During his lifetime, Pietro Santi Bartoli (1635–1700) was the most celebrated illustrator and copyist of ancient art and artefacts, working for and with the leading antiquaries and collectors of the day. He was known to a wider public through the many volumes of his engravings published with commentaries by Giovanni Pietro Bellori (1613–96) and others. His status as a copyist and engraver was already recognized in 1675 in an admiring note on his work at the end of the list of Italian artists published by the artist and biographer Joachim von Sandrart (Reference Sandrart and Peltzer1971: 294). Sandrart also briefly noted in this context another fine draughtsman, Giovanni Battista Galestruzzi, but neither he nor others at work in the burgeoning antiquarian scene of seventeenth-century Rome attained the reputation enjoyed by Bartoli. The volumes of engravings published during his lifetime were reprinted frequently thereafter, and the coloured copies he made of wall-paintings and mosaics became collectors' items in their own right. To the publications more directly associated with him may be added the celebrated Recueil de peintures antiques (Reference Mariette, de Tubières and de Caylus1757) published by the Comte de Caylus and Pierre-Jean Mariette on the basis of an incomplete set of Bartoli's coloured drawings that Caylus had acquired in rather unusual circumstances. Those drawings are now in the Bibliothèque Nationale de France in Paris,Footnote ¹ while others related to Bartoli's engravings, together with some that were never published, are scattered by the haphazard circumstances of acquisition and disposal in various collections of drawings after the antique, many of them in England (Connor Bulman, Reference Connor Bulman, Jones and Matthew2001). Only rarely do they survive in any systematic arrangement; one of the welcome exceptions is the volume compiled by Bartoli in 1674 for his first and most significant patron, Cardinal Camillo Massimo (1620–77). Rediscovered in the 1970s in Glasgow University Library, the volume was described in a paper by Claire Pace (Reference Pace1979) that provided an assessment of Bartoli's work in the context of the seventeenth-century reception of Roman painting.Footnote ²

The Glasgow volume contains 127 coloured drawings, with the newly-discovered paintings in the Tomb of the Nasonii, found north of Rome beside the Via Flaminia, given pride of place in the arrangement of the volume (Pace, Reference Pace1979: 130–1). It is perhaps to be identified as the bound collection of ‘pitture antiche miniate’ listed amongst the contents of the Cardinal's library in the inventory drawn up shortly after his death.Footnote ³ Two other albums connected with Camillo Massimo, but known by the names of subsequent owners, St Clair Baddeley and the first Marquess of Lansdowne (William Petty, Lord Shelburne), survive; neither is readily identifiable in the inventory (Pace, Reference Pace1979: 124–9; Pomponi, Reference Pomponi1992: 224 n. 111). The Baddeley Codex, now in the library of Eton College, is a heterogeneous compilation including copies of the Old Testament scenes from Raphael's Vatican Logge, as well as ancient works; it dates probably to the mid-1670s, and some 50 drawings may be missing from it, since the numbered folios begin at ‘li’ (Ashby, Reference Ashby1916: 48–51). The Lansdowne volume, now in the British Library, is a facsimile of the Vatican Vergil that was incomplete at Camillo Massimo's death and was bound posthumously (Claridge and Herklotz, Reference Claridge and Herklotz2012: 28 n. 14). The style of the coloured copies in this is comparable to Bartoli's work, but they are more hesitant in execution and colouring, and also on occasion more faithful to the originals than Bartoli's published engravings of the Vergil manuscript (Reference Bartoli1677). The likely source for these has been identified recently amongst drawings in the Paper Museum of Cassiano dal Pozzo, and the Lansdowne copies have been attributed tentatively to Cardinal Camillo Massimo himself (Claridge and Herklotz, Reference Claridge and Herklotz2012: 20, 23–4, 133); a more than competent draughtsman, he is said to have studied with both Nicolas Poussin and Bartoli (Pomponi, Reference Pomponi1992: 206).Footnote ⁴

The connection with Cardinal Camillo Massimo was an important element in establishing Bartoli's reputation as the foremost copyist of wall-paintings; he was noted especially for his rapid work in situ, recording these frail and often ephemeral relics of Roman antiquity, a work of salvage that the Cardinal, possessed of a keen antiquarian instinct, actively promoted (Pio, Reference Pio, Enggass and Enggass1977: 140; Beaven, Reference Beaven2010: 330–5). To the name of this first patron was sometimes added that of Cavaliere dal Pozzo: not Cassiano, founder of the Museum Chartaceum, that great systematic collection of drawings of antiquities, but his younger brother, Carlo Antonio (1606–89). During his tutelage of the Paper Museum, drawings by Bartoli were added to the collection, although the scale of his patronage can scarcely be compared with Camillo Massimo's, and the volume of Bartoli drawings reported by his biographers to be in the Paper Museum has not been located — the eighteen drawings after paintings, mosaics and the Vatican Vergil attributable to him in the large portion of the Paper Museum surviving at Windsor are possibly the disiecta membra of such a volume (Whitehouse, Reference Whitehouse2001: 48–50). A notable patron in the last decade or so of Bartoli's career was the Director of the French Academy in Rome, who acquired both drawings and engravings for the Cabinet of Louis XIV; the correspondence between Matthieu de la Teulière and his superiors in Paris provides significant information about the artist (Joyce, Reference Joyce1992: 236–7; and see below, pp. 274–5).

Bartoli's work reached its wider public, however, via ‘the archaeological coffee-table books of the seventeenth century’ (Herklotz, Reference Herklotz, Bell and Willette2002: 143), the publications in which his engravings were accompanied by the learned commentaries of Bellori.Footnote ⁵ His senior by 22 years, and Commissario delle Antichità di Roma from 1670 on, Bellori has been seen as a major influence on the style and documentary aim of Bartoli's work (Prosperi Valenti Rodino, Reference Prosperi Valenti Rodino, Borea and Gasparro2000: 133–4). In the early stages of their partnership, he steered him towards the Renaissance masters, above all Raphael (Borea, Reference Borea, Borea and Gasparro2000: 145–6, 149–50), an important factor in setting the aesthetic filter through which ancient paintings were viewed. He also fostered and helped shape Bartoli's enthusiasm for the archaeology of Rome: Bellori shared Camillo Massimo's concern for the investigation and preservation of its ancient remains, and was able to translate this into action as Commissario, exercising this office with great rigour (Montanari, Reference Montanari, Borea and Gasparro2000: 43–4); Bartoli's skills were the ideal complement to his knowledge and scholarship. The first of their publications dealing extensively with paintings was Le pitture antiche del Sepolcro de' Nasonii (1680), featuring the discovery of 1674.Footnote ⁶ Another, Gli antichi sepolcri, illustrating funerary monuments, notably the painted tombs discovered at Rome a few years earlier in the grounds of the Villa Corsini, was in the making when Bellori's death in 1696, before he could write the commentary, brought their collaboration to an end. The volume appeared in 1697, thanks to the editorial intervention of Ivan Paštrić, a process that has been examined in detail by Bruno Gialluca (Reference Gialluca2013).

Despite this setback, Bartoli, who was active in sketching and recording new discoveries in Rome until the late 1690s, planned further volumes, for which some engravings already had been made before he died in 1700. An expanded edition of the 1680 publication, Le pitture antiche delle grotte di Roma e del Sepolcro de' Nasonij, finally appeared in modified form in 1706 with the additional assistance of his son Francesco, and a commentary by Michelange de la Chausse. Francesco Bartoli (1670–1733), his only surviving son and his pupil, continued to produce copies of paintings and mosaics in the same vein (though with less refined style and colouring), both recording new discoveries and using the reference material accumulated by his father (Almagno, Reference Almagno2007). The noteworthy collection of his work formed by Richard Topham, now at Eton College, includes documentation that offers some insight into Francesco's business acumen — a letter from his agent Francesco Fernando d'Imperiali in response to Topham's complaint regarding the visible differences in detail between his copies and those acquired by Thomas Coke of Holkham (Ashby, Reference Ashby1914: 3–4; Connor, Reference Connor1993; Pomponi, Reference Pomponi1994). Apart from changes in pictorial content and the deliberate switching of colours, a particular shortcoming of Francesco's work was the erroneous information on provenance given on some copies, seemingly part of the secretive attitude manifested in this letter, and the strategies he allegedly employed to prevent fraud (Pomponi, Reference Pomponi1994: 261–2, 268–9).

His father's work also was not without its critics: the most palpable charge to be laid against him is his penchant for embellishing the subject-matter, manifest in those cases where it is possible to check the originals against his copies, and extending across the whole range of his work, not just the category of faded or fragmentary ancient paintings where the case for some restoration might have been urged; see, for instance, Gialluca's analysis (2013: 77–84) of the ‘improvements’ in his drawings of Etruscan urns. This drawback was perceived in his own time (Pace, Reference Pace1979: 122–3; and see below, p. 286), and in later assessments of the value of his work, the absence of systematic arrangement in the surviving sets of drawings perhaps also has contributed to the rather negative view of his achievements in recording archaeological discoveries that he saw at first hand; it imposes limitations on their use as evidence. This is all the more frustrating since he apparently left an extensive list of notable finds made in Rome in his day in the manuscript Memorie attributed to him; these were published by Fausto Amidei, in conjunction with a new edition of Famiano Nardini's Roma antica, in Reference Nardini1741, when the manuscript was in the possession of the Marchese Frangipani. They are more familiar, however, in the topographical rearrangement, numbered and lightly edited, subsequently published by Carlo Fea in Reference Fea1790, apparently working from the earlier published text rather than the manuscript. It would be satisfying if we could compile a pictorial record to accompany these archaeological notes, just as a comparison between the Memorie and the excavation licences issued for the relevant dates and sites would be illuminating — as recently demonstrated by Mirco Modolo (Reference Modolo2011) for one group of discoveries noted here (Appendix, 36–40). The role of Commissario delle Antichità passed from Bellori to Pietro Santi in 1694, so he was uniquely well placed to record the archaeological scene, both from his long association with Bellori, whose notes may be incorporated in the Memorie (Claridge, Reference Claridge and Bignamini2004: 40–2), and from his own tenure of the post.Footnote ⁷ On his death in 1700, it duly passed to his son (Ridley, Reference Ridley1992: 133–7; Almagno, Reference Almagno2007).

The Recueil published by Caylus, with a commentary by Mariette, exemplifies the problem posed by random arrangement. It contains reproductions of 33 coloured drawings ‘found by chance in Paris’, as Caylus himself told us in a note inserted in the album in which the mounted drawings are bound.Footnote ⁸ Further details of this discovery, which seems to have taken place around 1756, can be found elsewhere (Restout, Reference Restout1771: 6–7): Caylus was stopped in his tracks in a Parisian street by the sight of some children playing with coloured drawings in their father's shop. Upon enquiry, the man revealed that he had further sheets in the loft; Caylus offered him a louis for each one, secured the little collection, which he recognized as the work of Pietro Santi, engraved and published the pictures, and presented the set to the Cabinet du Roi in 1764, whence it passed eventually to the Bibliothèque Nationale.

The vicissitudes that the pictures have undergone is clear. Their condition is markedly less good than that of the pristine copies in the Glasgow volume, for instance — the paper is soiled, the colours have run in some places, and there are stains and blotches on both sides of the sheets. Given their history, however, it is surprising that they are not more damaged. The album in which they are mounted was assembled immediately after their arrival in the royal collection, and is arranged as a de luxe version of the printed Recueil, with the text of the latter handwritten on the pages preceding the mounted drawings.Footnote ⁹ The drawings have been cut down to within a few millimetres of their borders for mounting (the largest measures 421 × 277 mm), and they are inlaid on window mounts so that the reverse is visible. Appended to the volume is the coloured drawing of the Nile mosaic of Palestrina that Caylus commissioned for his publication from an unnamed Italian artist, and there are also six mounted drawings of paintings and mosaics that were added after the publication of the Recueil in 1757, five of them acquired by Caylus and the sixth presented after his death in 1765. Amongst them are four coloured drawings by Francesco Bartoli, which provide an opportunity to see the difference between his style and his father's:Footnote ¹⁰ Poussin was Pietro Santi's acknowledged teacher in drawing (Pascoli, Reference Pascoli1965: 229), but his fluid sketches seem a long way from the discontinuities of Poussin's workmanlike lines, and the graciously smiling faces that he conferred on his figures, who possess a genial sensuality, are apparent only intermittently in Poussin's work. As Pace noted (1979: 130), the elongated proportions of Pietro Santi's figures evoke the attenuated forms with which Claude Lorrain peopled his visions of classical antiquity. In Francesco's work, the figures retain — to the point of blandness — the smiling faces typical of his father's style, but the bodies lack the elegantly extended proportions. Pietro Santi's soft and subtle modelling of the limbs with tiny strokes of paint, and the contouring of the faces with delicate touches of pink on the rounded cheeks and crimson on the lips, become mannerisms in his son's work, and Francesco extended the crimson modelling to knees, ankle bones, soles of feet and toe joints, the latter articulated to an unusual degree. His pictures also show a significant shift to brighter colours, giving a harder and at times almost garish effect by comparison with his father's palette. He employed repetitive effects of shading, such as red on yellow garments or ochre with body-colour on green, giving a velvety lustre to drapery that complements the plasticity of the figures.

Although the original drawings have received some comment, the set is better known in the published Recueil — ‘one of the most beautiful memorials of French typography’,Footnote ¹¹ a limited edition of 30 copies in which the outlines of the pictures were reproduced in engravings that were then hand-coloured in each copy. A pencil note on the flyleaf of the volume now in the British Library records that three copies were coloured by Caylus himself.Footnote ¹² The book achieved further, if not wide, circulation in an expanded second edition (Mariette and Caylus, Reference Mariette, Caylus and de Tubières1783–7). The importance of these drawings was recognized by Rodolfo Lanciani in his pioneering study of drawings of Roman paintings and mosaics in their topographical context (1895: 166), and subsequently the plates of the Recueil were reproduced in their entirety, with brief notes decocted from the text, by Richard Engelmann (Reference Engelmann1909: 22–5, pls 24–9).

Given the circumstances in which they were found, the copies acquired by Caylus were obviously in total disarray, and they were arranged in a logical order for publication by utilizing the information contained in the inscriptions that many of the pictures carry, together with key numbers and letters. A feature of these drawings is the provision of plans or elevations that set the individual paintings or mosaics in context. This archaeological approach, so different to the work of other copyists who did not regularly work from paintings and mosaics in situ, is typified by Bartoli and Bellori's documentation of the paintings in the Tomb of the Nasonii, and the subsequent volume of Gli antichi sepolcri, in which the monuments or discoveries were treated systematically, with ground-plans, elevations and interior views keyed with numbers or letters that refer to details illustrated subsequently; Bartoli's pride in having assembled a number of plans by ‘diligent architects’ is reflected in his manuscript ‘Motivi’ relating to this publication (Gialluca, Reference Gialluca2013: 43–4). It seems likely that he intended to give similar treatment to other discoveries in the further publications he planned, and the Caylus set may display the remnants of such an endeavour. The drawings have manuscript captions, mostly in gold (which has sometimes oxidized to a dull green), interspersed with letters and Roman numerals in red; on some sheets, the writing is in brown ink. Several of the pictures are framed by red borders or show traces of red around the edge, indicating that a border has been trimmed away.

Caylus and Mariette did the best they could with the information provided by the inscriptions, but there are mistakes in the order devised for the publication –– detailed examination of the pictures, and comparison with surviving Bartoli copies elsewhere (notably those in the Glasgow volume), suggest some ways of reordering the sheets into more coherent (but still lacunose) groups of discoveries. This possibility is confirmed by the identification of further drawings related to the Paris set: the album of Bartoli copies in the drawings collection of the Royal Institute of British Architects (RIBA) in London (Harris, Reference Harris1972: 58–9).Footnote ¹³ Nothing is known of its history before its purchase in 1933, save for two inscriptions on the first flyleaf at the front: ‘F M Musgrave/Tourin’ in sepia pen,Footnote ¹⁴ subsequently erased with some chemical, and above, right, ‘T(?) Keele/ 36 Knights(?) Park/ Kingston on Thames’, in what is probably an early twentieth-century hand, rubbed out.

The volume was rebound in 1960 but retains its original boards covered in light-brown leather decorated with a simple blind-tooled design; the new spine is lettered ‘Bartoli: Roman Deco.’. The page edges are gilded and goffered with a design of alternating Greek key and oval patera, a style that accords with the date of 1813 in the watermark on one of the two flyleaves of Whatman paper. The volume contains 32 coloured drawings by Pietro Santi Bartoli, 26 of which (fols 1–26) are intact (but trimmed) paper sheets that measure 425 × 315 mm as bound (height × width); the lower edge of the drawing on fol. 15, the largest in the set, has been cut away. There follow two leaves of mounted drawings (fols 27 and 28, with two drawings on each) and two further drawings that are now loose but once formed fols 29 and 30.Footnote ¹⁵ The severed remains of these still can be seen in the binding, together with the remnants of the flyleaves and the endpaper, which have been removed from the back of the volume; fols 27 and 28 also bear the marks of aborted cutting. There are also traces of five extra drawings, added to the volume after it was bound but subsequently removed; the vestiges of paint and paper that they have left suggest that they were unrelated to the main Bartoli set.Footnote ¹⁶ The mounted drawings on fols 27 and 28, however, are certainly part of the original Bartoli set, as are the sheets of paper on which they are mounted (see below, p. 275). The drawings themselves are in quite good condition, with a few smudges and marks, and some foxing at the fore-edge of the sheets, where there is also evidence of handling.

The order of the pictures in the RIBA album is random. Almost half of them depict the tombs discovered from 1689 onwards in the grounds of the Villa Corsini at the Porta San Pancrazio in Rome, and partly published in Gli antichi sepolcri (Reference Bartoli1697; Gialluca, Reference Gialluca2013: 50 n. 5). These pictures, which are not themselves in a rational order in the album, are interspersed with other subjects, some explicitly identified by their captions, some enigmatically bearing key numbers, letters or cross-references to other drawings no longer adjacent to them. Like the Caylus set, they have manuscript captions in gold with key letters and numbers in red, or in brown ink; some also have red borders. The drawings in both sets are executed in brown ink with watercolour, body-colour and the occasional touch of gold; black chalk draughts are visible on many, both free-hand sketching and ruled setting-out lines, the latter sometimes impressed with a stylus, as are the guidelines for the inscriptions. In addition, they share two watermarks with the Caylus set: a lamb within a double circle, surmounted by ‘M’,Footnote ¹⁷ and a fleur-de-lis within a double circle surmounted by ‘V’, which appears only on the Corsini tomb series in either set.Footnote ¹⁸

That these drawings were in fact once part of the same set, or sets, is indicated by the presence on the reverse of many of the ‘ghost’ of another drawing, in the form of traces of paint, particularly the rich and thick-textured browns, blues and reds that are so characteristic of these coloured copies. As already noted by Pomponi, with reference to the RIBA drawings (Reference Pomponi1992: 224–5 n. 118), these indicate some earlier arrangement and the presence of other sheets; crucially, they help to confirm the relationship between the London and Paris sets. Sometimes the traces are too faint and dispersed to be diagnostic, but often the outlines of an individual picture are identifiable. In some cases, the ‘ghost’ is that of the next drawing, but more often it can be identified as that of another drawing now placed elsewhere, and in this way cross-links between the Caylus and RIBA sets, as well as transpositions within both sets, can be established. Occasionally the ‘ghost’ is a double image, that of the sheet that currently follows overlying that of another that formerly occupied this place.

Taking the evidence of the ‘ghosts’ in conjunction with the clues provided by the key numbers and letters, the cross-references in the captions, significant visual relationships between the pictures themselves, and the information provided by other Bartoli copies of the same works, it is possible to rearrange the combined Caylus and RIBA sets into orderly groups, the majority of which record discoveries made in Rome. The resulting rearrangement is presented in the Appendix to this paper, where the drawings have been assigned individual numbers (1–65), by which they are identified in the following discussion. Topographical and archaeological information is given in the notes to the Appendix.

In general the connections formed by the ‘ghosts’ do not extend across two different subject groups (unless the ‘ghost’ is that of the drawing that now follows), which suggests that these drawings originally existed as discrete sets, each devoted to a single discovery. This is supported by the distinction to be observed in the occurrence of gold-and-red or brown ink inscriptions: these are uniform within each group, with two exceptions — 5 to 19 (discussed below) and 29 to 32. The gold-and-red inscriptions seem to belong to a grander presentation than those in brown ink: the lettering has been executed carefully, often ending with a decorative flourish. By comparison, the ink inscriptions are less formal, even shakily written in a few cases.Footnote ¹⁹

The united RIBA and Caylus sets thus present a collection of Bartoli drawings apparently accumulated from distinct groups. In their subject-matter these groups reflect his lifetime's oeuvre, but there is a notable representation of discoveries made in the 1680s; the Corsini tombs (43–60, 1689–91) are the latest dated material. At least one set was prepared in response to an ongoing commission: the group devoted to the 1668 discoveries in the ‘Casa di Tito’ (5–19), where the captions change from gold to brown ink halfway through, and the first ink caption (11) begins by restating the details of the find-spot and ends with a reference to the ‘plan already sent with the other pictures’.

Where might such a collection have been formed, and when and where might these drawings have parted company? The dispersal perhaps occurred in the mid-eighteenth century, in the Parisian shop where Caylus purchased his drawings. He had no doubt that the set that he obtained so fortuitously was incomplete, and he may even have learnt from the vendor that others had been sold already, since he wrote: ‘I have reason to believe that the 33 drawings that I have had engraved once formed part of a more numerous collection; I was not lucky enough to reassemble the whole; but it is still quite something to have made such a discovery in France …’.Footnote ²⁰

A collection of 65, as represented by the united Caylus and RIBA sets, is certainly ‘more numerous’, but the total could be still higher, since items known from other Bartoli copies or publications to belong to the groups represented here are missing. Thus, there are no drawings of the mosaic and harbour scene that were part of the 1668 ‘Casa di Tito’ find (5–19: see note c on p. 311); the large Corsini group, 43–60, includes eight drawings that were not published but lacks twelve that were, and there are further preparatory drawings by Bartoli of Corsini material that would amplify the documentation of specific tombs here (see note j to the Appendix). In addition, a few drawings bear on the verso the ‘ghost’ of a picture that cannot be identified among the extant copies but was once part of the collection.

The potential total of drawings is of interest, because — as already noted by Joyce (Reference Joyce1992: 239 n. 99) with reference to the Caylus set — there is a tempting connection to be made with a collection of Bartoli's work that seems to have gone missing in France. These are the drawings bought in Rome through the second half of the 1680s by the director of the French Academy, La Teulière, and sent to his superior in Paris, François Michel Le Tellier de Louvois, Surintendant des Bâtiments du Roi, for the royal Cabinet (Gady, Reference Gady and Bonfait2002: 168–9). It seems they may never have reached it: Louvois died in July 1691, and two years later his successor, Edouard Colbert de Villacerf, was hunting in vain for them. In July 1693, he wrote to La Teulière: ‘From 15 December 1685 until 9 July 1690, I checked your accounts: you sent M. de Louvois 59 drawings of miniatures similar to the latter two which you have sent me. I found not a single one after his death’.Footnote ²¹

La Teulière sent a reassuring reply the following month, to the effect that the missing drawings were undoubtedly in the royal Cabinet, but we find no confirmation of this in the rest of the correspondence. The two new drawings sent to Villacerf were of Corsini tomb interiors, probably those published as plates 9 and 18 of Gli antichi sepolcri, to judge by the descriptions given by La Teulière (Montaiglon, Reference Montaiglon1887–1912: I, 401), and it is clear from the correspondence that some of the missing 59 also belonged to this series, as well as showing the 1683 discoveries near the Sette Sale (cf. 23–7) and some of the Orto Guglielmini find (cf. 36–40).Footnote ²² The ‘missing’ royal collection evidently covered some of the same subject-matter as the united Caylus/RIBA set, and a study comparing the subject-matter with the accounts of payment adds support to the likelihood of the relationship (Modolo, pers. comm.).

The drawings in the RIBA set, which are in better condition than the Caylus drawings, may have remained as a collection of unbound sheets until being made up into the present album early in the nineteenth century. The mount sheets used for fols 27, 28 and 29(1) are in fact surplus blank paper removed from some of the drawings.Footnote ²³ This generous provision of paper was evidently a feature of Bartoli's drawings: writing to Villacerf in May 1693, La Teulière sought to justify his purchase of the two new drawings despite the high price being asked, but noted candidly ‘They are on large sheets of paper, like those I have sent previously, and are not burdened with work …’.Footnote ²⁴

From the letters of La Teulière and Villacerf we learn something of the protracted publication of Gli antichi sepolcri, mentioned in correspondence from 1692 to 1697, and also Bartoli's plans for a third volume, for which twelve of the planned 40 pictures had been engraved, as of September 1697, and that he was optimistic that he had found a commentator to replace Bellori (Montaiglon, Reference Montaiglon1887–1912: II, 335–6, 342). In his last years it seems that his ideas had expanded to a tetralogy of volumes, including a reprint of the Sepolcro dei Nasonii (Gialluca, Reference Gialluca2013: 67 n. 2). But the project remained unfinished; with the loss of his old collaborator in 1696, Pietro Santi seemed to lose heart, and there was also his increasing age and infirmity to check his progress — ‘he is very stout, and is not young’, as La Teulière already had observed in 1693Footnote ²⁵ — as well as the amount of time devoted to engraving other subjects, and his need of immediately lucrative work to support his family. The piecemeal publication eventually produced by Francesco Bartoli and La Chausse in 1706 may be a pale reflection of what Pietro Santi had intended, and the Caylus/RIBA drawings probably give us a better idea of what might have been published if his plans had come to fruition. They include some notable groups of material hardly represented elsewhere: the Sette Sale discoveries of 1683 (23–7), the structures near San Lorenzo in Panisperna (33–5), and the building in the Orto Guglielmini (36–40). The detailed treatment of these, with plans, suggests that these discoveries would have been presented in the same ‘archaeological’ manner as the material in the other volumes.

The basis for such a presentation was the working sketches that Pietro Santi made in situ and for which he was particularly celebrated, and in assessing the merit of his work in recording Roman paintings and mosaics, one set of drawings in particular deserves comment for the light it sheds upon his working practices. This is the volume that once belonged to the writer, collector and aspiring artist Vicente Victoria of Valencia (1650–1709) and is now in the Royal Library at Windsor Castle. The contents were listed by Adolf Michaelis (Reference Michaelis1910: 111–22),Footnote ²⁶ and the volume was described briefly by Anthony Blunt (Reference Blunt1967) in a note on its former owner. Victoria arrived in Florence in 1675 as a member of a Spanish monastic order, but moved a few years later to Rome, where he remained until 1686, studying painting with Carlo Maratta and living in the Piazza di Spagna.Footnote ²⁷ He then returned to Spain as a Canon, but was back in Rome from 1698 onwards. Over the course of his two Roman sojourns, his circle of acquaintances included Bellori, Bartoli father and son, and La Chausse. His manuscript Academia de Pintura del Señor Carlos Maratti, conceived as a series of dialogues between Bellori, Maratta and the author, opens with an idealized description of Victoria's home in Rome, housing a collection of drawings and prints including ‘the most intriguing of all … the [volume] of ancient Paintings of pagan times copied in colour in miniature style from various ruins of Rome which the enthusiasm, and the vigilance of Pedro Santi Bartoli was able to accomplish over many years’.Footnote ²⁸ This is followed by a list of volumes of engravings of reliefs and sculpture, mostly by Pietro Santi and identifiable with his works published jointly with Bellori, such as those on the Colonna Traiana (1673) and the Colonna Antoniniana (1679). The terms in which the ‘ancient paintings … in miniature style’ are described, however, recall the kind of coloured copies commissioned by Camillo Massimo; Victoria perhaps aspired to ownership of such an album in his ideal collection — at the time at which the Academia was apparently composed (1686–90), the only published Bartoli volume of paintings was the Sepolcro dei Nasonii (1680), and Victoria did not as yet possess the album of drawings now in Windsor.

Most of the drawings in the Vittoria album, as it is known from the Italianized form of his name which appears on the title-page (Fig. 1), are preparatory material for finished work that appeared elsewhere. The title-page is itself the draught by Pietro Santi Bartoli and Camillo Massimo for the 1674 album of drawings now in Glasgow, converted by Victoria for his own use;Footnote ²⁹ his handwriting may be recognized in the inscriptions on the column plinths.Footnote ³⁰ As Blunt observed, the collection is heterogeneous but a substantial number of the drawings are the work of Pietro Santi, bearing out the title on the spine, ‘Pitture antiche Dise. da Piet. Santi’; Bassegoda I Hugas (Reference Bassegoda I Hugas1994: 53) suggested the likely presence of some of Victoria's own work, too. The binding of the album is Victoria's, and although eight pages have been removed and a few show signs of the removal or transposition of drawings,Footnote ³¹ the contents appear to be substantially the same as when it was bound. As noted by Blunt (Reference Blunt1967: 32), the latest dated drawing belongs to the year 1698.Footnote ³² In addition to the Bartoli working drawings, the album opens with a set of engravings of the wall-paintings in the Pyramid of Gaius Cestius, enhanced with wash and white heightening (Michaelis, Reference Michaelis1910: 112 fols 2–7). It also contains a group of cinquecento drawings, including sketches of decoration in the Domus Aurea (RL 9567–8; Dacos, Reference Dacos1969: 39), which could have formed part of Bartoli's reference stock, assuming that they were not collected by Victoria himself. The order followed for these early pages in the volume is, significantly, that of the chronological survey of surviving Roman paintings, Delli vestigi delle pitture antiche dal buon secolo de' Romani, compiled by Bellori and published in 1664 as an appendix to the notes on the galleries and collections of Rome (Zocca, Reference Zocca1976: 122–37, reproducing pp. 56–66 of the original).Footnote ³³ It seems that Victoria may have been planning a publication of his own on ancient paintings in conjunction with Domenico de' Rossi: in his manuscript Indice dell'opere di Raffaello Sanzio d'Urbino (dating to 1703), he refers to ‘my Book of Ancient Paintings, which is currently being engraved for publication’.Footnote ³⁴

Fig. 1. Title-page of the Vittoria album, drawn by Cardinal Camillo Massimo and Pietro Santi Bartoli, and adapted by Vicente Victoria. Inscribed: (left-hand plinth) L'Architettura e inventione, e Disegno del Em ^moSig. Card ^l. Massimi. (right-hand plinth) Le Vittorie laterali sono disegnate da Pietro Santi Bartoli. Pen and ink and wash; Vittoria album, Royal Library, Windsor Castle, RL 9566. (Reproduced courtesy of the Royal Collection Trust © Her Majesty Queen Elizabeth II 2014.)

Among the sketches in the Vittoria album are the original drawings executed in situ at the Tomb of the Nasonii, including one inscribed with the Incidatur for the printed edition (RL 9610; Whitehouse, Reference Whitehouse2001: 300 fig. 33). The 1702 publication of the tomb included on plate II a fulsome acknowledgement of Victoria's ownership of these that did not appear in the republication of the engravings in 1706 (cf. Rudolph, Reference Rudolph1990: 245–6 n. 81). The volume also contains a celebrated pen-and-ink drawing of a figure panel in the ‘Volta degli Stucchi’ of the Domus Aurea (RL 9573; Weston–Lewis, Reference Weston-Lewis1992: 313 no. 2). The drawing, identified in the seventeenth century as depicting Coriolanus and his womenfolk, traditionally has been attributed to Annibale Carracci and assumed to be the Carracci drawing of this subject known to have been in Bellori's possession. It has been suggested, however, that it is in fact a copy of Bellori's drawing, since the latter is documented as having passed from him to the collection of Padre Resta and thence via various collectors to Jonathan Richardson (Wood, Reference Wood1996: 11, 44–5, appendix I, 1).Footnote ³⁵ Victoria's drawing, however, enjoyed the reputation of being the Carracci ‘Coriolanus’ and was cited as such in Le pitture antiche (1706).Footnote ³⁶ Following Victoria's death, a large part of his collection of prints and drawings (45 volumes) was acquired by Pierre Crozat in Rome in 1715 (Montaiglon, Reference Montaiglon1887–1912: IV, 377 and cf. p. 368), and the Carracci ‘Coriolanus’ subsequently was reported to be in his possession (Du Bos, Reference Du Bos1719: 343) though neither the drawing nor the Vittoria album can be traced in the later dispersal of the Crozat collection.Footnote ³⁷ An alternative hypothesis is that the album was one of the 49 volumes from Victoria's collection that had already passed to the Albani collection in 1713, and thus subsequently reached the Royal Library at Windsor with the Dal Pozzo Paper Museum, purchased from the Albani family in 1762 (Blunt, Reference Blunt1967: 32; Lyons, Reference Lyons, Fejfer, Fischer-Hansen and Rathje2003: 483–4). The presence of the ‘Coriolanus’ drawing in the album, however, makes this otherwise plausible scenario problematic.

It seems likely that the drawings in the Vittoria album belonged substantially to Bartoli's studio stock, a large part of those ‘original sketches of everything that he had painted and engraved for [Cavalier del Pozzo and Cardinal de Massimi]’, mentioned in a letter of 1693 from La Teulière to Villacerf.Footnote ³⁸ They might have been transferred to Victoria at or just before Pietro Santi's death in 1700, though the transfer of such an important body of reference material contrasts strongly with the jealous attitude towards his notes displayed by Francesco Bartoli when his own career was established, as evidenced in the letter of 1730 to Topham (see above, p. 268). The removal of the acknowledgement of Victoria's ownership of the Nasonii drawings from the 1706 publication might be read as denoting some pique on Francesco's part, possibly also some controversy regarding Victoria's own plans for a publication, but relations between the two seem in fact to have been cordial — Francesco engraved some of Victoria's work (Bassegoda I Hugas, Reference Bassegoda I Hugas1994: 50 fig. 15), and seems to have had continuing access to the reference material in the Vittoria album after Pietro Santi's death, possibly with a view to a joint publication with Victoria.

The preparatory nature of these drawings was recognized by Thomas Ashby (Reference Ashby1914: 5): they are sketches with draught strokes, corrections and annotations, including colour references (some lost when the drawings were trimmed for mounting). The colour terms used by Bartoli are: bigio, cinabro, giallo, lacca, pavonazzo, rossiccio, rosso, torchino, verde, sometimes qualified by epithets such as chiaro; they are more often abbreviated as single letters or contractions than fully written out. A few comparable drawings are to be found in the volumes at Holkham, as Ashby observed, and may have belonged to the same reference collection.Footnote ³⁹ Two of them carry the enigmatic letter ‘m’ which appears on many items in the Vittoria album, on the drawings themselves (Figs 2 and 3), or occasionally on the mount sheet. The letter cannot stand for ‘M[assimi]’, as suggested by Michaelis (Reference Michaelis1910: 114 fol. 29), because some of the drawings with which it appears show discoveries made long after Cardinal Camillo's death and the end of active Massimi patronage, as observed by Beatrice Cacciotti (Reference Cacciotti and Palma Venetucci2001: 39 n. 43). Two Holkham drawings that have been attributed to Francesco Bartoli and apparently relate to wall-paintings found under the Palazzo Rospigliosi in 1709 would furnish the latest date associated with this letter, if the Rospigliosi attribution is correct (Holkham II, fols 53–4; Ashby, Reference Ashby1914: 46; Connor Bulman, Reference Connor Bulman1999: 205–6, fig. 2; contra: Modolo, Reference Modolo, Bruschetti, Giulierini, Swaddling, Gialluca and Reynolds2014: 165–70). The ‘m’ might signify ‘memorie’, or ‘museo’, two words used with some frequency in connection with the antiquarian records of the time, both of them present on Victoria's title-page for his album (Fig. 1). It thus might indicate some selection for publication being made by Francesco, possibly in conjunction with Victoria, though the latter died in May 1709, just as the first Rospigliosi discoveries were being made. The ‘m’ group (45 drawings in all, comprising both chalk and pen-and-ink draughts) does not consist solely of unpublished material: ten of the drawings show subject-matter in the Corsini tombs engraved for Gli antichi sepolcri, and six others correspond to plates in Le pitture antiche (1706). Footnote ⁴⁰

Fig. 2. Detail of a wall-painting, with colour notes; possibly part of the same discovery as 23–32. Inscribed: Pittura nella facciata di una stanza nel orto delle sette sale scoperta l'anno 1683; in the upper right corner, m. Pen and ink with black chalk draughting; Vittoria album, Royal Library, Windsor Castle, RL 9680. (Reproduced courtesy of the Royal Collection Trust © Her Majesty Queen Elizabeth II 2014.)

Fig. 3. Draught plan of the Villa Corsini cemetery, with annotations, on two sheets of paper with a horizontal join. Inscribed: (below the join, left-hand side, partly cut away) sepolcri scoperti il mese di Xbre 1689; (on the mount sheet, below the drawing) Queste piante con le seguenti pitture sino al N ^o106 furono trovate nella villa Corsina fuori la porta di San Pancratio L'anno 1690. Pen and ink; Vittoria album, Royal Library, Windsor Castle, RL 9644. (Reproduced courtesy of the Royal Collection Trust © Her Majesty Queen Elizabeth II 2014.)

The greater part of the Holkham drawings, whether finished coloured copies or preparatory sketches, may be attributed to Francesco, although there are some in either category that appear to be the work of Pietro Santi (Figs 4 and 5),Footnote ⁴¹ as well as a few that are the work of neither Bartoli. Distinguishing their hands in the preparatory sketches as opposed to the finished work is not so easy; Ashby suggested a simple (perhaps too simple) criterion, that the ‘pencil [i.e. black chalk] sketches with colour notes’ were Francesco's work (Ashby, Reference Ashby1914: 3). This would by default characterize the pen-and-ink sketches (which often have chalk draughting lines) as the work of Pietro Santi, typified by the series devoted to the Tomb of the Nasonii, a discovery made when Francesco was only four years old. At that time, Pietro Santi was living in the household of his father-in-law, Giovan Francesco Grimaldi ‘il Bolognese’, who may have been responsible for the picturesque frontispiece to the Sepolcro de' Nasonii set (Whitehouse, Reference Whitehouse2001: 298 fig. 31; attribution suggested by Lanciani (Reference Lanciani1895: 190)), and with whom he was involved in other enterprises (see below). It seems likely that Francesco, who lived at home until his first marriage at the age of 22 (Almagno, Reference Almagno2007: 454–5) would have assisted his father in the recording of discoveries, even though his name is absent from the contemporary records, a fact that may reflect the commercial desirability of branding all the work as Pietro Santi's, as well as the latter's tight control over the workshop and its output. Noting the huge volume of demands made on him, Maximilien Misson (Reference Misson1714: II.i, 172), an appreciative visitor to Pietro Santi at home, commented ‘… if he had an hundred Hands, they would scarce be enough for him’. We should consider the degree to which those hands were not Pietro Santi's own, albeit unacknowledged. As his father's pupil, Francesco's draughting style would have been similar to Pietro Santi's, and this might be true also of the handwriting with which the sketches are annotated, which is somewhat uniform whether they are executed in chalk or pen-and-ink. Adolf Michaelis (Reference Michaelis1910: 118 fol. 75) tentatively suggested Francesco's as the hand of one of the Villa Corsini pen-and-ink sketches (RL 9650: Fig. 6), in which he saw a difference in both the handwriting and the ‘rather coarse’ drawing; Pomponi (Reference Pomponi1992: 210) noted the possibility that Pietro Santi's brothers-in-law, as well as Francesco, assisted him.

Fig. 4. Quarter detail of the ceiling in tomb ‘Q’ of the Villa Corsini cemetery, with colour notes and details of the other three canephorae that support the central tondo (cf. 56). Inscribed: Pittura di una volta sepolcrale nella villa di Monsig ^reCorsini fuori Porta S. Pancratio / segnata nella pianta alla littera Q. Pen and ink with black chalk draughting; Library of Holkham Hall, vol. II. fol. 34 (85). (Reproduced by permission of Viscount Coke and the Trustees of the Holkham Estate.)

Fig. 5. A brick stamp and a cinerary urn with measurements in dita (inches), from the Villa Corsini cemetery (unpublished). Inscribed: (top right) Tevolone con il presente merco. trovato ne sepolcre della Villa Corsina; (bottom) Vasetto ove erano le cenerij di un picciolo putto. mede[si]mamente della Villa Corsina. Pen and ink with black chalk draughting; Library of Holkham Hall, vol. II. fol. 39 (106). (Reproduced by permission of Viscount Coke and the Trustees of the Holkham Estate.)

Fig. 6. Lunette in tomb ‘C’ of the Villa Corsini cemetery (cf. 47). Inscribed: pittura anticha; trovata in un sepolcro nella in una vignia fuori della porta di san Pancratio, posseduta dal ill ^moMo ^reCorsini: trovata ó scoperta lanno 1689 il mese decembre, largo pallmi [sic] .ii. longo 22 17. Colour notations with pointers; in the upper right-hand corner, m. Pen and ink with black chalk draughting; Vittoria album, Royal Library, Windsor Castle, RL 9650. (Reproduced courtesy of the Royal Collection Trust © Her Majesty Queen Elizabeth II 2014.)

From the sketches with colour notes of the Vittoria album to the highly-wrought and exquisitely coloured finished works exemplified by both the Glasgow album and the Caylus/RIBA set is a long step. The sketches are fluid but perfunctory — the elaborate décors recorded for the vaulted ceilings of the Corsini tombs, for instance, are regularly represented by quarter details only, with further elements sketched beside them for reference. The colour notes are meagre indications when compared with the subtle splendours of the finished versions, and the colours are in any case sometimes arbitrarily changed in the latter (see, for instance, Appendix 12, 13, 15, 38 and note c, and cf. Pomponi, Reference Pomponi1992: 208–9). It is worth in conclusion examining the implications of the expression ‘ancient pictures in miniature’.

In his finished drawings Pietro Santi did indeed miniaturize ancient paintings and mosaics that in themselves were often of considerable size. The addition of framing borders — gold in the albums of the 1670s with Massimi bindings, and red for some of the paintings shown in the Caylus/RIBA set — increases the miniaturized effect. The existence of framing or demarcating borders around the separate elements in Roman wall-paintings is an actual feature, and red with a white fillet is one of the most common forms;Footnote ⁴² but the addition of red borders to many of Pietro Santi Bartoli's copies brings them close to the red-bordered miniatures of the Vatican Vergil (Wright, Reference Wright1993: 1–2). Perhaps this authentic example of ancient miniature painting was a determining influence in the formation of his copying aesthetic.Footnote ⁴³ The use of red borders in Bartoli's later work may reflect also an economical step down from the ‘de luxe’ Massimi albums.Footnote ⁴⁴ Reduced and thus framed, wall-paintings and mosaics also became ‘ancient miniatures’, worthy to be assembled in fine volumes for connoisseurs; and where paintings in particular were concerned, the effect was heightened by his tendency to complete and embellish the often defective ancient original. This trait was already noted critically in his own time, famously by Misson (Reference Misson1714: II.i, 172) — ‘It is true, that he designs so well, that he cannot resolve with himself to design ill: and from thence it proceeds, that we cannot always be certain that the antick venerable Figures he has engraven have not been mightily embellish'd by his Tool’. Jonathan Richardson the Younger was shocked at the sight of the ‘altogether Gothic’ illustrations of the Vatican Vergil compared with Pietro Santi's engravings thereof, from which ‘One would imagine the Pictures to be of the Best Antique …’ (Richardson, Reference Richardson1722: 264–5).

Another potential influence on the way such subjects were treated was the use to which the copies might be put. Bartoli's records of the Corsini tombs are exquisite in their refinement of detail and colouring, notably in the rendering of the elaborate decoration of the vaulted ceilings. We may doubt that the originals shared the perfection of his coloured copies, but by his own admission in Gli antichi sepolcri, which also included architecture and reliefs, such monuments were fit to serve as ‘the most noble models, and exemplars for the disciplines of Architecture, Painting, and Sculpture …’. Footnote ⁴⁵ He himself may have experienced the utility of such sources earlier in his career, assisting his father-in-law with the decoration of the vaulted ceiling of the Galleria in the Palazzo Borghese.Footnote ⁴⁶ We know that at least one of the French purchases was intended as the model for a painted ceiling (above, n. 22), and the work of both Pietro Santi and Francesco certainly nourished neoclassical designs in eighteenth-century England (Bristow, Reference Bristow1996: 78–123; Aymonino, Gwynn and Modolo, Reference Aymonino, Gwynn and Modolo2013). The drawings were perhaps worked up with this function in mind.

In any case, Pietro Santi Bartoli was ever confident in perfecting and completing in his records the picture suggested by the fragmentary evidence before him. As Licia Luschi (Reference Luschi1992: 11) observed in her study of his drawings of the painted cupola of the ‘Mausoleo dei Gordani’, the intention was not to deceive but to offer a reasoned reconstruction. In looking at his fluent preparatory sketches, we sense his faith in his ability to recreate the original work from a knowledge of Roman art that was built on both instinct and experience. Yet the paradox remains — more than any other artist in this field, Bartoli, working in situ, was able to record the total context of a discovery and catch the elusive survivors in paint, stucco and mosaic before their destruction or decay, and much emphasis was placed on this aspect of his work by his admirers (Whitehouse, Reference Whitehouse1995: 241–2); yet once the record was captured, he seems to have transmuted it into an art form in its own right. The limitations thus placed upon the value of his drawings must be remembered whenever we use them to fill the gaps in our knowledge of ancient Rome. With this caveat in mind, we may none the less be grateful for the records that he left, amongst which the Caylus/RIBA drawings form a valuable and impressive series; moreover, they point to the desirability of examining in more detail other volumes and portfolios of mixed content from which further records may be reassembled, ‘lost’ drawings recovered, and original sequences restored.Footnote ⁴⁷ As La Teulière wrote to Villacerf in 1693, bemoaning the almost simultaneous destruction of wall-paintings at the moment of their discovery, ‘Without Pietro Santi's devotion, all but the memory of everything therein would be lost’.Footnote ⁴⁸