St George and the Dragon

WB.172     1550–75 (?) • Enamelled gold, precious gems • hat-ornament

Curator's Description

Oval medallion; gold; St George and Dragon; horse enamelled lilac; body of saint formed of angular diamonds; dragon enamelled pale green with emeralds set in wings and head; background chased with trees and buildings; raised border with cross-hatching in black enamel; five enamelled groups of fruit; five rubies in square settings; four attachment loops.

This object was collected and bequeathed to the British Museum by Ferdinand Anselm Rothschild.

How big is it?

5.9 cm wide, 7.3 cm high, 1 cm deep, and it weighs 75.1g

Detailed Curatorial Notes

Text from Tait 1986:-

Origin: Attributed to a French workshop, third quarter of 16th century.

Provenance: None is recorded.

Commentary: In Read 1902 this remarkable jewel was simply described as “German, 16th century”, an opinion repeated without any elaboration in Dalton 1927. In Evans (both the 1953 edition and again in the revised edition of 1970) the illustration of this jewel (pl. 62a) was printed in reverse - thereby creating a left-handed St George- and the caption read: “probably German, middle of 16th century”, although in the text (p. 91) Dame Joan Evans observed that this hat-jewel “looks Italianate in its design, but technical tricks such as the setting of the rubies to form the sleeve and the diamond set in the horse's flank, suggest a German origin”. Apart from the error about the rubies, this statement creates a seriously misleading impression, because in the absence of documented jewels of this type known to have been made in Italy or in France there can be no reliable information about the 'technical tricks' which might have characterised the Italian or French jewels. It now seems, after lengthy comparative study of the German jewels of this type, that the 'technical tricks' employed by the maker of this St George and the Dragon hat-jewel are not wholly consistent with known Germanic practice, and the differences are sufficient to leave the matter open for reconsideration.

Certainly, the Italian character of the composition is overwhelmingly convincing and, in all its essentials, recalls the brilliant solution created by Raphael when in 1506 he painted the historic picture for the Duke of Urbino to send as a present to Henry VII in London on the occasion of the Duke becoming a Knight of the Garter; the painting is now preserved in the National Gallery of Art, Washington, but is also illustrated in Oliver Millar, ‘The Queen's Pictures’, London, 1977, pl. 26. Raphael was a source of inspiration for many of his young contemporaries and followers, and a number of close variants in different media have survived, like the painting in Naples by Lelio Orsi (1511-87), which has been compared with the circular hat-jewel of St George and the Dragon in the Imperial Hapsburg Collections in Vienna (see Hackenbroch 1979, p. 44, figs 95 and 98). However, there is one major difference between the hat-jewel in Vienna and Orsi's painting in Naples, which has not been discussed hitherto: the horse and St George are shown moving to the right on the hat-jewel - not to the left, as in the Orsi painting. The goldsmith, by adopting the easier solution, has placed St George's right arm at the front of the relief; this arrangement has simplified matters because, by tradition, St George is always depicted wielding a sword or lance in his right hand, and, consequently, the goldsmith has not been obliged to depict the left arm at all but has simply shown the Saint's right arm in a raised position and his body without any contrapposto movement. However, Raphael and other Italian artists since Donatello's marble relief (c. 1420) in the church of Or San Michele, Florence, have frequently preferred to show the equestrian group moving towards the left and, hence, to show the right hand of the Saint coming across the chest to thrust downwards a death blow with a lance - not a sword; the lance must, therefore, pass downwards in front of the horse's neck and the twist of the body of the Saint must become greater, the closer he is shown to the Dragon.

Unlike the creator of this Waddesdon jewel of St George, neither the goldsmith of the Vienna hat-jewel nor of the famous Stonyhurst College Reliquary pendant traditionally associated with St Thomas More (beheaded in 1535) were faced with solving the problems created by an equestrian group moving to the left. They had both opted for the easier solution in the manner of their predecessor, the Flemish goldsmith who made the hat-jewel of St George, c.1510, which is now in the Royal Collection, Windsor Castle (see Evans 1970, col. pl. VA), and which was discussed in connection with the female bust (WB.192). They were probably making these two pieces (now in Vienna and Stonyhurst College) between c. 1520 and 1530 - several decades before the Waddesdon version was created. Both were gifted goldsmiths capable of brilliant repoussé work, from the subtle low relief details to the dramatic high relief of the figures; this technical skill is vividly revealed by a close examination of the reverse; unfortunately, only the back of the Stonyhurst College Man of Sorrows relief is visible because the most delicate part of the Reliquary - reputedly a much faded and damaged watercolour portrait of Sir Thomas More by Holbein - has never been removed to reveal the back of the St George relief. Both the Vienna jewel and the Stonyhurst College Reliquary have separate circular frames of enamelled gold, into which the repoussé reliefs are fitted. The very similar techniques used to create both the Vienna and the Stonyhurst jewels are strikingly different from those employed by the goldsmith who made this Waddesdon St George hat-jewel with its separately made figural reliefs 'pegged' to the gold ground plate. While discussion has continued over many years concerning the origin of the Vienna and the Stonyhurst St George jewels either in Italy or in Germany, there has been agreement that the Reliquary, despite its association with the More family, could not be described as an English work (see Tait 1962, p. 244; Evans 1970, p. 76, no. 6; D. F. Rowe, A "George in Gold" and Enamels in Chicago Collections, ‘Apollo’, June 1972, pp. 470-3, figs 13-15). Recently, the Reliquary has been published as originating in England, although there is no fresh additional evidence (see Hackenbroch 1979, pp. 272-6, fig. 731 A-B). In this publication it is misleadingly stated that it was “once owned by St Thomas More, c. 1535; frame 1550-60” and that it was “the work of some of those outstanding Netherlandish masters active in London, whose skill and experiences exceeded that of the local jewellers”. This new attribution is based on an incorrect understanding of the technique employed by the goldsmith in the construction of the two figural reliefs (p. 276): “the technique pursued by the goldsmith was once again that of casting the principal figures in gold and applying them with butterfly clips to a gold ground, on which all other forms are raised with infinite devotion to detail. This technique is typically northern and can be observed, for instance, on the enseigne with the portrait of Charles of Spain (later Emperor Charles V) dated 1519, which originated almost certainly from the Netherlands, and more particularly from Malines, the residence of Margaret of Austria. On the other hand, the floral borders of translucent 'basse-taille' enamel date from the later 16th century, when the two medallions were converted into parts of a reliquary pendant.”

In fact, no such technique was used on this Reliquary. None of the figures are cast, nor are they applied to the ground plate of gold. The entire relief is executed by embossing and is true repoussé work. The English and Netherlandish goldsmiths' technique of using butterfly clips to attach the separately made figural elements is not to be found on the Reliquary. Consequently, Miss Hackenbroch's new attribution is not tenable on those grounds.

Equally baseless is the supposition that the Reliquary's frame is a later addition (“1550-60”). Since there is no evidence to link the making of this object with a goldsmith's workshop in England, the basse-taille enamelling on the frame could again be accepted as contemporary with the two figural medallions provided that the talented workshop was located in one of the leading centres, probably Italy or Flanders, for these artistic centres were far in advance of England, both in style and in technique, at the time of Thomas More's appointment as Lord Chancellor in 1529. Wherever the Vienna St George hat-jewel and the Stonyhurst Reliquary were made on the Continent, there can be no dispute that a different and more ambitious workshop was responsible for this Waddesdon hat-jewel.

The popularity of the St George subject in Italian Renaissance jewellery is also reflected, for example, in the lists of jewels belonging to members of the Medici family in Florence. Thus, in the 1566 list (MS. no. 352, in the Biblioteca Nazionale Centrale, Florence) there is (item 13): “Una medaglia d'oro, aovata, entrovi un san giorgio col drago, smaltato, di sopra, nel festone, una diamante.” and in the 1585 list of jewels being given to Virginia Medici (Archivo di Stato, Florence, Miscellanea Medicea, 12.N.) there are two gem-set St George medallions:

“(a) Una medaglia d'oro smaltata entrovi un sangiorgio con 4 diamanti e 2 esmeraldi.

(b) Una medaglia d'oro dentrovi uno San’ Giorgio che amazza il drago di smalto et quatro diamanti a quadri e due esmeraldini.”

The ability of the Italian craftsmen, even lesser men than Cellini, to handle diamonds and set them into the gold reliefs of these historiated jewels cannot be questioned. In 1536 Paul III, wanting the particularly fine diamond from the Emperor Charles V to be set perfectly, insisted that Cellini should first obtain the opinions of three or four leading goldsmiths of Rome, including Gasparre Galli, who in 1548 was appointed the Papal jeweller. The lack of extant examples in Italy is largely a result of the upheavals of the succeeding centuries. Certainly, the freak survival in Vienna of the Cellini salt-cellar (made in Paris c. 1542-3) is as convincing a piece of evidence as one could wish to have, for it fully justifies the detailed technical explanation in Cellini's own accounts (the ‘Trattato’ and the ‘Vita’); although it is not set with gemstones, it is a tour de force of enamelling, especially of those subtly changing tones of translucent enamelling, which to a lesser degree are present on the landscape background of this hat-jewel. Cellini's ability to handle diamonds and other gemstones is fully discussed in his ‘Trattato dell' oreficeria’ and, fortunately, the British Museum's three coloured drawings made in the first half of the eighteenth century faithfully record Cellini's brilliant - but now destroyed - work on the morse of Pope Clement VII in 1530-1 with its eight table-cut stones and its one very large pointed diamond in the centre (first pub. by the Rev. Herbert Thurston, S. J., ‘The Burlington Magazine’, 1905-6, pp. 37-43, pls. 1-11; more recently, reproduced in ‘Tutta l'opera del Cellini’, Milan, 1955, p. 53). Cellini's design cleverly incorporates the precious stones so that they appear to be integral with the embossed figures of God the Father and the angels. Although Cellini does not actually use any of the stones to form the figures themselves, the idea of integrating the gemstones so that they form a part of the relief is essentially the same as on the Waddesdon hat-jewel, where they form the bright armour of St George and the evil green spikes of the dragon's wing.

The fashion for these jewels with a saint clad in armour composed of diamonds was clearly widespread. In France, where the 1560 Inventory of the Court Jewels also proves how fashionable the enseigne, or hat-jewel, had become, there are both St Michel and St George jewels: for example,

“470. Une enseigne de nacre de perle d'un Saint-George de relief, estimée viii.

524. Une grande enseigne d'or d'ung Saint-Michel orné de diamantz, qui combat un diable de nacre de perle enrichy de turquoises et rubiz dont il s'en deffault ung diamant, estimée IIC.”

The second of these two entries suggests that the St Michel was wearing armour composed of diamonds; indeed, the value would seem to confirm this interpretation. Another French Inventory - the contents of the Château de Pau in 1561-2 - is even more explicit:

“133. Ung Saint Michel d'or enrichy de dyamans, en facion d'Ordre.

134. Un autre saint Michel plus petit tout armé de dyamans et à l'entour cinq tables de dyamans.”

The second of these two jewels must have looked rather like the two documented figures of knights in armour composed entirely of diamonds, which have survived (though their settings are lost) in the Grünes Gewölbe, Dresden, and are traditionally thought to represent St George on foot. Certainly, they seem to resemble fairly closely the two late medieval St George pendants which are today known only from the wonderfully detailed illustrations made by Hans Mielich in 1552-6 for the Inventory of the Jewels owned by Anna, Duchess of Bavaria (Staatsbibliothek, Munich, Cod. Monacensis, icon. 429); these four valuable pieces of evidence are illustrated in Hackenbroch 1979, figs 329, 330 and 332 A-B, along with the undocumented and very improbable oval, gem-set St Michael pendant, sold by the Dowager Marchioness of Cholmondeley in 1972 (Christie's, London, 28 June, lot 16, col. illus; also Hackenbroch 1979, fig. 331, captioned “c. 1530”; see also Fritz Falk, ‘Edelsteinschliff und Fassungsformen’, Ulm, 1975, pp. 126-7, fig. 71). Apart from the complete uncertainty surrounding the origin of those four historic items (the two in the Munich Inventory may be as early as c. 1500 and the two in Dresden may be mid-sixteenth century), it is interesting to note that none of the four has the cuirass (or breastplate) composed of diamonds that bears any resemblance to the shape and cut of those on the Waddesdon St George. Furthermore, one of the two Dresden figures of St George has its head carved from mother-of-pearl, which is reminiscent of no. 524 in the Paris Inventory of 1560 (already quoted above), but is quite different from the Waddesdon jewel. Significantly, none of these four was mounted on a horse; they were always intended to be 'on foot' just like the two versions recorded in the Duke of Württemberg's hereditary treasures in the 1569 and 1617 Inventories: “ein Stammkleinod, der Ritter S. Georg zu Fuss von lauter dimant, darbei ein Schildlin darinnen ein Robinkreuz, aussen herun vier Demant Taflen, unten zwei grosse Perlen samt sechs Robinen, so die Altern ein Gespang genannt.” (“A house-jewel, the Knight St George on foot all of diamonds, with a shield set with a ruby cross, encircled by four table-cut diamonds, below two large pearls and six rubies, traditionally called a "Gespang".”)

Secondly, “ein Stammkleinod wieder ein runde guldine Gespang, der Ritter S. Georg zu Fuss von lauter Demant mit einer grossen Robinwaken.” (“A house-jewel, again a circular gold "Gespang", the Knight St George on foot all of diamonds with a large ruby.”) (See W. Fleischhauser, ‘Die Geschichte der Kunstkammer der Herzoge von Württemberg in Stuttgart’, Stuttgart, 1976, pp. 9 ff., 19 ff.).

The dating of these two diamond St George figures 'on foot' and the two examples in the Grünes Gewölbe cannot be precisely determined and, for that reason, it is fortunate that the same historic collection of the Electors of Saxony contains a remarkable pendant of David and Goliath (see Arnold 1964, pp. 52-3, no. 15, figs 23-4; Menzhausen 1968, p. 90, col. pl. 67 (i); Hackenbroch 1979, fig. 576 A-B, col. pl. XXIII). The importance of this pendant in this context lies in the representation of Goliath, standing erect in a suit of Roman-style armour, much of which is composed of diamonds shaped and faceted to indicate the different pieces of armour. It is particularly noticeable that the 'diamond armour' on the legs corresponds with the Waddesdon St George, but again the cuirass is rather differently treated. Because this pendant jewel is complete with its 'architectural' openwork frame, its gem-set bow-fronted platform and its elegant smaller figures, it can be more precisely dated on stylistic grounds to the late sixteenth century and attributed to a German workshop without much hesitation. It is a valuable document, demonstrating the continued use of this kind of 'diamond armour' in late sixteenth-century German historiated jewellery, and it opens up the possibility that the Waddesdon jewel could, also, have been created at a German court. It is, however, an isolated document today.

The rich Saxon court has preserved one of the largest and finest Renaissance gold enamelled pendants of St George in the round; the latter shows St George on a white horse ‘moving to the right’ and, with his lance in his right hand, he majestically slays the dragon beneath the rearing horse (see Arnold 1964, p. 57, no. 24, cover illus.; Menzhausen 1968, col. pl. 38; Hackenbroch 1979, fig. 437 A-B). Although the horse's harness is far more elaborately rendered with, for example, bridle and halter and even plumes between its ears, there are some similarities between the enamelled cartouche setting of the table-cut diamonds on the horse's flank and chest. In other respects there are probably as many differences as there are similarities, and, certainly, there has been no attempt to use diamonds to represent the armour of St George.

The evidence in support of a German attribution for this Waddesdon jewel, therefore, remains inadequate and consequently, there remains the possibility that this exceptional masterpiece may have been created at the French court, under very strong Italian influence. Paris would have had the necessary expertise in gem-cutting, and, undoubtedly, its goldsmiths by the middle of the sixteenth century had both the inventive skill and the lavish patronage to spur them on to excel in this field. Unfortunately, very little firm evidence has survived, but if the traditional attribution of the famous Battle Scene hat-jewel (in the Cabinet des Médailles) is accepted, then there is vivid evidence of a gifted goldsmith in Paris whose handling of miniature equestrian groups is both dramatic and skilful (see Evans 1970, pl. 46c).

Perhaps more can be gained by comparing the five bunches of brightly enamelled fruit on the frame of this hat-jewel with those that were being made at this period to decorate the best French items, like the famous locket of 1571 containing two miniatures by Clouet of Catherine de'Medici and Charles IX (now in the Imperial Collections in Vienna - see ‘Princely Magnificence’ 1980, no. 23; also Hackenbroch 1979, p. 96, fig. 246 A-D). The reverse of this locket is embossed and enamelled with the royal cipher of a crowned double C within a great garland of fruits in high relief and richly coloured in enamels.

Even more closely similar are the bunches of fruit surrounding the famous Mannerist square cameo frame of enamelled gold, which is today in the Cabinet des Médailles but which was formerly in the Cabinet du Roi and appears in an Inventory as early as 1664 (no. 196); for a detailed discussion of this exceptional piece see ‘L'École de Fontainebleau’, 1972, p. 457, no. 663, with illus.

Finally, the French goldsmiths under Catherine de'Medici, the Queen Mother, and her son, Charles IX (reigned 1560-74), produced one astounding achievement in enamelled gold that escaped destruction - the gold shield of Charles IX which since 1793 has been in the Gallerie d'Apollon, Musée du Louvre (see Pierre Verlet in E. Steingräber (ed.), ‘Royal Treasures’, Munich, 1968, p. 23, pl. 16, for an excellent detail view in colour; for an early publication of the shield see Edouard Lièvre, ‘Les Collections Célèbres’, Paris, 1869, pls 99-100, with an accompanying text by Henry Barbet de Jouy). This masterpiece of chasing and embossing, enamelling and burnishing is the work of the years 1566-72; in 1572 Marie de Foureroy, widow of the goldsmith Pierre Redon, was paid 5,000 livres for the shield. It depicts the victorious battle scene of 1565 between the French and the Turkish infidels in Malta. Made for Charles IX, this shield is indicative of the quality of French Mannerist goldsmiths' work long after Cellini's departure. Despite the vast difference in scale, the shield has a quality, both in the figure style and in the enamelling, which helps to reinforce the tentative attribution of this hat-jewel of St George to the Paris workshops of the 1560s.


  • Charles Hercules Read, ‘The Waddesdon Bequest: Catalogue of the Works of Art bequeathed to the British Museum by Baron Ferdinand Rothschild, M.P., 1898’, London, 1902, no. 172, pl. XLII
  • O.M. Dalton, ‘The Waddesdon Bequest’, 2nd edn (rev), British Museum, London, 1927, no. 172
  • Joan Evans, ‘A History of Jewellery, 110-1870’, London, 1953 (rev. edn 1970), p. 91, pl. 62a (printed in reverse)
  • Yvonne Hackenbroch, ‘Renaissance Jewellery’, Sotheby Parke Bernet Publications, London, New York and Munich, 1979, p. 164, fig. 442
  • Hugh Tait, 'Catalogue of the Waddesdon Bequest in the British Museum. 1., The Jewels', British Museum, London, 1986, no. 172, pl. VIA, figs. 48-53
  • Dora Thornton, 'A Rothschild Renaissance: Treasures from the Waddesdon Bequest', British Museum, London, 2015, pp.204-211.Hat ornaments were worn by men in their caps from the end of the fifteenth century and could be either purely decorative or symbolic. Those of the latter type are traditionally termed enseignes, since they either conveyed the personal intent of the wearer or carried a visible message. This type of jewel finds its origins in the medieval pilgrim badge, an object that was mostly mass-produced and often in base metal. It has been suggested that the transition from this type to a fashionable male ornament is attributable to the arrival of the French king, Charles VIII, into Naples in February 1495. On his cap, the king wore a gold circular jewel and his men had similar jewels (though not of gold) on their caps or sleeves. The Italians soon adopted this fashion and it then spread north reaching most of the European courts. The fashion lasted only until the late-sixteenth/early-seventeenth century, when the wearing of aigrettes became more popular.

    The hat ornament was usually commissioned of gold, and was enamelled or jewelled, or both. A group of gilt-bronze plaquettes in the British Museum’s collection that have the characteristic loops or pierced holes for attachment suggests that this was a fashion that trickled down to lower classes of society. This category of objects has been cast, which was a much cheaper and quicker mode of production than those that were commissioned. One of these plaquettes (1915,1216.133) has visible traces of enamel. This combined with the gilded decoration and placed at the apex of the body would have deceived any casual passer-by that this was a costly piece.

  • References

    1. Read 1902: Read, Charles Hercules, The Waddesdon Bequest. Catalogue of the Works of Art Bequeathed to the British Museum by Baron Ferdinand Rothschild, M.P., 1898, London, BMP, 1902
    2. Dalton 1927: Dalton, Ormonde Maddock, The Waddesdon Bequest : jewels, plate, and other works of art bequeathed by Baron Ferdinand Rothschild., London, BMP, 1927
    3. Tait 1986: Tait, Hugh, Catalogue of the Waddesdon Bequest in the British Museum; I The Jewels, London, BMP, 1986

Go to the Collection Online page for this object?