1The lidded phiale is a very rare find, found from the 4^th century BC to the Roman period. It is a composite vessel, consisting of a phiale and a plate-like lid. Only glass examples survive. There are several Roman finds, but the pre-Roman examples are very rare and most are without or of uncertain provenance.1

2The only early find from an scientific excavation is the lidded phiale from the Macedonian tomb at Makrygialos, in northern Greece.2 The Macedonian tomb had been used for three successive burials, dated to the 4^th and 3^rd centuries BC. The tomb had been recently looted, but small precious finds survived the looting. The lidded phiale was part of the last burial, on a built funerary bed, and was found along with the skeleton and a ceramic unguentarium, dated to the beginning of the 3^rd century BC.3 The glass vessel is probably older and dated to the 4^th century BC. The innumerable fragments of transparent colourless glass with a slight green tinge were reassembled to almost complete three vessels, two of which belong together to comprise the lidded phiale.4

3Both parts of the vessel are also atypical (figs. 1-2). The lower part, the phiale (Py 6435) is a simple bowl, 15.4 cm wide, 4.5 cm high and 3-4 mm thick. It has a convex profile, with an outsplayed rim and the rim lip entirely missing, and a slightly convex bottom without an added base. It is undecorated, but there may have been some decoration on the rim. Despite its simple shape, it has no parallels either in pottery or in metalware. Other phialae of that period are quite different, as they do not present a continuously curving body but a rather articulated profile. The upper part, the lid (Py 6436), is a shallow plate, 18 cm wide, 1.8 cm high and 3-4 mm thick; its rim is 2 cm wide. It has a convex body and a wide horizontal rim that ends in a vertical, but rounded, overhanging lip. Its shape seems to be familiar and similar to the existing ceramic fish-plates, mainly due to the existence of the vertical drop of the lip. Yet ceramic plates differ in profile, as those are conical rather than concave and they also feature an added ring base. Other glass plates from this period do not survive, if they existed at all. The first glass plates appear almost a century later and they are very different, as they result from the evolution of the articulated Achaemenid-type phiale.5

Fig. 1 — Lidded phiale, Pydna.

(Photo: Th. Stoupiadis)

Fig. 2 — Lidded phiale, Pydna. Profile and section of the lid, and section of the set.

Fig. 3 — Lidded phiale, Hermitage Museum.

Fig. 4 — Lidded phiale, Hermitage Museum. Reverse of the lid and drawing of the decoration.

4Several lidded phialae can be compared with the Pydna find. An almost complete example from, probably, the late 4^th or early 3^rd century BC is in the Kunstmuseum, Düsseldorf. It is undecorated but clearly shows the shape of this kind of vessel.6 A mid-Hellenistic example was found in the Tomba degli Ori in Canosa. It is fragmentary and undecorated, and the two parts present a critically close difference in diameter to be able to fit. Its identification as a lidded phiale is somewhat uncertain but rather probable.7

5Hellenistic examples, in the Hermitage and the Louvre, are decorated with painting and gilding. The Hermitage find is from the Black Sea region. Its angular contours and the phiale’s flat rim probably indicate a date well into the 3^rd century BC (fig. 3). It has decoration on the reverse of the lid. Gold lozenges on a red background are on the outer edge of the rim. Then, on a light blue background, a wreath of myrtle (or olive) circles the rim. The wreath is applied in gold leaf and shown tied with red ribbons. Then there is a band of short gold lines on a red background. The phiale preserves remains of a gold leaf decoration, perhaps a flower (fig. 4).8

6The Classical and early Hellenistic lidded phialae (Pydna, Dusseldorf) were probably made by rotary pressing, as is indicated by their angular contours and also by rotary marks evident on the surface of the Pydna find.9 From the mid-Hellenistic period lidded phialae appear to have been made by slumping. The Hermitage find is not only less angular but also has a very irregular rim-lip. In the late Hellenistic period the shape evolves to a non-angular, flowing version.

7Four finds associated with lidded phialae are in the Louvre collection: one set, and additionally one phiale and one lid.10 They all preserve traces of decoration. The set, of unknown provenance, has decoration both on the phiale and the lid (fig. 5). Their flowing contours and the phiale’s slightly curved rim indicates a date not before the 2^nd century BC (fig. 6). The phiale preserves a faint design executed in red and blue paint and gold leaf. The lid is reverse painted. The main frieze on it is composed of gold leaf palmettes (placed on the side) to form a kind of wreath. This is bordered by two sets of red and gold bands decorated with gold squares (fig. 7).11

Fig. 5 — Lidded phiale, Louvre Museum.

(After V. Arveiller-Dulong, M.-D. Nenna [n. 1], no. 197)

Fig. 6 — Lidded phiale, Louvre Museum. Section and drawing.

Fig. 7 — Lidded phiale, Louvre Museum. Reverse of the lid.

Fig. 8 — Lidded phiale, Pydna. The underside of the lid.

Fig. 9 — Lidded phiale, Pydna. The wave pattern as seen from the obverse.

Fig. 10 — Lidded phiale, Pydna. The wave pattern as seen from the reverse.

8The rim of the Pydna lid is also decorated. The decoration is executed on the underside so that it is visible from above, through the transparent glass (fig. 8). The design is that of a wave pattern, running clockwise (figs. 9-10). A purple triangle occupies the central part of each triangular unit of the wave pattern. Most triangles are filled with a smooth layer of pigment, but in a few places a thick layer, like a blob, is encountered (fig. 11). Initial examination of the pigment in a polarizing microscope,12 under incident light, showed it does not present a crystalline structure. It must therefore be an organic pigment, either purple or madder lake. Recently, the pigment was scientifically analysed and the results proved that it is definitely purple.13

Fig. 11 — Lidded phiale, Pydna. Details of the painted decoration.

(Photo: Th. Stoupiadis)

9The pattern is outlined by a white line, 1-2 mm wide, which is neither painted on nor incised on the glass. It is not very clear how this was made, or if it is intentional at all. It consists of a series of uneven pits and it looks like the result of some kind of deterioration on the borders of the wave pattern. It also appears on a sandwich gold glass from Canosa, on areas where the glass foil does not survive. Scholars have described it as a kind of “etching”.14 On the Canosa find, as well as on the Pydna find, the area outlined by this “etching” appears to be less weathered than its adjacent areas. This has led to the idea that the non-weathered surface was perhaps originally concealed under some kind of attached decoration. This could not have been a painted decoration, because it would have survived, even in traces. The only other possibility is the assumed existence of a thick gold foil, continuous over extensive lengths of the wave pattern. In that case the white line could have been created as a preliminary etching to define the area where the foil would be applied. There is also an alternative explanation. The foil alone would have protected the surface, causing only a local chemical or electrochemical reaction just underneath its borderline: what we see as a white line. Additionally, we may think that the surface to be covered with the gold foil was previously treated in some way or simply that glue somehow varnished and protected the surface. It is not clear whether the foil covered the pigment or if the pigment was applied on triangular openings cut in the foil. In any case the red triangles would have been visible from above on a gold background.

10The gold foil was not found in the tomb. It must have come off fairly intact and perhaps was picked up by the looters who entered the tomb in the 1990s. This perhaps explains also the absence of the phiale’s rim. Although even minute glass fragments were gathered from the tomb, we cannot identify among them any pieces from the phiale’s rim. If gold foil was attached to it and the phiale was already shattered, then perhaps the looters picked up the whole foil along with the vessel’s rim.

11Reverse painting on colourless glass was practised on a small scale in Assyria, where colourless glass was invented. In Nimrud, in the 8^th century BC, the decoration of couches and chairs consisted of ivory elements and colourless glass plaques, approximately 3 by 4 cm. Those were decorated with reverse painting in black outline with red and blue details. The theme was either vegetal or winged figures or winged sphinxes. It has been supposed that the painting was covered with gold foil, which created a background to it when seen from the obverse (fig. 12).15

12Four centuries later, in the 4^th century BC, that same principle resurfaced in Macedonia, where the combination of colourless glass and gold became very popular during the reigns of Philip and Alexander. Glass plaques, of similar size to that of the Assyrian finds, were used for the decoration of couches and were also backed with cut-out silver-gilt foil to be viewed from above (fig. 13).16 The plaque and the foil were not line painted, but there is evidence that the background on which the plaques were placed was painted red. This was either the wooden part of the furniture, or a second, quadrangular, silver-gilt foil. The manner is evident on a two-layered foil like that from Korinos in Pieria: two cut-out figures placed on the bigger foil, which has been used as a background and has therefore been painted red (fig. 14).17 The foil is thick enough to survive away from the glass plaques, even when it is very delicately cut; the same would be, therefore, possible for a wave pattern foil, if it existed on the Pydna vessel.

13The gold foil on the Pydna find would have been very similar to that on the Canosa bowls of the 2^nd century BC (fig. 16).18 A contemporary gold glass bowl was also found in Tresilico, Calabria, and its shape resembles that of the Pydna phiale, the only difference being the existence of a curved rim. Additionally, this one is also decorated with a double wave-pattern on the rim. The main decoration on the floor of the bowl is a hunting scene (fig. 15).19 The Tresilico and Canosa bowls are sandwich gold glass, with the gold foil encased between two layers of transparent glass. This technique appears one century later than the Pydna vessel, so it could not apply in the case of our find.

Fig. 12 — Painted glass plaque, Nimrud.

(After H. Tait [n. 15], fig.v41)

Fig. 13 — Glass inlay with gold foil, Vergina.

Fig. 14 — Gold foil from the decoration of a couch, Korinos, Pieria.

Fig. 15 — Gold glass phiale, Tresilico.

Fig. 16 — Gold glass bowl, Canosa.

Fig. 17 — Ivory shield, Vergina.

Fig. 18 — Glass situla, Metropolitan Museum of Art.

14Nevertheless, there is a line of evolution connecting the colourless glass production in 4^th century BC Macedonia to that of the late 3^rd century BC in Magna Graecia. We notice common shapes, common techniques of manufacture and decoration, commercial contacts and national identities but, above all, burial customs indicating the proximity of religious beliefs. This has led me to suggest elsewhere that the glassworking tradition, which flourished in Macedonia, but was discontinued there, was perhaps transplanted to Magna Grecia.20

15The combination of gold and glass in a wave pattern exists on a remarkable find in Macedonia, the Vergina shield. There are two bands of wave pattern. Its outline is cut in strips of ivory and on both sides of those strips the space left is filled with triangular plaques of colourless glass, placed over gold foil (fig. 17).21

16The only colourless glass vessel, nearly contemporary to the Pydna one, that is decorated with a painted wave pattern is an outstanding find fairly recently bought by the Metropolitan Museum of Art. It is a situla, with silver handles and painted decoration. A wave pattern “in purplish red, with a thin line of light (Egyptian) blue below” is reverse painted on the rim (fig. 18).22 The similarity of the fabric, shape, style, and decoration of this situla to Macedonian finds of both glass and metal is striking. It is a true masterpiece.

17So are the Pydna vessels, as the lidded phiale was not the only glass vessel in the tomb. It came as a set with another unique shape: a colourless kylix-kantharos.23 We do not know how and where the shape of the lidded phiale appeared, and we can only speculate about its use. It has been associated with a reference in the sources mentioning an exaleiptron, and, consequently, with a possible survival of the Classical shape into the Hellenistic period.24 But Hellenistic exaleiptra do exist and they are manufactured in the traditional shape.25 We could consider the lidded phiale a pyxis,26 but the typical pyxis has a convex not a concave lid. The only other vessels based on the same principle, of a receptacle with a concave lid, are the Macedonian silver so-called salt cellars.27

18The Pydna lidded phiale is the earliest decorated example of the very few lidded phialae surviving today, and one of the earliest examples of reverse painting. Its combination of shape and decoration is most probably the creation of a master craftsman, who was active in Macedonia in the late 4^th century BC; a production center that influenced the evolution of the craft throughout the Hellenistic period. The wave pattern on the underside of its rim is an expression of the desire for polychromy; desirable even by a society that valued very highly the transparency of colourless glass, and its prototype, rock crystal. Today’s technological advances permit us to identify and investigate exceptional manifestations of polychromy on colourless glass. It is my belief that these finds will not remain isolated and rare, and that many more will come to light.