Archaeological work at Arlington Court, Devonshire The Conservatory (and other glasshouses) at Arlington Court Building Survey and Historical Analysis Archaeological Report

Jeremy Milln

Throughout most of the history of the formal garden it has been reasonable to think of these ornamental adaptions of the local environment as extensions of our habitations. They might, in many situations, be considered 'rooms'. As such, they are amenable to the same archaeological techniques as the dwellings that they adorn. Features can be surveyed, planned and interpreted by analytical fieldwork. Sites can be discovered and interpreted by using air photographs, geophysical methods and excavation. Any disturbance of the ground is measurable. The archaeology of gardens is a relatively new discipline that is still developing its potential. In the past, the existence of gardens has not been appreciated and, therefore, they have not been looked for. Roman villas in Britain probably had some form of pleasure garden. However, until fairly recently, archaeologists have concentrated on the economic activities of such places. The villa at Apethorpe in Northamptonshire was arranged around a central courtyard. In the centre of the courtyard was a shallow stone-lined depression that was described as an impluvium. It now seems likely that it was a central water feature in a courtyard garden. Long narrow structures on two sides of the central courtyard of a Roman villa at Pitney, Somerset, positioned as they are facing on to the villa's façade, makes it likely that they were garden pavilions of some kind but they were initially interpreted as pigsties and slaves' quarters. The failure to appreciate the existence of gardens has led to confusion on sites of later date. At Steeple Gidding in Cambridgeshire complex earthworks were once wholly interpreted as the remains of a deserted Medieval village. However, the earthworks to the south of the church represent the terraces and ponds of a formal garden laid out on top of the abandoned village in 1648. They belonged to a large newly built house. In Bedfordshire at Old Warden the terraces of a late 16 th century garden partly overlay the extensive earthworks that are the remains of the great Cistercian abbey. None of these gardens was recognised until recently. There are probably hundreds of gardens in Britain which are preserved in pasture or hidden in woodland that remain undiscovered solely because they have not been recognised. There is considerable scope for archaeological techniques to be applied to the search for and examination of gardens. There is no evidence, at present, for pleasure gardens in Britain before the Roman occupation. There were almost certainly vegetable gardens in prehistoric times but these cannot be readily distinguished from small paddocks. In prehistoric contexts we find digging sticks and stone hoes. In late prehistoric times settlements begin to exhibit features like fenced and earthwork protected plots. In theory, fields are ploughed and gardens are not. In many instances, it is possible to see evidence of ploughing. Archaeological methods have become more sophisticated. It is possible to recover evidence of plough marks cut into the subsoil or underlying rock thereby distinguishing between fields and plots. However, modern agriculture activities destroy more and more of the prehistoric sites.

An archaeological investigation was requested due to garden landscaping at Upper Aston Farm house producing numerous artefacts. The area of investigation was located within an enclosed garden next to the house at Upper Aston Farm. The work consisted of establishing a series of test pits to investigate if any archaeological features may be related to the finds being produced. These works were requested by Greenpoint Creative Wellness CIC. The work was undertaken on 31/08/21 and 01/09/21. The work will establish if there are any underlying features associated with the historic property grounds that should be considered before further landscaping should commence. Whilst the investigation did produce some notable artefacts in context, no archaeological features of interest were identified. The artefacts being produced by works in the enclosed garden are likely a small amount of residual materials relating to the occupation of the house at Upper Aston Farm. As such it was concluded that further landscaping of the enclosed garden can be carried out without fear of disrupting any potential archaeological features.

This report presents the results of an archival and field survey of the walled garden at Acorn Bank, a National Trust property in Temple Sowerby, Cumbria. The aim of the survey was to establish a date for the walled garden and to provide information on the history and evolution of the sunken gardens and associated features. The research has established that the enclosure of a garden at Acorn Bank took place during the early to mid-seventeenth century, and that the sunken garden was created in the mid to late-seventeenth century. In addition, much information regarding the subsequent evolution of the gardens has been revealed. The project was designed by Heather Birkett, Head Gardener at Acorn Bank. It was undertaken by Marion Shiner, a WRoCAH/AHRC-funded doctoral researcher (Award no. 179474572) at the University of Sheffield, for a WRoCAH Researcher Employability Project.

Archaeological work at Arlington Court, Devonshire 1. The Conservatory (and other glasshouses) at Arlington Court Devonshire Building Survey and Historical Analysis Archaeological Report Jeremy Milln Archaeological Consultant December, 2013 Jeremy Milln Archaeological Report: glasshouses at Arlington Court Archaeological work at Arlington Court, Devonshire Building Survey and Historical Analysis Archaeological Report for the Conservatory and other glasshouses at Arlington Court Devonshire EX31 4LP NGR: SS 6126 4059 By Jeremy Milln 43, Villa Street, Hereford, HR2 7AU (28,794 words) Client: National Trust (Ana Chylak, Property Manager) PO Reference: TW436946 Ordered: 15th May, 2012 OASIS Project No: 168848 Listed Building Application No: 53676 North Devon Council Planning Application No: 53677 Planning Condition Discharge Reference: 54139 HES ref ARCH/DM/ND/19004 18 Dec 2013 © 2013 Jeremy Milln shall retain full copyright of this written report under the Copyright, Designs and Patents Act 1988 with all rights reserved; excepting that it hereby provide a full licence to the client for the use of this document by the client in all matters relating to the project as described Jeremy Milln Archaeological Report: glasshouses at Arlington Court Archaeological work at Arlington Court, Devonshire CONTENTS Page 1. SUMMARY, BACKGROUND & OBJECTIVES 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Executive Summary …........................................................................... Statement of Significance …................................................................. The Site ….............................................................................................. Objectives of the Project ….................................................................... Methodology & standards ….................................................................. Staffing, acknowledgements & archive …............................................. General references …............................................................................. List of Illustrations …............................................................................. 2. SPECIFYING the 1982 CONSERVATORY 2.1 2.2 2.3 2.4 Overview of the Specification …........................................................... Section 'A': General Conditions & Preliminaries ….............................. Section 'B': Trade Preambles – use of materials …................................ Section 'C': Schedule of Works ….........................................................       2.5 Demolitions New work to receive glasshouse Work to back wall Specialist's glass house construction Sundry builders work Sundry finishings Discussion: Specification and Implementation …..................................... ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ Brickwork Mortars & lime-washes Back wall Joinery Timber Floor Detailing, metal fittings etc Glazing 3. GENESIS and CONSTRUCTION of the 1982 CONSERVATORY 3.1 3.2 3.3 3.4 3.5 3.6 Summary ….................................................................................... Central Conservatory of c.1925 …................................................. The Trust in the early days at Arlington ….................................... Emerging debate about the glasshouse …...................................... Commissioning the Conservatory replacement ….......................... Appointment of Philip Jebb as Architect …............................. Jeremy Milln Archaeological Report: glasshouses at Arlington Court Archaeological work at Arlington Court, Devonshire 3.7 3.8 3.9 3.10 3.11 Contractual delays and challenging costs ….............................. Contract progress post Jebb …................................................... Horticultural preparations …...................................................... Final fitting and commissioning …............................................ Meeting costs …......................................................................... 4. DEVELOPMENT of the other GLASSHOUSES at ARLINGTON 4.1 4.2 4.3 4.4 4.5 4.6 4.7 Walled garden vinery range of the 1820s The glasshouse at Arlington Rectory, c.1845 Metallic conservatory proposals, 1849 Miss Chichester's conservatory, c.1925 The Pinery Small greenhouse After Jebb: the Salisbury Conservatory, 2012 Appendix I Appendix II Appendix III Appendix IV Obituary of Philip Jebb Obituary of Thomas Lee Archive Written Scheme of Investigation Jeremy Milln Archaeological Report: glasshouses at Arlington Court Chapter One: Summary, Background & Objectives 1.1 Executive Summary The chief objective of this report is to present an account of an old conservatory in the garden at Arlington which had to be taken down in 2012. The report records fabric and analyses development by reference both to historical documentation and physical evidence. The Conservatory itself, constructed in 1982, was to a design of the noted English architect Philip Jebb loosely based on the form of a conservatory which stood on the site between about 1925 and 1980 which for most of that period had been the centrepiece of a long low range of glass with vineries either side. The glass on this site was found to belong to three main periods. The first build, dating to around 1825-40, comprises a principal spine wall with a continuous lean-to range of glass against its south face and a series of back sheds against its northern. The second, which belongs to the inter-war period, around 1925, involved a substantial rebuild with a new central conservatory projecting forward of the line of the vineries. The final period is attributable to the National Trust's efforts to conserve and present a decaying site at a time of financial stringency. This involved the removal of all the historic glass and its replacement by a conservatory to Jebb's design, erected in 1982 on the site of the c.1925 structure. This report focuses chiefly on the present work involving the replacement of the 1982 building, but deals also with the evidence for the succession of earlier glasshouse structures. The site is now occupied by a conservatory by David Salisbury which was erected on the 1982 base in 2012-13. It is an aspiration for the future that the flanking vineries are also rebuilt for which evaluative excavation and further fabric recording and analysis is likely to be required. 1.2 Statement of Significance The Chichesters created here, probably under the eye of the Soane-inspired Thomas Lee, a remarkable range of timber-built vineries to front their kitchen garden. Were this glass to have survived, it would be of immense importance today for no complete three-part timber-built Georgian vineries now exist in anything like their original form. However, given the readability of its footprint and the very interesting succession of glass at the site, it enjoys a significance of a rather different sort. Appreciation of the glass at Arlington has taken many forms from the private pleasures of the 19th century country house estate to the spare utility of Miss Chichester's long life through the first half of the twentieth century to the popular visitor attraction of today. Although the National Trust failed fully to appreciate it at the time, association of the site with one of the most important English architects of the post-War period gives it a special place. The present conservatory, by David Salisbury and completed in 2012, is the fourth on the site; a developmental achievement in itself. 1.3 The Site Arlington Court was probably the fourth country house of the Chichesters to have stood since the sixteenth century1. Is replacement by the present one in 1820-3, to the designs of Thomas Lee (1794-1834), offers the likely context for the construction of the walled garden with a vinery/conservatory range set against its south-east facing main wall with back sheds behind. Today the wall overlooks the Victorian garden, abundant with spring and summer colour, a popular destination for visitors to this National Trust property. The original glasshouse structure was entirely of lean-to form and almost 60 metres long. It was timbered, stood upon a dwarf wall and divided into five sections. Only the central section today is occupied by a glasshouse, currently a conservatory by David Salisbury, the fourth on the site. The garden and designed landscape of which the vineries form a part is included in the Register of Parks and Gardens of Special Historic Interest at grade II*. Fig 01: Tithe map, 1842 – detail showing the relationship of the vinery/conservatory range (centre), pinery at top, walled and flowers gardens. Arlington Court is in the bottom left hand corner and Arlington Rectory in the bottom right hand corner. (left hand) Fig 02: Satellite photograph of the same view today (slightly enlarged) with the Conservatory at the centre of the picture and a more developed layout for both flower and walled gardens (right hand) 1 For an overview of the history of Arlington and its grounds, refer to the National Trust Guidebook Arlington Court (1996) Fig 03: First edition 1:2,560 scale Ordnance Survey, surveyed 1886 – detail showing the pleasure grounds 1.4 Objectives of the Project The objectives of the Archaeological Project are fully defined in the Written Scheme of Investigation (WSI, see Appendix IV). Briefly these are to:  record, research and to analytically document the principal glasshouse structures at Arlington, focussing on a conservatory by the architect Philip Jebb ahead of its demolition and replacement.  record, research and to analytically document ancillary glasshouse structures at Arlington, chiefly the remains of the flanking vineries. Based upon work by Lovie for the Garden Conservation Plan (2009), other glasshouse structures at Arlington are included where they are important to our overall understanding.  provide a narrative account of the design and creation of the 1982 conservatory to enable a full appreciation of its history and importance to be gained.  advise on the programme of recovery of the glass at Arlington; a delicate balance of conservation, restoration and new work. This to be inspired by archaeology, engaged by visitors, used by garden staff and admired by all.  create a readable narrative report, led by measured drawings and photographic images. Building upon existing and new historical research, to explain the development and significances of the sites. Further to provide information which may eventually be used in any interpretation of the sites, given the need to demonstrate public benefit. The objectives of the wider project, initiated by the National Trust, are to replace the 1982 Jebb Conservatory and rebuild the flanking vineries so they may be returned to beneficial use, enhance the setting of the Victorian Garden and communicate meaning to the history of the site. In so doing the glass may be managed cohesively with the Arlington Estate land-holding as a whole, held to benefit the local community, support enterprise, steward the historic environment and produce fruit or cut flowers. A further aim specific to the site is to provide low level interpretation, allowing visitors to appreciate the glasshouse(s) for what they are. 1.5 Methodology & standards Methods employed are outlined in the WSI and include: 1.5.1 Documentary Building upon existing research, a search was made of relevant and readily available published and unpublished documentary source material, including historic maps, early photographs, drawings and written descriptions. The principal primary source relating to the Conservatory occur in a series of business files held in the archive store at Wansdyke, Wilts. These were closely examined. Reference is made to comparative material for other local sites where relevant. 1.5.2 Measured survey Production of a full set of plans, sections and elevations of the structures was not found to be necessary. A new annotated 1:50 scale plan of the Conservatory was produced; otherwise existing drawings , mainly from the office of the late Philip Jebb, were found to answer the requirements by way of record and interpretation. Completion of a measured survey at 1:20 of the hot wall at the back of the glasshouses begun by a garden volunteer (Anja Fischbach) in 2009 was considered. This would depict breaks in plane, evidence for phasing and jointing, constructional detail and building character . However it was not found to be possible due to the amount of obscuring plant growth. 1.5.3 Archaeological recording & analysis The survey comprised physical examination of the safely accessible areas of the site and artefactual evidence, particularly in respect of structure and detail which might be impacted or informed by subsequent building or conservation work. The Conservatory due to be demolished and replaced was the principal focus . Its building record and assessment comprise a detailed examination of the structure. Physical recording of the fabric was undertaken to elucidate the sequence of construction, deposition and relationships between structures and surfaces. The photographic record comprises high resolution digital photography approximately commensurate with a Level 3 record as defined by English Heritage (2006, 14). Where possible, photographs included graded photographic scales. 1.5.4 Archaeological excavation No archaeological excavation was required or was carried out as part of this project. However in the event of it features exposed are recorded by written description, drawing and photography. Stratigraphic sequences are recorded, even where no archaeological evidence is present. Features are planned at a scale of 1:20 or 1:50, and sections drawn of all cut features and significant vertical stratigraphy at a scale of 1:20. A comprehensive written record is maintained using a continuous numbered context system on pro-forma cards. Written records and scale plans are supplemented by photographs using black and white monochrome and digital photography. The level and detail of this is determined by individual site circumstances at the discretion of the archaeological contractor , but may include a gazetteer of historic features. Allowance is made for recovery of artefacts and their cleaning, conservation and analysis for the report. Finds are cleaned, marked and bagged and stored in suitable conservation containers. All archaeologically significant stone, ceramic or metalwork are accurately plotted, recorded and added to the catalogue of moveable artefacts listed for the main building. Finds may be stored with the report and site archive either with the local Museum Service or the National Trust following specialist analysis and after reporting is complete. The exception is likely to be lithic and ceramic building material to be left on site. 1.5.5 Professional standards The project follows the requirements set down in the Standard and Guidance for Archaeological Desk-Based Assessment (IfA, 2001a) and Standard and Guidance for the Archaeological Investigation and Recording of Standing Buildings or Structures (IfA 2001b). Recording work aims to conform to guidelines set down by English Heritage’s Understanding Historic Buildings: a guide to good recording practice (EH 2006) and the Association of Local Government Archaeological Officers’ Analysis and Recording for the Conservation and Control of Works to Historic Buildings, (ALGAO 1997). 1.6 Staffing, acknowledgements & archive Documentary research, site assessment and report preparation has been undertaken by Jeremy Milln. Property Manager Ana Chylak managed the overall glasshouse project. National Trust Archaeologist for the Region Shirley Blaylock prepared the Brief for the archaeology and advised until her departure in 2012, after when James Parry has been the point of contact. I am grateful to my cousin, Mr Tom Winter, for permission to reproduce the family photographs in the section dealing with the glasshouse at Arlington Rectory, and to my Aunt, Anne Nye, for giving a recorded interview for the National Trust Sound Archive of her memories of staying at Arlington with her family in the 1930's. Sue Luker, Arlington's Head Gardener, and curator Jeremy Pearson are also thanked for their interest and support. Thanks are due to the staff of David Salisbury Ltd who won the contract to supply the replacement conservatory for the site, in particular Mike Chinn, Gail Hodge and Gareth Elvidge of David Salisbury Ltd. The evidence presented in this report is intended to be proportionate to the interest and significance of the conservatory and vineries at Arlington and to their garden context. Sources for that evidence are acknowledged, especially the Garden Conservation Plan by Johnathan Lovie. However any opinions expressed are the author's own. 1.7 References General Association of Local Government Archaeological Officers (ALGAO), 1997. Analysis and Recording for the Conservation and Control of Works to Historic Buildings. Campbell, Susan 2005, A History of Kitchen Gardening Grant, Fiona 2013, Glasshouses. Shire No 747 IfA, 2001a. Standard and Guidance for Archaeological Desk-Based Assessment IfA, 2001b. Standard and Guidance for the Archaeological Investigation and Recording of Standing Buildings or Structures North Devon Record Office, cited NDRO in footnotes Specific Berry, N, 2008, Archaeological and Historic Landscape Survey of the Arlington Estate Lovie, J, 2009, Arlington Court, Garden Conservation Statement National Trust, 1996, standard guidebook, Arlington Court National Trust archive File 323, Box 348, location BJ held at Wansdyke, Wiltshire National Trust archive File A1/4 'Arlington Court Garden 1983-1992', from Killerton, Devon (This contains a folder with correspondence, copies of plans, planting schemes and photos of the central section replaced by the 1982 glasshouse amongst other things) Website: philipjebb.com/Biography.htm 1.8 List of Illustrations Cover Photo View looking NE of the Jebb Conservatory shortly after completion in 1983 Fig 01: Tithe map, 1842 – detail showing the relationship of the vinery/conservatory range (centre), pinery at top, walled and flowers gardens. Arlington Court is in the bottom left hand corner and Arlington Rectory in the bottom right hand corner. (left hand) Fig 02: Satellite photograph of the same view today (slightly enlarged) with the Conservatory at the centre of the picture and a more developed layout for both flower and walled gardens (right hand) Fig 03: First edition 1:2,560 scale Ordnance Survey, surveyed 1886 – detail showing the pleasure grounds Fig 04: Graham Thomas' general planting plan for the Conservatory borders, May 1973 Fig 05: Graham Thomas' detailed planting plan for the Conservatory borders, March 1976 Fig 06: Sketch proposal for a 'bower' to replace the Conservatory, by John Dyke, 1978 Fig 07: Timber sections proposed by Room Outside Ltd for new Conservatory January, 1981 Fig 08: Fig 08: Ground Plan of the Conservatory on which Jim Marshall has identified the plants he wishes to see Fig 09: Garden writer Don Hoyle opening the 1982 Conservatory Fig 10: Philip Jebb's plan proposal for the 1982 Conservatory (original at 1:50 scale) Fig 11: Philip Jebb's front elevation proposal for the 1982 Conservatory (original at 1:50 scale) Fig 12: Selection of setting out details from Philip Jebbs' office, 1982 Fig 13: Archaeological record plan of the 1982 conservatory in 2012 before demolition Fig 14: Interior view of the 1982 Conservatory shortly after completion Fig 15: Views around the W end of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 16: Views of the projecting front porch of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 17: Views around the E end of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 18 Fig 19: Fig 20: Fig 21: Fig 22: Fig 23: Fig 24: Fig 25: Fig 26: Fig 27: Fig 28: Fig 29: Fig 30: Fig 31: Views of interior and floor of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Views of interior and roof structure of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Views of various details: window catch and stay, automatic vent mechanism, rope edged slate path, steel strapping to roof timbers and heron finial of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 View across the garden terraces towards the W end of the glasshouse range with Compartment 1 hidden behind the twin Araucaria araucana, c. 1890 Long view across the garden terraces towards the E end of the glasshouse range with Compartment 5 hidden behind the same Araucaria araucana, c. 1890 J C Loudon’s illustration (1824) of a range of hothouses 2 constructed at Skrivelsby, Lincolnshire is a close parallel to the form of the early 19 th century glasshouse range at Arlington East vinery range showing centre pivoted casements, late 19th cent photo Late 19th century view of the interior of the glasshouse, probably one of the flanking compartments looking east. This appears to show vines against the glass as one would expect, but palms beneath and a rear wall trellised for climbers Apertures in the dwarf wall for admitting vines Shows the Rectory front with glasshouse attached to its right lean-to against a high wall, perhaps of a service wing. A boy of about 10 years, identified as Michael Cardus, the rector’s son, is seated on a donkey in front. Written beneath are the words 'Daisy and self'. It is springtime, about 1920 with daffodils and primroses in the lawn. A small stone figure holding a basket above its head, is to be seen in the middle of the lawn (Original measures 102 x 77mm) Girl of about 14 years, identified as Rosie Cardus, the Rector's elder daughter, seated on pony in front of the glasshouses. A label beneath says 'Sunny Jim 1932', presumably the name of the pony. (Original measures 82 x 55 mm) View of the right hand end of the Rectory at its junction with the glasshouse. Shows shrub or rose border in front, mid 1920s. (Original measures 85 x 60) General view of front of Rectory with stone urn on a plinth in front. Glasshouses away to the right, early 1920s. (Original measures 82 x 58mm) Closer view of the glasshouse with two figures standing in front identified as Gertrude Marguerite Milln (née Cooper) and Alexander Lindsay Milln, c. 1927. (Original measures 82 x 58 mm) Fig 32: Fig 33: Fig 34: Fig 35: Fig 36: Fig 37: Fig 38: Fig 39: Fig 40: Fig 41: Fig 42: Fig 43: Fig 44: Fig 45: 2 Plan of proposed conservatory by Richard Turner of Dublin, 1849 Conservatory design by J W Thomson of London Proposed free-standing circular conservatory by William Walker of Manchester (with the domed conservatory at Farley, Hants inset for comparison) Design for a proposed conservatory to be attached to Arlington Court Interior view of the c.1925 conservatory shortly before its demolition in 1980 Photo of c. 1930 showing that Compartment 3 had been rebuilt as a conservatory with a projecting central bay, leaving the early nineteenth century vineries either side intact Page from the Messenger & Co catalogue of c.1921 showing a projecting plant house of very similar form to the c.1925 Conservatory at Arlington Interior views of the c.1925 Conservatory showing roof structure taken in 1978 Interior views of the c. 1925 conservatory, taken in 1978 Exterior views of the c. 1925 conservatory, taken in 1978 General view taken in 1978 of the interior of the c.1925 conservatory with the cast iron support which Jebb found was too defective to reuse in the 1982 conservatory Plan by ‘W.A.’ dated 1814 for the original pinery (NDRO) Shows its western compartment with its and flue system for the succession pit and part of the furnace house Section by ‘W.A.’ dated 1814 for the original pinery (NDRO) Pinery – remains of structure as altered in mid-nineteenth century, July 2012 - J C Loudon, Encyclopaedia of Gardening (1824, ed of 1830), pl 451 Fig 46: Fig 47: Fig 48: Fig 49: Fig 50: Fig 51: Pinery back shed steps down to furnace room Pinery - W end of back shed to left, site of pinery to right of the truncated wall The lean-to glasshouse in the kitchen garden, c 1900 The lean-to glasshouse in the reinstated kitchen garden today Salisbury Conservatory nearing completion exterior, November 2012 Salisbury Conservatory nearing completion interior, November 2012 Chapter 2: Specifying the 1982 Conservatory 2.1 Overview of the Specification 2.1.1 The present Conservatory was constructed in 1982 to a design by Philip Jebb (192795), arguably one of the best traditional architects working in Britain during the post-war period (see Appendix 1 for biography). It is built of timber on low brick walls against a high rear wall inherited from an earlier structure on the site. Its length is 12.70 and width 4.35 metres. In plan it is 'T'-shaped, having a central show area which extends the middle out to 6.25 metres. The dimensions are broadly those of an earlier conservatory on the site which was taken down in 1979 (see section 3). Jebb was instructed to design a replacement for the earlier conservatory and was provided with a drawing of it done by the National Trust Assistant Building Manager Michael Board3. The original of Philip Jebb's drawing for this commission, dated 2 nd August 1980 and revised for issue on 6th Jan 1982, is lost. At its Killerton office, the National Trust holds dyeline copies which are referred to here. 2.1.2 Jebb's commission can not be considered conservation sensu stricto, neither really was it to be an act of reconstruction, for modern materials and techniques were employed. It was accepted at the time that a contemporary structure broadly resembling a traditional one, would meet the Trust's requirements. 2.1.3 2.1.4 Although this was no like-for-like replacement; as a commission the job offered little scope for the talents of a fine architect by then becoming known for designing neo-classical country houses. This was not, in other words, to be the dramatic statement of a creative genius and the job seems to have failed to engage his enthusiasm. Misunderstandings arose and there were delays to the production of drawings and the appointment of contractors resulting in some illfeeling between him and National Trust Regional Director Peter Broomhead. Jebb quit and the building project was managed by the Trust's own surveyor. So Jebb's conservatory was neither particularly informed by historical precedent nor was it designed and executed with his characteristic flair. In those days archaeological staff employed by the Trust were few and the opportunity to formally record and advise from a position of informed understanding was missed4. At the time the evidence for earlier structures on the site was rather unselfconsciously sacrificed. Still less was there an awareness that, had a specialist garden archaeologist been hired, the result may have benefited from being better informed by that which had gone before. 2.1.5 2.1.6 So for one reason or another the result turned out rather different from its predecessor and not entirely as Jebb specified either. In this section we will look at the building 3 At his death Jebb's material was divided between the RIBA library and the NMR (now 'English Heritage Archive') in Swindon. This is yet uncatalogued so I am most grateful to Ian Leith and Nigel Wilkins in Swindon and to Alun Martin in Cambridge, for checking their holdings. Michel Board's original drawing, on which the Jebb scheme is based, is also untraced. 4 Expertise on horticultural aspects was was provided in the National Trust by its gardens advisers: Graham Stuart Thomas between about 1957-70, John Sales 1970-81 and Jim Marshall from 1981. Advice on aesthetics including colours, fittings and furnishings was provided by its Historic Buildings Representative (Hugh Mellor). Garden and architectural history and the techniques of archaeology were not applied. Jebb intended, recording that which was delivered and examining the reasons for it, feature by feature. Jebb's own December 1981 specification, which is largely reproduced here, is used as the starting point for this analysis. 2.1.7 As an aside Jebb specifies the work using Imperial measurements, which might have been considered odd fully ten years after decimalisation, but which makes perfect sense in an industry dealing with structures first set out in them. On the other hand his drawings are scaled in metric. The Specification of Works for the job5 which Jebb prepared for the National Trust carries the usual Section 'A' General Conditions and Preliminaries, Section 'B' Trade Preambles, Section 'C' All Trades and Section 'D' Provisional and Prime Cost Sum. 2.1.8 2.2 Section 'A' General Conditions and Preliminaries In Section 'A', it is explained that the Form of Contract is to be a J.C.T Agreement for Minor Building Works (1980 edition) with Supplementary Memorandum as published by the R.I.B.A. inclusive of a clause covering P.C. Sums. 'A4' goes on to define, for the purposes of the Contract, what will be the dates of commencement and completion, that damages for noncompletion will be £25 pw, that the defects liability period will be 6 months, the Retention will be 10% and that Final Certificate period will be 3 months. 2.2.1 2.2.2 Section 'A5' goes on to require a FIXED PRICE tender for the whole of the work necessary to complete in all trades contained or implied in the specification. Section 'A7' deals with insurance, 'A8' with detailed pricing of the specification, 'A9' with contractor's rates, 'A10' and 'A11' with the procedure for Architect's instruction, contract variations and additional cost claims. 'A12' deals with day work rates, incidental costs and overheads, 'A13', 'A14' and 'A15' deal with site access and care of the National Trust's track and garden areas. 2.2.3 Sub-sections 'A16', 'A17', 'A18' and 'A19' address issues of site safety, both in terms of protection of property and the general public, especially when bringing in materials and removing rubbish. 'A20' deals with water and power supply, 'A21' with temporary site huts and lavatories, 'A22' and 'A23' with the location and conditions for storage of materials and glasshouse components, 'A24' with mixing of materials, 'A25' with what would be expected of the contractor in terms of conditions for his sub-contractors. Sub-sections 'A26', 'A27' and 'A28' deal with cleanliness, in particular the removal of rubbish, keeping drains and gutters free of debris and cleaning glazing inside and out at completion. Finally sub-sections 'A29' and A30' deal with statutory notices and consents in respect of local authorities and the police where appropriate. 2.3 Section 'B' Trade Preambles – use of materials Section 'B' is intended to ensure that the materials and components supplied in the job are assembled in a correct and workmanlike manner. In brief this covers: 2.3.1 Hardcore, which is to be crushed material recovered from demolition or alteration works. 5 Jebb, Philip Esq., A.R.I.B.A., A.A.Dipl., Beenham Hatch, Chapel Row, Bucklebury Common, Reading, Berks, RG7 6NR, Dated December 1981. Job No. A457/S3190 Concrete, which is based on Portland cement, clean sharp >5 mm sand and 18 – 38 mm graded aggregate in a proportion four parts aggregate, two parts sand to one part cement. Not to be done when temperatures fall below 5 degrees centigrade 2.3.2 2.3.3. Brickwork Quoting the specification in full: The bricks are to be second hand common and facing bricks as recovered from demolitions being carried out by the employer nearby 6. The contractor is to allow for thoroughly cleaning off all faces free from mortar and also sorting out bricks suitable for use in fair face work. The cement mortar for brickwork to be composed of four parts of sand to one part of Portland cement. Jebb adds: The brickwork generally to be built in English Bond. 2.3.4 Jebb then continues with a standard passage, one of many: All cross joints to be well flushed up and grouted and the whole carried up regularly unless otherwise required of directed, no portion being raised more than 3”0” above another at any time. Bricks with frogs to be laid frog uppermost. Courses of brickwork to correspond with existing as much as possible and no bats or broken bricks are top be used except where required fr bond. All perpends to be perfectly perpendicular and all quoins, reveals and angles to be well and truly kept. All bricks to be well-wetted before being laid except in frosty weather. 2.3.5 Damp-proof course This to be provided as a 1.3 mm thick 'Hyload' pitch polymer supplied in rolls from Ruberoid & Co and used to control damp movement horizontally and vertically. 2.3.6 Lead Sheet is specified to BS 1178:1969, milled and laid in accordance with the Lead Development Association's technical guidance. It is later confirmed that Code 5 would be used. 2.3.7 Carpentry & joinery Imported softwood, generically European Redwood or Whitewood, is specified and is to be sawn die square. Home grown hardwood is not to be used. Treatment against decay or infestation to be 'Protim' to the ends of timbers. The specification provides for the use of nails, glue and plywood to BS 1455 2.3.8 The specification for joinery makes clear how the timber is to be converted, that it is to be primed and the standards of workmanship expected, but is silent on details such as the nature of joints, fixings or even the scantlings, mouldings and sections to be used. This is intended to be determined later between architect and the contractor or his joinery sub-contractor. 2.3.9 Plasterwork, is to be cement and sand as previously specified 2.3.10 Painting. There are standard clauses for preparation and workmanship including when and when not to paint, on moisture and on sanding between coats. Priming paint to be white and red lead for timber and zinc chromate or calcium plumbate for metalwork. Shellac knotting and hard stopping to be used. 2.3.11 Lime-washing. Clauses on lime-washing provide for the used of mature lime putty obtained either from freshly burnt rock lime or limestone hydrate. The matured putty is expected to be mesh sieved and thinned and a preference is expressed for lime-wash derived from the slaking of calcined rock, rather than hydrated lime. Interestingly Jebb goes on to say: Alternatively skimmed milk may be used instead of water for thinning down the putty; it has the effect of binding the wash and making it resistant to rubbing and moreover gives a particularly smooth, velvety finish. Jebb adds: To inhibit mould growth an inhibitor such as 1% solution of sodium pentachloraphenate (NaPCB) should be incorporated in the lime-wash. 6 A marginal note in the Trust's file copy of the specification reads: New bricks required. NT to source. 5000 required, say £1000, half common, half facings” 2.4 Section 'C' All Trades – Schedule of Works Section 'C' is effectively the Schedule of Works against which prospective contractors are asked to tender. Subject to confirmation by Architects' Instruction, the items of work in this section define what would actually be done on site. For the purpose of archaeological record and analysis, the Schedule is examined in detail. Completion of Demolitions C/1 Schedule Description Research notes (and see section 2.5) 1 Take up and set aside for reuse the slate threshold N/c slab some 5' x 10” Evidence seems to be that all new Delabole state was used in the 1982 work. 2 Ditto the slate slab inside last some 5' x 2' See note above 3 Take up and set aside for use as a template the It was originally intended present concrete foundation stone to old column to reuse the c.1920s cast base (now fractured beyond reuse) iron pier in the new work, but appears to have been damaged during dismantlement in 1979 No evidence the new pier is similarly supported – Tamlyn, the main contractors, cast their own. 4 Take up and stack on site all the bold roll flower border edgings (sic) to the projecting bay part of the old building complete with their cement mortar backing fillets (to be reset as needed by the Gardener) Surviving photos of the c.1920 Conservatory show a ceramic rope edging tile used for borders None of the edging instructed to be taken up was reused in the Jebb 1982 project. Instead a cast concrete slab with a rope edge supplied by Haddonstone following advice of Hugh Mellor (NT Hist Building Rep)was used. 5 Take up and set aside where directed the 4 no iron floor gratings set flush with the existing slate and stone pavings adjacent the concrete foundation stone got out previously. These would be cast gratings for use over piped heating ducts, but possibly not used as such in the c.1920s Conservatory which lacked this type of heating so far as we are aware. The piped heating ducts may belong to the 19th century glasshouse. The gratings were not reused in the 1982 Jebb conservatory, however there is a single such grating reused as a threshold at the entrance to the Garden Bothy (see photo). It is of 19th century type 6 Take up and set aside where directed the 4 no iron Comments as above floor gratings, two next to each end wall. Comments as above 7 Take up adjacent to each end sufficient of the stone/slate pavings in steps and landings together with their brick foundations to allow for getting out the old and getting in new foundations to the brick perimeter walls and set all aside for re-use by Gardener : in conjunction with ditto get out and clear away remains of old heating pipes built in below pavings The loss of the heating pipes and their sub-paving ducting, the dwarf walls and their footings from the earlier builds have not been confirmed by excavation as outside the scope of this project. 8 Beyond the panel of 4 no floor gratings lifted Not re-used, at least not in Items mislaid? adjacent the old column foundation take up a panel the Jebb conservatory of 3 no part slate/part cement in repairs pavings and set all aside for reuse N/c In practice all such paving was taken up and not reused, at least not in the Jebb Conservatory. Archaeology (and see section 2.5) Completion of Demolitions (cont'd) C/2 Schedule Description Research Notes Archaeology 1 Pull down and clear away the perimeter walls left from the previous greenhouse (timber-work already collapsed and cleared away) to at least 3' 0” below existing ground level measured on the outside of the wall and break up the foundations as found and leave trench for new foundations some 2'9”/3'0” wide. Trenching for 1982 Conservatory known not to be nearly so deep as specified. (archive item 55) Not verified, but no reason to doubt that all evidence of the earlier was destroyed while digging footings for the 1982 building. Investigation of the walling for the flanking vineries may indicate this was largely constructed on piers arched over and not as a continuous footing. 2 At one abutment dig a pit and cut off at least 3'0” Comments as for C/1.5 down the remains of an old cast iron heating pipe riser No evidence of this apparently now survives 3 At opposite end repeat ditto including pulling down It is not clear what this and clearing away also the stub of old walling left walling refers to, but in situ at abutment with the back wall. possibly a support structure for 19th cent heating pipes It is outside the scope of this project to excavate the conservatory site to determine. New Work to receive Glass House 4 Into trenches left when grubbing out the old base walls and foundations, cast 1:2:4 concrete strip foundations 3' 0” wide c 1' 0” thick for entire girth of the new building as proposed including across the 3 no door openings. 5 Level and ram the trench bottoms before laying last N/c concrete N/c 6 Where trench bottom is below the foundation level of the back wall, dig out 2 no pits and underpin same in say a 3' 0” x 2' 0” x 2' 0” deep block of concrete packed in tight to underside of wall. Allow for formwork in 2 no panels 3' 0” x 1' 0” high. Most unlikely to have been required, but it is not recorded whether done due to lack of monitoring during the work on site Of little archaeological interest in itself; the loss of historic fabric affecting the party walls within the Geo range, went unrecorded. 7 Build up in second hand bricks 9” foundation walls in English bond off the strip footings in cement mortar around entire perimeter including across the door openings: for pricing take an average height of 3' 6” for this brickwork ; level and prepare for D.P.C. Dense concrete blockwork Of little believed used below interest. ground level. Above the brickwork follows the spec. The use of D.P.C here likely to have exacerbated decay in the sills later as traps descending damp. It was suspected by the QS Could easily be verified by at the time that, in the excavation, but of very little absence of the architect, the archaeological interest contractor reduced the foundation spec, presumably deciding that the ground conditions were good for bearing the comparatively trivial load of a glasshouse. New Work to receive Glass House (con't) C/3 Schedule Description Research Notes Archaeology archaeological 1 Backfill in over spread of concrete footing on inside N/c face of new foundation wall with hardcore well rammed and consolidated up to 6” of ground level and leave for topping up with top soil by Gardener. Not verified by excavation as outside scope of this project. Of little archaeological interest anyway. 2 Repeat ditto on outside face of wall up to ground N/c level with run of agricultural drain pipes laid along outer edge of footings spread to act as French drain to take away rainwater. Provide 8 no 4” stoneware bends to take pipework around quoins and at angles As above 3 At the quoins dig out 2 no soakaway pits say a Cube Yard each below pipe invert and backfill each with graded hard material to allow for soakage : enlarge the foundation trench to spread out over the area of each soakaway and continue hard backfilling up to general ground level Not clear if this is As above intended to take rainwater from roof gutters and downpipes as well? 4 Protect the bottom three courses of brickwork at ground level against splash by applying around entire external perimeter two coats hot tar brush applied over brickwork with all pores etc filled in To guard against breach of low-level D.P.C, but not apparently carried out, presumably for aesthetic reasons 5 Put in 'Hyload' pitch polymer flexible D.P.Cs across N/c tops of foundation walls lapped 6” at junctions : continue the D.P.Cs across the doorways to pass under the thresholds. As above. No evidence this practice was ever done to glasshouse walls historically. Probably undesirable to impede vapour permeability of brickwork anyway. As above New Work to receive Glass House (con't) C/4 Schedule Description Research Notes Archaeology 1 Above D.P.C levels continue building up in second hand bricks laid in cement-lime-mortar to line of new cill in 9” work in English bond with the outer face built of selected bricks and fair-faced and flushpointed with a bagged joint to approval : for pricing take an average height of 2' 0” : level and prepare for topping course The cement content in the mortar, while unfortunate, does not seem to have adversely affected the condition of the bricks. Evidence is constructed as specified. Rises to 6 or 7 courses above ground level or just under 2' 2 At the 2 No. abutments of last brickwork with the existing back wall build in 2' 0” lengths of ¼” x 1¼” galvanised water bar bent to 90° with 18” tails split and fanged and 6” uprights twice drilled and galvanised screwed to Hilti plugs let into drilled mortices : space the ties every three courses up from the footings and staggered across width of 9” work. Modern technique for fixing new masonry to old without cutting in for a traditional bond. Unable to verify without dismantlement, but no reason not to suppose constructed as specified. 3 Where irregular angles occur along the front wall allow for overlapping of brickwork below ground level : above ground level allow for selecting the best of the second hand bricks and sawing to shape using a powered clipper saw to obtain neat angles : rub to a rustic texture with wire brushes at completion : reinforce the angles where such cut bricks used with strips of galvanised brickwork reinforcement or chicken wire meshing bedded in alternate courses and kept beck 2” from front face Modern technique for Evidence is constructed as coping with the four 45° specified. angles either side of the front projection without using specially moulded bricks as would have been the case in the past. 4 Finish off top of last 9” walls with a brick-on-flat cill course set weathering in mortar to overhang wall face below some 2” and with the space behind filled in with a 4” x 3” pressure-treated softwood wall plate so In spite of the emphasis given by Jebb, it would seem this sill, which soon decayed badly trapped in The Trust responded to decay in this wall plate by applying a painted cover board on the inside face, in set as to master the top of the cill course by some ½” its pocket, was not properly treated, but NB see discussion on choice of timber and inherent unsuitability of that specified by Jebb. an effort to deflect condensation moisture coming down inside the glasshouse from entering the timber. This seems to have trapped damp and exacerbated the problem. 5 At splayed angles of the brick cill course cut and form N/c, of little interest splayed and mitred meetings Evidence is constructed as specified. 6 Secure the backing timber wall plate every 2' 6” with N/c an 18” length of ¼ x 1¼ galvanised iron water bar bent over and let in flush to top edge and drilled and galvanised screwed on and the tail zig-zag bent back to face of wall and twice slot drilled and galvanised screwed to Hilti plugs morticed into inside face of brickwork. Put 2 no such cleats where lengths of wall plate are halved and mitred at headings and angles. Evidence is constructed as specified. New Work to receive Glass House (con't) C/5 Schedule Description 1 Level through the above wall plate to Specialists's satisfaction inserting hardwood packings if required dipped in timber fluid before driving home. 2 To provide for stabilising brick cill course drive in a 6” galvanised spike to face of timber plate at every fourth joint and build in same. 3 At the 6 No. door jambs and the 2 No. 90° quoins build in total 16 No. stout galvanised coping irons with 2” over wall face overhang reduced to ½” only next door jambs. 4 Cover down the entire perimeter of the brick cill course with a 5 lb lead weathering strip total 13 ½” girth and laid in lengths not exceeding 7' 0”, the back edge welted and set to fit into the ½” gap between top of brickwork and underside of Specialist's timber framing and the front edge bossed and dressed to act as overhanging drip and kept ¼” clear off nosing: where headings occur and at angles allow for 9” laps : allow for lead burnt returns at the jambs of 3 No doorways and the 2 No. 90° quoins and at the 2 No abutments allow for turning up some 6” to nearest joint and cutting out a groove and letting in some 2” and lead wedging in and making good pointing. 5 To secure last lead covering provide at 1'0” to 1'6” intervals (allow for pairs of fixings at overlaps and angles) stout copper tingles 1¼” wide by some 18” long copper tacked to wall plate and bossed and dressed down top of brickwork under leadwork and the last few inches doubled over and then turned up and clipped over the front overhanging drip edge to show some 1” deep on face. 6 Prior to finally positioning each length of leadwork paint a generous coat of bituminous paint on underside to prevent chemical attack from the cement lime mortar bedding of brick cill course New Work to receive Glass House (con't) C/6 Schedule Description 1 In conjunction with the assembling and fixing of the sundry Glass House components by Specialists, take from store and reset the existing slate thresholds, previously taken up in Items C/1/1 and 4 and set aside for re-use, to the pair front doors and the 2 No end doors, bedding ditto over D.P.cs on beds of fine concrete to exact levels as required. 2 At the 2 No. 90° quoins where remains of old dwarf walls to Glass House still in situ but not going to be rebuilt, carefully cut back existing brickwork with concrete cill over and requoin up to leave as a fair end with gap of some 3' 0” formed for passageway : form neat fair end and to the cills also and make good rendering to approval. Work to the Back Wall 3 In approximately 12 No. instances cut out old timber ends/clean out pockets where old timber ends already cut away and build up solid with salvaged bricks in mortar ready for re-rendering. 4 14 No. cut out sets of Vine eyes and wire from face of brickwork and hand to Gardener. 5 Go over whole of the wall face being enclosed by the boundaries of the new greenhouse and withdraw all old nails and fastenings in addition to the above and hack off the whole face free from traces of old cement backings etc and leave ready for re-rendering. 6 Take down carefully, working overhand at the doorway end of the wall the top 2'0” to2'6” of the existing mostly stone/part brick construction stopping off at the pronounced horizontal chase where an old plate member previously let in and set aside all materials for re-use. In conjunction with ditto break up the fine concrete capping and allow for extra cost of getting out sundry old ironwork in tie rods, truss ends etc cut off and left in situ when the old building collapsed: allow credit for lead being salvaged against this item. Work to the Back Wall (cont) C/7 Schedule Description 1 At each end of the upstand panel of walling still left standing, carefully pull down an area say 7'0”/8'0” long x a height of approximately 2'0”/2'6” and set all materials aside for reuse. 2 Using the materials set aside rebuilt up the wall top to profile as shown on “New Proposal” Drawing with both ends where taken down above rebuilt with sweeps to master changes in levels : where the rebuilt work is exposed to the rear face overlooking the potting shed and at both swept ends where visible on the front face execute this using selected stonework and leave with neatly pointed joints to match the work adjacent : for the work being covered over by the new Glass House rebuild with what suitable stonework remains and make up deficiencies with sound second hand bricks with the joints left raked out to form key for new rendering. 3 Finish the top of the wall as newly rebuilt with sweeps each end with a fine 1:2:4 mix concrete in situ capping some 2'3” wide x say 3” thick to match existing with a very slight fall across the wall width to throw off water. Incorporate stone dust to give a colour similar to existing. Provide all necessary wrought timber fillets to both edges to give throated overhangs : also incorporate 8 No thin timber slips to form expansion joints at junctions with existing, at foot of sweeps and remainder to the length at high level and subsequenty scrape out and point up with a mastic filler knifed off smooth. 4 Put in below last new concrete cappings 2'0” wide strip off flexible D.P.c as before to keep out damp from stonework and point in both edges with gun mastic. 5 To requirements as obtained from Specialist Glass House Sub-Contractors, provide and build in for depth of 6” a new 5 lb milled lead cover flashing some 2'0” total girth in lengths not exceeding 7'0” long to overhang the level and both raking abutments of the new patent glazing roof up to the rear wall. Work to the Back Wall (cont) C/8 Schedule Description 1 Where new raking abutments are against face of existing undisturbed walling cut raking grooves into thickness of brickwork/stonework for minimum depth of 3” and wedge in the leadwork with back edge welted and secured with lead wedges at about 18” intervals 2 Lap all junctions of the foregoing lead at least 6”. 3 In conjunction with the patent glazing fixers neatly boss and dress the lead overhangs down over the upstand bars and out onto the glass, trimming the bottom edge to allow for say a minimum 8” overlap. 4 Carefully dismantle the existing white painted arch headed door size some 3'6” v 6'3” high complete with its frame and set aside the whole off site in the dry for repairs : cut out and make good to the jambs and the arch where old iron fastenings to be removed : cut out and make good to the jambs where old timber fixing blocks are to be/have been removed and make good for re-rendering. 5 Repair the existing door as required stripping off all old paint at Shop and remaking the open joints : subsequently bring back to site and rehang on and including 1½ pairs 4” heavy cast iron butts and galvanised screws and Fix only a new mortice lock and furniture making good where al old ironmongery discarded. 6 Provide and fix into existing masonry opening a new treated softwood frame, rebated and machined and of the same sections as existing complete with arched head to accept last rehung door and secure back to masonry using 8 No. rawlbolts with large galvanised washers and the drillings filled up smooth with hard stopping and rubbed down smooth. 7 Pack in against masonry with Plumbers hemp behind frame to keep out weather and point both sides of opening with gun applied mastic. Work to the Back Wall (cont) C/9 Schedule Description 1 2 No. Provide pads of 5 lb lead from offcuts and isolate feet of frame from existing slate threshold to avoid decay. 2 Knot prime and stop and paint two undercoats all round (N.B. Bottom and top edges) on last repaired door and new frame at Shop : after fixing and stopping of drillings etc. touch up and paint a further two coats oil colour finished eggshell on al exposed surfaces. Specialist's Glass House Construction 3 The whole of this work is to be sublet in accordance with an Architect's Instruction – see P.C. Sum included in Section D. Attendance on Specialists 4 Provide for attendance on the Specialists whose work comprises:(i) Provision and erection of timber-framing to sides and slopes (ii) Provision and erection of doors, windows and roof ventilators (iii) Provision and erection of clear glazing fixed with beads to sides (iv) Provision and erection of patent glazing using aluminium bars flashed with lead and fitted with Georgian wired glass : also inclusive of 4 No. opening roof lights. (v) Provision and erection of aluminium gutters and 4 No. downpipes. Sundry Builders' Work C/10 Schedule Description 1 Clean off the entire delivery of timber components being received from the Specialists and apply one coat of aluminium primer on already pre-primed surfaces and two coats of oil paint before erection followed by touching up and making good and two further coats finished matt on components after erection by specialists : the work comprises:  Wall plates and ridge  Rafters, also hip rafters and valley rafters  Framing to sides consisting of head plate, mullions, transomes and cill as filled in with sashes  2 No. Single and 1 No. Pair half glazed doors complete with frames and transome over front pair doors 2 Provide to support a new central column (old column originally thought to be re-usable but found to be defective due to rusting) a replacement pre-cast concrete foundation stone size some 13” x 13” on plan x say 15” deep (as that set aside for copying) with circular recess let into top face say 10” diameter to accept column base plate. 3 Provisionally Allow for casting into last 3 No. 12” long x ¾ diameter galvanised holding down bolts each with large plate washer and washer and nut plus polystyrene block threaded over same to allow for subsequent loose fit. 4 Set last stone on and including a pad of 1:2:4 concrete say size 3'0” x 3'0” x 3'0” deep including digging pit some 4'0” deep for ditto with base levelled and rammed hard. 5 Provide and use sets of steel folding wedges and grouting in to get foundation stone to exact level and afterwards backfill around to within 6” of ground level inside with a splayed concrete fillet say 12” x 12” extreme trowelled smooth. Sundry Builders' work (con't) C/11 Schedule Description 1 Fix only replacement column of new hollow steel or second hand cast iron as found available some 16'0” tall complete with base plate and cap plate drilled to template as required by Specialists for his bolted fixings 2 Wire brush down and prime with anti-rust primer and paint three coats oil colour on last column say 18” girth together with cap and base plates. Sundry finishings 3 Around entire internal face f the new perimeter walls and across the whole of the back wall area as will be enclosed by the new Glass House, wire brush off to remove effloresce caused by drying out of brickwork and apply an anti-fungicide spray and then three coats of cement, lime and sand 1:1:6 rendering finished with a wooden float. 4 Make good rendering over cleats holding down wall plate : form slightly rounded arrises to external angles. 5 Along the bottom edge say 6” above bed level and up the sides of door openings put in a “Catnic” or similar galvanised render stop bead screwed to Hilti plugs let into drilled mortices. 6 Along the top edge next the wall plate around perimeter and up the abutments along the new wall put in a 'V' joint to mask cracking. 7 Put in 5 No. V – joints vertically to the back wall to try to eliminate cracking with panels put up alternately 8 At completion of drying out of rendering make a return visit to site and brush down and apply up to four coats of lime wash over rendering. 9 Prime and paint as Item 2 above on aluminium eaves gutters (inside and out) and 4 No. down pipes. 2.5 Discussion: Specification and Implementation 2.5.1 Brickwork The brickwork was relaid in 1982 to English Bond (alternate courses of headers and stretchers) the thickness of a brick (9” or c. 240mm) using a reclaimed 3” thick hand-made nineteenth century brick. The mortar is cementitious, this being a project dating from before the current revival of understanding of the superiority of lime mortars in conservation work. Visual inspection of the mortar which is hard and has clearly not weathered sacrificially in relation to the brickwork as it is supposed to do, would suggest that if it contains any lime at all this will have been used only to gauge or plasticise the cement. 2.5.2 Mortars & lime-washes The specification for lime-washing is one of the longer passages and is ahead of its time in terms of good conservation practice, although it is odd that the substrate on which it is to be applied is specified to be cement-based render (the same mix as the brick mortar) rather than a lime plaster or stucco Much attention is given to the quality, composition, surface preparation, brushing technique, drying time and number of coats for the lime-washing, so it is curious that Portland cement is still being required elsewhere in the job. Lime's renaissance in good building practice began in the early 1980s, but specifiers at the time were clearly not sufficiently confident to allow it to overtake Portland cement so specify both, unaware that the two materials are fundamentally incompatible. In practice cement-based mixes were predominantly used for the 1982 building as were modern alkyd paints. 2.5.3 Back wall The stone built back wall is, at the time of this survey, currently heavily obscured by plant growth. The wall has a high central section with scalloped ends against which the central conservatory is fixed. The lower flanking sections served to support the lean-to vineries either side and seem to have been flued, known as a hot wall. Rear (N) Elevation Some of its rear (north) face is visible and a detailed drawing at 1:20 scale was started on 16 th September 2009 by Anja Fischbach. Features observed on this N elevation to west of the back sheds include:  3 No. equally spaced circular iron stop ends, identified as having tied the wall plate to the c.1930 building  5 No. equally-spaced vertical slots, now stopped up. These may belong to rafters for a back shed range whose roof line was rather higher than the existing to the left. An alternative explanation for these is as passes for chains or ropes to work the roof sashes of the 19 th century glasshouse  3 No. equally-spaced holes in a level line, each blocked with a single brick header. Thesre are too small for hot wall flue service points,so may simply have been for inspection  a furnace point located at the far W end concealed within a small fairly modern timber shed. Early photographs show a small chimney atop the wall at this location. Front (S) Elevation Little of this is visible and that which is within the conservatory itself, has been obscured by cement render applied in 1981. The wall was provided in 1982 with ductile galvanised steel wire mounted on drive-in steel eyes set at 18” vertical spacing. Brickwork of the vertical chimney flue was observed immediately outside the east side door of the Conservatory and the flues of the hot wall beyond (to the E) can be verified [occupied by nesting birds at time of survey]. Jebb's reference to 'old timber ends' in Item 3 on page C/6 would be to sill and wall plates and presumably rafters and the ridge piece where they contacted the wall. Evidence lost or obscured in 1982, but for the top plate of the lean-to flanking ranges, some evidence survives, to research later if the flanking ranges are to be rebuilt. Jebb's requirement in Item 6 on page C/6 for the removal of old nails etc robs us of archaeological evidence for the way the exterior face of the hot wall was used within the glasshouse. However for the former flanking ranges some of this evidence survives, but is heavily obscured by modern climbing plants. Hand wrought iron nails were used with strips of cloth for tying back wall-trained fruit; in this case probably such as apricots. Later fixings are characterised by driven steel eyes or brackets for galvanised wire. 2.5.4 Joinery Overall the construction of the Jebb 1982 Conservatory differs little from its c.1925 predecessor. It is timber-framed with structural posts, sole plates and wall plates to wall areas. Windows are framed up with ¾ height transoms and filled with non-structural casements having top- and sidehung openers. The roof structure differs slightly in that the Jebb building opts to support its glass with rafters only while the c.1925 has rafters only every two glazing bars to the main roof section. The change probably reflects a precaution against defection caused by a heavier weight of glass for the 1982 building. Perhaps the main change from the c.1925 is in the decision to omit the end rafters and the lower diagonals which contacted with the flanking ranges and instead express the ends of the Conservatory as glazed hips. Jebb specifies a wall plate and provides for it to be thoroughly fixed to the wall using galvanised straps (see Item C/5.6). We would expect this of a wall since normally a roof sits atop and the strapping is provided to ensure the spread of the rafters resting on the wall plate do not force it outwards. In our situation a wall plate is not necessary. Indeed, given it is trapped on three sides at the base of the condensation stream from the internal glass face, a wall plate here is disastrous. It began to decay very quickly. The Trust responded by applying a facing to the interior front of the plate in an effort to deflect moisture from condensation, but this merely enclosed the plate completely guaranteeing that it would be unable to dry out properly. The disintegrating wall plate was the single most pressing defect of the building resulting in its replacement in 2012. 2.5.5 Timber Jebb's choice of timber is surprising. European Redwood AKA Red Deal, Pinus sylvestris from the Scot's Pine and European Whitewood, Picea abies from Norway Spruce, are both non-durable, lowdensity timbers unsuited for exterior joinery, at least not without comprehensive chemical treatment. Specifying only Protim treatment of the ends of timbers will have been insufficient to arrest decay once defects in paint protection had begun to appear. Normally Western Red Cedar, Thuja plicata, is used for domestic timber glasshouses as it is much more durable and resists warping and movement due to atmosphere changes. For larger glasshouses the stronger, denser Douglas-fir, Pseudotsuga menziesii, is preferred of the choice of commercially available softwoods. In practice the softwood was not used for the eight corner posts and here an imported hardwood was used which fared much better. Jebb proscribes the use of native hard woods, an attitude we seldom see today. Indeed so complete has our appreciation of the value of these in terms of their durability and environmental and economic sustainability over the imported timber prescribed is that today specifiers seek them out. Oak, for example, is an excellent choice for sills. Also distinctly anachronistic is Jebb's requirement of the use of white lead paint for putting together external joinery. For health reasons it is banned today for other than Grade I and II* listed buildings under licence. 2.5.6 Floor In practice no components of the earlier floor were retained in the 1982 work. The 1982 floor introduces all new beds, paths, steps and edging material. Internally the conservatory is arranged to beds and paths on two levels, stepped between.     Welsh grey slate, mechanically sawn to pieces measuring 3' x 2' (915 x 610 mm), makes up the path from the front entry. This divides around a central planting bed and leads to pairs of steps, each with two shallow treads, to reach an upper level walk. This is also done in grey slate the full length between doorways at E & W and of the Conservatory. Cast concrete, marketed as 'Haddonstone', is used for the path edging and this has a rope moulding to resemble traditional ceramic edging tiles. The concrete edging is itself concreted set in concrete A geometric patterned, predominantly red, quarry tile is used to define the lower standing areas in each of the two sides of the Conservatory Planting borders fringe the floor against the side and front walls and there are five planting pits against the rear wall which are also outlined with concrete rope edging. 2.5.7 Detailing, metal fittings etc The joinery for the 1982 Conservatory is without ornament. Timber sections are entirely utilitarian and without moulding. The exception being the dado plate subsequently applied to the face of the wall plate in an effort to arrest its decay. This has a simple ovolo to its upper surface. The joinery furniture consists of domestic brassware stays and catches for the casements (two to each to each window) and lever handles for the doors. Slot screws have been used. For the roof vents openers are automatically-operating, bi-metallic strip devices. Few still working at time of survey. Steel straps were retrofitted to strengthen the joints between rafters and purlins in the roof and between posts and rafters at eaves level. 2.5.8 Glazing Modern float glass is used throughout the 1982 Conservatory. For its c.1925 predecessor a drawn horticultural sheet was probably used and for the c.1840 original glasshouse on the site the glass will have been hand-made cylinder. The glass used today is typically twice the thickness of 19 th century glass which adds appreciably to the weight and largely explains the comparatively thick glazing bars used. For the Jebb house panes are cut to 12” width for the front wall glazing and 14” for the side, except for the side wall corner lights which are 19”. These widths are comparable to those of the c.1925 conservatory, but more than twice that for the original c.1840 glasshouse. In terms of length, Jebb's panes are very long indeed. Traditional laps are avoided and as a consequence panes to wall glazing upper lights are 12” and for lower 41”. For the roof panes are even longer and are jointed horizontally only at the purlins. For the c.1925 house the photographic evidence suggests panes lapped at about 18” and for the c.1840 at about 9”. The modern tendency to eliminate laps has advantages both horticulturally and from a maintenance point of view since it avoids the trapping of dirt between panes. Chapter 3 Genesis and construction of the 1982 Conservatory [NB: Bracketed numbers (00) refer to documents copied in full to the digital archive with this report. The source, in most cases, is from File 323, Box 348, location BJ held at Wansdyke, Wiltshire and file A1/4 'Arlington Court Garden 1983-1992, from Killerton, Devon] 3.1 Summary 'A new conservatory was constructed in 1982 to serve as the focal point of the flower garden. Although it has been stated that this structure incorporates elements of the original building, it must be noted that the present conservatory does not replicate the early C19 conservatory in either plan or form' (Lovie, 2009) 3.2 Central Conservatory of c. 1925 The National Trust's general file7 for Arlington starts with a letter from Miss Rosalie Chichester to the Secretary, Mr S.H. Hamer dated May 3rd 1921 in which she requests a visit to discuss her will and the future of her property. It marks the start of a period of regular correspondence almost to her death in 1949 at the age of 84. Having finally cleared the debts inherited from her father Sir Bruce Chichester on his death in 1881, she had begun to turn her attention to some undoubtedly overdue restoration and to the future of the property generally. About this time Miss Chichester seems to have had the projecting central section of the old vinery, whose condition had presumably deteriorated more than its wings, rebuilt. The new structure, although broadly of the same scale, was to a different plan and glazing arrangement. The precise date is uncertain, but it certainly came after 1903 and before 1949 as the OS map evidence attests. From the photographs of it we have stylistically and structurally it looks about 1925, and perhaps the rest of the structure was simply repaired 8. To Jonathan Lovie (2009) the projecting canted bay stylistically resembled the work of Messenger & Co, a leading glasshouse manufacturer based in Loughborough, Leicestershire, although Miss Chichester is not listed among that firm's clients in its 1926 catalogue9. In fact most major glasshouse builders of the day including Richardsons of Darlington, Foster & Pearson of Beeston, Boulton & Paul of Norwich, or Halliday & Co of Middleton included generically very similar conservatories in their trade catalogues. Although the authorship of Miss Chichester's conservatory remains unknown and it is perfectly likely to have been the work of a local joinery firm, we can say that it was entirely of the day. It was designed to link with the pre-existing vinery wings and was accordingly provided with second diagonals to its ends at the level of the lower flanking wings, but it was not intended to emulate them stylistically. Towards the end of her long life the condition of the fabric of the Estate declined. The Trust was trying to get a sense of how to cope when the day came. A memorandum from Freddie Reeks, the 7 File ref 323 for 1921-1948, held at Wansdyke (previously at QAG, London), Box 348, location BJ - Messenger & Co are known to have undertaken similar work, including adding new sections to existing glasshouses at, among other sites, Chiswick House, London. 8 9 - Messenger & Co Ltd, Horticultural Section, 5th ed (c 1926) Area Agent based at Holnicote, reveals that it was not the idea Arlington would be open for public benefit at all: I have not allowed for upkeep of the gardens as we have never decided on what scale these are to be maintained. In any case their upkeep costs would be put on the tenant of the Mansion House. 3.3 The Trust in the early days at Arlington At Miss Chichester’s death in 1949, the Estate passed to the National Trust. Photographic and other evidence suggests that after years of War-time privation, and a degree of neglect towards the end of her life, the buildings and landscape had declined significantly from their late nineteenth century heyday. The National Trust's files are silent on the subject of its glasshouse range in the first few years of its management. The Trust was too busy elsewhere rationalising buildings in poor repair and downsizing Miss Chichester's vast collections. The Trust was forced to demolish Sir Bruce Chichester’s dining room in order to deal with dry rot. Other structures including the ice house and the pinery behind the walled garden were demolished or abandoned in the 1950s. The outer wings of the glasshouse range in the flower garden followed in the late 1960s. At the same time, the Trust undertook important work in clearing the excesses of growth which had arisen during Miss Chichester’s latter years and developing the horticultural interest of the pleasure grounds. Improvements were also made, including the construction of the south and west ha-has in the 1970s and the re-instatement of the Victorian character of the flower garden in the 1980s. 3.4 Emerging debate about the glasshouse Comments from Graham Stuart Thomas, the Trust's Gardens Adviser, visiting in May 1957, are written in the context of emerging debate about the future of the gardens and the glasshouse in particular. The Adviser, having recommended the felling of two large 19 th century monkey puzzles from the terraces to south of the conservatory portion, writes: The greenhouse needs reglazing and painting but it is in sound condition otherwise. It is large and acts as an embellishment to the design. It produces some flowers for the house and as the land available for this purpose is hardly adequate I think we can make greater use of the greenhouse It is however, rather an expensive embellishment for a property where there is little actual need for it. It would seem a pity to pull it down as it is strong and well made. … This whole garden and its greenhouse are now open to view, owing to removal of trees and shrubs during felling, and the sight of the garden from the drive strikes one as sympathetic with the rest of this spacious place. (08) Freddie Reeks, still the Trust's Area Agent, replies to Thomas on 10 th July 1957 to point out that Arlington has only one gardener: but such things as the painting and repair of the Conservatory will in my view have to be weighed against the needs of the painting and repair of the farms and cottages on the Estate, many of which have not received attention for over 35 years Discussion continued in the early days. James Lees-Milne in a memo of 27 th June 1960 to Rodney Feddon wrote: We both agree that the greenhouses in the garden should be retained; and this means that money ought to be spent on them. They are deteriorating. I hope the Trust will consider them under general maintenance of the Estate, and that they will not be looked upon as a superfluous luxury. (09) Fedden follows this up on 29th June with a memo to Mr G. Michael Trinick, the recently appointed agent: As I believe you know, Lees-Milne thinks that the greenhouses should be retained and I hope you can arrange for money for their repair to be included in the estimates. Sadly the Trust locally felt it was faced with an impossible choice. The 1960s was a bleak decade for historic buildings with rising costs and diminishing central support. The decision was taken to demolish the flanking vinery ranges, relict of the first phase of glass. An ornamental planting scheme, devised by Graham Thomas which included wall climbers and flowering shrubs, was implemented for the site of the two ranges. By 3rd November 1972 Rodney Fedden is writing to Michael Trinick of the central conservatory which had been spared: I shall be sad to see the last of the Conservatory go, but I accept your arguments about scale. I hope you will plant some new monkey puzzles! With regard to your idea for a bower or largish summer house on the site of the Conservatory, you might get something original and more to your purpose from a stage designer or painter such as John Craxton. The design could probably be carried out by a competent builder without resort to an architect. One the other hand, if you want an architect I would suggest Philip Jebb ARIBA (140 Sloane Street, London, S.W.1 tel: 01-730 6383 and Francis Pym ARIBA (11 Buckingham Street, London, W.C.2 tel: 01-930 0171) (10). Michael Trinick replies to this suggestion in a memo to R. E. Meyrick dated 15th March 1973: I have been able to have a useful talk with John Craxton in London last week, through the good offices of Roger Meek, and we were able to discuss in detail the summerhouse which he is is to design for us to replace in due course the conservatory. I envisage this might happen when the monkey puzzles finally have to be felled We have in mind a summerhouse with a Regency umbrella roof and the possible tying in of this feature with the very extensive and high brick walls at the top of the garden by rows of pleached lime. Anyway, Croxton has asked if we can provide him with a scale plan of the whole of this enclosed garden, which should have noted down on it the height between the several levels and anything else which will enable him to make a model.(11) . Little progress was made over the Conservatory in the ensuing four years. However a sketch plan, dated May 1973 (see below), was duly produced of the site of vineries etc so as to make a start on the garden planting. (12). Fig 04: Graham Thomas' general planting plan for the Conservatory borders, May 1973 A further plan, apparently also the work of Graham Stuart Thomas, is dated March 1976. This deals in greater detail with the planting of the so-called 'Victorian Garden' borders adjacent to the Conservatory. (13). Fig 05: Graham Thomas' detailed planting plan for the Conservatory borders, March 1976 On 8th July 1977 David Bett ARICS, then Assistant Agent, wrote to Michael Trinick: I must confess to having made no progress whatsoever over drawing up a better plan for replacing the Victorian Garden Greenhouse with a bower. Looking at it again the other day I was struck by the fact that the greenhouse really does seem very appropriate to the garden: indeed having demolished the monkey puzzles, to go on to demolish the greenhouse would really drastically alter the garden. We have some money approved and I honestly feel this would be better spent in thoroughly overhauling the greenhouse so that it can be retained indefinitely. (14) John Sales, now National Trust Gardens Adviser, reports on 6th October 1977: Conservatory. I gather that this has to be demolished and I am sure that we should retain a building at this point (a) as a focal point on the axis, about which this garden is arranged; (b) as a shelter, which is to necessary in this garden. No doubt we could find a new structure of appropriate style designed for the site and believe that there is a scheme on the files; alternately Mr Feesy, who designed the pavilion at Knightshayes, would oblige for a fee. On the whole I would favour restoration of the central bit only if the conservatory is to be planted with permanent subjects and furnished with seats for visitors to use. (15). Trinick, who had seen off pleas for retention of the Victorian vinery wings some years previously, is impatient in his response to J. Brunner dated 9th December 1977: I shall be writing to John Sales soon with my suggestions for a replacement of the conservatory with an open-fronted arbour. His (John Sales) suggestion of repairing the central portion only of the conservatory would be pretty expensive, both in first cost and subsequent maintenance. I don't like it because the proportions and scale of the building are so poor in relation to this garden. They were all right when there were long glass houses on either side but are now not right. (16). Trinick then secures a sketch proposal for his arbour or 'bower' from John Dyke (17) Fig 06 Sketch proposal for a 'bower' to replace the Conservatory, by John Dyke, 1978 To counter Trinick's objections to the conservatory on grounds of cost, his colleagues seek to raise funds through sponsorship. With breath-taking audacity the then Assistant Agent M.D. Cook,ARICS (David Bett having moved to Mercia Region), writes on 15 th September 1978 to Crittalls (Windows) Ltd and to Boulton & Paul the distinguished old Norwich firm of glasshouse builders and horticultural equipment suppliers: I am writing to you because as specialists in widow construction it could just be that your firm may be interested in financing the reconstruction of this building (the Arlington Conservatory) as a prestige project. Detailed plans of any such reconstruction have not yet been drawn up, but the Trust has estimated that the cost of providing the replica would be in the region of £12,000 - £15,000. (18) John Sales, the Gardens Adviser, on 15 th November 1978 remains determined to retain a conservatory albeit recognising that, with lack of maintenance, the fate of the existing one was sealed. Copying in Dudley Dodd, the Historic Buildings Secretary, he writes: 1.1 The present structure is deemed unsafe and clearly must be demolished; this work to be started next week. It is essential that an accurate measured drawing be made before demolition. 1.2 We discussed the future of the site and the Historic Buildings Representative and Gardens Advisers are agreed that a building of similar shape and size to the present glasshouse is required to give a focal point to the main axis of the garden. 1.3 Of several alternatives we would favour rebuilding a conservatory for the following reasons: ◦ It would be historically correct and true to the tradition and function of the garden ◦ A glasshouse of some kind is required for pot plants and cut flowers for the house and for some propagation and overwintering of plants ◦ It would provide shelter for visitors in the (frequent) event of rain ◦ It would be an attractive feature for visitors in its own right 1.4 I recommend that every effort should be made to raise the necessary funds by special appeal; by approach to benefactors; by appeals to members' centres; through special projects (windfall) funds. I will raise the matter at the next Gardens Panel if necessary (?) 1.5 Meanwhile we should find out what sum would be required. Messrs Richardson (of Darlington), Amdega Ltd, 35 Godalming Avenue, Wallington, Surrey, SM6 8NP, manufacture cedarwood glasshouses of similar style and it may well be possible to have one made up from their standard sections. Would you please investigate. Unless and “off the peg” design is available and significantly cheaper the least controversial approach would be to rebuild on exactly the same pattern. This could be done gradually over a period using the funds available this year to make a start.(19). But in Michael Trinick's corner the Regional Chairman, citing Lord Morley's views, weighs in with an absolutely priceless memorandum which demonstrates no understanding of the place whatever. It is dated 10th January 1979: The Lord Morley has given me his views about replacing the glasshouse at Arlington and it so happens that I entirely agree with what he says. Neither of us thinks that £10,000 on a new glasshouse could possibly be justified unless a donor gave the money for that specific purpose, which is unlikely. Apart from the huge cost what would we do with a glasshouse when we had it? I suppose we could put exotic and tender plants in it, but I doubt whether many people would look at them or would sit in it. We think that the old wall should be levelled off and the white plaster work removed from the position of the old glasshouse. Then we should plant creepers up the wall... (20). Sensing danger, but clearly unaware of the details, Sales follows up his recommendations with a memo to Cook dated 15th January 1979: What please was the outcome of your decision to take down the conservatory? Has it now gone and if so did you make a measured drawing? I would be glad to know of your plans for its replacement, assuming that you will be rebuilding substantially on the same pattern. (21). Cook, replies on 18th January 1979: Yes the conservatory has gone. We carried out an extremely detailed survey and our Architectural Draughtsman, Mr Board, has done an excellent scale drawing. Then, with a certain understatement, goes on to report: I raised the matter of a replacement at the last Regional Committee meeting and as this was held at Arlington itself we all inspected the site. I must say that the feeling of the Committee seemed to be that a replacement was an expensive luxury and that perhaps a pergola or something similar would be the better proposition.(22). 3.5 Commissioning the Conservatory replacement Nevertheless Sales prevails and progress is made towards the replacement of the conservatory likefor-like. The buildings department at Killerton sends copies of Mr Board's drawing off to Amdega Ltd10 for an estimate of costs. Amdega's reply, dated 24 th July 1979 states that a reproduction could be made in Western Red Cedar and erected (not including builders work on the base, floor, drains etc) for £15,100. Amdega hears no more from the Trust on the subject until 6th May 1980 when its draughtsman, Mr Board writes to invite Amdega to consider revising its quotation. This they do and in a reply dated 15th May, the price has increased to £17,940 (23 & 25). On the 6th June 1980 Cook reports to Sales: We have in this year's estimate plus a top up from the Arlington Fund enough money to build the replacement Conservatory. I have obtained a quotation from the specialists in Darlington which comes to about £18,000 and there will be additional incidentals on top of this (26). By this time Cook is concerned about the details, not the principle and is also consulting Dudley Dodd: The particular question which vexes us at the moment is whether or not to go for treated cedar wood and paint at some later date if we do not like the finish or to consider anodised aluminium or even plastic coated timber. On Sales behalf, Jim Marshall, the Assistant Gardens Adviser, replies: 10 Amdega Ltd is the new name of the firm which manufactured Richardsons greenhouses from about 1870, based in Darlington I am sure it would be wrong to have the conservatory built of any other material than wood. Ideally teak would be the best material, but Western Red Cedar is certainly the next best alternative. With my experience with such glasshouses and conservatories Western Red Cedar should only be oil treated and it should never be painted (27). On the 17th June 1980 Dudley Dodd replies and his memorandum is so cardinal to the project it is quoted in full: As agreed last week I consulted the Architectural Panel about our plan for reinstating the Conservatory. Certainly they welcomed the decision to put back the Conservatory and agree that it should be in softwood, painted white, not cedar. The Panel agree that the design looks truncated and that some device – such as a pergola – is needed to express the missing wing(s). Rather than tacking on a pergola to the existing design, the Panel has asked us to consult a sensitive architect with a view to redesigning the Conservatory, in the Victorian idiom and at no greater cost than the estimate already obtained. Bobby Gore and I suggest we ask Philip Jebb to redesign the Conservatory. He is a very sensible architect and, as you probably know, travels quite often to Lundy to work for the Landmark Trust. He is also a member of the Architectural Panel. Like all good architects he is busy and I suspect that he would be only too grateful for help from Michael Board both in surveying the site and in supervising the work. With your agreement, I would like to invite Philip Jebb to undertake the project and arrange a preliminary site meeting with him. Meanwhile, could you please consult John Sales asking him to specify what will be grown in the Conservatory and the requirements for heating, staff etc. (28) John Sales is triumphant and on 25th June 1980 writes to Cook with a copy to Dodd: I am of course delighted to hear that it seems possible that a conservatory will be rebuilt on the site of the old. I agree with you about the desirability of the building being painted white, if possible, but this will increase maintenance costs considerably and I wonder whether it would be better to try the building in its 'raw' state as pressure-treated timber first to see what it looks like? (30) Sales continues by expressing the view that because the conservatory was only the central portion of a long range it looks 'oddly truncated'. Then, disappointingly from an archaeological point of view, he advocates omitting the lower diagonals from the ends necessary in the event of recovery of the flanking vineries. This resulted in an amendment of Michael Board's drawings which, when built, meant that the Trust would be stuck with a stand alone conservatory where it would be impossible to recover the attached vineries later if desired. 3.6 Appointment of Phip Jebb as architect Philip Jebb is commissioned to take on the project and on 2 nd August 1980 issues new drawings which amend those from Michael Board also in that the E & W roof ends become hipped, a change praised by Dudley Dodd in his letter to him dated 18th August 1980. Dodd continues by inviting Jebb to 'proceed with the specifications and obtain a price from Richardsons' , providing for heating and automatic ventilation. Bobby Gore asks that the plans are submitted to the Architectural Panel for its September or October meetings (31). In the event Jebb goes back to Amdega Ltd of Darlington (who previously quoted £15,100 and £17,940). On 28th October 1980 Amdega provides their third quote, this on the basis of the amended scheme, for £19,470. Theirs is to be a Western Red Cedar glasshouse not including 'builders work'. Jebb proposes getting a second quote from Room Outside Ltd, but they need a local contractor to erect (33). Dudley Dodd, in a letter to Jebb dated 30 th October, 1980, confirms the building to be painted white rather 'than left in raw cedar'. Dodd goes onto say: It was entirely to be expected that the Architectural Panel should have been so very complimentary about your design and we all look forward to the finished Conservatory! (34). This re-energises the project and already from the 27 th June 1980 we have a quotation provided to Michael Board from the Barnstaple firm of R.J. Tamlyn & Son for the builders work (rebuilding the dwarf walls etc) of £2,107.50 plus VAT. Michael Cook, the Trust's local land agent, writes to Jebb on 5th November 1980 to confirm that, so far as the builders work and erection is concerned, R.J. Tamlyn & Son of Barnstaple are recommended (29 & 35). Mr G.M. Burton for Room Outside Ltd submit their quote of £6,432.00 to Jebb on 21 st January 1981 after a delay occasioned by the time taken to obtain quotes from prospective subcontractors. Room Outside allow for the use of 7mm Georgian wired glass, for 3 coats of Sikkens Rubol DSA, for lead flashing against the back wall and for pole-operated ventilation gear. Room Outside's quotation includes a drawing of the timber sections they would use (36, 37 & 38): Fig 07: Timber sections proposed by Room Outside Ltd for new Conservatory January, 1981 Almost a year passes while the Specification, which is dated December 1981, is awaited from Philip Jebb, a delay in part occasioned by his breaking his collar bone in September 1981 although there was a period early in the year when chasing letters and phone calls to Jebb are not answered. When the specification arrives it is found to cover the building and erection work of the main site contractor but not the specialist glasshouse construction itself which is very much left to Room Outside Ltd to provide. To move things on contract administration is let by Jebb to Mr T. G. Williams of Bare, Leaning and Bare, a firm of quantity surveyors based in Swindon. On 4 th January, 1982 Williams writes to Michael Board at Killerton to canvass for names and addresses of firms he would like to invite to tender suggesting Woodman & Son of Exeter and a Mr Gist of Torrington. It seems that Board replies to recommend Tamlyns again and R Harris & Son Ltd, also of Barnstaple for the builders work and Amdega and Room Outside for the glasshouse construction. Michael Cook, the land agent, also replies (6th Jan 1982), in his case to ask Williams to seek Building Regulations and, if necessary Planning Permission from the local council (39 & 40). For the supply of the glasshouse Theo Williams on 15 th January 1982 writes to Room Outside Ltd, apparently omitting Amdega Ltd who have already provided three quotes in succession, the latest being: Amdega Ltd £19,470 (including erection) Room Outside :Ltd £9,922 (excluding erection), whose appointment is assumed For builders work and erection including for submission for Building Regulation approval, Williams on 18th January 1982 writes to Harris, Woodman and Tamlyn for prices. Quotes come in as follows: Harris & Son, Barnstaple £16,248 Woodman & Son, Exeter £21,633 R.J. Tamlyn & Son £16,873 Room Outside Ltd reply to Philip Jebb on 20th February 1982 confirming their revised quotation at £9,922 but excluding lead coverings to hips and valleys and allowing for 4mm float glass rather than Georgian wired glass offered previously. Room Outside query Jebb's provision for sixteen roof vents and offer ten, omitting the three each on the side hips. The firm points out, perfectly reasonably, that these six vents, as arranged by Jebb, cannot be made weather tight, since they rely on a lead flashing from the wall which in their case cannot be provided. It is an indictment of Jebb's design which nobody at the Trust had spotted. Room Outside Ltd also spot that the long glazing bars need more support than provided by Jebb, and accordingly allow an extra purlin (41). Williams replies on 22nd February 1982 and confirms that Room Outside Ltd's new quotation has been accepted. The firm is appointed with Tamlyns as the main contractor. The total cost of the work to both firms including fees and ironmongery is budgeted to be in the region of £20,466 (42). 3.7 Contractual delays and challenging costs The first whiff of difficulties with the contract is sensed in correspondence between Michael Cook, the Trust agent, and Theo Williams, the contract administrator from Bare, Leaning and Bare, over the question of professional fees. Williams letter of 11th March 1982 explains why these are likely to be higher for Jebb than the usual scale. A figure of £2,450 is quoted which represents about 12% of the project cost: The detailing of the Glasshouse has proved a major headache as, with a purpose made object the total cost is excessive and, with adapting a series of standard modules to reduce the cost, problem after problem has been thrown up and it takes an inordinate amount of time to convince the producers of the standard bits and pieces that they (a) can and should vary their designs and (b) that the costs need not mount excessively because they have departed from their usual patterns (43). Curiously Williams then claims rather more credit for his firm than he is probably strictly entitled to: My firm's part has been to take Mr Jebb's design and write up the schedules of work, hopefully covering all the details so that when the job is started nothing is found wanting in the Specifications prepared to cover both the Specialist and the Site Contractor. Mr Jebb from now on will deal with the periodic visits to check that the design and appearance is as he wants. Subsequently we then agree the money side with both Specialist and Site Contractor and the proof of the adequacy of the Specifications will only then be seen (44). Soon afterwards, in early June, the Trust receives from Jebb an invoice for interim payment for professional services covering the period January 1981 - 31st May 1982 amounting to £1498.78. Of this Jebb is claiming £507.50 and Williams £774.76, the remainder being expenses and VAT (47). Michael Cook scribbles a note on the invoice to say pay reluctantly: 2 years for Jebb to sort out this scheme – 2 months to send in an interim account. Although these charges seem very high I suppose they are right but I'm afraid that we have received a sub-professional performance from Jebb. Had I been allowed to use Michael Board as I originally intended you could have deducted £1000 from this (47). So far as the actual building work is concerned G.R. Tamlyn begins with a letter of 26 th March 1982 placing the order for the glasshouse with Room Outside Ltd and requesting detailed working drawings including construction details at valley gutters in time for a meeting with Michael Board on 7th April. In particular Tamlyn is: concerned that because the timber will be treated under pressure with an aqueous solution, the paint coatings may fail very quickly due to the fact that the moisture content of the timber was too high prior to priming. I am at this time writing to Mr Jebb for his advice on this subject. Please await his advice before priming your frames (45). Also of 26th March 1982 Williams writes to Jebb and in response Jebb on 2 nd April furnishes Michael Cook at the Trust with copies of drawings showing typical hips, ridges and valleys. Jebb confirms that the new 9” brick base will be built on the line of the existing, correcting if the existing is found to be out of square or unsymmetrical. It is clear that progress has been slow for foundation digging did not occur until August. There are concerns about day-to-day client supervision, never mind overall project management. Theo Williams, writing on 19th August to Michael Board, by now Assistant Building Manager, observes tactfully that the contractor has not had to take the foundations as deep as specified on page C/2 Item 7 and that a saving could be secured if the matter is pointed out directly (50). It seems that Jebb's £1498.78 invoice submitted in early June had not been paid which caused him to write to Michael Cook on the subject on 24 th September. Regional Director Peter Broomhead replies on 14th October 1982 to say: As far as we can trace, we have not received your June invoice and I am afraid that I am unhappy with the account submitted because from the reports I have received to date we have not received the standard of professional service which might have been hoped for and expected and there do appear to have been lapses of control which are resulting in higher expenditure (52). dignified reply, dated 21st October shows him to have been a man of unusual integrity and humility. 'I am somewhat shattered to receive your letter' he begins with understatement. 'I do wish I had known of this earlier, in which case I would not have dreamt of sending an account'. In similar vein in the next paragraph he continues: ' I only feel a sense of disappointment with myself at having allowed such a situation to arise without being aware of it'. Jebb goes on by holding up a candid hand: Jebb's I am aware, most acutely, of having issued a setting out drawing on which there was a dimensional error of 2' 0” which did not come to light for more than three months, but it has never been my intention that the Trust should bear ant additional cost arising from this error and I assume your letter expresses a more deep-seated dissatisfaction. It is easier for me to have views on what best be done now to repair the matter. They are these: 1. You should disregard my account dated 17 th June but settle that part of it which is attributable to Bare, Leaning and Bare direct, unless your dissatisfaction extends to them as well, in which case you should be more specific. 2. I should withdraw, keeping in touch only in so far as it is necessary to ensure that any additional cost arising from the dimensional error on the setting out drawing is not passed on to the Trust and, of course, attending to or clearing up any other matter which your phrase “lapses of control which are resulting in higher expenditure” may imply. 3. That you should accept my abject apologies for having been such a disappointment. As I was originally asked to help with the Glass House by Dudley Dodd and because my failure to do so leaves me with a very nasty sense of having let him down I am sending him a copy of your letter of 14 th October and this letter (53). Peter Broomhead replies at length and his letter, which is openly copied to Martin Drury and Dudley Dodd and is dated 28th October, 1982 is paraphrased here. His complaints of Philip Jebb are threefold: 1. Delay. It is observed that since Jebb was asked to help in June 1980 there have been long periods of little progress with work not actually starting on site until August 1982 2. Escalating cost. The Regional Director also observes that the original quotes from Room Outside and Tamlyn were £6,432 and £2,107 respectively. Although he recognises that some of the increase in cost is due to changes in specification, he attributes most of it to delay concluding that 'the whole job now looks to be heading for somewhere between £23,000 and £25,000. 3. Professional negligence. Lastly the matter of the 2ft dimensional error is touched upon and there is a reference to lack of provision for fixing details of the galvanised anchors to the vertical posts. Broomhead concludes his letter by accepting Jebb's resignation and enclosing a cheque in settlement of Bare, Leaning and Bare's part of the fee account. The gesture can scarcely be regarded as conciliatory, but Broomhead was able to say at the end of his letter: I am very sorry that you were the architect who has been the object of my dissatisfaction. I am afraid that at present I have one or two other disappointing performances by architects in this Region and I regret that you may have got more of the backlash than you necessarily deserve (56 & 57). Jebb is both beaten and bowed, but there is a curious end to his part in the story for, due to a clerical error at the Trust, the cheque Broomhead enclosed had been made out for the full £1498.78 amount of the June invoice, not just the Bare, Leaning and Bare part of it. So, with a short hand-written reply written on 4th November 1982 Jebb nobly writes out a cheque for his part of the invoice (£607.81) and sends it to the Trust. His estate remains unremunerated for his work at Arlington, an inglorious incident in an otherwise celebrated career (58). 3.8 Contract progress post Jebb By July 1982 the Trust had already been thinking about publicity and who might officially open the new Conservatory. The garden writer Don Hoyle is approached about the latter in a letter from the Killerton office dated 19th July 1982 with a view to reopening in March or early April 1983. Pinned to site from July 1982 onwards is a notice which informs the visitor: You will see from the many photographs on display that the centre portion only is being replaced and, in fact, the building was very much larger in the early part of this century. We hope that this has been a faithful copy of the original in everything but size and will create much interest. The work has to be carried out in the summer months because of the notoriously bad weather in the winter at Arlington whan an intricate structure like this could easily fall foul of a North Devon gale (48 & 49). By October 1982 the Trust had effectively assumed direct day-to-day management of the project and Theo Williams at Bare, Leaning & Bare a better grasp of the costs. It seems not a moment too soon because it is clear from correspondence between the contractors that further shortcomings had been identified in the design. Burton of Room Outside Ltd writes to Tamlyn on 20 September and 8th October 1982. Tamlyn's response to Theo Williams dated 14th October identifies a costly list of 'extras' including £338 for “add on” sections to widen the purlins and £210 for timber to form roll joint on hips. Williams identifies cost variations to date: 1) Saving on foundations 2) Extra £1000 found by NT to supply 2nd hand bricks 3) Dimensional error 2 ft under meant that additional lengths of hip timbers were required to be manufactured to suit span 4) Purlins had to be added to satisfy structural engineer 5) No provision for fixing details to vertical posts entailed additional 8 No. galvanised wall anchors 6) Additional glazing 7) Additional timber for hips and leadwork 8) Nothing allowed for door & window furniture (ironmongery or brassware) We find Michael Board is dealing directly with the contractor; for example on the 8 th October 1982 in the matter of the choice of fittings: Mr Hugh Meller, the Trust's Historic Buildings Adviser, has confirmed his intention of using the brassware, all as catalogue upon which your quotation is based, in the sum of £550.54 + 5% profit = £578.11 (51). Michael Cook summons a meeting with Jim Marshall and Hugh Meller on 16 th November, summarised in a note of 18th November 1982. There is a sense of urgency: I feel it necessary to outline the areas of responsibility for getting work done in time for the opening next season. It is imperative that we press on as fast as possible with the work as there is a considerable amount to do once the structure has been completed. Can we please take action on the following points, set out below: 1. Can J. Brunner please obtain as soon as possible a report and estimate for the provision of electricity, both to the Conservatory and to Paul Amos' workshop 2. Michael Board to earmark the necessary days work for Tommy and Brian for the building work to the paths inside the Conservatory, the approach path across the lawn to the Conservatory and all the ancillary work in connection with the dwarf walls adjoining the building. 3. P. Amos to concentrate in liaison with Tommy and Brian on the formation of the borders and the wiring of the rear wall for climbing plants 4. M.D. Cook to chase the quantity surveyor regarding the completion of the building by Christmas which is imperative so that the ancillary work can proceed. 5. Jim Marshall undertook to supply most of the plants to be planted once the foregoing work has been completed. Jim Marshall's report will follow as soon as possible but we should start to at least plan the work now so that there is no hold up later on. It is hoped that Jim Marshall's report will also include an outline plan of the conservatory for us to work to (60). To move matters on in an effort to secure completion by Christmas Michael Cook has Theo Williams at Bare, Leaning and Bare chase Messrs R.J. Tamlyn to obtain missing glazing bars from Room Outside Ltd and henceforward prepare for the glazing sub-contractor appointed by Mr Burton. On 29th November 1982 Williams writes: Would you please as a matter of urgency look at the programming of this work. The operation as we see it falling into three distinct operations viz:(a) Fixing glazing bars, timber packings and hip rolls (b) Glazing windows, doors and roof (c) Fixing leadwork to valleys and hips Mr Burton has appointed a firm of glaziers in Barnstaple to carry out the glazing – Messrs Andre Wartha. When considering your programme please look very carefully at the method and difficulty which may occur in fixing the leadwork to hips and valleys; it is clearly a very delicate operation. Presumably you will also consider the possibility of getting the glaziers to commence the window and door glazing a day or so before the roof glazing commences (63). 3.9 Horticultural preparations In his report, dated 22nd November, 1982, Marshall sets out what he requires for the Conservatory interior once the glazing has been completed: 1. General principles: only sufficient heat to be provided to raise the temperature above freezing; a continuous border to be made outlining the shape of the conservatory; climbers to be trained on the back wall and seats to be placed in front of the back wall. I will supply a detail plan showing position of borders and paths. 2. Heating: Ideally electricity should be installed; obtain 2 Camplex New Thermo Matic Fan Heaters. If this proves impossible, obtain propane type gas heaters i.e. 2 Tempest TX3000. 3. Seats: Obtain 2 three-seater metal seats and 1 two-seater metal seat; all to be of the Arlington design and painted white 4. Paths and steps: All to be slate and the borders edged with rope tiles as per present type 5. Borders: Top soil to be added, double dug and well-rotted farmyard manure or garden compost added. Prior to planting apply 'Growmore' at the rate of 50g per m² (1½ oz per yd²). 6. Back wall: To be wired as per following specification – Use 1.6mm galvanised wire (16s.w.g) and old type galvanised vine eyes which are obtainable from the Building Manager at Lanhydrock. All wires spaced 450mm (18”) apart and vine eyes 2.7m (9 ft) apart approximately 7. Planting: Climbers to be trained on the back wall, the central support pole and the four principal rafters. The majority of the border planting not to grow above the base wall. I will report separately on the plants required. 8. Two central gravel areas: Grade and colour to be the same as that used on the plants either side of the Conservatory. 9. Top ventilators: These to be automatic. You will have to check with “Room Outside” that these automatic ventilators can be fitted to their conservatories. 10. Gutters: Iron guttering and downpipes will be necessary; all water being collected in a tank on the nursery side of the back wall. The report continues with a few recommendations for the area immediately outside the Conservatory:     Take down the single brick return walls either side of the Conservatory and erect a pier at each end of the main walls. Make two new gravel paths to the side doors of the Conservatory Train the Loniceras either side of the Conservatory to cover the new rendering on the back wall We pegged out the width of the new front path from the middle steps to the Conservatory. It is essential to use the same surface material as the present path (61 & 62). Jim Marshall, the Gardens Adviser writes a memorandum on 6 th January 1983 with a plan identifying the plants he wishes to see in the Conservatory. For the five island borders against the back wall: Canna iridiflora, a high latitude S American plant with pendulous pink flowers Lippia citriodora, lemon verbena, also from S America Coronilla glauca, fragrant yellow flowering plant, native of Spain Salvia Grahamii, Graham's Sage, a crimson flowering shrub from Arizona and Mexico Teucrium fruticans, a type of mint called Shrubby Germander from the Western Mediterranean Hedychium gardnerianum, the ginger lily, with fragrant pale yellow flowers, a native of the Himalayas For the centre bed Marshall required: Daphne odora aureomarginata, the gold-edged winter daphne, a leathery-leaved native of China and Japan Rhododendron fragrantissimum, scented with pinkish-white flowers, a native of N India. Rhododendron daviesii, a fragrant Ghent azalea with yellow-flashed cream flowers Dierama pulcherrimum, known as Angel's fishing rod or wandflower, a native of S Africa For the borders round the sides, the plan was to have: Rose 'Cécile Brunner', a tea rose with exquisite little blush-pink scented flowers from China Geranium anemonifolium, a purple-flowered geranium originally from Madiera Cyrtomium fulcatum, a type of fern, native of South Korea Polystichum setiferum 'Herrenhausen', soft shield shuttlecock fern, a native evergreen Polystichum setiferum divisilobum, a woodland fern similar to above Woodwardia radicans, another shade-loving evergreen fern Calceolaria integrifolia, shrub with clustered yellow flowers, native of Chile Saxifraga fortunei Rubrifolia, a white flowered woodland plant native of China Agapanthus Headbourne hybrids, clump-forming perrenial with umbrils of blue flowers Zantedeschia, Arum lilly, a poisonous rhizomal plant from South Africa As climbers for the centre pole and cross members: Lonicera x tellmanniana, a scentless honeysuckle hybrid Passiflora incarnata 'purple passion', herbaceous vine, a native of southern USA Solanum jasminoides 'Album', vigorous evergreen climber with starry white flowers, native of Brazil Jasminium polyanthemum, vigorous climber with pinkish five-petalled flowers, native of China Lapageria rosea, evergreen climber, a forest-bellflower, native of Chile (64). Marshall follows this up with a memorandum on 14 th March 1983 consolidating this list of plants with the following additions to go against the walls: Abutilon megapotanicum 'Kentish Bell', a woody flowering shrub maple from South America Camellia japonica 'Rubescens Major', dark green shrub with pink flowers, from South Korea & Japan Acacia baileyana, shrub with yellow flowers, from New South Wales, Australia Fuchsia 'Ballet Girl', upright shrub with pink & mauve flowers, originally from central America Cestrum 'Newelii', vigorous evergreen shrub with bright red flowers from central & S America Also for the borders round the sides he wanted various forms of: Liriope, low herbaceous flowering plants known as monkey grass, from east Asia (67 & 68). Fig 08: Ground Plan of the Conservatory on which Jim Marshall has identified the plants he wishes to see 3.10 Final fitting and commissioning Work on fitting out the Conservatory continues into 1983. A note on file dated 19 th December 1982 records that seats were quoted to Michael Board at £85 or £120 each, copying existing ones at Arlington. For the path edging Hugh Meller, the Historic Buildings Representative, chooses Haddonstone's rope-moulded concrete edging at £4.75 per 18” length. Meller is consulted about the question of the finial and in a memo dated 2 nd February 1983, enthusiastically endorses the idea of the heron which he said is a 'splendid idea' and 'should be painted dark grey'. For the interior path and steps would the Trust use slate from the Mill Hill Quarry at Tavistock or Delabole from Camelford in Cornwall? In the event Delabole is ordered and invoiced on 23 rd February 1983 at a cost of £1143 plus VAT (65). The opening for the new conservatory was arranged for 7th April 1983 and on 28th March invitations went out from the Killerton office. Don Hoyle had kindly agreed to perform the ceremony at 11.30 am. The letter from Anthony Adam, the Public Affairs Manager, was somewhat inaccurate on the historical details but nobody seemed to mind that the new design was not in fact at all like the original or that it was one of the designs presented to the Chichesters in 1849 as claimed (68). Fig 09: Garden writer Don Hoyle opening the 1982 Conservatory Later in 1983 the matter of the electricity supply was revisited. South-West Electricity Board quoted for this on the 17th October 1983 (70). In all this was likely to add another £1000 and it was put to the local National Trust Centre for fund-raising. Michael Cook, on the 9 th November 1983, defends to Peter Broomhead the 'high cost' thus: The installation of electricity is always expensive and in this particular instance because it comes from a transformer in the walled garden we have to use SWEB. Furthermore as this, in the long term, will provide a cheaper and less polluting form of frost control (it is not heating in the accepted sense of the word) for the conservatory and we shall also have the beneficial spin off of being able to install electricity in Paul Amos' shed/workshop for the first time which means that he will be able to carry out all his maintenance work in one place without having to run any cables or use any other ancillary forms of power. It is likely that this cost will be in total just under £900 which I believe represents a good long-term investment (71). 3.11 Meeting costs However the formal opening (or even the question of electricity) was not to be the conclusion of the protracted saga of the Conservatory, as there were some outstanding claims. Williams from Bare, Leaning and Bare wrote to Michael Cook on the 8 th July 1983 with their 'Valuation No 2 in the sum of £9,656'. The letter continues: We have not included in the Valuation the sum of £1,728 which Room Outside are claiming as extras. Their account to Tamlyn dated 23rd March 1983 was presented to us by Tamlyn at a meeting on the 1 st July – noone had: (a) bothered to send us a copy or (b) notify either the Trust or ourselves of these extras as and when they became apparent. As far as the final account is concerned the best position is £20,495 if the extras are not allowed or £22,223.47 if they prove to be acceptable (69). The question of the 'extras' rumbles on for months and eventually solicitors become involved. George Ide, Phillips & Co acting for Room Outside Ltd write on 6 th January 1984 with a final demand 'before issue of court proceedings'. The bill for 'extras' including timber purlins following consulting engineers' advice and metal angle brackets at junctions between purlins and plates now amounts to £1,185 claimed. In the opposing corner solicitors Messrs Pitts Tucker of Barnstaple are acting for Bare, Leaning and Bare for the Trust. Theo Williams' letter of 16 th January 1984 to the latter firm make clear Bare, Leaning and Bare can only act for the Trust as intermediary with the contractor and not on behalf of one contractor or another. It seems the Room Outside claim is rejected (72). Fig 10: Philip Jebb's plan proposal for the 1982 Conservatory (original at 1:50 scale) (top) Fig 11: Philip Jebb's front elevation proposal for the 1982 Conservatory (original at 1:50 scale) (bottom) Fig 12: Selection of setting out details from Philip Jebbs' office 1982 Fig 13: Archaeological record plan of the 1982 conservatory in 2012 before demolition (top) Fig 14: Interior view of the 1982 Conservatory shortly after completion (bottom) Fig 15: Views around the W end of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 16: Views of the projecting front porch of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 17: Views around the E end of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 18: Views of interior and floor of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 19: Views of interior and roof structure of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Fig 20: Views of various details: window catch and stay, automatic vent mechanism, rope edged slate path, steel strapping to roof timbers and heron finial of the Jebb 1982 Conservatory shortly before demolition, taken on 15th June 2012 Chapter 4 Development of the other Glasshouses at Arlington11 4.1 Walled garden glasshouse range of c.1820s. 4.1.1. Introduction We know little of the Georgian glasshouse which originally occupied the site. Construction of the present house in the 1820-3 period provides a likely context for their genesis and they had certainly arrived by 1842 when they are shown on the Tithe map. Fig 21 View across the garden terraces towards the W end of the glasshouse range with Compartment 1 hidden behind the twin Araucaria araucana, c. 1890 11 This section builds upon the brief overview in Conservation Statement of the Garden by Jonathon Lovie (2009) Fig 22 Long view across the garden terraces towards the E end of the glasshouse range with Compartment 5 hidden behind the same Araucaria araucana, c. 1890 4.1.2 General description The early glasshouses at Arlington follow the classic Georgian form and are built in timber leant against a hot wall. It is clear from early photography (fig ) that they were vented by sliding sashes in the roof and pivoted casements in the sides. The sashes themselves are alternate, that is every other one is fixed in the upper half and all are fixed in the lower. They were probably standard English lights measuring 8 ft by 3' 6” suitable for use in cold frames as well as in the glasshouses. Each are divided by five glazing bars to six panes across; each pane (or 'quarry') measuring 6 inches wide and around 10 inches long and lapped over one another down the length by about ¼ of an inch. The glass will have been hand-made probably by the cylinder rather than the Crown process to a thickness of around 1/16th of an inch (about 2mm). This is half the thickness of modern horticultural glass and is a result of the fact glass at the time was taxed by weight. Fig 23: J C Loudon’s illustration (1824) of a range of hothouses 12 constructed at Skrivelsby, Lincolnshire is a close parallel to the form of the early 19th century glasshouse range at Arlington Fig 24: East vinery range showing centre pivoted casements, late 19 th cent photo 12 - J C Loudon, Encyclopaedia of Gardening (1824, ed of 1830), pl 451 4.1.3 Plan and layout The glasshouse in plan is divided into five sections which were probably compartmented from one another by internal glazed cross partitions, each with a door. Working from left to right, the five sections have a bay rhythm of: 7 . 13 . 11 . 13 . 7, as defined by the number of lights to each compartment. In terms of lengths, measured in feet to the nearest inch or two, respectively: 27 . 50 . 42 . 50 . 27, giving a glasshouse range of almost exactly 196 feet or 59.74 metres. In terms of internal span, compartments 2, 3 & 4 measure 13' 6” while 1 & 5 measure at 12' (for the full width add 9” for the thickness of the wall). 4.1.4 The Compartments The central compartment, occupied by the present conservatory, can be seen clearly in surviving 19th century photographs. Of its 11 bays of lights, six in the roof had moveable sashes. The front wall seems to have been glass almost to the ground so it clearly did not share the same dwarf wall as the ranges either side. The three bays of lights at the centre of the front were occupied by a pair of glazed doors with narrow fixed 2-pane lights either side. The glass in these doors has been treated differently, ie with horizontal glazing bars rather than laps and the inference is they may have been introduced later, perhaps to make more of a show house of the central compartment in anticipation of its eventual metamorphosis into a conservatory. The 13 bay ranges flanking the high central section were identical. They stood on on the extant dwarf brick walls and comprised a low glazed front resting on the wall top. This had alternate centre-pivot opening casements. In line with the front wall openers are the sliding roof sashes, six to each of the 13 bay sections. Such sashes will have been operated manually and individually on counterbalancing weights mounted against the rear wall, where some evidence survives, albeit too obscured by climbing plants at the time of survey to interpret properly. Of the 7-bay compartments at either end, very little can be deduced for they remained stubbornly obscured in the early photographs. We know their length and width and that they had existed in 1842 (Tithe Map) and survived at least through the date of the 1 st edition 1:2,560 Ordnance Survey map of 1886 on which they are clearly marked. The photography seems to suggest that its glazing came almost to the ground and that, like the centre compartment, they had little or no dwarf wall 4.1.5 Hot Wall. Of the earliest build on the site only the hot wall survives to its full height; the hot wall being the back wall which was heated by flues running up inside from one or more furnaces situated on its north side. The furnace positions have been filled and are not apparent; neither are the line of the flue runs, due to growth of obscuring climbing vegetation. Only where some damage had recently occurred to the masonry of the wall to right of the Conservatory and where occupation by bees was apparent at the time of the survey, could the internal flue be detected. Such hot wall flues normally comprise flues arranged in panels running horizontally to right and left, a small ascent at each turn until finally venting through a short chimney or chimneys (here missing) positioned atop the wall. Such walls are generally, but not invariably associated with early glasshouses and this one may originally have been constructed simply to keep the frost off wall-trained fruit. 4.1.6 Function Such a range as Arlington's in the late Georgian and Victorian periods will have served a range of functions and it is likely the compartments could be heated to varying temperatures according to need and weather. Likewise their individually operated roof sashes and wall casements will have enabled garden staff to manage ventilation very closely. The central section, Compartment 3 probably served as something of a show house from the start. Originally perhaps for oranges and lemons, later as the fashion changed in the Victorian period for such as camellias. Once it had its large hinged glazed panels for access from the front, it would have been possible to bring potted plants out onto the terrace in summer. The back wall may have been used for climbers or wall trained fruit such as figs or nectarines. So the centre compartment was clearly treated as a display house by the time of the late 19 th century photographs and probably contained a mixture of plants kept for their colour or scent in summer with potted citrus brought in here for protection in winter. Fig 25: Late 19th century view of the interior of the glasshouse, probably one of the flanking compartments looking east. This appears to show vines against the glass as one would expect, but palms beneath and a rear wall trellised for climbers The flanking compartments 2 and 4 served as vineries. The evidence for the growing of grapes is to be found in the dwarf walls which are perforated for vines. These were grown outside and trained in on rods or later wires to spread and fruit under the glass. Commonly Georgian vineries doubled as pineries, that is they grew pineapples in pits under the vines. The archaeological evidence of this at Arlington is wanting and in any case the garden had its own separate pinery to the north of the walled garden. Fig. 26: Apertures in the dwarf wall for admitting vines Bays 1 and 5 are likely to have been considered too narrow for this use and may have been used simply for peaches or apricots on cage frames. The photographs seem to show that not only were these compartments narrower, they were lower too. It is possible therefore that their original floor levels were also lower and if below ground level may have acted as pit houses for such as melons or cucumbers. 4.2 The glasshouse at Arlington Rectory, c.1845 4.2.1 For comparison it is instructive to examine the evidence of the glasshouse which graced the adjacent rectory. This was supplied to the Rev. James H. J. Chichester, younger son of Colonel Chichester and the uncle of Sir Bruce Chichester. Arlington Rectory (now known as the Glebe House) is the beautifully proportioned residence sited between Arlington's walled garden and St James' Church. It had been built by the Rev Chichester soon after he assumed the living of Arlington and Loxore in 1825. Like Arlington Court, the architect would seem to be Thomas Lee of Barnstaple (1794-1834), perhaps the most influenced of the pupils of Sir John Soane. 4.2.2 A conservatory had been added to the east side of the Rectory a few years after it was built. It is not shown on the 1842 Tithe Map, but is of a 'Georgian' form, so cannot have been much later. For its construction minor modifications were made to the house, including the adaptation of a window in the bay to give access and the construction of a high wall at the back with a pair of tall fan-lit windows, presumably to give light on a service yard behind. This conservatory sadly no longer survives but we know something of it from five tiny photographs found in the albums of the family of Thomas Cardus, Rector there between 1909 and 194013. These are: Fig 27: Shows the Rectory front with glasshouse attached to its right lean-to against a high wall, perhaps of a service wing. A boy of about 10 years, identified as Michael Cardus, the rector’s son, is seated on a donkey in front. Written beneath are the words 'Daisy and self'. It is springtime, about 1920 with daffodils and primroses in the lawn. A small stone figure holding a basket above its head, is to be seen in the middle of the lawn (Original measures 102 x 77mm) 13 I am grateful to Mr Tom Winter of Piddington, Northants, grandson of the Rev Thomas Cardus, for permission to reproduce these images Fig 28: Girl of about 14 years, identified as Rosie Cardus, the Rector's elder daughter, seated on pony in front of the glasshouses. A label beneath says 'Sunny Jim 1932', presumably the name of the pony. (Original measures 82 x 55 mm) Fig 29: View of the right hand end of the Rectory at its junction with the glasshouse. Shows shrub or rose border in front, mid 1920s. (Original measures 85 x 60) Fig 30: General view of front of Rectory with stone urn on a plinth in front. Glasshouses away to the right, early 1920s. (Original measures 82 x 58mm) Fig 31: Closer view of the glasshouse with two figures standing in front identified as Gertrude Marguerite Milln (née Cooper) and Alexander Lindsay Milln, c. 1927. (Original measures 82 x 58 mm) 4.2.3 From close examination of these images the glasshouse at Arlington Rectory can be described as a lean-to structure of eight bays standing on a high rendered masonry, probably brick, plinth. The five bays to the left (west) are at a slightly higher level than the three to the right. The first bay at the W end is shown as a partial one ingeniously shaped around the pronounced rounded end of the house. 4.2.4 Timber-built and white painted, the glasshouse is of typical Georgian form with small quarries of hand-made glass which contrast markedly with the larger panes of presumably crown glass used for the house itself. Typically such panes measure 6 inches across and perhaps no more than 8 or 9 inches long, although those used in the front wall can be seen to be square. Glass is cut with straight edges, rather than cusped, or 'beaver-tailed' with very small over-laps. Each light is 6 panes across and 6 high, the laps being lined up (other than where broken panes have been economically replaced) 4.2.5 Ventilation in the roof is by means of sashes. The photographs seem to show that the sashes of bays 2, 4, 6 and 8 moved, probably on counterbalance weights. Low level ventilation is provided by top hinged casements in the wall glazing in bays 2 and 4. For the front wall glazing of bays 6 and 8, the casements are centre pivot. The odd numbered lights would thus all have been fixed. 4.2.6 A gutter of half ogee section, presumably cast iron, took the rain from the roof to collection at its far right hand end 4.2.7 The prints are very small and it is not possible to define much internal detail, but Image 1 clearly shows a pair of very tall Georgian type fan-light openings in the rear wall opposite bays 2 and 4. There is a doorway opening (where there is today a window) in the wall to the left giving direct access from the house. 4.2.8 No apertures in the front wall, such as would suggest use of the house for vines, are visible so we may suppose this was used as a conservatory for delicate and specimen plants to provide outof-season colour and fragrance. The form is, however, of a service glasshouse such as we would find in the productive parts of a garden and would have been considered distinctly old-fashioned later in the Victorian period when lavishly designed conservatories became the norm. 4.2.9 In conclusion the structure is so very similar to those in the walled garden we must suppose it to be by the same hand and very likely contemporary. It is therefore important evidence for any contemporary recreation of the lost ranges flanking the central conservatory. 4.3 Metallic conservatory proposals, 1849 (adapted and extended from J. Lovie Conservation Plan) 4.3.1 Four designs for conservatories exist in the Chichester papers (NDRO: 50/11/151-55; 50/11/171/1-2)14. They have in the past been held to represent proposals for a building close to the walled garden or on the site of the glasshouses recorded on the Tithe map. However these do not relate to the range of glasshouses in the walled or flower garden areas. It is clear that these plans significantly post-date the construction of the glasshouses in the walled garden area and that the site under consideration for the construction of a fancy conservatory as part of Sir John’s improvements was immediately adjacent to or attached to the house. 14 - NDRO: 50/11/151-155 and 50/11/171/1,2 4.3.2 The designs are all for elaborate, highly ornamental metallic structures, and clearly originate from a variety of sources although only three of them are signed and identified. One plan is dated (June 1849)15, while another is drawn on paper watermarked 184716. The inference is that the plans were obtained c 1849 in connection with a project which was not realised. 4.3.3 These are all members of that rare breed of glasshouse made substantially of iron rather than timber, a material used until about 1800 only for fittings such as piers, astragals and stays. Being largely curvilinear their designs descend from an influential paper read by Sir George Stuart MacKenzie on 1st August 1815 at a meeting of the Horticultural Society of London which advocated this as ‘the Form which the Glass of a Forcing-house ought to have, in order to receive the greatest possible quantity of Rays from the Sun’. 4.3.4 Thomas Andrew Knight, President of the Horticultural Society, offered ‘Suggestions for Improvement of Sir George Stuart MacKenzie's Plan for Forcing houses, publishing these in the Society's Transcations in 1817. Loudon, the Scottish botanist, agriculturalist and garden designer, took up the challenge. He vigorously promoted the curvilinear form and the use of iron. In his Remarks On The Construction of Hothouses, published in 1817 he dismissed the objections (eg breakage of glass through expansion and contraction), citing the fact that there were ‘houses entirely of iron which have been in existence from five to fifteen years’. Examples given were Mr. Blackburne’s vinery near Preston, a conservatory at Windsor and a stove house for Thomas Hope at East Sheen. In 1816 Loudon invented a wrought iron bar which was lighter and less clumsy than cast iron and which could be heated and then curved. In 1818 Loudon erected a large, trial group of curvilinear horticultural structures in Bayswater, London and transferred his right in the invention of the bar to Messrs W & D Bailey of High Holborn, key co-operators in a revolution of metallic hot-house design. 4.3.5 By 1824 W & D Bailey had manufactured 24 curvilinear, horticultural structures for sites around the country (and two in Antwerp) including one for Thomas Andrew Knight, at Downton Castle, Herefordshire. Two experimental wrought iron curvilinear hot-houses and a curvilinear forcing house were built in the garden of the Horticultural Society in 1823. The most spectacular surviving examples attributed to Baileys are the Palm House at Bicton Park, Devon (c1825) and that at Dallam Tower, Milnthorpe, Cumbria. 4.3.6 The establishment in 1818 of the firm of Jones & Clark and later Clark & Hope of Lionel Street, Birmingham did much to promote the form in the south and midlands, supplying over 120 such structures for example to Wollaton Hall, Notts (1823), The Grange, Hampshire (1824) and Arundel, Sussex, this last having lately been restored by architects Carden & Godfrey. 4.3.7 Another Birmingham manufacturer, perhaps also starting as Jones & Clark, was Richards and Jones who were responsible for structures at Syon House, Middlesex; Eaton Hall, Cheshire and Alnwick Castle, Northumberland. Other firms which seem to have been producing metallic glasshouses in the 1820s and 1830s include William Crosskill of Beverley, Yorks who was described by a contemporary as having built ‘a considerable number’ of hothouses in ‘various parts of England’ and always ‘uses wrought iron for his curvilinear roofs’. Others include Edward Weeks of Chelsea, James Watts of Manchester, William Sutton and Thomas Jones and Sons but little is yet known of their output. 4.3.8 Iron-framed glasshouses exist at the forefront of industrial and horticultural revolutions where, with the technological advance and the introduction of exotics, country house owners would invest in the added durability of metal with confidence. The examples proposed for Arlington 15 16 - NDRO: 50/11/152 -NDRO: 50/11/155 slightly post date the repeal of glass tax in 1845 and the designs all benefit from the larger, heavier panes made possible by Robert Lucas Chance's improved process of producing cylinder glass. 4.3.9 Turner’s design17 bears the note. ‘NB This may be made either to stand alone or against the wall of a house’; a pair of unsigned designs clearly show the proposed conservatory attached to the corner (probably the north-west corner) of the house 18; other designs are for an entirely freestanding structure19. It thus appears much more likely that the conservatory designs were produced in the late 1840s for Sir John Chichester as part of his wider scheme for extending the house and providing new service quarters20, which were abandoned on his death in 185121. 4.3.10 Turner Conservatory The design dated 184922 has an elaborate domed roof with Anthemion cresting and rolling sash vents. It is is signed by Richard Turner of the Hammersmith Iron Works, Dublin, and is captioned ‘Sketch of Proposed Conservatory for Sir Bruce Chichester Bart’. Richard Turner (1798-1881) was one of the foremost manufacturers of iron-framed glasshouses in the nineteenth century. He understood the inherent structural properties of wrought and cast iron and successfully combined them to create vast spectacular structures that have stood the test of time. 4.3.11 The curvilinear glasshouses Turner built in the botanic gardens of Ireland were effectively the prototypes of the Palm House at Kew (1844-48) which he built in collaboration with Decimus Burton (Grant 2012, p32). The Palm House for Belfast's Botanic Garden, begun in 1839 to a design by Charles Lanyon, was an early Turner work and he went to to supply the curvilinear range for the National Botanic Gardens at Glasnevin between 1843 and 1848 and the conservatory for the Horticultural Society Garden at Regent’s Park (1845). Any of these commissions could have drawn Turner to Sir John’s attention having encountered his work in Ireland. Fig 32: Plan of proposed conservatory by Richard Turner of Dublin, 1849 17 - NDRO: 50/11/152 - NDRO: 50/11/155 19 - NDRO: 50/11/153, 50/11/154 20 - Plans for new service quarters by James Howell and dated 1849 survive in the house. 21 - The site of the dining room, built for Sir Bruce in 1865, seems the most likely location for the proposed conservatory. 22 - NDRO: 50/11/152 18 Fig 33: Conservatory design by J W Thomson of London 4.3.12 Thomson Conservatory The second signed plan23 is for an elaborate structure with a pair of domed compartments flanking a central range. It is captioned, ‘Designed by J W Thomson Landscape Gardener, Hammersmith for Thomas Sandes Esq, Elmer Wood, Liverpool’, and is endorsed on the reverse, ‘Mr Thompson, Garden Architect, Hammersmith, London’. There is no indication how this design, made for Thomas Sandes of Liverpool, came into the possession of Sir John Chichester. John Thomson (1805-1895) had trained at Kew and subsequently served as Gardener to the Duke of Northumberland at Syon House, before establishing a nursery business at Hammersmith (1835-60). 4.3.13 Walker Conservatory The third signed design is for as circular free-standing conservatory24. This was produced by William Walker, Engineer, of Low King Street, Manchester. Nothing further has been traced with regard to his work but it is not dis-similar to the domed metallic curvilinear conservatories at Hilton, Staffs and Farley, Hants (inset). 4.3.14 Unidentified conservatory. The fourth design is recalls the greenhouse temple form popular in the 1760s with glazing set between stone columns supporting a heavy stone cornice. However its roof is double-domed of curvilinear metal sections recalling the form of Paxton's Great Stove at Chatsworth which had such a profile. 23 24 - NDRO: 50/11/151 - NDRO: 50/11/153 Fig 34: Proposed free-standing circular conservatory by William Walker of Manchester (above). Inset the surviving metallic domed conservatory at Farley Hall, Alton for comparison Fig 35: Design for a proposed conservatory to be attached to Arlington Court (below) 4.4 Miss Chichester's conservatory, c. 1925 4.4.1 It is clear from Miss Chichester's photographs that the glasshouse range had not changed significantly since it was built, apparently in the 1820s. That a timber, small-paned glasshouse should have lasted for a century unscathed is a tribute to the care and maintenance lavished upon it up until the 1914-18 war. The original choice of first class timber such as pitch pine and the fungicidal properties of traditional lead-based greenhouse paint will also have played a part in ensuring its longevity. 4.4.2 The Tithe map indicates that the range of glasshouses was constructed with a straight front façade facing the flower garden, and with no projecting central portion. This configuration continues to be recorded on the 1886 and 1903 OS, and is confirmed by late C19 photographic evidence. 4.4.3 The old central compartment, which had glazing to the floor and a higher roof ridge than the flanking ranges which were supported by dwarf walls, was taken down and rebuilt with a projecting middle bay in about 1925 possibly by the firm of Messenger & Co of Loughborough although as Miss Chichester's name has not been traced in the company's records and as other firms supplied similar glasshouses, this is not certain. 4.4.4 The apex of the roof of the projection was held by an ornamental cast iron column and the glazing moved to broader panes to admit more light. The early 20 th century conservatory had doubled rafters at either ends both to support its own roof and to link to the lower level roof of the adjacent vineries which were retained. 4.4.5 The condition of the flanking vinery wings deteriorated after Miss Chichester's death and were finally taken down in the late 1960s although their dwarf brick supporting walls were retained to enclose ornamental planting laid out to a scheme by Graham Thomas. In a modified form this scheme survives today. 4.4.6 The c.1925 Conservatory was taken down in 1980 to make way for the Jebb Conservatory as discussed in Chapter Two. Fig 36: Interior view of the c.1925 conservatory shortly before its demolition in 1980 Fig 37: Photo of c. 1930 showing that Compartment 3 had been rebuilt as a conservatory with a projecting central bay, leaving the early nineteenth century vineries either side intact (above) Fig 38: Page from the Messenger & Co catalogue of c.1921 showing a projecting plant house of very similar form to the c.1925 Conservatory at Arlington (side) Fig 39: Interior views of the c.1925 Conservatory showing roof structure taken in 1978 Fig 40: Interior views of the c. 1925 conservatory, taken in 1978 Fig 41: Exterior views of the c. 1925 conservatory, taken in 1978 Fig 42: General view taken in 1978 of the interior of the c.1925 conservatory with the cast iron support which Jebb found was too defective to reuse in the 1982 conservatory 4.5 The Pinery (from J. Lovie Conservation Plan) 4.5.1 The pinery at Arlington is, unusually, a detached building set on its own at some remove from the walled garden. Currently in very poor condition, research on this structure is needed ahead of its restoration. However this research falls outside the scope of the present report. An edited version of the account given by Lovie (2009) is reproduced here to complete the suite of glasshouses at Arlington: Historic Development 4.5.2 The pinery or pineapple pits were originally constructed to a plan of 1814 signed by 25 ‘W.A.’ . The identity of the designer has not been ascertained with certainty. It is possible, on the one hand, that the initials refer to the Royal Gardener, William Townsend Aiton (1766-1849), who was noted for his work refining the design of the pine pits at Kensington Palace26; on the other hand, an in the absence of any additional evidence, it is perhaps more likely that the designer was a local nurseryman such as William Addiscott of Exeter (fl c 1820-1850) or a gardener at a neighbouring estate27. 4.5.3 W.A.'s plans for the pinery represent a bespoke structure of the period intended for the specialised raising of pineapples. The production of this fruit had been a status symbol to which owners of country estates had aspired since the late seventeenth century, and through the Georgian period the science of its growth became a matter of controversy and rivalry between country house head gardeners28. The Arlington pinery appears to differ from the designs discussed by Loudon, but broadly follows those advocated by William Speechley (1735 - 1819)29, but it is not unlileky to be the work of a local designer30. 4.5.4 The original structure had a sloping timber and glass roof supported on a low brick wall at the front and a higher back wall. A shed for fuel and stoke holes was situated at the centre of the rear wall and flues ran from two boilers round two divisions within the house: the left-hand division being a succession house31, and the right-hand division a fruiting house. In having only two rather than three divisions the Arlington pinery differed from the standard pattern which would have included a house for the crowns and suckers. This suggests that the stock plants were either kept elsewhere, perhaps in one of the glasshouses in the Flower Garden, or that new plants were boughtin on a regular basis. At a combined length of forty-five feet, the two divisions at Arlington correspond approximately to the proportions of Nichol’s pinery as described by Loudon, which was 100’ in length (comprising three divisions), which ‘would give pine-apples enough for a large family’32. 4.5.5 The pinery is recorded on the Tithe map (1842) as a single, free-standing structure on the northern side of an approximately elliptical-shaped enclosure designated the ‘Outer Garden’. It 25 - NDRO: 50/11/145-150 - J C Loudon, Encyclopaedia of Gardening (1824, ed of 1830), p 503 27 - The fact that Colonel Chichester turned to a local architect, all be it one with more than local experience, to design his house perhaps suggests that it is unlikely that he would have solicited a design for his pinery from the Royal Gardener. In addition, the form of pinery adopted by Aiton at Kensignton (and illustrated by Loudon) differs from the Arlington pinery in that it was sunk into the ground in order to conserve heat. 28 - Loudon (1824, ed of 1830), pp 502-506 refers to differing designs for pine pits produced by Walter Nicol (d 1811), gardener at Weymss Castle and Thomas Baldwin (fl 1810s), gardener at Ragley Hall, Warwickshire, W T Aiton at Kensington Palace, T A Knight owner of Downton Castle, and Henry Scott , gardener at Chiswick (d 1760s). 29 - W. Speechley, A Treatise on the Culture of the Pineapple, 1779 30 - It is closest to Nichol’s design and may have been inspired by discussion in the horticultural press. 31 - A house for plants being brought into fruiting condition; the division allowed each stage of the plant’s development to be provided with optimum growing conditions. 32 - Loudon (1824, ed of 1830), p 502 26 is not known with certainty what was grown in this area, but it seems likely that it was an auxiliary kitchen garden, perhaps with an emphasis on the production of soft fruit. 4.5.6 At some point during the nineteenth century, most probably after the succession of Sir Bruce Chichester in 1863, the original pinery was altered through the raising of the roof height, the formation of an ornamental terrace to the south with a pair of Irish yews flanking the central axis of the glasshouse, and the formation, at the end of the century, of a range of frames or pits below the terrace. Evidence of an ornamental rockwork finish to the internal back wall of the remodelled glasshouse survives (2008). 4.5.7 It is not known what was originally cultivated in the remodelled glasshouse, but the opportunity for high levels of heat suggests the existence of a collection of exotic plants, perhaps including orchids which were becoming highly fashionable plants among country house owners in the second half of the nineteenth century. Latterly, Jan Newman reported that the house was used by Miss Chichester as a fernery33: the ornamental rockwork within the house would accord well with such a purpose, whether the structure was still heated or not. 4.5.8 The pinery, terrace and Irish yews are recorded on the 1886 Ordnance Survey. The frames to the south of the terrace were constructed between 1886 and 1903 when they were recorded on the second edition 25” OS. The 1886 OS does not indicate with any clarity whether the Outer Garden remained in cultivation. An open area of ground to the south and south-west of the pinery may correspond to a substantial remnant of this feature, but its elliptical shape, presumably defined in 1842 by fences or hedges, had been lost by 1886. Recommendations 4.5.9 Consideration should be given initially to the consolidation (to conservation standard) of the remains of the pinery to be based upon the results of specialist archaeological recording and analysis. The objective of the archaeological research is to obtain a thorough understanding of its condition and conservation requirements, engendering an appreciation of heritage quality, considering in detail and by reference to similar sites facing similar challenges, the options for conservation, adaptation, change or even abandonment. By reference to similar examples elsewhere it would seek to assess the impacts of these options on heritage significance and the opportunities for public benefit, offering pragmatic and informed solutions. 4.5.10 Given the archaeological evidence, reinstatement should be considered in the longer term. This may be either in its original form (as shown by the 1814 plans) or in its latter form as an ornamental house depending on the quality of the evidence and funding. 4.5.11 The setting of the pinery should be conserved, the terrace reinstated and the remains of the frames consolidated. The wider setting of the pinery also requires consideration. A longerterm objective could be its reinstatement as an ‘Outer Garden’, perhaps planted with fruit trees and soft fruit along the lines suggested by the evidence of the Tithe map (1842). Significance 4.5.12 The pinery is a most interesting and important horticultural structure. Its consolidation and ultimate reinstatement as an area for public access would allow the explanation of the whole range of nineteenth century horticultural endeavour at Arlington and, with appropriate interpretation, would give visitors a fascinating insight into the scientific approach adopted by gardeners to the cultivation of exotics such as pineapples. 33 - J Newman, 1997 – pers comm. Noted on file Fig 43: Plan by ‘W.A.’ dated 1814 for the original pinery (NDRO) Shows its western compartment with its and flue system for the succession pit and part of the furnace house (above) Fig 44: Section by ‘W.A.’ dated 1814 for the original pinery (NDRO) Fig 45: Pinery - remains of structure as altered in mid-nineteenth century, July 2012 (above) Fig 46: Pinery - back shed steps down to furnace room (side) Fig 47: Pinery - W end of back shed to left, site of pinery to right of the truncated wall (below) 4. 6 Small greenhouse A small glasshouse is recorded on the 1886 Ordnance Survey 1:2,560 map and by late nineteenth and early twentieth century photographs. It stood towards the north-western corner of the walled garden. It is not shown on the 1842 Tithe Map and is in any case of later 19th century form. Entered at its W end, the house is of four bays of lights alternately ventilated by hinged casements in the roof and front wall. The National Trust has recently substantially rebuilt this structure. Fig 48: The lean-to glasshouse in the kitchen garden, c 1900 Fig 49: The lean-to glasshouse in the reinstated kitchen garden today 4.7 After Jebb: the Salisbury Conservatory, 2012 4.7.1 The question of how to manage the deteriorating condition of the imperfect 1982 conservatory vexed the Trust almost from the start. Re-painting was not carried out sufficiently regularly and deterioration of sills became apparent within a few years. However the thirty years which followed its design have been marked by a renaissance for the Conservatory to the extent that pastiche 'Victorian' reproductions are now available 'off the shelf' to a very suburban market. Numerous companies now make them both in timber and aluminium and there is a much greater level of understanding of their strengths and weaknesses, so that it would be inconceivable today to procure something so bespoke as was the 1982 Conservatory or to have to relearn the same lessons. 4.7.2 On the subject of the demise of the 1982 Conservatory, the Garden Conservation Plan (Lovie 2009) comments: Careful consideration must be given to the replacement of the existing conservatory. It is important to understand that the present building does not replicate the early C19 conservatory, but it does provide a focal point to the axial layout of the Victorian Garden. A structure is therefore required in this location; but its form must be considered with care. While it may be tempting insert a building of modern design, it has to be considered in relation to the strongly Victorian character of the flower garden to which it relates; conversely, does sufficient of the original structure survive (supporting wall and the heated wall) to justify a reconstruction of the early nineteenth century range? 4.7.3 Once the Trust had accepted that replacement of the decaying Jebb Conservatory was necessary and affordable, the decision was taken to return it to something more like its c.1925 than the 1982 predecessor albeit with the longer panes of the 1982. The recently established local firm of David Salisbury Conservatories34 won the contract with a set of drawings dated 2012. The conscious decision to return to the form which Michael Trinick had decried as 'oddly truncated' is attributable to a renewed appreciation of historic form and a recognition that the flanking vineries represent more of an opportunity than a liability. 4.7.4 The Salisbury Conservatory restores the lower timber diagonals in the glazed sides seen as the link with the flanking vineries in the c.1925 house; it restores the gable ends from the hipped form adopted by the 1982 house, and it restores the squared off (where had been the hot wall chimneys) wall ends above the Conservatory from the scalloped form created for the 1982. In terms of the design and ambitions for the site, the reasons for approval of the 2012 scheme represent a remarkable volte face for the those given by the National Trust Architectural Panel in its approval of the 1982. Whereas in 1982 the aesthetic judgement of a few members carried the day, in 2012 informed practical and garden historical considerations weighed more heavily. 34 Based on the Isleport Business Park, Bennett Road, Highbridge, Somerset, TA9 4PW, tel 01278 764400 Fig 50: Salisbury Conservatory nearing completion exterior, November 2012 (above) Fig 51: Salisbury Conservatory nearing completion interior, November 2012 (below) Appendices to Archaeological Report on Conservatory and other Glass at Arlington Appendix I BIOGRAPHY of Philip Jebb 1927-1995 Philip Jebb was one of the leading private-client architects working in Britain during the second half of the 20th century. He practised from 1953 to 1994, first as a freelancer in London, New York, and San Francisco. Then between 1956 and 1958 he shared an office with his brother-in-law Francis Pollen. From 1958 he ran his own firm through eras of austerity (the 1950s), high taxation (the 1960s) and a depressed building industry (the late 1970s) before a final boom in country-house work (the 1980s and 1990s). Jebb’s versatility in an age of increasing specialisation, the celebrity of some of his clients, and his reputation in painstaking historic buildings work have perhaps diverted attention from his success in executing new buildings (in both contemporary and classical idioms) during what was a fruitful and fulfilling career. Some accounts of Jebb’s work have dwelt on the frustrations he suffered in turning a client’s proposal into a building – that he “stood out of the mainstream” during architecturally “darker” decades. And yet his experiences in having work built seem typical of the 20th-century architect, from giants such as Sir Edwin Lutyens, Frank Lloyd Wright and Le Corbusier down. An apposite comparison was suggested by a note sent to Jebb in 1993 by his friend and client Sir John Smith. It was found at Coutts Bank and is in the hand of the architect CFA Voysey: Total number of clients in 40 years up to end of 1921 – 224, of which 53 have employed me more than once. Have built 117 houses in the same period, in addition to many other works. Total amount of fees for same period – £31,172-13-10. Jebb answered by return: Thank you for riveting Voysey statistics which caused me to waste a lot of time calculating from my “Job book”... that in the 35 years to 1993 I have had (approx) 180 clients of which 57 have employed me more than once. In the same period some 440 jobs were given a number (i.e. had a file started) of which (approx) 224 led to building work and 217 did not. Interesting Jebb felt the influence of his immediate great predecessors, including Lutyens, Wright and Le Corbusier, as much as that of his favourite classical architects, William Kent, John Soane and Nicholas Hawksmoor. In the late Fifties he admitted to the casino-owner John Aspinall – for whom he was working on restoring two first-rate Palladian buildings, Howletts, in Kent, and 44 Berkeley Square, in Mayfair – that he still longed to design a substantial modernist building. His most satisfactory built project in that respect was the Castlerosse Motel, at Killarney, Co Kerry (1960), his most openly Wrightian work. Arne Jacobsen, then at the height of his réclame, paid Jebb an compliment exquisite in its brevity – “Thank you for your nice hotel” – written on the reverse of a business card. Yet Jebb was no avant-gardiste, nor a polemicist in his own cause. He saw no good in promoting his own work through advertising or giving interviews to magazine journalists, advising younger associates and collaborators, such as the interior designer David Mlinaric, to follow the same course. At the same time, he was happy to help those researching studies of those he worked with, including biographers of the decorator John Fowler, the garden designer Lanning Roper, and his lifelong friend and brother-in-law Francis Pollen. He cared for the history of his profession, admired in particular the work of the historians John Harris and Howard Colvin and had developed a synopsis in the early Fifties, while completing his diploma at the Architectural Association, for a biography of Hawksmoor. Jebb’s reluctance to encourage writing on himself or his buildings, meant that critical reception of his work really starts with the publication of obituaries in the British national newspapers after his death in 1995. The first extended newspaper article on one of Jebb’s new buildings was published in 2001, the first full-scale magazine profile of Jebb and his work in 2004, and the first Country Life feature on one of his new houses in 2005. To Francis Pollen’s biographer Alan Powers, Jebb and Francis Pollen (though very different as architects) formed part of an alternative narrative of British architecture in the postwar years; one little celebrated in the architectural press during their lifetimes. To Powers’s contemporary John Martin Robinson, author of Jebb’s unsigned obituary in the Daily Telegraph, Jebb was the most distinguished traditional architect practising in postwar Britain and his work at 44 Berkeley Square (1961-63) and Curzon House, the best Georgian restorations in Mayfair. Stephen Gardiner, the architect and journalist and author of Jebb’s unsigned obituary in The Times, particularly admired Jebb’s ability to marry the demands of property development and classical propriety in a new building at 12/14 Cheyne Walk, London, in the early 1970s, without suppressing his own authorial verve or resorting to architectural pastiche. “For sheer architectural expertise…” Gardiner writes, the building “is excellently achieved”: his is an 18th-century row of considerable quality. Perfection of detail and proportion was essential to produce a work in harmony with the rest. But the development was for a block of flats, not a house, which made the aesthetic problems far harder to resolve. Yet Jebb managed to do so, and in such a way that his façade shows up the horribly inept Victorian addition to Rossetti’s house, a near neighbour, and other ugly alterations from the period elsewhere. Jebb, by picking up salient elements of the best in the terrace… pulled off a singular feat in this work of the 1970s: few passers-by would realise it was not built two hundred years ago. Among the succeeding generation of architectural historians, Christopher Woodward, Jeremy Musson, Mary Miers and the late Giles Worsley (author of a full-page profile of Jebb’s neoGeorgian house at Oakingham, “A modern classic?” Daily Telegraph, 13 October 2001) have carried the flame for Jebb’s work. In Jebb’s lifetime, the private-client architect was a rare creature. What gave Jebb extra rarity was his willingness to take on large amounts of commercial work – night clubs, casinos, hotels, property development in London and in Spain. He produced much of his best work to a tight business brief, where space was limited and time short, all without losing sight of the need for a humanising fancy in the final building. He worked best for clients whose sense of proprietorial freedom and visual flair meant they might well have done the work themselves. Perhaps his finest work – El Cuartón (1966-70), a new village near Tarifa overlooking the Straits of Gibraltar – is one where his personal stamp as an architect is most apparent. It was conceived as a set of holiday apartments on multiple levels running up the side of a cork-tree lined hill, commissioned by Hugh Millais, actor, writer, restaurateur and man of many talents. It was described as Annabel’s-by-the-Sea, as its intended customers were to come from the same tranche of London café society who danced at Annabel’s night club, which Jebb had created for Mark Birley 10 years previously. The brief was for a new 18th-century Spanish village. Jebb’s attention to detail meant the brief was met elegantly and without collapsing into pastiche. At the same time, the positioning of the “blocks” of apartments – in relation to each other and to the steeply sloping site – is a triumph of witty three-dimensional planning. And the main façade of the village, approached from below, has an almost Cubist quality (expressing one plane and many at the same time), quite unlike anything else in Jebb’s work. The villas around the village show Jebb at his most playful and elegant. El Cuartón was admired by François Spoerry, architect of the celebrated, and exactly contemporaneous canal town of Port Grimaud, St Tropez. More than 30 years after its completion, when Cuartón had developed into a local village, it stood out for architectural historians, including Giles Worsley, as a first-rate example of 1960s Mediterranean resort architecture. In the 1980s and 1990s, Jebb continued to produce a high volume of historic buildings work for the Landmark Trust, notably on Lundy, in the Bristol Channel. It was also the period when he executed two big restoration projects –Badminton House, Gloucestershire, for the Duke of Beaufort, and Curzon House as a new casino for John Aspinall – and designed his four largest built houses: Las Irlandesas, a town palace in the centre of Madrid for the March family; Montebello, a magnificently sited villa at Totteridge Common, north London, for the record producer Micky Most, with a giant order Colonial entrance front; La Cañada, Guatemala City, a clifftop urban villa for Similiano Garcia; and Oakingham House, near Henley-on-Thames. One of his last complete works – the restoration and enlargement of Poston Court, Herefordshire, a hilltop rotunda by Sir William Chambers – called on all his strengths as an architect: familiarity with the classical language, sympathy for an historic building and ingenuity at planning in confined spaces. To Marcus Binney, writing in The Times, the revitalised house was “immaculate baby grand, the suavest conceivable expression of the new taste for grandeur in miniature”. When Country Life featured the house in 2005, they described it in a cover line as “Modern masterpiece: Philip Jebb’s last country house”. Towards the end of his life, Philip Jebb was in correspondence with the Royal Institute of British Architects, who wished to take a large part of his archive. When the RIBA realised they lacked the storage space to house them, a number of Jebb’s drawings was passed instead to the National Monuments Record (NMR), the public archive of English Heritage at Swindon, in Wiltshire. Before this transfer was made, clients were offered their drawings. Drawings for completed projects for the National Trust and Landmark Trust were returned to the trusts. Drawings for active projects were retained by his associates, who continued work on them during his final illness and after his death on 7 April 1995. Some of his drawings have been preserved at the English Heritage Archive (formerly NMR), Kemble Drive, Swindon SN2 2GZ. Appendix II BIOGRAPHY of Thomas Lee (Jnr) 1794 - 1834 Thomas Lee (Jnr) (1794 – 5 September 1834), was a n architect, the son of Thomas Lee (Snr) (1756 – 1836) of Barnstaple, Devon, also an architect. The younger Thomas was educated at Barnstaple Grammar School and left to train briefly in 1810 at Sir John Soane's office, where his father no doubt placed him, but left for the office of David Laing before becoming articled. He was also admitted to the Royal Academy School in 1812 and won a Royal Academy silver medal in 1816, for a drawing of Lord Burlington's villa at Chiswick, and a gold medal from the Society of Arts, for a design for a British Senate House.[2] His first major work was the Wellington Monument, Somerset. Lee's further work was characterised as "eclectic" by Howard Colvin, who instanced the pared-down Soanean neoclassicism of Arlington Court, Devonshire (1820-23 for Col. J.P. Chichester), the Tudor Gothic Eggesford House, Devon (1822 for Hon. Newton Fellowes; now a ruin), several "Commissioners' Gothic" churches in Worcestershire, Cheshire and Staffordshire, as well as an unusually early neoNorman one. In 1826 he designed the Guildhall in Barnstaple (finished 1828 and recently restored) which makes an impressive frontage for the later Pannier Market. Colonel Chichester occupies a key position in the development both of the family, the buildings at Arlington, and the designed landscape. In 1790 he remodelled the existing manor house, creating a fashionable, if poorly built, late Georgian country house. Thirty years later his construction of the new house by Thomas Lee on a fresh site marks a development of the improvement in the landscape at Arlington. Colonel Chichester’s choice of architect, Thomas Lee (1794-1834)35, and architectural style influenced by Sir John Soane, marks him out as a man of sophisticated taste. The contemporary Rectory at Arlington is, like the house, almost certainly also by Lee. James Rothwell comments: 'I certainly came to the conclusion that it was likely to be by Lee - it is so Soanian and very comparable to Lee's other work, albeit that the diocesan records only refer to the builder.' Rothwell continues: “As to the glasshouses, I suppose Lee could have designed them but I did not come across any references. I do not know the kitchen gardens at his other sites except Eggesford's, which is a garden centre. Plans of the house do survive at the Portsmouth estate office at Farleigh Wallop, Hants. I don't remember anything on glasshouses but I was not looking for them. Lee did include a curving conservatory on an early plan, for adapting the old house by the church, but that is pretty sketchy. As Lee was the most Soanian of Soane's pupils it might be worth looking at Soane sites too.” (pers comm, July 201236) In 1834 Thomas Lee died in a swimming accident at Morthoe, near Barnstaple. 35 Lee’s other commissions include the Wellington Monument, Blackdown Hills (1817), Eggesford House (1822) and Priory Hall, Dudley (1825) 36 Lee's life and work has been researched by James Rothwell for his MA. James is currently curator for the National Trust in Hampshire. Appendix III: Arlington Archive The project archive comprises nine timber samples and two metal component samples recovered from the 1982 conservatory at the time of its demolition. These are listed in the table below, plus: Archaeological report (pdf) Digital Archive of source documents, drawings and photographs on CD Copy of the original specification for the 1982 Conservatory Copies of all original drawings for the 1982 Conservatory, dated 2nd August 1980 Copy of Planting Plan drawing dated December 1982 Site plan on drafting film of the 1982 Conservatory as in 2012, scale 1:50 Set of 39 No. 7” x 5” full colour photographic prints and negatives, taken 15th June 2012 Set of 27 No 7” x 5” black & white photographic prints and negatives, taken 15th June 2012 Conservatory Components Item no Identification description 1 Window moulded rebated, white 2 & Material cill, Softwood, & drib close grain painted Dimensions Use & date Condition cut from 120 x 70 1982 Jebb, ran atop wall Poor, decay in mm (slightly less plate and under middle where damp than 5” x 3”) windows right round trapped Window fixed Softwood, Cut from 70 x 60 1982 Jebb, to windows Fair, some decay in right round rebate damp trap frame, moulded close grain mm and rebated for glass, painted white 3 Window mullion, Softwood, chamfered medium section, glazing grain rebates 4 Casement opening frame Cut from 70 x 60 1982 Jebb, to all Fair, slight decay in mm windows right round rebate damp trap Softwood, Cut from 70 x 40 1982 Jebb, to all Very poor, extensive close grain mm windows right round decay where no timber protection (not fully primed before assembly) 5 Casement Softwood, Cut from 70 x 40 1982 Jebb, to all As 4, above opening frame, close grain mm windows right round form as 4, above but shorter section 6 Glazing bar, with Softwood, Cut from 40 x 22 1982 Jebb, to all Good, and the putty top & bottom close grain mm (slightly less windows right round has generally held. than 2” x 1”) putty for 4mm float glass 7 Bead, with half- Softwood, Cut from 22 x 9 1982 Jebb, round mould close grain mm throughout trim Good, although not apparently primed preassembly 8 Cornice, with Hardwood, Cut from 52 x 52 2012 Salisbury, dec for Excellent, new good ovolo mould non-native, mm (approx 2” eaves? paint protection fast grown square) 9 Nosing, with Hardwood, Cut from 38 x 32 2012 Salisbury Excellent, new good rebate and mould drip non-native, (1½” x 1¼”) fast grown paint protection 10 Handle to lever type scroll including and key hole door, Brass with copper end, alloy plate or Plate is 152 x 49 1982, Jebb (chosen by Excellent. A good mm NT) quality fitting (6” x 1½”) 11 Stay to casement Brass with five settings, copper scroll end alloy or Stay length is 305 1982, Jebb (chosen by Excellent. A good mm (1 ft) NT) quality fitting

RELATED PAPERS

RELATED TOPICS

Log In

Archaeological work at Arlington Court, Devonshire The Conservatory (and other glasshouses) at Arlington Court Building Survey and Historical Analysis Archaeological Report

Archaeological work at Arlington Court, Devonshire The Conservatory (and other glasshouses) at Arlington Court Building Survey and Historical Analysis Archaeological Report

Related Papers

RELATED PAPERS

RELATED TOPICS