antiquaries Research Reports 40

WHEELER, R.E.M., ed. 1930, London Museum Catalogues No.3: London in Roman Times. London: HMSO

WHEELER, R.E.M., 1943, Maiden Castle, Dorset. London: Society of Antiquaries Research Reports, 12

WHEELER, R.E,M. and WHEELER, T.V., 1928, “The Roman Amphitheatre at Caerleon, Monmouthshire’, Archaeologia 78, 111-218

YOUNG, C.Y., 1977, Oxfordshire Roman Pottery. Oxford: British Archaeological Reports 42

Wiltshire Archaeological & Natural History Magazine, vol. 93, 2000, pp. 233-254

Notes

EXCAVATIONS OF BRONZE AGE AND ROMANO-BRITISH SITES ALONG THE CHIPPENHAM WESTERN BYPASS A4 TO A350 LINK

by Clifford Bateman and Dawn Enright

with contributions by Jane Timby and Graeme Walker

Evidence of Bronze Age occupation and Romano- British agricultural activity was revealed in 1997 during excavations along the route of the Chippenham Western Bypass. In addition lithic material retrieved during a watching brief on the remainder of the road scheme suggests Mesolithic and Bronze Age activity within the immediate area.

In September 1997 Cotswold Archaeological Trust (CAT) undertook a programme of archaeological recording, comprising excavation and a watching brief, along a section of the Chippenham Western Bypass between the A4 and the A350, a distance of 3.9km (Figure 1). The work was commissioned and funded by Wiltshire County Council Environmental Services Department.

Archaeological evaluation of the road corridor in 1991 had revealed two sites of archaeological interest (Dyer 1991). The more northerly site was Area E (ST 8970 7195), where fieldwalking had revealed a concentration of Romano-British material, including pottery and tile. The other site was Area F (ST 8986 7176), where a concentration of worked flint broadly dated as prehistoric was located. No further archaeological deposits are known from the road corridor itself. To ensure continuity in the project

archive the site identification codes (Areas E and F) are reproduced here. The archive will be deposited with Devizes Museum.

LOCATION AND SETTING

The underlying geology of the area is dominated by Middle Jurassic Cornbrash with a surviving outcrop of Kellaways Clay at the northern limit of the scheme (Geological Survey of England Sheet 265). The road corridor crossed generally flat, mixed agricultural land ranging from approximately 50m OD at its southern extent, to 60m OD at the northern limit.

Within the immediate vicinity a number of archaeological features have been identified. A Neolithic pit, Romano-British enclosure ditch and Saxon sunken-featured building were revealed 30m north of Area E (ST 8980 7270) (Anon 1991, 143). Archaeological evaluation undertaken at Showell Nursery (centred on ST 9130 7140), 800m beyond the southern limit of the road corridor, revealed late Neolithic/early Bronze Age activity including pits, postholes, gullies and ditches; a middle Iron Age pit and ditch; Romano-British enclosure ditches; and a medieval gully (OAU 1991; Anon 1993, 159). Archaeological excavation undertaken by Wessex Archaeology 2.3km to the north of the study area (centred on ST 8987 7415) revealed a pit containing late Bronze Age pottery (N. J. Oakey, pers. comm.).

METHODOLOGY

Machine excavation of the topsoil within Areas E and F (both 100m by 25m) revealed a number of archaeological features. A site meeting he!d with representatives from Wiltshire County Council Archaeological Service and Environmental Services Department agreed that archaeological excavation should continue by hand in both areas. All intrusive

Cotswold Archaeological Trust, Headquarters Building, Unit 9, Kemble Business Park, Cirencester, Glos GL7 6BQ

234 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

® Berkshire

Chippenham

Wiltshire

Diamonds denote Archaeological Sites within Immediate Vicinity = eae . \ \

NN

f& eg PR e ~

Sa

¢ r- Hl Tes | i, SN @ e *\ ony fig} BE 7 i NNe fie es ea Sal ee i Milbourne Farm

eo

t / a Farm

SS ~/@\ Holywell House LN

|

Showell Farm | |

eed | Area of watching brief

Figure 1. Location Plan

NOTES

groundwork along the remainder of the road corridor was subject to a watching brief.

RESULTS OF THE ARCHAEOLOGICAL RECORDING

Area F

Area F was located on flat ground and the natural substrate largely comprised Kellaways Clay, although Cornbrash was revealed at the northern limit of the site. A cluster of truncated sub-oval pits, stakeholes, and a gully was revealed towards the southern end of the excavation (Figure 2).

Pits

Three contiguous pits, 104, 106, and 108, aligned north-west to south-east, were revealed 35m from the north-western limit of the excavation. Due to the

235

similar nature of the orange-brown clay fills the relationship between the pits remains undetermined and although they are likely to be intercutting, contemporary use and/or abandonment remain possibilities. Two sherds of Bronze Age pottery and three worked pieces of flint were retrieved from 106. Two further contiguous pits, 112 and 124, were located 15m from the south-eastern limit of the site. The pits were aligned east-west and were of similar dimensions. The relationship between the pits remains undetermined, but may be similar to that of pits 104, 106, and 108 above. Two worked flint flakes were retrieved from pit 112. Eighteen pieces of worked flint were retrieved from fill 111 of pit 110, located 10m north- west of pit 112. At the south-eastern limit of the excavation was pit 116. It contained a later Mesolithic microlith (an edge-blunted point). A further 15 heavily truncated pits, typically 0.08m in depth and ranging in length from 0.3m to 2m, were excavated, but no artefactual material was retrieved from them.

Plan 15m to end of site ; Fae 148) Modern field drai ug iis 0134 130). 142 0, wy oa Use 116 150 110 1400 D €/2 section 3 is 5D 2, gt 124 8120 Section 4 > & Section 2 igat60 $128 S Section 1 Oe otne fo) A ° 164 Uo 158 a 4 152 178 ace —166 3 104106 108 =a 126 bal Geotechnical test pit 176 | 3D 174 172 170 8 156 2 ) 1222 0 20m es > Section 1 109 108 KR 198 aon — 62m AOD se log OF ea ee Be a Section 2 Section 3 Section 4 aut - 62m AOD = {61,.5m AOD ——— = Alisly7. ae 3 WH 113 AG 111 y 2 110 = — 61.5m AOD O Im fm OS

Figure 2. Area F: Plan and Sections

236 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Stakeholes

Of the total of 13 stakeholes recorded, all but two formed a distinct group to the west of the pits. The stakeholes ranged in size from 0.1m to 0.2m in diameter, had an average depth of 0.05m, and contained orange—brown clay fills from which no artefacts were recovered.

Gully

Gully 122 was revealed at the western edge of the excavation. Orientated north-south, it was at least 2m long, 0.6m wide and 0.3m deep. Its uniform linear shape, steep profile and flat base were noticeably distinct from that of the contiguous pits. Interpretation of the feature is limited by the absence of artefactual material and, while it may be contemporary with the adjacent pits and stakeholes, the possibility that it is a natural feature should not be overlooked.

The Finds

Pottery by Jane Timby

Two handmade bodysherds with a combined weight of 11g were retrieved from pit 106. The sherds were in relatively fresh condition and probably derived from the same vessel. The exterior surface and outer core were orange, while the interior surface and inner core were black. The sherds consisted of a fine sandy fabric with fine elongated voids on the surface arising from the former presence of organic matter. Larger ovoid voids were visible in a fresh fracture, possibly from chalk/limestone inclusions, with sparse, rounded, dark red-brown iron compounds also present. The character of the fabric and the firing is typical of ceramic material of Bronze Age date.

Worked flint by Graeme Walker

Twenty-four pieces of flint were recovered from stratified contexts, with a further five unstratified pieces. Material from the stratified contexts is very fresh and largely undamaged, indicating relatively rapid incorporation into pit fills. The small assemblage is characterised by small secondary and tertiary flakes, a few struck from blade cores. Raw material is variable and the artefacts are generally small, both aspects consistent with exploitation of derived material. It is notable that there is no primary flaking waste and

few tools, which suggests that final manufacture and repair of existing tool-kits was the main activity taking place within the areas examined. Tool use and discard were presumably taking place off-site.

The general characteristics of the assemblage would suggest a late prehistoric date, perhaps Bronze Age being appropriate for most of the collection, athough a late Mesolithic edge-blunted point does suggest earlier activity within the area.

Area E

Area E was located immediately to the south of Pudding Brook at the northern limit of the bypass corridor. The underlying geology consisted of Kellaways Clay.

Two truncated ditches forming a “T’ shaped junction were revealed at the southern limit of the excavation (Figure 3). Ditch 003, orientated east- west, was on average 1.2m wide and 0.2m deep. Two small Romano-British sherds (weighing 6g) were retrieved from its fill. Both consisted of fine grey sandy ware from wheelmade closed forms (identified by Jane Timby). Ditch 005, orientated north-south, was 1.5m wide and 0.15m deep. The ditches are interpreted as contemporary, perhaps forming agricultural boundaries and/or drainage ditches.

The Watching Brief

No archaeological features were revealed during the watching brief, although a small assemblage of 106 pieces of unstratified worked flint was retrieved from the ploughsoil throughout the bypass corridor. None of the recorded material is diagnostic, but the general characteristics suggest that this assemblage is of similar date and composition to the stratified material from Area F, although the presence of numerous small snapped blades provides further evidence of Mesolithic activity.

CONCLUSIONS

The earliest activity detected, particularly during the watching brief, is represented by the Mesolithic component to the flint assemblage. No settlement or core of activity was identified and consequently little further comment can be made on this activity save to note that it is typical of the scatters recorded in the Chippenham region (Tucker 1985, Anon 1993, 159;

NOTES

Plan

80m to end of site

fee # mode™

Section

Figure 3. Southern portion of Area E: plan and section

Bateman 2000), and across the southern Cotswolds generally (Saville 1984), which presumably indicate sites of temporary camps utilising woodland and riverine resources.

By contrast, late Neolithic and late Bronze Age settlement has previously been suggested both to the north and south of these sites and, although no evidence of structures has been identified the quantity of domestic refuse retrieved from ditches and pits has been interpreted as indicative of nearby occupation (OAU 1991; N.J. Oakey, pers. comm.).

Although no definitive habitation features were identified at Area F, Bronze Age occupation may be suggested both by the density of pitting, which is indicative of extended use rather than a single episode of activity, and by the close grouping of stakeholes, possibly representing a basic shelter or windbreak. The paucity of domestic refuse, such as pottery, retrieved from the excavation may indicate the pits are peripheral to the main settlement area. Alternatively it may suggest short term, maybe seasonal occupation. The lithic evidence also suggests use of the site as a temporary settlement as the final manufacture and repair of tool-kits rather than primary reduction was being undertaken, perhaps prior to the exploitation of the immediate area. The full extent of the site was not revealed, and the likelihood of further settlement features to

237 Section oe = = — 58m aop O Tm

the south and east of Area F must be considered high.

It remains unclear whether this occupation site was deliberately located upon the Kellaways Clays, rather than the Cornbrash which dominates the local geology, or whether it was located to take advantage of the local topography (overlooking the Pudding Brook and afforded some protection from the prevailing weather by the slightly higher ground to the north, east and west). However, it is worth noting that the prehistoric activity previously identified immediately beyond the northern and southern limits of the road corridor was also sited upon Kellaways Clay rather than the Cornbrash (Anon 1991,143; 1993, 159).

The Romano-British activity revealed at Area E confirms that the Romano-British agricultural activity previously identified to the north of Pudding Brook (Anon 1991, 143) continues to the south and it is noteworthy that the activity is restricted to the lower slopes rather than the floor of the valley. Obvious limitations are placed upon the interpretation of such seemingly isolated features, but they add to our knowledge of the spatial distribution of Romano- British activity within the Chippenham area, where an increasingly dense settlement pattern, notably along the Cotswold dip slope and the immediate environs of the North Wiltshire Clay Vale, is becoming apparent (Bateman 2000).

238 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Acknowledgements

The archaeological recording was funded by Wiltshire Civil Engineering Consultancy. The authors would like to thank Duncan Coe (Wiltshire County Council Archaeological Service) and Peter Hanson (Wiltshire Civil Engineering Consultancy) for their assistance during the course of this project, and Niall Oakey (Wessex Archaeology) for making information from previous work along the bypass corridor readily available during the compilation of this report.

The project was managed by Dawn Enright. The excavations at Areas E and F were directed by Clifford Bateman and the watching brief supervised by Franco Vartuca. The illustrations were drawn by Richard Morton.

References

ANON 1991, ‘Excavation and Fieldwork in Wiltshire 1989: Chippenham’, WAM, 84,143

ANON 1993, ‘Excavation and Fieldwork in Wiltshire 1991: Chippenham’, WAM, 86, 159

BATEMAN, C., 2000, ‘Excavations Along the Littleton Drew to Chippenham Gas Pipeline’, WAM, 93, 90-104

DYER, C. A., 1991, The Archaeological Works on the Chippenham Bypass Route Spring 1991, Unpublished client report: Thamesdown Archaeological Unit

OAU (OXFORD ARCHAEOLOGICAL UNIT) 1991, Archaeological Evaluation: Milbourne Farm and Showell Nurseries, Chippenham, Wiltshire. Unpublished client report.

TUCKER, J.H., 1985, ‘Upper Palaeolithic and Mesolithic Sites in the Chippenham Area’, WAM, 79, 226-8

A MINIATURE FLAT AXE OR CHISEL FROM BROAD TOWN, NORTH WILTSHIRE by Bob Clarke

A small bronze axe was discovered in May 1998 whilst returning an area of land back to garden at ‘The Laurels’, Broad Town, North Wiltshire (SU 09057792), by the owner of the house Mr Michael Broomfield. No other features or items were found. The object has been identified as a miniature flat axe or chisel.

The axe has suffered slight damage to both the cutting edge and butt, and exhibits a dark green/brown patina with some slight pitting and corrosion on both sides. The original outline is only slightly reduced by deterioration of the artefact. It has a very slight stop- bevel positioned exactly half way along the overall length of the axe and a gently curved butt profile. The sides descend from the butt almost vertically to the stop-bevel then expand in a smooth curve to produce a cutting edge just over twice the width of the butt. The sides are bowed and a slight rise in edge section thickness is evident between the stop bevel and the cutting edge, which probably results from final working of the axe during manufacture. The dimensions are: length 85.5mm, width of cutting edge 32.5mm, maximum thickness 5.5mm and weight 50.53 gms.

]

'S rie

CENTIMETRE

3

Q

Figure 1. Bronze miniature flat axe from Broad Town (DERA Boscombe Down, © Crown Copyright)

Bob Clarke, c/o DERA, Boscombe Down, Amesbury, Salisbury SP4 7RE

NOTES

Discussion

Miniature axes often parallel larger forms, and the example from Broad Town is no exception, with similar larger forms being a relatively common type in England, mainly concentrated in the southern half of the country (West pers com). Clusters are present in Cornwall, Wessex and South Wales, as well as examples in East Yorkshire and Ireland (Needham 1983, 167-8).

The presence of a stop-bevel separates the axe from Class 3 axes in the British Bronze Age Metalwork Series. The blade width differentiates it from Class 4A axes and the absence of flanges along the sides separates it from Class 4C and later axes. This places the Broad ‘Town example in Class 4B of the series and gives it a date of c.2000-1800 BC (Needham et al 1985, iii).

One miniature example is known from the local area which parallels very closely that from Broad Town, discovered at Tan Hill, All Cannings (SU 090645) around 1864. It is currently held in Salisbury and South Wiltshire Museum (Moore and Rowlands 1972). It has a length of 81mm and is considered to be in the Migdale-Marnoch tradition as described by Britton (1963). The find spot on Tan Hill is 14 km south of the discovery at Broad Town.

Other miniature flat axes are recorded at Devizes Museum, one from an unknown locality being 70mm long (Annable and Simpson 1964, 52), whilst another was recovered from a primary deposit in a bowl barrow at Collingbourne Kingston G4 (SU 21385179). The latter had a length of only 45mm (Annable and Simpson 1964, 57). Further afield, a miniature flat axe was located as part of a hoard of 11 bronze objects during the excavation of the Iron Age hillfort of Danebury, Hampshire (Cunliffe 1984, 335). This axe was smaller than the Broad Town example and has been interpreted by Dennis Britton as an example of a light woodworking tool (1963, 271).

A GOLD FINGER RING FROM THE RUDGE ROMANO-BRITISH VILLA SITE, FROXFIELD, WILTSHIRE

by Bernard Phillips! and Martin Henig?

239

Unfortunately, the Broad Town axe falls into the un-provenanced category of find, there being a number of possibilities as to how it got to its last position. It could have been found elsewhere and brought to the house, or it may have been part of an in situ deposit disturbed during the many phases of building carried out over the centuries.

Acknowledgements

Thanks go to Mike Broomfield for his prompt reporting of the find and Alan West at the Department of Prehistoric and Romano-British Antiquities, the British Museum, for his useful comments and assistance in producing this note. Any errors are naturally my own.

Bibliography

ANNABLE, F. and SIMPSON, D. 1964 Guide Catalogue of the Neolithic and Bronze Age Collections in Devizes Museum, Devizes: Wiltshire Archaeological and Natural History Society

BRITTON, D. 1963 Traditions of Metal-working in the Later Neolithic and Early Bronze Age of Britain: Part 1, Proceedings of the Prehistoric Society 29, 258- 325

CUNLIFFE, B. 1984 Danebury: An Iron Age Hillfort in Hampshire: Vol. 2, The excavations, 1969-1978: The Finds, London: Council for British Archaeology Research Report 52

MOORE, C. and ROWLANDS, M. 1972 Bronze Age Metalwork in Salisbury Museum, Salisbury: Salisbury and South Wiltshire Museum

NEEDHAM, S. 1983 The Early Bronze Age Axeheads of Central and Southern England, Unpublished PhD, University College Cardiff

NEEDHAM, S., LAWSON, A and GREEN, H.S. 1985 British Bronze Age Metalwork: Al-6 Early Bronze Age Hoards. London: British Museum Publications

The Site

In the late 1980s a gold finger ring was recovered from plough-soil overlying the Rudge Romano-British villa site by a metal detector user. Being the gamekeeper for the Littlecote Estate, on whose land the villa lay, Mr. M. Goodfield, the finder, brought the ring to the first author who was then directing an archaeological excavation on the Littlecote Romano- British villa site. The author contacted Martin Henig

1. Roman Research Trust 15 Yiewsley Cres., Stratton St Margaret, Swindon SN3 4LT 2. Institute of Archaeology, 36 Beaumont St.,Oxford OX1 2PG

240 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

who examined the ring. His account forms the second part of this report. Following drawing by the Littlecote excavation’s draughtsman, Luigi Thompson, the ring was returned to the finder.

Evidence that a Romano-British villa had existed at Rudge first occurred during clearance of woodland in 1725 (Pugh and Crittall 1957). Then foundations, a well and a tessellated pavement were located. The damaged central emblema of the pavement depicted the lower parts of two figures and a jar from which water flows. Late Roman coins, an enamelled cup and five human skeletons were found in the well. The cup shows and names military forts. Identification of these has been made with forts located on Hadrian’s Wall. Ploughing in about 1875 revealed a stone statuette thought to be of Atys. The discovery of tesserae occurred during the second world war in the course of digging foundations for Nissen huts.

During an aerial survey in 1976 of local villa sites, involving the first author, cropmarks of a south facing rectangular structure were seen, presumably the villa- house, fronted by a corridor and with a room protruding at the east-end. On the west side of an apparent courtyard fronting the house a further rectangular building, divided widthways into three rooms, was also revealed.

Figure 1. Rudge: Roman gold ring (scale = 10mm)

The Ring

The ring is very small with a diameter of about 12 mm and a weight of 2.6g, presumably intended for a girl. Itis made from sheet gold approximately 0.75mm thick and is faceted, probably octagonal, a form which is widespread in the 3rd century. The topmost facet which is rectangular with bowed sides (8mm by 7mm) serves as a bezel with an engraved intaglio device of a

palm-branch with four projecting leaves on each side, all of them terminating in pellets. When found the ring was badly distorted, but this may have been recent damage.

For gold rings of the same type engraved with the palm-frond device, comparison may be made with two even lighter in weight (respectively 1.8g and 1.0g) from a Severan hoard at Lyons (Comarond 1844, 40, pl. 3 nos. 25 and 26). There is no exact parallel from Britain but I know of a number of faceted rings from the province, for instance two from Vindolanda (Birley 1973, 119-20, pl. XVIII nos. 17 and 18), one set with a gem, the other plain, and one from Carrawburgh (Allason-Jones and McKay 1985, 19-20, no. 30 pl. XID), also plain.

The device of a palm with its connotations of victory and success (including victory in love) is widespread. A gold ring from Verulamium of somewhat different form, but perhaps also 3rd century may be noted (Henig 1984, 19. no. 1, pl. la, b) as well as a gold ring from Carlisle inscribed ‘AMA ME’ (Dalton 1912, 4, no. 15,) while for the form of palm with pellets, the device on a silver ring, probably a 2nd century type, from Southwark comes to mind (Henig 1976). Incidentally, although such pelleting is to be seen on some Iron Age coins of the Dobunni, it is hardly conceivable that the two types of objects can be linked over two centuries and, besides, such pelleting is to be seen on palms engraved on two rings from the Continent (Hoffman and Von Claer 1968, 184, no. 124).

Gold rings were the prerogative of such people of high social rank and, however expanded the definition of such an honour had become by the 3rd century, we clearly have here an object likely to have belonged to the child of a villa-owner: The device hints at her future prospects in finding a suitable husband.

References

ALLASON-JONES, L. and McKAY, B., 1985, Coventina’s Well. A shrine on Hadrian’s Wall. Chesters Museum

BIRLEY, R., 1973, ‘Vindolanda-Chesterholm 1969-1972: some important material from the vicus’, Archaeologia Aeliana, 5th series 1

COMARMOND, A., 1844, Description de L’Ercin d’une dame Romaine trouve a Lyon en 1841. Paris

DALTON, O. M., 1912, Catalogue of the Finger-Rings; Early Christian, Byzantine, Teutonic, Medieval and Later. London: British Museum

HENIG, M., 1976, ‘A silver finger-ring from Winchester Wharf, Southwark’, Transactions of the London and

NOTES

Middlesex Archaeological Society 27, 256

HENIG, M., 1984, in Frere, S., Verulamium Excavation III, Monograph, no. 1, Oxford University Committee for Archaeology

HOFFMAN, H. and VON CLAER, V., 1968, Museum fur

241

Kunst und Gewerbe Hamburg Antiker Gold-und Silberschmiuck. Mainz

PUGH, R. B. and CRITTALL, E., 1957, Victoria County History of Wiltshire, volume 1, part 1. Oxford University Press

EXCAVATIONS AT VALE’S LANE, DEVIZES, 1996-7

by Phil Andrews and Lorraine Mepham

In 1996 Wessex Archaeology was commissioned by Lovell Partnerships (Southern) Ltd to carry out an archaeological excavation prior to redevelopment at the former Old Joinery Works, Vale’s Lane, Devizes (centred on SU 61370061). The report published here is asummary of the principal discoveries as financial support for a full programme of post-excavation work was not obtained. However, the pottery is considered at greater length as there is a dearth of material so far published for the town. Few finds other than pottery were recovered and further details of these, the archaeological deposits and the environmental remains can be found in the assessment report (WA 1996) which forms part of the site archive, shortly to be deposited at Devizes Museum.

The known archaeology and historical setting of Devizes have been summarised by Haslam (1976). The town is of medieval origin, first recorded as ‘Divisas’ in 1135, and the evidence available, especially the street pattern, indicates that it developed as a result of construction of the castle. The earliest known castle, which was burnt down in 1113, was replaced in c. 1120 by one of multiple bailey type (Figure 1a). This comprised a motte, inner bailey, and outer bailey, with an outer arc of streets and properties (the area refered to as Old Port) surrounded by a ditch and bank comprising the ‘town defences’. Haslam has argued that the ‘town defences’ were created at the same time as the castle in c. 1120, and integral with these was the construction of the earthworks surrounding the deer park to the south-east (Haslam 1980, 64-5). At some point, probably in the later 12th century, the outer bailey of the castle was abandoned and this set of defences was either removed or left to decay naturally. The town then developed within the former outer bailey (contra Butler 1976, 45), the area subsequently referred to as New Port.

With the exception of the excavations south of Hare and Hounds Street (Haslam 1980) and at New

Park Street (Russell 1993), recent archaeological work in Devizes has either taken the form of evaluations or watching briefs, or has not been fully published. Redevelopment of the Old Joinery, Vale’s Lane, provided an opportunity to establish the position of the outer bailey ditch and investigate the evidence for medieval and later settlement in this part of the town.

The redevelopment site covered an area of approximately 3800m? (Figure 1b), lay at c. 129.00m aOD and, at the time of the excavation, was covered with concrete slabs and tarmac. The underlying geology is Cretaceous Upper Greensand. An evaluation undertaken by the Cotswold Archaeological Trust in January 1996 (CAT 1996) revealed archaeological deposits of medieval and later date in the northern part of the site, but none were present or survived to the south. On the basis of the evaluation results the County Archaeological Officer requested that a further programme of archaeological work be undertaken.

The Excavation

The fieldwork strategy comprised the excavation of a single area measuring c. 15m by 10m (incorporating the majority of the northern evaluation trench) at the north end of the redevelopment area, and a watching brief during subsequent groundworks associated with the housing development. The excavation was carried out in October-November 1996 and the watching brief in February-March 1997. The archaeological features recorded are shown in Figures 1b and Ic.

Medieval features

The earliest feature was probably a large ditch (514), approximately 8m wide and up to 4m deep. It was aligned roughly north-south and lay towards the west side of the site, immediately to the west of the excavation trench (see Figure 1b). The location of this feature was indicated by a series of geotechnic boreholes prior to excavation, and was confirmed during the watching brief. Only a small part of this feature was revealed in plan and section and it was

Wessex Archaeology, Portway House, Old Sarum Park, Salisbury SP4 6EB

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

[] Medieval features [__] Post-medieval features

—-— — Probable kne of ditch s Recorded section

Approximate line of Outer Bailey ditch

Evaluation trench

4 Excavation trench Watching Brief Area

50m

Figure 1. Site Location.

NOTES

not possible to ascertain its profile. The upper fill was a homogeneous dark greyish brown loam containing post-medieval (18th — 19th century) debris, and it is suggested below that this feature was the outer bailey ditch.

The excavation revealed six shallow pits (29, 36, 51, 53, 67, 71), all less than 0.5m deep, which have been assigned on ceramic evidence to the 13th century. These pits varied in shape and size, with an apparent concentration of larger examples towards the east end of the trench. Towards the west end was a homogeneous spread of dark greyish brown sandy loam up to 0.3m thick which overlay the natural greensand and was indistinguishable from the fills of pits 51 and 53. Pottery recovered from the spread was also predominantly of 13th century date, but both it and several of the pits contained some residual 12th century sherds.

Three further shallow pits (24, 35, 48) were of 13th or early 14th century date. Also assigned to this phase was a short length of wall footing represented by a single course of small limestone lumps (8) aligned approximately east - west, and an oval, clay hearth (33) some 5m to the north of the wall footing. Two pits (58, 64) are the only features assigned a late medieval (15th — 16th century) date.

Post-medieval and modern features

The western two-thirds of the trench was partly covered by a large, shallow scoop (28), of probable 16th century date which had truncated the earlier features and deposits in this area. This was only partly excavated. Its fill comprised a generally homogeneous dark greyish brown sandy clay loam with a concentration of coarse shelly limestone fragments at the east end; these included at least one fragment of stone rooftile. Limestone building material recovered to the east of the excavated area during the watching brief may have derived from the same source, perhaps from a nearby medieval structure.

Five pits (4, 15, 17, 41, 80) and as many as five post-holes (9, 11, 13, 19 and 31) have been assigned a 17th century date, although few of the post-holes contained dating evidence. The post-holes were generally insubstantial, but may represent the corner of a timber building aligned approximately north - south. Pit 80 was a large, shallow, sub-circular feature which contained a concentration of stone building material, and pit 4 was deeper than any of the other pits on the site: it was sub-rectangular in plan, at least 1.5m deep (not bottomed) and had vertical sides.

The 17th century and earlier features were sealed by a substantial deposit of very dark greyish brown

243

sandy clay loam. This homogeneous deposit, most of which was removed by machine, was up to 0.7m thick at the west end of the site and appears to represent a well-mixed agricultural soil. This soil most likely reflects the existence of an orchard on the site from the 18th century until the construction of the joinery works in the 20th century.

The Pottery by Lorraine Mepham

The pottery assemblage from the Vale’s Lane site amounts to 224 sherds (4994 g) of medieval to post- medieval date. This has been analysed using the standard Wessex Archaeology pottery recording system (Morris 1994) by which fabrics are defined and coded on the basis of dominant inclusion type. In addition, some fabrics can be identified as of known type or source. Medieval vessel forms are defined following nationally recommended nomenclature (MPRG 1998).

Medieval

On the basis of known or suspected source/source area the medieval assemblage falls into six main groups:

1. sandy wares probably deriving from the Nash Hill kilns at Lacock

2. micaceous sandy wares probably from the Crockerton area

3. coarsewares and finewares of Laverstock type from south-east Wiltshire

4. calcareous and flint-tempered ‘Kennet Valley wares’ 5. calcareous wares from north Wiltshire (Nash Hill or Minety)

6. other miscellaneous wares

Predominant amongst the medieval assemblage are wares which can be tentatively identified as originating from the Nash Hill production centre approximately 10 km north-west of Devizes. Four fabrics were identified, although visual similarities suggest that all are merely variants of a single type (McCarthy 1974, fabric B):

Q401 Hard, moderately coarse matrix, slightly micaceous; common, fairly well sorted, subrounded quartz <0.5 mm (rarely <1 mm); handmade; generally oxidised with unoxidised core.

Q402_ Finer version of Q401 with quartz <0.25 mm (rarely <0.5 mm); wheelthrown.

244 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

‘Table 1. Fabric totals and diagnostic forms

Fabric No. sherds

Nash Hill

Q401 35 787 74.0

Q402 19 186 V7.5

Q404 2 64 6.0

Q405 2 26 2.5

sub-total 58 1063 - 36.0

Crockerton

Q400 10 139 61.8 Q403 5 86 38:2 sub-total 15 225 - 7.0 Laverstock E422a 1 24 4.6 E422b 6 152 29.0 E420 3 70 13.3 E421 1 16 3.) 11 262 50.0 sub-total 22 524 - Dh

‘Kennet Valley’

C401 31 734 98.7

F400 1 10 1.3

sub-total 32 744 - 25.2 N. Wilts

C400 3 48 14.9

C402 2 52 16.2

C403 4 222 68.9

sub-total 9 322 - 10.9 Q406 2} 54 - 1.8 ‘Tudor Green 2, 22 - 0.8 TOTAL MED 140 2954 - . E600 25 380

E601 55 1582

Verwood 2 2

Stoneware 1 4

Porcelain 1 1

TOTAL

POST-MED 84 2040

Weight % of group % of total med.

Vessel forms

tripod pitcher, frying pans (Fig. 2, 3), jars (Fig. 2, 1), glazed and white-slipped jugs

glazed and white-slipped jugs

glazed jugs

jar

frying pan, jars glazed jugs glazed jug

jars (Fig. 2, 2

glazed and decorated tripod pitcher

glazed tripod pitcher

glazed jug handled cup

includes slipwares dishes, bowls, jars

Q404 Hard, moderately fine, slightly micaceous matrix; sparse, well sorted, subrounded quartz <0.5 mm; rare limestone <0.25 mm; sparse iron oxides; ?wheelthrown; single example is unoxidised but might be overfired. Q405 Coarser version of Q401 with quartz <1 mm; also rare limestone <0.5 mm; handmade.

Vessel forms present are summarised in Table 1. The jars and glazed jugs are comparable to published examples from the 13th century kilns (McCarthy 1974), but the tripod pitcher is likely to be earlier. The type is generally dated to the 12th century, and an example found in a non-kiln context at Lacock

NOTES

was dated no later than the beginning of the 13th century (ibid., fig. 34).

Fabrics originating in the Crockerton area are characterised by their visibly micaceous clay matrix. Two fabrics are defined here:

Q400_ Hard, fine, micaceous matrix; sparse subrounded quartz <0.5 mm; rare subangular flint and irregular limestone <5mm; sparse iron oxides; handmade; generally oxidised (often pale-firing) with unoxidised core.

Q403 Hard, fine, micaceous matrix; moderate, well-sorted, subrounded quartz <0.25 mm; moderate iron oxides; wheelthrown; oxidised (often pale-firing) with unoxidised core.

These two fabrics can be compared respectively with fabrics D and H at Emwell Street, Warminster (Smith 1997, 20-1). Both have a lengthy currency from at least the 12th century at Warminster, but in general the finer fabric does not feature prominently until the later 13th century. Fabric D is predominant throughout the medieval period (ibid., fig. 13). Only one rim sherd, from a jar in fabric Q400, is present here; the form is not closely datable.

The Laverstock-type wares are represented by three fabric types, two coarse (E422a, E422b) and two fine (E420, E421). These are defined following the type series established for Salisbury (Mepham 2000) and are not therefore described in detail here. The finewares derive from glazed jugs and the coarsewares from jugs, with one frying pan also recognised.

Two fabrics, one calcareous and one flint- tempered, fall within a widespread ceramic tradition found across west Berkshire and north-east Wiltshire, and recently redefined as ‘Kennet Valley wares’:

C401 Hard, moderately coarse matrix, very slightly micaceous; sparse, poorly sorted limestone <1 mm; sparse, fairly well sorted, subrounded quartz <1 mm; sparse iron oxides; handmade; firing irregular.

F400 Hard, moderately coarse matrix; moderate, poorly sorted, subrounded quartz <0.5 mm; sparse, subangular flint <2 mm; rare limestone <1 mm; sparse iron oxides; handmade; irregular firing.

One kiln site is known for these wares near Newbury (Mepham forthcoming), but given such a widespread distribution more must still await discovery; one putative source has been identified on place-name evidence in the Savernake Forest (Vince 1997, 65). One sherd has been thin-sectioned as part of a small- scale programme of petrological analysis on these ‘Kennet Valley’ wares (see Mepham forthcoming), and

245

the sample from Vale’s Lane was found to differ quite significantly from samples from sites further east in the Kennet Valley in terms of matrix (coarser) and range of inclusions (relatively high content of shell but no flint). The potential date range of these wares is wide (see Vince 1997), but there is nothing amongst the vessel forms here (jars only) to suggest a date range later than the 13th century.

Three fabrics, all calcareous, may not derive from a single source, but all are likely to originate in north Wiltshire, where calcareous wares are known to have been produced at Nash Hill as well as at Minety and Lyneham (McCarthy 1974, fabric A; Musty 1973; Annable 1960).

C400 Hard, moderately fine matrix, slightly micaceous, slightly soapy feel; moderate, poorly sorted, irregular limestone <2 mm; rare quartz and iron oxides; handmade; unoxidised with oxidised internal surface.

C402 Hard, moderately fine, slightly micaceous matrix, soapy feel; moderate, fairly well sorted limestone (including ooliths) <0.5 mm (rarely <1 mm); rare subrounded quartz <0.5 mm; handmade; unoxidised with oxidised external surface.

C403 Hard, moderately fine, very slightly micaceous matrix; moderate, fairly well sorted limestone (including ooliths) <1 mm; very rare quartz <0.5 mm; handmade; generally oxidised with unoxidised core.

Sherds of glazed tripod pitchers occur in fabrics C400 and C402; the example in C400 has combed decoration; there are no other diagnostic sherds.

Two other fabrics were identified:

Q406 Hard, moderately coarse matrix; moderate, well sorted, subangular/subrounded quartz <0.125 mm; sparse limestone <0.125 mm; sparse iron oxides; handmade; unoxidised with oxidised (pale-firing) surfaces.

“Tudor Green’ ware: for full description see Pearce and Vince (1988).

Fabric Q406 is represented only by two sherds from one context, probably both from the same glazed jug. The source of this fabric is unknown, although it could be from the Bristol area, possibly Ham Green.

Post-medieval

The post-medieval assemblage consists largely of coarse earthenwares, mostly glazed. These have been broadly divided into wares which are at least partially unoxidised, often resulting in an olive-green internal glaze (E601), and wholly oxidised wares (E600). The

246 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

former are all probably Crockerton area products, and at Warminster tend to occur in earlier post- medieval contexts (pre-18th century), superseded later by the oxidised wares (Smith 1997, 29). A few examples of each type are slip-decorated.

Other wares are very scarce, and are represented by a handful of sherds of Verwood type earthenware, stoneware and porcelain, all of which are 18th century or later.

Ceramic sequence

It is difficult to identify any demonstrable sequence within the medieval ceramic assemblage, even when viewed against the background of the stratigraphic evidence, since most of the wares identified could have covered a relatively long timespan. A 12th century component, in the form of tripod pitchers, is certainly present, but all examples are found with 13th century or later wares. All the other medieval wares identified (with the exception of “Tudor Green’) would fit within a 13th century date range. The ‘Kennet Valley’ wares were certainly in use earlier elsewhere in the Kennet Valley (eg. Vince 1997, 64), but in this instance, apart from pits 51 and 53, they are associated with definite 13th century or later wares.

Wares which might extend the sequence into the 14th century include the finer glazed wares of Laverstock and Nash Hill type, although there are no closely datable forms present here, nor are there any forms comparable to the wasters found at Minety and provisionally dated to the 14th/15th century (Musty 1973).

It seems, then, that there is a real hiatus in the ceramic sequence between, at the latest, the early 14th century and the late 15th/16th century, which is very sparsely represented by a few sherds of “Tudor Green’ ware.

An early post-medieval ceramic phase can be defined (16th/17th century), dominated by Crockerton area products, in particular the partially unoxidised wares (E601). The almost complete absence of other wares suggests that there was little activity on the site after the 17th century, although some late deposits were removed by machining. Wares which can be certainly be dated later than this are restricted to two sherds of Verwood type earthenware, one sherd of stoneware and one of porcelain.

Discussion

The range of medieval wares identified at Vale’s Lane is comparable to that described for the site at New

Park Street (Gardiner 1993), and confirms the location of the town within the overlapping distributions of several pottery production centres. Potential 12th century wares probably derive from the nearest known source, Nash Hill, as well as other kilns elsewhere in north Wiltshire. From the 13th century the dominant local wares from Nash Hill are supplemented by a wider range of sources, including the Salisbury area (Laverstock types), the Kennet Valley, and the Warminster area (Crockerton types), as well as possible sources to the west, towards Bristol.

“"-! Green glaze 9 50mm a

Figure 2. Medieval Pottery.

List of illustrated vessels (Figure 2)

1. Jar rim, fabric Q401. PRN (Pottery Record Number) 50, context 34, hearth 33.

2. Jar rim, fabric C401. PRN 23/96, layers 6/66.

3. Frying pan handle, fabric Q401. PRN 30, layer 6.

Discussion

The excavation at Vale’s Lane, although limited in extent, has provided a welcome addition to the rather meagre archaeological evidence from Devizes. The outer bailey ditch was recorded in approximately the position postulated by Haslam (1980, fig. 3; Figure

NOTES

1b), although no controlled excavation was possible and the ditch could not be closely dated. The outer bailey ditch has been recorded on three other occasions since 1980 (Figure la): near to St John’s church to the south-west where part of the associated bank survived (IAFAU 1991); to the north-west beneath the Corn Exchange (WA 1994); and to the north-north-west between the Market Place and New Park Street (TVAS 1999). None of these sites produced any secure dating evidence, but all three indicated that the ditch was approximately 8m wide and 4m deep, similar in size to the outer ‘town ditch’ (Haslam 1980, fig. 2). The Corn Exchange site produced no evidence for the existence of a large ‘moat’ separating the castle inner bailey from the outer bailey as postulated by Haslam (1980, fig. 3). Instead, it appears that the outer bailey ditch may have continued around to form a complete circuit enclosing the kidney-shaped outer bailey. The recorded ditch fills on all three sites comprised mainly homogeneous slightly loamy sands, suggesting that the ditch had been regularly cleaned out, or little debris deposited in it, and that it may have been backfilled largely with bank material in a single operation. When the backfilling took place is uncertain, but it is unlikely that the ditch was maintained after the 12th century when the outer bailey fell out of use and the inner market place developed there. Very small quantities of pottery recovered from the top of the ditch at Vale’s Lane suggest that in places it may have remained as a slight hollow and not been finally infilled until the 18th-19th century. Indeed, its line appears to be still evident as a very shallow linear hollow in the graveyard belonging to the Baptist Chapel immediately to the north of the development area (Figure 1b).

The outer bailey ditch at Vale’s Lane would have served to demarcate properties along Sheep Street to the east from those along Long Street to the west. The excavated area lay to the rear of properties along Sheep Street, and thus outside the castle bailey and within the area of the original planned borough laid out in the early 12th century. No features of 12th century date were identified, although 12th century pottery was present as residual material in later features, and the majority of the medieval features comprised shallow pits of 13th century date. The nature and date of these features, and the limited structural evidence, is likely to reflect the location of the excavated area in backlands 70m or so from the street frontage. No property divisions were apparent, nor any evidence for specific crafts or industries, and the finds recovered are likely to reflect the disposal of domestic rubbish. The ceramic evidence indicates an

247

apparent hiatus in activity in this area between the early 14th century and the late 15th/ early 16th century. This was followed by a further phase of pit digging, with a possibly associated post-built structure, in the 17th century, prior to the establishment of an orchard in the 18th century.

Acknowledgements

Wessex Archaeology would like to acknowledge the assistance and facilities provided by the staff ot Lovell Partnership (Southern) Ltd, and in particular Mr Richard Cruse and Mr David Milner. The collaborative role of the Archaeology Section of the Libraries and Heritage Services, Wiltshire County Council and particularly Mr Roy Canham, County Archaeologist, and Mr Duncan Coe, Assistant County Archaeological Officer, is also acknowledged. We are grateful to the County Archaeologist and Mr Will Bryant, Housing Development Officer, Kennet District Council, for arranging grants from Wiltshire County Council and Kennet District Council which have enabled the publication of this report.

The project was managed for Wessex Archaeology by Niall Oakey, with the excavation carried out by Phil Andrews, Richard May, Angela Batt and Rosie Edmunds, and the watching brief undertaken by Nicholas Wells and Rosie Edmunds. The illustrations were produced by S.E. James.

References

ANNABLE, E.K., 1960, ‘A possible medieval pottery site at Lyneham’, WAM, 57, 403-4

BUTLER, L., 1976, ‘The evolution of towns: Planned towns after 1066’, in M.W. Barley, (ed.), The plans and topography of medieval towns in England and Wales, Council British Archaeological Research Report 14

CAT [Cotswold Archaeological Trust] 1996, ‘Old Joinery Works, Vales Lane, Devizes: Archaeological evaluation’, unpublished client report, CAT 96341

GARDINER, M., 1993, “The pottery’, in Russell 1993, 94-6

HASLAM, J., 1976, Wiltshire Towns: The Archaeological Potential. Devizes

HASLAM, J., 1980, ‘The excavation of the defences of Devizes, Wiltshire, 1974’, WAM, 72/73, 59-65

IAFAU (Institute of Archaeology Field Archaeology Unit] 1991, ‘An evaluation of the Norman outer bailey defences at Estcourt Hill, Devizes’, unpublished client report

MCCARTHY, M.R., 1974, “The medieval kilns on Nash Hill, Lacock, Wiltshire’, WAM, 69, 97-160

248 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

MEPHAM, L., 2000, ‘Pottery’ in M. Rawlings, ‘Excavations at Ivy Street and Brown Street, Salisbury, 1994’, WAM 93, 29-37 [this volume]

MEPHAM, L., forthcoming, ‘The medieval pottery from Enborne Street and Wheatlands Lane’, in V. Birbeck and C.W. Moore, Archaeological Investigations on the A34 Bypass, Berkshire/Hampshire

MORRIS, E.L., 1994, The analysis of pottery, Salisbury, unpub. Wessex Archaeology guideline 4

MPRG [Medieval Pottery Research Group] 1998, A Guide to the Classification of Medieval Ceramic Forms, Medieval Pottery Research Group Occasional Paper 1

MUSTY, J., 1973, ‘A preliminary account of a medieval pottery industry at Minety, North Wiltshire’, WAM, 68, 79-88

PEARCE, J. and VINCE, A., 1988, A dated type-series of London medieval pottery. Part 4: Surrey Whitewares. London Middlesex Archaeological Society Special Paper 10

RUSSELL, M., 1993, ‘Excavations at New Park Street, Devizes, 1990’, WAM, 86, 88-101

SMITH, R.W., 1997, Excavations at Emwell Street, Warminster: the early economy and environment of a Wiltshire Market Town. Salisbury: Wessex Archaeology

TVAS [Thames Valley Archaeological Services] 1999, ‘Land to the rear of Cromwell House and 33 Market Place, Devizes, Wiltshire’, unpublished client report, MPD 99/ 20

VINCE, A., 1997, ‘Excavations at Nos. 143-5 Bartholomew Street, Newbury, 1979’, in A.G. Vince, S.J. Lobb, J.C. Richards and L. Mepham, Excavations in Newbury, Berkshire 1979-1990. Wessex Archaeological Reports 13, 7-85

WA [Wessex Archaeology] 1994, ‘Corn Exchange, Devizes, Wiltshire: Archaeological watching brief’, unpublished client report, W687

WA [Wessex Archaeology] 1996, ‘Old Joinery Works, Vale’s Lane, Devizes: Assessment report on the results of the archaeological excavation including proposals for post- excavation analysis and publication’, unpublished client report, W2658

A MONKEY’S HEAD KNIFE FINIAL FROM NEAR TROWBRIDGE

by Paul Robinson

In the 16th century the style arose of applying to the end of the handle of a scale-tang knife a small decorative finial (or ‘knife cap’) generally made of copper alloy. The most common designs were those in the form of the head of a hammer (either a carpenter’s claw hammer or a shoeing hammer), a horse’s hoof or two addorsed horse’s hooves. These often include detail showing the horse-shoe and its nails. Complete knives with finials of this form are not uncommon. One example in Salisbury Museum with a horse’s hoof finial may have been made by a Salisbury cutler (Saunders 1986, 7). Another, found in London, is believed to be the work of a Flemish cutler (Hayward 1957, i3, pl.le). The use of finials in the form of a horse’s hoof remained popular with cutlers until well into the 17th century. Much less often the decorative finial was in the form of a human head, the head of an animal or a bird. When a human head is used, generally there is insufficient detail to show for certain whom the figure was intended to depict. Possibly female heads were intended to represent St Barbara, the patron saint of gunners, and male heads St Lawrence, the patron saint of cutlers. Terminals in the form of a bird or pair of birds may

Figure 1. Monkey-head knife finials from near Trowbridge (left) and Sussex (below)

be intended to depict a pigeon or dove. Those in the form of an animal’s head may depict what appears to be a horse’s head, while some certainly show the head of a monkey.

The style of the finials does not closely imitate the range of decorative finials in use at the same time on silver or base metal spoons. In the main, decorative finials were not used on spoons until well after the fashion of decorative knife finials had begun. It is noticeable too that some of the subjects of knife finials are light-hearted in nature, such as the head of a hammer and the monkey’s head, which contrast with the more formal subjects on 16th- and 17th-century spoons. It is possible too that the use of the designs of the horse’s hoof or the head of a claw-hammer may be to do with tapping the end of the knife on the table.

Perhaps not surprisingly the monkey’s head design

NOTES

of knife finial is among the least common found in England. The example illustrated is the first to be recorded from Wiltshire. It was found near Trowbridge in 1992 and acquired by Devizes Museum (accession number 1993.619). It is smaller in size than most other examples of the type and shows less detail of the head. For comparison, a similar knife finial, now in a private collection, found in Sussex, is shown. Both can be dated to c. 1500-1550.

It is doubtful whether more than a few people in Wiltshire in the first half of the 1 6th century would have seen a live monkey. They could, however, have seen images of monkeys on objects ranging from misericords to public house signs. They would know about monkeys from the Bible, from classical literature and from medieval writers and bestiaries. In medieval thought monkeys have been seen as symbols of evil, sin, particularly in relation to the fall of man, and of the

249

waywardness of Christian Europe. They were a ludicrous approximation of humanity, a prop for the derision of women, as well as a demonstration of the idiocies of all human endeavour. But to the common man then as in post-medieval times, the monkey was seen as a boisterous comical rebel, an unrestrained mockery of mankind which it so closely resembled, and a mischievous roguish creature of insatiable curiosity, which inevitably made it vulnerable to a smarter opponent. It is as such a creature it is shown on knife finials.

References

HAYWARD, J.F., 1957, English Cutlery 1500-1800. HMSO (Victoria & Albert Museum Catalogue)

SAUNDERS, P.R., 1986, Channels to the Past: the Salisbury drainage collection. Salisbury & South Wilts Museum

YET MORE ABOUT CUMBERWELL by Kenneth Rogers

This note describes the owners, occupiers and history of Cumberwell, a vanished house near Bradford-on- Avon, from the late 17th century. In the light of more recently available sources it supplements and corrects earlier accounts, published in 1950 and 1952.

In 1950 G. J. Kidston of Hazelbury wrote an article in this magazine about Cumberwell, which he described as ‘a quite insignificant and uninteresting place tucked away between Bradford-on-Avon and South Wraxall’. In 1952 he added new information in a further article ‘More about Cumberwell’.' Since then Cumberwell has developed a higher profile as a golf club and restaurant. It is now possible to add more to what Kidston knew, mainly from papers of the Clark family in the large deposit made by Mann and Rodway of Trowbridge in the Wiltshire and Swindon Record Office.? Rather than trying to fill gaps in Kidston’s articles it seems best to present a narrative from the late 17th century.

The Button family of Alton, later of Tockenham Court in Lyneham, had acquired Cumberwell in 1530.’ Sir Robert Button, the second of three brothers who in turn succeeded to the baronetcy, died c. 1679 without issue, and left Cumberwell by will to his

nephew Charles Steward, the son of his sister Jane, who had married Richard Steward, Dean of St Paul’s and Westminster. At that time it was described as a farm and lands occupied by Robert Foote.’ Steward moved to Cumberwell, and was probably the builder of a mansion house of some size on the estate. By 1691 he was married to Mary Compton of a family seated at Hartpury, Gloucestershire; his uncle had married one of this family too. Steward died in1698 as a result of injuries received from a fall from his horse, as is told on the elaborate memorial, featuring a full-length figure of the dead man in the costume of the time, which Mary set up in the chancel of Bradford church.’

The couple left no issue, so Mary had only a life estate with remainder to John Walker of Hadley, Middlesex and his heirs. He had married Mary, another sister of Sir Robert Button who had left it to Charles Steward. John Walker died in1703, probably before Mary Steward, whose death has not been found. By 1718 it was in the possession of John’s son Heneage Walker, who executed a deed of settlement which names no occupier of the mansion house but describes the farm of 265 acres as occupied by John Newton. The effect of this would probably have resulted in Cumberwell passing in due course to Heneage’s brother John, the ancestor of the Walker- Heneage family of Compton Bassett, for Heneage was unmarried. However in 1721 he revoked the settlement, evidently with a view to selling Cumberwell, which he did in 1723.°

250 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

The purchaser was John Cooper of Trowbridge, who paid £2,925 for a property (fully described for the first time in the deed) consisting of the mansion house, a house in Little Cumberwell, and about 230 acres of land. He had made money as a clothier, as had his father Thomas, and Thomas bought the manor of South Wraxall, adjoining Cumberwell, in 1722. Other purchases in the same area followed.’ They owned the large and stately house in Fore Street, Trowbridge now occupied by Lloyds Bank, and a few letters from John in the last years of his life in the late forties show that he moved between there and Cumberwell. Racked by gout so that he sometimes had to be carried by two men, he went from Cumberwell to Trowbridge in February 1747, a time of wet and unwholesome weather, and after a fortnight there felt much better for the change of air and the company.

John Cooper died early in 1749, and Cumberwell passed to his son Thomas, who had married Frances Bathurst of Clarendon Park. After his father’s death Thomas wrote to his sister, Ann Fleming:

I have (tis true) arrived to a very good estate, but then

tis all entailed on my eldest son and I am only a tenant

for life so that my daughters’ fortunes must depend on what I can save out of my estates. If my poor father had

thought proper to have shewed me his will I believe I

shou’d have prevailed with him to have altered it in many

particulars, but he was so Reserved that I never knew

the contents of it till after his decease. In these circumstances it was essential for Thomas to live to a good age, but in fact he died at the age of about 46 in 1756, leaving the estate to bear the costs of a widowhood until Frances’s death in 1779. Daniel Clutterbuck, the Bradford lawyer, wrote to Thomas’s brother-in-law, Edward Fleming, ‘Mr. Cooper’s concerns are so circumstanced as to render it necessary to call in all the debts’. Fleming did not pay up, and in 1758 Frances had to tell him, ‘Our affairs are left so that it is unavoidable . .. I must not let my children hazard so considerable a loss .. ., Life is very precarious’.®

The next heir was Thomas’s son, John Cooper, who married Mary, daughter of Edward Baynton of Spye Park in 1759.’ He was living at Cumberwell in 1762 while his mother appears to have moved to Salisbury. In 1763 his huntsman, very much in liquor, was killed by being thrown from a mettlesome young horse of his master’s between Bradford and Cumberwell. In 1765 John wrote from London to his uncle Fleming that he could not meet him in town because he had to be at the races at Newmarket. In 1766, when he died, aged about 30, the estate was mortgaged for over £6,000.'°

Mary Cooper re-married in the same year. Her new husband was a captain in the Royal Volunteers in which her first husband had been a lieutenant; he was Charles Cooper, an illegitimate son of Henry, Lord Holland, and not known to be related to the Cumberwell Coopers. On their marriage she made arrangements for financial provision for her only daughter, Frances Sarah; her failure to honour them led her and Charles into what appears to have been an acrimonious dispute with her father and brother culminating in a Chancery cause in which they acted as ‘next friends’ to the girl."! Cooper may indeed have been an unsuitable man to be step-father to her, as is known from the memoirs of William Hickey. He first met Charles and Mary when they were living at Sheerness and sailed with them on their yacht The Porpoise. Later entries depict Charles as a dissipated gambler and drunkard.

What was happening at Cumberwell meanwhile? Andrews and Dury’s map of 1773 shows Sir Edward Bayntun as occupier. The significance of this is not clear; he may well have been acting as guardian of the next heir, his grandson, John Allen Cooper, but is hardly likely to have set up an establishment for a boy of about 12 away from his own seat at Spye Park.

Nor is it likely that he fitted the house with new furniture, which happened in c. 1776, as the following advertisement!” makes clear:

BRADFORD, WILTS.

To be SOLD BY AUCTION by HENRY HILL, on

Wednesday the 24th of September instant, and the

following day._The HOUSEHOLD FURNITURE of

CUMBERWELL HOUSE, removed for the

convenience of sale, to a commodious room at the Green

Dragon in Bradford aforesaid; consisting of handsome

four-post bedsteads, with silk, mohair, cotton and other

furnitures, and window curtains, fine goose feather beds, mattrasses, blankets, quilts, and counterpanes, Wilton and Turkey carpets, mahogany chairs, and a set of drawing room armchairs, dining, card, and Pembroke tables, chests of drawers, oval, pier, and other glasses, prints, cast iron Bath stoves, and other useful furniture.

The whole was put in new about twelve months since,

and exceedingly well kept.

The whole may be viewed on Tuesday the 23d, and each morning of the sale, which will begin at three

0’ clock.

Catalogues will be delivered in due time, at the

King’s Arms, Devizes; Woolpacks at Trowbridge; at the

place of sale; and of Henry Hill, upholder, in

Marlborough.

*** The room is upwards of sixty feet long, in which seats will be compleatly fitted up for the company.

It seems more likely that this furniture was put in by a tenant whose term ended abruptly in some way.

2511

NOTES

its demolition

r10r to 1ts

Cumberwell in 1903 p

to Ui bo

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

He

Cumberwell in 1903 prior to its demolition

The heir, John Allen Cooper, was a soldier who had just returned from service in America in May 1781, when he was almost of age, and was than at Spye Park. His mother, Mary, had died the previous year.'’ It seems possible that he moved into Cumberwell for a time, for another advertisement for furniture" sheds light on the situation:

BRADFORD

TO BE SOLD by AUCTION, by WILLIAM PITMAN,

on Thursday next, the 18th instant, and following days,

The HOUSEHOLD FURNITURE, Linen, China,

Glass, and other Effects, at Cumberwell House, near

Bradford; consisting of bedsteads, with cotton, check,

and other furniture; feather beds, blankets, quilts, and

counterpanes ; the Mahogany articles consist of tables, chairs, night stools etc large marble slab in mahogany and other frames; handsome pier glasses in carved and

gilt frames; carpets, an excellent eight-day clock in a

japan case; an elegant time-piece (by Evil and Co.) a

neat mahogany sopha covered, and festoon window

curtains of the same; a billiard table complete; Bath stove and other grates; smoke and other jacks, with all kinds of useful kitchen furniture, brewing utensils, etc.

N.B. A quantity of exceeding good HAY will be sold the second day of the sale. The whole may be viewed the day preceding the sale, when catalogues may be had at the place of sale ; at the George and Woolpack, Trowbridge; King’s Arms, Melksham; the New Bear, the Printing-Office, and of W.Pitman, Auctioneer and Undertaker, Bradford. To begin each day precisely at eleven o’clock.

In 1786 the Cooper property in Bradford and

South Wraxall was put up for sale by auction.'? Lot 1 was:

Cumberwell House; containing on the ground-floor two handsome parlours, a breakfast parlour, and study, a large kitchen and all other proper offices for servants; on the first floor, a handsome drawing room with bed- chambers and dressing-rooms; and good rooms in the attic story; with stables, coach-houses, a walled garden, a rich piece of pasture ground, and a spacious lawn in the front, containing about five acres, with an an inclosed park, containing about thirty-three acres, well stocked,with deer.

With the house went a farm of 230 acres let to Thomas Gerrish, who lived at Little Cumberwell.

NOTES

The house was bought by Robert Taunton LL.D., a clergyman in his forties who held the livings of Sydling St Nicholas, Dorset, North Perrott, Somerset, and Alton Barnes. He had apparently resided at Sydling, where he improved the vicarage, until he came to Cumberwell.'® Kidston quotes a letter from Eliza Purbeck describing a visit to Cumberwell, probably in 1787:'’ We returned to Bath... after spending a week at Cumberwell Park, Dr. Taunton’s new purchase; from its situation it must be in the summer a very agreeable residence ... The prospects round it are beautiful... but the house is old, and too large to be comfortable; there was something very gloomy in the idea of a number of uninhabited rooms, which, large as their family is, you will imagine to be the case, when I tell you that there are more than thirty apartments in the house. The Doctor is at present undetermined what he shall do with this great pile of building; he sometimes talks of dividing it, at others of building a new house in the Park. This implies that the house was larger than the description of 1786 suggests. Perhaps there was an older and less genteel part? - However, the Doctor’s problem as to the size of the house was solved on 9th December 1790:'*

259

Thursday morning, 9th, very early, a dreadful fire broke out at Cumberwell House, near Bradford, the seat of the Revd. Dr. Taunton which raged with such fury that Mrs. ‘Taunton and her little family with great difficulty escaped with their lives and with scarcely any clothing; the conflagration was soon general, and totally destroyed the house, except one wing, with nearly all the furniture, plate, cash, notes, books, etc. to the amount of nearly £5000 no part of which was insured. Thomas Underwood the butler, a faithful servant was lost in his attempt to save a second maid-servant, after having brought one safe out of the flames. Two female servants were greatly hurt by jumping out of the window, one having had her thigh broke. This melancholy catastrophe is said to have begun in the laundry, where some linen took fire that was hung up to dry. Kidston had heard only a tradition about a fire, which lay the blame on a drunken butler, so it is pleasing to be able to rescue the reputation of the heroic Thomas Underwood. Affidavits made many years later about the loss of the deeds of the property in the fire assert that Taunton did not live in the house again, but in fact at least part must have been reinstated, for he died there in 1797.'° His heir was his son William Leonard Thomas Pile Taunton, who was still in his teens. However, it seems certain that the Tauntons soon left, for in 1800 a lease was made to a Mrs. Moncaster.

Cumberwell in 1903 prior to its demolition

254 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Cumberwell: the interior in 1903 prior to 1ts demolition

In 1802 the whole estate was put up for auction in London. The description of the house shows that what was described in 1787 had been restored completely. The stone-built house was described as ‘exceedingly well fitted up and furnished in a neat genteel manner and in perfect and compleat repair’. On the attic floor were six chambers and a seed room, on the first floor four bed-chambers and an elegant drawing room, 28ft. by 18ft., and on the ground floor dining room, library, parlour and entrance hall. There were stables for 11 horses, coach houses, dove-cote, walled kitchen garden, fish ponds, and various pleasure grounds.””

No sale was made, and the Tauntons let it for a number of years. After Mrs. Muncaster we know of Mary Ann Rundell and Charles Ponsonby Butler as tenants, then in 1820 a lease was made to Samuel Staples. He was of a banking family in London.

Finally in 1832 W. L.T. P. Taunton, then of Stoke Bishop near Bristol, sold the Cumberwell estate to John Clark of Trowbridge, clothier. He was of the firm of J. and T. Clark, and lived at Bellefield House in Hilperton Road, Trowbridge. Cumberwell House was still let to Staples at that time.”

In 1836 it was said that Clark farmed the land but did not reside,?*> and no more is known of the mansion house being occupied. Kidston was told that when Dorcas, widow of Thomas Clark, sold the property to Erlysman Pinckney in 1903 the house was a ruin unsafe to enter, and this is borne out by the photographs reproduced here. It was then pulled down and the materials, Pinckney’s son remembered, were used to build some farm-workers’ cottages. However, it will be seen that the door and window cases on the photographs are numbered and lettered, so it looks as though they were intended to be sold.

The house was built of unsquared stone laid in courses and covered with stucco on the front and one

side. The front facade shows only two storeys and attics, but the ground against it was some feet higher than on the other three sides, which reveal a basement level: Kidston was told there was a basement kitchen. The window and door features all suggest a late 17th- century date, which agrees with Charles Steward being the builder. So, too, do the gate piers now at Avebury Manor. The extra rooms which existed until the fire probably formed a wing of which the remains can be seen on the right of the back frontage. Why the Clarks let a house of standing and evidently in reasonable condition in 1831 fall to ruin is a puzzle. Kidston suggested that it was haunted!

Notes

1. WAM 53, 471-485, and 54, 279-288. This article is concerned only with the mansion and does not attempt to trace the history of the farms on the estate.

WRO 2153; at present only partially catalogued, and no

final item numbers allotted.

. WAM 53, 481.

4. G.E.C., Complete Baronetage ;\WWRO 2153, abs. title; W. H. Jones and J. Beddoe, Historv of Bradford-on-Avon (1907).

. Ibid.

. WRO 2153, abstract of title.

. WRO 947/1371, 1378, 1380.

. Alistair Rowan, ‘Sibdon Castle, Shropshire’, Country Life 1 and 8 June, 1967, which gives more detail of the Cooper- Fleming alliance. The letters, then in the possession of Major H. Holden of Sibdon Castle, were kindly shown to me in 1967.

9. Gentleman’s Magazine, 1 Nov. 1759.

10. Sibdon Letters; Hunnisett, R.F. (ed.) Wiltshire Coroners’

Bills 1752-1796 (Wilts. Rec. Soc., vol 36), no 360.

11. Gerald Hamilton-Edwards, ‘Should Fanny have been sent to France? Revelations of a Chancery Proceeding’, Genealogists Magazine, vol. 16, 10-14.

12. S(alisbury and Winchester) J(ournal), 22nd Sept. 1777.

13. Sibdon letters.

14. SJ, 15 Aug 1785.

15. Ibid:, 12 June 1786.

16. WRO 2153, deed; J. Hutchins, History of Dorset, (3rd. ed.), vol. 4, pp. 507, 510; J. Foster, Alumni Oxonienses, 2nd series, 1390

17. Quoted at more length by Kidston, WAM, 54, 280.

18. SJ, 20 Dec 1790.

19. WRO 2153; SJ, 24 July 1797.

20. WRO 947/1363.

21. WRO 2153.

22. WRO 866/12.

23. WRO 2153, case and opinion about rights of way. This, incidentally, shows that Robert Taunton had sold a lot of stone from an old quarry on the site. Two quarries, one being worked, are mentioned in the sale particular of 1802.

i)

Qo

OANA WYN

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 255-264

Excavation and Fieldwork in Wiltshire 1998

Alton

Sewerage Scheme (centred on SU 108 621); Romano- British to Medieval

A programme of archaeological work, comprising excavation of selected lengths of pipe-trench and a watching brief, was undertaken by AC archaeology in conjunction with the construction of Wessex Water’s Honeystreet and Alton Barnes/Alton Priors Sewerage Scheme. Existing archaeological records, pre- construction desk-based study, earthwork survey and geophysical survey previously defined an archaeological interest. Prehistoric and Roman finds are present within the area; the settlement of Alton Barnes and its parish church, St. Mary’s, are believed to have Saxon origins; and significant lengths of the pipeline route were seen to cross or pass adjacent to earthworks believed to relate to medieval settlement at Alton Barnes and Alton Priors. Some 57 ditches and 15 pits were recorded, the majority of medieval date (probably 13thcentury). Throughout the course of the investigations, it was noted that the layout of the earthworks (either plotted or as visible on the ground) did not correspond with the location or orientation of the excavated (medieval) ditches. Finds included Romano-British and late Saxon pottery, both in limited quantities.

Amesbury

DERA, Boscombe Down (SU 145 412); Late Iron Age or Romano-British

Wessex Archaeology undertook an archaeological watching brief during the demolition of two buildings within the base at Boscombe Down, Amesbury. A number of archaeological features were visible in the chalk beneath the buildings, indicating that similar remains may be preserved under standing buildings elsewhere within the base. Twelve small pits and post- holes were excavated, but only one yielded any dating evidence, pottery of Late Iron Age or Romano-British date. The dating of this one pit is consistent with the evidence for extensive occupation at this time at the adjacent site of Butterfield Down.

Boscombe Down Airfield; Romano-British to Modern Fifteen watching briefs and two excavations were carried out at the Defence Evaluation and Research Agency (DERA) site at Boscombe Down during 1998. The watching briefs were carried out during a number of development projects as part of the on-going assessment of the archaeological potential of the airfield. These recorded mostly modern or natural features. However, two produced archaeological features that were excavated and fully recorded. The first (at SU 1834 4088) was a ditch or field boundary. No dating evidence was recovered, but a small amount of weathered human bone was present. The second (at SU 1862 3950) produced evidence of a multi-phase field boundary. The ceramic assemblage from this ditch indicated a lst to 3rd century AD date. Both excavations were managed by Bob Clarke, whilst the watching briefs were carried out by Bob Clarke and Colin Kirby for Boscombe Down Conservation Group.

Queensbury Bridge (SU 152 413); Post-Medieval Two trial holes excavated for engineering assessment on Queensbury Bridge (built 1775) were monitored by AC archaeology in accordance with the requirements of a scheduled monument consent. It had been anticipated that the trial holes might encounter early surfaces but, although various (undated) flint and sandy mortar make-up layers were encountered, no detail of the bridge construction was observed.

72 London Rd (SU 1584 4188); Modern

An archaeological watching brief was undertaken by Wessex Archaeology during residential development at 72 London Road, immediately south of Ratfyn Barrow, a Bronze Age burial mound and Scheduled Monument (Wiltshire No. 28931). No archaeological features or deposits were recorded, but as the foundation and service trenches represent c.5% of the total site area there remains a high likelihood that other archaeological features are present on the site.

Former Pitt’s Garage site (SU 1550 4156); Modern Wessex Archaeology was commissioned by Primary

256 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Health Care Centres (Amesbury) Ltd to undertake an archaeological evaluation of an area of land at the former Pitts Garage site, Amesbury, situated within the core of the town. No archaeological features or deposits were noted.

Avebury

National Trust Estate (centred on SU 102 699); Prehistoric, Medieval and Post-Medieval

Since 1995 National Trust staff have been recording all ground disturbances resulting from work carried out by staff and contractors on the property. Surface artefacts collected by the staff (both targeted and casual finds) have also been recorded as archaeological interventions. A total of 65 were recorded in the period from September 1995 to December 1998. A full list of these and their archive reports are held at the Alexander Keiller Museum.

Nearly half of all interventions were recorded when excavating pits for fence posts and sign posts. Only two of these interventions have produced worked flint, and only post-medieval and modern disturbance has been encountered. Four redundant display panels have been removed from the henge, and although there are no records of their post placement pits having been recorded when erected in the 1980s, it appears they were sited in areas of post-medieval and modern disturbance.

A number of service trenches have also been excavated and recorded around the property, including two sewage pipe trenches within the henge itself. Both revealed considerable post-medieval and modern disturbance, and no prehistoric features were encountered. Other service trenches east of the henge and along the winterbourne have demonstrated the nature of the sometimes difficult geology around Avebury.

The erection of signs and improvements to the surface of the southern car park have continued to produce evidence for medieval activity west of the henge. A small quantity of residual medieval pottery has been collected from the topsoil and further evidence for the medieval ploughsoil, observed during excavations by the Wiltshire Rescue Archaeology Project in 1988, has been recorded.

The Winterbourne Team Rector, Warren Sellars, has kindly given permission for the archaeological staff to observe and record excavations for graves within the churchyard of St. James, Avebury. Two possible early features, a post-hole and a linear feature, have been recorded outside the area of the (now flattened) henge bank. Although no artefactual dating evidence

was recovered from them, they do predate the earliest grave cuts, and may relate to the Saxon settlement or prehistoric activity.

Surface artefact collection involved the recovery of prehistory pottery and worked flint prior to erosion repairs to the Overton Hill barrow cemetery, and the chance find of a barbed and tanged arrowhead near to the agger of the Roman road.

11 Kv Cable refurbishment (SU 098 690); Prehistoric to Post-Medieval

Wessex Archaeology was commissioned to carry out an archaeological watching brief during the refurbishment of parts of the 11 Kv supply network within the World Heritage Site at Avebury, Wiltshire. The observed works included the excavation of open-cut trenches for replacement cables and new sockets for replacement poles, along with the removal of existing poles in areas of archaeological sensitivity. The works occurred in both privately owned and National Trust land, but did not cross any area currently designated as a Scheduled Monument.

Medieval and post-medieval material was recovered from topsoil/subsoil deposits in the vicinity of Avebury Trusloe, and prehistoric and Roman material from ploughsoil deposits to the south of West Kennet. In Butler’s Field, to the east of Avebury Trusloe, material was recovered consistent with the suggestion that this was formerly a post-medieval water meadow. A large, buried sarsen with associated medieval pottery was noted to the north of this field. Archaeological observations by the National Trust as part of this project resulted in a low-density recovery of finds representing prehistoric, Romano-British and Saxon activity.

Blunsdon St Andrew

Groundwell Farm (SU 1513 8902); Roman and Medieval

Trenches cut during an evaluation, carried out prior to selling land for development, revealed a probable Romano-British cambered stone road and traces of 13thto 15thcentury settlement. Worn into the road surface were a number of wagon ruts. Nearby, a vast bowled depression cut into the Corallian escarpment and extensive linear features indicate that stone quarrying had taken place. This activity may have been the reason for the road’s existence. A trench cut into the top of the bowled depression produced a few unworn Romano-British sherds, a box tile fragment and many medieval sherds. The medieval

EXCAVATION AND FIELDWORK IN WILTSHIRE, 1998

settlement remains included wall traces, yard surfaces and ditches. The work was undertaken by Bernard Phillips.

Bishopstone (north)

Bury Mill, Hinton Parva (SU 2264 8540); Roman and Medieval

Linear features noted on an aerial photograph suggested the presence of a deserted medieval village. Examination on the ground revealed that the area adjacent to a stream had been ploughed since the photograph was taken. Fieldwalking produced 13th and 14th century sherds, animal bones and building stone. The hedgerow either side of the stream also contains many sarsen blocks, presumably removed from the field during ploughing. Two Romano-British sherds were also found. The work was undertaken by Bernard Phillips.

Bishopstrow

Home Farm (ST 895 444); Late Mesolithic/Early Neolithic and Post-medieval

Machine-trenching by Asi revealed parallel linear ditches of post-medieval date and scatters of late Mesolithic/early Neolithic worked flint. Whilst the linear ditches are not remarkable, the flint assemblage indicates the survival here of an episode otherwise poorly represented within the chalkland landscapes of the south Wiltshire downs, potentially a tool manufacturing site or seasonal camp, and is therefore of considerable archaeological significance. Comparison with lowland Mesolithic sites revealed in the adjacent Kennet watershed suggests that this material is likely to be distributed for some distance across the site.

Bradford-on-Avon

The West Barn, Barton Manor Farm (ST 8230 6045); Medieval

Compilation of archive and excavation data, together with an historic building survey conducted by Archaeological Site Investigations (Asi) and the Bradford on Avon Preservation Trust, established that the ruinous West Barn adjacent to the Great Tithe Barn represents the remains of a much modified, and unparalleled, agricultural structure, possibly based upon components of a 13th century monastic grange farm. The building, extant until a fire in 1982, is now in a hazardous state of dynamic collapse.

257

Brixton Deverill

The Manor House (ST 8640 3865); Undated

A single machine-excavated trench was located across elements of a network of hollow-ways and house platforms following detailed earthwork survey. This revealed an intermittent linear depression within the soliflucted chalk, containing animal bone and fragments of Red Pennant sandstone. It is concluded that, although no longer visible in detail due to recent topsoil dumping, earthworks relating to the medieval village of Bristicii do extend across the rear of the Manor House gardens. The work was undertaken by Asi.

Broad Chalke

Water main relining works (centred on SU 041 253) Monitoring was undertaken by AC archaeology during groundworks to facilitate water mains relining. The works involved the excavation of several access pits sited at various locations around the village. Observations revealed no archaeological deposits or finds.

Broad Town

Mesolithic and Medieval

Two surface artefact collections were carried out during 1998 by Bob Clarke. These were part of an ongoing project into the development of the village of Broad’Town. The first (centred on SU 0960 7827) identified a concentration of Mesolithic worked flint, comprising blades, blade cores, and scrapers, along with burins. The second (centred on SU 0988 7793) located a large concentration of 13thto 14th century pottery including examples from Minety, Wootton Bassett and Naish Hill.

Chippenham Without

Sheldon Manor (ST 8865 7414); Medieval

An archaeological investigation at Sheldon Manor entailed an earthwork survey at a scale of 1:1000 of the site of a deserted medieval settlement, and a detailed investigation of the remainder of the manor. The work is part of a personal research project on the Hundred of Chippenham being undertaken by Graham Brown.

The earthworks are contained within three fields to the north and north-west of the manor house. In the latter are the remains of a deserted settlement, the most prominent feature being a

258 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

hollow way (210m in length and up to 8m wide) that extends in an east-west direction from Sheldon Wood. This can be traced further west in Corsham Wood where it merges with the woodland edge and probably continued to Biddestone. Sited along the northern side of the hollow way are earthwork remains of up to ten sub- rectangular building platforms, ranging in size from 5 x 5m to 17 x 10m. Covering much of the field to the north of the hollow way is a swathe of what appears to be ridge- and-furrow, but is more likely drainage, probably cut in the early 16th century. The majority of the furrows are sharply incised, particularly near the building platforms.

To the south of Sheldon Wood, three ditches extend in a southerly direction with traces of further probable building platforms on the east side of the field. Another hollow way extends in an east-west direction from Sheldon Wood and can be traced for some 100m: it is aligned on a modern field boundary that ultimately leads to Allington.

Corsham

Heywood Preparatory School, Priory Lane (ST 8723 7062); Medieval

During groundworks associated with construction of a new building at the rear of Heywood Preparatory School, observed by Asi, the northern terminal of a 2m broad ditch was revealed, sealed beneath a sterile clay subsoil. Unweathered sherds of 10th-13th century pottery were recovered from the charcoal- rich fill. The school occupies the site, and many of the buildings, of the post-Dissolution ‘Rectory Manor’ of Corsham, itself the remains of a short- lived Benedictine Priory founded in the 12thcentury. It is likely that the features revealed relate to the Priory.

Cricklade

Horse Fair Lane (SU 1017 9376); Medieval and Post- Medieval

Watching briefs and a small excavation were undertaken by Bernard Phillips prior to house construction. These revealed an occupation layer, ditches and traces of two buildings. Associated pottery attests to occupation from the 10thto the 1 6thcentury. One building was sunken floored and the other had a paved stone floor and a drainage ditch on its south side. Later occupation, contemporary with the former Three Horseshoes Inn, in whose garden the construction site lay, included an

early 18th century cess pit and a late 19th century stone lined well.

High Street (SU 1012 9383); Roman and Medieval

A watching brief undertaken by Bernard Phillips in advance of house construction, revealed a 2nd century occupation layer. Overlying this were traces of two buildings floored in clay and dated by pottery to the 11th century. A hearth sealed by the floor of the eastern building suggests that it had an earlier phase. The other building’s floor preserved traces of intense burning in association with iron slag. Pottery fragments show that occupation on the site continued into the 13thcentury.

Devizes

Wayside Farm, Nursteed Road (SU 016 603); Romano-British

An evaluation of a proposed residential development was carried out by AC archaeology. The site is located to the southeast of Devizes, adjacent to Wayside Farm, and covers approximately 7.2 hectares. Work initially comprised the machine-excavation of 15 trenches, each 30 x 2m, plus an additional 90m’ contingency trenching excavated to pursue specific features. The trenching revealed extensive evidence for Romano- British occupation, including stone structures, possibly ovens, cut features, and evidence for a buried soil horizon containing significant quantities of Romano-British artefacts. Finds indicate a mid 4th century emphasis for the settlement activity. Further excavation in advance of development is proposed for 1999,

Downton

Land adjacent to 136 The Borough (SU 1793 2152); Medieval and Post-Medieval

A machine-excavated trench revealed pits and post- holes from which medieval and post-medieval pottery, building materials, iron slag or clinker, nails and animal bones were recovered, cutting through a sequence of alluvial soils, the lowest of which contained only medieval pottery. Though situated within the medieval core of Downton and containing residual materials, the activities represented by the bulk of the deposits appear to be of post-medieval date and of uncertain, though non-domestic, function. The work was undertaken by Asi.

EXCAVATION AND FIELDWORK IN WILTSHIRE, 1998

Heytesbury and Imber

Park Street Gates, Heytesbury (ST 9315 4265); Iron Age and Medieval

Machine-excavated trenches, located over the footprints of two proposed dwellings adjacent to the former Park Street gates of Heytesbury House, revealed ditches, gullys, post-settings and other sub- soil deposits from which pottery of Iron Age and early to late Medieval date was recovered, along with animal bone and burnt stone. The results of the work suggest that the site overlies archaeological deposits associated with the medieval precursor of Heytesbury, possibly re-located after landscaping works associated with Heytesbury House and its parkland gardens, and hitherto unsuspected prehistoric activity. The work was undertaken by Asi.

Imber Village Silt Lagoons, SPTA (ST 965 486); Medieval, Post-Medieval and Modern

Observations by Asi maintained during construction of silt lagoons at Imber, within the Salisbury Plain Training Area, recorded residual medieval pottery from within a range of post-medieval and modern deposits. Whilst construction work does not appear to have had a deleterious archaeological impact itself, the results suggest that in situ deposits representative of the medieval village may well survive elsewhere within the vicinity of the site.

Latton

Latton Lands (SU 0800 9670); Bronze Age and ?Medieval

A one hectare area was stripped by Cotswold Aggregates in December 1998 and January 1999 in the north-east corner of the gravel extraction area. The Oxford Archaeological Unit monitored the machining as part of the watching brief and recorded a stream course and a waterhole. These contained no dating evidence but are likely to be medieval. Subsequent excavations encountered a trackway and several ditches, possibly early medieval, and probable Bronze Age pits and waterholes. Excavation continues in 1999.

Lakeside, The Street (SU 0922 9544); Post-Medieval

Evaluation by the Cotswold Archaeological Trust (CAT) located the foundations of a post-medieval building fronting The Street, and revealed that the majority of the site had been subject to previously unrecorded gravel extraction.

Maiden Bradley

Church Street and High Street/Back Lane (ST 8020 3878 and 8045 3918); Post-Medieval and Modern Hand excavated test pits, located over the footprints of proposed new dwellings, revealed deposits and features of post-medieval and modern date, likely to be the by-product of recent gardening activities, but containing relatively large quantities of residual medieval pottery. The work was undertaken by Asi.

Marlborough

Waitrose, High Street (SU 1885 6905); Medieval and Post-Medieval

An evaluation was undertaken by Wessex Archaeology at the rear of the Waitrose supermarket within the medieval town of Marlborough. Two evaluation trenches revealed sequences of post- medieval deposits to a depth of at least 1.4m below existing ground levels. Soliflucted chalk was sealed below a series of soil, yard, and make-up layers and modern disturbances. A small quantity of residual medieval material was recovered, although the stratigraphically earliest deposit, possibly a cobbled surface, contained pottery of 15th or early 16th century date.

A homogeneous soil layer was recorded in the second trench. It was at least 1.4m thick, contained post-medieval material, and is consistent with the agricultural use of an area some distance from the street frontage.

Chandler’s Yard (SU 1875 6925); Post-Medieval Evaluation by CAT identified a make-up layer dating to the post-medieval period and two post-holes probably of the same date. No evidence for the late Saxon or Medieval town was found.

Mildenhall

‘The Bothy’, Werg Mill (SU 2145 6955); Romano- British

An archaeological evaluation was undertaken by AC archaeology on the site of a proposed house extension at ‘The Bothy’, Mildenhall. The location of the evaluation was very close to the known north-western perimeter of the Roman town of Cvnetio. The evaluation comprised a small trench some 5.5m? in area, dug within the confines of the extension. This demonstrated the presence of deeply stratified Roman deposits, probably dating to the late 2nd century. These deposits generally consisted of flint rubble in a

260 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

matrix of clay or clay loam soil, and can be interpreted as demolition rubble or levelling layers.

Ogbourne St Andrew/Wroughton Barbury Castle (SU 149 763); Prehistoric and Modern

Barbury Castle, an Iron Age hillfort on the scarp edge of the Marlborough Downs overlooking Wroughton airfield, was surveyed by the RCHME in 1998 at the request of, and with the help of, Swindon Borough Council (the site owners). Though Barbury 1s a prominent fort, little research has been done here and our knowledge of it is slight. Cursory excavations between the 1870s and the 1930s produced less information, probably, than the considerable disturbances to the site by the US Air Force during the Second World War, which revealed pits containing Iron Age pottery and human skeletons. Geophysical survey by the Ancient Monuments Laboratory in 1996 revealed a density of sub-surface features which was confirmed by our surface survey. This new survey recorded traces of approximately 40 hut circles and revealed many other interesting features. The forework outside the east entrance sits at a strange angle to the main ramparts and has clearly been cut by the outer ditch, suggesting it could have been an earlier enclosure re-used as an outwork. The scarp around the outside of the northern defences, possibly an unfinished third rampart, might alternatively be the remnant of another earlier enclosure. Also found within the fort were traces of one probable and two possible round barrows, suggesting that those surviving on the slopes to the west are the tail of a barrow cemetery which covered the ridge in the Bronze Age. Several large hollows around the periphery of the fort’s interior are anti-aircraft gun pits of Second World War vintage, showing that Barbury was used for defensive purposes in the 20th century for a type of warfare which could never have been dreamed of by its Iron Age builders.

Salisbury

Dairy Meadow Lane (SU 1562 2929); Undated and Modern

Archaeological evaluation on land to the rear of Harcros Timber Merchants, Dairy Meadow Lane, was carried out by AC archaeology. The one-hectare site is situated on low-lying ground less than 50m from the Anglo-Saxon settlement close to Dairyhouse Bridge. Trenching demonstrated the presence of deep

former quarries to the south, with variable degrees of truncation evident elsewhere. No intact archaeological stratigraphy or any pre-modern finds were encountered.

Waitrose, Old Livestock Market (SU 140 307); Modern

An intermittent watching brief was maintained during groundworks associated with the construction of a shopping complex on the former site of a cattle market north of the historic core of Salisbury. The level of the site had been raised by 1.5m when rubbish and topsoil were dumped before the construction of the Cattle Market in the 1950s. Few of the groundworks penetrated below this dumped material, but it was seen to directly overlay natural gravels. Part of the brick and concrete structure of a bridge constructed in the late 1950s to cross a former course of the River Avon was observed, but no features or deposits of archaeological significance were revealed during the watching brief.

High Street Enhancement Scheme (SU 14275 29785); Modern j

In April 1998 Wessex Archaeology undertook an archaeological watching brief during the enhancement of Salisbury High Street. The scheme involved the replacement of the existing tarmacadam carriageway and pavement, and the shallowness of the works did not penetrate any underlying archaeological deposits.

1 The Rings, Old Sarum (SU 13465 32895); Undated and Modern

Wessex Archaeology was commissioned to undertake an archaeological watching brief during renovations to 1 The Rings, immediately to the north-west of the Scheduled Monument of Old Sarum and within the presumed extent of the later medieval settlement of Nyweton Westyate. One undated ditch, one possible tree-throw hole and two modern features were recorded. The ditch was aligned south-south-west to north-north-east. No finds were recovered.

Fisherton Manor Middle School, Highbury Avenue (SU 132 307); Modern

A watching brief by Wessex Archaeology during the groundworks for a new entrance lobby to the school revealed that the land had been raised with redeposited topsoil, probably as part of the terracing for the construction of the school. The underlying river gravels were not observed in any of the trenches. Although no archaeological features or deposits were observed during the watching brief, it is possible that

EXCAVATION AND FIELDWORK IN WILTSHIRE, 1998

archaeological deposits may be preserved below the made ground in this area.

The Close (SU 14350 29711); Post-Medieval and Modern

Wessex Archaeology was commissioned to undertake the field evaluation of a proposed extension to 22 The Close, a largely 18th century cottage with possible medieval beginnings. The existing north-east wall of the building is probably of medieval construction and was recorded during evaluation.

A probable 18th century courtyard surface with soakaway, consisting of at least two phases, was observed sealing deep deposits of building rubble. A number of disturbances of the courtyard may well have been the result of a major renovation of the building known to have taken place in the mid 18th century. These included a linear trench running the length of the wall, associated with red brick underpinning of the medieval wall fabric. The full northern and western extent of the courtyard still remains unclear, although the eastern limit is suggested by surviving traces of kerbing. The courtyard appears to have remained in use until the early—mid 19thcentury, when the general ground level within the plot was raised by the deposition of large quantities of rubble and garden soils. A low background of residual medieval material was observed, with small quantities of residual potsherds and tile recovered from the topsoil, linear trench fill and gravel deposits at the base of one test pit. No significant medieval features or horizons were encountered.

Salisbury Plain Training Area Sites Wessex Archaeology has undertaken a series of excavations associated with the construction of the Southern Range Road within the Salisbury Plain Training Area.

East of Quebec Barn (ST 9704 4410); Late Bronze Age and Romano-British

Excavations to the east of Quebec Barn, c.2km west of Chitterne, on the north-facing slope of a large dry valley, produced a small assemblage of Late Bronze Age and Romano-British pottery, worked flint and four possible quernstone fragments, all from within the topsoil. Two pits, two post-holes a gully and a hearth, all of Late Bronze Age date, were also recorded.

East of Knook Castle (ST 9620 4388); Undated Excavation c.100m east of Knook Castle, close to the crest of a ridge between two large dry valleys, revealed

261

a positive lynchet and a small hearth, neither of which produced any dating evidence.

Willis’s Field Barn (ST 9473 4366); Late Neolithic/ Early—Middle Bronze Age, Late Iron Age

An area of 640m? was excavated on the crest of a ridge on the north side of the Wylye Valley 2km north- east of Heytesbury. A ‘ditch and pit complex’ had been identified during an earlier evaluation. Two ditches and three pits represented at least two distinct phases of activity. Three other features may be the vestigial remains of small pits or post-holes. A moderate-sized ditch of Late Neolithic/Early Bronze Age date and a small pit containing a small assemblage of Beaker pottery were recorded. Two other small pits may be contemporary. A substantial ditch, possibly part of an enclosure cut both the earlier ditch and pit, and was itself later recut. A large assemblage of Middle Bronze Age pottery was recovered from both phases of this ditch, along with large quantities of animal bone, burnt and worked flint and a small quantity of quernstone fragments. A complete cattle skull and a group of articulated cattle bones were recovered from the terminal of the enclosure ditch. Two iron objects, a small flat fragment and a Late Iron Age brooch, came from the uppermost ditch fill.

Horse Down (SU 02100 48300); Late Neolithic/Early Bronze Age and Undated

An area of c. 1250m? was excavated across an east— west chalk spur to the west of the village of Tilshead where two linear features and three possible pits had been identified during evaluation. The ditches probably represent prehistoric field boundaries; extensive field systems have been recorded in aerial photographic surveys on high ground to the south (on Copley Down) and west (Tilshead Down). However, a small quantity of Neolithic/Early Bronze Age pottery in one of the ditches may indicate an earlier date.

South of Foxtrot Crossing (SU 10985 48550); Romano-British

Three possible intercutting ditches of Romano- British date had been recorded in an evaluation on a low ridge to the north-west of Tilshead. Excavation revealed at least nine intercutting quarry pits producing samian and Romano-British coarsewares. The upper fills were cut by a curvilinear ditch which produced a single sherd of Romano-British pottery.

262 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Shrewton

Old Coal Yard (SU 0681 4340); Medieval and Modern

An archaeological watching brief by Wessex Archaeology during the construction of six residential properties recorded two ditches and a post-hole, all undated. Two sherds of 12th or 13th century pottery were recovered from a pit (probably modern), but may have derived from one of the ditches through which the pit had been cut. If of medieval date, these features would constitute the first buried archaeological remains of medieval Shrewton.

Stanton St Bernard

Manor Farm (SU 0934 6231); Undated

An archaeological evaluation was undertaken by AC archaeology on the site of a proposed housing development at Stanton St Bernard. The evaluation comprised three trenches amounting to a total area of over 75m’. A large ditch, possibly medieval or post- medieval, was encountered in one trench; with a second trench containing a number of irregular, shallow features of indeterminate date.

Sutton Benger

58 High Street (SP 9475 7880); Post-Medieval

A previous topographic survey had revealed a variety of medieval earthwork features immediately to the west of land at the rear of 58 High Street. Evaluation by CAT to the rear of this property found no further medieval features. The area had been subject to quarrying in the 19th century.

Tilshead

‘M’ and ‘N’ Crossings, A360 near Tilshead, SPTA (SU 0392 4685 and 0432 4590); Later Prehistoric Observations maintained by Asi during groundworks associated with the upgrading of two road crossings on the A360 south of Tilshead revealed that crossing point ‘M7’ is situated over the line of the ‘Old Ditch’ — otherwise known as the Breach Hill linear earthwork — one of a series of prehistoric land boundaries preserved on Salisbury Plain. The watching brief was able to record, however, that the upgrading works had been minimally intrusive, with no significant archaeological impact.

Upavon

Widdington Farm (SU 125 540); Modern

An evaluation within a proposed agricultural deveiopment at Widdington Farm was undertaken by AC archaeology. This revealed only modern re- deposited material, probably from the construction of an existing adjacent barn.

Wanborough

Earlscourt Manor (SU 2166 8558); Roman and Medieval

Two trenches, cut prior to granting permission for tree planting, revealed evidence of substantial medieval occupation. Three Romano-British and many 13thand 14thcentury pottery sherds were also found in soil disturbed by ploughing and in the back- fill of a recent pipe trench. The work was undertaken by Bernard Phillips.

Warminster

Battlesbury Bowl (ST 8986 4610); Late Bronze Age/ Early Iron Age

Wessex Archaeology undertook an archaeological excavation of a c.418m long strip (0.62ha) of land immediately to the north of the Iron Age hillfort of Battlesbury Camp, near Warminster. The work identified an extensive spread of Late Bronze Age/ Early Iron Age activity along the length of a north— south chalk spur between the hillfort and the modern military buildings of the Harman Lines works to the north. A group of ditch and gully features appear to delineate the southern extent of the activity, close to the hillfort.

More than 900 archaeological features were recorded, including c.725 post-holes, c.170 pits and seven ditches/gullies. Most date to the 8th—7th centuries BC, although a small number of 10th—9th century BC features were also recorded. Three features of 6th—4th century BC date occurred in the southern part of the excavation area. Most of the pits occurred within distinct clusters. The fills of the pits from the two most southerly clusters were distinctly different from those in the two northerly clusters; 34 of 38 pits containing ‘structured deposits’. All three 6th—4th century features occurred in the two southern clusters. Human skeletal remains occurred in 19 features, including ten of those with otherwise ‘structured deposits’. Six complete inhumation burials were recorded from four pits in close proximity,

EXCAVATION AND FIELDWORK IN WILTSHIRE, 1998

comprising two double inhumation and two single inhumation burials with a further six pits containing human skeletal fragments. All but one were in the southern part of the site.

Ten sub-rectangular/square post-built structures were recorded: six 4-post structures, two 5-post and two 6-post. Seven of these were again in the southern part of the site. Three possible round-houses were recorded, including one with internal hearths.

South-east of Battlesbury Wood (ST 9008 4488); Late Bronze Age, Romano-British and Undated

Wessex Archaeology undertook an archaeological excavation of an area of 560m? 400m south-east of Battlesbury Wood, at the foot of the low ridge between Battlesbury Hill and Middle Hill. This was targeted on two channels, a series of pits and stake-holes and a possible linear mound identified during an earlier evaluation. Four pits, a ring-gully and a possible post- hole were excavated, as well as three probable erosion channels, potentially of prehistoric and Romano- British date.

Only one of the pits contained datable material, a small assemblage of worked flint and Late Bronze Age pottery. Small quantities of burnt flint were recovered from the other pits but no datable material. The shallow ring-gully, c.9m in diameter and no more than 1.1m wide, lay in the centre of the excavation area, between two erosion channels and produced a very small assemblage of fired clay fragments, animal bone and burnt flint. The erosion channels were probably caused by seasonal ‘run off’ from the higher ground to the north-east. One may be confidently dated to the Romano-British period, the others are possibly 1st millennium BC.

Stake-holes identified during the evaluation are in fact probably rootholes, and the possible linear mound was found to consist of a localised subsoil deposit of post-medieval or modern date. This may be the result of relatively recent erosion from upslope of the site.

Boreham Farm Bungalow (ST 8951 4566); Mesolithic/Early Neolithic, Late Bronze Age and Early Iron Age

Wessex Archaeology undertook an excavation of an area of 340m? in the base of a broad valley immediately south of Battlesbury Hill on the north bank of a small culverted stream, a tributary of the river Wylye. This was targeted on a group of pits and stake-holes associated with a channel filled with a black clay deposit identified during a previous evaluation.

263

A large stream channel, two pits and two possible post-holes were examined. The channel probably represents the original course of a small stream, almost certainly that which currently flows in a deep, narrow ditch immediately to the south of the site. A small assemblage of abraded Late Bronze Age pottery and residual worked flint of Mesolithic/Early Neolithic date were recorded from the lower fills, with two sherds from a furrowed bowl of Early Iron Age type from the upper fills. This material appears to be derived from elsewhere, possibly further upstream. The other apparent features may be of natural origin, but included charred seeds and Late Bronze Age pottery.

Water pipeline, Furnax Lane (ST 8662 4617); Iron Age, Romano-British, Medieval and Post-Medieval

Wessex Archaeology was commissioned to undertake an archaeological watching brief during construction works along c.300m of the south-eastern slopes of Brick Hill. A pit containing pottery of Early to Middle Iron Age date (700-100 BC) was recorded. Deposits of colluvium in the south of the field and a layer of crushed chalk, probably a recent levelling layer, were also recorded. Finds of Iron Age, Roman, medieval and post-medieval date occurred in the topsoil.

Westbury

Westbury Quarry Chalk Pit (centred on ST 890 503); Prehistoric and Romano-British

Some 46ha adjacent to the existing chalk pit were the subject of an evaluation by AC archaeology. An earlier fieldwalking exercise had suggested the presence of low-density scatters of worked flint, and the layout of trenches was designed to concentrate on those areas, whilst providing even coverage of the remaining parts of the site.

In fields to the north-east of the existing quarry (Area A) there were no obvious archaeological features. A number of probable tree-root hollows and periglacial features were investigated, but none produced archaeological finds. Limited quantities of unstratified Late Neolithic-Earlier Bronze Age worked flint were, however, recovered from the spoil heaps. In areas south of the existing quarry (Area B) positive findings were limited to two linear features (one re-appearing in four trenches). Both contained small quantities of worked flint and one a small sherd of Roman pottery.

Proposed Northacre Business Park (ST 857 522 and area); Prehistoric and Medieval Following initial evaluation of part of the site in 1997

264 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

(see WAM 92, 142) further investigation was undertaken by AC archaeology. This comprised twenty machine-excavated trenches across the site and an area of some 900m? dug in the area of the intended access road. No archaeological sub-soil features were encountered, and the only artefacts comprised very small quantities of medieval pottery and prehistoric worked flint, neither in sufficient numbers to suggest settlement activity in the vicinity.

Wingfield

Olid Timber Yard, Church Lane (ST 822 567); Modern

An evaluation was carried out at The Old Timber Yard by AC archaeology during December 1998. Evidence from early maps indicates probable medieval settlement concentrated alongside this, the main road of the village. The evaluation comprised four trenches, all of which contained evidence of truncation and revealed modern made-ground immediately overlying natural subsoil. With the exception of a modern pit and a ditch of post-medievai date, no subsoil features or archaeological deposits were encountered, and no pre-modern finds were recovered.

Winterbourne Stoke

Hill Farm (SU 0844 4086); ?Prehistoric Observations were undertaken by AC archaeology in conjunction with groundworks to construct a telecommunications mast at Hill Farm. Although the site lay within the known extent of a prehistoric field system, no archaeological features were observed during the work. A limited quantity of burnt flint was recovered.

Wroughton

Overtown House (SU 1543 7970); Modern

Wessex Archaeology carried out an archaeological watching brief during machine stripping for the construction of a new drive to the west of Overtown House. The site is immediately north of the former medieval village of Overtown (SM2859/01 and 02), represented by a complex of earthworks. No

archaeological features or deposits of note were recorded. One sherd of Romano-British pottery and several of 13th century date, along with pieces of animal bone, were recorded from the subsoil and the remnants of a former pathway and driveway.

Yatton Keynell

Church Farm (ST 8655 7655); Medieval

An earthwork survey was carried out at a scale of 1:1000 in a field immediately north of Church Farm, Yatton Keynell. The earthworks comprise the remains of six tofts defined by a bank and ditch. The largest measures 80 x 25m, whilst the length of the others has been reduced by the construction of the modern road. At the eastern end of three of the tofts are probable building platforms, measuring up to 25 x 15m, whilst along the western side of the tofts is a hollow way with ridge-and-furrow cultivation in the remainder of the field. The work is part of a personal research project on the Hundred of Chippenham being undertaken by Graham Brown.

Various

Wiltshire barrows; prehistoric

The Ancient Monuments Laboratory undertook several magnetometer surveys in Wiltshire as part of a pilot study of severely eroded barrows. Six sites were used to assess the response from, and condition of, any remains. Only one site, at Littkecombe Down produced results suggestive of significant surviving prehistoric features. On other sites interesting anomalies were recorded, such as a square-shaped feature at Mere Down Farm and a large amorphous response at Liddington Castle, but neither of these is likely to be ploughed out barrows. On the other three sites (The Park, West of Court Hill Plantation; Smeathe’s Ridge, South Burderop Down; and Coombe Down, South-East of Smeathe’s Plantation) only minor anomalies of possible archaeological origin were located. The lack of positive geophysical identification of barrows is unlikely to be due to geological conditions: either they have been totally eliminated from the landscape by cultivation, or the original locational information was in error.

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 265-272

Reviews

S. E. Kelly (editor). Anglo-Saxon Charters V; Charters of Shaftesbury Abbey. Oxford University Press for the British Academy, 1996; xxxvili + 151 pages. Price £30.00, hardback, ISBN 0 19 726151 5.

A joint committee of the British Academy and the Royal Historical Society was set up in 1966 to publish a new critical edition of the whole corpus of Anglo- Saxon charters. The magnitude of the task may be judged from the list of some 1500 charters in P. H. Sawyer’s book on the subject. A glance at the list emphasises how great was the loss of these documents from the 8th century onwards in those parts of the country devastated by the ‘fury of the Northmen’. This fifth volume of the series will be welcomed, in the words of N. Brooks’s foreword, as setting new standards in the editing of these charters.

The introductory section of Dr. Kelly’s book deals with the history of the abbey and its estates and with the nature of its Cartulary (B.L. Harley 61) which preserves the only surviving copies of the abbey’s thirty known pre-conquest charters, ranging from one of around 670 A.D. to a grant by King Cnut in 1016. The text of this early 15th century manuscript, much corrupted by repeated copying over the centuries, sets editors a formidable task to recover as far as possible the original charters and then to evaluate their authenticity. Dr. Kelly’s conclusions are most carefully set out in her commentaries on each, with a wide ranging discussion placing the charter against its own probable background. Here too is an analysis of the minutiae of the standard formulae in the Latin of the charters, and their development over the Saxon centuries. Without a detailed knowledge of the subject it is not easy to assess this work, but we seem to have here a large step forward in elucidating these charters.

Nearly all the charters take the same form, albeit _ with wide variations. A Latin introduction is followed by a statement of the grant itself, often with an anathema calling down damnation on anyone who might break its terms. This leads to a description in Old English of the bounds of the land granted, and the charter ends with a list of witnesses who confirm the grant. In the introduction, an invocation ‘In the

name of God’ or a more florid phrase usually leads to a section of a religious nature and sometimes of great length. The king styles himself ‘of the West Saxons’ in the earlier charters but ‘rex Anglorum’ or even ‘king of all Britain’ in later ones.

With the continuing development of this subject it is natural that there will be new suggestions on the interpretation of details. In the Bradford charter (no. 29) for instance, recent work has convincingly proposed that the unidentified Alvestone (on p. 120) was Calvestone, that is, Kelston, Somerset, and that a small stream running from a point near ST821582 on the B3109 road down to join the River Frome at Stowford could be the unidentified wigewen brook. These however are minor matters.

Many readers will not find reading the Latin of the texts easy. It would obviously not be possible in a book of reasonable size and cost to translate the texts, but the Glossary in an appendix overcomes many of the problems which arise largely from the florid and obscure style. Lack of translation does not, in any case, affect one’s appreciation of the treasures to be discovered in this book.

R. HARVEY

Peter Tolhurst. Wessex, a Literary Pilgrimage. Black Dog Books, 1999, 264 pages; 200 black and white illustrations (line and photographic), 8 colour photographs. Price £19.95, hardback, ISBN 0 9528839 1 0.

It is through its literature that many of us first came to know the landscapes which we now love and visit regularly. Shakespeare was often our first introduction to the Avon and Warwickshire, the grandeur of the Lakes was invoked by Wordsworth before many of us had set foot there. Further back in time children’s authors may have been the inspiration for visits in later life to the places where they had set their stories.

Closer to home few can think of Dorset without an image from Thomas Hardy, the Powys brothers

266 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

or, more recently, John Fowles coming to mind. Indeed the combination of literature and the cinema has made the Cobb at Lyme Regis an enduring image of Dorset for many. Unlike Hardy, with his vision of an extended Wessex, Peter Tolhurst concentrates on the historic core of Wessex, Wiltshire and Dorset. Naturally Dorset dominates. Wiltshire cannot lay claim to a novelist of the first rank although some have dwelt here for a while and used the county as a setting for a book or two.

Most notable of these is Nobel prize-winner William Golding who, as carefully chronicled, used his life and experiences at both Marlborough and Salisbury for two of his important early novels. Golding visited Figsbury Rings with E.M. Forster many years after Forster published The Longest Journey, which was partly inspired by the Rings and their landscape but mostly by a chance meeting with a lame Wiltshire shepherd boy in those Rings.

Tolhurst has a sympathetic understanding of Francis Kilvert, a writer whose deep Wiltshire roots and local writing are often overlooked in favour of his beloved Clyro. He recognises that the eligible young socialite moving in the upper reaches of rural society is fully compatible with the earnest curate bringing comfort to poor people in their broken down hovels, without railing against the social order which created their pitiful situation. Although in full agreement with the writer’s views on the poetical works I would take issue with the somewhat summary dismissal of the later prose works of Edward Thomas. There is a strong sense of locality in these works and, although condemned by economic necessity to hack work, Thomas put far more into the lowest paid task than many a better paid writer has managed to achieve.

It is always interesting to see the links between writers associated with one’s own country. Edward Thomas was inspired by Richard Jefferies, for whom a sympathetic evaluation is provided in this book. Thomas went on to write a biography of the mentor he never met while later, outside the geographical scope of this book, Gloucestershire country writer John Moore wrote a biography of Thomas by whom he was inspired but never knew. Three generations of writers linked to one another by common themes and beliefs. Another link is provided in that the eyes of Kenneth Allsop were fully opened to the value of the countryside, its birds and other wildlife by Henry Williamson. There is a good section on Allsop’s writings and his fight to preserve part of Dorset’s ecology.

This is an excellent book by a literary man with great insight into the way landscapes have shaped their

writers and how, in turn those writers have shaped their landscapes. Thomas Hardy is dominant but, apart from those already mentioned, Alfred Williams, Charles Sorley, John Betjeman, Henry James, Stevie Davies, Virginia Woolf, A.G. Street, Edith Olivier and W.H. Hudson represent those who have been influenced by Wiltshire and have themselves influenced public perception of that county. There are omissions, Geoffrey Grigson and Maureen Duffy could have been included, but there is much here to savour.

It is through its writers that a country achieves an immortality comparable with that given by the great structures of the past. Most of what we do will be soon forgotten; that which is written in earth, stone and words will endure. Take this book and be captivated by the literary landscapes of Wiltshire and Dorset. Then visit both new and familiar places and see them through the eyes of their writers.

MICHAEL MARSHMAN

Mere Papers, Numbers 1 -9, edited by M. F. Tighe. The Friends of the Church of St Michael the Archangel, 1996 -1999, 204 pages; illustrations, maps. Paperbacks.

In the relatively short period of three years this on- going series has produced a substantial body of over two hundred pages of research into the town’s history. Subjects covered include the Textile Industry, Inns, Congregationalism, Probate Records, Domestic Buildings, Enclosure, Edge-Tool Making, the Families of Walton, Goldsborough and Edmunds, Doctors and T. H. Baker, Historian. All but the last two, which were written by Dr. David Longbourne, for many years a leading figure in the Mere Historical Society, were written by the editor himself. His industry and enthusiasm are as striking as his ability to write in an informed manner.

His piece on Mere in 1851 describing what a visitor to the town would have seen is an extremely clever blend of imaginative writing and sound historical knowledge of the subject. The description of the self-penned will of Margaret Harding in the 1630s as a ‘DIY’ will is a fine example of his engaging and informative style which accurately describes the idiosyncratic spelling of the document.

The preface in each volume in which the editor seeks contributions from others is not a sign of flagging on his part. It is a call from an enthusiast

REVIEWS

eager to engage others to share his passion. It is to be hoped that it may fall on receptive ears and that a series, now firmly planted, will continue to flourish. That would be no mere achievement.

STEVEN HOBBS

Hazel Gifford. The Biography of a Country Church; Berwick St. John. Winkelbury Publications, 1999; 101 pages, illustrations. Price £6.00, paperback, ISBN 0 9535893 0 7.

Every parish has one. It is the background to village life, accepted as part of the landscape by local people and an influence on the lives of most of them even if they do not attend or only enter its doors for christenings, marriages or funerals or at Christmas and Harvest Thanksgiving. Most churches have a printed guide; some are good, some are indifferent. Others have only information sheets pasted on boards. It is a rare thing to have a book devoted to a village church but that is what Hazel Gifford has given us.

This is a good biography of a village institution. The book takes full account of the effects of national events and movements and their influence upon a small rural parish. Nearly half the book is taken up with descriptions of rectors and lay benefactors, fleshing out those characters who are often only names on a board inside the church. One such was Richard Downes (1826-1855), an energetic rector who repaired and improved the church, spending nearly £400 of his own money. He also took great interest in employment during times of agricultural depression, built a school with his own money and both he and his wife were responsible for many other charitable works.

It is rectors like Downes who bring church history alive. Although information is sparse in earlier centuries, one rector notably stands out. Edmund Audley (1465-1480) held several livings and so visited Berwick infrequently but he did become, first, Bishop of Rochester and then Bishop of Salisbury. His family were powerful and well connected and Audley was made Chancellor of the Order of the Garter.

Chapters on the buildings, furnishings and memorial stones and effigies provide the reader with a picture of the church, even though they may never have visited it. I would have liked to have seen a chapter on the relationship of the church with its community; for it is this relationship, spiritual, economic and social, which has sustained the church

267

over the centuries. Apart from that quibble I found this a well researched and annotated book which serves as an excellent introduction for anyone wishing to learn about the history of rural parish churches, as well as being the story of the Berwick church. Very readable, it contains good and interesting illustrations and provides a definitive history.

MICHAEL MARSHMAN

Gwyneth F. Jackson (compiler). A tale of two manors; Zeals, a Wiltshire village. Dickins Printers, 1997; 208 pages, illustrations, maps. Price £12.50, paperback ISBN 1 902247 00 0.

The effects of a bypass on a community can be far- reaching. To the people of Zeals it provided the spur for a celebration of the history of their village, now no longer divided by the relentless flow of traffic along the A303. The result of their efforts is an attractively produced, well illustrated and informative history for which all involved in its production can feel justifiably satisfied. Drawing on a rich fund of oral history, supported by a wide range of excellently captioned photographs, the reader is presented with a good balance of administrative and personal histories which make up the story of the village. The final chapter is a good example of the blend of both. Its content somewhat belies its somewhat discouraging title, Local Government and Facts and Figures. Lists of parish overseers of the poor and parish council chairmen and an analysis of occupations in the village in 1891 compared with those of today are useful and informative. Similarly, lists of surnames found in archives for 1332, 1648, 1891 and 1991 are fascinating, not least because the name Martin occurs in each list.

The overlying strengths of the book lie firmly in the last 150 years which comprise the bulk of the text. A few weaknesses occur in the brief section dealing with the earlier history which sadly detract from the pleasures to be enjoyed later on. The suggestion that there is a difference between Cottagers and cottagers in the Domesday Book is a woeful misconception. The notion that Geoffrey de Seles escaped royal wrath in establishing a park without first obtaining permission was probably as a result of him being a hunting acquaintance of the king reveals a naive misunderstanding of the relations between king and tenants-in-chief. These lead to the view that the book might have had a better balance if the earlier section had been excised. This would have enabled a more

268 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

appropriate title to have been selected. However the unpublished thesis on the settlement and landscape of Zeals, from which only a few interesting points were included, could have provided the basis of the early section and allowed an interesting piece of original research to reach a wider audience.

Proof-reading was a little awry. Mention of the researches of John Bratton on p.14, correctly referred to as John Batten on p. 24, is an unfortunate slip since the noted Somerset antiquarian had a real link with the village as his daughter married a Troyte Bullock, thus making him a great grandfather of Bill Woodhouse, who contributed an informative piece about Zeals House. It is worth noting here that the sources of Batten’s work, together with that of T.H. Baker (not J.H. Baker as on p. 204) relied on by the authors are in volumes 28 and 29 of this journal.

These are, however, minor quibbles which should not detract from the success of the book in recording the life of the village over the last two centuries ensuring that it will be of value to all those will an interest in Zeals.

STEVEN HOBBS

Books also noted

It is intended that there will be a section in WAM 94 for those works on a parish published to mark the millennium in that community. Some books already published are not noted below as they are being held over for next year.

Norman Beale. Is that the Doctor; a history of the Calne GPs. N. Beale, 1998; 95 pages, illustrations. Price £6.95. paperback, ISBN 0 9533992 0 6. Fascinating account, by a current Calne GP of the medical men and women who have practised in Calne since the mid-17th century. Sixty three biographical entries with interesting insights on the development of medicine.

Keith Berry. Bradford on Avon’s Schools; the story of education in a small Wiltshire town. Ex Libris Books, 1999; 239 pages, illustrations. Price £8.95, paperback, ISBN 0 948578 96 3. Covers the history of schools in the town for the last 300 years relating national developments to local practice. The main part of the book is a detailed history of Fitzmaurice Grammar School.

John Chandler. Great-grandmother’s Footsteps; a stroll through Victorian Salisbury. Salisbury and South Wiltshire Museum, 1999; 45 pages, illustrations. Price £4.99, paperback, ISBN 0 947535 18 7. Skilful juxtapositioning of the paintings of Louise Rayner and others from the 1870s with maps and modern photographs interwoven with the expected masterly text.

T.S. Crawford. Wiltshire and the Great War; training the Empire’s Soldiers. DPF Publishing, 1999; 181 pages, illustrations, maps. Price £12.95, paperback, ISBN 0 9535100 0 X. Suprisingly the first book to be published specifically on this period which so greatly affected Wiltshire. Substantial section on the preparation for, and the reality of, war and a very helpful section on the histories and descriptions of individual camps.

Jane Freeman and Aelred Watkin. A History of Malmesbury. Wiltshire County Council and The Warden and Freemen of Malmesbury, 1999; x + 230 pages, illustrations, maps. Price £9.75, paperback, ISBN 0 86080 444 5. All the articles on Malmesbury and its abbey from volumes 3 and 14 of the Victoria History of Wiltshire re-formatted into a more user- friendly format.

T.E. Holt. Travelling Folk: Itinerant Mission in the Diocese of Salisbury, 1882,1883. Transcribed and edited by Rosemary Church. Wiltshire Family History Society, 1999; 68 pages, maps. Price £6.50, paperback, ISNB 1898714 44 4. The log books of this mission to the travelling population of drovers, showmen, hawkers and gypsies who frequented fairs and race meetings in south Wiltshire and Dorset. Complements A Parish on Wheels (1897) by J. Howard Swinstead.

Danny Howell. Wylye Valley Folk Volume 1; an album of memories by senior citizens who lived and worked in the Wylye Valley during their younger days. Recorded and edited by Danny Howell. Bedeguar Books, 1999; 288 pages, illustrations. Price £18.00, paperback, ISBN 1872818 35 8. Five people with extensive memories which have been collected and transcribed by an expert in the oral history field.

Peter Lavis. A Century of Nestle at Staverton 1897 - 1997. Nestle UK Ltd, 1998; 56 pages, illustrations. ISBN 0 9532792 0 0. One of a small number of business histories for Wiltshire provides a welcome account from the time the Anglo-Swiss Condensed Milk Company bought the Staverton cloth mill.

Ruth Marshall. Trowbridge Voices; recollections of local people compiled by Ruth

OBITUARIES

Marshall. Tempus, 1999; 128 pages, illustrations. Price £9.99, paperback, ISBN 0 7524 1644 8. Interesting collection of memories dating back to the early 1900s, complemented by largely unpublished photographs.

Michael Marshman. The Wiltshire Village Book. Countryside Books, 1999; 256 pages, illustrations, map. Price £9.95, paperback, ISBN 1 85306 583 8. Covers over 180 villages with snippets of history, anecdotes, description, personalities and events.

Terence Meaden. The Secrets of the Avebury Stones. Souvenir Press, 1999; 152 pages, illustrations. Price £12.99, ISBN 0 285 63501 8. Interesting account linking Avebury to the Neolithic Earth Goddess which provides symbolic meanings for every stone.

Max Milligan. Circles of Stone; the prehistoric rings of Britain and Ireland. Text by Aubrey Burl. Harvill Press, 1999; 232 pages, mainly colour photographs. Price £30.00, hardback, ISBN 1 86046 661 3. Only a small amount of Wiltshire material but

Obituaries

269

these superb photographs show the wide variety of stone circles constructed in these islands over a 2,000 year period.

Lynda J. Murray. A Zest for Life; the story of Alexander Keiller. Marren Books, 1999; 134 pages, illustrations. Price £9.99, paperback, ISBN 0 9536039 0 3. A general biography which gives generous coverage to the two decades of Avebury excavations in Keiller’s full and varied life in this well written and readable book.

Andrew Sewell. Aldbourne; the Present Past. A. Sewell, 1998: 78 pages, Illustrations, maps. Paperback. Comprehensive coverage of the parish in a well researched and informative book. Strong on prehistory.

Doug Small. The Wilts and Berks Canal. Tempus (Images of England Series), 1999; 128 pages, chiefly photographs. Price £9.99, ISBN 0 7524 1619 7. Photographic exploration of the canal in images both old and new. An evocative journey from Semington to Abingdon also shows new restoration projects and plans for the future.

Desmond Hawkins, writer, producer and founder of the BBC’s Natural History Unit, died on 6 May 1999. He was born on 20 October 1908.

Desmond Hawkins was born a Londoner, later moving to Guildford, but at heart always seemed to be a countryman. The family firm was an ironmonger’s shop near the Elephant and Castle and after leaving Cranleigh at 16 he spent five years working in the hardware business. While delivering electrical equipment to theatres he became familiar with life in the West End, seeing plays, attending lectures and developing a love of literature. This self education led to his decision to become a writer, at first a rather precarious living, writing for such magazines as Purpose, The Listener, The New Statesman and Time and Tide. He became part of the cafe and literary culture of the 1930s, discussing art and literature with the fascinating novelists and poets of that decade. He also took great pleasure in supporting them when he was literary editor of The New English Weekly. These times and his early life are well chronicled in his autobiography, When I Was.

His first link with Wiltshire occurred in this period when living just over the border in Berkshire. Johnny

Morris was then a farm agent living at Aldbourne and the two friends often visited one another. In later years they managed to spend one day each year together at a Test match, at Lords or the Oval; a practice they continued to 1998 for they died on the same day before the 1999 Tests began.

With a wife, Barbara, and two children writing was very much a hand to mouth existence but from 1936 onwards he had ideas for programmes accepted by the BBC and was also editing selections from writers such as Donne and Lawrence for publishers. In 1939 his first novel Hawk Among the Sparrows was published. It is the story of the disruption caused by an intellectual who has taken rooms with a middle class couple, the sexual awakening of a young girl and a destructive conflict between her and a worldly wise aunt. A second, and lesser, novel, Lighter than Day, was published in 1940 but by then Desmond had realised that the war had ended the society and mores about which he had just begun to write.

Disqualified from military service by ill health he worked as a freelance for the BBC particularly on the weekly programme Country Magazine and the daily War Report. He worked with George Orwell

270 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

in the Far Eastern Service and wrote scripts with Louis MacNeice on the bombing of the capital. He became a permanent member of the BBC staff in 1945 and soon became a features producer based in Bristol. This period was to be the high point of regional radio which was an important opinion former and shaper.

Through the medium of radio Desmond introduced many writers to a far wider public than could have been achieved by other means. Most notable was Dylan Thomas who was a friend from the 1930s when the two used to drink in Soho and talk of poetry and showgirls in equal measure. Until only a year or two before his death, when he sold them at auction, Desmond preserved an important collection of material on Dylan including a letter in which the poet announced his marriage.

Under Frank Gillard the West Region of the BBC took a specialist interest in wildlife programmes and Desmond, whose first broadcast in 1936 had been on birdsong, took his opportunity with a series called The Naturalist. In the early 1950s he moved into television with a series, Look, and also produced some of Peter Scott’s early wildlife programmes. These led, in 1957, to him founding the BBC’s Natural History Unit and setting out its strong principles of parallel status of radio and television, high standards of scientific accuracy and use of the best available technology.

At the same time literature was not forgotten and he dramatised the novels of Thomas Hardy which were broadcast as Sunday serials and greatly boosted both Hardy’s readership and reputation. In 1955 Desmond had become head of programmes in Bristol and later became the region’s last controller in 1967 until his retirement in 1970. At this point, with nearly 30 years of active life to come he decided that he would produce at least one book, one television film and one radio programme each year. Amazingly he very nearly achieved this.

It was in the latter third of his life that his involvement with Wessex, always strong, became the cornerstone of his literary and ecological life. He wrote what are arguably the best books on Hardy by anyone of his generation. These were followed by articles and presentations for the Thomas Hardy Society, broadcast anthologies of poetry and prose and a televised book, Hardy’s Wessex. He later confessed that as far as Thomas Hardy was concerned he had written all he could and so, fortunately for us in Wiltshire, he turned his attentions to other areas. He produced a definitive work on Cranborne Chase and later brought out new editions of rare early works on the Chase by William Chafin and Wake Smart.

In 1973 while researching the friendship of Hardy and Agnes Grove he became aware of the Grove family habit of writing and preserving diaries. After his book on this friendship had been published he researched and wrote another on the relationship between Harriet Grove and her cousin, the poet Percy Bysshe Shelley, Shelley’s First Love. Many of the diaries are in the Wiltshire and Swindon Record Office and many were the trips made from Blandford in Dorset to Trowbridge. It was then that we, in Wiltshire, became aware of the detailed and meticulous research that preceded every book, article, broadcast and lecture that were a part of the man and writer of integrity that he was.

His energy was prodigious. Before the publication of The Grove Diaries and at the age of 86 he undertook an exhausting trip to the U.S.A. to further research diaries held there and to raise funding for the publication of the American edition. He was not happy unless he had one project in progress and another one or two ready to start when the current one had been completed. Only two weeks before he died he was planning a trip to Trowbridge to discuss a small edition of a travel journal of Tom Grove.

With an enviable lifetime of achievement Desmond was a very modest man and had to be pressed to speak of writers he had known and things he had done. He much preferred to talk of current projects and discuss avenues of research. He was a great friend of the Wiltshire and Swindon Record Office and the County Local Studies Library and we all became accustomed to his erudite questions on genealogy, heraldry and local history which extended our own knowledge and competence whilst finding the answers. Wessex and south western England have lost a great advocate of their history, literature and beauty and we in Wiltshire feel that we have lost a great gentleman and a friend.

MICHAEL MARSHMAN ,

Eve Machin, archaeologist, poet, linguist, died 1 November 1999. She was born 17 January 1915.

Although born in Bristol to an English father, Jim Baxter, Eve’s mother came from a well-to-do Austrian Jewish family, and with her younger brother Eve was brought up in Vienna from the age of eight. There Eve got to know the friends and relatives of her Viennese family, a wide and and often brilliant circle of artists, entepreneurs, scientists and writers. Although she was to return several times to school in England, the Vienna of that era was in many ways to remain Eve’s spiritual home. She went hiking in the Alps, she was introduced to the glittering group of poets

OBITUARIES

and dramatists who gathered around her distant cousin Richard Beer Hofmann, and she must often have been invited to the dazzling parties and balls which were still such a feature of Viennese life. She was soon fully bilingual in German and English, and learnt excellent French from her Swiss governess. She acquired a life- long love of German literature and poetry.

At the age of 17 Eve was back in England, living in Surrey and preparing for university. Her fluency in English, German and French was not in doubt, but one other language — Latin — was a requirement. Luckily, a master at the nearby Cranleigh School, Max Machin, was able to tutor her in this. In 1933 Eve went up to Oxford to study Philosophy, Politics and Economics at Lady Margaret Hall, but left a year later to marry Max. Although in many ways a natural academic, it was a decision that she never regretted.

Max and Eve settled in Cranleigh and employed a young German architect to build them a house there. The resulting structure caused some consternation among the good people of Surrey, but it gave their children the rare distinction of having grown up in a Bauhaus. The field around the house was soon populated with an assortment of animals including goats, geese, and guinea-pigs. The couple’s first child, Tess, was born in 1935, followed by Noel in 1939 and Blaise in 1946. Max, who had already served in the First War, did not have to go on active service after 1939, but for Eve, who was by now also working as a teacher at Cranleigh, bringing up two young children during the war was very stressful.

Max and Eve had long shared a love of archaeology; they spent holidays in Avebury and Brittany, and in the Dordogne they visited the painted caves of Lascaux and Font de Gaume. And even though a schoolmaster’s salary was no fortune, they took their children abroad every other year to Austria, France, Spain and other parts of Europe. Eve went on to teach German to sixth-form students at her old school of St. Catherine’s, and in the early 1960s she gained an extra-mural diploma in archaeology from the University of London.

In 1968 after Max had retired from Cranleigh School he and Eve were able to move to a part of the country that they had always loved; they had already bought the house Chancel End in St. Johns churchyard. When they moved to Devizes they took an active part in WANHS and were fortunate that several of their friends from Surrey also moved to Wiltshire. But Max died suddenly in 1970 and Eve had to resign herself to what she knew would be long years of widowhood. Her love of working with young people and her happy memories of the Girl Guides in her youth led to her

271

becoming a District Commissioner for Guides, much to the astonishment of her family, who had never seen her in a uniform before (or since)!

She became very active on behalf of this Society, where her speciality was the classification of flint implements. She particularly enjoyed leading field parties of flint hunters for the young people’s section of the Museum. Her passion for archaeology was reinforced when the eminent archaeologist, Peggy Guido, moved into 44 Long Street next door to Chancel End. In this way, a close friendship began which was to last until Peggy’s sudden death five years ago. Eve helped Peggy in the massive task of classifying all the beads found in European archaeological sites, an area previously neglected, and travelled with her to Europe on several occasions where her knowledge of German and French often came in useful. Although Peggy herself had not lived to see her work published Gn 1999), Eve was enormously encouraged by the fact that it had been so well produced and that her own part in it had been generously acknowledged.

Another strand in Eve’s life was her poetry. She had written poems when she was young, but late in life the muse spoke to her again — and to great effect. Her poems were published in small but highly regarded magazines, and she was very fortunate that her friends in Devizes included the artists Graham and Ann Arnold, who helped her publish two beautifully printed volumes with illustrations by themselves and other members of the Brotherhood of Ruralists. At the time of her death Eve was particularly delighted that her latest poem had just appeared in Agenda, and that there had been talk of her collected works being published.

In later life Eve, who was a believing but broad- minded Christian, became a member of the Society of Friends, and through them she acquired a wide circle of friends in and around Devizes who were much valued by her and a great support to her. Although to avery large extent she retained her sight, her hearing, and her fierce and inquiring intellect to the end of her life, she was becoming progressively less mobile as the result of an inner-ear infection that had affected her balance. Peggy’s death had hit her hard, and it was less and less easy for her to visit friends and family. Her great comforts were her home in Chancel End, and the kindness of the people of Devizes. When it came, her death by heart attack was sudden, and quite painless. She leaves behind her two surviving children, six grandchildren and one great-grandchild.

Adapted from the funeral address given by her son, BLAISE MACHIN

212 THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Michael Lansdown, newspaper editor, writer, historian, died 20 December 1999. He was born 25 November 1916.

Michael Lansdown was born at Trowbridge in 1916, the son of Charles Lansdown and great grandson of Benjamin Lansdown who founded the Trowbridge Advertiser (predecessor of the Wiltshire Times ) in 1854. Another great grandfather was William Millington, the artist whose paintings and prints of Victorian Trowbridge and its people are well known. Michael was educated at Trowbridge Boys’ High School and read Modem Languages at Corpus Christi College, Cambridge. After war service in the Royal Signals which took him to India and Burma he joined the staff of the Wiltshire Times in 1946, and succeeded his cousin Leonard Lansdown as editor in 1957. In 1961 ownership of the Wiltshire Times was transferred from the Lansdown family to the Westminster Press, Michael remaining editor until his retirement in 1981. He died after a brief illness on 20 December 1999.

As a local historian Michael Lansdown naturally concentrated on Trowbridge as revealed by the files of his own newspaper, and later, after their rediscovery in Bath, those of its rival, the Trowbridge Chronicle. His knowledge of Victorian Trowbridge was encyclopaedic, and he was frequently able to quote verbatim from the files from memory. Many finished little pieces of local history appeared in the columns of the newspaper during his time there, and he joined as co-editor in three books of old photographs of the town. He also published pamphlets on Trowbridge’s fight for pure water, on the Trowbridge Chartists and the stained glass windows of St. James’s Church, ‘Trowbridge.

Michael Lansdown served as honorary treasurer of the Wiltshire Record Society for 48 years, and managed its finances with a success probably unparalleled by any similar body. The enthusiasm and humour with which he reported the intricacies of postage, packing, and covenants and the quirks of printers and booksellers will remain a vivid memory with all who served on the Society’s committee. He was photographed at the Record Office only a few days before his death, reading a file of the Wiltshire Times, in preparation for a memoir to be inserted in a future volume to mark his half century in office.

He was long a member of the West Wiltshire branch of the Historical Association and in recent years its president. He regularly attended the tours in this country and abroad which the Historical Association organized and it was through them that he met his wife, Dorothy, whom he married in 1972. They were regular attenders at the concerts of the Trowbridge Philharmonic Choral Society and the Trowbridge Orchestra, and took an active part in the affairs of the Trowbridge Civic Society and the Friends of the Trowbridge Museum. Both attended St. James’s Church, and Michael served for some years on the P.C.C.

Regularly to be seen round the town (he did not drive a car), he never failed to have some new fact he had discovered, or some happening which had amused him to report. As a raconteur he was, in the writer’s experience, unequalled. His recall of events from childhood, schooldays, army, newspaper office, holidays, seemed detailed and perfectly complete. He will be sadly missed.

KEN ROGERS

Wiltshire Archaeological & Natural History Magazine, vol. 93 (2000), pp. 273-282

Index by Philip Aslett

NOTE: Wiltshire places are indexed or referenced under civil parishes.

Abingdon (Oxon), 167, 269

AC Archaeology. evaluations, 259-60, 262, 263-4; excavations, 255, 257, 258

Addington, Henry, Viscount Sidmouth, 10

Adshead, Stanley Davenport, 210

adzes: Neolithic, 153; Roman, 227

aerenchyma, remains, 171

aerial photography, 90-2, 96, 103, 140, 240, 257, 261

Agenda, 271

agriculture, Romano-British, 233, 237

Agrostemma githago (corn cockle), remains, 42, 43, 44

Ainsworth, James, 55

Albania, 16

Albert, Prince, 16.

Aldbourne, 269; Aldbourne Chase, 108; Lewisham Castle, 108

Alderbury, 197; kilns, 37

Alfred, King, 183, 201

Alington, Cyril, 211

Alismataceae (water plantains), 171

Alisma spp. (water plantains), 171

Allan, Edith, 228

Allason-Jones, L., 225

All Cannings, Tan Hill, 239

Allen, Michael J.: note on buried soil at St John’s Hospital and South Street, Wilton, 200; note on marine shells from Ivy Street/ Brown Street, Salisbury, 45

Allsop, Kenneth, 266

Alton, 249; Alton Barnes, 253, 255; Alton Priors, 255; church, 255; Honeystreet, 255

Alvestone (Avon), 265

America, 252

Amesbury, 66; Boscombe Down, 255; Boscombe Down Airfield, 255; Butterfield Down, 76, 77, 79, 81, 255; Coneybury, 166, 167, 168; Gauntlet pipe factory, 38; London Road, 255; Normanton Farm, 67; Pitt’s Garage, 255—6; Queensbury Bridge, 255; Ratfyn Barrow, 255, see also Stonehenge

amphibians, bones, 45

Ancient Monuments Laboratory, 2, 260, 264

Andover (Hants), 66; faunal remains, 49, 50

Andrews and Dury, map (1773), 250

Andrews, Phil: note on excavations at Vale’s Lane, Devizes, 241-8; report on excavations at St John’s Hospital and South Street, Wilton, 181—204

Anethum graveolens (dill): seeds, 42; uses, 44

Anglo-Swiss Condensed Milk Company, 268

animal bone see bone, animal

animal remains see faunal remains

Broad Hinton, 67, 228, 229, 230; Bicknoll Castle, 114, 115

Broad Town, 257; bronze axe, 238-9

Bromham: church, 87; Spye Park, 250

Bromus spp. (brome-grass), seeds, 42

brooches: Late Iron Age, 261; Roman, 220-5; Romano-British, 100, 102, 185, 198; Aucissa type, 223; Birdlip Brooch, 223; Colchester derivatives, 220-2; dolphin type, 222; Drahtfibel derivatives, 222; Late LaTeéne, 222—3; Nauheim derivatives, 222; penannular, 225; Polden Hill type, 220; Trumpet Type, 223-5, see also fibulae; strip brooches

Brooks, Mr, 56

Brooks, N., 265

Broomfield, Michael, 238

Brotherhood of Ruralists, 271

Browne, William, 132

Brown, Graham, 257, 264

Brown, John, 132

Brussels (Belgium), 65

Bruton (Somerset), 112

Buccinum undatum (whelk), 45

buckles, Roman, 225

Bucks see High Wycombe

Budden, Thomas, 53

building materials: medieval, 22, 257, 258; post-medieval, 243, 258; ceramic, 37-8, 196-7, see also brick; chalk blocks; tiles; wattle and daub

buildings: Late Bronze Age/Early Iron Age, 263; Roman, 231; Romano-British, 90, 92, 99; Saxon, 90, 185, 233; medieval, 26-8, 57-9, 257, 258; post-medieval, 259, 261; 16th century, 203; as war memorials, 210, see also villas; walls

Bulford, 18

Bull family, 125

Bull, John, 125

Bullock, Troyte, 268

Bupleurum rotundifolium (thorow-wax), 44

Burcombe Without, 185, 202

Burghal Hidage, 183, 185, 201, 202

burials see inhumations

Burl, Aubrey, work noted, 269

Burma, 272

Burton (Glos), 83

Burton, Thomas, 83

Bury, Lady Charlotte, 12

Busshel, William, 53

butchery: Neolithic, 154, 156, 158, 162, 166— 7, 175; Beaker, 78-9; Romano-British, 198, 200; medieval, 50; post-medieval, 51

Butler, Charles Ponsonby, 254

Button family, 249 Button, Sir Robert, 249 By Brook, 103, 219 Byron, Lord, 16

Caerleon (Gwent), 222

Calluna vulgaris (heather), 42

Calne: burial registers, 120, 122; Castle House, 120, 125-7; Castle Street, 125; church, 120, 122-8; doctors, 268; fortifications, 108; Ingen Housz in. 120-30; Quemerford, 128

Calne Without: Calstone, 125; Derry Hill, 124; Quemerford, 128, see also Bowood

Caltha palustris (marsh marigold), 44

Cambridge: Corpus Christi, 272; Magdalene College, 212-13; Trinity College, 16

Cambridgeshire see West Wickham

Camden, Ist Marquess of, 10-12

Camerton (Avon), 223

Campbell College (Ireland), 210

camps, temporary, 237

Camulodunum (Essex), 227

Cannabis sativa (hemp), seeds, 43

Canterbury, Archbishop of, 181

Canterbury (Kent): cess pits, 45; Prerogative Court, 54

Carex spp. (sedge), 44, 200

Carlisle (Cumbria), 240

Caroline, Queen, 12

Carrawburgh (Northumberland), 227, 240

Case, Humphrey, 76

Castle Combe, earthwork, 114

Castlereagh, Robert Stewart, Viscount, 10-12

castles, 261; Norman, 105, 107; medieval, 241-8; classification, 116-17; early, 105— 19

Catesby, Sir John, 87

cats, bones, 45,51, 198

cattle: bones, 45, 48, 49, 50, 51, 59, 79, 157, 160-5, 166, 198, 200, 261; teeth, 102, 155, 160, 162, 165

Caundel, John, 52, 53

celts, Neolithic, 134

cemeteries: Bronze Age, 260; Romano-British, 90, see also inhumations

Centaurea cyanus (cornflower), charred grains, 43, 44

ceramics see pottery

Cerastium spp. (chickweeds), charred seeds, 200

cereals: charred, 168, 172, 173, 174; remains, 42,43, 44, 45, 131, 153-4, 168-9, 170

cess pits: medieval, 24, 28, 39, 43, 45, 48, 49, 50, 59; post-medieval, 51, 191

274

Chaffin, Mr, 55

Chaffyn, Mistress C., 54

Chaffyn, Thomas, 53, 54

Chafin, Thomas, 54, 56

Chafin, William, 54, 66-7, 270

Chafyn, Charles, 54

Chafyn, Dorothy, 54

Chafyn family, 57

chalk blocks, 24, 141, 144

Chandler, John: documentary report relating to excavations at Ivy Street/Brown Street, Salisbury, 51-7; work noted, 268

Channel Islands, 50, see also St Clement

charcoal, 80, 141, 144, 168, 170, 173-4, 176, 258

Charlton Musgrove (Somerset), Cockroad Wood, 112

Charterhouse (Surrey), 210

charters, Anglo-Saxon, 185, 265

Chartists, 272

Chedworth (Glos), 228, 229

Chenopis atrata (black swan), 70

Chenopodium murale (goosefoot), seeds, 170, 174

Cherhill, 167; Yatesbury Field, 134

Chester, cess pits, 45

Chesterholm (Northumberland), 240

Cheverell, Robert, 87

Chicklade, manor, 84, 86, 87

Chilmark, stone, 188

Chippenham, 84; Bell Inn, 86; Cock, 70; gas pipeline, 90-104; hundred, 264; Market Place, 108; Swan, 70;Western Bypass, 233— 8

Chippenham College Archaeology Group, 103

Chippenham Without: Allington, 258; Corsham Wood, 258; Sheldon Manor, 257-8; Sheldon Wood, 258

Chiseldon, Burderop Down, 148

Chitterne, great bustards, 67

Chokke, Richard, 83, 84, 85

Chrysanthemum segetum (corn marigold), charred grains, 43, 44

Church, Rosemary, work noted, 268

Chute, William, 10

Civic Arts Association, 209

Clarence, Duke of, 12, 15

Clarendon Park, 250; Ashley Hill, 31; Clarendon Palace, 29, 37; Fussell’s Lodge long barrow, 166; Petersfinger, 31; tile industry, 38

Clark, Dorcas, 254

Clarke, Bob, 255; note on bronze axe from Broad Town, 238-9

Clark family, 249, 254

Clark, John, 254

Clark, Thomas, 254

clay pipes, 6, 7, 38

Cleal, Rosamund M. J., report on excavations

at Crescent Copse, Shrewton, 71-81

cloth trade, Salisbury, 53

Clutterbuck, Daniel, 250

Clwyd see Prestatyn

Clyffe Pypard, 115; Stanmore, 115

Clyro (Powys), 266

Cnut, King, 265

Cobbett, William, 9

(Vindolanda)

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Codford, 87; East Codtord, 86; West Codford, 86

coffins, lead, 123-4

coins: Iron Age, 240; Roman, 220, 228-30, 240; Romano-British, 228-30, 240

Colchester (Essex);. 29, 223, Camulodunum, 227

Colerne: Ashley Barn, 220; Bury Wood Camp, 90, 103; Euridge [Ewerigga/Iwerugge/ Uridge], 219; Euridge Manor Farm, 218— 32

Cole, William, 70

Collingbourne Kingston, barrows, 239

Collingwood, R. C., 228

Colnbrook, Sir George, 125

Columella edentula (mollusc), 78

combs: medieval, 197; bone, 38—9

Comper, Sir Ninian, 214

Compton Chamberlayne, 66

Compton, Lord, 10

Compton, Mary, 249

conger eels, bones, 49, 51, 59

Conium maculatum (hemlock): charred seeds, 200; uses, 44

conservation, and fungi, 80-1

Constantinople, 16

Coombes, John, 55

Cooper, Charles, 250

Cooper, Frances Sarah, 250

Cooper, John, 250

Cooper, John Allen, 250, 252

Cooper, Mary, 252

Cooper, Thomas, 250

copper alloy objects: Roman, 220-7, 227; medieval, 29; waste, 197

cores, 134, 148-9, 151, see also flintwork

Coriandrum sativum (coriander), uses, 44

Corio, 103

Cornburgh, Avery [Alfred], 85

Cornhill Magazine, 212

Cornwall, 85, 239, see also East Looe; Penryn; ‘Trevelgue

Corsham, 16; church, 87; Corsham Court, 183 Hartham Park, 115; Heywood Preparatory School, 258; Neston, 83, 85; Rectory Manor, 258

Cotswold Aggregates, 259

Cotswold Archaeological Trust, 77; evaluations, 241, 259, 262; excavations, 90-104, 233-8

Cotswolds, 102, 103, 237

Country Life, 209

Country Magazine, 269

Cowley (Oxon), 219, 228

Cranborne Chase, 132, 270

Cranborne Downs (Dorset), 70

Cranleigh (Surrey), 269, 271

Crataegus monogyna (hawthorn), seeds, 169

Crawford, OG: S., 132

Crawford, T. S., work noted, 268

Creighton, Oliver H., paper on early castles in medieval landscape, 105-19

Cricklade, 196; fortifications, 108; High Street, 258; Horse Fair Lane, 258; Three Horseshoes Inn, 258

Croft Ambrey (Hereford and Worcester), 222

cropmarks, 90—2, 94, 96, 103, 240

Cross, Mr, 123

PYAR

Crummy, N., 225, 227 Crutcher, Thomas, 56, 57 culverts, Romano-British, 96 Cumbria see Carlisle Cunetio, evaluations, 259-60 Cunnington, Maud, 2 Cunnington, William, 67, 132 cups: Late Neolithic, 146; Romano-British, 240 cutlers, 16th century, 248 Cyperaceae (herbs), 171

Daily Telegraph, 63

Dale, Mr, 132

Danebury (Hants), 239

Danes, campaigns, 183, 201

Darvill, T., 77

David, Andrew, 2

Davies, Jessica, report on excavations at Windmill Hill causewayed enclosure, 131— 80

Davies, Stevie, 266

Dawson, W. R., 70

de Dunstanville family, 114

deer: antlers, 162, 166; bones, 45,51, 96, 103, 156, 198; teeth, 101

Defence Estates Organisation (Lands) South- West, 71

Defence Evaluation and Research Agency, 255

defences: Late Saxon, 181—204; medieval, 241

de Keynes family, 111

Delmé, Peter, 9

Deneman, Margaret, 52

Denham, Thomas, 53

Dennis, Ian, report on excavations atWindmill Hill causewayed enclosure, 131-80

Devizes, 10, 12, 13, 18, 67; Baptist Chapel, 247; Bear Hotel, 15; Bear Inn, 14; Black Swan, 70; Castle, 241; Chancel End, 271; Corn Exchange, 247; Corporation, 18; Hare and Hounds Street, 241; King’s Arms, 250; Long Street, 247; Market Place, 247; New Park Street, 241, 246, 247; New Port, 241; Pelican, 70; ringwork, 107; St John’s church, 247; Sheep Street, 247; Vale’s Lane, 241-8; Wayside Farm, 258; White Swan, 70

Devizes Museum, 68, 132, 134, 219, 220, 227, 230, 233, 239, 241, 249, 271

Devizes and Wiltshire Gazette, 13-15

Devon see Exeter; Plymouth

Dew, 67

Dewel, John, 52

Dinton, Baverstock, 66

Discus rotundatis (mollusc), 78

dishes, medieval, 33

Disraeli, Benjamin, Ist Earl of Beaconsfield, 16

Disraeli, Mary Anne, 16

ditches: Neolithic, 2-4, 6, 175; Early Neolithic, 146; Late Neolithic, 233, 261; Early Bronze Age, 233; Late Bronze Age/Early Iron Age, 262; Iron Age, 92, 96, 259, 260; Middle Iron Age, 233; Roman, 102; Romano-British, 92—4, 96, 99, 100, 103, 233, 236, 261; Saxon, 183, 185, 186-8, 201; medieval, 189, 191, 202, 255, 262; post-medieval, 257; ?modern, 189; ring,

INDEX

92, 141, see also gullies; pits

Dobunni, 240

dogs, bones, 45, 51, 158, 198

Dole, hundred, 114

Domesday Book, 52, 114, 183, 267

Donne, John, 269

Donyatt (Somerset), pottery, 36, 37

Dorchester (Dorset), 181

Dordogne (France), 271

Dorn (Glos), 228, 229

Dorset, 268; literature, 265—6; pottery, 31, see also Blandford; Cranborne Downs; Dorchester; Folke; Maiden Castle; Poole; Purbeck; Ridgeway; Shaftesbury Abbey; Sydling St Nicholas; Verwood; Waddon Hill; Woodcutts

Douai, Walscin of, 115

Douglas, Mr, 66

Dove, Peter, 53

Doves, Mr, 55

Downes, Richard, 267

Downman, E. A., 105

Downside School (Somerset), 210

Downton, 80; The Borough, 258; Castle Meadow, 108; Charlton, 53; Moot House, 107; Old Court, 108; ringwork, 107-8

Driesch, A. von den, 45

Driver, J. T., paper on career of Thomas Tropenell, 82-9

Drummonds (bankers), 122, 123

Dubois, Mr, 66

Duckett family, 125

Duckett, Thomas, 125

Duffy, Maureen, 266

Dumont, Rev, 122

Durdall, Thomas, 54

Durdall, Widow, 55

Durnford, Little Durnford, 86, 87

Durrington, Durrington Walls, 100, 166

Dycar, Nicholas, 53, 54

Eames, George, 10

earthworks, 107, 111; medieval, 264; interpretation, 105, see also mottes; ringworks

East Anglia, beakers, 76

East Anton (Hants), 228, 229

Easterton, Eastcott, 67

East India Company, 125

East Knoyle, Milton, 86

East Looe (Cornwall), 12

East Sussex see Hastings; Seaford

East Yorkshire, 239

Eaton Socon (Beds), 114

Ebbesborne Wake, Fifield Bavant, 100

Edinburgh (Scotland), 16, 120, 210; Academy, 210

Edington, 87;Tinhead, 67

Edmonds, M., 166

Edmonstone, Mr, 15

Edmunds family, 266

Edward IV, King, 84, 85

Edwards, Brian, paper on Marlborough College Memorial Hall, 205-17

eels: bones, 45, 49; in medieval period, 50, 59

Eleocharis spp. (spike-rush), 44

enclosures: Neolithic, 1-8; Iron Age, 103; Romano-British, 233; causewayed, 131—

80; palisaded, 131, see also hillforts; settlements

English Heritage, 2

Enright, Dawn, note on excavations at Chippenham Western Bypass, 233-8

environmental materials, from Ivy Street/ Brown Street, Salisbury, 39-51

Erlestoke: Erlestoke Park, 9-19; prison, 9, 19; Senior Officers Training School, 18-19; ‘Stoke Park’, 10; village improvements, 10

Erley, John, 83

Essex see Camulodunum; Colchester; Harlow; Woodham Ferrers

Estcourt, Mr, 12

Estcourt, T. G. B., 15

Ethelred II, King, 112

Eton College (Berks), 211

Euphrasia spp. (eyebright), seeds, 43

Evans, John, 1, 2, 132

Evelyn, John, 65

Evil and Co., 252

Eward, Suzanne, 51

excreta: medieval, 26, 43; fruit stones in, 44

Exeter (Devon), 90

fabricators, Neolithic, 151

faeces see excreta

Fairbairn, Andrew S., report on excavations at Windmill Hill causewayed enclosure, 131-80

Falconer, William, 121

Farleigh Hungerford (Somerset), 87

faunal remains: classification, 45; from Crescent Copse, Shrewton, 78; from Ivy Street/Brown Street, Salisbury, 45-51, 59; identification, 45

Fawconer, Jasper, 53

Fettes School (Scotland), 210

fibulae: Roman, 220; Alesia type, 223; Aucissa type, 223

Ficus spp. (fig), seeds, 42

field surveys: Chippenham Western Bypass, 233; Euridge Manor Farm, Colerne, 231; Westbury, 263; Windmill Hill causewayed enclosure, 137

field systems, 261; Romano-British, 99, 103

Figheldean, Robin Hood’s Ball, 77, 79, 80, 153

Findon (West Sussex), 66

finials, medieval, 37-8

First Pembroke Survey (1563), 203

First World War see World War I

fish: bones, 48, 49, 50; in medieval period, 50, 59; recipes, 50

Fisher, Mr, 15

flagstones: 17th—18th century, 28; limestone, 27

flakes, flint, 90, 102, 134, 149, 151, 152, 236; notched, 149; serrated, 149, 153; utilised, 149, 152, 153, see also flintwork

Fleming, Ann, 250

Fleming, Edward, 250

flints: burnt, 73, 152, 154, 261, 263, 264; nodules, 188, 189; raw materials, 148, 236; scattered, 92, 102, 131, 137, 148, 151

flintwork: prehistoric, 90, 102, 256, 264; Mesolithic, 102, 103, 235, 236, 257, 263; Neolithic, 4, 5, 103, 131, 132-7, 148-54, 176; Early Neolithic, 141, 257, 263; Late

275

Neolithic/Early Bronze Age, 263; Beaker, 73, 77-8, 79, 80; Bronze Age, 235, 236; Middle Bronze Age, 261; Late Bronze Age, 263; classification, 271, see also adzes; arrowheads; awls; axes; blades; celts; cores; flakes, flint; hammerstones; Knives; scrapers; tools

Foch, Marshall, 213

Foeniculum vulgare (fennel), uses, 44

Folke (Dorset), 84

Font de Gaume (France), 271

food, processing, 131, 166-7, 172-3, 176

Foote, Robert, 249

Forster, E. M., 266

Fortescue, John, 83, 84

Fosse Way, 90, 92, 102

Fowler, Richard, 7

Fowles, John, 266

fowls, bones, 198

Fox, Caroline, 126

Fragaria spp. (strawberry), seeds, 42, 44

France, 271, see also Arles; Brittany; Dordogne; Font de Gaume; Lascaux; Loos; Lugdunum; Lyons; Morvan; Paris

Franco-Prussian War, 68

Franklin, Benjamin, 121

Freeman, Jane and Watkin, Aelred, work noted, 268

French Revolution, 121

Fripp family, 125

Frome, River, 265

Froxfield, Rudge Romano-British villa, 239— 41

fruits, 59, 168-75; analyses, 43—4

fruit stones, remains, 42, 43, 59

fuels, plant, 173

funerary monuments, prehistoric, 90

fungi, arboreal, effects on archaeological remains, 71, 80-1

furnaces, Late Saxon, 197

Gadebridge (Herts), 223

Galium spp. (herbs), 169-70

Gauntlet pipe factory, 38

geese, bones, 198

Gentleman’s Magazine, The, 126

geophysics: Barbury Castle, 260; Euridge Manor Farm, Colerne, 220;Windmill Hill causewayed enclosure, 137—41

George III, Kung, 10

George IV, King, 12, 15

Germany see Trier

Gerrish, Thomas, 252

Gibbes, Bridgett, 54

Giffard, Osbern, 114

Gifford, Hazel, work reviewed, 267

Gifford and Partners, excavations at Crescent Copse, Shrewton, 71-81

Gilbert, Sarah, 10

Gillard, Frank, 270

Gillings, Mark, report on excavations at Beckhampton Avenue, 1—8

Gilpin, 65

Girl Guides, 271

Glamorgan, castles, 117

Glasgow (Scotland), Academy, 210

glass, medieval, 24, 38

Gloucester, Robert, Earl of, 114

276

Gloucestershire, 103, 266, see also Burton; Chedworth; Dorn; Hartpury; South Cerney; Tewkesbury

goats: bones, 96, 157, 158, 160—5, 166, 198; teeth, 162

Goddard, E. H., 105, 132

Godshill (Hants), 108

Golding, William, 266

gold objects, Romano-British, 239-41

Goldsborough family, 266

Gomme, Alison, 19

Goodfield, M., 239

Gosport (Hants), 70

Gower, Widow, 55

gradiometers, 137

gradiometer surveys, 137-41

Grant (farmer), 67

grasses, seeds, 42, 45

gravel pits, Saxon, 108

Graves, 67

Gray, Thomas, 214

Great Bedwyn, 83

great bustard, in Wiltshire, 63-70

Great Bustard Trust, 68

Great Chard (Kent), 70

Great Somerford, 117; church, 114; The Mount, 112-14

Great War see World War I

Greece, 16

Green & Co, Messrs, 18

Greenwood, Thomas, 122

Griffin (farmer), 7

Grigson, C., 165

Grigson, Geoffrey, 266

Grime’s Graves (Norfolk), 132

Grimes, John, 53

Grimes, Mary, 53

Grimstead, West Grimstead, 31

Grinsell, L. V., 140

Grist, Stephen, 54

Grittleton, 102, 115; Green Barrow Farm, 92

Grove, Agnes, 270 :

Grove, Harriet, 270

Grove, Tom, 270

Guernsey, 18

Guido, Peggy, 271

Guildford (Surrey), 269

Guild Merchant, 183

gullies, 235, 236; Late Bronze Age, 261, 263; Late Bronze Age/Early Iron Age, 262; medieval, 233, see also ditches

Gumbleton, John, 55

Gwent see Caerleon

Habsburg family, 121

Hadley (Middlesex), 249

Hadrian’s Wall, 240

Hamilton-Dyer, Sheila: note on animal bones from Crescent Copse, Shrewton, 78-9; note on animal bones from St John’s Hospital and South Street, Wilton, 198-9; note on faunal remains from Ivy Street/ Brown Street, Salisbury, 45-51

Hamilton, Michael A., report on excavations at Windmill Hill causewayed enclosure, 131-80

hammers, 248

hammerstones, Neolithic, 151, 152, 153,154

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Hampshire: pottery, 36; tile industry, 37, see also Andover; Danebury; East Anton; Godshill; Gosport; Kings Somborne; New Forest; Owlesbury; Romsey; Solent; Southampton; Sparsholt; Wherwell; Winchester

Handle, Cristina, 53

handles, medieval, 197

Harcros Timber Merchants, 260

Hardell, Robert, 86

Harding, Margaret, 266

Hardy, Thomas, 265, 266, 270

hares, bones, 45, 48, 50, 198

Harlow (Essex), Harlow Museum, 225

Hartley, B., 219, 227

Hartpury (Glos), 249

Harvey, R., review by, 265

Haslam, J., 181, 183, 201, 203, 241, 246-7

Hastings (East Sussex), cess pits, 45

Hather, Jon, 168

Hatt, Robert, 54

Haughton Court (Jamaica), 12

Hawkes, C. F. C., 225

Hawkins, Barbara, 269

Hawkins, Desmond, obituary, 269-70

Hay, Lady Charlotte, 18

Haydon Wick, Moredon, 115

Hay, Lady Julia, 16

hazelnuts, remains, 169, 170

hearths: Late Bronze Age, 261; Saxo-Norman, 191, 197, 201

hearthstones, Neolithic, 154

heathers, seeds, 42, 43

Heaton, Michael, report on excavations at Crescent Copse, Shrewton, 71-81

Helbaek, H., 169

Helix spp. (mollusc), 78

henge monuments, 269

Henig, Martin, note on ring from Froxfield, 239-41

Henry I, King, 108, 114

Henry VI, King, 84, 86

Herbert, Thomas, 84

Hercules (god), sculpture, 219

Hereford, Viscountess, 125

Hereford and Worcester see Croft Ambrey

Hertfordshire see Braughing; Gadebridge; Puckeridge; St Albans; Verulamium

Heytesbury, 52, 261; Heytesbury House, 259; hundred, 114; Park Street Gates, 259

Heywood: Down Farm, 90; Lodge Farm, 90

Hibberd, John, 55

Hickey, William, 250

Higgins, Peter, note on palaeoenvironmental remains from Crescent Copse, Shrewton, 78

Highworth, Swan, 70

High Wycombe (Bucks), 125

Hill, David, 183

hillforts: Iron Age, 90, 108, 116, 260, 262, see also enclosures; specific sites

Hill, Henry, 250

Hillman, Gordon, 168

Hinde, Thomas, 211

Hindon: manor, 84, 86, 87; Swan, 70

Hinton, Pat: note on plant remains from Ivy Street/Brown Street, Salisbury, 39-45; note on plant remains from St John’s Hospital

and South Street, Wilton, 199-200

Historical Association, 272

Hobbs, Steven, reviews by, 266-7, 267-8

Hobhouse, Charlotte, 16

Hobhouse, John Cam, Ist Baron Broughton de Gyfford, 9, 16-18

Hobhouse, Julia, 16, 18

Hobhouse, Sophia, 16

Hodding, John, 66

Hofmann, Richard Beer, 271

Holgate, R., 2

Holland, 126

Holland, Henry, Lord, 250

Holland (Jamaica), 12

Holt: Great Chalfield, 82—9; Little Chalfield, 84, 85

Holt, T. E., work noted, 268

Hordeum spp. (barley), charred grains, 42, 43, 200

Hordeum vulgare var. nudum (naked barley), grains, 169, 173

horncores, 24

horses, bones, 166, 198

Houghton Hall (Norfolk), 12

Houghton (West Sussex), Bury Hill, 6

Howell, Danny, work noted, 268

Hudson, W. H., 68-9, 266

Hullavington, 115

Hull, M. R., 225

Hundred Days, 16

Hungerford, 68

Hungerford, Sir Edmund, 84

Hungerford family, 52, 83, 84, 86

Hungerford, Robert, 2nd Baron Hungerford, 84, 86

Hungerford, Robert, Baron Moleyns and 3rd Baron Hungerford, 84, 85, 87

Hungerford, Thomas, 52

Hungerford, Sir Thomas, 85, 87

Hungerford, Thomas de, 52

Hungerford, Sir Walter, 83

Hungerford, Walter, 85

Hunstanton (Norfolk), 65

Hunter, John, 121

Hunter, William, 121

Hunt, Thomas, 38

Hurst Fen (Suffolk), 148, 149

Hussey, William, 66

Hydra, sculpture, 219

Hyoscyamus niger (henbane): charred seeds, 200; uses, 44

Ide, Isabel, paper on Erlestoke Park, 9-19

Idmiston: Gomeldon, 35; Porton Down, 68

Imber, 84; silt lagoons, 259

India, 272, see also New Delhi

infusions, 44

Ingen Housz, Agatha (née Jacquin), 122

Ingen Housz, Jan, biographical notes, 120-30

inhumations: Beaker, 2; Late Bronze Age/Early Tron Age, 262-3; Saxon, 116; medieval, 116; vault burials, 123-8

inn names, ornithological, 70

inoculation, smallpox, 121

insect remains, from Crescent Copse, Shrewton, 78

Ireland, 10, 239, see also Belfast; Campbell College; Knowles collection

INDEX

iron objects, 261; Roman, 222-3, 227; Romano-British, 185, 198; Saxo-Norman, 198; medieval, 29, 198; post-medieval, 198; modern, 140

iron slag, 181, 189, 191, 197, 202

iron smelting, ?Saxon, 183, 191, 197

Isle of Wight see Newport

Italy see Milan; Ticinum

Jackson, Gwyneth F., work reviewed, 267-8

Jacquin, Josef, 122

Jacquin, Nikolaus, 121

Jamaica, 15, 16, see also Haughton Court; Holland; Llanrhumney; Lyssons; Sauls’ River

James, Henry, 266

James, M. K., 181

jars: Romano-British, 94; Late Saxon, 196; medieval, 31, 35, 37

Jatt, Thomas, 53,55

Jefferies, Richard, 266

Jersey see St Clement

Jerveys, Isabelle, 86

Jerveys, William, 86

jewellery see beads; brooches; rings

Jews, in Wilton, 183

Jope, E. M., 7

Jordan, Temys, 125

Joseph I], Emperor, 126

jugs, medieval, 33, 35

Keiller, Alexander, 6-7, 131, 149, 151, 269; and Kendall, 132-4

Kelly, S. E., work reviewed, 265

Kelston (Avon), 265

Kemp, Mr, 13

Kendall, H. G. O., 131; biographical notes, 132-4

Kennet, River, 108, 115, 257

Kennet Valley, pottery, 243, 245, 246

Kent see Canterbury; Great Chard; Romney Marsh; Sheerness; Wye

Kent, Duchess of, 12, 15

Kew Gardens, 18

Keynsham (Avon), 220

Kidston, G. J., 249, 253, 254

kilns: Romano-British, 94, 147; medieval, 30— 1, 36, 243, 244-5; post-medieval, 35

Kilvert, Francis, 266

King, Mr, 68

Kings Somborne (Hants), 202

Kingston Deverill: Court Hill Plantation, 264; The Park, 264

Kingston upon Thames (Surrey), 181

Kington Langley, 102

Kirby, Colin, 255

knife finials: horse-head, 248; monkey-head, 248-9

knives, 198; Neolithic, 134, 149-51, 152, 153; Beaker, 151; Roman, 227; medieval, 29; 16th century, 248-9

Knook: Knook Castle, 261; pottery, 35; Quebec Barn, 261;Willis’s Field Barn, 261

Knowles collection (Ireland), 134

Lacock: kilns, 244—5; Naish Hill, 38, 243, 245, 246, 257; Showell Nursery, 233 Lake District, 265

Lakenheath (Suffolk), 227

Lamb, George, 16

Lamiaceae (mints), seeds, 170

Lancaster, Duke of, 111

Lancastrians, 83, 84, 85, 87

Langley Burrell Without, 67

Lansdown, Benjamin, 272

Lansdown, Charles, 272

Lansdown, Dorothy, 272

Lansdowne, Ist Marquis of see Petty, William, 1st Marquis of Lansdowne

Lansdowne, 3rd Marquis of see Petty- Fitzmaurice, Henry, 3rd Marquis of Lansdowne

Lansdowne family, 16, 124

Lansdown, Leonard, 272

Lansdown, Michael, obituary, 272

La Plante, Linda, 19

Lark Hill (Worcs), 225

Lascaux (France), 271

Latton: Latton Lands, 259; The Street, 259

Laverstock: kilns, 30-3, 195; pottery, 35, 37, 193, 196, 243, 245, 246

Lavis, Peter, work noted, 268

Lavoisier, Antoine Laurent, 121

Lawrence, D. H., 269

Lawrence, Margarett, 54

Lawrence, Sir Thomas, 15

lead objects, medieval, 29

leather, medieval, 39, 198

Legge, A. J., 165

Leicester, University, 1

Lemna spp. (duckweeds), charred seeds, 200

Leonard, William, 253

Leopold, King, 16

LEstrange, Household book, 65

Leyden (Netherlands), 120

Liber Niger (1455), 53

Liddington Castle, 264

Lincoln, 90

Lincolnshire see Torksey

linears, 92, 257; prehistoric, 261; Neolithic, 139-40; Bronze Age, 141, see also ditches

lingulae, Roman, 225

Linnean Society, 63—5

Linum usitatissimum (flax): seeds, 42; uses, 44

Listener, The, 269

Litten, J., 123

Little Bedwyn, Chisbury, 115

Littleton Drew—Chippenham gas pipeline, excavations, 90-104

Little, William, 54

Llanrhumney (Jamaica), 12

Loader, Emma, notes on finds at Ivy Street/ Brown Street, Salisbury, 29, 37-9

Locking (Avon), 228

Lodge, Sir Oliver, 18

London, 120, 121, 125, 126, 254; British Museum, 225; Cavendish Square, 12; Cenotaph, 210; Charing Cross, 122; Cheapside, 65; Drapers’ Hall, 126; Elephant and Castle, 269; Holland House, 16; Horse Guards, 16; House of Commons, 10, 12, 16; Knightsbridge, 16; Lambeth, 9; Lincoln’s Inn, 10, 83; Lords, 269; Newgate Prison, 16; Oval, 269; Royal Academy, 209, 213; St James Palace, 15;

Dil.

St James Street, 15; Soho, 270; Southwark, 240; Surveyors’ Institute, 206; Theatre Royal (Covent Garden), 10; University College, 168; University of, 271;West End, 269; Westminster, 83, 206; Westminster School, 16; Whites’ Club, 15

London, Hugh, 86

Longbourne, David, 266

Longbridge Deverill: Crockerton, 35, 36, 193, 195, 196, 243, 245, 246; Hill Deverill, 83

Long family, 87

Long, Henry, 83, 84, 85

Look, 270

Loos (France), 206

Louis the Dauphin, 108

Louis XVIII, King, 16

Louvain, University of (Belgium), 120

Low Countries, 65

Luckett, Larry, report on excavations at Euridge Manor Farm, Colerne, 218-32

Ludgershall, 83; Castle, 107

Ludlow family, arms, 86

Ludlow, William, 83,86

Lugdunum (?France), 230

Lychnis flos-cuculi (ragged robin), 44; charred seeds, 200

Lyme Regis, 266

Lyneham: Hillocks Wood, 116; pottery, 245; Tockenham Court, 249

Lyons (France), 230, 240

Lyssons (Jamaica), 12

Lywood, E., 68

MacGregor, A., 39

Machin, Blaise, obituary by, 270-1

Machin, Eve, obituary, 270-1

Machin, Max, 271

Machin, Noel, 271

Machin, Tess, 271

McKinley, Jacqueline, 81

Mackreth, D. F., note on finds from Euridge Manor Farm, Colerne, 220-5

MacNeice, Louis, 270

Maerla, barrow of, 108

magnetometer surveys, 264

Magnus Maximus, 229

Maiden Bradley: Church Street, 259; High Street, 259; manor, 86, 87

Maiden Castle (Dorset), 223

Malmesbury, 268; Cam’s Hill, 108, 117; Castle, 114; Swan, 70

Malmesbury Abbey, cemetery, 114

Malus spp. (apple), seeds, 42, 44, 169

Man, 132

Mann and Rodway, 249

Maria Theresa, Empress, 121

marine shells, from Ivy Street/Brown Street, Salisbury, 45

Markes, Edmund, 54

Market Lavington, 65, 66, 68; pottery, 35, 196

Marks, Mr, 55

Marks, Thomas Chafin, 54

Marlborough, 266; Bath Road, 205; Castle, 108; Chandler’s Yard, 259; High Street, 259; pipemakers, 38; pottery, 77

Marlborough College: Cloistered Garth, 209; College Chapel, 207, 209; Council, 207 9, 211, 213; Marlburian Club, 209;

278

Memorial Gardens, 211; Memorial Hall, 205-17; Old Marlburians, 205, 206-10, 211, 213; Roll of Honour, 212; war casualties, 206; War Memorial Committee, 207-9

Marlborough Downs, 79, 132, 260

Marlburian, The, 206, 209, 210, 213

Marshall, Ruth, work noted, 268—9

Marshfield (Avon), Ironmonger’s Piece, 225

Marshman, Michael: obituary by, 269-70; reviews by, 265—6, 267; work noted, 269

Martin family, 267

Martin, Mary, 53

Martin, Thomas, 53

Mary Rose (ship), 49

Maton, G., 67

Maton, Martha, 55

Matthews, Rev. Murray, 68

Maundrell, Henry, 122

May, Thomas, 219

Meaden, Terence, work noted, 269

meat, processing, 131, 166-7

Melbourne, Lord, 16, 18

Melksham, King’s Arms, 252

Membury, hillfort, 116

memorials, war, 205-17

Mentha cf aquatica (water mint), charred seeds, 200

Mepham, Lorraine: note on excavations at Vale’s Lane, Devizes, 241-8; notes on finds from Ivy Street/Brown Street, Salisbury, 29-37, 39; report on excavations at St John’s Hospital and South Street, Wilton, 181-204

Mercury, Temple of, 107

Mere: Mere Down Farm, 264; papers, 266-7; Swan, 70

Mere Historical Society, 266

Mervyn, 85

Mervyn, John, 84

Mespilus spp. (medlar), 45

metalwork: Romano-British, 198; Saxo- Norman, 198; medieval, 29, see also arrowheads; blades; copper alloy objects; gold objects; iron objects; knives; lead objects; nails; steel objects

Methuen family, 16

mice, bones, 45, 48, 50

Middlesex see Hadley (Middlesex)

Milan (Italy), 230

Milbanke, Anne, 16

Mildenhall, 85; Werg Mill, 259-60, see also Cunetio

Milligan, Max, work noted, 269

Millington, William, 272

mill leat, post-medieval, 108

Mills, John, 53

Milvus migrans (black kite), bones, 49

Milvus milvus (red kite), bones, 49

Minety, pottery, 243, 245, 246, 257

minimum number of anatomical units (MNAU) method, 155-6

minimum number of individuals (MNI) method, 155-6

Mitchell, A., 55

Mitchell, Mrs F., 55

MNAU method, 155-6

MNI method, 155-6

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

mollusc remains: Crescent Copse, Shrewton, 78; Windmill Hill, 166, 168

Mompesson, Henry, 87

Mompesson, John, 84—5

Moncaster [Muncaster], Mrs, 253, 254

Montacute (Somerset), 112

Montagu, 67

Moore, George, 55

Moore, John, 266

Moore, Thomas, 15, 16

Moorhead,T. S. N., note on coins from Euridge Manor Farm, Colerne, 228-30

Morales, A., 45

Morgan, John, 38

Morres, A. P., 68

Morris, Johnny, 269

Morrison, H., 219, 220,231

Morrison, Mrs, 220

mortar, 198

Mortimer, Ralph of, 115

Morus nigra (mulberry), 45

Morvan (France), 228

mottes, 108-16, 116-17; use of term, 107

Mozart, Wolfgang Amadeus, 120

Muncaster [Moncaster], Mrs, 253, 254

Murray, Lynda J., work noted, 269

Mustin, George, 53, 54

Mustion, Elizabeth, 53, 54

Myulus edulis (mussel), 45

Nadder, River, 181, 183, 201

nail cleaners, Roman, 225

nails, medieval, 29, 198

Naish, W., 52, 55

Napoleon I (Bonaparte), 16

Napoleonic Wars, 12

National Trust, 256

Naturaliste (ship), 70

Naturalist, The, 270

needles, medieval, 29

Nesovitrea hammonis (mollusc), 78

Netherlands, The, 126, see also Breda; Leyden

Nettleton, 223, 225;The Gibb, 102; Nettleton Shrub, 90

Neville, George, 85

Newbury (Berks), 245

New Delhi, 63

New English Weekly, The, 269

New Forest (Hants), pottery, 103, 227

Newmarket (Suffolk), 250

Newport (Isle of Wight), 12

New Sarum see Salisbury

New Statesman, The, 269

Newton, John, 249

Newton, William, 211

New Zealand, 70

NISP method, 155-6

Norborne, Mary, 125

Norborne, Walter, 125, 126, 128

Norfolk see Grime’s Graves; Houghton Hall; Hunstanton

Normans, 112, 114, 115

North America, fungi, 80

Northampton, 197

Northants see Sulgrave

North Perrott (Somerset), 253

Northumberland see Carrawburgh; Vindolanda

North Wraxall, 103; Truckle Hill, 90

Norwood, Cyril, 211

Nottingham High School, 210

number of individual specimens (NISP) method, 155-6

nuts, remains, 169, 172

Oaksey, Norwood Castle, 111

Odontites verna (red bartsia), seeds, 43

Ogbourne St Andrew: Barbury Castle, 260; Coombe Down, 264; Dean Bottom, 73, 79; mound, 116; Smeathe’s Plantation, 264; Smeathe’s Ridge, 264

Olivier, Edith, 266

organic materials, 198, 202

ornithology, and inn names, 70

Orwell, George, 269-70

Ostrea edulis (oyster), 45

Otis tarda (great bustard): decline, 66—8; protection, 65; reintroduction, 68; scarcity, 63-5; sightings, 70; in Wiltshire, 63—70

ovens, Romano-British, 258

Owlesbury (Hants), 228, 229

Oxford: Ashmolean Museum, 134; defences, 201; Lady Margaret Hall, 271; Magdalen College, 126; St Mary’s Hall, 10; University, 15

Oxford Archaeological Unit, 259

Oxfordshire: pottery, 196, 219, 227, 228; tile industry, 37, see also Abingdon; Cowley; Wayland’s Smithy

Oxychilus alliarius (mollusc), 78

palace, Saxon, 108

palaeoenvironmental materials, Beaker, 78—9

parenchyma, remains, 171

Paris (France), 120, 121

Parker Pearson, M., 7

Parliamentary Reform Bill (1832), 16, 18

Passmore, A. D., 134

Peacock, Thomas Love, 16

pebble hammers, 100, 103

Pecten spp. (scallop), 45

Pembroke, Earl of, 65

Pennant, Thomas, 66

Penruddocke family, 66

Penryn (Cornwall), 10

Penselwood (Somerset), Balland’s Castle, 112

Percy family, 83, 85, 86

Percy, Katherine, 83, 85

Percy, Sir William, 85

Petty-Fitzmaurice, Henry, 3rd Marquis of Lansdowne, 127

Petty, George, 54, 55

Petty, Lord Henry, 122

Petty, Richard, 54, 55

Petty, William, 1st Marquis of Lansdowne, 120, 1122; 1233124, 125, 126

Pevsner, Nikolaus, 210

Phillips, Bernard, 257, 258, 262; note on ring from Froxfield, 239-41

Phipps, T. H., 15

photosynthesis, Ingen Housz’s studies, 120, 121

Pierce, Dr, 126

piercers, Neolithic, 149

pigs: bones, 45, 48, 49, 50, 51, 59, 79, 157, 160-5, 166, 198; teeth, 155, 160, 162, 165

INDEX

pinbeaters, Late Saxon, 197

Pinckney, Erlysman, 254

pins: Roman, 225, 227; medieval, 29

pipemakers, 38

Pisum sativum (pea), remains, 43

Pitman, William, 252

pits, 235; Neolithic, 79, 151-3, 233; Early Neolithic, 131, 140, 141-4, 151, 154, 165, 166, 167, 168, 173-4, 175-6; Late Neolithic, 131, 141, 144, 154-5, 165, 166, 167, 169, 174, 176-7, 233; Beaker, 71- 81; Bronze Age, 259; Early Bronze Age, 79, 233; Late Bronze Age, 261; Late Bronze Age/Early Iron Age, 262; Middle Iron Age, 233; Romano-British, 94, 99, 102; Late Saxon, 191; Saxo-Norman, 191; medieval, 22-4, 26-7, 42-3, 189, 191, 243, 255; post-medieval, 27, 39, 243, 247; 18th century, 191; 19th century, 27; 20th century, 27; plant remains in, 173-4; stone destruction, 1, 4, 6; test, 137, 140-1, 148— 54, see also cess pits; ditches; postholes

Pitt, William, the younger, 10

Plantago major (greater plantain), seeds, 170, 174

plant remains: from Crescent Copse, Shrewton, 78; from Ivy Street/Brown Street, Salisbury, 39-45, 59; from St John’s Hospital and South Street, Wilton, 199— 200, 202; from Windmill Hill causewayed enclosure, 168-75, 175-6

plants: applications, 44—5; burning of, 173

Plymouth (Devon), 50

Poa spp. (meadowgrasses), seeds, 170

Pobeman, William, 53

Polden Hills (Somerset), 220

Pollard, Joshua: note on worked sarsen at Windmill Hill causewayed enclosure, 154; report on excavations at Beckhampton Avenue, 1-8

pondweeds, uses, 44

Ponting, C. E., 209

Poole (Dorset), 70

Poore, Sir Edward, 15

Porpoise, The (yacht), 250

Portugal, 16

postholes: Neolithic, 152; Late Neolithic, 144, 233; Early Bronze Age, 233; Late Bronze Age, 261; Late Bronze Age/Early Iron Age, 262; Romano-British, 99; Saxon, 188-9, 200-1; Saxo-Norman, 191; medieval, 26, 27; post-medieval, 243, 259; modern, 137, see also pits; stakeholes

postpipes, 144

Potamogeton spp. (pondweeds), seeds, 43

Potamogeton polygonifolius (bog pondweed), 44

Potter, Mr, 18

pottery: prehistoric, 71, 147, 191, 192, 256; Neolithic, 4, 261; Early Neolithic, 141, 144-6, 147, 175; Late Neolithic, 146-7, 148; Beaker, 73, 74-7, 79, 261; Bronze Age, 77, 233, 234, 235, 236; Early Bronze Age, 261; Middle Bronze Age, 261; Late Bronze Age, 147, 148, 261, 263; Iron Age, 90, 94, 102, 103, 259, 260; Late Iron Age/ Romano-British, 255; Roman, 219, 220, 227-8, 263; Romano-British, 92, 94, 103,

147, 185-6, 191, 192, 196, 200, 233, 236, 255, 256, 257, 261, 262, 264; Saxon, 255; Late Saxon, 181, 189, 191, 192-3, 195-6; Saxo-Norman, 189, 191, 192-3, 195-6, 201-2; medieval, 6, 22, 24, 26, 27, 29-37, 108, 147, 189, 190, 191, 195, 195-6, 202, 243-5, 246, 247, 256, 257, 258, 259, 262, 264; post-medieval, 6, 24, 26-7, 29, 35-7, 102, 147, 195, 245-6, 258, 259; 17th-18th century, 24, 27, 28; Black Burnished Ware, type 1, 227; Bristol-type slipwares, 36; Cheddar type, 192, 193, 196; chronologies, 36-7; classification, 30; coarse redwares, 36; coarsewares, 30, 31—3, 33-5, 36, 192, 195; Crockerton type, 193, 195, 196, 243, 245, 246; Dorset black burnished ware, 94; finewares, 33, 36, 192, 245; ?Frechen ware, 36; German wares, 36; Grooved Ware, 4, 6, 77, 144, 146-7, 148, 154; industrial wares, 36; Kennet Valley wares, 243, 245, 246; Laverstock type, 30-3, 36, 37, 193, 195, 196, 243, 245, 246; New Forest ware, 103, 227; North Gaulish roughcast wares, 192; Oxfordshire ware, 94, 103, 219, 227, 228; Peterborough ware, 77; Raeren ware, 36; Samian, 94, 100, 103, 192, 219, 227, 261; Savernake ware, 94, 100; slipwares, 36; Staffordshire-type slipwares, 36, 195; stonewares, 36; tinglazed earthenwares, 36; “Tudor Green’ ware, 35, 37, 196, 245, 246; Verwood type, 31, 35-6, 37, 195, 196, 246; Westerwald ware, 36; ‘Wiltshire Brown’ ware, 35, see also beakers; bowls; clay pipes; cups; dishes; jars; jugs; kilns; saucers; tiles

Powys see Clyro

Powys brothers, 265

Prehistoric Society of East Anglia, 134

Prestatyn (Clwyd), 220

Priestley, Joseph, 121

Primary Health Care Centres (Amesbury) Ltd, 255-6

Pringle, Sir John, 120

Prothero, R. E., 206

Prunus spp. (plum/sloe), stones, 42, 44, 169, 170

Pteridium aquilinum (bracken), 42

Puckeridge (Herts), 223

Pudding Brook, 237

Puddlehill (Beds), 165, 167

Punctum pygmaeum (mollusc), 78

Pupilla muscorum (mollusc), 78

Purbeck (Dorset), 31

Purbeck, Eliza, 253

Purbeck Marble, 22, 38

Purpose, 269

Pympol, Richard, 53

Queensberry, Duke of, 66

querns: Neolithic, 154, 176; Eaily Neolithic, 141; Middle Bronze Age, 261; ?Romano- British, 197, 261; Saxon, 189

rabbits, bones, 45, 51

radiocarbon dating, Avebury area, 131—2, 175 Rajasthan, great bustards, 63

Ralf, 115

Ramilisonina, 7

279

Ramsbury, Littlecote Estate, 239

Ranunculus spp. (buttercups), charred seeds, 200

Rattew, George, 55

Rawlings, Mick, report on excavations at Ivy Street/Brown Street, Salisbury, 20-62

Rayner, Alan, 80

Rayner, Louise, 268

recipes, fish, 50

Reece, Richard, 230

refuse disposal, medieval, 50-1

resistivity meters, 139

ridge and furrow, medieval, 4

Ridgeway (Dorset), 132

rings: Roman, 225; Romano-British, 239-41

ringworks, 107-8, 108-16, 116-17; use of term, 107

roads: Roman, 90, 112, 256, see also trackways

robber trenches, 19th century, 191

Robertson, A. S., 225

Robinson, A. M., 31

Robinson, Paul, note on knife finial from ‘Trowbridge, 248-9

Rochester, Bishop of, 267

Roche, Walter, 83

Rogers, A. G., 134

Rogers, Kenneth: note on Cumberwell, 249— 54; obituary by, 272

Romney Marsh (Kent), 70

Romsey (Hants), 202; faunal remains, 49

Rosaceae, seeds, 170

Roseaman, Judith, note on fieldwork at Euridge Manor Farm, Colerne, 220

Rosenlund, K., 45

Rous, William, 85

Route, John, 53

Royal Commission on the Historical Monuments of England, 260

Royal Historical Society, 265

Royal Institute of British Architects, 211

Royal Navy, 15

Royal Signals, 272

Royal Society, 121

Royal Volunteers, 250

Rubus spp. (blackberry/raspberry), seeds, 42, 44

Rumex spp. (dock), seeds, 170

Rundell, Mary Ann, 254

Rusteshale, Alice, 52, 53

Sagittaria sagittifolia (arrowhead), 171

St Albans (Herts), Verulamium, 227, 240

St Barbara, 248

St Clement, Jersey, 70

St Joseph, J. K. S., 140, 219

St Lawrence, 248

Salisbury, 67, 83, 84, 134, 193, 195, 196, 197, 202, 245, 246, 250, 266; medieval, 52-3; post-medieval, 53-5; modern, 55-7; Antelope Chequer, 21, 51, 52, 54, 56; Balle’s Place, 52; Barnwell Cross, 53; Bird in Hand, 70; Blue Lion, 51, 53,54; Brown Street, 20-62; Catherine Street, 52; Chough, 70;The Close, 53, 54, 261; cloth trade, 53; Colletts Garage, 57; Dairyhouse Bridge, 260; Dairy Meadow Lane, 260; East Harnham, 86, 87; Falcon, 70; Faraway Garage, 55, 57; Fisherton Anger, 86;

280

Fisherton Manor Middle School, 260-1; Gigant Street, 49, 53; Guilder Lane, 31; Harnham Bridge, 183; High Street, 260; itinerant mission, 268; Ivy Street, 20-62; kilns, 30-1; London Road/Rampart Road, 30-1; Maiden Head, 51,53; Market ward, 52; Martins ward, 51, 55; Milford Bridge, 30; New Street, 51, 52, 53; Nyweton Westyate, 260; Old Livestock Market, 260; Old Sarum, 35, 107, 117, 181, 183, 260; Pheasant, 70; pipemakers, 38; Poultry Cross Market, 65; Queen Arms Inn, 51, 53, 54, 55, 56-7; Red Lion, 67; Rollestone Street, 52; St, Martin’s Street, 52; St Thomas, 51, 54; Spread Eagle, 70; Swan, 70;Three Swans, 70;Victorian, 268; White Bear, 51; White Swan, 70; Winchester Street, 52

Salisbury, Bishop of, 267

Salisbury Cathedral, 87; Archives, 51; Dean and Chapter, 51, 52, 53, 57; Ecclesiastical Commissioners, 55

Salisbury City Council, 51

Salisbury District Council, 21

Salisbury Local Studies Library, 51

Salisbury Museum, 248

Salisbury Plain, 262; great bustards, 63, 65, 66, 68, 69, 70

Salisbury Plain Training Area, excavations, 71-81, 261

Salisbury and South Wiltshire Museum, 68, 239, 268

Sambucus nigra (elder), 172, 200

Sandford, 219

Sandy Lane, 103, see also Verlucio

Santo Domingo slave uprising (1797-1804), 12

sarsen stones, 1-2, 7, 141, 144, 176, 256; grinding, 173; worked, 154

Sassoon, Siegfried, 205, 214

saucers, medieval, 33

Sauls’ River (Jamaica), 10

Savernake Forest, 108, 245; kilns, 94, 147, 227; Knowle Farm, 132-4

Savory, William, 122

Sawyer, P. H., 265

Scammell, E. L., 55

Scammell, Mrs, 55

Scandix pecten-veneris (shepherd’s needle), 44

scanning electron microscopy, plant remains studies, 168, 170

Scotland see Edinburgh; Fettes School; Glasgow

Scott, Peter, 270

scrapers: prehistoric, 90; Mesolithic, 257; Neolithic, 134, 153; Early Neolithic, 149, 151; Late Neolithic, 149; Beaker, 77-8; Early Bronze Age, 149

sculpture: Roman, 219, 220; Romano-British, 240

Seaford (East Sussex), 12

Secale cereale (rye), charred grains, 43

Second World War see World War II

seeds: charred, 42, 43, 44, 45, 168-75, 263; identification, 42

Segrim, Dr, 15

Selby, John Prideaux, 67

Seles, Geoffrey de, 267

250)

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Sellars, Warren, 256

SEM, plant remains studies, 168, 170

Semington, 269

Senior Officers’ Training School, 18-19

settlements: Late Neolithic, 237; Bronze Age, 233-8; Late Bronze Age, 237; Iron Age, 90; Romano-British, 90, 103, 233-8; Anglo-Saxon, 260; Saxon, 256; Late Saxon, 107, 181-204; Saxo-Norman, 196; medieval, 20-62, 241-8, 256-7, 257-8; post-medieval, 241-8, see also castles; enclosures; villages

Severn Valley, 222

Sewell, Andrew, work noted, 269

Seymour, Hugh, 219, 220

Shaftesbury Abbey (Dorset), charters, 265

Shakespeare, William, 265

Shaw Mellor, Alfred, 219, 227

sheep: bones, 45, 48, 49, 50, 51, 59, 96, 155, 157, 160—5, 166, 198; Leicester Longwool, 51; Southdown, 51; teeth, 162

Sheerness (Kent), 250

Shelburne family, 124

Shelburne, John, Earl of, 125

Shelburne, Lord, 121

Shelburne, William, 2nd Earl of, 125

Shelley, Percy Bysshe, 270

shells (marine) see marine shells

Sherrington: church, 115; motte, 114-15, 117

shoes, medieval, 39

shovels, Roman, 227

Shrewton, 68; Bustard Inn, 68, 69; Crescent Copse, 71-81; Maddington, 84 (Manor Farm, 68); Old Coal Yard, 262; Robin Hood’s Ball, 77, 79, 80, 153; Rollestone Bake Farm, 68

Shuter, Thomas, 54

sickles, Roman, 227

Sidmouth, Lady, 15

Sidmouth, Lord, 15

S., J., 67

skates/runners, medieval, 197

Slater, William, 127-8

slavery: end of, 15; public opinion, 12

slaves, treatment of, 12

Small, Doug, work noted, 269

smallpox, inoculation, 121

Smart, Wake, 270

smelting, iron, 183, 191, 197

Smith, A: 'C:; 132

Smith, Cecil, 68

Smith, Charles, 10

Smith, I. F., 6-7, 77

Smith, John, 9

Smith, Joshua, 9-10, 12, 19

Smith, Mr, 126

Smith, Rachael Seager, report on excavations at St John’s Hospital and South Street, Wilton, 181-204

Smith, Stephen, 68

Society of Friends, 271

soils, buried, 200

Solent (Hants), 50

Somerset, 84, 225; great bustards, 68; pottery, 195, see also Bishops Lydeard; Bruton; Charlton Musgrove; Donyatt; Downside School; Farleigh Hungerford; Montacute; North Perrott; Penselwood; Polden Hills;

‘Taunton; Wanstrow

Soper, Thomas, 54

Sorley, Charles Hamilton, 206, 266

Southampton (Hants), 59, 197; Brokage Books, 50; faunal remains, 48, 50; Lower High Street, 50; port, 50; University, 1

South Cerney (Glos), 111

South Wales, 239

South Wraxall, 87, 249, 250, 252

Spain, 16, 271; great bustards, 65

Sparsholt (Hants), 228

spatulas, medieval, 197

Sphagnum spp. (bog moss), 44—5

Spoerry, P. S., 31

spoons, 248

Staffordshire, slipwares, 36, 195

stakeholes, 235, 236, 263; ?modern, 137, see also postholes

Stanton St Bernard, Manor Farm, 262

Stapleford, 115; ringwork, 111

Staples, Samuel, 254

stater, gold, 103

statues, Romano-British, 240

Staverton, Nestlé at, 268

Steedman, Mrs, 67

steel objects, medieval, 29

steelyards, Roman, 227

Stellaria spp. (chickweeds), charred seeds, 200

Stephen, King, 111

Stephens, Robert, 53

Stephens, Widow, 54

Steuart, George, 9-10

Steward, Charles, 249, 254

Steward, Jane, 249

Steward, Richard, 249

Stickler, Tracey, notes on animal bone from Littleton Drew—Chippenham gas pipeline, 96, 100-1

Stock, Jan, 63

Stokes, Eleanor, 126

Stokes, Judith, 125

Stokes, Richard, 125, 126

stone circles, 269

Stonehenge, 6, 15, 65, 76, 79, 132, 153, see also Amesbury

stones: burial of, 1, 4, 5, 7; Chilmark, 188; destruction, 7; rubbing, 154, 176; sockets, 6, 7, see also sarsen stones

stones (fruit) see fruit stones

stonework: Early Neolithic, 154; ?Bronze Age, 100, 103; Roman, 219, 220; Late Saxon, 197; medieval, 38, 197, see also querns; sculpture

Stourton with Gasper, Castle Orchard, 111— 12

Street, A. G., 266

Street, George, 18

strip brooches: Roman, 223; medieval, 197—8

Struthionidae, 63

studs, Roman, 225

Stukeley, William, 1, 2, 6, 7

Suffolk see Hurst Fen; Newmarket; West Stow

Sulgrave (Northants), 114

Surrey, 271, see also Charterhouse; Cranleigh; Guildford; Kingston upon Thames

susceptibility meters, 139

Sussex, knife finial, 249

Lakenheath;

INDEX

Sutton Benger High Street, 262 Sutton, James, 53

Sutton Veny, Littkecombe Down, 264 Svein, 115

Swindon, 84

Swindon Borough Council, 260 Swindon Museum, 134

Swinstead, J. Howard, 268

Sydling St Nicholas (Dorset), 253

Tahiti, 12

tanning, evidence for, 24

Taunton family, 254

Taunton, Mrs, 253

Taunton, Robert, 253

Taunton (Somerset), 68; cess pits, 45

Taunton, Thomas Pile, 253

Taunton, W. L.T. P., 254

Taylor, Anna Susanna, 12

Taylor, Sir John, 12

Taylor, John, 55

Taylor, Simon, 12

Taylor, Simon Richard, 12

Taylor, Sir Simon Richard Brisset, 12

Taylor, Simon Watson, 18

tea, infusions, 44

Tede, Dominique, 121, 122

teeth: cattle, 102, 155, 160, 162, 165; deer, 101; goats, 162; pigs, 155, 160, 162, 165; sheep, 162 :

temples, Romano-British, 90

tessellation, Romano-British, 240

Tewkesbury (Glos), 84

Thackeray, William Makepeace, 16

Thomas, 2

Thomas, Dylan, 270

Thomas, Edward, 266

Thomas Hardy Society, 270

Thomas, James, paper on great bustard in Wiltshire, 63-70

Thompson, Luigi, 240

Ticinum (Italy), 230

Tidworth, Tedworth House, 18

Tighe, M. F., work reviewed, 266-7

tiles: Roman, 220; Romano-British, 92, 96, 103, 147, 233, 256; Late Saxon, 196-7; medieval, 24, 26, 27, 37-8; floor, 38; ridge, 37-8; roof, 37

Till, River, 71, 111

Tilshead, 66, 67, 262; Breach Hill, 262; Copley Down, 261; Foxtrot Crossing, 261; Horse Down, 261;Tilshead Down, 261

Timby, Jane: note on pottery from Chippenham Western Bypass, 236; notes on pottery from Littleton Drew— Chippenham gas pipeline, 94—5, 100

Times, The, 206

Time and Tide, 269

Tocotes, Sir Roger, 85, 86

Tolhurst, Peter, work reviewed, 2605-6

Tollard Royal, 86

tombs, chambered, 131

tools, 198; Neolithic, 134; Roman, 227; flint, 236, 257

Torksey (Lincs), 228

trackways, medieval, 259

Transco, 90, 92

‘Treadgold, Mr, 66

tree holes, 2modern, 189

Trevelgue (Cornwall), 228, 229

Trier (Germany), 230

Triticum spp. (wheat), charred grains, 42, 43

Triticum aestivum (bread wheat), charred grains, 200

Triticum dicoccum (emmer wheat), grains, 169, 173

Triticum monococcum (einkorn wheat), grains, 169

Triticum spelta (spelt), charred grains, 200

Tropenell, Agnes, 83, 86

Tropenell Cartulary, The, 87

Tropenell, Christopher, 87

Tropenell, Edith, 83

Tropenell family, arms, 86

Tropenell, Henry, 83

‘Tropenell, Margaret, 83, 86, 87

Tropenell, Thomas, career of, 82—9

Tropenell, Walter, 83, 85-6

Trowbridge, 86, 249, 268-9, 270, 272; Bellefield House, 254; Bustard Club, 68; Castle, 107; Fore Street, 250; George and Woolpack, 252; Hilperton Road, 254; knife finial from, 248-9; Lloyds Bank, 250; pottery, 193, 196; St James’s Church, 272; Swan, 70; White Swan, 70

Trowbridge Advertser, 272

Trowbridge Boys’ High School, 272

Trowbridge Chronicle, 272

‘Trowbridge Civic Society, 272

Trowbridge Museum, 272

Trowbridge Orchestra, 272

Trowbridge Philharmonic Choral Society, 272

Trowbridge Reference and Local Studies Library, 51

Tso-Hsin, Cheng, 68

tubers, 168-75

‘Tweeddale, 7th Marquis of, 16

Tweeddale, 8th Marquis of, 18

Underwood, Thomas, 253

United States, 270

University of Sheffield, Department of Archaeology and Prehistory, 155

Upavon, 67; Widdington Farm, 262

Urtica spp. (nettles), charred seeds, 200

Urtica dioica (stinging nettle), uses, 44

US Air Force, 260

vault burials, 123-8

Vellonia costata (mollusc), 78

Verlucio, 103

Vernon, Elizabeth, 126

Verulamium (Herts), 227, 240

Verwood (Dorset), pottery, 31, 35-6, 37, 195, 196, 246

Vicia faba (broad bean), seeds, 42, 43

Victoria History of Wiltshire, 268

Victoria, Queen, 12, 15, 16

Vienna (Austria), 121, 122, 270-1

Vikings, 112

villages: medieval, 259, 264; deserted, 257

villas; Roman, 218-32; Romano-British, 90, 103, 239-41

Vindolanda (Northumberland), 240

Vinynge, George, 53, 54

Vitis vinifera (grape), seeds, 43, 44

Viveash, Samuel, 122 Vousden, Aaron, 55 Vousden, Ann, 55

Vulpes vulpes (red fox), 68

Waddon Hill (Dorset), 223

Wales, University, 1

Walker, Graeme, note on flintwork from Chippenham Western Bypass, 236

Walker, Heneage, 249

Walker-Heneage family, 249

Walker, John, 249

Walker, Mary, 249

walls: Roman, 220; Saxon, 188; ?medieval, 191; medieval, 26, 28; post-medieval, 24, 191

Walpole, Sir Robert, 12

Walters, Bryn, 220; note on Euridge Manor Farm, Colerne, 231

Walton family, 266

Walton, Isaak, 50

Wanborough, Earlscourt Manor, 262

Wanstrow (Somerset), 36

war memorials, 205-17

Warminster, 54, 68, 84; Battlesbury Bowl, 262-3; Battlesbury Camp, 262; Battlesbury Hill, 263; Battlesbury Wood, 263; Boreham Farm Bungalow, 263; Brick Hill, 263; Cock, 70; Emwell Street, 245; Furnax Lane, 263; Harman Lines, 262; market, 67; Middle Hill, 263; pottery, 35; Swan, 70; White Swan, 70

Warmwell, William, 52, 57

War Office, 19

War Report, 269

Warriner, Mr, 15

Warwickshire, 265

Waterloo (1815), 67

Waters, John, 67

Watson, Anna Susanna, 12

Watson, George, 10

Watson Taylor family, 18

Watson Taylor, George, 9, 10-16, 18

Watson, William, 121

wattle and daub, 197, 198, 201

Wayland’s Smithy (Oxon), 131

Weeks and Son, Messrs, 55, 56

wells, 27; Romano-British, 99, 100; ?post- medieval, 24, 51

Wessex, 183, 239, 270; literature, 265-6

Wessex Archaeology: evaluations, 71, 259, 260-1, 264; excavations, 233 (Amesbury, 255; Avebury, 256; Devizes, 241-8; Salisbury, 20-62; Salisbury Plain, 261; Warminster, 262-3; Wilton, 181—204)

Wessex, King of, 181

Wessex Water, 255

Westbury: chalk pit, 263; Northacre Business Park, 263-4

Westby, Gregory [George], 85

West Dean: Bentley, 66; church, 111; earthwork, 111, 117

Western Front, 206

West Indian Planters and Merchants, 12, 18

West Indies, 12

West Lavington, 67; Littleton Pannell, 115

Westminster Press, 272

West Stow (Suffolk), 197

West Sussex see Findon; Houghton

282

West Tisbury, Pythouse, 53

West Wickham (Cambs), 70

Weyhill Brewery, 55

Wheatley, David, report on excavations at Beckhampton Avenue, 1-8

Wherwell (Hants), 66

whetstones, 100

Whipsnade Wildlife Park, 68

White, Gilbert, 66

Whittle, Alasdair, 1, 7; report on excavations at Windmill Hill causewayed enclosure, 131-80

Whittokesmede, John, 83, 84

Wigmore, Thomas, 55

Wilkes, Elizabeth, 53

Wilkes, John, 53, 55

William IV, King, 15

Williams, Alfred, 266

Williamson, Henry, 266

Williams, Widow, 55

Wilsford cum Lake, 67

Wilson, Wynne, 206

Wilton, 65, 87; Bull Bridge, 189; church, 183, 189, 202, 203; Cross Bridge, 202-3; Ditchampton, 185, 202; fortifications, 108; Kingsbury Square, 183, 202; legal status, 181-3; Market Place, 183; Netherwells Lane, 185; North Street, 202; Russell Street, 202; St John’s Hospital, 181-204; South Street, 181-204; West Street, 185, 202

Wilts and Berks Canal, 269

THE WILTSHIRE ARCHAEOLOGICAL AND NATURAL HISTORY MAGAZINE

Wiltshire: early castles, 105-19; tile industry, 37

Wiltshire Archaeological and Natural History Magazine, 105, 134, 219

Wiltshire Archaeological and Natural History Society, 87, 134, 228, 271; Archaeology Field Group, 220

Wiltshire County Archaeological Officer, 183, 241

Wiltshire County Council: Archaeological Service, 92, 233; Environmental Services Department, 233; Local Studies Library, 270

Wiltshire Gazette, 18

Wiltshire Record Society, 272

Wiltshire Rescue Archaeology Project, 256

Wiltshire and Swindon Record Office, 51, 249, 270

Wiltshire Times, 272

Winchester (Hants), 66, 183, 193, 196, 202; bishops of, 86

windmills, 116

Wingfield: Church Lane, 264; Stowford, 265

Winterbourne, Figsbury Rings, 266

Winterbourne Bassett, 132

Winterbourne Monkton: Hackpen Hill, 132, 134; Millbarrow, 131

Winterbourne Stoke, 67; Hill Farm, 264

Winterslow: Easton Down, 80, 131, 132; Winterslow Hut, 66-7

Wodehull, John, 86

Woodcutts (Dorset), Down Farm, 166, 167, 168

wooden objects, Saxo-Norman, 198

Woodham Ferrers (Essex), 70

Woodhouse, Bill, 268

Woodspring Museum (Avon), 228

woodwork, medieval, 39

Woolf, Virginia, 266

Wootton Bassett, pottery, 257

Worcester, cess pits, 45

Worcestershire see Lark Hill

Wordsworth, William, 265

World War I, 268; casualties, 206; memorials, 205-17

World War II, 18; anti-aircraft gun pits, 260; memorials, 205

Wroughton: Barbury Castle, 260; Overtown House, 264

Wye (Kent), 70

Wyles, Sarah F., note on marine shells from Ivy Street/Brown Street, Salisbury, 45

Wylye, River, 115, 181, 183, 201, 263

Wylye Valley, 261, 268

Wyndham, Rev., 67

Yarrell, William, 63—5, 67

Yatton Keynell, 103; Church Farm, 264; Grove Farm, 101; Park Farm, 92

York, 44

Yorkshire, great bustards, 68

Zeals, 267-8; Bypass, 267; Zeals House, 268

WILTSHIRE HERITAGE ia

= MUSEUM GALLERY LIBRARY

Published by The Wiltshire Archaeological and Natural History Society

ISSN 0262 6608

WILTSHIRE HERITAGE

MUSEUM GALLERY LIBRARY

Published by The Wiltshire Archaeological and Natural History Society

ISSN 0262 6608