Investigating Colors and Techniques on the Wall Paintings of the ‘Tomb of the Philosophers’, an Early Hellenistic Macedonian Monumental Cist Tomb in Pella (Macedonia, Greece)

Abstract

The ‘Tomb of the Philosophers’, owing its name to the iconographic theme depicted on the interior of its walls, offers a rare example of high-quality early Hellenistic painting, which enhances our knowledge on the use of painting materials and techniques in Greece during the late 4th–early 3rd century BC. The tomb was excavated in 2001 by the 17th Ephorate of Prehistoric and Classical Antiquities, in the area of the east cemetery of the ancient city of Pella and is among the largest built cist graves to have been found in the region to date. This article presents the results of a scientific investigation of the painting materials and techniques used in the tomb’s decoration by means of high-resolution visible- and raking-light imaging, broadband imaging (IRR, UIL, UVR, VIL), X-ray fluorescence spectroscopy and SEM-EDAX microanalysis on a representative number of samples. The examination of the paintings testifies to the application of elaborate painting techniques and to the use of a varied and rich chromatic ‘palette’ (iron oxides, Egyptian blue, malachite, conichalcite, cinnabar, lead white, carbon-based black, a purple organic colorant, yellow-orange arsenic and vanadium-based pigments). Furthermore, iconographic elements of the paintings were better visualized, allowing for a more accurate description and interpretation of the decorative program of the tomb.

1. Introduction

Twentieth-century archaeological research in Macedonia has given us significant insights into the history of Greek painting [1,2,3,4,5]. From the first discoveries of monumental painted tombs at Mieza [6] to the astonishing royal tombs at Aigai [7,8] and the most recent findings at Agios Athanasios [9] and Pella [10], a few pieces of the jigsaw puzzle of Greek painting have started to fall into place. On the plastered walls of monumental tombs and the polished surfaces of sumptuous marble furniture, colorful compositions still preserve a glimpse of the artistic micro-contexts in which they were conceived and produced. They reveal a painter’s preference for a specific pigment [11], the painter’s personal markers of style and technique, the artistic gestures and pictorial textures that allow us to visualize the disparate information on Greek painting preserved in textual sources [12].

As part of a larger research project on the study and documentation of ancient Greek painting led by Hariclia Brecoulaki and hosted by the Institute of Historical Research/NHRF, a non-invasive analytical campaign was undertaken in 2018 to study the materials and techniques used in the creation of the wall painting scheme. The techniques used in this study are high-resolution visible- and raking-light imaging, broadband imaging in the infrared and ultraviolet ranges, and X-ray fluorescence spectroscopy. Our scientific investigation was complemented with a comprehensive evaluation of the SEM-EDAX microanalysis data on a representative number of samples taken from flaking areas of the wall paintings shortly after the discovery of the tomb [10,13]. A preliminary presentation of these results, written in Greek, was published in 2007 in an appendix of a monograph dedicated to the tomb’s wall paintings [14]. Our objectives were to explore the making of the tomb’s wall paintings in order to enrich our knowledge of Greek painting of the early Hellenistic period and to determine specific features of local Macedonian workshops; to deepen our appreciation of the original iconography of the tomb and to fine-tune earlier interpretations; and, finally, to properly document as much information as possible on what is still preserved on the tomb’s walls for future studies, given the risk of rapid decay of the painted decoration of the monuments that are preserved in situ, mostly due to adverse environmental conditions.

2. Materials and Methods

2.1. The Tomb and Its Painted Decoration

The ‘Tomb of the Philosophers’, owing its name to the iconographic theme depicted on the interior of its walls—a gathering of intellectuals, the earliest of its kind to have been actually preserved on a Greek funerary monument—offers a rare example of high-quality early Hellenistic painting. Moreover, it significantly enhances our knowledge of the use of painting materials and techniques in Greece during the late 4th–early 3rd c. BC [1,10]. The tomb was excavated in 2001 by the 17th Ephorate of Prehistoric and Classical Antiquities, by its then Director Maria Lilimpaki-Akamati and her collaborators [10,13], in the area of the east cemetery of the ancient city of Pella (see Supplementary Materials Figure S1) and is among the largest built cist graves to have been found in the region to date, together with the well-known ‘Tomb of Persephone’ at Aigai [8]. Its external dimensions are 5.50 m long by 3.50 m wide (see Supplementary Materials Figure S2). The burial chamber is 4.40 m long, 1.95 m wide and 3.00 m high. Covered by eight (8) stone slabs (length, 2.30–2.75 m; width, 0.40–0.80 m; height, 0.33–0.55 m), it housed the burials of a man (c. 45–50 years old), a woman (c. 29–34 years old) and a child (up to 6 years old) [15]. The fragments of plaster/wall paintings found collapsed from the east section of the north wall, as well as from the east wall, that are currently kept in the Archaeological Museum of Pella could not be accessed for this study. Underneath the covering slabs, traces of a wooden roofing are preserved (wooden planks resting on wooden beams). A mound might have covered the tomb, but the evidence for its existence is inconclusive [10,14]. The tomb had already been looted in antiquity, at the end of the 3rd century BC [10], and the few finds retrieved from the burial chamber of the tomb are dated by its excavator to the end of the 4th–beginning of the 3rd century BC [10].

The interior of the tomb is decorated with colorful wall paintings, combining architectural elements, ornamental motifs and more copious figurative compositions (pp. 32–76, [10]) (see Supplementary Materials Figures S3–S6). The choice of the main iconographic theme of the tomb is associated with the representation of figures of literati, or philosophers, reflecting both the high cultural level of the Macedonian society during the late Classical and early Hellenistic period, and their specific interest in problems of cosmology and astronomy, as revealed by the depiction of a deep blue sphere on the west wall (pp. 88–91, [10]). Philosophical currents of the 4th century BC and Aristotelian teaching, in particular, enjoyed a considerable reputation in elite Macedonian circles, and interest in astronomy was particularly explicit in Aratus’s Phaenomena, a famous work on constellations and weather signs [16], written during the poet’s stay in Macedonia (c. 276 BC), in the court of Antigonus II Gonatas.

The main components of the tomb’s interior decoration are the following:

• The toichobate. This is the lowest part of the wall, which was painted dark blue (0.30–0.35 m high);
• Off-white, unpainted plaster band. 1.00–1.06 m high;
• Two tainiae. Two narrow bands, one light brown (0.035–0.038 m high) and one off-white (0.036 m high);
• A kosmophoros. This is a frieze (0.135–0.15 m high) that is decorated with elaborate floral patterns on a blue background. The decorative scheme of this frieze includes scrolls, acanthus leaves and lilies, smaller flowers and perhaps anemones;
• Tainia. This narrow off-white band is 0.04–0.043 m high;
• Egg-and-dart painted Ionian molding, approximately 0.018 m high;
• Main figural frieze. This is the highest of the decorated sections, measuring 0.75–0.82 m, where figures of philosophers are represented (two figures on the south wall, two on the north, one on the east and one standing figure on the west wall; see below for more details of the figures);
• Recessed socket. Unpainted, approximately 0.09–0.10 m high; it would have perhaps supported a wooden lining;
• Figural frieze. About 0.32 m in height, this protruding section shows riders on galloping horses, rosettes and, according to the original publication, stelai with palmette crownings. Each stele appears to be on top or behind a mound. The background of the frieze is light blue;
• Egg-and-dart painted Ionian molding. As the previous one, this was approximately 0.018 m high;
• Conclusive elements. A painted tainia in light yellow, 0.30 m high. Painted mutules and guttae are found on the north and south sides, between the beam sockets and along the upper edge of the uncut sections of the blocks.

The main figural scenes as understood after the excavation:

West wall: A young man stands alone against the unpainted plaster. With short brown hair, he wears a bright red himation, leather sandals (krepides) and a wreath with green leaves. With his left leg leaning on a mossy green rock, the man points with a wooden rod at a blue globe in a box. This figure has been identified as an ‘astronomer’ (see Supplementary Materials Figure S3 and Figure 1).

North wall: A bearded man sits on a rock on the west side of the north wall. With short hair, he wears a chiton, a himation and a wreath. He is represented as being absorbed in reading a papyrus roll. Two rolled papyri are represented to the east of the man. Large sections of the wall were found on the ground during the excavation and were partly reassembled in the Museum of Pella. According to the original publication, one more figure of a wreathed man was shown seated in the partially collapsed area, dressed in a chiton, himation and krepides. Now split between the museum and the site, this man probably also sat on a rock and held a staff (see Supplementary Materials Figure S4 and Figure 2).

East wall: As for the north side of the wall, large sections were affected by plaster loss. On collapsed plaster fragments that were transferred to the Museum of Pella, parts of the upper body and head of a man were preserved. The man appeared to have dark hair and beard. Still in situ is preserved the man’s lower body; he was possibly seated on a rock, dressed in a himation and holding a staff (see Supplementary Materials Figure S5).

South wall: Two figures are represented on the south wall. On the west side, a bearded man with short hair and a wreath sits on a rock. He is shown wearing a himation and krepides; he looks to his left and supports his head with his right arm, elbow pressed against his leg. On the east side, another man was seated on a rock, wearing a himation, wreath and krepides, and held a bent arm across the torso (see Supplementary Materials Figure S6 and Figure 3).

State of Preservation of the Main Figural Scenes

Although extensive losses of the pictorial layer and plaster deformation were documented by the time of the tomb’s discovery in 2001 (see Supplementary Materials Figure S7), the general condition of the preserved wall paintings was good. ‘First-aid’ conservation interventions were carried out shortly after the tomb was unearthed, as described by the then director of the Department of Conservation of Ancient and Modern Monuments, N. Minos [17]. Earth accretions and dirt deposits were removed mechanically in the course of an initial cleaning of the painted surfaces, while saline incrustations were dissolved and removed using both mechanical and chemical methods. Consolidation of the detached plaster was achieved with the use of hydraulic lime and inorganic fillers. For the re-adhesion of loose paint layers, acrylic resins were used. Despite the fact that the interior of the tomb has remained inaccessible to the public and was protected by a modern shelter, the in situ preservation during the past two decades has caused further damage to the mural paintings, mainly due to environmental and biological factors, such as moisture, salts and fungi (see Supplementary Materials Figure S8). Such deterioration is particularly visible on the faces of the figures, in which paint layers have lost their original chromatic intensity and cohesion (see Supplementary Materials Figure S9).

2.2. Scientific Methodology

2.2.1. High-Resolution Digital Photography and Broadband Imaging

High-resolution digital images were taken using a Nikon D800 camera body. An X-Nite CC1 filter (325 nm, 645 nm) filter was applied in front of the lens, and Profoto B2 flashes were used for illumination. A modified Nikon D7000 camera that was equipped with filtered Nikon SB-80DX xenon flashtubes was used to capture visible reflected, infrared- (IRR) and ultraviolet-reflected (UVR), ultraviolet-induced luminescence (UIL), visible-induced luminescence (VIL) and raking-light images. (For details on the experimental setup, see Refs. [18,19].)

2.2.2. Handheld X-ray Fluorescence Spectroscopy (XRF)

An XG-Lab Elio spectrometer with a Rh tube and silicon drift detector was used with the following settings: 50 kV, 40 mA and 200 s integration time. The nominal spot size has a c. 1.3 mm diameter. While care was taken to analyze areas where pigment was as isolated as possible, some overlap with adjacent areas may have occurred.

2.2.3. Cross-Sections

The samples analyzed in this study come from selective areas of the painted decoration of the tomb where there was already some damage. The colors chosen were representative of the artist’s ‘palette’ in both the architectural decoration (moldings) and the extensive figural representations on the tomb’s walls. In all cases, sampling required particular care so that each sample contained the entire microstratigraphy of the paint layers from the top surface down to the substrate. A total of twelve submillimetric samples, with occasional traces of their substrate (plaster), were selected to be presented in this study.

All samples were examined under a Zeiss stereoscopic optical microscope (magnification = 10×) in order to observe the individual elements of the microstratigraphy, the top surface colors and the presence of incrustations (salts and soil). The samples were then embedded in Serifix resin and were polished with sandpaper down to 4000 so that smooth cross-sections were prepared. The cross-sections were then cleaned with water (ultrasonic bath) and alcohol and were examined under a Leica DFC 295 petrographic/metallographic optical microscope, equipped with a camera, at different magnifications.

2.2.4. Scanning Electron Microscopy Coupled with Energy-Dispersive X-ray Microanalyzer (SEM-EDAX)

The samples were subsequently coated with a thin layer of carbon in a BALZERS CED 030 carbon evaporator to make the surface conductive for examination in the scanning electron microscope. They were mounted on special stubs and were analyzed in a scanning electron microscope (Philips 550) that was equipped with an energy dispersive X-ray microanalyzer and EDAX software. The analysis with EDS in the SEM was performed first on rectangular areas of several hundred microns in horizontal length and 5–30 microns width as the thickness of the pigment layer under examination permitted in order to obtain an average (bulk) analysis of the whole pigment layer. This was followed by a detailed point analysis with a beam size of 3 μm at high magnifications on most of the individual grains contained in the paint layer in order to identify the particles and binding medium producing the overall color effect.

3. Results and Discussion

The technological examination of the wall paintings of the ‘Tomb of the Philosophers’ brought to light significant information that testifies to the application of elaborate painting techniques and to the use of a varied and rich chromatic ‘palette’. Furthermore, iconographic elements of the paintings were better visualized, allowing for a more accurate description and comprehensive interpretation of the decorative program of the tomb. While different pictorial techniques and a variety in the use of color are observed in the different decorative zones of the tomb, it remains uncertain whether one or more painters were actively involved. Therefore, at this stage, we conventionally refer to one artist only, leaving the issue open for further investigation.

3.1. Preliminary Stages

The walls of the tomb were constructed with a local porous limestone. The plaster in cross-sections (sample Pel-4) consists clearly of lime as a binding medium and crashed marble or calcite particles to give it extra sheen and transparency. Moreover, the coarse mortar (Pel-39) consists of lime as a binding medium and clay. A coarse clay-based mortar was used at the junction of stones and over the masonry, in the areas that required leveling, in order to obtain a regular surface. Of irregular thickness, this mortar was rich in clay minerals, quartz and feldspars. A thin calcium-based plaster layer was subsequently applied over the walls and served as the paint layers’ support [15].

By means of visible- and raking-light photography, it was possible to document the state of preservation of the mortar, deformed and cracked in many areas, likely due to changes in the environment, which is monitored but not controlled (see Supplementary Materials Figure S7). Snapped and incised lines into the fresh plaster were used to set out the horizontal borders of the different pictorial fields and provided guides for the design of the moldings and the friezes (Figure 4).

The horizontal lines, however, are only roughly parallel; they slightly vary in distance from the edges and from each other and are uneven in depth, betraying an inconsistent and hurried craftsmanship. Freehand incisions made with a stylus systematically served for the setting out of compositional details, representing a major stylistic marker of the artist that was involved in the decoration of the tomb (Figure 5). In many instances, these incised lines were not precise, but rather improvised and summary, recalling the economic and effective manner in which the preliminary incised drawing was rendered in the famous painting depicting the abduction of Persephone by Hades in the interior or the ‘Tomb of Persephone’ at Aigai [1,8,20] (see Supplementary Materials Figure S10).

The expressive pictorial process of the artist is evident throughout the painted walls; in particular, the rendering of the equine bodies with multiple quick and curved lines on the upper frieze of the south wall of the tomb exemplifies a dynamic way of painting (Figure 6 and Figure 7).

These rough outline sketches were rarely respected when the paint layers were applied. In many instances, the incised lines remained visible even once the painting was completed. Incisions were also used for a compass-drawn guideline around the rosette circles on the south wall (Figure 8). A change in the position of the motif is observed, reflecting the hasty execution of the preliminary sketch and the absence of any kind of means to transfer a finished design onto the surface of the wall.

A quite complete freehand incised sketch of the scroll frieze was remarkably rendered with the use of a stylus, capturing the confident and dynamic gesture of the artist (Figure 9 and Figure 10; see Supplementary Materials Figures S11 and S12).

The lack of accuracy in the rendering of the pictorial elements is counterbalanced by the rhythmic expression of the wavy and disobedient lines, eloquently visualizing the frieze’s theme of fertility and vitality. Floral elements do not reiterate themselves as in a calculated and rigorous composition; rather, each element is unique in its spontaneous and nervous execution, underscoring the role of this preliminary sketch in the overall aesthetic and stylistic character of the painting. Incisions here are not quite as deep as in the horse-race frieze, and they remain barely visible under the thick paint layers applied over the plaster. It seems likely that these incisions were not created directly on a wet plaster but rather on a pictorial layer composed of carbon black that served as an undercoat to the subsequent paint layers of the scroll frieze (Figure 11 and Figure 12).

Incised lines and preparatory drawing were also employed by the artist to mark some of the principal outlines of the large-scale figures. Some of the incised lines were not followed up with colors, and, in consequence, they remain visible. The incisions for the head and body of the bearded figure on the north wall have a spontaneous quality that suggests the freehand technique used (Figure 4). Incisions were redrawn by way of correction, as if the artist was trying to grasp the desirable form directly on the wall. Traces of a preliminary drawing are visible with infrared photography in the figure’s right hand, the papyrus roll and the drapery folds. The underdrawing in these cases picks out the outlines of the various figural elements in a light greyish color. A similar treatment is observed in the figure of the west wall (Figure 13).

3.2. The Painter’s Colorful ‘Palette’ (Table 1, Table 2, Table 3, Table 4 and Table 5)

The use of color in the different sections of the wall paintings shows variety and likely intention. A more restricted gamut of pigments was employed for the large-scale figural compositions and the horse-race frieze compared to the striking diversity that characterizes the floral frieze, while different uses of similar pigments are also observed. By varying the thickness and texture, transparency and superimposition of the paint layers, different pictorial effects were achieved. Due to the lack of the final color layers in certain areas, it was not always possible to deduce the exact intention of the painter vis-à-vis specific pictorial effects.

The rich ‘palette’ is composed of inorganic and organic materials, already attested in Greek painting and sculptural polychromy: calcium carbonate white; lead white; iron oxides (red, brown and yellow ochres); cinnabar; a purple organic colorant; Egyptian blue; copper-based green minerals, alongside yellow-orange arsenic and vanadium-based pigments; and likely a carbon-based black. These results were confirmed both by X-ray fluorescence in situ measurements (Figure 14, Figure 15, Figure 16 and Figure 17; Table 1, Table 2, Table 3 and Table 4; see Supplementary Materials Figure S13a–q) and by a microanalysis of samples with SEM-EDAX (the cross-sections analyzed with SEM-EDAX are shown in Figure 18, and the results are given in Table 5), as reported here, and are further supported by the results of previous studies on Macedonian funerary paintings [1,2,21,22].

Table 1. XRF results related to Area 1 (south wall; basemaps and XRF spectra; see Supplementary Materials Figure S13a–h). Elements in bold are predominant; otherwise, they are present in minor amounts. Major elements are indicated in bold.

Location	Elements	Tentative Identification	Comments
(a) Plaster	Ca, Fe, Cu, Sr, Pb, Ba?	Calcium carbonate
(b) Hand, flesh tone	Ca, Ti, V, Fe, Cu, Sr, Pb	Vanadinite, lead-based white, Egyptian blue	High concentration of Pb, suggesting the use of a lead-based white. Egyptian blue also shows high IR luminescence.
(c) Green leaf over yellow	Ca, V, Fe, Cu, Sr, Ba, Pb	Malachite, vanadinite, Egyptian blue	V may relate to vanadinite in the yellow underlayer (see location e). Cu may relate to both Egyptian blue (see also VIL images) and, especially, a Cu-containing mineral, such as malachite.
(d) Red lips	Ca, V, Fe, Cu, Sr, Ba, Pb	Red ochre, lead-based white, vanadinite	As the main paint layer appears to be missing from this area, the red pigment may correspond to a colored underdrawing/preparation.
(e) Yellow temple/yellow leaf	Ca, V, Fe, Cu, Sr, Ba, Pb	Vanadinite, lead-based white, Egyptian blue	V, Cu and Pb are elements already found in the skin tones, probably in the form of vanadinite, a Pb-based white and Egyptian blue (see also VIL images). The yellow paint layer may be related to the leaves of the wreath, rather than the skin tones (see location c). The presence of litharge could not be excluded or confirmed.
(f) Dark brown mantle	Ca, Ti, V, Fe, Cu, Sr, Pb	Yellow/brown ochre, lead-based white	This layer contains considerable amounts of Fe, suggesting the use of ochres, possibly applied with a Pb-based white and a darker pigment (perhaps carbon, given the strong absorption in the IR-reflected image).
(g) Yellow mantle	Ca, Ti, V, Fe, Cu, Sr, Pb	Yellow/brown ochre, vanadinite, lead-based white	This layer contains considerable amounts of Fe, suggesting the use of ochres, possibly applied with a Pb-based white and vanadinite. The presence of litharge could not be excluded or confirmed.
(h) Black Hair	Ca, Ti, V, Fe, Cu, Sr, Ba, Pb	Carbon-based black, lead-based white	No distinctive elements aided the identification of the black pigment. The strong absorbance in the IR-reflected image may point to the presence of a carbon-based black.

Table 1. XRF results related to Area 1 (south wall; basemaps and XRF spectra; see Supplementary Materials Figure S13a–h). Elements in bold are predominant; otherwise, they are present in minor amounts. Major elements are indicated in bold.

Location	Elements	Tentative Identification	Comments
(a) Plaster	Ca, Fe, Cu, Sr, Pb, Ba?	Calcium carbonate
(b) Hand, flesh tone	Ca, Ti, V, Fe, Cu, Sr, Pb	Vanadinite, lead-based white, Egyptian blue	High concentration of Pb, suggesting the use of a lead-based white. Egyptian blue also shows high IR luminescence.
(c) Green leaf over yellow	Ca, V, Fe, Cu, Sr, Ba, Pb	Malachite, vanadinite, Egyptian blue	V may relate to vanadinite in the yellow underlayer (see location e). Cu may relate to both Egyptian blue (see also VIL images) and, especially, a Cu-containing mineral, such as malachite.
(d) Red lips	Ca, V, Fe, Cu, Sr, Ba, Pb	Red ochre, lead-based white, vanadinite	As the main paint layer appears to be missing from this area, the red pigment may correspond to a colored underdrawing/preparation.
(e) Yellow temple/yellow leaf	Ca, V, Fe, Cu, Sr, Ba, Pb	Vanadinite, lead-based white, Egyptian blue	V, Cu and Pb are elements already found in the skin tones, probably in the form of vanadinite, a Pb-based white and Egyptian blue (see also VIL images). The yellow paint layer may be related to the leaves of the wreath, rather than the skin tones (see location c). The presence of litharge could not be excluded or confirmed.
(f) Dark brown mantle	Ca, Ti, V, Fe, Cu, Sr, Pb	Yellow/brown ochre, lead-based white	This layer contains considerable amounts of Fe, suggesting the use of ochres, possibly applied with a Pb-based white and a darker pigment (perhaps carbon, given the strong absorption in the IR-reflected image).
(g) Yellow mantle	Ca, Ti, V, Fe, Cu, Sr, Pb	Yellow/brown ochre, vanadinite, lead-based white	This layer contains considerable amounts of Fe, suggesting the use of ochres, possibly applied with a Pb-based white and vanadinite. The presence of litharge could not be excluded or confirmed.
(h) Black Hair	Ca, Ti, V, Fe, Cu, Sr, Ba, Pb	Carbon-based black, lead-based white	No distinctive elements aided the identification of the black pigment. The strong absorbance in the IR-reflected image may point to the presence of a carbon-based black.

Table 2. XRF results related to Area 2 (scroll, north wall; basemaps and XRF spectra; see Supplementary Materials Figure S13j–o). All identifications are tentative and will need to be confirmed with molecular identification techniques in the second phase of this study. Elements in bold are predominant; otherwise, they are present in minor amounts. Major elements are indicated in bold.

Location	Elements	Tentative Identification	Comments
(i) Blue	S, Ca, Ti, Fe, Cu, Sr, Pb	Egyptian blue	The presence of S is unclear. Possible explanations include the use of gypsum or indicative of sulphation, which is a likely explanation.
(j) Purple	S, Ca, Ti, V, Mn, Fe, Cu, Sr, Pb	Unknown colorant likely of organic origin, with Egyptian blue and lead-based white	No traces of Br were detected.
(k) Yellow	S, Ca, Ti, V, Fe, Cu, Sr, Pb	Vanadinite, lead-based white
(l) Orange	S, Ca, Ti, V, Fe, Cu, As, Sr, Pb	Arsenic-based orange, ochres, lead-based white	Options include mimetite, orpiment, realgar, pararealgar.
(m) Red	S, Ca, Fe, Cu, Sr, Zr, Ba, Hg, Pb	Cinnabar, lead-based white
(n) Green	S, Ca, Ti, V, Mn, Fe, Cu, Zn, As, Sr, Zr, Pb	Conichalcite	The presence of Zn may be related to adelite minerals.
(o) Black	S, Ca, V, Mn, Fe, Cu, Sr, Ba, Pb		No distinctive elements aided the identification of the black pigment. Possibly a carbon-based black, based on the strong absorption in the IR-reflected images.

Table 2. XRF results related to Area 2 (scroll, north wall; basemaps and XRF spectra; see Supplementary Materials Figure S13j–o). All identifications are tentative and will need to be confirmed with molecular identification techniques in the second phase of this study. Elements in bold are predominant; otherwise, they are present in minor amounts. Major elements are indicated in bold.

Location	Elements	Tentative Identification	Comments
(i) Blue	S, Ca, Ti, Fe, Cu, Sr, Pb	Egyptian blue	The presence of S is unclear. Possible explanations include the use of gypsum or indicative of sulphation, which is a likely explanation.
(j) Purple	S, Ca, Ti, V, Mn, Fe, Cu, Sr, Pb	Unknown colorant likely of organic origin, with Egyptian blue and lead-based white	No traces of Br were detected.
(k) Yellow	S, Ca, Ti, V, Fe, Cu, Sr, Pb	Vanadinite, lead-based white
(l) Orange	S, Ca, Ti, V, Fe, Cu, As, Sr, Pb	Arsenic-based orange, ochres, lead-based white	Options include mimetite, orpiment, realgar, pararealgar.
(m) Red	S, Ca, Fe, Cu, Sr, Zr, Ba, Hg, Pb	Cinnabar, lead-based white
(n) Green	S, Ca, Ti, V, Mn, Fe, Cu, Zn, As, Sr, Zr, Pb	Conichalcite	The presence of Zn may be related to adelite minerals.
(o) Black	S, Ca, V, Mn, Fe, Cu, Sr, Ba, Pb		No distinctive elements aided the identification of the black pigment. Possibly a carbon-based black, based on the strong absorption in the IR-reflected images.

Table 3. XRF results related to Area 3 (philosopher east wall; basemaps and XRF spectra; see Supplementary Materials Figure S13p). All identifications are tentative and will need to be confirmed with molecular identification techniques in the second phase of this study. Elements in bold are predominant; otherwise, they are present in minor amounts. Major elements are indicated in bold.

Location	Elements	Tentative Identification	Comments
(p) Green	K, Ca, V, Mn, Fe, Cu, Sr, Ba, Pb	Copper-based green, lead-based white	This spectrum is similar to (c)

Table 3. XRF results related to Area 3 (philosopher east wall; basemaps and XRF spectra; see Supplementary Materials Figure S13p). All identifications are tentative and will need to be confirmed with molecular identification techniques in the second phase of this study. Elements in bold are predominant; otherwise, they are present in minor amounts. Major elements are indicated in bold.

Location	Elements	Tentative Identification	Comments
(p) Green	K, Ca, V, Mn, Fe, Cu, Sr, Ba, Pb	Copper-based green, lead-based white	This spectrum is similar to (c)

Table 4. XRF results related to Area 4 (purple stain, south wall; basemaps and XRF spectra; see Supplementary Materials Figure S13q). All identifications are tentative and will need to be confirmed with molecular identification techniques in the second phase of this study. Elements in bold are predominant; otherwise, they are present in minor amounts.

Location	Elements	Tentative Identification	Comments
(q) Purple	S, Ca, Fe, Br, Sr, Ba, Pb	Brominated compound	Likely mainly 6,6′-dibromoindigo

Table 4. XRF results related to Area 4 (purple stain, south wall; basemaps and XRF spectra; see Supplementary Materials Figure S13q). All identifications are tentative and will need to be confirmed with molecular identification techniques in the second phase of this study. Elements in bold are predominant; otherwise, they are present in minor amounts.

Location	Elements	Tentative Identification	Comments
(q) Purple	S, Ca, Fe, Br, Sr, Ba, Pb	Brominated compound	Likely mainly 6,6′-dibromoindigo

Table 5. SEM-EDAX results related to cross-sections of samples (Figure 18) taken from paint layers of the wall paintings. Major elements are indicated in bold.

Samples	Top Surface Color/ Location	Stratigraphy of Paint Layers from Top to Bottom	Elements	Suggested Pigments
P.1 Figure 18a	Light brown/coarse grained plaster layer		Ca, Si, Al	Calcium carbonate, clay
P.2 Figure 18b	Light yellow/band from the upper zone	Semi-transparent yellow layer with orange grains	Si, Ca, Fe, K, Al, Cl, Ti	Yellow ochre
P.3 Figure 18c	Yellow/rolled-up papyrus from the north wall	Yellow layer with black particles	Ca, Fe, Pb, Si	Yellow ochre mixed with lead white and carbon black (?) (inferred from strong absorption in the IR range)
P.4 Figure 18d	Bright red/himation of the male figure, west wall	Red layer	S, Hg, Ca, Si	Cinnabar
P.5 Figure 18e	Bright red/flower from the scroll frieze, north wall	1. Red-orange layer 2. Black layer	1. S, Hg, Ca, Si 2. Si, Ca, Al	1. Cinnabar 2. Carbon black (inferred from strong absorption in the IR range)
P.6 Figure 18f	Bright red/egg-and- dart Ionian molding, south wall, upper part	Red-orange layer	S, Hg, Ca, Si	Cinnabar
P.7 Figure 18g	Red/Egg-and- dart Ionian molding, south wall, lower part	Red layer	Fe, Ca, Si, Al	Iron oxides (hematite)
P.8 Figure 18h	Orange-brown/flower from the scroll frieze, north wall	1. Yellow-orange layer 2. Black layer 3. Blue particles	1. Ca, Fe, Al, Si 2. Si, Ca 3. Cu, Ca, Si	1. Hematite/goethite 2. Carbon black (inferred from strong absorption in the IR range) 3. Egyptian blue
P.9 Figure 18i	Brown/garment of the figure (left), north wall	Yellow layer with orange particles	Ca, Fe, Si, Al	Iron oxides (hematite, goethite)
P.10 Figure 18j	Brown/garment of the figure (right), south wall	Yellow layer with black particles	Ca, Fe, Si, Al, Pb	Iron oxides (goethite) mixed with carbon black (inferred from strong absorption in the IR range)
P.11 Figure 18k	Blue/egg-and-dart Ionian molding, south wall, lower part	Light blue layer	Cu, Ca, Si	Egyptian blue
P.12 Figure 18l	Blue/background of the frieze with the horse race, south wall	Blue, semi-transparent layer	Cu, Ca, Si, Al, Fe	Egyptian blue
P.13 Figure 18m	Blue/Egg-and- dart Ionian molding, south wall, upper part	Blue layer	Si, Ca, Cu	Egyptian blue
P.14 Figure 18n	Green/flower from the scroll frieze, north wall	1. Green layer with yellow particles 2. Traces of a thin black layer	1. Cu, As, Ca, Si, Al, Fe 2. Si, Ca, Cu, C	1. Conichalcite 2. Carbon black (inferred from strong absorption in the IR range)
P.15 Figure 18o	White/flower from the scroll frieze, west wall	Opaque white layer	Pb, Na, Al, Ca	Lead white
P.16 Figure 18p	Blue/background from the scroll frieze, west wall	1. Blue particles in a semi-transparent white matrix 2. Traces of a thin black layer	1. Cu, Ca, Si 2. Si, Ca, Cu, C	1. Egyptian blue 2. Carbon black (inferred from strong absorption in the IR range)
P.17 Figure 18q	Purple/flower from the scroll frieze, west wall	Dark purple-black layer	Ca, Na, Mg, Si, P, Pb	Organic colorant

Table 5. SEM-EDAX results related to cross-sections of samples (Figure 18) taken from paint layers of the wall paintings. Major elements are indicated in bold.

Samples	Top Surface Color/ Location	Stratigraphy of Paint Layers from Top to Bottom	Elements	Suggested Pigments
P.1 Figure 18a	Light brown/coarse grained plaster layer		Ca, Si, Al	Calcium carbonate, clay
P.2 Figure 18b	Light yellow/band from the upper zone	Semi-transparent yellow layer with orange grains	Si, Ca, Fe, K, Al, Cl, Ti	Yellow ochre
P.3 Figure 18c	Yellow/rolled-up papyrus from the north wall	Yellow layer with black particles	Ca, Fe, Pb, Si	Yellow ochre mixed with lead white and carbon black (?) (inferred from strong absorption in the IR range)
P.4 Figure 18d	Bright red/himation of the male figure, west wall	Red layer	S, Hg, Ca, Si	Cinnabar
P.5 Figure 18e	Bright red/flower from the scroll frieze, north wall	1. Red-orange layer 2. Black layer	1. S, Hg, Ca, Si 2. Si, Ca, Al	1. Cinnabar 2. Carbon black (inferred from strong absorption in the IR range)
P.6 Figure 18f	Bright red/egg-and- dart Ionian molding, south wall, upper part	Red-orange layer	S, Hg, Ca, Si	Cinnabar
P.7 Figure 18g	Red/Egg-and- dart Ionian molding, south wall, lower part	Red layer	Fe, Ca, Si, Al	Iron oxides (hematite)
P.8 Figure 18h	Orange-brown/flower from the scroll frieze, north wall	1. Yellow-orange layer 2. Black layer 3. Blue particles	1. Ca, Fe, Al, Si 2. Si, Ca 3. Cu, Ca, Si	1. Hematite/goethite 2. Carbon black (inferred from strong absorption in the IR range) 3. Egyptian blue
P.9 Figure 18i	Brown/garment of the figure (left), north wall	Yellow layer with orange particles	Ca, Fe, Si, Al	Iron oxides (hematite, goethite)
P.10 Figure 18j	Brown/garment of the figure (right), south wall	Yellow layer with black particles	Ca, Fe, Si, Al, Pb	Iron oxides (goethite) mixed with carbon black (inferred from strong absorption in the IR range)
P.11 Figure 18k	Blue/egg-and-dart Ionian molding, south wall, lower part	Light blue layer	Cu, Ca, Si	Egyptian blue
P.12 Figure 18l	Blue/background of the frieze with the horse race, south wall	Blue, semi-transparent layer	Cu, Ca, Si, Al, Fe	Egyptian blue
P.13 Figure 18m	Blue/Egg-and- dart Ionian molding, south wall, upper part	Blue layer	Si, Ca, Cu	Egyptian blue
P.14 Figure 18n	Green/flower from the scroll frieze, north wall	1. Green layer with yellow particles 2. Traces of a thin black layer	1. Cu, As, Ca, Si, Al, Fe 2. Si, Ca, Cu, C	1. Conichalcite 2. Carbon black (inferred from strong absorption in the IR range)
P.15 Figure 18o	White/flower from the scroll frieze, west wall	Opaque white layer	Pb, Na, Al, Ca	Lead white
P.16 Figure 18p	Blue/background from the scroll frieze, west wall	1. Blue particles in a semi-transparent white matrix 2. Traces of a thin black layer	1. Cu, Ca, Si 2. Si, Ca, Cu, C	1. Egyptian blue 2. Carbon black (inferred from strong absorption in the IR range)
P.17 Figure 18q	Purple/flower from the scroll frieze, west wall	Dark purple-black layer	Ca, Na, Mg, Si, P, Pb	Organic colorant

Egyptian blue was securely identified here through the correlated presence of Si, Ca and Cu (Table 5, samples P.11, P.12. P13, and P.16; Figure 18k,l,m,p; see Supplementary Materials Figure S13i) and the strong infrared luminescence (Figure 19). A pigment rightfully deserving the Homeric adjective πολύτροπος, (a Homeric adjective meaning multimodal) it has been identified up to now in most paintings of the Classical, Hellenistic and Roman periods that have retained their blue hues [23,24,25]. It fulfilled a variety of purposes in the decoration of the tomb; in its pure form, Egyptian blue was used to paint the darts of the ionic moldings, the backgrounds of the scroll frieze and the frieze with riders, the blue globe depicted inside a box, all appearing as ‘glowing white’ on the VIL image. In mixtures with other pigments, it was used for several purposes: to produce a variety of subtle tones, including the shading of the skin tones, the shadows of the main figures; to give a cooler hue to the elements of the landscape, both vegetal (moss on stone and plant wreaths) and mineral (cool grey tint of the rocks); and to create the cooler tones of the purplish garments (Figure 20).

Cinnabar and red ochres were employed to produce red and orange hues. Each pigment was chosen for specific elements in the tomb’s decoration. The bright and saturated tint of cinnabar was identified on the red himation of the figure occupying the west wall of the tomb (Figure 13; Table 5, sample P.4; Figure 18d), on the red petals of the scroll frieze (Figure 9) and in the egg motif of the Ionian egg-and-dart molding, crowning the frieze with the riders (Figure 6). The high concentration of Hg and the signal of S in the EDAX spectrum confirm the identification of cinnabar in three samples (Table 5, samples P.4, P.5, and P.6; Figure 18d–f) [26]. The detection of Hg in the XRF spectrum also suggests the use of cinnabar for the red petals of the scroll frieze (see Supplementary Materials Figure S13m), possibly mixed with red ochre due to the presence of Fe, a practice that has already been attested to on other Macedonian wall paintings [1,21]. Interestingly, the SEM-EDXA revealed a high concentration of Fe, indicating the use of a red ochre (hematite based, due to the high concentration of Fe in the paint layer; Table 5, sample P.7; Figure 18g) for the rendering of the egg motif of the Ionian egg-and-dart molding crowning the scroll frieze, instead of cinnabar, which was chosen instead for the upper molding (Table 5, sample P.6; Figure 18f). Was this difference related to a different ‘palette’, reflecting a painter’s personal choice, or was it dictated by the will to economize the more expensive cinnabar, restricting its use only to the molding that would be viewed from below and therefore requiring a brighter pigment to grab the spectator’s eye (Figure 6)?

Based on the presence of Fe, a variety of ochres producing yellow, orange and brownish hues were found to have been used for the rendering of the figures’ garments (Table 1; see Supplementary Materials Figure S13f,g) and facial features (Table 1; see Supplementary Materials Figure S13b,d,e,h), as well as for various objects, such as scrolls and boxes. A carbon-based black [27], tentatively identified by its strong absorption in the infrared-reflected images, was added in the areas where a darkened hue was desired, as, for example, on the brownish layers of the mantles of the two bearded figures of the south and north walls (Table 1; see Supplementary Materials Figure S13f; Table 5, samples P.8, P.9, and P.10; Figure 18h–j). The presence of Fe together with V and Pb in yellow layers examined with XRF may suggest a possible mixture of yellow ochre and vanadinite.

The detection of vanadium (V) as a major element with XRF in multiple yellow areas (Table 1, Table 2 and Table 3; see Supplementary Materials Figure S13b–e,h,k) may suggest the use of orange/yellow vanadinite (Pb₅(VO₄)₃Cl), a mineral occurring in the oxidation zone of lead (Pb)-bearing mineral deposits [28]. However, given its proximity to the tube’s rhodium (Rh) L-lines, the presence of chlorine (Cl) could not be ascertained with enough certainty. Vanadinite has already been identified on a few ancient Greek and Roman artefacts, with its earliest recorded occurrence being found on a pale yellow layer on the Artemis from the Maison des Cinq Statues in Delos [29,30]. More recently the use of vanadinite has been attested in greenish and bluish paint layers, mixed with Egyptian blue on a ‘Campana’ relief in Copenhagen, Ny Carlsberg Glyptotek [31]. The presence of copper (Cu) in paint layers associated with V could be related to the ubiquitous presence of Egyptian blue, but the unlikely presence of Cu vanadates cannot be fully excluded at this stage [28,32,33]. This holds true for the green pigments discussed below. The detection of arsenic (As) in a yellow-orange layer (Table 2; see Supplementary Materials Figure S13l,n) may be related to As-bearing minerals, such as orpiment, realgar or pararealgar. As it is found in the vicinity of a yellowish area in which V is present, it may therefore also correspond to mimetite (Pb₅(AsO₄)₃Cl), a secondary lead arsenate mineral which forms in the oxidation zone of Pb–As ore deposits and can be associated with vanadinite [34,35] However, Cl was not identified. As noted in the literature, mimetite may be present as a pigment or as a phase of alteration related to the decay of realgar and orpiment [36,37]. The presence of litharge could not be excluded or confirmed.

The considerable presence of Pb in many of the layers examined with XRF, in both friezes and large-scale figures, supports the use of Pb-based whites, likely to brighten the color, in mixtures with other pigments, even for most paint layers in which Pb possibly coexists with V (Table 1, Table 2 and Table 3; see Supplementary Materials Figure S13b,d,e,g,h,k,p). Applied pure, Pb-based whites served to create highlights due to their significant tinting strength, as may be observed in the cross-section of the sample P.15 (Figure 18o).

Two Cu-based greens of slightly different hues were identified by means of XRF measurements and SEM-EDAX. The high signal of Cu in two similar XRF spectra (Table 1; see Supplementary Materials Figure S13c; Table 3; see Supplementary Materials Figure S13p) may suggest the use of malachite (Cu₂CO₃(OH)₂) already identified in other Macedonian wall paintings of the same period (1). It was used to depict the mossy rock on the west wall (Figure 16) and the warm green leaves on the wreath of the bearded man on the south wall (Figure 21a). This pigment, applied in thick layers, as is revealed in raking-light photography (Figure 21b) seems to have been laid over a layer composed of vanadinite, as is suggested by the presence of V and Pb on the same layer, visible in the microphotography of the analyzed area. The high concentration of copper, in the presence of limited amounts of Egyptian blue, also supports this tentative conclusion. The VIL image confirms the presence of grains of Egyptian blue on the top layer (Figure 21c).

The detection of copper and arsenic in an apple-green layer on the scroll frieze of the north wall (Table 2; see Supplementary Materials Figure S13n; Figure 15) likely points to the use of conichalcite, a copper arsenate (CaCu(AsO₄)(OH)) that was identified for the first time as a deliberately chosen pigment, and not as a random associated mineral, on the gold and ivory couch of Tomb II at Aegae [21,38,39]. The SEM-EDS analysis confirmed this result (Table 5, sample P.14; Figure 18n). The presence of Zn in the XRF spectrum may be related to the presence of associate adelite minerals, such as austinite (CaZn(AsO₄)(OH)) [40]. The yellow grains that are visible in the cross-section of the green sample point out to the presence of iron oxides, supported by the signal of Fe in both the XRF and EDAX spectra.

The examination of purplish layers on the scroll frieze did not yield significant results regarding the possible identification of the purple pigment, since no Br was detected. The predominant element was Pb, suggesting the use of lead white (Table 3; see Supplementary Materials Figure S13j; Table 5, sample P.17; Figure 18q). The use of lead white as an undercoat to organic colorants such as murex purple and madder lake seems to have been a well-known practice among Greek painters, most probably serving to enhance the colorant’s application and chromatic intensity [41]. On the contrary, a clear sign of Br was detected with XRF on a purple stain on the south wall of the tomb (Table 3; see Supplementary Materials Figure S13q; Figure 17), suggesting the presence of an organic colorant (figure), possibly deriving from murex or, less likely, given the robust Br signal, orcein [42]. It is unclear whether this stain originally belonged to a piece of fabric that was dyed in purple or to a paint layer.

Carbon black was interpreted as indirect evidence because (1) no color bearing minerals or oxides was detected in the black layers; (2) the most representative and not-contaminated black layer is on sample Pel-24 on top of a blue layer, which contains minute traces of Si and Mg and a lot of C together with Ca from the lime and Cu from the underlying blue layer; and (3) the absorbance of IR. All of this evidence excludes any other possibility for a black color and leaves no doubt that it is a carbon black layer.

All the aforementioned pigments have already been identified on other Macedonian paintings of the same period. However, the use of cinnabar, vanadium and copper-based minerals remains rather restricted [1,2]. Purple organic colorants are also less common and still the subject of analytical studies [43,44]. Here, below, we highlight the major elements of color and technique which shaped the paintings’ style on the three main decorative zones of the interior walls of the ‘Tomb of the Philosophers’.

3.3. Color and Painting Technique

3.3.1. The Frieze with the Horse Race

Against a light blue background, twelve mounted riders are depicted on the walls of the tomb. The painter of this frieze painted from back to front, filling in the background with a thin and semi-transparent layer of Egyptian blue in large oblique strokes that are visible in VIL images (Figure 7 and Figure 19). It seems likely that this was never conceived as a uniform background, but rather as a faux marble surface, imitating monumental masonry. On top of this background, broad and fluid brushstrokes of carbon black were applied to depict the horses’ outlines, effectively suggesting the volume and movement of the equine bodies. The composition was completed with very few pigments: carbon black was used for shading, yellow ochre for the middle tones and lead white for the highlights. The wheel-shaped rosettes’ oblique cast shadows are projected diagonally towards the upper right to mimic depth, suggesting a light source coming from the left (Figure 22a). The same shading was applied on the bright red button in the center of the rosettes. The painter adopted a different system of shading for the riders and the horses of the composition, which did not reproduce the trompe-l’oeil effect of the rosettes but relied on an interior shading, aiming at emphasizing the plasticity of the forms. The yellowish luminescence in the ultraviolet-induced luminescence images that is visible in the illuminated sides of the rosettes, the riders and the horses corroborates the use of lead white [45] (Figure 22b), a pigment identified in the highlighted zones of the floral frieze and the main figures of the tomb (see below).

3.3.2. The Scroll Frieze’s (Kosmophoros) Polychromy

Defined by a light brown strip along its upper edge, the scroll frieze carried a blue background, against which developed the multicolored floral decoration. As can be seen in the painting’s texture under raking light, the paint in this composition was applied vigorously in at least two superimposed layers (Figure 10 and Figure 11). Although in terms of style this frieze shows similarities to the frieze with the horse race, mainly due to the rapid and improvised application of paint, regarding their techniques, notable differences can be observed. The background here consisted of two superimposed layers that were applied in different stages. In the first stage, a uniform layer of carbon black was laid upon the surface of the plaster, on top of which was incised the preliminary freehand sketch of the floral motifs (Figure 9, Figure 10, Figure 11 and Figure 12). In the second stage, bright pigments composed of cinnabar, conichalcite, vanadinite (and possibly mimetite), brown ochre and a purple lake were mixed with lead white in order to obtain thick opaque paint layers that the painter employed to shape and model the forms of the floral motifs. A thick layer of Egyptian blue was subsequently applied on the background, following the predefined contours of the motifs, reserved very schematically on the black undercoat before the application of any other color (Figure 23). A rich and vivid polychromy characterized this frieze, based on the superimposition of colored layers and the juxtaposition of saturated hues of red, blue, yellow, purple, green and orange, whose chromatic intensity was underscored by the existence of the black undercoat. The stratigraphy of the paint layer is well illustrated in the photomicrographs of the cross-sections of samples taken from the scroll frieze (Table 5, samples P.5, P.8, P.14, P.16; Figure 18e,h,n,p). Highlights of pure lead white were added over the colored layers at a final stage in order to emphasize the volume of the polychrome petals.

3.3.3. Cast Shadows and Skin Tones on the Figures of the Main Frieze

A more elaborate application of the pictorial layers is observed in the rendering of the male figures of the main frieze, denoting an experienced artist in command of his technique, despite the obvious hastiness of execution which characterizes the entire decoration of the tomb. The figures’ silhouettes are sketched, in essence, in fluid outlines, defining the interior space of the bodies and the garments. Thin semi-transparent layers of greyish tones served to model the attached shades, which give the figures a three-dimensional quality, and the cast shadows connected to the figures. On the south and north walls of the tomb, cast shadows surrounding the proper left sides of the figures show the typical fluorescence of Egyptian blue, attesting to the use of this pigment for the creation of a cooler hue, which makes the figures stand out and enhances the illusion of spatial recession (Figure 19 and Figure 20). On the face of the bearded man on the south wall, skin tones were produced through an interplay of warm and cool grey tones, applied in thin paint layers visible on the face’s proper left side, composed of Egyptian blue, lead white and vanadinite. The role of Egyptian blue in the creation of formal shadows is visible in the VIL image (Figure 21c), in the illuminated areas of the cheek and the ear and in lesser quantities on the front, the lower part of the nose and the temple.

3.4. New Iconographical Evidence Revealed through Imaging

Based on the new evidence revealed through the examination of the tomb’s wall paintings with high-resolution photography and broadband imaging, an enhanced understanding of the tomb’s iconography was possible. The most significant new evidence is summarized here, below.

3.4.1. The Frieze with the Horse Race

The riders are set on a terrain, which is defined by mounds combined with stone monuments (‘stelai’). On the north wall, it was possible to detect another mound-and-stone monument set that was shown between the first pair of rosettes (viewer’s left), to the right of the first rider. On the south wall, the execution of the last (west) rosette and mound-and-stone monument set was better understood: it was confirmed that the rosette was depicted over the mound-and-stone monument combination [10] (p. 71) (Figure 8); incisions on the plaster attest that this stone monument (‘stele’) also carried an additional element on its topside (‘crowning’; see below). New evidence was retrieved regarding the stone monuments, previously thought to be stelai with palmette crownings [10] (p. 65): imaging allowed us to better visualize the form of their ‘crownings’ and to identify them probably with vessels, and more particularly lebetes (a type of cauldron) probably adorned with griffin protomai (head and upper torso) [46,47]. Painted vertical/upright elements, discernible on the south wall, could suggest the presence of two more such monuments represented behind the first and second horses (from the viewer’s left to right).

3.4.2. The Figure of the Man on the West Wall

The suggestion that the globe was depicted with its base, both placed in a box and shown emerging from it, has been reconsidered [10] (p. 49). The fact that the element recognized as the globe’s base has the same color as the box’s main body, in a darker hue to denote a shaded interior side, corroborates the idea that this is, in fact, the box’s lid (Figure 13 and Figure 19).

3.4.3. The Figures of the North Wall

The new high-resolution photography has made it possible to ascertain that the himation of the ‘reader’, on the west part of the wall, was drawn high up so as to cover his nape, but it was not drawn over the back of his head (Figure 20a,b and Figure 24a,b). Of the figure’s wreath, the excavator discerned a double incision for its rim and meagre vestiges of green color from its foliage close to the man’s face [10]. The VIL imaging confirms that the figure bore a wreath, an observation corroborated by the three figures on the west and south walls, where Egyptian blue was likewise used for the rendering of the leaves of their wreaths. On the east side of the north wall, what the excavator had originally identified as a seated figure seems to be a standing (female?) figure dressed in a long garment, considering the layers of Egyptian blue revealed with VIL imaging (see Supplementary Materials Figure S14a,b). Traces of Egyptian blue were also detected farther to the right of the standing figure, allowing us to suspect that more iconographical elements were originally depicted on this side of the wall.

3.4.4. The Figure of the Man on the East Wall

Traces of Egyptian blue revealed by VIL in the area corresponding to a seated figure’s head clearly suggest that the man would have worn a plant wreath (see Supplementary Materials Figure S15a,b).

3.4.5. The Two Figures of Men on the South Wall

Prior to our investigation, it was unclear whether the bearded man on the right was meant to be wearing a chiton (Figure 3). However, a more accurate evaluation of the color allowed us to recognize a dark brown chiton and a more lightly colored himation: dark brown covering the area between the himation’s hem and the figure’s right wrist and palm stops to give place to the white color used to render the figure’s right upper arm, thus possibly corresponding to a chiton’s short sleeve. The impression that the man bore no moustache and had a schematized beard on the chin [10] is due to color loss, and it is therefore reconsidered. He originally wore a typical beard and moustache. The figure on the east part of the south wall seems to be holding a staff, as is discernible in the new photographs, which would serve to characterize him as an accomplished man (Figure 25a–c).

3.4.6. The Rocky Landscape

The men represented in the main figural scene appear to inhabit a unified landscape, primarily defined by rocks. In the case of the ‘reader’ of the north wall, VIL images revealed further elements of scenery (Figure 20a,b): the curving contours surrounding the figure, thus rendering a feature in front of which this figure was meant to be seated, could correspond to the mouth of a grotto. On the south wall, UV-induced luminescence images significantly enhanced the visibility of the rock on which the bearded figure is seated and the plant which grows on the rock’s proper right upper edge (Figure 26a,b).

4. Conclusions

The technological investigation of the painted decoration of the ‘Tomb of the Philosophers’ offers new insights both into the chromatic materiality of the paintings and their aesthetic dimension. There still remain significant questions that need to be answered, particularly regarding the exact mineralogical composition of the Cu and V-based pigments identified with elemental techniques and the nature of the organic binders used to produce the paint layers. Considering the ubiquitous presence of egg and plant gums in a representative number of samples from the corpus of ancient Macedonian painting (pp. 400–404, [1]), it seems likely that the wall paintings of the ‘Tomb of the Philosophers’ were also executed with a tempera technique.

While there are marked differences in terms of technique and color choices between the three major decorative zones of the tomb, the overall effect is determined by the quickness and freedom in execution, reflecting a confident artist, working in an improvised and creative way. A similar treatment has already been observed and thoroughly discussed with regard to the famous composition from the ‘Tomb of Persephone’ at Aigai [8]. In this extraordinary painting, the high quality of execution is reflected in the masterly use of the preliminary incised drawing, the vigorous outlines of the bodies and the audacious treatment of the draperies. Although the Pella paintings do not reach the exceedingly high artistic level of the Aigai masterpiece, they echo, nonetheless, a similar stylistic attitude in terms of execution, relying on the restriction to the necessary for the completion of a painting, with the use of shortcuts and freehand drawing. The archaeological evidence seems to support the literary information on famous Greek painters of the fourth century BC who invented and adopted a quick way of painting without depriving their works of their artistic value, as is suggested by the terms compendiaria and celeritas [20,48].

While it remains uncertain by which criteria color and specific pigments were chosen to be applied on the different friezes of the tomb, it seems plausible that the restrained palette of the upper frieze with riders, in accordance with the trompe-l’oeil rendering of the wheel shaped rosettes and the faux marble effect of the background, may have alluded to a three-dimensional frieze in relief—as was the case of the painted metopes of the ‘Tomb of the Judgement’ in Mieza [49,50]—while the more varied and vibrant polychromy of the lower scroll frieze was intended to produce the impression of real garlands of flowers, like those that adorned the walls of the tombs during the funeral.

Differences in the use of pigments must have been related to their visual impact in relation to the paintings’ iconography: the bright red cinnabar applied to the figure’s garment on the west wall, presumably identified with the deceased, stands out and creates a harmonious connection with the vivid polychromy of the scroll frieze, in contrast to the reduced intensities of brown and purple hues used for the folds of the figures’ drapery on the long walls of the tomb. As far as the composition of pigments is concerned, it would be interesting to further investigate their physicochemical characteristics and their provenance, in particular, for the copper-based greens (malachite and conichalcite) and lead-based yellows (vanadinite, mimetite), in order to estimate their material value as painting materials and the formation of local trends within Macedonian workshops. Mimetite and vanadinite have also been attested as painting materials on Hellenistic gravestones from Alexandria, most probably obtained from local mineral sources [51,52].

The Pella tomb seems to be the earliest preserved pictorial representation of such a gathering of intellectuals [53] and of the celestial globe, commemorating the literatus status of the tomb’s occupant and his more specific interest in astronomical questions. Wooden and painted celestial globes almost certainly existed by the early fourth century BC, if not earlier [54], and Claudius Ptolemy instructs how to make a wooden sphere to resemble the night sky, with stars painted in yellow and red on the dark background [55]. The use of Egyptian blue for the depiction of the celestial globe in the ‘Tomb of the Philosophers’ may have alluded to the color of the precious mineral lapis lazuli, whose ‘starry’ quality due to the golden flakes of pyrite it contained was a metaphor for the night sky and its constellations, as is documented by ancient sources [22,56].

The high social ranking of the deceased and of those who commissioned his grave was further emphasized by the choice of expensive and unusual pigments for its painted decoration, such as cinnabar, conichalcite, vanadinite and murex purple. Translucent and opaque paint layers, bright colors and pale shades, bright and dark backgrounds, decorative motifs in trompe-l’oeil and figural, and realistic representations, all of these images were meant to create a coherent whole, a unique pictorial narrative sealed inside the tomb’s universe, with metaphors and allusions to the erudite life of the deceased and his glorious passage into the afterlife.