- Museum number
Studies for the Virgin and Child with St Anne and the infant Baptist, and some studies of machinery
Pen and brown ink, with grey wash, heightened with white, over black chalk, indented for transfer
Verso: Variant of recto group, and man's head in profile to right
Black chalk, and the group transferred from recto
Popham & Pouncey 1950
A. The largest study (enclosed in a frame), with the Virgin seated to the 1. on the lap of S. Anne, holding the Infant Christ, Who reaches out to S. John standing on the right. B. A wheel, through which passes an axle-shaft, c. A mill-wheel with water pouring on it from a conduit. D. An inscription: 'pagol data vechia per / vedere le machie de / le pietre tedessce'. e. A large wheel on a shaft on which are smaller wheels (?). F. The whole group (variant of A), g. Slight scrawl of a child leaning back to the right. - his 1. leg bent. H. The Child, in much the same position as in A, leaning over to the right. and extending His r. arm. I. The Child seated in profile to the right., His r. arm raised. J. Very slight sketch of the Virgin seated on the lap of S. Anne, her legs crossed (preliminary to k). k. The Virgin, her legs crossed, seated on the lap of S. Anne, whose legs and head only are indicated. L. Part of a bridge, weir, or dam. M. Inscription: 'fa coche dove lacq' (another line below?). N. Part of a sketch of a (?) dam.
[See Popham & Pouncey 1950, cat. no. 108 for a diagram of the positions of parts A-N on the BM sheet]
- Production date
- 1505-1508 (circa)
Height: 265 millimetres
Width: 199 millimetres
- Curator's comments
- The figure studies are connected with Leonardo's cartoon in the National Gallery, London (NG6337). The produced at least four versions of the subject of the Virgin and Child with St Anne and the Infant Baptist and/or lamb: three cartoons (only one of which survives) and a painting now in the Louvre. The chronology of these variants depends on stylistic analysis of the surviving works (and in one case of a contemporary copy), as well as taking into account the few descriptions of the works written by Leonardo's contemporaries (a good overview of this problem is given by Martin Clayton in the Palazzo Grassi catalogue pp. 242-3).
There is a consensus among scholars that the National Gallery cartoon dates from the first decade of the 16th century, but considerable disagreement as to exactly when in this period it was executed. Wasserman, for example, is of the opinion that it belongs right at the beginning of the period; Popham and Pouncey to 1501-5 and Kenneth Clark and Carlo Pedretti to 1508-10. In style it is perhaps closest to drawings of around 1505: the chalk underdrawing and the quick-fire revisions in some of the Leda drawings, particularly 12337 at Windsor (no. 74 in the Hayward show), are certainly comparable.
The main study enclosed within a rectangular frame with a series of equally spaced dots at the bottom giving an indication of the scale is generally agreed to be preparatory for the National Gallery cartoon. In the main study the artist's alterations to the figures, rapidly drawn in black chalk, pen and ink with touches of white heightening and grey wash added with the brush, are so extensive that their forms are difficult to make out. To combat the illegibility of the study on the recto, Leonardo traced the preferred arrangement of the figures by means of a stylus onto the verso of the sheet. The composition of this is very close, albeit in reverse, to that of the finished cartoon. In subsidiary drawings below the principal one, the artist explored variant poses for the Virgin and Child some of which are executed only in black chalk while others have been worked up in pen and wash. Wasserman (1970) proposed that these sketches date from a few years later than the main one, based on the appearance of a lamb by the Baptist in the right-hand study (he maintains there are two animals but the wash drawing is so nebulous that it is difficult to see a second animal). In his view this indicates that these drawings are for a cartoon described by Vasari of the Virgin and Child with St Anne and the Infant Baptist playing with a lamb, a lost work which he suggests was executed in 1503. The Baptist in the study does seem to be leaning on a lamb, yet quite apart from the doubt surrounding the reliability of Vasari's description (he might have been conflating the NG cartoon with the later Louvre painting; neither of which he had actually seen), there is no reason to think that Leonardo could not have explored this idea in the course of his preparation for the other cartoon especially as a lamb is a standard motif in representations of the Infant Baptist. Comparison of handling between the main study and the smaller ones would strongly indicate that they are coeval.
The truncated studies of the dam at the lower left and the water wheel on the right shows that this sheet has been cut; Pedretti observed that a sheet of water studies at Windsor (12666) was once part of the present one. The mechanical studies have been dated to c. 1506-8.
Lit.: B. Berenson, 'Drawings of the Florentine Painters', London, 1903 (and later editions), II, no. 1028; A.E. Popham, 'The Drawings of Leonardo da Vinci', London, 1946 (and later editions), no. 175; A.E. Popham and P. Pouncey, 'Italian drawings in the BM, the fourteenth and fifteenth centuries', London, 1950, I, no. 108, II, pls. CVI, CVII (with previous literature); A.E. Popham, in exhib. cat., London, Royal Academy, 'Leonardo da Vinci quincentenary exhibition', 1952, no. 104; J. Wasserman, 'A re-discovered cartoon by Leonardo da Vinci', 'The Burlington Magazine', CXII, 805, April 1970, pp. 194-204; idem, 'The Dating and Patronage of Leonardo's Burlington House Cartoon', "Art Bulletin", LIII, 3, September 1971, p. 316, fig. 7; M. Kemp, 'Leonardo da Vinci: The Marvellous Works of Nature and Man', London and Cambridge (Mass.), 1981, p. 223; N. Turner, in exhib. cat., BM, 'Florentine drawings of the sixteenth century', 1986, no. 5; M. Kemp, in exhib. cat., London, Hayward Gallery, 'Leonardo da Vinci', 1989, no. 77; M. Clayton, in exhib. cat., Venice, Palazzo Grassi, 'Leonardo & Venice', 1992, nos. 23-23a, p. 246; C.C. Bambach, in exhib. cat., New York, The Metropolitan Museum of Art, 'Leonardo da Vinci Master Draftsman', 2003, no. 96 (with further literature); idem in exhib. cat., Paris, Musée du Louvre, 'Léonard de Vinci: Dessins et manuscrits', 2003, no. 82; C. Van Cleave, 'Master Drawings of the Italian Renaissance', London, 2007, p. 90, illustrated p. 97
Popham & Pouncey 1950
The sheet is a document in the evolution of one of Leonardo's most important compositions, the cartoon at Burlington House (Bodmer 280). The smaller studies (F, G, H, I, J, and K) are for the whole or parts of the same group.
A was presumably drawn first on the sheet, very roughly in black chalk. The composition must then have been built up in chalk and pen and ink, and the small marginal jottings made in the course of its evolution, A has been so worked over by the artist as to make impossible any exact analysis of the various stages of this evolution; but among the many 'pentimenti' can be made out: two positions for the Virgin's 1. leg, a position for the Child corresponding closely with H, and what might be a foot corresponding with the Virgin's 1. foot in K.
The date of the drawing must depend on the date of the Burlington House cartoon, which is usually considered to belong to Leonardo's first Milanese period, before 1499, for the following reasons:
(a) Its supposed identity with the Arconati 'schizzo' mentioned by Resta in a letter to Bellori (before 1696: Bottari-Ticozzi, iii, p. 481): "Lodovico XII, re di Francia, prima del 1500 ordinò un cartone di s. Anna a Lionardo da Vinci, dimorante in Milano al servizio di Lodovico il Moro. Ne fece Leonardo un primo schizzo, che sta presso a' signori conti Arconati in Milano".
There is, however, no reason for equating the cartoon with the 'schizzo', of whose history nothing is known after 1721 (when it was seen by Edward Wright, 'Some Observations made in . . . France and Italy', London, 1730, who significantly describes it as "the same which is painted ... in the sacristy of St Celsus", i.e. a repetition of the Louvre composition), while the history of the cartoon is unknown before 1791 (when it is first mentioned in the R.A. records).
(b) The supposed identity of composition between the Louvre 'Virgin and Child with S. Anne' and the cartoon made for the Servites, seen, in an unfinished state, by Novellara and described in his letter of 3 April 1501 ["Ha facto solo dopoi che è ad Firenze uno schizo in uno cartone: finge uno Christo bambino de età cerca uno anno che uscendo quasi de bracci ad la mamma, piglia uno agnello et pare che lo strìnga. La mamma quasi levandose de grembo ad Santa Anna piglia el bambino per spiccarlo da lo agnello (animale immolatole) che significa la Passione. Santa Anna alquanto levandose da sedere pare che voglia ritenere la figliola che non spicca el bambino da lo agnellino, che forse vole figurare la Chiesa che non vorebbe fussi impedita la Passione di Christo. Et sono queste figure grande al naturale, ma stanno in piccolo cartone, perche tutte o sedono o stanno curve et una stae alquanto dinanci al l'altra verso la man sinistra: et questo schizo ancora non è finito"]; leading to the inference that the Burlington House cartoon, less evolved formally, would be earlier in date.
But the attitude of S. Anne in the Louvre picture does not fit with Novellara's account; in our opinion all the available evidence points to a Milanese origin for the Louvre picture, and a date about 1510. The mystical conceit of the Infant Christ embracing the lamb, the symbol of the Passion, and being restrained by His mother, who in her turn is restrained by S. Anne, is not due to Novellara's ingenuity in interpretation: it is precisely this elaborate symbolism which forms the central idea of Casio's sonnet (published about 1525), 'Per S. Anna che dipinse L. Vinci, che tenea la M. in brazzo, che non volea il figlio scendessi sopra un Agnello'. A contemporary example of its literal translation into pictorial terms is the picture by Brescianino at Berlin (1946,0713.8) which is almost certainly inspired ultimately by Leonardo, but is not necessarily a direct copy of a lost composition, as Suida and Clark suppose; it seems to us more likely that Brescianino knew a Leonardo composition at second hand, possibly through Raphael or Fra Bartolomeo (Raphael uses the lamb motive as early as 1506 or 1507, in his Prado 'Holy Family'; and a 'Holy Family' by a follower of Fra Bartolomeo (Christie's, 26 Nov. 1948, lot 45) embodies the essential points of the symbolism described by Novellara).
The possibility should not be overlooked that the Burlington House cartoon may belong to the second Florentine period. A drawing by Michelangelo at Oxford (BB 1561), of the 'Virgin and Child with S. Anne', in which the Burlington House group is used as the basis for a compact sculptural composition, with the figures turned inward instead of out, suggests that the cartoon was known in Florence about 1505: studies on the verso are very close to studies for the 'Bathers' and can be dated at the same period.
If we accept the hypothesis that the Burlington House cartoon is Florentine, the question arises of its relation to the '1501 cartoon'. Douglas (pp. 25 ff.) conjectures it to be a "first cartoon" for the Servites, made soon after Leonardo's arrival in April 1500, and rejected because it did not lay enough stress on the part played by S. Anne in the Redemption. The Burlington House cartoon is, however, much more than "just a beautiful presentation of happy family life". It is, surely, a subtle development of the theme of the '1501 cartoon': instead of embracing the Lamb, Christ blesses the Infant Baptist, and the Virgin, instead of restraining Him, smiles indulgently as though amused at His behaviour; while S. Anne, with a mysterious and searching smile, points upward to remind her daughter of the Child's divine mission, which is also alluded to by the presence of the Baptist (whose usual role is to foretell the Passion by indicating the Lamb of God with the words, "Ecce Agnus Dei"). The infinitely delicate nuances of expression which convey this idea are almost entirely lost in a reproduction, however good. Without knowing what the '1501 cartoon' (or cartoons?) was like (there is no evidence that the Berlin Brescianino is anything but a free, and possibly second-hand, adaptation of the motive), there is no stylistic basis for comparative dating. But it should not be overlooked that Vasari's description of the cartoon for the Servites fits perfectly with the Burlington House cartoon in all but one respect (he describes the Baptist embracing a lamb), far more closely, in fact, than with Novellara's description.
The argument outlined above contains nothing amounting to absolute proof; but, in our opinion, the balance of probabilities inclines towards Florence: the fact that it seems to be a refinement of the theme described by Novellara suggests that its date, and consequently the date of our drawing, is probably between April 1501 and about 1505 (the approximate date of the drawing by Michelangelo, BB 1561), while the style of our drawing is close to that of studies for the 'Battle ofAnghiari' (e.g. 109 below and Bodmer 292), the cartoon of which was commissioned by October 1503.
Heydenreich suggests (loc. cit.) that the transference on to the verso was by Leonardo himself. It is not possible to determine which of the many solutions and 'pentimenti' in A was the final one, but that transferred differs from the Burlington House cartoon in the position of the Virgin's hand and arm supporting the Child, and the absence of S. Anne's pointing hand. Heydenreich's theory, that this reversed composition, transferred to another sheet, was the basis of the Louvre drawing (Bodmer 278), and consequently the germ of the Louvre painting of the 'Virgin and Child with S. Anne', is ingenious: it is true, as he himself points out, that the composition has been radically altered (the Virgin and S. Anne have been transposed), but it is possible to make out traces of a head and blessing hand in the Paris drawing in more or less the same positions as the Child's head and hand in our transfer. The motive of the Virgin's 1. leg crossed over her r., which is an important feature in the Louvre drawing, is paralleled on our sheet in K. Heydenreich's theory rests largely on the assumption that the London
cartoon and the Louvre picture are very close in date. If, as we are inclined to think, following Clark, the latter was executed in Milan about 1510, it seems unlikely that Leonardo would have had recourse to a messy little scribble drawn several years earlier.
It is possible that Leonardo may have made use of a clay or wax model; Cardinal Borromeo (Gori, 'Symbolae, Decas Secunda', vii (1754), p. 123) mentions a clay model used by Leonardo in designing the cartoon from which Luini's picture was copied, presumably the Burlington House cartoon.
Of the other drawings on the sheet, not connected with the 'Virgin and Child and S. Anne', the water-wheel (c) may be compared with a series of drawings on fol. 165 r of the Arundel MS., which seem to illustrate a method of draining marshes by mill-wheels. Cf. a similar wheel on Windsor no. 12328 r (Bodmer 319) which also has studies of horsemen, dated by Clark, on account of their resemblance to studies for the 'Battle of Anghiari', about 1505.
The head of an old man on the verso is one of Leonardo's types (cf. Windsor no. 12607 and for style Windsor no. 12599). Windsor no. 12494, which is certainly not by Leonardo himself (as Clark recognized), is a copy from it.
Literature: BB 1028; Marks, p. 41 (recto); Bodmer 279 (recto), 312 (verso); Popham 175 (recto), 144 (verso); L. H. Heydenreich, Gazette, x (1933), pp. 205 ff., repr. p. 209 (recto), p. 211 (verso); A. Pettorelli, Rassegna d'arte, xx (1920), p. 197; Richter, ii, no. 1457; Bottari, pl. 100; Giglioli, p. 122, pl. CV (recto), p. 145, pl. CLXIV (verso); Popp 44; Venturi, p. 119; Seidlitz, ii, p. 33; Suida, p. 129; Poggi, pls. 105,157; H. Chapman and M. Faietti, exhib. cat., BM, London, `Fra Angelico to Leonardo: Italian Renaissance Drawings`, 2010, no. 56, pp.216-7 (cat. entry by H. Chapman).
- Not on display
- Exhibition history
1952, London, Royal Academy, no. 104
1972, BM, 'The Art of Drawing', no. 117
1986, BM, Florentine Drawings, no. 5
1989, London, Hayward Gallery, 'Leonardo', no. 77
1992, Venice, Palazzo Grassi, 'Leonardo & Venice', no. 23
2003 Jan-April, New York, Met Mus of Art, Leonardo da Vinci, Master Draftsman
2003 May-July, Paris, Musée du Louvre, Léonard de Vinci: Dessins et manuscrits, no.82
2010 April-July, BM, `Fra Angelico to Leonardo`, no.56
2011, March-June, Uffizi, Florence, 'Figure, Memorie, Spazio: Disegni da Fra'Angelico a Leonardo', no.56
2011/12 Nov-Feb, London, National Gallery, Leonardo da Vinci
2012 March-May, Paris, Louvre, La Vierge a l'Enfant avec Sainte Anne
- Acquisition date
- Acquisition notes
- Sold at the Galichon sale for 13,000 francs (being the top price for a drawing in the sale)
- Prints and Drawings
- Registration number