Cork Street and Savile Row Area: No. 7 Burlington Gardens

No. 7 Burlington Gardens: The Royal Bank of Scotland

Formerly No. 1 Old Burlington Street

The nucleus of the present building is formed by the surviving carcase of a house built in 1721–3 and known as Queensberry House. Its site comprises the westernmost seven bays of the southern (Burlington Gardens) front and the southernmost 53 feet of the western (Old Burlington Street) front. Queensberry House was rated in Old Burlington Street but its front faced south. It was altered and enlarged to its present extent c. 1785–9 and was thenceforward known as Uxbridge House until its sale to the Bank of England in 1855, when further alterations were made. Thereafter it was rated in Burlington Gardens.

Queensberry House

This large town house was first occupied in 1723 or 1724 by Charles Douglas, third Duke of Queensberry and second Duke of Dover, who remained here until his death in 1778. It had been begun in 1721 by the bricklayer, John Witt, to the design of Giacomo Leoni. It is certain, however, that this had been not for the Duke but for the Irish Member of Parliament, John Bligh, who in September 1721 was created Baron Clifton, in 1723 Viscount Darnley and in 1725 Earl of Darnley. (fn. 2) His name had not occurred among the subscribers to the important architectural publications of 1715–16, the first volume of Vitruvius Britannicus and the first volume of Leoni's own Palladio, but he had subscribed to the second volume of Vitruvius Britannicus published in 1717 and later subscribed to Leoni's Alberti in 1726. He was also, like Burlington, an original subscriber to the Royal Academy of Music. (fn. 3)

In October 1719 Burlington's agents informed him that they had 'lett' this site. (fn. 4) It appears, however, that what had been concluded was not a lease but an agreement for building to be followed by a lease. Whatever the precise nature of the transaction the other party was presumably the bricklayer Witt. The evidence for this and for the following stage of the site's history is chiefly in a manuscript 'Conveyancer's Guide' of 1739 in the British Museum. (fn. 5) This contains specimen forms of various legal instruments, based on actual deeds. The parties and other persons mentioned are sometimes represented only by initials and the other features of the deeds are sometimes abbreviated but enough survives to make these formularies good evidence of the substance of the deeds on which they are based: the possibility must be borne in mind, however, that some of the covenants and provisions may have been introduced simply as exemplars. Two of the sites providing specimen instruments are on the Burlington estate, one in New Burlington Street and the other that of Queensberry House. Two deeds are given for this site, of which the first is dated 12 April 1721 and is a building agreement between John Bligh, esquire, and John Witt, bricklayer. (fn. 6) This begins by stating that Witt had 'lately assigned' the site to Bligh. No precise information about this assignment is given but it presumably concerned a conditional title to the site comprised in an agreement between Burlington and Witt and had been made before 4 March of that year, when 'Mr. Blye' is mentioned as a tenant on the Burlington estate. (fn. 7) The agreement now concluded was that Witt would build 'one good and substantial Brick Messuage or Tenement 70 foot by 50 foot', for which Bligh would pay him £3000 in seven instalments as the work progressed, with a later addition or subtraction adjusting the price to the value set on the completed house by two surveyors. The work was to be 'in such manner and form as the plans now drawn and designed by Ja. L. for the said Building and which are annexed to these presents shall direct'. No such plans are given with the copy, but fortunately plans of the completed house were to be published by the architect. This was the Venetian, Giacomo (anglice James) Leoni, who appended two plates of the design, including a view of the main elevation in an Italianate country setting, to his edition of Alberti published in 1726, where he described himself as 'Inventor and Director of the Building there of'. (fn. 8) The design is theredated 1721 (Plates 76, 77a).

The accompanying text indicates a measure of active interest by Lord Burlington in the appearance of the house to be erected on the part of his estate nearest to his own mansion. Leoni states that Burlington himself designed 'the portal on that side of the House which opens to the Court-yard', that is, in Old Burlington Street, and the relevant drawings still exist in the BurlingtonDevonshire collection in the Library of the Royal Institute of British Architects. (fn. 9) Leoni goes on to claim Burlington's approval of his design. 'When I laid this Design before him for his Approbation, his Lordship gave leave to the Person who executed it, to set the Front towards his own Garden: A Privilege denied to all the other Houses there'. As will be seen, when the interior came to be completed one room was modelled on an interior at Burlington House. Some sixteen years later when the site was extended to Savile Row, Burlington again made provision that a design of his own should be used for any gateway erected in that street: this was, however, his common practice in leasing frontages on the west side of Savile Row. Burlington had subscribed to the third volume of Leoni's Palladio which appeared in 1718 (fn. 10) and later subscribed to the Alberti of 1726, which contained a tribute to him in the introduction. But like the other works of Leoni the design for Queensberry House was not included in the third (1725) volume of Campbell's Vitruvius Britannicus, possibly because by then Campbell or Burlington did not approve of Leoni's modified Palladianism.

The executant of Leoni's design, John Witt, was a builder of some prominence. He had been responsible in about 1717 for the development of sites in Cheyne Walk, Chelsea, (fn. 11) and took more than one building site on the Burlington estate. In c. 1722–3 he was receiving payments from the Commissioners for building Gibbs's new church of St. Martin in the Fields. (fn. 12) As Mr. John Witt, bricklayer, he subscribed to the second (1721) edition of Leoni's Palladio and the subscription list of the Alberti of 1726 contains his name, although he was then deceased, having died by June 1724, leaving his father Andrew Witt, gentleman, administrator of his estate. (fn. 13)

In the agreement of April 1721 Witt undertook to complete the house in eighteen months. A lease on 31 October of the next site northward mentions the new messuage of Lord Clifton (as Bligh had then become) to the south. (fn. 14) By the following year the house was sufficiently complete externally to be commended by Macky in the second edition of his Journey Through England where he praised the new houses behind Burlington House. 'That which fronts the Garden-Wall is a very noble one, the Columns and Windows of the Corinthian Order, of Free-Stone, belonging to Mr. Blythe, who married the Earl of Clarendon's Daughter'. (fn. 15) The omission of Bligh's title perhaps carries this observation back to 1721, either before or not very long after his advancement to the barony of Clifton in September of that year.

The building agreement with Witt had left Bligh free to refuse the house, within three months of its completion, when he would 'procure' a lease of the site to be made by Burlington to Witt. He would also accept a mortgage of the site to secure the repayment of the sum paid by him to Witt. In 1722 Lord Clifton decided not to take the house. Leoni's comment on his published designs states that 'the Person for whom I designed and finished this Structure, in order to save Stone, wou'd not execute the Rustic of the Basement, which wou'd have added a great deal of beauty to the Front: But this may easily be done still'. Leoni's failure to name 'the Person' makes his comment read complainingly, but may have been intended tactfully to conceal Lord Clifton's desire for economy: Clifton, in his new guise as Earl of Darnley, was in fact a subscriber to the volume in which the design was published. The surrender of the new house, which was finished only externally, occurred in 1722, (fn. 16) and it may be that it was caused by no dislike of the house but by the death of Clifton's wife, aged twenty-six, in July of that year.

She had been the second cousin of Catherine, the young Duchess of Queensberry (both being great grand-daughters of the first Earl of Clarendon), and it was the third Duke of Queensberry, an Anglicized Scot aged twenty-four, who now took over the house. He was himself a cousin of Burlington, being a son of the second Earl of Burlington's sister Mary. (fn. 2) With his Duchess he acted a role of patron to artists and men of letters, and both he and the Duchess, like the Burlingtons, seem to have had some talent for the arts. He had been in Italy a little before his cousin's second visit and had become acquainted there with William Kent who produced 'a picture figurs as big as ye life' for him and reported to a correspondent that 'my Lord Duck Queensbery … draws very prettely himself'. (fn. 17) Queensberry had not subscribed to the volumes of the first edition of Leoni's Palladio but was a subscriber to the onevolume edition which had appeared in 1721 and also to his Alberti of 1726.

The second item relating to the house in the 'Conveyancer's Guide' is given only a year-date in 1722, and was between Witt and Queensberry. (fn. 16) It rehearsed that Clifton (the initials J.B. are used with reference to Bligh's name at the time of the 1721 agreement) had 'signified … his Intention of parting with the said Messuage' and that Queensberry had 'a desire to purchase the same'. Witt therefore covenanted that as soon as he received authority from Clifton he would procure a lease of the site and the messuage 'now in building thereon' to be made by Burlington to Queensberry 'at and under the rent and covenants reserved and mencioned in and by certain Art[icle]s of Agreement made bet[ween] the said Earl of the one part and the said J[ohn] W[itt] of the other part', presumably those concluded in the autumn of 1719. Witt further undertook to 'finish' the house according to detailed specifications which comprised the whole of the interior work including floors and staircases. Queensberry undertook that as soon as Bligh re-assigned his interest in the house, he would pay Witt £2000 (which was perhaps the sum which Bligh had already paid for the building of the house and which Witt had to repay to him) and that as soon as the lease from Burlington was executed he would pay Witt a further £4300, making the whole cost of the house to him £6300.

The lease from Burlington to Queensberry was executed by 4 March 1722/3, as on that day it was submitted for registration in the Middlesex Land Registry. The registration was made on the 23rd of that month. (fn. 18) The lease survives in the possession of the present owners of the site. (fn. 19) It bears the fictitious date of 8 December 1719 for its signing and sealing, and runs, like almost all the other sites in Old Burlington Street, for 61 years from Michaelmas 1719, at a peppercorn rent for the first two years and then at £35 per annum. It was said to be 'in part performance of a certaine contract or agreement made by the said Earle with John Witt and to which the said Charles Duke of Dover and Queensberry is now intitled for the letting to him the peice or parcell of Ground hereafter mencioned', but no reference is made to Bligh's interest in the site. This is said to front towards the west; by the time the lease was actually drawn up this was, as Leoni's design of 1721 shows, no longer true, but the description may derive from the agreement actually made in 1719. The covenants by Queensberry included an undertaking to lay the pavement before the house and to pay a share of the cost of maintaining the common sewer in Old Burlington Street. He also undertook, like other lessees from Burlington at that time, not to let the house 'for a Butchers house or shop slaughter house poulterers house or shop Tallow Chandler Melter of Tallow Soapmaker Tobacco Pipe Maker Brewhouse Distiller Farrier or Blacksmith'.

On the day on which this lease was registered, 23 March 1722/3, Queensberry's man of business, Alexander Burne, wrote from England to a lawyer in Edinburgh, to say that money was needed: 'There will be no room for delay of payments, especially as this plaguy new house is a devouring article.' (fn. 20) In the Westminster Commissioners of Sewers ratebook for 1723 Witt appears as the ratepayer but no rate was stated as the house was said to be 'not finished'. (fn. 21) In the parish poor ratebook for the same year Queensberry's name appears but again no rate is stated, the house being first rated in the following year. (fn. 22) Burne's comments on outlay for building continue for some years; in December 1725 £6000 was owed to tradesmen and a large sum to Burne himself, who wrote: 'The occasion of this extraordinary expense is owing to a spirit of building which at present governs us'. (fn. 20) Work was in hand on the Queensberrys' country seat at Amesbury in Wiltshire about this time and must have occasioned some of this outlay. In December 1726 Burne reported that the Queensberrys were 'still in the country, purely to avoid the clamorous demands of Tradesmen'. (fn. 23) In May of the following year, however, Burne could hope that the situation would improve as 'our building is over'. The records of Burne's bank account are preserved, and contain payments clearly made on Queensberry's behalf. There are payments from 1725 onwards to the Burlingtonian architect Henry Flitcroft (as there are in Queensberry's own account at Drummonds Bank from 1730 onwards), but they are probably for work at Amesbury. There are also payments to Christopher Horsenaile, presumably one of the masons of that name, and to Robert Sparke, perhaps the brazier. Payments were made in 1726–7 and later to James Slater, possibly the plumber who was concerned in the first lease of a house in Savile Row. (fn. 24)

Architectural Description of Queensberry House

It may be observed of Leoni's disposition of the house that although there is some evidence that in 1719 the site was intended to face Old Burlington Street and he speaks of Burlington's grant of permission for the front to face south when the design was laid before him, the 'privilege' so conceded was essential to the execution of Leoni's plan. This, while making the best of the site, presented only a side elevation and court-yard wall towards the principal street of the estate. The front looked on to the garden wall of Burlington House. This met with the disapproval of Ralph, writing in 1734, when he could 'find no other fault with the Duke of Queensborough's house, but that 'tis badly situated, overagainst a dead wall, and in a lane that is unworthy of so grand a building: to which we may add, that it wants wings, and must ever do so, because there is not room to make so necessary and graceful an addition'. (fn. 25)

The site was roughly a rectangle fronting some 104 feet to the street on the west (now Old Burlington Street) and 70 feet 7 inches to 'Vigo Passage' on the south (now Burlington Gardens). The house, a double pile measuring externally 70 feet east to west and 53 feet north to south, was placed to front the south boundary of the site, leaving space for a large court-yard to the north (Plate 76). It will be noted that Leoni shows no provision for stables or coach-houses on the site. Beneath the court-yard the great groin-vaulted kitchen and other offices were placed, lit and ventilated by a long and narrow area against the east boundary of the site. The cellar or basement of the house contained a smaller kitchen with offices and servants' rooms, and this accommodation was supplemented by a 20 feet-deep range beyond the south front area, containing a large groin-vaulted room, perhaps a brewhouse, some cellars, and a three-seat privy.

The plans show that the house was solidly built, a substantial wall dividing the front and back ranges which were, respectively, 22 and 21 feet deep. In front, on the ground or parlour floor, was an east room 28 feet wide, with three windows, a west room 18 feet wide, with two windows, and between them an entrance hall 17 feet wide, with a window on the west side of the doorway. The north range was correspondingly divided, with a 27 feet-wide room to the east of the two-storeyed staircase compartment, immediately north of the entrance hall. The northwest compartment, however, was subdivided to accommodate a secondary staircase, placed between a small closet and a lobby which led to the 9 feet-wide west room. This arrangement was repeated on the first floor except that the largest room on the south front was in the middle, with three windows. The secondary stair continued to the second floor, where the principal room was the gallery at the east end, 47 feet long and 17 feet wide, with two windows between two fireplaces in the long east wall, and two windows in each end wall. A corridor reduced the depth of the three south rooms to 15 feet 6 inches; two of these were 18 feet wide, and the third, a closet next the gallery, was 9 feet 6 inches wide. The north rooms were all 21 feet deep, the largest one was 18 feet wide, being above the great staircase, one adjoining the gallery was 8 feet wide, and one to the west of the secondary stair was 9 feet wide. There is no plan of the garret storey, but since Leoni's elevation shows no windows in the roof it is probable that the rooms were lit by dormers looking out on to a valley between two parallel roofs.

The principal front to the south was three storeys high and seven windows wide, all equally spaced (Plate 77a). A painting in the Marquess of Anglesey's possession and now at Plas Newydd (Plate 75a) confirms Leoni's complaint that the ground storey was given an ordinary brick finish, and his engraving shows that he intended it to be of smooth-faced masonry with rustic jointing. The window openings, each a double square, were to be plain with voussoired heads, but the central doorway was to be set in an arch dressed with longand-short rustics, plain imposts, and a maskheaded keystone. The ground storey, which finished with a plain stone bandcourse, was unbroken according to the painting, but Leoni intended the face embracing the five middle bays to project sufficiently to sustain the Composite order rising through the two upper storeys. This order was of stone and comprised six plain-shafted pilasters, placed on pedestals without bases but with moulded cappings, and supporting an entablature having a moulded architrave, plain frieze, and a modillioned cornice. The wall face was of brick, but stone was used to dress the windows of the principal and chamber storeys, and for the pedestals below the former. Although the cappings of these pedestals lined up with those of the pilaster pedestals they had moulded bases which reduced the height of the dies. The principal-storey windows, again double square, were all dressed with moulded architraves, broken out and returned at the feet, but only those in the five pilastered bays were finished with pulvino-friezes and cornices. The chamber-storey windows were square, and all were completely framed with moulded architraves. The crowning entablature, which was returned and continued across each end bay of the front, was intended to be finished with a blocking-course, broken forward over each pilaster to form pedestals for statues (presumably of lead). For this the builder substituted an atticpedestal, with projecting dies of stone over the pilasters and plain brick dies between.

The Plas Newydd painting shows the east side wall as a brick face without adornment save for the narrow return ends of the stone bandcourse, pedestal and entablature of the south front. The west side probably survives as part of the present frontage to Old Burlington Street. In this the plain window-openings are set in a brick face, simply dressed with a stone plinth and finishing bandcourse to the ground storey, a continued sill below the first-floor windows, and a frieze-band and shallow cornice continuing the lines of the modillioned cornice crowning the south front. The same simple treatment probably served for the north front towards the court-yard. The plans show that this front had the same disposition of windows as the south elevation, and suggest that the middle opening of the ground storey, a doorway to the staircase hall, was dressed with a doorcase having engaged columns.

Leoni's statement that the court-yard gate was designed by Lord Burlington is confirmed by two drawings in the Burlington-Devonshire collection. (fn. 9) The first is a sketch elevation inscribed in Burlington's hand 'Gate for the back court', and the second is Flitcroft's finished version for engraving, with the verso inscribed 'for the D. of Queensbury' (Plate 77b). The design is for an arched opening constructed of long-and-short vermiculated stones, with a plain impost carried across the opening to form a lintel below a tympanum carved with a garlanded cartouche. The brickwork of the high wall in which the arch is set has no dressing except for the continued stonework of the plinth, impost and cornice-coping, the last being surmounted by three stone balls above the piers and centre of the gateway.

The building agreement of 1722 between Queensberry and Witt, already referred to, provides enough evidence to show that the interior was well finished, though not elaborately decorated. Except for the front hall and principal staircase, the rooms on the ground and first floors were floored with 'clean' deal boards, dowelled together. The walls were fully lined with yellow deal wainscot, the panels 'raised with a Bead upon the rising' and the framing moulded with 'a quarter round and Oger' (ogee), all the panelling being painted 'in the usual Common Colours'. The doors and window-shutters were of oak ('right Wainscot') and presumably unpainted, the former having brass furniture and the latter having iron. The sashes were stoutly constructed of 'two inch Stuff glazed with Crown Glass'. Every fireplace was to have a marble chimneypiece 'as the said Duke shall direct', not costing more than £15 in price. In the 'Great parlour' on the ground floor, and in the middle front room on the first floor, the walls were to be finished with an 'Architrave and Modillion Cornish of ye Corinthian Order' executed in plaster, but generally the rooms had 'an Architrave and a Coving … with a Moulding in the Ceiling', which in every room was to be floated. The entrance hall was 'to be done in such manner as the Earl of Burlington's front Hall is already done', and the stair compartment was 'to be all plaistered for painting' presumably by a history painter. Both the hall and the stair compartment were to have 'Portland paving with black spots', and the grand stairs, constituting a feature of Leoni's house which still survives in altered form (Plate 79b), were to be 'six foot going with portland Astrigal Steps, and iron Rails'. The back stairs were to be 'wainscotted rail high with plain bead work', and the passages leading thereto were to be 'wainscotted as high as the top of the doorheads with the same work'. All the second-floor rooms, except for the gallery, had floors of 'second-best boards, straight joints and nailed' and panelling of 'plain bead deal wainscot', the doors and windowshutters being of the same material. The gallery (called the 'Long Roome' or 'great Roome'), however, had floors of 'clean Deals dowled' and was 'to be done with plain bead and raised pannells'. All the fireplaces on this floor were to have 'flat Chimney pieces of white and veined Marble'. The garrets were to be finished in the simplest manner and so were the servants' rooms in the basement. These last, however, were to have floors of yellow deal boards laid on oak joists, the remainder of the basement being paved with Purbeck slabs. The offices were to have Portland stone chimneypieces, and a lead pump was to be fixed in the middle room.

Later history of Queensberry House

In July 1732, at the period when the Pollen family was disposing of some of its freeholds, Queensberry purchased, subject to the Burlington leasehold interest expiring in 1809, the freehold of the site and also that of the ground to the east abutting on Savile Street (now Savile Row) which was then being laid out. This additional site extended twenty-one feet north, behind No. 2 Old Burlington Street. It also extended some fifteen feet south of the building-line of the house. (fn. 26) Burlington made a lease to Queensberry of the same eastward site (which the latter had by then walled round) in March 1736/7 for 44 years to expire, like the lease of the house site, in 1780. (fn. 27) This excluded the south-east corner which was 'made circular for the more convenient passing of Coaches and other Carriages to and from Savile Street'. Burlington exacted certain covenants evidently common to this side of Savile Row. No building was to over-top the fourteen-feet-high wall, and Queensberry was not to put up any 'Stall Bulk Shop Shed Shewboard or other Erection whatsoever of that nature' on the Savile Row frontage. Burlington also required that if Queensberry wished to make a doorway into Savile Row he should 'accept a Model or Design thereof from the said Earl … and shall … erect and finish such Door Way with Bath Portland or other sort of Stone exactly conformable and agreeable in every respect [to] such Model or design and not otherwise'. Whether a gateway of Burlington's designing was made here as well as in Old Burlington Street is not known.

Queensberry later secured his tenure of the combined sites from Michaelmas 1780 to the coming into effect of his freehold in 1809 by a lease for the additional term, at £460 per annum, obtained from Burlington's heir the Duke of Devonshire in June 1775. This repeated the covenant regarding the height of building on the Savile Row frontage but not that respecting Burlington's design for a gateway. (fn. 28)

The vivacious life of Queensberry House during its long occupation by the Duke and Duchess cannot be described here. We are taken inside the house for a moment on a day in March 1729 when the Queensberrys' protege Gay was writing to Swift from 'the room next to our dining-room', surrounded by the sheets of his opera Polly which 'two people from the binder' were busy folding. (fn. 29) In this house Gay died, in December 1732. Some years later, in 1748, the Duchess, 'opening her house to the town with several balls and masquerades', had a stage erected where the noble family of Argyll performed several plays. (fn. 30)

The Duke died in October 1778, a little over a year after his wife, (fn. 2) and by his will (fn. 31) left the house to his kinsman, the Earl of March, henceforward fourth Duke of Queensberry and subsequently famous in the history of gossip as 'Old Q'. (fn. 2) The following month a neighbour on the opposite side of Old Burlington Street, Mary Townshend, wrote to a friend of the fourth Duke, foretelling that he would prefer to stay at his house in Piccadilly and fearing that if he did move to Burlington Gardens 'there would be brick and mortar without end' immediately opposite her. (fn. 32) In fact the Duke showed no interest in the house, but 'brick and mortar' was not very long postponed.

Uxbridge House

From 1780 at latest, when the Duke purchased his villa at Richmond, Queensberry House stood empty. (fn. 22) But by the spring of 1785 it had passed, presumably on lease, to Henry, Lord Paget, who had in the previous year been created Earl of Uxbridge. (fn. 2) No record of this transaction is known, but in April 1785 Lord Uxbridge had sufficient security of tenure to have estimates made for the alteration and enlargement of the house, (fn. 33) which was begun by the summer. Apart from the sale of the site (the leasehold in being and the freehold in reversion) by Queensberry to Lord Uxbridge on 19–20 July 1790, (fn. 34) by which time the rebuilding was substantially completed, nothing is subsequently known of any connexion of Queensberry with the house. Lord Uxbridge was rated for Uxbridge House, as it was henceforward known, from 1787; the structural work is reflected in a rise of the rateable value from £250 in 1785 to £600 in 1788. (fn. 35) The commencement of the alterations is shown in the Marquess of Anglesey's painting at Plas Newydd (Plate 75a). The architect was John Vardy, doubtless the son of the architect of Spencer House. (fn. 36)

When writing his account of the house for The Public Buildings of London (fn. 37) J. B. Papworth mistakenly assumed that a complete rebuilding had taken place, and he criticized the plans and elevations accordingly. In this he was most unfair to Vardy, who adapted and extended Leoni's building with considerable skill, as a comparison of the 1726 and 1825 engraved plans and elevations will show (Plates 76 77, fig. 79). The estimate of April 1785 had been for 'Repairs, Alterations, and Additions to Queensberry House', and it allowed for a complete refurbishing of the existing house with the minimum of structural alteration, the building of an extension some 30 feet eastwards up to the frontage of Savile Row, and the addition of a north wing over part of the court-yard, fronting to Old Burlington Street and presumably utilizing the existing basement.

Vardy removed the entrance from the south front to the new wing in Old Burlington Street, where a three-bay porch marked the position of the new hall, a deep oblong apartment. On its north side were rooms for the porter and groom, and on the south were doors opening to Lord Uxbridge's private suite. A wide doorway at the east end of the hall led to an ante and thence, through the old north doorway of the house, into the great staircase. Retaining the old stone stair, which was 'cleaned and repaired', Vardy heightened the compartment in order to obtain toplighting. To enlarge the room on the west side of the great stair, the old secondary stair was demolished and a 'new back staircase of Portland stone' was built on the east side of the great stair. The old front hall became a waiting-room, serving Lord Uxbridge's suite on the west, and the library on the east. Beyond the library, in the new extension, was a large dining-room with a screened recess at its north end, and behind this was the servants' staircase.

On the first floor an open screen replaced the wall between the front and back west rooms, which now became a double ante-room serving the great music room in the north wing and the drawing-rooms on the south front. The estimate makes it clear that the two drawing-rooms shown on Leoni's plan had already been united to form one great room, seven windows wide, but to enable this to be used en suite with the west ante-room and the new east drawing-room, the fireplaces in the side walls had to make way for centrally placed doorways, and a new fireplace was formed in the middle of the long wall opposite the windows.

The estimate proposed improving the south front, somewhat on the lines of Leoni's frustrated intention, by adding 'a rustic arched Basement of Portland Stone', this being the present arcaded face of the ground storey, and by substituting a 'Stone Blocking' for the old brick and stone parapet. It also allowed for 'the Windows of the principal Floor to be cut down within Six inches of the Floor' and, presumably, for the stone balcony and iron railing to be added. The estimate, however, makes no mention of replacing the brick facing with ashlar work, in fact it specified for 'the Brickwork and Stone to be cleaned and repaired', and for the 'New Building next Saville Row' it intended 'the front to correspond with the old one … rustic Arched Basement, Composite Pillasters and Dress to Windows of Portland Stone'. In the surviving documents relating to the work there is nothing to show when or why this expensive change was made, nor, indeed, any mention of mason's work, but the refacing was probably carried out under Vardy's direction. A water-colour drawing by T. Malton, presumably made some time in advance of the aquatint view dated 1800, shows the front to be of one material, presumably stone as some joints are indicated (Plate 75b). Moreover, the engraved elevation in The Public Buildings of London (Plate 77c) shows jointing throughout the upper face, and the accompanying description states that 'The south elevation is of the Composite order, consisting of nine pilasters, supported by a rustic arched basement, and executed in Portland stone. The remainder of the building is of brick work, with stone cornices and accompaniments'. Papworth concludes with the statement that Vardy was 'assisted in the disposition of the south front by the late Mr. Joseph Benomi [sic], the Architect'. (fn. 37) Bonomi's name does not, however, figure in such records of the rebuilding as survive. It may well be that this was because he acted solely for or under Vardy who alone dealt with the owner.

Vardy's name is, however, associated with that of another artist in the letters of Lord Uxbridge's agent, who constantly speaks of his consultations with 'Vardy and Linnell'. The latter was John Linnell of Berkeley Square, the eminent and versatile cabinet-maker and furniture designer. (fn. 38) He here seems to have been so closely connected with the direction of the work as to suggest that, as well as designing specific furnishings, including steel grates (fn. 39) and an organ case, (fn. 40) he was acting more generally as 'interior decorator'. He was described by the architect C. H. Tatham as 'in the first line of his Profession,' and was doubtless expensive. By September 1790 Lord Uxbridge was lamenting that 'that fellow is certainly making a property of me and will never let me out of his books if he can help it—I heartily wish he cou'd be finally discharged.' (fn. 41) But Linnell's bills came relentlessly in, and by June 1793 Lord Uxbridge had paid him at least £5894. (fn. 42)

Some of the other workmen's names are known. No mason's or carpenter's name occurs, but the main bricklayers were the partners Thomas Poynder and Edward Wix. (fn. 43) The latter was dead by January 1793 when Poynder wrote menacingly from Bishopsgate for payment of his bill. (fn. 44) The smith was James Messenger (fn. 45) and the plumber Lancelot Burton of Newcastle Street, Strand. (fn. 46) The painter, who paid a tribute to the excellent work of the unidentified joiner, (fn. 47) was Jonathan Sadler of Lothbury. (fn. 48) Difficulty arose over the plasterer's work. The first to be employed was named Pritchard; by September 1786 he had been dismissed and his place taken by the well known craftsman Joseph Rose. Disagreements with him led to a lawsuit which was evidently determined in the plasterer's favour. (fn. 49)

Chimneypieces were provided by Richard Westmacott, doubtless the elder statuary of that name (d. 1808) and Vardy's brother-in-law. (fn. 50) Payments to him included some £360 in March 1790 (fn. 42) and £147 in March 1793. (fn. 51) He was perhaps responsible for the ball-room chimneypiece which is now, somewhat altered, in the ante-room to the court room in the Bank of England. (fn. 52) The Builder, when recording the removal of this 'magnificent mantelpiece of white alabaster' by the Bank, which then occupied the house as its West End branch, stated that customers had offered up to a thousand guineas for it. (fn. 53)

Joseph Bramah provided 'Water Closets etc', (fn. 49) and other aspects of the furnishing and fitting-up of the house are probably represented by the names Duesbury, Messrs. Gillow, Messrs. King & Co., Martyr, Joseph Spode, Thomas Waring, and Wedgwood in Lord Uxbridge's bank account.

The commencement of work had followed promptly upon the estimate of April 1785: smith's work began in July (fn. 45) and the bricklayers' work had begun at some time in the same year. (fn. 43) Letters in the Marquess of Anglesey's possession at Plas Newydd, written to Lord Uxbridge by his self-important and rather unsatisfactory agent, Thomas Harrison, describe the alterations. His communications alternated between unspecified forebodings of trouble and unsubstantiated assurances that all could confidently be left in his hands, but they at least suffice to indicate the progress of the work. In August 1786 he reported that Vardy and Linnell 'seem to be going on with Spirit [and] Dispatch'. Evidently the structural work was then already finished as 'the Scaffolding in the Front' was about to be taken away. (fn. 54) By the following month work was proceeding inside, starting at the top, where the garret story was 'pretty well completed.' (fn. 55) Later in the autumn Harrison was reporting 'laudable Bustle'. Upward of £8000 had already been spent and he prophesied that the total cost, including furniture, would not be less than £25,000: 'It is a serious sum and has staggered me, but, My Lord, there is no stopping, no looking back, & it must be weathered through'. (fn. 56) By this time Lord Uxbridge was evidently becoming alarmed at his own expensive proclivities, and by January 1787 had suggested a saving of £5000 by leaving part of the house unfinished: (fn. 55) perhaps the belated intention to face all the south front in stone had been carried out and its cost occasioned this dismay. How far the saving now suggested by Lord Uxbridge was effected is not clear: some deceleration of work perhaps followed and painter's and plasterer's work was still in progress in September ot that year. Lord Uxbridge seems then to have been at once urging expedition and already talking of a sale of the house. Harrison himself felt some trepidation but replied roundly, 'what purpose can possibly be answered by the Idea of selling does not, I own, occur to me. The more I see of it the more its Magnitude & the Greatness of the Design strikes me, and of consequence, the more I dread the Day of reckoning. For great it is, and tho' there are some things, of inferior moment, which a nice eye may discern not so well executed, the principal and main parts are executed in the most masterly stile both as to Substance and Beauty.' (fn. 57) Work was still going on in February 1788 (fn. 58) but by May the smith's, plumber's and plasterer's work was finished, (fn. 59) and the bricklayers' was concluded some time in the same year. (fn. 43) By that year the assessment of the rateable value, which had been progressively increased during the alterations, had reached its final figure. (fn. 22) In October Harrison still considered it necessary to strengthen his master against vain afterthoughts. 'Your Lordship's Reflexions upon the London House … should not be indulged but resisted. They are too late to produce any Effects but such as will do harm'; (fn. 58) and by March 1789 Lord Uxbridge was in his new house. (fn. 60) An organ was installed in that year, (fn. 61) doubtless that for which a design by Linnell is preserved in the Victoria and Albert Museum. (fn. 62) As has been said, the outright purchase of the house from the Duke of Queensberry followed in July 1790.

In 1791 some bricklayer's repairs were executed. (fn. 63) The 'finishing' of the interior seems to have been leisurely. In August 1792 the painter reported himself 'very forward with respect to Beautifying the several rooms in Uxbridge House', where he was finishing work on Lord Uxbridge's dressing-room, the library and the breakfast-room. (fn. 64)

Little of the interior as altered for Lord Uxbridge survives as direct evidence of the character of this work. The estimate of 1785 had included the replacement of Leoni's wainscot by paper or hangings, and records of interior painting in 1796 demonstrate a taste for much 'dead white'; (fn. 65) the estimate had also included stuccoing the dining parlour pea-green or French grey.

As has been indicated, the payment for the work produced difficulties. In September 1788 Harrison had had to tell of 'no little Squabbling' between Vardy, Linnell and Westmacott. Vardy was 'making a very firm & probably a very proper stand' against charges he thought exorbitant. (fn. 53) The receipted bills of workmen testify that Vardy had disallowed a proportion of most of their claims. (fn. 66) This resistance seems to have collapsed, however, perhaps because of the judgment favourable to the plasterer Rose in his lawsuit against Lord Uxbridge in the autumn of 1789, (fn. 49) and the same workmen's receipted bills show that the sums disallowed by Vardy were in fact paid. But dunning letters from tradesmen continued until 1793, by which time most of the debts were probably discharged.

It is not certain what the total cost was, or whether it much exceeded Harrison's estimate of at least £25,000. By January 1789 when the main work was drawing to a close the debts incurred amounted to £19,000, not including the sums owed on the bricklayers', plasterer's and coppersmith's bills: (fn. 67) the first two of these latter probably came to nearly £6500. (fn. 68) Additional to this were Vardy's fees as architect in charge, and the internal painting and finishing remained to be paid for.

In May 1789 Lord Uxbridge had taken a lease of a stable-yard off Swallow Street. (fn. 69) It is not clear whether he also had stabling erected behind his enlarged house, where Horwood's map of 1792 shows an opening on to Savile Row between gateway lodges. (fn. 1)

The Paget family had its town residence in Uxbridge House tor some sixty-five years, under the ownership successively of Lord Uxbridge, who died in March 1812, and his son, commander of the cavalry at Waterloo where he lost a leg and earned his advancement to the Marquessate of Anglesey in July 1815. The extravagance of Lord Uxbridge had made the upkeep of the house a matter of some difficulty and in 1805 Lady Uxbridge was thinking gloomily of a sale. (fn. 70) The situation does not seem to have become easier under the liberal instincts of the Marquess whose hospitality here was on the pattern of older and less urban times. (fn. 71) In 1816 the house was mortgaged for £16,000 to the Duke of Beaufort and Lord Charles Fitzroy and in 1821 to the Marquess's bankers, Andrew and John Drummond, for £10,000. (fn. 72) In 1830 and again in 1833 there were renewed thoughts of a sale, (fn. 73) and in 1844 another £12,500 was raised when the mortgage was assigned to the Earl of Essex. (fn. 74)

The Marquess of Anglesey died in the house in April 1854, and in the following year it was sold. At this time the West End pattern of private houses and retail shops was just beginning to be modified, with a few large private houses passing into the hands of the Government, insurance offices or banks. It was the Bank of England which now took Uxbridge House for conversion as its West End branch. The sale was not made direct by the Paget family to the Bank. On 19 July 1855 the trustees under the Marquess's will, together with the Earl of Essex as mortgagee, conveyed the house to a lawyer, N. W. J. Strode, of Albany, esquire, for £31,250 (of which £8750 was paid to the trustees and the rest to Lord Essex). (fn. 75) On 28 July The Builder announced the sale of the house to the Bank of England 'for £45000 or £47000.' (fn. 76) A conveyance by Strode to the Bank was in fact executed on 11 August, for a purchase price of £42,500. (fn. 77)

The necessary work of conversion, including the construction of a wide stone Doric portico in the centre of the Burlington Gardens front (Plate 78a), and the conversion of the dining-room into the 'principal office', was rapidly carried out by Messrs. William Cubitt and Company in the months of August and September, from designs of Philip Hardwick, and the Bank was opened on 1 October. Some work then remained to be completed, perhaps in the upper part which, it was announced, 'will be appropriated to the residences of Captain Tindal, R.N., the manager, and Mr. Miller, the sub-manager'. (fn. 78)

Probably at this time, and certainly by 1869, Vardy's entrance porch in Old Burlington Street was moved to a more southerly position, as is shown on the 1869–70 Ordnance Survey map: it was restored to its original and present position in 1934. (fn. 79) In 1878 'extensive alterations', projected some two or three years previously, were carried out under the direction of P. C. Hardwick. The Builder reported that the work would comprise the conversion of the upper rooms into offices, and would cost about £12,000. The banking hall, however, was enlarged also and the total cost considerably exceeded the sum stated by The Builder. (fn. 80)

The Bank had previously, in December 1876, bought the adjacent house, No. 2 Old Burlington Street. (fn. 81)

The Bank of England occupied the former Uxbridge House until 1930 when it was taken as the Western Branch of the Royal Bank of Scotland, (fn. 82) to whom the Bank of England sold the freehold, together with that of No. 2 Old Burlington Street, in 1933. (fn. 83)

Architectural Description of Uxbridge House

Reconstruction of the interior, to provide a spacious two-storeyed banking hall, has left only the western part of Uxbridge House reasonably intact. This, fortunately, includes the great staircase, the north and south ante-rooms, and a part of the great drawing-room, all on the first floor.

The staircase rises in three flights against the west, north and east walls of the oblong compartment, to a gallery landing on the south side (Plate 79). The steps and landings are of stone, the steps being finished with bracket-profiled soffits, and the landings panelled on the underside. The handsome wrought-iron railing, possibly Leoni's plain original work with added enrichments, is composed of alternate standards, one consisting of two vertical bars joined by a series of small circles, the other a highly wrought standard with lyreshaped scrollwork at the foot, scrolls and an anthemion in the middle, and a smaller lyre-shape or foliage at the top. The handrail is finely veneered with mahogany. The compartment walls are plain below the wide architrave that marks the first-floor level, but in the principal stage each wall contains a shallow recess, arch-headed, between wide piers. The north and south arches are semi-circular, and the wider east and west arches are segmental, but all have moulded archivolts with one fluted fascia. These archivolts rise from an enriched impost, featuring a flowered guilloche, the impost being returned inside and across each arched recess. Raised mouldings frame a panel on each pier and there are smaller panels of oblong form above the impost. This stage of the compartment is finished with an enriched modillion cornice, above which rise the pendentives of the oval dome, framing a panelled tympanum of segmental form on each wall. Each pendentive is decorated with a panel containing one of the 'Trophies with Shields, Fasses—Leaves etc.' for which Joseph Rose charged £29 10s. in all. Round the base of the dome runs a cornice of two members, the most prominent decorated with raffle leaves and water leaves in alternation. The surface of the dome is divided into three rings of quadrangular coffers, eighteen in each ring, by ribs decorated with Rose's 'Diminished Goloss'. To frame the eye, Rose used 'Circular Reed Moulding with 36 Leaves laid on and 18 double Ribbon Ties', and the low drum of the skylight is divided by '18 Trusses with Cups and 3 Flutes in each', with festooned garlands of 'Husks and Berries' between the trusses.

The door on the west side of the staircase leads to the oblong ante-room, now the Bank's committee room. In the east wall is an arch-headed recess, originally open to the staircase, framed by an enriched archivolt resting on pilasters with panelled shafts. In the west wall, centred between two windows, is the fireplace, with a richly carved marble chimneypiece, possibly lacking its plinths. The architrave to the opening is carved with oakleaf, and the jambs have panels filled with a double wave-scroll guilloche below large acanthus leaf brackets. These support the cornice-shelf, which has a deep fascia carved with a wave-scroll. The large doorcase in the south wall, framing a twoleaf door opening to the south-west room, is designed to accord with the chimneypiece but may be later. Rose charged £47 7s. 4½d. for the ceiling 'with frames, foliage, garlands, flowers, etc' arranged in compartments divided by ribs ornamented with guilloches (Plate 78b). The large central compartment, an elongated octagon, contains a chandelier-boss within an oval ring of interlacing garlands. In each angle compartment is a lyre flanked by foliage scrolls and surmounted by an anthemion, and a similar motif, without the lyre, is placed in each side panel.

The south-west ante-room has now been united with the truncated great drawing-room to form a large office of two compartments. The ceiling in the west compartment is probably Rose's original work, for which he charged £68, and that in the east compartment is probably an altered portion of a ceiling costing £188 16s. 10d. The west ceiling is a composition of three oblong compartments, that at each end containing a lozenge flanked by crossed branches. In the middle is a circle with a surround of formal leaves radiating to fill an oval frame.

The most elaborately decorated room was the new 'Great Room' for music, added by Vardy. Rose's account suggests that the walls were divided into bays by twenty pilasters with Corinthian capitals '23½ inches high', supporting an entablature having an enriched architrave, a frieze of 'rich double foliage' and a 'modillion cornice with 7 members carved'. The 'various ornaments to the level part of the ceiling' cost £214 12s. 2d., and the cove ornaments cost £232 6s. 2½d.