1 The main contract drawings for Worcester Lodge, an elevation and plan among the Beaufort drawings at Gloucester CRO (4.28: see n. 4 and 6 below) and a section still in the Badminton muniment room, are in fact in the hand of Stephen Wright, an identification 1 owe to John Harris. As will be shown below Wright was Kent’s aide-de-camp at Badminton and stayed on to complete Kent’s works after his death.

2 Country Life, 4 April 1968, pp. 800-04.

3 The Buildings of England: Gloucestershire: the Cotswolds, p. 256.

4 The notebook, described below, is in the Bodleian Library, English MS f. 556; the bank accounts are at Child’s (first and second dukes) and Hoare’s (first, third, and fourth dukes); the remaining material is in the Beaufort muniments, some — notably letters — still at Badminton, but most now on deposit at the Gloucestershire County Record Office (ref. D. 2700). I acknowledge with warm thanks the permission of their Graces the Duke and Duchess of Beaufort to reproduce and summarize from the Badminton muniments and the readiness with which they allowed me to spend long hours in the house studying and photographing; likewise the permission of Bodley’s librarian and the directors of Williams & Glynn’s and Hoare’s banks to use material in their care or possession. I am also most grateful for the willing help given to me by Mr S. Bywater at Badminton, by Maurice Lloyd at Hoare’s and by Margaret Richards and Nicholas Kingsley at Gloucester, and by Geoffrey Beard. Much of the research for this paper and surrounding studies was undertaken while I held a Leverhulme fellowship, without the generous assistance of which I could have made very little progress indeed.

5 The total cost was £29,760 13s. i^/id. (Glos. CRO 303.1.16).

6 See below: the plan is on the recto of no. 1.4 in the collection of architectural drawings belonging to the Duke of Beaufort now on deposit in the Gloucestershire County Record Office and referred to hereafter as ‘Beaufort drawings’.

7 Possibly the range on the east side of the stable courtyard, remodelled in the seventeenth century, substantially survives from the early house.?

8 Glos. CRO D.2700.401.2.6. Excerpts from these muniments will be referred to hereafter simply by their catalogue number.

9 Hoare’s Bank archives.

10 Though only Bache’s and Gibbons’s Christian names appear in the accounts (the latter’s as ‘Grymlin’, ‘Grimlin’, Grimblin’ or ‘Grimlen’), their trades and the collocation of surnames identify them with certainty.

11 Recorded payments are as follows: Atherton £27, Bache £100, Bankes £435 (in two payments), Emmett £330, Gibbons £148 (including £98 ‘for pictures’), Grove £115, Streater£i40 (two payments). King (see below) has two payments recorded (£280), Rawson three (£103) and Mansfield fourteen (£737).

12 It is perhaps worth observing that Emmett had in 1661-62 been employed at Somerset House (History of the King’s Works, v, 255).

13 All dates are given in new style. The recent Guide to Badminton prepared by the Extra-Mural department of Birmingham University gives a helpful brief account of the family but differs from the present paper in its' interpretation of the interiors of the house.

14 See Colvin, Biographical Dictionary of British Architects, p. 602.

15 Ibid., p. 494.

16 Sumsion, Thomas the mason of Colerne who was previously credited with the design of Swangrove, may perhaps have been its builder or simply the proprietor of the Swangrove quarry mentioned under his name in CRO 401 11, p. 39.Google Scholar

17 It may have been Power who drew the exploded elevations of a small room (Beaufort drawings 1.6) to show the replacement of old-fashioned panelling by up-to-date wainscot. A fourth contractor (401.1.1, p. 273) is James Martin of Castle Combe who is to do ‘all plasterer’s and tiler’s work consequent on the above’, i.e. the operations of Hayward, Bennett, and Power; he was to receive £24.

18 The tri-pedimented almshouses in the village, for which a childish (though measured) elevation survives (1.7), may be by Hayward alone.

19 Dated 1709; another, slightly more interesting, is by ‘Halwood the bricklayer’ (1.14).

20 9.1-49. Some of the detailed studies of windows, doors etc., may have been copied from Rossi; Scamozzi of course he would have already found in his own library.

21 In addition £100 was paid in 1742 to ‘Charles Eborall’: no such name appears in the Warwick baptism records, and ‘Charles’ may be a slip for George or Thomas.

22 Another Beaufort property, Netheravon House, Wiltshire, which Smith was building while at work on Badminton, also figures in the notebook, pp. 1-4. It is not surprising that the Beauforts were much in debt in the middle 1730s, as evidenced by letters from the bank asking for payment; in fact the account seems to have been suspended between 1734 and 1736.

23 Hayward, Bennett and others continued working at least through 1729 and 1730 alongside the new craftsmen: Smith apparently did not bring a mason of his own until Bloxham came in 1739.

24 For which compare the overmantel in the morning room at Chicheley. The panel was originally elsewhere in the house, for Pocock mentions a picture above the library fireplace.

25 The evidence is in a plan (6.3) by Capability Brown who worked at Badminton in 1752 (the date comes from a payment from Hoare’s and a signed and dated sketch for the Mount, 6.1) and suggested rebuilding to make room for a ‘State Bed Chamber where the Great Stairs now are’. His ambitious but gawky plan was not followed. Brown’s evidence is confirmed by three drawings of Wyatville’s showing the plan in 1807 with proposed alterations (8.1, 8.2 and 8.4) and displaying the imperial staircase which he intended to replace. Curiously Brown had also proposed a single-flight staircase to replace Eborall’s, a very narrow one to the south-west of the hall: see 6.3.

26 The closest parallel I know to the Badminton doorcases is one in the Bowling Green House at Wrest, which has been implausibly claimed for Batty Langley. Gibbs might also have proposed the form of the staircase which he had used at Cannons.

27 For the temple see 3.6 and 3.7, also Colvin, art. cit., fig. 9. Its duplicated appearance on either side of both the Bridgeman drawing (see below) and the Wootton painting which was perhaps based on Bridgeman (ib. figs 4 and 5) is no proof of building: the obelisks and Gibbsian pediment were almost certainly not built either.

28 3.3 is a related design for a pedimented gate with alternative iron grilles.

29 The ‘Bridgeman’ drawing is — so far as the house is concerned — essentially a perspective version of 4.6 with a plain pediment added.

30 This letter, one of the few known from the great building period, is in Correspondence Drawer 32 in the Badminton muniment room.

31 Payments for work at Badminton were, as with so many eighteenth-century estates, made in a bewildering variety of ways — by cheque or in cash, by an agent or steward, as well as by the duke or duchess. The bank accounts are therefore only evidence of minimum payments.

32 Cf. the similar technique of the elevation for the north front of Gopsall Hall (RIBA Hiorne 12, reproduced in Harris, The Palladians, pi. 98).?

33 Payments amounting to £1,978 are reported from Flitcroft’s own account at Hoare’s, though there seem to be no equivalent debits on the Beaufort side. The likelihood is that these were for the reconstruction of Grosvenor House, Mayfair, which the Beauforts had just then taken on lease. Of course, if Flitcroft did work at Badminton, this would make comparisons with Ditchley even less discriminating, for then all four main architects involved would have worked at both houses.

34 Related, as Colvin pointed out (art. cit., p. 804), to one of Kent’s Houses of Parliament designs: it was revived by Paine a few years later for the now lost front of Serlby.

35 Was Kent in fact responsible also for the tiny but extraordinarily telling recession of the outer bays of the centrepiece in the upper two storeys? Neither the early engravings nor, more significantly, the survey drawing (4.6) shows it.

36 See, e.g., Harris, Artist and the Country House, pi. 351a and b. The visit is referred to by Vertue, writing in the summer of 1749: see Constable, W. G. Canaletto, 11, 405.Google Scholar

37 The half-elevation, again in an Office of Works hand, reproduced by Colvin (art. cit., fig. 7) must be interpreted in the same way. Incidentally William Smith’s last payment is as late asjuly 1747, three months after his death: it was doubltess overdue from two or three years before.

38 These were also the years when the great furniture collection was being made: John Phillips ‘cabinet-maker of Hammersmith’ was kept active between 1728 and 1730 making many very large pier glasses with dolphins, sphynxes, eagles and so forth (304.11.1); and William Linnell seems to have worked throughout the earlier 1750s with an occasional contribution from William Vile (Hoare’s archives). The famous Chippendale pieces must date from the fifth duke’s time, for which no bank accounts have yet been found.

39 Something may also have gone in 1787-88, for accounts (403.2.6) include payments of£2,115 for carpenter’s work on the house and offices.

40 The park buildings make a story of their own for which there is no space here: it belongs as much as anyone to Thomas Wright, many of whose attractive sketches for them survive (5.1-34). Wright also designed the reconstruction of the Beaufort house at Stoke Gifford (see 10.1-7) and worked in the park at Netheravon (5.18 and 40). Mention must finally be made of surviving designs for the church which nestles so intimately against the house. Francis Smith produced two versions of a conventional Georgian church — one arcaded, one with colonnade and straight entablature — attached to an elementary Gothic steeple, the tower of which may be intended to represent that of the medieval church (12.5-11). Much more interesting is a group of drawings (12.1-4) sketching plans, elevations, and sections of a church with lateral entrance opposite a central altar: one (12.4) includes a tremendous Jonesian screen of two superimposed hexastyle orders with a pedimented triumphal arch above. The drawings suggest a knowledge of Webb’s theoretical experiments and are perhaps by Vardy. Nothing came of any of these schemes, and the church which finally replaced the medieval one in 1783—85 was built by the fifth duke’s London carpenter Charles Evans who is firmly identified in the accounts as ‘architect and carpenter’. The dull exterior (12.12-15) cannot have seriously taxed Evans’s architectural powers, and the sumptuous interior comes, with minimal neo-classical trimming, straight from StMartin-in-the-Fields. The guiding spirit of Gibbs still haunted Badminton thirty years after his death.