Vetusta Monumenta: Ancient Monuments, a Digital Edition

Plate 2.9: Roman Pavements Found at Winterton and Roxby

Plate: Engraved by George Vertue (1684-1756) in 1752 after original drawings by Charles Mitley (1705-1758), the sculptor and woodcarver based in York. Mitley sketched all but one of the pavements in situ when they were first uncovered in 1747 and his finished drawings in full color were exhibited at the Society of Antiquaries on 7 February 1751 by Francis Drake, FSA (1696-1771), who commissioned them. The engraving was commissioned by ballot on 14 February (SAL Minutes VI.75, 78). Vertue presented his first proof on 7 May 1752 (VII.18r.). He then engraved the caption, written by John Ward, and delivered finished prints on 9 November (VII.25v.) The labels I-IV refer to the following mosaics and fragments:
I – Winterton: “Orpheus” Pavement
II – Winterton: “Ceres” Pavement
III – Winterton: Fragment adjoining the Orpheus Pavement
IV – Roxby: Fragment

Objects: A Romano-British mosaic fragment (IV) discovered at Roxby, North Lincolnshire in July 1699, together with three others (I-III) discovered at Winterton, a short distance to the north, in August of 1747. The flower with four heart-shaped petals at the top of Fig. IV later proved to be the central panel of a much larger geometric design. Figs. I-III show pieces of two figured mosaics: a long rectangle with an apparently female bust (Ceres) in the middle (II); an octagonal design inside another rectangle (I) featuring a circular border of animals around a central figure apparently playing the lyre (Orpheus); and a fragment with a roundel and a stag (III) originally joined to the Orpheus design. Both sets of mosaics belonged to substantial Roman villas. Construction at the Winterton site began sometime in the first century CE, but the mosaics at both sites date from the fourth century CE, with a specific date of “about AD 350” assigned to the Winterton mosaics by Neal and Cosh (2002, 199, 201). The size of the Roxby mosaic, more fully exposed in 1799, is given by Neal and Cosh as 5.95 square meters (187). They give the dimensions of the Orpheus (I) and Ceres pavements (II) as 7.47 x 4.57 meters and 9.16 x 2.20 meters, respectively, diverging significantly from the measurements according to the scale on the print, 29 x 20.5 ft (Orpheus) and 39 x 13 ft (Ceres). The Ceres pavement has been reconstructed and is on display at Scunthorpe (University Campus North Lincolnshire).


Left: Tria pavimenta tessellata Romani operis, animantium figuris, aliisque diversi generis ornamentis, variis coloribus eleganter depicta; quae in agro Wintertoniae, Lincolniensis comitatus oppidi, prope fluvios Trentam et Humbram anno MDCCXLVII primum reperiebantur. His quartum est adjunctum, apud Roxby vicinum oppidum jacens, sed nondum penitus retectum.

Right: Tria priora ex his pavimentis contigua fere sibi invicem locabantur, quorum primum triclinio fuisse substratum conjectari licet. In medio enim Orpheus sedet citharam pulsans, quem bruta circumdant animalia cantu delinita [sic]; et in quatuor angulis totidem crateres vinarii conspiciuntur. Secundi autem partem mediam Ceres spicas, ut videtur, dextera tenens occupat. Tertium vero cervum exhibet currentem. Quae omnia copiae et hilaritatis signa non immerito censeantur.

Caption: C. Mitley Eboracensis delineavit, curante Fr. Drake, R.S.S. Publishd according to Act of Parliament, Octo 12, 1752. Scala pedum. Aeri sumptu Societatis Antiquar. Lond. Incidit G. Vertue, ann. 1751.


Left: Three tessellated pavements of Roman manufacture, decorated with figures of animals and other diverse kinds of ornaments, painted elegantly with various colors, which were first discovered in Winterton fields in the county of Lincoln near the rivers Trent and Humber in the year 1747.

To these has been added a fourth [mosaic] found in the neighboring town of Roxby but not yet thoroughly uncovered.

Right: The first three of these were each placed contiguously; it can be inferred that the first of these was the floor for a dining room. For Orpheus sits in the middle [of the first mosaic], strumming his cithara, surrounded by brute animals captivated by his song; and in the four corners, as many wine bowls can be seen. And Ceres is seen holding in her right hand a sheaf of wheat in the middle of the second [mosaic]. And truly, the third one shows a deer running. All of these should be justly be considered symbols of abundance and cheerfulness.

Caption: Drawn by C[harles] Mitley of York, commissioned by Fr[ancis] Drake, R.S.S. Scale of feet. Engraved in copper at the expense of the Society of Antiquaries of London by George Vertue, 1751.

Commentary by Noah Heringman and Yoonjae Shin:

Discovery of the Mosaics

These three fourth-century mosaics in Lincolnshire were among eight Romano-British mosaics engraved for Vetusta Monumenta between 1737 and 1787. An anxious remark by George Stovin (1696-1780), who first reported the 1747 discovery at Winterton (Figs. I-III), helps to explain the high priority placed on mosaics by highlighting their vulnerability: “It is certainly the most curious peece of antiquity that ever was found in this county, and I wish you could see it, for [I] am affraid the country people will deface it” (Stukeley 1883, 2.291). William Stukeley (1687-1755) had visited a similar site at Cotterstock, Northamptonshire in 1736 and assisted with the publication of the corresponding print (Plate 1.48), but apparently declined Stovin’s invitation in this letter. The first pavement was discovered by chance on 20 August 1747 and reported to Stovin, a local landowner and antiquary, who visited the spot two days later and sent invitations to Stukeley and to the York physician and antiquary Francis Drake the next day. Fortunately, Drake accepted and engaged the services of noted sculptor and architect Charles Mitley, who made the drawings on which this large engraving is based. Both the high concentration of mosaics in the series as a whole (Plates 1.48, 1.50-52, 2.9, 2.43, 2.44) and the details of this particular story demonstrate the importance of the visual record for preserving these largely accidental discoveries. What makes this print unique is that it includes one earlier discovery, the Roxby fragment found in 1699 (Fig. IV), thereby offering a perspective on the long and gradual process by which antiquaries came to realize the crucial importance of illustration.

The Winterton mosaics are among several important Roman finds made in Lincolnshire in the 1740s. At the beginning of the decade, George Vertue made a special trip from London to record the Lincoln hypocaust in situ (Plate 1.57), but in most cases—especially where highly vulnerable mosaic pavements were concerned—it was more practical to rely on the work of local artists and antiquaries. In this instance, Drake and Mitley made the much shorter journey from York twice to record the Winterton mosaics. Drake had been elected a Fellow of the Society of Antiquaries of London (SAL) on the strength of his History of York (1735), which included a map that was re-engraved for Vetusta Monumenta (Plate 1.47). He seems to have had the same publication venue in mind as a possibility when he engaged Mitley to accompany him to Winterton. Drake shared Stovin’s anxiety about the “country people,” which is echoed in much of the eighteenth-century literature surrounding Romano-British mosaics; a warning about the “unlearned public’s carelessness” was even engraved as a part of the caption on the Cotterstock plate (Plate 1.48). Although Mitley’s accuracy has been challenged by modern archaeology (Cosh and Neal 2002, 1.197), these images are outstanding for their time, and the discovery generated an especially rich verbal record of the state of Roman archaeology in eighteenth-century Britain. As often, aspects of the verbal and visual record that appear outmoded by archaeological standards are of particular value for understanding the practice of antiquarianism itself.

Stovin recorded the names of the “countryman” who discovered the first pavement while “pitching corn,” Richard Cowper (Stukeley 1883, 2.290), and the gardener who assisted with the excavation, Thomas Perfect (Peacock 1866, 241). According to another first-hand account, there were numerous witnesses to the discovery of the Ceres pavement (II-III in the engraving), including “several labourers,” Mitley, and four named as “gentlemen”: Stovin, Drake, and his two companions from York—a Mr. Flowers and the author of the account, identified only as B. E. in the long narrative he submitted to the Grand Magazine (B. E. 1760, 232). Because Stovin, Drake, and B. E. all recorded their observations, there is an unusually strong complement of primary source material in addition to Mitley’s drawings, which provided the foundation for Vertue’s engravings and for a new set of engravings based on them by William Fowler in 1798 (Cosh and Neal 2002, Fig. 166). The Roxby mosaic also depicted here (IV) had been discovered a half-century earlier and then reburied, like the first pavement at Wellow (Plates 1.50-52). Several of the Winterton accounts refer to the earlier account of Abraham de la Pryme, who discovered the Roxby mosaic, originally published in the Philosophical Transactions of the Royal Society of London (De la Pryme 1700). This journal was also considered by Drake as a possible venue, but the appearance of these mosaics in Vetusta Monumenta is more typical for the mid-eighteenth century, a time when the number of antiquarian papers read at the Royal Society was decreasing and the SAL received its own royal charter (1751) to advance “the study and knowledge of the antiquities and history of this and other countries.”

The extensive primary literature around this find, including several newspaper and magazine reports, highlights profound differences between early modern curiosity and modern science, even as those curious about Roman antiquity began to ask some of the same questions that concern archaeologists today. This print and the surrounding texts not only offer us an irreplaceable record of two mosaics that now survive only as fragments; they are also of value for understanding eighteenth-century convictions about the ancient past in their historical uniqueness. B. E., for example, looks for military imagery in these mosaics because he assumes that Roman legions in Britain traveled with wagons full of tesserae, so they could lay ornate mosaic floors for the tents of their officers during long-term encampments (B. E. 1760, 234). Today, these mosaics are understood as “a display of wealth . . . likely to have graced the main reception areas and triclinia of high-status houses,” the villas built for a civilian colonial elite (Neal and Cosh 2002, 36).

The one surviving room of Roxby villa has been measured at 10.6 square meters, making it “one of the largest rooms in a villa in Britain”—probably an aisled building like the one at Winterton, which has been more fully excavated. Jesnick notes that “certain Romano-British pavements may have been laid in rooms used both for public receptions and for dining,” and lists the Winterton pavements among her specific examples of “room function” that was “fluid” in this manner (Jesnick 1997, 104). John Ward, who authored the caption on this engraving, came closer to the modern understanding when he assigned these mosaics to “the first dining room,” but the military bias that colors B. E.’s interpretation was still very dominant in the mid-eighteenth century. The effort to preserve a record of the mosaics in Vetusta Monumenta is worth recovering partly because it shows how historical understanding of the Roman empire both shaped and was shaped by excavation and discovery, and partly because it shows how radically this understanding has changed over time.

The Discoverers: Stovin, Drake, and Mitley

The story of these North Lincolnshire mosaics sheds light on the evolution of historical and archaeological understanding and also provides a clear view of the informal collaboration and networking that were instrumental to its progress, as Rosemary Sweet has shown (Sweet 2004, 60-69). Mitley had collaborated with Drake two years before on another Romano-British subject, the Hovingham hypocaust; an engraving based on Mitley’s drawings, with Drake’s notes, was published as The Plan of a Roman Hypocaust in 1747. A large mosaic pavement found near the hypocaust features prominently in the composition. This engraving, also by Vertue, was apparently ordered by the SAL (Alexander 2008, 376-77) but not included in Vetusta Monumenta. Drake’s letters testify that he considered approaching Lord Burlington, who had sponsored this earlier project, as a patron for the Winterton engravings (Stukeley 1882, 2.294), and B. E. notes that the Royal and Antiquarian Societies were both considered as well (1760, 232). In the event, the drawings were selected for Vetusta Monumenta at a meeting of the SAL on 7 February 1751, more than three years after the original discovery (SAL Minutes VI.75, 78).

The Ceres pavement (II) was fully uncovered in the presence of Stovin, Drake, Mitley, and the others near the end of August 1747, and the Orpheus pavement (I) was discovered just a few days later by Stovin’s son James. It is clear from at least two primary sources that Drake and Mitley recorded only the first discovery in August and were not present for the initial discovery of the Orpheus pavement (B. E. 1760, 233 and Peacock 1866, 241; but cf. Cosh and Neal 2002, 1.197). This pavement was prudently reburied until Drake and Mitley could make the three-day journey from York to Winterton again in October. In a letter to Stukeley of 2 November, Drake claims the credit for sponsoring the second excavation along with a second set of drawings:

I have been lately tempted, over again, into your county to view another discovery, made near the last, in Winterton Fields. I was at the expense of laying bare a noble pavement, 30 feet long, by 20 broad, of most curious tessellated tracery, in the centre of which is the enclosed figure, and round the sides of the octagon, are the figures of a lyon, a stag couchant, a boar, an elephant, a Pegasus, a dog, a bear, and a fox, whose outlines are extremely well executed for such kind of work. (Stukeley 1883, 2.293)

As often in antiquarian correspondence, a drawing was included, in this case so Stukeley could help Drake identify the figure and the instrument—identified in the caption on the print, and since, as Orpheus playing a kithara.

This letter is the main primary source concerning the Orpheus pavement, but the discovery in August of the Ceres pavement was widely noticed in the press. The fullest account appeared in the London Magazine for September 1747:

About three Feet deep was discovered a most curious tessellated Pavement, wrought in elegant Knots, Circles, &c. with a Busto in the Center, representing a Man, but not in a Military Habit, of the same Mosaick Work as the Pavement. The Whole is about 12 Feet wide and 35 long, and is perfect except in some few Places. (Kimber 1747, 434)

Some of the same details—in some cases verbatim—were printed at the same time in the Newcastle General Magazine and Old England. A brief account by Drake himself reportedly appeared in the York Courant that same autumn. Several other period accounts, anchored in the military history of Roman colonization, also assume this bust to be male, but the caption on the print favors the interpretation of this figure as a female divinity, Ceres. The stability of this consensus reflects the long-term influence of the print as well as more recent archaeological findings. Almost all the period accounts devote some attention to artifacts found on and around the pavement, especially Roman glass, and also mention the fragment of a second mosaic (III), later found to be a portion of the Orpheus pavement. The London Magazine further notes that both the Ceres pavement and the fragment “were carefully plann’d out upon the Spot, by an able Artist, Mr. Charles Mitley of York, in the Presence of several Gentlemen.”

Description of the Mosaics

The discovery of the Winterton pavements revived interest in the pavement discovered nearby at Roxby (IV) forty-eight years earlier. Situated near the River Humber and only about 1.5 km south of the villa at Winterton, Roxby Roman villa was discovered in summer 1699 by a farmer who struck on the mosaic while repairing a fence (De la Pryme 1700). Although it was partially recorded by Abraham de la Pryme in July of that year, the Roxby pavement was still “not thoroughly uncovered” forty-eight years later, according to the caption on the print. It was more fully exposed only in the late 1790s when William Fowler, an antiquary and artist based in Winterton, reopened the investigation of both sites. Neal and Cosh make use of Fowler’s 1799 engraving of the Roxby mosaic as well as archaeological findings in their description:

The central panel is preserved and contains an elaborate four-petalled flower formed from four inward-pointing, heart-shaped petals with red tips, yellow bands and white cusps. Around the outer edges of the petals is a yellow line, which separates pointed excrescences above each petal. The four large squares towards the corners have mats of guilloche outlined blue-grey and with red, yellow and white strands. (2002, 187)

Figure IV in the print shows the central panel at the top with a few of the “poised squares” immediately beneath it and the right-hand edge of one of the large squares below and to the left in the original design. Some fragments of this mosaic, together with two more panels found in 1874, “remain in situ, under the garden of a bungalow in Roxby” (Nicholson 2020).

The discoveries at Winterton in 1747 caused greater excitement both because several pavements were uncovered, and because the two large ones both featured human figures, as opposed to the purely geometrical Roxby pavement.1 However, the central panel of the Orpheus mosaic is surrounded by a very substantial geometric border, and the central bust occupies only a small fraction of the Ceres pavement, which is surrounded by a large geometric design described by Neal and Cosh as follows:

A rectangular mosaic is divided into three rectangular panels, each defined by a thin blue-grey line. The panels at each end contain an overall pattern of intersecting circles, outlined blue-grey, forming concave squares and spindle-shapes. . . . The central rectangular panel is subdivided by a thin red line into three compartments. . . . The central compartment has a large circle of round-tongued double guilloche outlined blue-grey and shaded red, and two rows of white, containing a red linear circle tangent to a pair of interlaced squares, one blue-grey and the other white, occupying the centre. . . . The interlaced squares enclose a circle of simple guilloche containing the fragmentary remains of a female bust. (2002, 199)

Several early accounts see this figure as male, and early commentary on both mosaics is occupied with identifying and describing the figures, while the geometry of the designs plays a secondary role. In the case of “Ceres,” these questions remain unresolved, though the name “Ceres mosaic” has become conventional.

According to Neal and Cosh, “it is notable that very few mosaics in the county [of Lincolnshire] are figured” (2002, 134), suggesting another reason for the particular interest in these two figured pavements (of which there eventually proved to be three overall, as against one geometric one here and two geometric pavements total at Roxby). However, one element interpreted as geometric by Mitley and Vertue, the stylized heart shapes in the four corners (spandrels) surrounding the bust, are actually figurative according to Neal (who helped excavate the site in 1958) and Cosh: “The spandrels . . . [contain] canthari with S-shaped handles and stalks with red buds that droop down on either side of them” (200). These buds are interpreted as “military caps” by B. E. (1760, 233).

Archaeological scholarship, while adopting the conventional name of “Ceres,” has found additional grounds for uncertainty surrounding the figure’s identity. Neal and Cosh describe the head as “pale buff, with dark red tesserae used to accentuate the cheekbones, the cleft of the chin, the upper lip and the brow” and with “white piercing eyes with circular blue pupils” (2002, 199-200). Only a fragment of the key attribute survives, “a cluster of blue-grey tesserae, possibly the tip of an attribute shown on the Vertue and Fowler engravings as being carried across her shoulder and interpreted at the time as a sheaf of corn and therefore identifying the figure as Ceres.” Modern interpretations of this attribute range from blue grapes (Bacchus) to a cornucopia (Fortuna) to a fan (Venus). Describing the same bust as “a female with cornucopia,” Jesnick identifies her as the goddess Ge, linked with similar manifestations of abundance, but notes that “another bust in a corridor mosaic, so-called Fortuna, is more likely to be Bacchus with a thyrsus” (Jesnick 1997, 96). This third large mosaic, discovered just to the east of the Ceres pavement by Fowler in 1796-97, is now in the North Lincolnshire Museum.

The Orpheus pavement, found shortly after the Ceres pavement, but not recorded until Drake’s and Mitley’s second visit in late October, initially posed similar challenges of identification. In his letter to Stukeley of 2 November, cited above, Drake not only describes this mosaic but encloses a drawing: “I shall be extreamly obliged to you for your opinion of the instrument near the figure in the centre, which I have sent you; I cannot take it for an ensign, there being no military dress of the man to correspond with it” (Stukeley 1883, 293). Drake makes an analogy between this instrument and the attribute featured in the Ceres pavement, the presumptive sheaf of grain, noting further that both figures are of indeterminate sex. Echoing Stovin’s fears concerning “the country people,” Drake places a strong emphasis on his employment of Mitley, and this source is not alone in suggesting that several copies of Mitley’s initial sketches were circulating within days of each of the two site visits: “By the assistance of an excellent architect I had with me, I have got exact plans of both these curiosities, in their natural colours, which I must prize, because, by their situation in an open corn field, and subject to every fool’s enquiry, they will soon be utterly destroyed” (293-94). The animals surrounding Orpheus soon made it clear who was depicted here, but the head was in fact destroyed sometime after the print was published, and the 1958 excavation found an eighteenth-century chamber pot in its place.2

Here the figures occupy a substantial portion of the mosaic, marked off by a large square. Once again, Neal and Cosh work from both the Fowler image and the surviving fragments (most of the right-hand rectangle and small fragments of the central panel) to show that bands of guilloche (a pattern of interweaving lines), subtending the angles of the large square, create “a large octagon, which contains a large circle of round-tongued double guilloche” (2002, 201). The circle between this and a smaller central octagon is subdivided into eight trapezoidal compartments containing eight “animals captivated by [Orpheus’s] song,” as the Latin caption has it. Of the eight animals “charmed by Orpheus’s music and moving in an anticlockwise progression around him,” fragments of six survive (202-03).

These fragments prove that Mitley took some liberties with the placement of the animals, and perhaps even with the kinds of animals depicted. The engraving shows a stag (above Orpheus’s right shoulder), a lion, a fox, an elephant, a dog, a winged horse, a boar, and possibly a bear. The reproduction in Neal and Cosh clearly shows a leopard instead of a fox, but curiously, a fox is mentioned specifically in the primary sources (B. E. 1760, 233 and Peacock 1866, 241), and Neal and Cosh note elsewhere that Orpheus appears in many other pavements—including the famous one at Woodchester (Plate 2.44)—“seated with a lyre resting on his left knee, with a fox leaping close to the instrument” (2010, 10). The engraving also departs from the surviving fourth panel, which is “the best preserved and depicts an animal, possibly a tigress with an over-large head with a white eye and a black pupil giving a piercing gaze” (2002, 203), where the engraving shows an elephant. Neal and Cosh suggest that the elephant may originally have occupied the missing seventh panel. The engraving also replaces the boar shown in the surviving final panel with “an animal more like a bear.”

Neal and Cosh note a number of omissions and inaccuracies in Vertue’s and Fowler’s engravings, attributing them reasonably enough to the Mitley drawings (though these themselves are untraced). The engravings displace at least two of the original animals and also omit two altogether, replacing them with other animals. One possible explanation is simply that Mitley was faced with a very large pavement to record during his limited time at the site. Fowler, by contrast, lived locally and had an opportunity to make corrections to Vertue’s depiction of the guilloche pattern. Apart from these four small insertions, however, and the addition of color, Fowler’s print of 1799 is an exact copy of Vertue’s, right down to the caption, though no acknowledgment of the original is given. Neal and Cosh point out that neither Mitley nor Fowler seem to have noticed an inscription in the Orpheus mosaic, minute traces of which remain in the surviving edges of the central panel (2002, 202). The heart shapes substituted for the canthari in the Ceres pavement (II) may have been another shortcut on Mitley’s part.

The mosaic appearing in Vertue’s print as a fragment (III) with a stag and a blank roundel has been referred to as “Panel B” since I. M. Stead discovered that it occupied “an annexe at the west end of the Orpheus mosaic” (Neal and Cosh 2002, 204). Stead also provides clear evidence that Fowler re-exposed this mosaic in 1796-97 by pointing out that in addition to the four guilloche patterns, he also added a tree behind the stag in this figure (Stead 1976, 21). This tree is absent in the Vertue engraving. Neal and Cosh interpret the blank roundel as a medallion formerly containing a bust and cite Stovin’s narrative to support the idea that there might originally have been four of these, perhaps representing the four seasons. They suggest that this panel “should be placed above, and contiguous to, the panel over Orpheus and that the lozenge border at this point should be omitted”—indicating that Vertue (or Mitley) might have placed this border on the right-hand side of Fig. I to create a sense of completion or “restoration.” In 1748, perhaps remembering what this fragment looked like before it could be properly recorded, Stovin noted that “in one part to the west end was a Bust of a Lady (perhaps Diana) with a stagg at her feet curiously don; in the middle and more to the east was the Bust of some Emperor, with a Sceptre or some such thing in his hand” (Peacock 1866, 241).

Origins of the Mosaics

Much more is known today concerning Winterton villa, “one of the few villa sites in Britain that have been totally excavated” (Neal and Cosh 2002, 134). The first “scientific excavations” there were done by Stead in 1958-67, uncovering the footprints of several Roman buildings on three sides of a courtyard (198). Roundhouses on the site date to the early second century and construction on the villa began about 180, with three new buildings, in which Stead found traces of second-century mosaics. The mosaics depicted here, together with the Fortuna pavement and a geometric one discovered later, were created when one of the existing buildings was renovated in the fourth century. Beyond the renovation of this “developed aisled building,” new buildings were added at this time.

The mosaics at Winterton and Roxby were created by itinerant groups of artisans, and may be linked to other mosaics known in the region and the province of Britannia as a whole. As indicated earlier with respect to the fox, the Orpheus myth was a popular motif in mosaics, most famously in the “Great Pavement” at Woodchester (Plate 2.44), which likely predates the one at Winterton and may have influenced it: “such an important pavement at Woodchester, presumably with a very influential owner, was likely to have been emulated” (Neal and Cosh 2010, 15-16). Jesnick surveys the broad geographic range of this motif, also found in Roman mosaics in modern-day Libya and Switzerland, among other locations (1997, 128-47). Orpheus mosaics in Britain, mainly in the west, are thought to originate from the “civitas of the Dobunni and are believed to date from the mid-fourth century” (Neal and Cosh 2010, 9). Neal and Cosh note that the animals at Winterton were “naively depicted” by a mosaicist who may have lacked the understanding of the myth or (more likely) the skill to display the calming effect of Orpheus’s music, more evident in the Woodchester pavement uncovered fully in 1793. They also observe that the Orpheus mosaic found near Winterton at Horkstow is executed “very differently,” rendering problematic the attribution of both these mosaics to the Northern Group of itinerant mosaicists (2002, 23). There is more evidence to attribute the Roxby mosaic to the Midlands Group (24) because the small birds discovered in the otherwise geometric pavement there resemble precisely the birds found in another North Lincolnshire mosaic uncovered in 1976 (188).3

Reception History: Roxby

Today’s students of mosaics in Roman Britain have many resources at their disposal, including Neal and Cosh’s magisterial four-volume survey. This literature has made it possible to augment our account of Vertue’s print with extensive information about these mosaics that was unavailable to Stovin, Drake, Mitley, Vertue, and the other eighteenth-century actors discussed in the first portion of this commentary. The print provides us with a line of connection between the ancient artifact and current archaeological knowledge. This particular print, even more than the other prints of Roman subjects in Vetusta Monumenta, provides access to a substantial cross-section of the reception history of Roman antiquity in Britain. This reception history begins with Abraham de la Pryme, who first recorded the Roxby pavement in 1699, and concludes a century later with Fowler’s amended copy of Vertue’s print. This is still a relatively brief episode in the 1600-year history of these monuments, but it is a critical period for the rediscovery of Roman mosaic pavements in Britain, which accelerated rapidly in the course of the century. By retroactively including the Roxby pavement and anticipating that it too will soon be "thoroughly uncovered," this print vividly captures the growing realization that visual recording was essential for a knowledge of antiquity.

Although a few Romano-British mosaics were known from William Camden’s Britannia (1586), de la Pryme still responded to the newly discovered pavement as a curiosity best suited for display in a private collection. Writing to Thomas Gale in July 1699, de la Pryme states his intention of going to see the recently uncovered pavement in order to “begg or buy it,” “take it up whole,” and put it in a “table frame” in his personal museum (1870, 209). In preparation for the event, he also intends to order a copy of Giovanni Ciampini’s Vetera monimenta (1690-99), which includes numerous mosaics. This letter is undated, but the month may be inferred from the 22 July date of de la Pryme’s second letter to Gale, in which he reports that he became too impatient to wait for the property owner or for the arrival of the Ciampini volumes (which, it turned out, were “not to be had in all London”). This letter contains his enthusiastic first-person narrative of discovery:

I have inclosed herein an exact draught of as much of this Rom[an] pav[ement] as we bared and discover’d, with ye colours of ye little stones as they stand in ye work, which I took upon ye place; and when that I discover and take ye rest, I shall make bold to present ye same unto you, with some of ye very figures, if I might be so happy as to know that this and they would be acceptable unto you. (De la Pryme 1870, 213)

The author was a country curate at the time, eager to please Gale, who was Dean of York, and sent “specimens” of the colored tesserae along with the bearer of his letter (212).

Thomas Gale (1636-1702) was also a prominent Fellow of the Royal Society and communicated de la Pryme’s letter about the mosaic to the Society. A copy of the letter, significantly altered for publication and re-dated 2 August 1700, was published in the Philosophical Transactions (De la Pryme 1700) and de la Pryme was duly elected FRS in 1702. The edited letter combines much of the original of 22 July 1699 with an account added later of the northernmost section of the Roman road from London to the River Humber. At this time, the Philosophical Transactions regularly included letters on archaeological discoveries from Roman Britain, which fit into the journal’s domain of “natural knowledge” as it was conceived at that time. By 1747, Vetusta Monumenta was the more obvious choice for such a publication. The published version of the letter makes no mention of de la Pryme’s original intention to take up some or all parts of the pavement for display, and it even seems to alter the facts by stating that it would be redundant to enclose an image because he does “not find that [the Roxby pavement] is so fine as some in Cambden, or Ciampini, that has lately writ upon this subject” (1700, 567). From the original letter, we know that he did send Gale a drawing (as well as specimens). This revision might be due to a refusal by the journal’s editors to publish an illustration, as high-quality illustrations were not a focus of the Philosophical Transactions at this time. When a similar article did include images (Lyster 1707), they were crude and hastily done.

Reception History: Winterton

By the time the Winterton pavements were discovered, publication of high-quality images was a standard practice and pavements were being deliberately reburied for the purposes of preservation. These profound changes in antiquarian practice are belied by the uniform appearance of Vertue’s print, which brings the Roxby discovery up to date by grouping it with the Winterton pavements. The contrasting narratives of de la Pryme and Stovin, who first drew attention to the Winterton pavements in 1747, show how much change had actually taken place over these forty-eight years. Stovin sought institutional sanction for his discovery by inviting Stukeley and Drake to the scene and helped to make sure that the pavements were properly illustrated and preserved in situ. Stovin may even have supplied Mitley with de la Pryme’s original drawing of the Roxby pavement, as he acquired the papers of de la Pryme sometime after the latter’s death. Stovin wrote his own account of the 1747 discovery on a blank leaf in de la Pryme’s manuscript history of Winterton. There is no record of Mitley’s source for the Roxby pavement, but it seems unlikely to have been re-exposed in 1747. Although the attribution to de la Pryme must remain speculative, the difference between their approaches is clear: Stovin was more interested in preservation than collecting, and through his more extensive network he worked toward publication in a more specialized venue, Vetusta Monumenta, in which the pavement could be presented as a work of ancient art rather than a natural curiosity. In his manuscript note, he also gave an accurate date for the pavements of “at least 1400 year old” (Peacock 1866, 241).

Stovin’s interest in de la Pryme’s research shows that there was continuity amidst these changes, particularly in the passion for local antiquities that both men shared. In 1747, shortly before the discovery at Winterton, Stovin published a letter concerning his archaeological investigation of a local legend together with an original poem of de la Pryme’s about this same legend of a hermit living near Hatfield in South Yorkshire (De la Pryme 1747). Both made repeated excursions to the wetlands of Hatfield Chase near the border of Lincolnshire and Yorkshire, which included the peat bog known as Thorne and Hatfield Moors. De la Pryme made a special study of wood preserved in the peat bog, described as a “subterraneous forest” in his second article in the Philosophical Transactions (De la Pryme 1701). Stovin, following in his footsteps, discovered a perfectly preserved pre-Roman female corpse in this bog (Stukeley, 2.292). These details are relevant because they provide an archaeological context for the discovery of the two North Lincolnshire pavements, both discovered by antiquaries eager to trace the deep histories preserved in local soil. Tellingly, Stukeley places these discoveries side by side in the one letter of his that mentions the pavement, which suggests that he had received a copy of Mitley’s first drawing (of the Ceres pavement). Reporting on his visit to the Spalding Gentlemen’s Society in October 1747, he states: “I entertained ‘em with a drawing of the mosaic pavement lately found in our county, near the Humber. Mr. Stovin, who sent it me, tells me, last summer, he took up the body of a woman in the moors near Crowl, in Yorkshire, with the hair on her head, nails on her fingers” (292). Stukeley, who shared de la Pryme’s interest in the theory of the earth, devoted considerably more space to the peat bog than to the pavement. A week later, Drake wrote to Stukeley to say that Mitley was now recording the Orpheus pavement (293-94), but Stukeley’s replies to Stovin and Drake are not preserved. In his letter to Stukeley of 23 August, Stovin had made a special point of mentioning the Roxby pavement discovered earlier, claiming that it was “yet almost intire” (Stukeley 1883, 2.291).

Production of the Print

The mid-century literature on the Winterton pavements is more extensive, reflecting the broad network that linked Stovin to the SAL, but the Roxby pavement is mentioned in almost every case, and Vertue’s print effectively fused the two reception histories into one. Many other pavements had been discovered in Britain in the intervening years. As opposed to de la Pryme, who looked to Italy for publications, antiquaries at mid-century had numerous publications on Romano-British pavements to consult, including the earlier plates in Vetusta Monumenta. At the SAL, the Winterton and Roxby pavements were selected over several other potential contenders for engraving, as can be seen by comparatively frequent Minute Book references to drawings of Roman pavements being shown around this time, including pavements at Chichester (SAL Minutes VI.35); Rudge, Wilts. (VII.23); Yeovil (72, 76); and Leicester (162). The Mitley drawings are first mentioned in the minutes on 7 February 1751: “Mr. Drake shewed several curious drawings colloured, of Tesselated Pavements found near Winterton in Lincolnshire about three miles from the Sea Shore; also a Fine Large Brass Roman eagle, which had formerly been a Standard” (VI.75). A motion was made at this meeting to “Engrave Rookseby in Lincolnshire, and the Pavements found in Winter Fields,” suggesting that the drawings of both sites must have been presented together, and the decision was finalized week later (78). Drake donated the drawings, but asked for “two or three copies” of the recently issued print of the warrant for the execution of Charles I (Plate 2.6) in exchange (79).

The minutes show that Vertue’s considerable care with this large print (591 x 441 mm) extended to the color scheme that was evidently a highlight of Mitley’s drawings. The final reference to the plate in the minutes, after the proof (7 May 1752) and then the finished prints were delivered (9 November), occurs on 30 November 1752: “Mr Vertue brought the Winterton Pavements Coloured according to Order of the Society from that done by Mr Drake of York” (SAL Minutes VII.29r). This hand-colored version of the print may have been unique, and was almost certainly not known to Fowler, whose “antiquarian interest began in 1796 when he copied George Vertue’s uncoloured engraving based on Mitley’s drawing of the Winterton mosaics, and excavated this site close to his home in order to ascertain the colours” (Neal and Cosh 2010, 415). By the time Fowler published his engravings of Winterton (1798) and Roxby (1799), the production of colored prints in significant numbers had become more technically feasible; color prints were introduced in Vetusta Monumenta itself in 1821 with Charles Stothard’s images of the Bayeux Tapestry (Plates 6.1-16). Neal and Cosh emphasize the limitations posed by Mitley’s drawings, but photographic accuracy was not the goal of antiquarian illustration in 1750, and Vertue (in spite of probable inaccuracies in the drawings) was able to produce a representation of lasting value, without which most of the details of these pavements would have been permanently lost.

This engraving inspired Fowler and has informed all subsequent discussion of these mosaics. The head of Orpheus faces right and that of Ceres faces left, a feature that seems designed to encourage active participation from the viewer, who would turn the print 180 degrees in the course of looking at it. John Ward’s caption, drafted between May and November, offers an engaging narrative in its own right, and although a range of interpretations were available to contemporaries, this caption seems to have established a lasting consensus concerning the identity of the figures as well as the function of the mosaics. When the print was distributed on 9 November 1752, Ward received six copies, Drake received a dozen “for his great pains about it,” members received two each, and the price for additional sales was set at three shillings, about £24 today (SAL Minutes VII.25v.). Coincidentally a new letter from Drake (based in York) was read aloud on 7 May, the day that Vertue presented the first proof of this plate. A significant cache of Roman objects had been discovered during drainage work in Micklegate, and Drake included a transcription and rubbing of what he termed an “altar stone,” which were copied into the minutes (18-19).

Reception History: Roman Civilization

Vertue’s strong emphasis on the geometry of these compositions, while minimizing the fragmentary character of the discovery, reflects ideas about the refinement of Roman civilization and about classical aesthetics that conditioned the reception of classical antiquity at this time. As noted by Neal and Cosh (2002, 204), the lozenge border on the right-hand side of the Orpheus pavement (Fig. I) was likely added to complement the three sides that were visible and is unlikely to have been part of the actual mosaic. Similarly, we know from Stovin’s narrative that “the plough had tore up part of” the east end of the Ceres pavement (Stukeley 1883, 2.291), but Vertue depicts it as complete (Fig. II). This “restoration” of the two large designs, together with the fine-grained depiction of individual tesserae, corresponds closely with the assumptions about Roman civilization expressed by one of the eye-witnesses of the original discovery. B. E., in a redacted version of his original memorandum on the occasion, notes admiringly that the mosaics “were made with precision according to geometrical rules” (1760, 233). He surmises that the three panels are perfect squares, in total thirty-six feet by twelve exactly, and gives a very detailed account of these “quadrangles,” their symmetry, and the exact sizes of the tiles that made up each design. The implications of this emphasis become clear near the end of the piece, when the author declares that the five “Grecian, columnal orders of building” and their “geometric rules” are “fundamental” to all art (234). In this way, the formal beauty and longevity of the pavements signal the high degree of Roman civilization, unequaled in the intervening centuries: “It is evident that the mechanical arts, and particularly architecture, were not only arrived at, but practiced in perfection by the Romans in those early times” (234). The Saxons and the Goths, on this account, were “very ignorant of human arts,” and although Inigo Jones and Christopher Wren have finally emerged in modern times as “our two great masters” of architecture, the technology of Roman glass (also found on the site) still evades “our modern ignorance” (233).

This catastrophist history of civilization, echoing the arguments made by some conjectural historians, also helps to explain the contemporary expectation that the imagery and function of the mosaics would be found to illustrate the military aspect of Roman colonization. According to B. E., the “arts of fortification and architecture” both vanished with the Romans (234), and the expectation of a strong correspondence between military and cultural history may also be seen in many of the questions cited earlier concerning the gender and habit of the figures in the two Winterton mosaics. Even B. E.’s explanation of the military origins of these mosaics begins to seem less implausible in this context: “These pavements are supposed to have been the floors of the chief General’s tent, called Praetorium, and of the pavilions of some other officers of high rank in the Roman army: For the Romans carried with them mechanics and tesserae with their military baggage for that purpose.” Thus, B. E. imagines that the Romans “sailed into the Humber, landed about Wintringham, and marched into the field of Winterton, which they made their first station or place of encampment.” Even Richard Gough, describing the sites in his 1789 edition of Camden, notes that “no station had yet been discovered near this place” (Camden 1789, 2.278). Julius Caesar, one of the most readily available sources on Roman campaigns, cites no instances of mosaics having been laid at a camp, but his description of the “luxury” on display in the tents of Pompey’s officers at Pharsalus (Caesar 1856, BC 3.96) might have inspired B. E.’s idea.


Admittedly, B. E. gives freer rein to speculation than any of the other primary sources, but his account shares much in common with the others, above all his emphasis on the visual record initiated by Mitley and Drake. A corresponding emphasis is placed on Vertue’s print by later sources beginning with Gough (1768, 253) and culminating with Neal and Cosh (2002), though they give the preference to Fowler’s version for understandable reasons. Reconstructing the trajectory that led from these discoveries to the print sheds light on many details of antiquarian practice, including even the local knowledge of the “countryman,” Richard Cowper, who became the first discoverer by noticing the non-native character of the brick and tile fragments that he was turning up in Winterton fields in August of 1747 (B. E. 1760, 232). Fowler, too, was guided by local knowledge in re-examining these finds and improving on the visual record. Like Mitley, he was an architect-builder and was self-taught as an engraver, but Fowler achieved successes to rival those of Samuel Lysons, who became director of the SAL in 1798 on the strength of his work on Roman mosaics. According to one recent scholarly account of Fowler’s life, “it had all begun with the discovery of a Roman mosaic floor locally at Winterton, of which his drawings were so admired that he was persuaded to take them to London” (Franklin 2004, 385).

By the time that Fowler re-exposed the Winterton and Roxby pavements, the archaeology of Roman Britain was much more fully explored. Ironically, Fowler found the pavement that de la Pryme dreamed of “taking up whole” still largely undamaged, whereas the Winterton pavements had suffered considerably—including the displacement of Orpheus’s head by an eighteenth-century chamber pot finally documented by Stead in 1958. The Vetusta Monumenta print constitutes a vital chapter in this long reception history, both by preserving a record of the mosaics in a relatively intact state and by contributing to an increasingly vigorous discussion of Roman Britain that is now of historical interest in its own right.


[1] Neal and Cosh observe that there are two small figures of birds, not noticed by Mitley or Fowler, in this otherwise geometric design (2002, 187).

[2] The chamber pot is now on display at the North Lincolnshire Museum.

[3] For more general background on Roman villas and mosaics, see Scott, this volume.

Works Cited:

B. E. 1760. “To the Author of the Grand Magazine.” The Grand Magazine of Universal Intelligence 3: 232-35.

Caesar, C. Julius. 1856. Commentaries on the Civil War, edited by William Duncan. Perseus Digital Library.

Camden, William. 1789. Britannia: or, a chorographical description of the flourishing kingdoms of England, Scotland, and Ireland, and the islands adjacent; from the earliest antiquity. Edited and expanded by Richard Gough. 3 vols. London: John Nichols.

“Country News.” 1747. Old England 176 (12 September): 917.

Franklin, Colin. 2004. “William Fowler of Winterton.” The Book Collector 53, no. 3: 381-412.

Gough, Richard. 1768. Anecdotes of British Topography. London: T. Payne.

“Historical Register.” 1747. Newcastle General Magazine XX (September): 3.

Jesnick, Illona Julia. 1997. The Image of Orpheus in Roman Mosaic. Oxford: Archaeopress.

Kimber, Isaac. 1747. “The Monthly Chronologer.” The London Magazine 16 (September): 434.

Lyster, John. 1707. “A Description of a Roman Sudatory, or Hypocaustum, Found at Wroxeter in Shropshire, Anno 1701.” Philosophical Transactions of the Royal Society of London 25: 2226-27.

Neal, David S. and Stephen R. Cosh. 2002. Roman Mosaics of Britain, Volume 1: Northern Britain. London: Illuminata Publishers for the Society of Antiquaries.

------. 2010. Roman Mosaics of Britain, Volume 4: Western Britain. London: Illuminata Publishers for the Society of Antiquaries.

Nicholson, Rose. 2020. Personal Communication.

Peacock, Edward. 1866. “History of Winterton, in the County of Lincoln, by Abraham de la Pryme.” Archaeologia 40: 225-241.

Pryme, Abraham de la. 1700. “A Letter . . . Concerning Some Roman Antiquities in Lincolnshire.” Philosophical Transactions of the Royal Society of London 22: 561-67.

------. 1701. “Part of a Letter . . . Concerning Trees Found Under Ground in Hatfield Chace.” Philosophical Transactions of the Royal Society of London 22: 980-92.

------. 1747. “A Poem on the Said Hermit.” Gentleman’s Magazine 17 (January): 23-24.

------. 1870. The Diary of Abraham de la Pryme, The Yorkshire Antiquary. London: Andrews & Co.

Society of Antiquaries of London. 1718-. Minutes of the Society's Proceedings.

Stead, I. M. 1976. “The Roman Villa at Winterton.” Department of the Environment Archaeological Reports: Excavations at Winterton Roman Villa 9: 20-79.

Stukeley, William. 1883. The Family Memoirs of the Rev. William Stukeley, M.D., and the Antiquarian and Other Correspondence of William Stukeley, Roger & Samuel Gale, Etc. Vol. 2. London: Andrews & Co.

This page references: