Painted circa 1738, this picture is a supreme example of a distinct phase within Canaletto’s production that lasted from the late 1730s until circa 1742, in which the warm sunshine so characteristic of his early maturity gives way to a cool, clear light, bringing with it a greater clarity and precision. Never before sold at auction, the painting has remained largely inaccessible to scholars, hidden in private collections and exhibited only once, almost forty years ago.\n\nThe view shows the upper reaches of the Grand Canal with the almost concealed entrance to the Cannaregio Canal just to the left of centre. On the far left is the Palazzo Flangini, begun to the designs of Giuseppe Sardi in 1644 but left unfinished. Beside it is the Scuola dei Morti, which was to be completely rebuilt in a different form after the Austrian bombardment of 1849. The lantern surmounting the cupola of San Geremia can be glimpsed behind it. Beyond the short garden wall is the priest’s house (canonica) of San Geremia, with a gable breaking the line of its roof, and, immediately adjacent, we see the low garden wall of Palazzo Labia (which itself faces the Cannaregio). Beyond the mouth of the Cannaregio is a large building, possibly a warehouse, behind the site of the still unfinished Palazzo Querini (only finally completed in 1828) with, in the distance, the low buildings bordering the Campo San Marcuola. Further on, half cut off by the buildings on the right bank, is part of the façade of the 18th-century Palazzo Vendramin-Calergi. On the right of the composition is the Riva di Biasio, with the 16th-century Palazzo Corner and Palazzo Gritti (both greatly altered today), the 18th-century Palazzo Zen (later Donà Balbi) crowned by twin obelisks (this building was also badly damaged in 1849 by the Austrian bombardment). The projecting side wall of the Palazzo Bembo (now demolished) is seen in profile at the furthest point.\nThe quay on the corner of the Cannaregio is shown without the balustrade and the statue of Saint John Nepomuk by Giovanni Marchiori erected in 1742. Canaletto was quite attentive to this particular detail. Indeed, in his earlier view of The Entrance to the Cannaregio in the Royal Collection the statue was added to that painting, presumably at the request of its then owner, Joseph Smith, in order to bring the picture up-to-date topographically, as Sir Michael Levey observed (M. Levey, "Canaletto’s Fourteen Paintings and Visentini’s Prospectus Magni Canalis", in The Burlington Magazine, vol. CIV, no. 713, August 1962, p. 338). Its absence thus constitutes a probable terminus ante quem for the execution of the painting. An even more solid terminus ante quem is provided by the inclusion of an engraving after the painting in the second edition of Antonio Visentini’s celebrated book of engravings after Venetian views by Canaletto, the Prospectus Magni Canalis Venetiarum, published in 1742 (see fig. 1). Although Constable initially favoured a slightly larger version, now in the Minneapolis Institute of Arts, as the prototype for the engraving (his no. 257(a); 61.2 by 92.4 cm.; 24 1/8 by 36 3/8 in.), Links and all subsequent writers give this painting precedence. The correspondence in details such as the figure on the balcony of the Palazzo Flangini, the figure in the alleyway next to the far corner of that building, and the awning on the Palazzo Bembo, all absent from the Minneapolis version, would seem to confirm this. The other topographical detail possibly relevant to the dating of the painting is the low, brick building on the Campo San Marcuola, with a door facing the canal; this presumably replaced the hut of rough boards shown in Canaletto’s drawing of the campo in the Cagnola sketchbook in the Venetian Accademia (fol. 48v; G. Nepi Scirè, Il Quaderno di Canaletto, Venice 1997, pp. 144-5, reproduced; here fig. 2) and in Canaletto’s painting The Grand Canal, from the Palazzo Bembo to the Palazzo Vendramin-Calergi at Woburn Abbey (Constable, see Literature, reproduced vol. I, plate 51; vol. II, no. 256). If the replacing of the hut with a permanent structure is related to the completion and consecration of the Church of San Marcuola in 1735, this would provide a firm terminus post quem.\n\nMore precise dating is, however, possible on stylistic grounds. the mood of Canaletto’s work changes in 1738-9, a colder light replacing the warm sunshine which pervades his earlier production, and it retains this distinctive character until around 1742, when the sunshine returns to stay. The style produced its own masterpieces in the frosty panoramic view of The Bacino di San Marco, looking East (Boston, Museum of Fine Arts; ibid., reproduced vol. I, plate 32; vol. II, no. 131), and The Grand Canal at San Simeone Piccolo (London, National Gallery; ibid., reproduced vol. I, plate 52; vol. II, no. 259). Smaller examples include the Campo SS. Giovanni e Paolo sold London, Christie’s, 13 December 2000, lot 104, for £7,000,000. In its translucence, precision and subtlety, this painting clearly belongs to this moment in the artist’s development, indeed to the earlier part of it. This is confirmed by the existence of a slightly larger version of this painting by Bernardo Bellotto, datable to circa 1740/1, when he was nearing the end of his period in Canaletto’s studio (Constable, op. cit., vol. II, no. 257(b)). Identified as the work of Bellotto by Charles Beddington and Bozena Kowalczyk in 1998, that was sold London, Christie’s, 16 December 1998, lot 78, for £3,050,000, and is now in a private collection, U.S.A. (see fig. 3; see Kowalczyk, under Literature, 2001, reproduced fig. 11). That should not be confused with the anonymous derivation sold London, Christie’s, 11 July 1980, lot 121, as ‘B. Bellotto’, offered unsuccessfully New York, Sotheby’s, 21 May 1998, lot 134, as by Bellotto, and now in a private collection, Brescia (see Succi, op. cit., 1999, p. 61, reproduced fig. 45, and Succi, loc. cit., 2001). Another copy was sold New York, Christie’s, 15 January 1985, lot 237.\n\nAt the time of publishing his monograph Constable only knew this exceptional painting from a poor photograph in the Witt Library, taken when it was with Leggatt in 1925, and did not even know its size. He even wondered if it might be by Bellotto (according to private notes in the Constable-Links archive in the possession of Charles Beddington). He first saw it in 1964 when it was with Knoedler’s, to whom he wrote on 22 November 1964 confirming its identification as his no. 257(d) and adding ‘I am also, from having seen the original, convinced that the painting is by Canaletto himself’ … ‘I now think that the ex-Leggatt picture has claims to be the original of the engraving’ (carbon copy in the Constable-Links archive). Noting that Leggatt’s records were destroyed during the war, he never makes any suggestion as to its earlier provenance.\nThis painting was, however, added by Links (see Literature) to the ‘Harvey group’ of twenty Venetian views of similar size, depicting the Grand Canal, churches and campi, generally dated to the mid-1730s. The group is named after Sir Robert Grenville Harvey, 2nd Bt. (1856-1931), of Langley Park, Buckinghamshire. It was sold by the Harvey Trustees around 1957, and the paintings are now widely dispersed, curiously not a single one being in a public collection. The pictures were first seen by Constable at Langley Park in the 1920s, before they were placed on public exhibition at Oxford (1936-8) and subsequently in Birmingham (1938-c.1957). Constable (op. cit., vol. I, pp. 111-12; vol. II, pp. 262-3, under no. 188) inexplicably refers to twenty-one pictures while identifying only twenty in his catalogue, describing the series as consisting of eleven views of the Grand Canal and ten of churches and the campi in which they stand, but listing only ten canal views. He also refers to nine of the Harvey series as being engraved by Visentini for his Prospectus Magni Canalis Venetiarum, but identifies only eight. Links clearly found these inconsistencies tantalising, and the stylistic similarities with the Harvey paintings (with four of which it fortuitously formerly hung in the Wrightsman Collection) sufficient to warrant identification of this painting, a canal view engraved by Visentini, as the missing twenty-first component of the group. He was encouraged in this by information provided by F.J.B. Watson, from an unnamed source, that ‘one of the Harvey group was found damaged in an attic and sold separately some time between 1918 and 1936’. This might be taken as an explanation of how this painting, despite showing no signs of damage, came to be with Leggatt’s in 1925. The identification of this painting as the twenty-first of the Harvey group has been accepted by all subsequent writers up to the present day (see, for instance, Kowalczyk, loc. cit., 2005). If it might seem strange for a set of this kind to have an odd number of components, Links even on one occasion suggested that it originally comprised twenty-two views (J.G. Links, Canaletto and his patrons, London 1977, p. 45).\nWhile Constable recorded a family tradition that the paintings were bought in Venice by Harvey’s uncle-in-law, the 3rd (and last) Duke of Buckingham and Chandos (1823-1889), Links (op. cit., 1976, vol. II, p. 277, under no. 188) pointed out that ‘it seems unlikely that the group would have been intact in Venice during his lifetime’ and an earlier date of acquisition is supported by the matching frames of a type favoured by Consul Smith in which all the paintings appear to have been displayed at Langley Park.\nMore recently, Francis Russell has proposed an alternative early provenance for the group, which would appear to provide a solution to this problem: ‘Here the recorded history of one of a coherent group of Woottons, also formerly at Langley, may be relevant: this was acquired with the house by Sir Robert Harvey when he purchased it from the financially embarrassed fourth Duke of Marlborough in the 1780s. Houses at the time were often sold furnished and if the duke was prepared to sell pictures of his father, the third duke’s favourite horses, he would surely have found Venetian views equally disposable. It cannot be a coincidence that the third duke’s sister, Diana, who died in 1735, was the first wife of John, fourth Duke of Bedford, who acquired the only comparable series of views by the artist, the earliest of the three bills associable with which is of 27th February 1732-33’ (F. Russell, ‘A Supplement to W.G. Constable’s Canaletto, Giovanni Antonio Canal, 1697-1768’, book review, in The Burlington Magazine, vol. CXLI, no. 1152, March 1999, p. 181). Documentary evidence in the form of an apparently unique copy of a printed list of the paintings at Langley was recently discovered by John Harris in the Buckinghamshire Record Office, attached to an insurance valuation of the 1890s, and would appear to confirm the Marlborough provenance for the series. Five views of Venice are listed in the Drawing Room, followed by a note stating that “These pictures and others by Canaletto were bought with the house from the Duke of Marlborough in 1788. In an old inventory they are described as ‘20 views in fine frames’” (see E. Fahy, Paintings and Drawings in the Wrightsman Collection, publication due May 2005). This would seem to confirm that there were only ever twenty Harvey paintings, and indeed this painting would seem stylistically slightly later in date. It probably originally did have a companion or companions but these remain to be identified. Visentini’s engraving in the Prospectus Magni Canalis Venetiarum indicates that the painting passed through the hands of Joseph Smith, Canaletto’s great patron and agent, but the identity of his client also remains to be ascertained.\nA drawing of the unfinished Palazzo Querini and the adjacent buildings from a slightly different viewpoint is in the Cagnola sketchbook in the Venetian Accademia, fol. 49v (see fig. 5; Nepi Scirè, op. cit, pp. 146-7, reproduced); that corresponds more closely with a significantly earlier painting of the view from a different viewpoint in the Wallace Collection (Constable, op. cit., vol. I, reproduced plate 52; vol. II, no. 257; Hedley, loc. cit., reproduced in colour).\nWe are grateful to Charles Beddington for his assistance in preparing this catalogue entry.