The Gallery holds the most extensive collection of portraits in the world. Search over 215,000 works, 150,000 of which are illustrated from the 16th Century to the present day.

Advanced Collection search

3. Lawrence's materials and processes

Unlike Joshua Reynolds, the leading portraitist of the previous generation, Lawrence favoured traditional painting techniques and materials. He maintained an interest in obtaining the best colours and from about 1810 he enlisted the help of George Field, the leading experimental chemist engaged in preparing colours. Lawrence's technical practice has not so far been examined in detail, unlike that of his contemporaries, of whom J.M.W. Turner and William Blake have been studied by Joyce Townsend and John Constable by Sarah Cove.

Here, much of the available information has been gathered on Lawrence’s suppliers, his supports, his colours and paint media, his brushes and his recommendations on varnishing. For more detailed information on individual suppliers, see British artists’ suppliers on the National Portrait Gallery website.

Canvases and canvas suppliers

Lawrence used commercially primed canvas, normally with a twill weave, as was common practice with portrait painters at this period. Less usual was the scale of his work. Whereas the standard size for a full-length portrait was 94 x 58 ins, or for a large full-length 96 x 60 ins, Lawrence used a canvas size of 100 x 60 ins for his ‘Largest size, whole length’, a size for which he was quoting 700 guineas in 1829, 100 guineas more than his standard whole length (Layard 1906 p.233). Lawrence used a still larger size of 106 x 70 ins, what John Constable in 1825 called ‘very large King size’, for his series of portraits of European leaders for the Waterloo Chamber and for a few other portraits. When another portrait painter took up this size to Lawrence’s disadvantage, Constable wrote that Lawrence had ‘made a rod for himself in inventing that size canvas’ (R.B. Beckett (ed.), John Constable's Correspondence, Suffolk Records Society, vol.4, 1964, p.419, vol.12, 1968, p.220). In the sheer size of his more ambitious single figure canvases and of their frames, Lawrence reached an extreme rarely surpassed in Britain.

Like many of his contemporaries, Lawrence preferred a pale ground as can be seen from some of his unfinished works, such as Lady Cahir, 1803/4 (Trustees of the Chevening Estate) or Elizabeth Inchbald (Private coll., both repr. Lawrence 2010 pp.194-5). His chalk portrait drawings on canvas, such as William Sotheby, c.1807, and The Duke of Wellington, c.1820 or earlier (both National Portrait Gallery) have warm off-white grounds which fill the canvas grain to present a smooth surface for drawing. That of Countess Czernin, 1819 (Private coll., repr. Lawrence 2010 p.168) has a slightly greener tone.

George IV by Thomas Lawrence, c.1814 (National Portrait Gallery).

Fig.9. George IV
by Thomas Lawrence, c.1814
(National Portrait Gallery).
Click on image to view larger version

Some of Lawrence’s incomplete paintings exhibit a further warm priming: buff in the case of Lady Callcott, 1819 (fig.6) and William Wilberforce, c.1828 (fig.5), pinkish for George IV, c.1814 (fig.9) and brown for his Self-portrait, c.1825 (Royal Academy). Unusually, the ground of William Wilberforce has been rubbed down to reveal the canvas weave.

Lawrence depended on a number of leading London artists’ colourmen for his canvases and colours. In 1842, ‘Young’ Thomas Brown claimed that he, his father (also Thomas Brown, d.1840), and his father's predecessor (James Poole), had between them supplied all the Royal Academy’s Presidents up to that time (The Art-Union January 1842 p.18). The Brown business was situated at 163 High Holborn, at the St Giles end, not far from Lawrence’s studio in Russell Square.

From his stamp on the canvas reverse, it is possible to document Brown as supplying Lawrence with canvas for various works between about 1807 and 1823, namely William Sotheby, c.1807 (National Portrait Gallery, see Walker 1985 p.467), Emma Cunliffe, begun 1809 (Huntington Art Collections, see Robyn Asleson and Shelley Bennett, British Paintings at the Huntington, 2001, p.238), Baron Crewe and Lady Crewe, c.1810 (Sotheby's 22 March 2005 lot 64, personal observation), Thomas Campbell, c.1820 (National Portrait Gallery, see Walker 1985 p.91) and Frederick Duke of York, c.1822 (Christie’s 6 July 2007 lot 209, personal observation). In 1823, Lawrence instructed ‘Wood’, presumably John Wood (1801-70), to obtain canvas from Brown when making him a copy of the King’s portrait (British Museum Print Room, Whitley papers, vol.3, p.298). Perhaps as late as 1835 'Thos Brown', presumably the colourman rather than the restorer by the name of Thomas Boden Brown, was paid £147.15s from Lawrence's estate towards a debt of more than £291 owing him (V&A National Art Library, MSL/1938/1922, p.26).

Smith & Warner's canvas stamp on reverse of George Canning, begun 1817? (Private coll.).

Fig.11. Smith & Warner's canvas stamp on reverse of George Canning begun 1817?
(Private coll.).
Click on image to view larger version

Lawrence occasionally used Smith, Warner & Co for the supply of canvas. Their unusual mark (fig.11), in white rather than black, can be found on Lawrence's George Canning (Private coll., see Lawrence 2010 p.266), a rare example of a marked canvas from this business. The address, 211 Piccadilly, was that used by Smith & Warner until about 1820. For fuller details of this leading company, see below under Colour suppliers.

Many of Lawrence’s canvases have been relined and little has been published concerning his unlined canvases, making it difficult to identify other suppliers. However, it is worth recording that in 1801 Lawrence was indebted to James Poole (d.1801), Brown’s predecessor, in the sum of £40 for unspecified supplies, and that he also owed the very substantial figure of £400 to another colourman, John Middleton (d.1818) of 81 St Martin’s Lane (Farington, vol.4 p.1525). Both Poole and Middleton were known canvas suppliers. ‘Middleton’, whether the colourman or not, crops up again in Lawrence’s finances in 1807 as being due £256 on a loan (Layard 1906 p.54).


Another supplier, Robert Davy (c.1771-1843?) of 16 Wardour Street, from 1822 of 83 Newman Street, specialised in prepared panels and millboards, with ‘Genuine Flemish Grounds’, according to his label (fig.12). But while his panels and boards were widely used by artists such as Landseer, Turner and Clarkson Stanfield, none of Lawrence’s works have been found on supports marked by Davy despite his prominent claim, 'Patronized by Sir Thos Lawrence, President of the Royal Academy'. Nevertheless, from time to time, Lawrence chose to work on quite large mahogany panels, such as his Baron Brougham, 1825, which he had enlarged in the course of painting, and John Fawcett, 1828 (both National Portrait Gallery, neither marked with source of supply, see Walker 1985 pp.66, 180).

Robert Davy's label on reverse of John Linnell's panel, William Mulready, 1833 (National Portrait Gallery).

Fig.12. Robert Davy's label on reverse of John Linnell's panel, William Mulready, 1833
(National Portrait Gallery).
Click on image to view larger version

Colours and Lawrence’s palette

‘Lawrence’s greatest gifts… were in the handling of paint and in the use of colour’, according to the late Sir Oliver Millar (Burlington Magazine, vol.122, 1980, p.79). The reception of Lawrence’s high-toned pictures at the Royal Academy has already been discussed. Here, his range and choice of individual colours is set out in a chronological survey, using the limited available documentation.

In 1790 Lawrence identified various colours for use by Lady Malden, the wife of one of his patrons, and these included Prussian Blue, Ivory Black, Van Dyck Brown (also known as Cologne Earth), Terra di Sienna (or Raw Sienna), Brown Ochre, Naples Yellow and Indian Red, standard colours of the time (Royal Academy Archive, LAW/1/32, draft letter to Lord Malden). Indian Red was later picked out by George Field as a colour Lawrence held in high esteem (George Field, Chromatography, 1835, p.96).

Subsequently in 1807, as Farington reveals, the sight of a picture by Titian led Lawrence to decide to abandon the use of Terra di Sienna, which he called an impure colour which had ‘often led him off from that purple freshness which should be attempted and caused him to fall into heavy & leathery colouring’ (Farington, vol.8, p.3008).

In 1809, Lawrence provided the young Henry Hamilton, son of his late artist friend, William Hamilton, with a drawing showing the layout for a palette (Beinecke Rare Books Manuscript Library, Yale University, repr. Lawrence 2010 p.160). The palette is marked with ten colours, including Antwerp Blue, Ivory Black, Brown Pink, Lake, Indian Red, Light Red, Light Oker, Naples Yellow, White and Blue Black. Beneath the palette Lawrence has noted, 'With(?) the Colors which have this mark, X mix a little sacrum(?) to dry them, and use in working only linseed oil’. The colours marked with a cross, X, are Ivory Black, Brown Pink, Lake and Blue Black. Subsequently, in 1819, Lawrence gave a similar palette drawing to the amateur painter, Countess Lulu Thürheim (see Albinson in Lawrence 2010 p.158).

Colour samples from Lawrence's own palette were apparently copied by Thomas Sully, as recorded in the student notebook, 1824, of the Philadelphia portrait painter, John Neagle (American Philosophical Society, Philadelphia, John Neagle papers).

Evidence as to other colours used by Lawrence comes from his suppliers. As is discussed in more detail below, Lawrence obtained Purple Lake from George Field in 1810, among other colours, and Italian Pink (a yellow pigment), Venetian Lake, Vermillion, Mountain Blue and Ultramarine from the Roberson business in the 1820s.

Sir Thomas Lawrence, c.1825, later copy by Richard Evans of an unfinished self-portrait by Lawrence, the palette and brushes added by Evans, c.1868 (National Portrait Gallery)

Fig.13. Sir Thomas Lawrence c.1825,
later copy by Richard Evans of an unfinished self-portrait by Lawrence, the palette and brushes added by Evans, c.1868
(National Portrait Gallery).
Click on image to view larger version

Lawrence’s approach to setting his colours on his palette developed with time, as is apparent from his letter to the Irish artist, Thomas Mulvany, in 1821, when he explained how he now used his palette, '... you should not mix up tints upon it. Blend them, for your immediate purpose, with the pencil. In my earlier practice, I did the former; but you have the very highest authority for the latter, that of Sir Joshua Reynolds, whose palette was very simple' (Williams 1831, vol.2, p.304). Lawrence left a set of notes, apparently dating to 1825, perhaps sketched out with one of his Academy discourses in mind, on the use of various colours when working out a composition, making mention of Reynolds’s example (Royal Academy Archive, LAW/4/337).

Lawrence’s range of colours was not unusual for the time. However, little technical analysis has been undertaken on his paintings when compared with the wealth of research on Blake, Constable and Turner. One of the few Lawrence portraits where analysis has been published is his Queen Charlotte, 1789 (National Gallery) where Cochineal, a carmine pigment, has been identified in the brownish red foliage in the background (National Gallery Technical Bulletin, vol.28, 2007, p.88). A more considered analysis of Lawrence’ practice will only be possible once further technical information becomes available.

Colour suppliers

Little is known about the supply of Lawrence’s colours in the first half of his career beyond the names of some of his colourmen in London, including John Middleton, James Poole and Thomas Brown, as already discussed. From about 1810, he was in touch with George Field and in the 1820s, he was buying some colours from Charles Roberson.

George Field, mezzotint by David Lucas, after Richard Rothwell, published 1845 (National Portrait Gallery))

Fig.14. George Field,
mezzotint by David Lucas, after Richard Rothwell, published 1845
(National Portrait Gallery).
Click on image to view larger version

George Field (1777-1854) was one of a number of leading chemists and colourmen in the first half of the 19th century who were seeking more permanent colours, whether naturally occurring or, increasingly, chemically produced. He had a good working relationship with Lawrence, and examined colours for him, as he did for other prominent artists (R.D. Harley, ‘Field’s Manuscripts: early 19th-century colour samples and fading tests’, Studies in Conservation, vol.24, 1979, pp.79-83). Lawrence was interested in new colours, but cautious about their effect, so that he asked Field to test pigments he had brought back from Rome, from Mengs’s painting box, in 1820. The yellow, a sulphurate of arsenic, Lawrence rejected, but he welcomed the red, a sulphurate of Mercury, which differed in hue from available vermilion (Gage 2001 p.8). Lawrence also brought Field samples of ‘Paris Yellow’ (Chrome Yellow, a new inorganic colour), Ultramarine and ‘Brown Pink’ (a greenish or orange brown) (Gage 2001 p.16). It is reported that Chrome Yellow has been detected in Lawrence’s Portrait of a Gentleman, 1800s/1810s (Bayerisches Staatsgemäldesammlungen, see Robert L. Feller (ed.), Artists’ Pigments: A Handbook of their History and Characteristics, 1986, vol.1, p.213).

Lawrence tried Field’s Purple Lake in 1810, a colour which Field had perfected after many tests (Farington, vol.10, p.3820, see also vol.11, p.3924, vol.12, p.4166) and he used a madder yellow which Field prepared specially for him (Gage 2001 p.68, n.129). Field explored the idea of an elaborate apparatus for restoring Lawrence’s stained and discoloured Raphael and Michelangelo drawings (John Gage, George Field and his Circle: From Romanticism to the Pre-Raphaelite Brotherhood, Fitzwilliam Museum, 1989, p.67), while Lawrence entered into discussions with him on Newton’s theory of the origin of colours in light. Lawrence was the first subscriber to Field's celebrated Chromatography, but the book was not published until 1835, several years after his death (Gage 2001 p.16). Many years later, in 1868, one of Lawrence's pupils, Richard Evans, described Field as the colourman that Lawrence employed, in a conversation with George Scharf, Director of the National Portrait Gallery (National Portrait Gallery Library, note by Scharf, 16 July 1868, in his annotated copy of Williams 1831, vol.2, p.568; Evans thought that many of Lawrence’s colours in the pictures at the National Portraits exhibition at South Kensington that year had deadened or flown).

Lawrence also dealt with another prominent business, Smith, Warner & Co of 211 Piccadilly, a partnership between Peter Warner (d.1824) and the experimental chemist, Charles Smith (d.1845). In their trade catalogue of c.1811-12, they listed Lawrence as one of many artists approving and patronising their colours. Lawrence later used one of their canvases for his portrait, George Canning, probably begun 1817 (see fig.11).

There were other suppliers with whom Lawrence came in touch at this period. The leading Glasgow chemical manufacturer, Charles Macintosh, known for his Prussian Blue, wrote at length to Lawrence in 1810 concerning the preparation of permanent colours for painters (Royal Academy Archive, LAW/1/240; for Macintosh, see The New Statistical Account of Scotland, vol.7, 1835, pp.166-7, accessed through Google Book Search). A year later, Lawrence obtained Ultramarine from an individual by the name of Marshall (Farington, vol.11, pp.3884, 3897), possibly the artist Ben Marshall (see below). In 1823, Lawrence sought a supply of fine Ultramarine from Rome through Pietro Camuccini, a picture dealer and restorer, but it took a year before his request was met (Royal Academy Archive, LAW/4/98, 201).

In 1821 Lawrence began patronising the young Charles Roberson (1799-1876), ‘Colourman to Artists and hair pencil maker’ and successor to Henry Matley at 54 Long Acre. Initially, Lawrence purchased materials to the value of £4.3s.6d, including Italian Pink, Venetian Lake and Vermillion in one or two ounce quantities, Italian chalk totalling six ounces and Mountain Blue, that is Azurite, totalling 19 ounces (about 550 grams); of these the Venetian Lake was the most expensive pigment at 6s an ounce (see Appendix for full transcript).

Roberson & Miller’s trade sheet (detail), Materials for Drawing and Painting, c.1828-39.

Fig.15. Roberson & Miller’s trade sheet (detail), Materials for Drawing and Painting, c.1828-39.
Click on image to view larger version

There is then a gap in the sequence of surviving ledgers in the Roberson archive before the next record of Lawrence’s purchases in 1828. That year Charles Roberson entered into partnership with Thomas Miller (c.1800-54), trading as Roberson & Miller at 51 Long Acre. Lawrence’s most remarkable purchase was made on 15 April 1829 when he bought 7 oz 3 drams (204 grams) of Ultramarine at £71.17s.6d, a very significant sum at a time when few individuals earned as much in a year. Lawrence’s account for £75.11s.6d was settled by his executor on 3 September 1830 (V&A National Art Library, MSL/1938/1922). In Roberson & Miller’s trade sheet (fig.15), Ultramarine was by far the most expensive pigment, costing fourteen times as much as any other when supplied in bladders as a finely ground colour for oil painting. Lawrence, however, purchased the colour in pigment form, as one of the colours that Roberson & Miller offered ‘finely ground in Powder, for Oil or Water Painting, or in their Crude state’.

Following Lawrence’s death, his executor made a payment of £13.3s.6d to another leading business, Rowney & Forster (V&A National Art Library, MSL/1938/1922, 16 August 1830).

Paint media

In 1790 Lawrence told Lady Malden that the vehicle he used to mix his colours was composed of nut oil and mastic varnish (Royal Academy Archive, LAW/1/32), whereas when writing to William Hamilton’s son in 1809, he recommended linseed oil. Analysis of Lawrence’s Queen Charlotte, 1789 (National Gallery) has identified linseed oil as the paint medium in the three positions which were sampled (National Gallery Technical Bulletin, vol.16, 1995, p.91). Both nut oil and linseed oil were common vehicles at the period.

Like other artists of the time, Lawrence was faced by a number of practitioners who claimed to have some magical ingredient or formula which would produce wonderful effects in painting. They included the amateur artist, Ann Jemima Provis, the colour merchant, Sebastiano Grandi, and ‘Mr Marshall’, possibly the artist, Benjamin Marshall. However, whatever Lawrence may have tried out, there is no evidence that he changed his painting medium in response to such claims.

Titianus Redivivus; or the seven-wise-men consulting the new Venetian oracle, etching by James Gillray, published by Hannah Humphrey, 2 November 1797 (National Portrait Gallery)

Fig.16. Titianus Redivivus; or the seven-wise-men consulting the new Venetian oracle,
etching by James Gillray, published by Hannah Humphrey, 2 November 1797
(National Portrait Gallery).
Click on image to view larger version

Ann Jemima Provis, and her father Thomas, offered to sell information on a secret medium as the basis of a system of painting. According to Joseph Farington, a number of artists tried the process in 1797, believing it would enable them to emulate the effects achieved by the great Venetian masters, hence talk of the ‘Venetian secret’, but the process proved to be disappointing (see John Gage, ‘Magilphs and Mysteries’, Apollo, vol.80, July 1964, pp.38-41). As a result of his interest, Lawrence was among the Royal Academicians caricatured by James Gillray in 1797 in his elaborate etching, Titianus Redivivus (fig.16), albeit in a very minor role. From a diary note made by Farington in March 1797, Lawrence 'surrendered himself to Provis and his daughter as if he had never held a pencil, and afterwards was surprised at having done so to two fools who knew so much less than himself'. Three months later, Lawrence told Farington that he had been ‘painting in body colour on Provis’ dark ground & thought it a great advantage' (Farington, vol.3, pp.788, 859). Lawrence does not appear to have made much further use of such a ground.

Sebastiano Grandi (active 1789-1822?) claimed to have discovered how to prepare canvas, copper, or panel, in the old Venetian style, citing ‘the peculiar harmony, brightness, and durability’ of the work of Titian, Veronese, the Bassani and other Venetian masters. He was described by George Field as ‘a most ignorant Italian quack in Colours, absorbent grounds, and Vehicles. A Mountebank and droll’ (Carlyle 2001 p.39 n.7). Nevertheless, he won the Society of Arts’ Silver Medal for colours and materials for painting and for a preparation of grounds on panels for painters; he also submitted a method for purifying oils for use by painters. Lawrence was among the artists providing the Society of Arts with certificates in Grandi’s favour in 1806 (Transactions of the Society of Arts, vol.24, 1806, pp.85-9; see also Farington, vol.7, p.2721, Carlyle 2001 p.33).

That a sense that success in painting could lie in a particular medium or process is evident from a recipe supplied by ‘Mr Marshall’, possibly the sporting artist Ben Marshall, to Lawrence, apparently in 1820. He provided directions for making a 'Menstruum' for painting, which was to be followed exactly if Lawrence wished to have the same success as enjoyed ‘for thirty years with Mr. Marshall’ (Royal Academy Archive, LAW/3/130).

Brushes and brush suppliers

Lawrence’s purchases from Roberson in 1821 included various brushes, namely six Hog Hair Tools at 4s and six ‘French Tin ditto’ at 6s, as well as two porte crayons at 2s.6d, and 18 French Tools for 12s later the same year. A tool was a type of brush. The porte crayons were intended for holding chalk. In 1828, Lawrence purchased a gross of Best Sable brushes at £3.12s. Roberson & Miller’s trade sheet (fig.15) features ‘Fine French Hog Hair Tools, flat and round’, as well as ‘Flat Sable, Fitch and Camel Hair Brushes in Tin, from 1 to 6 inches wide’ (wide flat brushes would be used for varnishing, as the colourman, Lechertier Barbe, makes clear in his 1851 price list). Brushes in ferrules, at this time made of tin, became common in the 19th century and could readily be made both flat and round (fig.17). It was said that the popularity of the flat brush in England originated with its adoption by Lawrence (W.T. Whitley, Artists and their friends in England 1700-1799, 1928, vol.1, p.102, kindly drawn to my attention by Sally Woodcock).

‘Flat Hog’s Hair Brushes in Tin for Oil Painting’, page from Winsor & Newton’s Catalogue and Price List. For the Trade Only, c.1857-61.)

Fig.17. ‘Flat Hog’s Hair Brushes in Tin for Oil Painting’, page from Winsor & Newton’s Catalogue and Price List. For the Trade Only, c.1857-61.
Click on image to view larger version

As early as 1803 Lawrence was ‘wishing for Paris brushes’, according to Joseph Farington (Farington, vol.5, p.1988), at a time when French brushes enjoyed a high reputation. The following year, Farington noted how between January and September Lawrence had run up a huge bill totalling £14 for turpentine for the sole purpose of cleaning his brushes, so confirming Lawrence’s reputation for carelessness with money, whereas Farington himself compromised by using soap to clean his brushes (Farington, vol.6, p.2404). Much later, Lady Grosvenor, remembering sitting to Lawrence as a child, added that she had been told that he was very extravagant with materials and never used the same brush twice (Lord Ronald Gower, Romney and Lawrence, 1882, p.69). However, this story hardly chimes with Farington’s report of Lawrence cleaning his brushes. Lawrence’s studio sale included 142 dozen brushes, sold mostly in lots of three to six dozens (British Museum, Whitley papers, vol.3, p.298). This may be compared to the 35 dozen brushes in John Hoppner’s studio sale in 1823 (W.T. Whitley, Art in England 1800-1820, 1928, p.162).


It was common practice to postpone final varnishing by a year or more after a painting had been finished, to allow the paint to dry properly. ‘About a year or two hence, and not sooner, I shall varnish it. Let no one else touch it’, Lawrence told a patron, perhaps Lord Dudley, in 1827. On occasion, Lawrence would arrange for a portrait to be varnished at a patron’s house, as is evident in 1822 when Lord Kidbrooke told Lawrence that it would not be convenient for his picture of Lord Frederick Campbell to be varnished since his house was going to be full of company (Royal Academy Archive, LAW/4/51).

If Lawrence could not supervise the varnishing process himself, he would provide detailed instructions, as when sending a portrait to America in 1810. ‘It should be varnish’d with the best Mastic Varnish…. The brush should be a soft flat brush and it should be pass’d evenly downwards and across the picture thus + The operation is however so simple, that any artist or picture cleaner knows perfectly well how to do it. The Mastic should be the purest picked gum dissolv’d in Turpentine.’ (Kenneth Garlick, Sir Thomas Lawrence, Regency Painter, exh. cat., Worcester Art Museum, 1960, p.38). Lawrence gave similar advice when sending a further two portraits to America in 1820, suggesting that mastic varnish should be obtained from a good chemist, rather than trusting to ‘a common color-shop’, and recommending that the work should be carried out by ‘some ingenious artist..., for he would do it with more care and delicacy than others' (Thomas Sully, Hints to Young Painters, Philadelphia, 1873, p.51).

In 1825, Lawrence asked the picture dealer, Samuel Woodburn, then in France, if he would be prepared to give a portrait of the King, which he was sending to the British Embassy in Paris, a coat of varnish, using the ‘best mastic varnish, with some of the very purest drying oil, a very small portion of which, mixed thoroughly with the mastic, will preserve it from chilling’ (Williams 1831, vol.2, p.406). It was necessary, Lawrence explained to Woodburn, ‘to examine the whole of the picture with your finger, to ascertain if it is thoroughly dry’, an essential test before varnishing (Williams 1831, vol.2, p.407).

Introduction (for works cited in brief)

It is hoped to develop this study of Lawrence’s portrait practice further as and when the results of detailed technical examination of his paintings and chalk on canvas drawings become available. Please provide feedback to Jacob Simon at [email protected]

Jacob Simon
October 2010