X Close

The Survey of London


Recording the history of London's built environment since 1894


Park Crescent: John Nash, Charles Mayor and one of the severest bankruptcies in London building history

By the Survey of London, on 25 May 2018

The Park Crescent of today is a post-war replica. A combination of poor original building methods, wartime damage and heavy-handed reconstruction policies has left little or no old fabric from the ‘semi-circus’ conceived by John Nash and built with much difficulty between about 1812 and 1822. Even that was only half of the full circus Nash had planned. Yet despite those early failures, despite also its present lack of authenticity, Park Crescent remains one of London’s most memorable episodes of urban planning.

Park Crescent. Eastern crescent. (© Historic England, Chris Redgrave)

Three colonnaded circuses featured in Nash’s immortal scheme to link Whitehall and Westminster with Marylebone Park by way of what became Regent Street, as first published in 1811. All occupied critical junctures or points of transition, where traffic-ridden east-west routes broke the new line of procession, for Nash believed that the circus form invested such crossings with dignity and lessened the psychological sense of a barrier – something he strove to overcome so as to make the Crown’s developments around Regent’s Park feel accessible and eligible to Londoners of high standing. Confusingly, all three rond-points went at first by the informal name of Regent or Regent’s Circus. Piccadilly Circus and Oxford Circus, carved laboriously out of existing building fabric by compulsory purchase under the New Street Act of 1813, had to be quite small and in the event lacked the colonnades Nash had hoped for. Both have long gone, and though Oxford Circus survives as a concept, it is higher and slightly wider than Nash’s original creation.

The third circus was to be different altogether. Over 500ft in diameter, it came at the point where the city eased into a gracious modern suburb, beyond the final obstacle of the New (now Marylebone) Road. But the shape and texture of that suburb shifted greatly over the years of its creation, with implications for the third circus. So much can be gathered from the fate of a fourth circus, the largest of all, devised by Nash as a central focus for Regent’s Park. Marked the Great Circus on the plan of 1811, when bands of continuous housing were anticipated within the park as well as on its periphery, it was then reserved for concentric terraces. Ultimately, it acquired a bare smattering of villas and became the Inner Circle, a low-key feature in the park landscape. With the opening-out of that landscape in the 1820s and the failure to develop the northern sector of the third circus, Park Crescent turned into a prelude to the park rather than the transitional interlude Nash intended.

Park Crescent and Portland Place area from the west in 2011. (© Historic England, Damian Grady)

If the circus was the most striking of the crop in town-planning terms, it was also among the first elements of the Office of Woods and Forests’ plan for Regent Street and Regent’s Park to be implemented. Its site, Crown farmland, was in hand, so that it could be started even before the New Street Act authorised the complete scheme in 1813. Not surprisingly, Nash invested energy and urgency in getting it along. By the end of 1811 the hunt was on for a single builder brave and resourceful enough to take on both the whole of the circus, where Nash envisaged large first-rate houses, and the extension of Harley Street, for which he proposed brick-faced houses on narrower frontages. By March 1812 he had hooked his fish. ‘Mr Mayor is willing to adopt the elevation I had proposed to him, which is to encircle the whole with a collonnade of coupled columns surrounded [sic] by a ballustrade’. [1] Charles Mayor had started out as a jobbing carpenter and undertaker. In 1800 he successfully took over the north side of Brunswick Square from James Burton and built other houses near by. More to the point, he and Nash had been collaborating over a house in Foley Gardens south of Portland Place.

By May 1812 Nash and Mayor had worked out a detailed plan and schedule. Together they increased the diameter of the circus to 724ft, giving some houses frontages of up to 100ft but in consequence making them shallower and cramping the mews spaces behind. The timetable stipulated the first roofings-in and issuing of leases at the southern end to be by August 1814 and the final ones in the northern sector by August 1816. Nash projected Mayor’s overall outlay at about £300,000 and urged the Crown authorities to buy the improved ground rents so as to guarantee his liquidity. Mayor got the final go-ahead in July. Soon afterwards Nash was arranging to show his whole plan to the Prince Regent, telling Alexander Milne of the Woods and Forests: ‘It will be very impolitic not to pursue this course if we wish HRH to take up the measure con amore’. [2]

92–98 Portland Place. Original plan of houses as leased to Charles Mayor in 1813. From plans in the National Archives, CRES2/752. (Survey of London, Helen Jones)

Mayor got going fast enough to ask for his first leases between December 1812 and August 1813. These concerned the southernmost houses on both sides of Regent’s Circus and also some smaller brick-fronted ones that were to connect Portland Place with the circus (Nos 77–81 on the west and 92–96 on the east). Though these leases were duly issued, little finishing work can have taken place within the carcasses. For financial reasons Mayor evidently hoped to secure as many leases as he could as fast as might be. To that end he appears to have embarked upon the cellars of almost the whole of the southern semi-circus and even to have done some excavation on the northern side of the New Road.

Then things started to unravel. With the Napoleonic Wars reaching their climax, it was a bad time for builders. Few people were in the market for houses, yet building materials were going up in price. Mayor seems to have done nothing elaborately tricky: with Nash egging him on, he just miscalculated and was drawn into one of the severest bankruptcies in London building history. His slow-down during 1813 must have been well-noted. ‘Every stimulus should be applied to Mr Mayor to complete his Circus,’ urged Nash in a report to the Commissioners on changes planned for the park that December, ‘which will carry that character of respectable building into the heart of the Park and tempt the higher classes of society to come there’. [3]

Park Crescent. Western crescent. (© Historic England, Chris Redgrave)

The first half of 1814 saw little progress apart from an intended pub built by James Smith of Norton Street on back land between Harley Street and the circus. In the autumn Mayor appealed to the Woods and Forests for a loan, admitting: ‘I am absolutely at a Stand Still’. Nash, asked for advice, reported just before Christmas with his usual optimism that with eleven houses covered in, another six up to the chamber floor, excavation in some state of forwardness for the whole of the circus and a million and a half bricks on site, the developer ought to have enough security to win through – ‘if the half circus can be completed in the course of the next year (I mean externally)… his speculation will have a favourable issue’. [4] With the Crown declining to lend, vary terms or purchase ground rents, the creditors deemed otherwise. Mayor was carted off to the Fleet Prison early in 1815 and the grindstones of bankruptcy began to turn.

Like other bankrupts, Mayor was allowed out on day release, and made efforts to realize his assets. He managed to let the Foley Gardens house, but a severe injury, incurred while inspecting one of the Regent’s Circus houses in February 1816, cannot have helped. One bankruptcy commission superseded another. So persistent was the post-war slump that Mayor’s assignees were wary of trying to finish the speculation themselves and disappointed when they tried to sell off assets. Meanwhile the carcasses in sundry states of scruffiness and dereliction scared off buyers.

Park Crescent and the New Road, looking west towards Marylebone parish church. From Ackermann’s Repository of Arts, 1822. (Wikimedia Commons, from ianvisits.co.uk)

John Shaw, the architect and surveyor to the Eyre Estate, claimed that it was thanks to ‘a very extraordinary man, Mr. Farquhar’, who put money in as a speculation, that building resumed at what was to become Park Crescent. This was the former East India Company’s gunpowder contractor, John Farquhar. He died in 1826 leaving property in the vicinity of Park Crescent, but no further evidence of his involvement has come to light. [5] In 1817 the Great Portland Street builder William Richardson, apprised that Mayor’s assignees were not going to complete the development themselves, sought terms for both the Harley Street continuation and for finishing the western half of the crescent, where foundations were partly laid. A bigger, City-based contractor, Henry Peto, then bought the three carcasses on the east side of Portland Place (Nos 92–96), completing them ‘in the first style of elegance’, and went on to take most of the eastern quadrant, excepting its tip, assigned to Samuel Baxter and the corner house No. 15. Peto bought the ground rent from this last from a mortgagee, but the house itself was evidently completed by Mayor’s assignees, who decided to sell it in 1820.

In the early stages Peto was hampered by the collapse of a party wall between two of the Portland Place houses during a storm in March 1818. There had been a bad fire in one of these in 1814, on which Nash blamed the fall. Peto was adamant that the houses had been built with bricks ‘mostly of the very worst description and totally unfit for use’, and laxly supervised – accusations that have haunted the Nash developments and Park Crescent in particular ever since – and commissioned an independent report to prove it. ‘I must beg to be allowed to treat the insinuations of my inattention and that of my clerks with contempt’, riposted Nash, adding that Peto ought to have noted the state of the houses when he bought them. [6]

On both sides of the crescent, finishing off Mayor’s carcasses generally preceded the building of the other houses. The western crescent went up mainly in 1819–20; at the same time Richardson undertook the east side of Upper Harley Street, essentially following the original Nash-Mayor plan. Judging from a sketch of Park Crescent in 1820, Peto had as yet made little or no progress with his side. Beyond Mayor’s houses at Nos 13–15, the eastern quadrant was just excavated ground and derelict-looking pavement vaults, while No. 13 was in a ruinous state, with bare roof timbers and fallen brickwork. Perhaps the last of the houses to be started were Baxter’s at Nos 1–6, the first two of which were still in carcass in July 1821. The whole can have been barely finished when Ackermann published his prospect of The Crescent, Portland Place in 1822.

Park Crescent. Eastern Crescent. (© Historic England, Chris Redgrave)

The name ‘Park Crescent’ is first encountered in the year 1821. By then it was abundantly clear that the half-circus north of the New Road would never go ahead. Ideas about the form of the park had radically changed, with a greater openness of landscape prevailing. Around 1820 Nash therefore scrapped the northern crescent altogether in favour of Park Square, square-sided with an open north end. The crescent furnished the park, the New Road and indeed London as a whole with a fresh and uplifting piece of unified urban scenery. Its semicircular profile, albeit only half of what had been planned, offered a decisive depth of sweep that an elliptical crescent would have lacked. On closer inspection the composition, as often with Nash, was sleek but shallow. Minimal projections and pediments at the ends do something to relieve the simple economy of the plastered elevations (originally in Roman cement, Bath-stone coloured, with the jointing perhaps sharper than it is today). But the main compositional trick is in the enfilade of the continuous ground-storey colonnade with coupled Ionic columns, a device integral to Nash’s overall New Street plan and surviving only here. The original agreement with Mayor lays some weight on the balustrades topping the parapets and colonnades, which were to be of stone, and on the iron fencing, which both fronted the deep continuous basements and ran laterally to shield the entrance bridges. As for the houses themselves, they were not the monster 100ft mansions Nash had first dreamt of, just good first-raters with fronts typically of 32ft, backs several feet wider, and ample stone staircases.


[1] The National Archives (TNA), CRES2/748.

[2] Ibid.

[3] TNA, CRES2/742.

[4] TNA, CRES2/752; B3/3365.

[5] Report from the Select Committee on Crown Leases, p. 50.

[6] TNA, CRES2/763.

‘Portland House’: Robert Adam’s unexecuted designs for the Duke of Portland’s London residence

By the Survey of London, on 7 April 2017

The Adam brothers’ celebrated street improvements at Mansfield Street and Portland Place, carried out from the 1760s on the Marylebone estate of the Dukes of Portland, are among the many significant buildings covered by the Survey of London’s forthcoming volumes on South-East Marylebone. Less well-known, however, is the detached mansion that Robert Adam designed around 1770–2 as a new London residence for the 3rd Duke, to stand on a large site on New Cavendish Street, looking down Mansfield Street. Though it was never built, its story can be pieced together from designs in the collection of Adam office drawings at Sir John Soane’s Museum – the principal resource today for anyone wishing to study the work of the Adam brothers.

William Henry Cavendish Cavendish-Bentinck, 3rd Duke of Portland. An engraving of 1785 after Sir Joshua Reynolds’ portrait of the Duke. British Museum, Prints & Drawings Dept (museum no. 1902,1011.3545) © Trustees of the British Museum

William Henry Cavendish Cavendish-Bentinck, 3rd Duke of Portland (1738–1809), had recently married Lady Dorothy Cavendish, daughter of the 4th Duke of Devonshire, and was already embarked upon a career as a statesman that would see him appointed 1st Lord of the Treasury (the equivalent of today’s prime minister) on two occasions, in 1783 and 1807–9. But although he had succeeded to his father’s title in 1762, the 3rd Duke did not immediately inherit all his estates. By the terms of his father’s and grandmother’s wills, the Duke’s mother, Lady Margaret Cavendish-Harley Bentinck, the Dowager Duchess (1715–85), retained a life interest in the family’s lucrative Cavendish lands, and she also held on to her husband’s house in Whitehall – leaving her son short of funds and without a London residence. The situation was exacerbated by strained relations between the two. They argued over country seats, in the end engineering a ‘house swap’ (she favoured Bulstrode in Buckinghamshire, he preferred Welbeck, Nottinghamshire), and failed to see eye to eye on politics as well as family finances. The Duke was a Rockingham Whig, intent on curbing what he perceived to be an increase in royal powers under George III; she numbered the king and queen among her friends, and was especially close to John Stuart, 3rd Earl of Bute, the royal favourite and prime minister in 1762–3, a man whom her son vehemently distrusted. The Duke complained to anyone who would listen that he was required to pay rent for a London house when he should have had access to the ducal residence in Whitehall. And so a new, Adam-designed townhouse at the head of Mansfield Street would suit his intended station as a leading politician and also act as a focus for his fast-improving Marylebone estate.

Portrait busts of William Bentinck, 2nd Duke of Portland, at centre, his wife Lady Margaret Cavendish-Harley at left, and Lady Mary Wortley Montagu at right, in ovals, with coats of arms below, allegorical objects between, curtains at left and above, in ornamental frame. From a drawing by Vertue after a painting by Zincke, 1739. British Museum, Prints & Drawings Dept (museum no. 1849,1031.70), © Trustees of the British Museum

In its size and scale the house that Adam designed, intended to be known as Portland House, was more like a country pile than a townhouse. In this and in certain elements of its internal planning it shared similarities with Lansdowne House, Adam’s first big private London commission, designed initially for Lord Bute but finished after his fall from favour in 1763 for his rival, another future prime minister, William Petty Fitzmaurice (1737–1805), 2nd Earl of Shelburne, later Marquess of Lansdowne.

Adam’s designs for Portland House were for a rectangular, two-storey block set within extensive grounds, with a garden to the rear and an entrance courtyard in front. The house itself would have been a fairly standard neo-Palladian affair, with seven central bays recessed behind projecting three-bay end wings. The entrance front was marked by a central portico with columns of what look like Adam’s favourite ‘Spalatro’ order – an invention of his own, based on a late-Roman capital he had seen in the Peristyle of the Emperor Diocletian’s Palace at Spalato on the Dalmatian coast (now Split, Croatia).

‘Principal Front of a House for His Grace the Duke of Portland’. Adam office elevation of c. 1770–2. Sir John Soane’s Museum, Adam vol. 29/2. Reproduced by courtesy of the Trustees of the Soane Museum

Portland House was to have had a lower ground floor given over mostly to servants’ rooms and storage, but with a gentleman’s library in a bowed room at the rear, and a bedchamber for the Duke and ‘Book room’ alongside. The principal state rooms were placed centrally on the floor above, with dressing rooms for the Duke and Duchess to either side. These connected with the little single-storey bays shown in shadow at each side of the mansion in the elevation, where there were to be powdering and retiring rooms, privies and water closets. There would have been further rooms on the first floor and in an attic within the hipped roof.

One of a pair of Adam office plans shows a proposed design for the house’s lower ground floor, with a rectangular courtyard in front, lined with coach-houses and stables on one side, kitchens, sculleries and more service buildings on the other. This plan matches the elevation, and shows how the portico served as a porte-cochère, with a curved ramp for coaches leading up to the main entrance. Also, in this plan an entrance screen wall and gateway is set quite a way back from the road, with more stabling and coach-houses in front.

Plan of the Ground Story of a House for His Grace the Duke of Portland’. Adam office design of c. 1770–2. Sir John Soane’s Museum, Adam vol. 29/4. Reproduced by courtesy of the Trustees of the Soane Museum

A first-floor plan offers an alternative arrangement, for a far more dramatic circular courtyard, surrounded by a roofed and colonnaded walkway. An accompanying section shows how this colonnade connected directly to the house, dispensing with the portico. This arrangement required further alterations to the design of the house, with windows at a higher level on the piano nobile, to allow light to enter the main rooms above the courtyard structure. Apparently this was the design chosen by the Duke.

‘Plan of the Principal Story of a House and Offices, For His Grace The Duke of Portland’. Adam office design of c. 1770–2. Sir John Soane’s Museum, Adam vol. 29/5. Reproduced by courtesy of the Trustees of the Soane Museum


‘Section through The Gateway, Circular Court and Body of the House, For His Grace The Duke of Portland, Fronting Mansfield Street’. Adam office design of c. 1770–2. Sir John Soane’s Museum, Adam vol. 29/3. Reproduced by courtesy of the Trustees of the Soane Museum

The Adams had been experimenting with colonnaded courtyards in house designs since the 1750s. Although their prime inspiration was always Italy, and in particular ancient Rome, there is a heavy debt to French plan-types, particularly Parisian hôtels, in their mansion schemes of the early 1760s. An unexecuted house of c.1764, intended for Lord Shelburne near Hyde Park Corner, was set behind a large front court, as was another design of the same date for a house for Lord Holland at The Albany, Piccadilly. Also, an early Adam brothers’ plan of around 1767 for the house they built for General Robert Clerk at the south end of Mansfield Street, facing Duchess Street, had the lower part of the house arranged in a curve and fronted by a semicircle of columns forming a carriage-way, in a similar manner to Portland House.

‘Gateway for Portland House’, Adam office design of c. 1770–2. This worked-up office version, with doors in the centre of the curved linking walls rather than windows, probably matches the rectangular courtyard plan for the house as shown in the ground-floor plan reproduced above. Sir John Soane’s Museum, Adam vol. 29/6. Reproduced by courtesy of the Trustees of the Soane Museum

Other surviving drawings include an office elevation of a screen wall and gateway to stand on New Cavendish Street in front of the courtyard, in the form of a triumphal arch, closing the vista up Mansfield Street. A second version, in pen and pencil, apparently in Robert Adam’s own hand, has detailed measurements added to it, in preparation for drawing up estimates.

Design for a gateway for Portland House. This rendition in pen and pencil, in Robert Adam’s own hand, has had measurements added to help with working out an estimated cost, as mentioned in Robert Adam’s letter of February 1772 to the Duke, quoted above. Sir John Soane’s Museum, Adam vol. 51/98. Reproduced by courtesy of the Trustees of the Soane Museum

As the Duke was both short of funds and overindulgent in his spending, a house on such a scale was evidently beyond his means. Unfortunately, the scheme also coincided with a reversal in the Adam family’s own fortunes, brought on by their attempts to develop the Adelphi and Portland Place at the same time. The Portland House project was still in hand in February 1772, when Robert Adam wrote to the Duke with a price for the ‘great gate’, porter’s lodges and some of the circular walls, and he sounded hopeful of further progress:

as Your Grace was so good as say, you would do every thing that should be necessary, to finish the end of the street, towards Your Grace’s House. I have therefore got an Estimate made of the great gate & porter’s lodges, with the circular walls that form the Entrance, & now take the liberty to send it enclosed, that Your Grace may consider it & if approved of, it will be of great Service, both to your Grace’s estate & to us, to be allowed to proceed with it this Season. [1]

But within a year the project had been dropped and the Duke was happy to let part of the site to the builder–developer John White for houses on the east side of Harley Street. The ground fronting New Cavendish Street was then leased by the Duke and the Adams to the architect-builder John Johnson, who erected the present Nos 61–63 there in 1775–6 (these will be the subject of a future blog post).

For a time around 1773–4 the Adams seem to have considered re-siting their ‘hotel’ for the Duke of Portland to the west side of Portland Place, where they had been planning at least two other very large aristocratic houses, for the Dukes of Kerry and Findlater, but this plan also failed to materialize, and the Duke of Portland when in London continued to live mostly at Burlington House, courtesy of the Duke of Devonshire. In 1807, when he was made 1st Lord of the Treasury for the second time, the Duke moved to 10 Downing Street, which was then as now the official residence of the 1st Lord of the Treasury (not the prime minister, though in modern times the same person has usually occupied both posts).

The Adams must have remained on reasonably friendly terms with the Duke, as they were allowed to continue to work on Portland Place, even though its completion was delayed until the 1790s by unfavourable economic conditions and the Adam brothers’ own financial problems; by then both Robert and James Adam were dead. Their cause may have been aided by the Duke’s friendship with their nephew, the Rt Hon William Adam of Blair Adam, the son of Robert and James’s older brother John Adam. A lawyer and advocate by training, and later a judge, he was one of the 3rd Duke’s great allies in the Whig party when it came to boosting party morale and raising funds in preparations for the general election of 1790.


Special thanks are due to Dr Frances Sands, Curator of Drawings at Sir John Soane’s Museum, for supplying the images of the Adam designs for the Duke of Portland’s house; these are reproduced here by courtesy of the Trustees of the Soane Museum. The catalogue entries for these drawings in the Adam office collection at the Soane can be found by following this link. A discussion of the designs also features in Fran’s book Robert Adam’s London, published to accompany the exhibition of that name recently held at the museum. See here for further details.


1. University of Nottingham, MSS and Special Collections, PwF 35



Cavendish Square 4: No. 20 (the Royal College of Nursing)

By the Survey of London, on 29 April 2016

This is the fourth instalment in an occasional series of posts about Cavendish Square. Outward appearance belies the fact that there is a substantial early Georgian house at 20 Cavendish Square. It is enclosed within a shell of the 1930s that was part of an extensive redevelopment of a larger corner site as premises dedicated to the nursing profession.

P_1_34 Cavendish Square

20 Cavendish Square (to centre), c.1910 (Royal College of Nursing Archives).

The plain brick-fronted three-storey house was built in 1727–9 by George Greaves, a Clerkenwell carpenter. Its first resident was Francis Shepheard, a former MP who had moved from the Whigs to the Harleyite Tories. He was a wine trader and East India merchant in the steps of his father Samuel, who had been Robert Harley’s financier and was Deputy Governor of the South Sea Company from 1713. Francis inherited a fortune when his father died in 1719.


Plans of the ground floor and first floor of 20 Cavendish Square, showing the layout of the eighteenth-century house and later extensions of 1921-6 (© Survey of London, Helen Jones) Please click here to download a pdf version of the plans.

A broad five-bay front permitted a central entrance, but in an almost square plan there is an unusual layout. The house’s finest feature, a painted staircase, among the best of its kind in London, is oddly situated in the south-east corner. The great expense of such decoration may have obliged Shepheard to avoid an open-well entrance-hall form in favour of a dog-leg stair walled off from the entrance hall. This meant orienting the stair to rise from back to front, forcing a landing against the front wall and so causing irregular fenestration. The town-house aesthetic was evidently, and even more than usual, internal. The main staircase walls display two large architectural capricci, attributed by Edward Croft-Murray to John Devoto, the principal scene painter at the Drury Lane Theatre at the time. There are also trophies and grisaille figures of the Arts on the landing and of Music below the southern capriccio, these more in the style of James Thornhill. The ceiling, in contrast, is painted in the manner of William Kent, with gilded hatching and grisaille caryatids supporting a trompe-l’oeil dome.

Howard de Walden Project. Royal College of Nursing, 1a Cavendish Square, Marylebone, Greater London. Detail of stair. View from east.

View of the painted staircase at 20 Cavendish Square, showing the southern capriccio attributed to John Devoto (© Historic England, Chris Redgrave).


Howard de Walden Project. Royal College of Nursing, 1a Cavendish Square, Marylebone, Greater London. Detail of stair. View from west.

View of the painted staircase at 20 Cavendish Square, looking north-east from the first-floor landing (© Historic England, Chris Redgrave).

The next most impressive space in the house, to the north-west on the ground floor, is a fully pilastered room. The quality of this room, and the fact that its Ionic pilasters are sunk at the corners, hint at involvement on the part of James Gibbs, who was stationed just around the corner on Henrietta Street at the time. An imposing marble chimneypiece has a swagged frieze with a Medusa-like mask, plaits tied under her chin, a type widespread in the 1730s. However, flanking eagle-headed terms suggest the fireplace could be datable to the 1750s as double-headed eagles featured in the coat of arms of William Wildman Barrington, 2nd Viscount Barrington. He moved in when he became Secretary at War in 1755. During his tenure the house became known as a place for parties. The Barrington family continued to occupy the house up to 1888.


Measured drawing of the central part of the east side of the pilastered room (marked lounge on the plan) at 20 Cavendish Square (© Survey of London, Helen Jones).

In 1894 Sir Charles Tennant bought the property as a wedding present for his daughter, Margot, and the Liberal politician Herbert Henry Asquith. The couple lived here with their children and servants before and after Asquith served as Prime Minister from 1908 to 1916. Margot Asquith was a political hostess with a reputation for extravagance. On departing for 10 Downing Street she noted, ‘All the colour, furniture, grates, curtains, and every chair, table, and rug in Cavendish Square I had chosen myself.’ This precedes the disarming admission that ‘It is a constant source of surprise to people of moderate means to observe how little a big fortune contributes to Beauty.’ (The autobiography of Margot Asquith, ii, 1922, pp.103 and 106) After their wartime return, their income much diminished, the Asquiths were obliged to sell in 1920, marketing the property as suitable for ‘a nobleman, embassy or family of distinction’ (Royal College of Nursing Archives, RCN/28/3).

But Annie Pearson, Viscountess Cowdray, bought 20 Cavendish Square to be a clubhouse for nurses. She followed up in 1921–2 by funding rebuilding along Henrietta Street to form a headquarters building for the College of Nursing, established in 1916 in an office on Vere Street as a response to wartime exigencies, to support and advance the nursing profession. This was part of a phased campaign of redevelopment with Sir Edwin Cooper as architect throughout. The first phase included some internal remodelling of the house; Cooper’s neo-Georgian detailing is hard to distinguish from the eighteenth-century work.

20 Cavendish Square (left), as refaced and raised in 1932-4 (© Historic England, Chris Redgrave).

20 Cavendish Square (left), as refaced and raised in 1932-4 (© Historic England, Chris Redgrave).

The last rebuilding phase in 1932–4 followed the acquisition and demolition of a corner house. Cooper refaced and, through the ingenious introduction of bridging structural steelwork, heightened No. 20, which was used as the Cowdray Club up to 1974. The RCN has since carried out three refurbishments, the most recent of which in 2012–13 saw the opening on Henrietta Street of a public heritage centre with retail and museum elements.

30 Portland Place: London’s Guggenheim Museum that never was

By the Survey of London, on 26 February 2016

In the summer of 1939 plans were almost complete for the wealthy American heiress and socialite Peggy Guggenheim to open an Institute of Modern Art in London, in a house at 30 Portland Place specially rented for the purpose from Sir Kenneth Clark, who had been living there for some years with his family and his own remarkable collection of paintings and furnishings. The art historian and critic Herbert Read was already lined up as museum director on a five-year contract and Ms Guggenheim had left for Paris to acquire more artworks, with a list of desiderata drawn up by Read. But then war broke out…

30 Portland Place, view from the north-west (© Historic England, Chris Redgrave).

30 Portland Place, view from the north-west (© Historic England, Chris Redgrave). If you are having trouble viewing images, please click here.

No. 30 was one of a block of fashionable terraced houses, built in the 1770s–90s as part of the Adam brothers’ development at Portland Place. With the neighbouring properties to its south at Nos 26 and 28 it forms an elegant trio of period houses, all with good Adam interiors. In plan the house was straightforward, though typical of the care Robert Adam took in varying room shapes, with partitioning used to introduce curves at the far end of the rear ground-floor breakfast parlour and at both ends of the second drawing room, echoed by other shallow curves in the rear closet wing – features that are still traceable in the house today. Several Adam office designs for chimneypieces, friezes and ceilings for No. 30 survive in the collection at Sir John Soane’s Museum, but only the drawing-room ceilings remain in situ. [1] Much of the eighteenth-century décor was lost during a comprehensive refit in 1901 for the Liberal politician and merchant banker William Charles Heaton-Armstrong, who lived there with his family from 1898 until around 1911, when the failure of his bank forced him to move. His successor in 1911, Lady Margaret Jenkins, spent heavily reversing much of this, reintroducing ‘correct’ Georgian panelling and mantelpieces and stone hall flooring which she thought more in keeping with the date of the house. [2]

In 1934 No. 30 was taken by Sir Kenneth Clark, then newly appointed as Director of the National Galley and Surveyor of the King’s Pictures, and his wife Jane. Clark remembered the house as being ‘far too big’, and the piano nobile of Adam rooms as ‘completely unnecessary’, but he and his wife made use of the space to display artworks and to entertain on a grand scale. Photographs now in the RIBA show the house during their residence, with curtains, rugs and other fittings that the Clarks commissioned from contemporary artists and friends like Marion Dorn, Duncan Grant and Vanessa Bell; the painter Graham Bell lived for a while in an upstairs room. [3]

Kenneth Clark by Howard Coster, 1934 (© National Portrait Gallery, London)

Kenneth Clark by Howard Coster, 1934 (© National Portrait Gallery, London).

30PP Kenneth Clark's house RIBA72130

The dining room at 30 Portland Place in 1938, during Kenneth Clark’s residence, with curtains designed by Duncan Grant, and George Seurat’s Le Bec du Hoc, Grandcamp (now in the National Gallery) above the chimneypiece. (Photograph by Alfred Cracknell, © RIBA)

30PP Kenneth Clark's house RIBA72131

The sitting room at 30 Portland Place in 1938, with a rug designed by Marion Dorn, and what appears to be one of Paul Cézanne’s studies of Château Noir above the chimneypiece. (Photograph by Alfred Cracknell, © RIBA)

But by the summer of 1939, with war looming, the Clarks were keen to move their children out of London, and Sir Kenneth informed the Howard de Walden Estate that a ‘Mrs Goggenheim’ [sic] had expressed an interest in taking his house on a five-year lease to establish an Institute of Modern Art. By this date Peggy Guggenheim had closed her short-lived commercial gallery in Cork Street, Guggenheim Jeune, and was turning her attention to establishing a museum of modern art in Europe, preferably in London. [4]

The choice of Herbert Read as museum director was a canny one. As editor of the leading art magazine Burlington Review, he was at the time probably Britain’s best-known advocate of modern art. But it took the offer of a five-year contract from Ms Guggenheim to persuade him to relinquish his editorship. With such a large house intended for what, initially, would be a small museum, the plan was for Read and his family to reside on one upper floor, Guggenheim on another, though apparently they quarrelled about who should have which floor. There seems to have been genuine affection in their relationship: she regarded Read as a bit of a father figure, later reminiscing: ‘he treated me the way Disraeli treated Queen Victoria’. For his part, Read often signed letters to Ms Guggenheim as ‘Papa’. [5]

Herbert Read by Howard Coster, half-plate film negative, 1934 (© National Portrait Gallery, London).

Herbert Read by Howard Coster, 1934 (© National Portrait Gallery, London).

With everything decided, towards the end of July 1939 Peggy Guggenheim began to make plans for a trip to Paris to buy suitable works for the new museum. Read gave her a list of possible loans and acquisitions, and had already told the press that the opening show would trace the development of modern art since 1910, beginning with Matisse. The collector and gallery owner Roland Penrose, later a co-founder of the ICA, offered to lend some of his Picassos for the first show. The exact content of Read’s list is not known, but Matisse was one of the names that Peggy Guggenheim crossed off as being not ‘modern’ enough for her tastes; others to suffer the same fate included Cézanne and Rousseau.

But the outbreak of war early in September put paid to the project, even though the lease to the Portland Place house had been agreed (but not signed). Sir Kenneth Clark later suggested letting it rent-free as a centre for artists to meet and exhibit their work, but it was requisitioned in 1940 and damaged by bombing in 1941, since when it has seen a variety of commercial uses. Herbert Read was dismissed, never having held office. Once the dust had settled, Read rather ungenerously wrote to the artist Ben Nicholson about the Guggenheim affair: ‘Never in business matters rely on a single patron particularly if that patron is a woman and an American’ – this despite the fact that he had already been paid and allowed to keep an advance of £2,500, half his five-year salary.

The aborted Portland Place museum did, however, have one lasting legacy. Peggy Guggenheim held on to Read’s ‘shopping list’, which she later revised with the aid of artist friends such as Marcel Duchamp and Theo van Doesburg’s widow, Nellie; and the works that she then acquired – including masterpieces by Georges Braque, Salvador Dalí, Piet Mondrian and Man Ray, were to form the basis for her now world-famous private collection of abstract and Surrealist art.


[1] Sir John Soane’s Museum, Adam drawings, vol. 13/122–3; vol. 24/139–44; vol. 53/64

[2] Much of this and the succeeding paragraphs is based on correspondence files belonging to the Howard de Walden Estate Archive

[3] Kenneth Clark, Another Part of the Wood. A Self Portrait, 1985 edn, pp. 238–9, 251, 276

[4] Howard de Walden Estate Archive, property files for 30 Portland Place

[5] This and the succeeding paragraphs are largely based on Mary V. Dearborn, Mistress of Modernism: The Life of Peggy Guggenheim, pp. 158–63: Peggy Guggenheim: A Celebration, exhibition catalogue, 1998/1999, pp. 47–9: http://www.guggenheim-venice.it/inglese/museum/peggy.html

Cavendish Square 3: Nos 15 and 16 (with 2–14 Harley Street)

By the Survey of London, on 19 February 2016

This is the third in an occasional series of posts about Cavendish Square. At the corner of Cavendish Square and Harley Street is ‘les 110 de Taillevent’, a recently opened branch of a Parisian restaurant. This is the latest twist in the convoluted history of a building that despite serial alterations stands as a remnant of the square’s aristocratic origins.

16 Cavendish Square, Marylebone, Greater London. View from the south west. (© Historic England, Chris Redgrave)

Nos 15 and 16 Cavendish Square on the corner with Harley Street, from the south-west in 2014 (© Historic England, Chris Redgrave). If you are having trouble viewing images, please click here.

After Cavendish Square was laid out in 1717–18, the Duke of Chandos intended to build himself a palace across the whole of its north side. He was obliged to back-pedal after the South Sea Bubble burst, and in 1724–5 began building a pair of houses at either end of the frontage, to designs by Edward Shepherd. Dilatory and indecisive, Chandos left the carcasses incomplete. He eventually decided to take that to the west for himself, and returned to complete it in 1733–5. Entered from Harley Street, the house had a painted imperial staircase, expensively decorated by Gaetano Brunetti and Jacopo Amigoni.

by Herman van der Myn, oil on canvas, 1725 or before

James Brydges, 1st Duke of Chandos, by Herman van der Myn, c.1725 (Reproduced by kind permission of the National Portrait Gallery, London).

by Philip Mercier, oil on canvas, 1733

‘The Music Party’ by Philip Mercier, 1733, depicting (from left to right) Princess Anne, Princess Caroline, Prince Frederick and Princess Amelia reading from Milton. In the background is a depiction of the Kew Palace, or the Dutch House at Kew Gardens (Reproduced by kind permission of the National Potrait Gallery, London).

From 1762 Princess Amelia, George II’s second daughter, lived here and there was it seems a decline from Chandos’s opulence. After she died in 1786, The Times (21 Feb 1787, p.3) reported, ‘Of all the ill-furnished houses – perhaps that of the late Princess Amelia was the worst. With the exception of one large glass, it was much of the same sort as might have been expected at a plain Esquire’s in the country.’ James Hope, the 3rd Earl of Hopetoun, then undertook substantial improvements with Robert Adam as his architect. The entrance moved northwards within a tetrastyle stone porch (now gone) and under a pedimented aedicule (extant) and the north wing was much enlarged. In 1795 in moved Henry Hope, a distant cousin, Europe’s pre-eminent merchant banker, a great art collector and a refugee from Amsterdam. Poussins faced the front door and the front drawing room was graced by a Titian, two Veroneses and several Van Dycks. The last occupant of the whole house from 1816 to 1824 was George Watson Taylor MP, another art collector and, as an heir to a Jamaican fortune, a defender of Caribbean vested interests in Parliament.

by Charles Howard Hodges, published by John Boydell, published by Josiah Boydell, after Sir Joshua Reynolds, mezzotint, published 1 January 1788 (1787)

Henry Hope by Charles Howard Hodges, after Sir Joshua Reynolds, c.1787 (Reproduced by kind permission of the National Portrait Gallery, London).

by Pieter Christoffel Wonder, oil on canvas, 1826-1830

Study for ‘Patrons and Lovers of Art’ by Pieter Christoffel Wonder, 1826-1830, depicting George Watson Taylor kneeling in front of Titian’s ‘Bacchus and Ariadne’ (Reproduced by kind permission of the National Portrait Gallery, London).

Such a vast house in this location was no longer tenable and in the later 1820s Thomas Hardwick oversaw the separation of the capital mansion from its back parts with remodelling that included a top-lit staircase that is still extant. Hardwick added what is now 15 Cavendish Square and, on the garden, 6–14 Harley Street. The largest dwelling, on the corner, was taken by William Carr Beresford, Viscount Beresford, a hero of the Peninsular campaign. Dr Henry Herbert Southey, an eminent physician and younger brother of the poet Robert, took what became 4 Harley Street. Further division and alterations followed in 1863 and further eminent doctors and merchants moved in. The last private resident of the corner property was Edward Berman, a German button importer. No. 15 saw Beaux Arts alterations for the Jockey Club in 1913 and the corner block was converted with ground-floor rustication in 1926–7 to be a branch of Coutts Bank that closed in 2012.

(© Survey of London, Helen Jones)

Phase plans of Nos 15 and 16 Cavendish Square and 2-4 Harley Street. Please click on the picture to expand (© Survey of London, Helen Jones).

Mews Views

By the Survey of London, on 5 February 2016

Back in Marylebone, the subject of today’s post is the numerous mews that are significant hyphenations of the great regular grid of streets that is bracketed by Wigmore Street, Marylebone High Street, the Marylebone Road and Portland Place – the core, that is, of the Howard De Walden Estate. It is an irony that the latter-day charm of these places, which were designed in effect for the parking of coaches and horses, rests largely in their residential calm and relative freedom from vehicles.

Laid out with the streets in the later decades of the eighteenth century, they were characteristically sett-paved and originally lined by low (almost invariably two-storey and plain stock brick) rows of stables and coach-houses with living space above for associated servants. If there was architectural finesse, it faced the gardens of the houses, not the mews. There were piecemeal early rebuilds, but change appears to have been humble and in keeping until the 1890s (Thomas Woolner’s studio of 1862 in Marylebone Mews being an interesting exception). Around then a new type appeared, a variation where access to the upper living space was made separate by virtue of external stairs across the front, facilitating occupancy by those who had nothing to do with the horses. Soon after, motor garages appeared, as conversions and in some cases as purposeful rebuilds.

The early and middle decades of the twentieth century saw increasingly ambitious and concerted interventions, more expressly residential if always above parking. A good deal of this was due to two property developers, William Willett and Henry Brandon, who insinuated stronger elements of architectural style, ranging from variations on the Neo-Georgian to the Neo-Tudor.

After the Second World War residential and gentrifying conversions that began to diminish ground-floor vehicle cover led to gradual prettification through paint, stucco, glazing bars, carriage lamps and window boxes. Through the same post-war decades there were a few substantial Modernist redevelopments, occasionally for office or institutional use. Residential use remains the rule and basements are being excavated. The most recent replacement buildings are yet more self-consciously architectural than any of their predecessors.

(© Historic England, Chris Redgrave)

Devonshire Mews West, Nos 27 to 36 on the east side from the south, showing a frontage that has its origins in the 1820s, the London Clinic beyond (© Historic England, Chris Redgrave). If you are having trouble viewing images, please click here.

(© Historic England, Chris Redgrave)

29 Devonshire Mews West, a rare unreconstructed garage (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

37-38 Devonshire Mews West, 1956–7, Basil Hughes and Bonfield, architects, garages under two flats for the Howard De Walden Estate (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

39 Devonshire Mews West, 1908–9, with independent access to an upper-storey flat by means of an external staircase (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

Devonshire Close, showing Neo-Georgian brick at No. 15 (left), 1924 for William Willett with Amos Faulkner, architect, and Neo-Tudor black-and-white at Nos 12 and 14, 1935-9 for Henry Brandon with Alfred and Vincent Burr, architects (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

30-35 Devonshire Close, rebuilt for clients ranging from Major H. A. Wernher at No. 30 in 1922-3 (right) to Vincent Gluckstein in 1926 at No. 34. Willett and Faulkner were responsible for the gabled profiles of Nos 33 (1910-12) and 32 (1920), Moore-Smith & Colbeck, architects, for No. 31 in 1930-1 (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

Devonshire Mews South, sett paving and two-storey rows that have their origins in the 1780s, with the London Clinic again looming up to the north (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

107 Devonshire Mews South, a 1900ish rebuild with external stairs, altered in 1954 and since (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

34 Devonshire Place Mews, another stabling rebuild with a former forage-store opening to an attic that has been converted. Mansard roofs are fairly ubiquitous (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

Wimpole Mews, the east side from the south, showing No. 6 (right) of 1904 with shaped parapet, built for Walter Hamilton Hylton Jessop, an opthalmic surgeon of 73 Harley Street, with William Henry White as architect. For No. 8 beyond, see below (© Historic England, Chris Redgrave)

(© Historic England, Chris Redgrave)

No. 8 Wimpole Mews,  rebuilt in 2011–12 for the Howard De Walden Estate to plans by Urban Mesh Design Ltd (© Historic England, Chris Redgrave).

(© Historic England, Chris Redgrave)

33 Weymouth Mews, 1898, by and for Waterhouse & Son, the architects Alfred, then elderly, and his son and partner, Paul, who were based in an adjacent New Cavendish Street property (© Historic England, Chris Redgrave).

Cavendish Square 2: Nos 11-14

By the Survey of London, on 8 January 2016

This is the second in what will be an occasional series of posts about Cavendish Square. The north side of Cavendish Square has symmetry about its centre. This reflects the prominence of the elevation in relation to the square’s common axis along Holles Street to Hanover Square, a coherent piece of town-planning across two landholdings that is unusual for early eighteenth-century London. But the axis was the extent of the coherence then achieved. The Duke of Chandos’s plans for a full-width palace across Cavendish Square’s north side fell by the wayside when much of his fortune evaporated. Instead he built two big houses at either end and left the middle of the square’s show frontage empty save for use as a rubbish dump. This failure was excoriated by James Ralph in the Critical Review in an account of the square in 1734:

‘… there we shall see the folly of attempting great things, before we are sure we can accomplish little ones. Here ’tis, the modern plague of building was first stayed, and I think the rude, unfinish’d figure of this project should deter others from a like infatuation. When we see any thing like grandeur or beauty going forward, we are uneasy till ’tis finish’d, but when we see it interrupted, or intirely laid aside, we are not only angry with the disappointment, but the author too: I am morally assur’d that more people are displeas’d at seeing this square lie in its present neglected condition, than are entertain’d with what was meant for elegance or ornament in it.’ [1]

Howard de Walden Project. General view of Cavendish Square, Marylebone, Greater London. View from south.

Cavendish Square in 2013, with a view through the trees to Nos 11-14 (© Historic England, Chris Redgrave). If you are having trouble viewing images, please click here.

Extract from the Ordnance Survey First Edition map of Middlesex XVII, showing Cavendish Square and Hanover Square c. 1870 (Reproduced by permission of the National Library of Scotland)

Extract from the Ordnance Survey First Edition map of Middlesex XVII, showing Cavendish Square and its environs, including Hanover Square, c. 1870 (Reproduced by permission of the National Library of Scotland)

So things rested until the Society of Dilettanti acquired the site. Famously bibulous but seriously antiquarian, the Dilettanti began here in 1753 to build an academy for the improvement of painting, sculpture and architecture, shipping in Portland stone for what would have been an important public building and an early essay in neo-classicism, based on the Temple of Rome and Augustus at Pola in Istria, and designed by a sub-committee, Robert Dingley, Sir Francis Dashwood and Col. George Gray. Competition and want of money caused the project to be abandoned and the stone was sold to John Spencer. The columns on the Green Park front of Spencer House of 1755–9 are those intended for Cavendish Square in recycled and cut-down form.

(© Historic England, Chris Redgrave)

The north side of Cavendish Square in 2014, showing Nos 11-14 (© Historic England, Chris Redgrave)


Detail of the Corinthian portico at Nos 13-14 Cavendish Square, photographed in 2012 (© Historic England, Chris Redgrave)

Cavendish Square’s embarrassing gap mouldered and it was not until 1768–70 that the present buildings went up. Their Portland stone fronts with what Sir John Summerson called ‘magnificent Corinthian porticos’ appear to be a conscious if somewhat mysterious reflection of the abandoned Dilettanti project. [2] This was no academy, merely two pairs of semi-detached houses, a speculation for an MP, George Forster Tufnell, whose family gave its name to Tufnell Park – that’s another story.

11-14 Cavendish Square from the LCC GLC Historic Building Survey drawings collection (© Survey of London, Helen Jones)

Plans of 11-14 Cavendish Square, adapted from a Greater London Council survey of 1966 (© Survey of London, Helen Jones)

The inner houses were entered from a central dividing roadway that led to a stable yard to the rear. At the end of the nineteenth century three of the houses were adapted for the Society of the Holy Child Jesus, a Roman Catholic institution, to be a convent school. After war damage, repairs that included a linking bridge over the road were topped off with Jacob Epstein’s Madonna and Child, erected in 1953. The buildings are now the headquarters of the King’s Fund health charity.

(© Historic England, Chris Redgrave)

Jacob Epstein’s Madonna and Child of 1953, commissioned by the Society of the Holy Child Jesus, a Roman Catholic institution. The Society moved to and took the freehold of No. 11 in 1888-9, added No. 12 in 1891, and then No. 13 in 1898. Photographed in 2014 (© Historic England, Chris Redgrave)

For a fuller account of the eighteenth-century history of this site, see: Peter Guillery, ‘Cavendish Square and Spencer House: Neo-classicism, opportunity and nostalgia’, in The Georgian Group Journal, vol. 23, pp. 75-96.


[1] James Ralph, A Critical Review of the Publick Buildings, Statues and Ornaments in and around London and Westminster, 1734, p. 106

[2] John Summerson, ‘The Society’s House: an architectural study’, Journal of the Royal Society of Arts, vol. 102, Oct 1954, p. 924

Cavendish Square 1: No. 1A

By the Survey of London, on 27 November 2015

This is the first in a short series of posts about hidden or obscure features from the early years of Cavendish Square, a place that has more than its share of surprises for those with an interest in eighteenth-century architectural and art history.

Cavendish Square was laid out in 1717–18 with palatial aristocratic habitation a primary intention. As landowners, the Harley family sought to capitalize on top-echelon contacts to give development of its Marylebone estate the lustre of wealth. Plans were set tolerably fair, but went awry with the bursting of the South Sea Bubble in 1720. Money vanished and so did most of the aristocrats. By 1730 only two aristocratic town houses had been completed; one (Bingley House, to the west) has wholly gone, the other survives in forgotten part at the south end of the square’s east side, as No. 1a or Flanders House, where the Flemish representation of the Belgian Embassy has been housed since 2003.

No. 1A Cavendish Square in 2013 (Historic England, Chris Redgrave)

No. 1A Cavendish Square in 2013 (© Historic England, Chris Redgrave). If you are having difficulty viewing images, please click here.

This building is the southern third of what went up in almost cubic form in 1720–6 as Harcourt House, for Simon Harcourt, 1st Viscount Harcourt and a former Tory Lord Chancellor, who, suffering cataracts, was all but blind. This fact as much as short money might explain why richly Baroque first designs by Thomas Archer [1] were pared down by Harcourt’s first executant architect, Edward Wilcox; he died and in 1721 was succeeded by Francis Smith of Warwick. In 1724 Harcourt was three months a widower before marrying Elizabeth, née Vernon, the widow of a former friend, Sir John Walter. Lavish furnishing of the house, including Genoese damasks patterned with leopards, suns and elephants, was attributed to her influence. The internal fit-out was overseen by George I’s cabinet-maker James Moore. Harcourt did not enjoy the house long, dying in 1727.

3D Reconstruction of Harcourt House

Reconstruction of the former Harcourt House, as in 1827 (© Survey of London, Helen Jones)

His grandson, Simon, 1st Earl Harcourt, enlarged a middle back parlour with a canted bay in 1758 and gave it an interior by James ‘Athenian’ Stuart. This was Stuart’s first London commission and an important early ‘Greek’ neoclassical work. It was not universally liked – ‘God Damn my Blood My Lord is this your Grecian Architecture what villainy what absurdity If this be Grecian, Give me Chinese give me Gothick, Any thing is better than this, For Shame My Lord Pull it down & Burn it . . . not to expose your own ignorance for it is the most Wretched miserable affair ever was seen by Mortal.’ [2] This round deprecation by Lord Delaware was famously relished by Robert Adam and Paul Sandby. The more stolid Stiff Leadbetter was responsible for plainer alterations in the early 1760s, including the addition of a room to the south-east along Margaret Street.

Detail from plan of Cavendish Square, showing Nos 1-2

Detail from plan of Cavendish Square showing Nos 1 and 2, c.1810 (© Survey of London, Helen Jones)

The house was divided into two dwellings in 1825–7 and remodelled with new stucco facing by Sir Jeffry Wyatville. Sir Claude Scott, who had made a fortune as a wartime government grain contractor, bought the southern section and enlarged the Margaret Street room to be a banking hall, which, extended again in 1873, survives as a NatWest branch. Further works in 1928 that were overseen by Wimperis, Simpson and Guthrie gave the surviving part of the building its current appearance. The last aristocrats departed from the five-bay northern dwelling during the First World War. In came Queen Mary’s Needlework Guild and then, in the 1920s, the Mixed Arbitral Tribunals set up by the Treaty of Versailles to settle disputes arising from the war. Demolition and replacement with an explicitly commercial building followed in 1933–4.


  1. In the British Library – King’s Maps xxvii.6
  2. As quoted by Richard Hewlings in ed. Susan W. Soros, James ‘Athenian’ Stuart 1713–1788: the rediscovery of antiquity, 2006, pp.195–7