In the Lanciani collection, there is a drawing by Francesco Patrizi (1826–1905) (Fig. 1), who, with his brother and father, rebuilt the Villa Patrizi after 1849 and who was an important patron of contemporary painters. With his business partner Antonio Bonfigli he constructed a large building composed of artist's studios on Via Margutta (Moncada di Paternò, Reference Moncada di Paternò2012). One of his most notable commissions was the redecoration of the main room of the second piano nobile at Palazzo Patrizi opposite S. Luigi dei Francesi in Gothic revival style, with illusionistic paintings by Pietro Aldi (1852–1888) (Marshall, Reference Marshall2015: 252–7). He was an amateur draftsman, if not a particularly proficient one, making, for example, various drawings for projects for the subdivision and redevelopment of the land around the Villa Patrizi in the years before it was swallowed up in the urbanization of post-unification Rome (Marshall, Reference Marshall2015: 362–74). The Villa Patrizi was the site of important antiquities, including a cemetery described by Antonio Bosio in 1601, which appears frequently in Francesco's maps of the villa (Marshall, Reference Marshall2015: 91–3).Footnote 1

Fig. 1. Francesco Patrizi, Tomb Excavated in the Vigna Casali, 1871–2. 266 × 189 mm. Rome, Biblioteca di archeologia e storia dell'arte, Lanciani Collection, Roma XI.20.XII.3. Catalogued as Vigna Casati, but the inscription clearly reads ‘Casali’. Online at Stanford University, http://purl.stanford.edu/mw456nd6113. [Images can only be printed by using this page and for personal/scholarly use. Any other use must be requested in writing to Istituto Nazionale di Archeologia e Storia dell'Arte: inasa@inasa-roma.it]
Patrizi was clearly interested in other archaeological excavations in nearby villas, and the drawing in the Lanciani collection, catalogued as ‘Via Appia, Sepolcreto nella Vigna Casati’, shows an excavation of the tomb at the Vigna Casali.Footnote 2 This vigna was inside the walls near Porta San Sebastiano (Figs 10 and 11 further below). The Casali had another property within the walls, a villa on the Caelian near S. Stefano Rotondo, the Villa Casali, which is not well known to students of Roman villas as it was destroyed in 1884 (Santolini Gordiani, Reference Santolini Giordani1989: 38–49 and fig. 38).
The drawing, according to an inscription in pencil in Francesco Patrizi's hand,Footnote 3 was made ‘on site’ (sul posto) ‘around 1870’ (the excavations at Vigna Casali actually took place in 1871–2).Footnote 4 This inscription is clearly later than the drawing, and similar inscriptions designed to inform others about the nature and purpose of a drawing are found in his drawings of projects for the Villa Patrizi and date from the late 1880s and probably later. The excavations were undertaken by Marchesa Carlotta Faustina Casali and Marchese Raffaele Casali del Drago (Santolini Giordiani, Reference Santolini Giordani1989: 72). There is a contemporary description of the excavation by F. Brizio (Reference Brizio1873). In a pen inscription evidently made at the same time as the image, Patrizi wrote that ‘e come con errore inqualificabile fu denudato asportando altrove ogni cosa’ (by an unspeakable mistake it was denuded of its contents with everything being taken elsewhere). In fact, the antiquities were taken to the Villa Casali al Celio. A large marble sarcophagus, abandoned by earlier excavations, had been partly exposed for some time. In the attempt to extract it, the tomb was discovered (Brizio, Reference Brizio1873: 14; Santolini Giordiani, Reference Santolini Giordani1989: 72).
The tomb, the final stage of which dates from the third century AD, belonged to the Sempronii family and had originally been constructed above ground (Borg, Reference Borg2013: 126, 127). Many, but not all, of the antiquities drawn by Patrizi survive today and were catalogued by Santolini Giordani (Reference Santolini Giordani1989) in Antichità Casali: la collezione di Villa Casali a Roma, which draws on the Casali archive.
A watercolour plan with elevations of the site made at the time of the excavations in 1871–2 in the Casali (Fig. 2, with the section lines emphasized) shows the space drawn by Patrizi at the upper left of the long elevation, with an elevation drawing of the end wall corresponding to the stepped line C–D on the plan that corresponds approximately to Patrizi's viewpoint at upper left. The diagram in Figure 3 shows the Patrizi drawing with the short elevation and surviving antiquities to scale. The details of Santolini Giordani's reproduction are difficult to read, so for clarity I have combined freshly photographed details of the excavation watercolour with photographs of the pieces in their present state in the illustrations that follow.Footnote 5 By comparing the two with Patrizi's drawing I shall attempt to establish the accuracy of Patrizi's drawing as a record of the tomb and its contents at the moment of its discovery. This is no simple task as Patrizi was not a particularly skilled draftsman and it is not always clear what he intended his pen strokes to describe. More importantly, Patrizi on occasion shows something that was clearly not the case, suggesting that his drawing may have been done at least in part from memory, in spite of his (later) assertion that it was done sul posto.

Fig. 2. L. Stoll, Tomb excavated in the Vigna Casali in 1871–72. Watercolour. Archivio Casali. (Author, with section lines emphasized.)
The contents of the tomb
The strigilated sarcophagus
A case in point is the strigilated sarcophagus in the middle of the drawing (Fig. 4).Footnote 6 This is now in the garden of St Paul's American Episcopal Church in Via Nazionale, and was in the hands of the dealer Scalambrini in 1888.Footnote 7 This has a temple tomb motif that signifies the Porta Inferis, or entrance to the underworld. In the sarcophagus today these doors are closed, and the watercolour also shows this. Patrizi, however, seems to show them open. In the sarcophagus the door panels are set forward of a ground plane which generates a shadow between them. Possibly Patrizi interpreted this to signify that the left panel was in front of the right, and hence open. This might be the result of working from memory; but on the other hand, he is careful to differentiate between the upper right panel, which has a rectangular relief, and the left, which does not. On the lower panels there are marks that may be rudimentary notations of the representations of metal rings on the sarcophagus, which the watercolourist, who was presumably not working on site, has omitted, turning a rather messy four-panel set of doors into a tidy six-panel set. The watercolourist displays a greater conceptual understanding of the pediment, identifying that it has a fastigium and two reclining figures. Patrizi clearly indicates the central wreath of the fastigium and there are some marks that may have been prompted by the reclining figures, but these become involved with lines indicating the strigilation. These lines are done as fine lines, probably first, which are reinforced by thicker lines, which could have been done at a later time but are primarily an attempt to indicate shadows and so may have been done at the same time. The curvature of both sets of lines is the reverse of what they should be: is this a reliance on memory, or is it the result of a conceptualization triumphing over observation? For a right-handed person, drawing them this way is slightly easier than the correct way, and it may have been enough for Patrizi, working in a hurry, to indicate that the sarcophagus was strigilated; the precise form of the strigilation may not have mattered to him. Similarly, the existence of the hole in the strigilation at the left may have mattered more than its precise shape, which would explain why this differs from the shape shown by the watercolourist, which may be equally arbitrary.

Fig. 4. Unknown, Strigilated Sarcophagus with the Porta Inferis, Roman, mid-third century AD. Rome, Church of St Paul's within the Walls, garden. Santolini Giordani, Reference Santolini Giordani1989: 140–1, cat. 93. Top: detail of Figure 1. Centre: photograph, after Santolini Giordani, Reference Santolini Giordani1989: tav. XVII. Bottom: detail of Figure 2.
The sarcophagus of the Muses
This type of strigilated sarcophagus was in use by both pagans and Christians in the third century AD, and because of its position Borg (Reference Borg2013: 130) argues that it predates the sarcophagus with the Muses that was situated above (Fig. 5). The latter sarcophagus, again in 1888 with the dealer Scalambrini, is now in the Nelson-Atkins Museum in Kansas City.Footnote 8 Borg (Reference Borg2013: 128) dates this to not earlier than the 260s AD on the basis of the women's hairstyles.Footnote 9
There are eleven figures in the relief (five and four muses respectively to the left and right of Apollo and Minerva), which makes it hard to keep track of them all, and Patrizi only manages to show nine. One of the few that are clearly identifiable is the fifth muse from the left (the fourth in the drawing) who holds a kithara, which Patrizi indicates with a reverse curve like the strigilation in the sarcophagus below. The graph-like zigzag lines below seem to be an attempt to indicate the drapery folds over the muse's raised knee. Two notations consisting of broken arcs centred on a circle are evidently two of the masks, the lower one corresponding to the mask between either the first and second or second and third muses, while the upper is between the third and fourth. There is an explosion of lines in the upper part of Patrizi's second figure, which may be a response to the complexity of the tight grouping of two heads of the second and third muses on either side of a mask. It is possible that a raised prong between Patrizi's sixth and seventh muses is a response to the second kithara between the ninth and tenth figures, or perhaps Minerva's staff. The remaining three figures to the right are similar and help us to understand his graphic process, which is a kind of doodling that only occasionally, when prompted by distinctive forms like a kithara or mask, develops into something specific. There are round loops at the top indicating heads, doubled loops below to indicate shoulders or breasts, more vertical squiggles with some curvature for the body below the waist, and finally a downward or up-and-down movement of the pen to indicate the legs. This may be contrasted with the careful, economical and accurate notations of the watercolourist, who was probably distilling a careful and unhurried drawing or photograph of the sarcophagus.
The sarcophagus of the lion hunt
At the left of the sarcophagus of the Muses in the Patrizi drawing is the right half of a sarcophagus with a lion hunt, now in the Ny Carlsberg Glyptothek (Fig. 6).Footnote 10 This is dated by Borg to the 240s AD on the basis of its patron's portrait, which resembles the Emperor Balbinus. Because of the placement of the sarcophagi, Borg (Reference Borg2013: 128) proposes that this area ‘was used by the senatorial family of M. Sempronius Proculus Faustinianus over a couple of decades, from at least the mid- to late-Severan period to the 260s’. In the excavation plan is visible what is evidently the plan of this sarcophagus, showing exactly the same degree of embeddedness in the ground (Figs 2 and 3).
In Patrizi's drawing, the figure of the lion is clearly identifiable, with a circle to indicate facial features. The arc of the lion's hindquarters is repeated above but without a clear referent, although the forms above may have been suggested by the horse's head facing right. In the watercolour this relief is in shadow and obscured by the statues in front and it is hard to decide whether it attempts to represent the figure at the far right of the sarcophagus or whether the figures are generic.
The headless ‘Victory’
In front of the lion hunt sarcophagus, Patrizi shows two headless statues, one in front of the other, both of which are visible because of his high viewpoint. These two statues are given catalogue entries by Santolini Giordani, who draws both on Brizio's account and on descriptions in Friedrich Matz and Friedrich Karl von Duhn's Antike Bildwerke in Rom mit Ausschluss der grösseren Sammlungen, published in 1881. It is evident that Matz and von Duhn's descriptions are informed by a knowledge of the statues provided by sources other than Brizio. Brizio places these statues in the long middle room, but it is possible that this is an error on his part.Footnote 11
The representations and descriptions of the first statue (Fig. 7, centre and right) are somewhat contradictory, which raises the question of whether all are dealing with the same piece. Although the two representations seem to differ at first sight, they can be reconciled, except for the fact that the excavation watercolour shows the statue with a head, and Patrizi does not (Fig. 7, centre and right). On Patrizi's drawing we see two arcs near the armpits, which are perhaps his attempt to indicate the crossed straps of the statue's mantle. The watercolour shows the hands emphatically cut off at the forearms not too far from the elbows. Patrizi continues his right arm beyond this, at a somewhat awkward angle. Does he mean to show an intact forearm? The lines of his left arm continue downwards, as if the arm is intact and hanging straight down, and there is a form that spreads out behind. This must be a confused attempt to render the drapery hanging over the left arm of the statue in the watercolour. A couple of lines correspond to the diagonal folds of the mantle, while vertical lines beneath seem to correspond to the chiton beneath. The curvature of the lower body at the left corresponds to the shadow line of the watercolour, which suggests that Patrizi is responding to the same shadows as the watercolourist; the awkwardly bent right forearm might be a consequence of this as well. The watercolour makes it clear that under the hem of the chiton the statue becomes a tapered rectangular or conical shaft on a rectangular base.
Brizio describes an ‘archaizing’ Victory ‘di mediocre lavoro’ lacking head and arms, in the form of a draped female figure who is about to take flight and whose feet are placed on a base in the form of an inverted cone.Footnote 12 While the inverted cone could be identified with the above-mentioned shaft, the watercolour shows no feet. Matz and von Duhn state that the legs were close together and the feet were placed on a ball (Kugel) rather than Brizio's ‘rialzo a forma di cono rovesciato’.Footnote 13 This sounds like the common image of a flying Victory on a ball, but the excavation watercolour gives no hint of this, and seems to be a more static kind of statue. The statue in Patrizi's drawing, by contrast, could more plausibly be read as a Victory about to take flight, as the lower parts are obscured by the statue in front. Matz and von Duhn state that she wore a chiton with small folds and short sleeves that fitted snugly to the lower parts of the body but spread out behind, which is compatible with the representations, as is their statement that the outstretched left forearm hung a himation that covered the lower part of the body as far as the knees. They note that two deep holes behind the shoulders indicated the attachment points of the wings, which were lost, and which were held by two crossed bands on the chest, perhaps those noted above.
Santolini Giordani (Reference Santolini Giordani1989: 105) states that the antique portrait head was reattached, which explains its absence in Patrizi's drawing and implies that a certain amount of time elapsed between the excavation and the making of the excavation watercolour, and raises questions about the reliability of this drawing as a record of the tomb as found. According to Matz and von Duhn ([1881] Reference Matz and von Duhn1968: 248), the hairdo indicated a date in the second century AD and the head had enlarged eyes, which, they say, was a sign of the ‘archaism’ mentioned by Brizio (Reference Brizio1873, 16), although enlarged eyes are more commonly associated with Late Antique works than archaistic ones. The excavation watercolour gives no particular emphasis to the eyes, but shows hair parted in the centre as can be found in late Antonine and Severan sculpture.
The headless herm of a divinity
The statue in front as drawn by Patrizi seems to be a headless draped figure, possibly male, with its left arm slightly raised (Fig. 7, left) (Santolini Giordani, Reference Santolini Giordani1989: 107, cat. 31). This must be the headless herm of a divinity wrapped up in a cloak described by Brizio.Footnote 14 Given Brizio's information, it is apparent that the lower part was a tapering shaft, with the hem of the cloak arranged diagonally. It can be identified with the object in front of the ‘Victory’ in the excavation watercolour, which clearly has a tapering shaft, although the draperies are quite different and it hardly looks like a human figure. According to Matz and von Duhn ([1881] Reference Matz and von Duhn1968: 1638), the sex of the figure was not visible, and although the statue was headless when found, it was reunited with its antique head of the Antinous type, which was young and beardless, and had a wreath in the soft-looking hair. The implication is that the head was found on site. It is curious that although the excavation watercolour showed a restored head for the other statue, it does not do so here.
Brizio (Reference Brizio1873: 16–17) observed a bag hanging from the statue's left arm which allowed him to identify the figure as Mercury, who is sometimes shown carrying a purse or money-bag. Matz and von Duhn considered this to be a misunderstanding and argued that this feature was in fact the hem of the cloak that the figure holds in its left hand, which was also held back by the right hand across his chest.Footnote 15 Patrizi's drawing, however, shows something that does look like a bag hanging from the left arm (which is cut off at the wrist), while the right hand seems to hang down vertically, but Matz and von Duhn were presumably correct in stating that there was no bag. Does this mean that Patrizi's representation is somehow shaped by an interpretation shared with Brizio? Or is this an instance of his exploratory doodling producing a ‘bag’ accidentally?
The Dionysiac sarcophagus of Maconiana Severiana
The sarcophagus shown by Patrizi in perspective at the right is the Dionysiac sarcophagus of Maconiana Severiana, now in the Getty Museum and extensively published (Fig. 8).Footnote 16 It is placed on a brick or tile shelf in a recess in front of three round-headed niches. This bears little relation to the more complex scheme shown in the excavation plan, where the sarcophagus is set hard against an exedra between side walls that obscured the ends, corresponding to Brizio's statement (Reference Brizio1873: 19) that the ends were not properly visible, although the back (which has a scene of wine-making) was accessible.Footnote 17 In the elevation, however, the ends seem to be free of the side walls and the sarcophagus is raised above the ground by three supports. (The sarcophagus today is supported by two nineteenth-century marble lions.)Footnote 18 Brizio states (Reference Brizio1873: 17) that it was placed on the ground, which corresponds better to Patrizi's drawing.Footnote 19 It is difficult, nonetheless, to reconcile all this with Patrizi's view, which shows one end exposed. Possibly, when he saw it, the exedra was being, or had been, demolished to get at the sarcophagus, which might explain his problematic wall treatment, which seems to be based on the wall recesses in the second (main) space (although the excavation drawing does not show round tops here) (Fig. 2). Also, Patrizi squares off the ends of what is a round-ended sarcophagus. Another anomaly in Patrizi's drawing is that although the tabula (which contains the inscription referring to Maconiana Severiana) on the front of the lid is recognizable, and with a little imagination the various squiggles he places on the sides and ends can be made to correspond with the reliefs, he omits the lid itself and shows an empty interior.
This is the most problematic part of his representation, and is very difficult to reconcile with the drawing being an optical view of the whole complex seen from one point sul posto. Possibly, when Patrizi saw it, the sarcophagus had been extracted from its findspot and had been placed in the next room in front of the recesses, but the representation would still have to be an additive composition based on memory or individual studies. Certainly, the placement and lack of perspective of the marble altar nearby suggests a composite representation.
The marble altar
This circular marble altar, now in the Ny Carlsberg Glypthothek, was acquired from the Villa Casali in 1884 (Fig. 9).Footnote 20 Brizio (Reference Brizio1873: 14–15) describes the reliefs in detail, but Patrizi only shows three generic standing figures. The side shown in the excavation elevation corresponds to the first of Santolini Giordiani's illustrations (tav. X,a).

Fig. 9. Unknown, Circular Altar with Dancing Figures, Roman. Ny Carlsberg Glypthothek, I.N. 513. Left: photograph, after Santolini Giordani, Reference Santolini Giordani1989: tav. X,a. Centre: detail of Figure 2. Right: detail of Figure 1.
The feature on top of the altar is confusing to read in Patrizi's drawing as the hatching corresponds to that on the cylindrical altar below, but must here be intended to indicate concavity as the excavation drawing shows a candelabrum with concave sides. It also has corners that are vertical, not inclining inwards as Patrizi shows them. This piece is now lost, and Santolini Giordani's description (Reference Santolini Giordani1989: 121, cat. 66) is based on Brizio. Brizio (Reference Brizio1873: 15–16) refers to single figures on each side: Minerva with helmet, shield in left hand and right raised, who is on the other side from Ceres who holds a sheaf of wheat and wears a simple chiton. On the other two sides are Silvanus in a short tunic and with items in both hands, and Apollo on a base with items in both hands. The excavation elevation shows a figure with its left elbow bent and right arm raised that fits Minerva, yet it appears to be male, with bare legs. Patrizi's representation is too sketchy to convey much information but does seem to indicate a bent left elbow, and an optimistic reading might see it as a draped standing female figure. Brizio (Reference Brizio1873, 15) points out that each of the corners was richly decorated, as Patrizi also indicates. There were three figures on each corner, one above the other. At the top was a griffin with spread wings and raised beak, which must be the complicated scrawls at the top. Below this were herms of a beardless youth with a mantle, sketchily indicated by Patrizi, and at the bottom a sphinx with its two feet on the ground corresponding to the flourishes at the bottom of Patrizi's drawing.
Location
The location of the tomb is not known exactly (Avetta, Reference Avetta1985: 37). As Patrizi's drawing makes clear, the tomb was located at the boundary between two properties, which he carefully identifies: the ‘Proprietà Volpi’ at the left, and the ‘Proprietà Casali’ at the end of the tomb at the top (Fig. 1). The line delineating the boundary makes clear that the boundary ran through the middle of the lion sarcophagus, which is only half excavated. Brizio laments the fact that, at the time of writing, the lion sarcophagus could not be fully excavated, so that ‘one cannot clearly see the whole composition’ (Reference Brizio1873: 21).Footnote 21
The line Patrizi draws demarcating the two properties bends left, which seems to imply that what lies below this line belonged to the Casali. Yet the excavation remains in line with the middle of the sarcophagus, and Brizio observed that the second room, which is mostly out of sight in the foreground, could not be fully explored because one side lay within the boundaries of the Volpi property (Brizio, Reference Brizio1873: 14).Footnote 22 In Lanciani's Forma Roma Urbis the label ‘SEP. SEMPRONIOR’, referring to these excavations, is situated near the bottom left of a section of the Vigna Casali that, in plan, forms a vertical prong (Fig. 10). This prong appears in earlier maps, including Nolli's 1748 map (Fig. 11), an agrimensore plan of the 1660s (Fig. 12) (Santolini Giordani, Reference Santolini Giordani1989: fig. 10 (Archivio Casali)) and a catasto plan of 1873 (Fig. 13) (Santolini Giordani, Reference Santolini Giordani1989: fig. 11 (Archivio Casali)). Lanciani's indications are placed near where the boundary turns west, near where, since at least the 1660s, there were two small buildings (case). Neither Brizio nor the excavation drawing gives any indications of orientation, so it is likely that the location was somewhere near one of the two positions marked in Figure 10.Footnote 23

Fig. 10. Detail of R. Lanciani, Forma Urbis Romae, Rome 1893–1901, facsimile reprint, Rome: Quasar, 1988, plate 46, with additions by the author.

Fig. 11. G.B. Nolli, Pianta di Roma, 1748, detail of the Vigna Casali, with additions by the author.

Fig. 12. Agrimensore plan of the Vigna Casali, 1660s, Archivio Casali, with additions by the author.

Fig. 13. Catasto plan of the Vigna Casali, 1873, Archivio Casali. After Santolini Giordani, Reference Santolini Giordani1989: fig. 11, with additions by the author.
Santolini Giordani (Reference Santolini Giordani1989: 62, citing Archivio Casali, tomo L, 1) notes that in the nineteenth century the Volpi usurped more than a pezza (2640.6 sq. m, or about a quarter of a hectare) of the Casali land. The land in question may be a section to the left of the prong where there is a curved line in the agrimensore plan of the late 1660s (Fig. 10) and in Nolli (Fig. 11). In the late 1660s plan, this area lies clearly within the Casali boundary, which follows a straight line connecting the ends of the curve, like a bowstring, but in the catasto plan of 1873 (Fig. 13) the Casali boundary follows this curve. If there was bad feeling between the families as a consequence of this, it might explain why Marchesa Carlotta Faustina Casali and Marchese Raffaelle Casali del Drago found it difficult to pursue their excavations beyond the boundary, although they must have done so to some degree to extract the lion sarcophagus fully.
Conclusion
Patrizi's drawing provides a useful addition to our understanding of the state of the tomb of the Sempronii at the time of its discovery in 1872–3. Although it is inscribed as having been made sul posto, and there is no reason to doubt that Patrizi observed it for himself from above in the direction shown, it seems to be unlikely that it is a sketch that records dispassionately an optical view from a single point made in a short space of time. Elements, especially at the right, may have been assembled from separate observations or done from memory, and cannot be fully relied upon. As an amateur draftsman Patrizi was not entirely in control of his representational means, and seems to have employed a process of exploratory doodling that only intermittently resolves itself into a considered representation of what was, or had been, before his eyes. Yet it is not often that we find so immediate a record of an excavation made by a disinterested observer.