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,