Salamis Tomb 79: A Royal Tomb with Chariots in the Dromos

The kings of Salamis, on the east coast of Cyprus, were buried on the plain west of their city, in the lands of the village of Enkomi. The area is called the Royal Necropolis. About 150 tombs have been found there, among them nine large built tombs with wide approach passages, and in six of those (Tombs 1, 2, 3, 47, 50 and 79) the passage held sacrificed horses and the vehicles they had drawn (Karageorghis 1969). Tomb 79 is the richest of them: the final catalogue, published by Vassos Karageorghis in 1973, runs to 1,107 entries (Barnett 1976, p. 2).

The room for the dead is small. The chamber measures 3.20 by 2.40 m inside, under a gabled ceiling 1.80 m high at the ridge, and it was built, like that of Tomb 50, “of two very large blocks of stone” (Karageorghis 1969, p. 76). Its roof stone was so close to the surface that it could be seen in the 19th century. In front of the chamber is a porch, the propylaeum: a recess shaped like the Greek letter Π, 7.00 m wide and 3.20 m deep, its walls 2.40 m high in well-dressed hard limestone, its floor paved with limestone slabs. The side walls of the porch turn at right angles to meet the walls of the passage (Karageorghis 1969, p. 77).

The passage, the dromos, is by far the largest part of the tomb. It runs 16.80 m east from the façade, 12.80 m wide where it meets the porch and 9.00 m at its outer end (its north wall runs straight and the south wall closes in), with a floor of lime concrete that slopes down towards the chamber. Its walls are dressed blocks laid dry, without mortar, three courses high at the façade, falling to one course and then to none as the floor rises to the old ground surface (Karageorghis 1969, p. 77; Barnett 1976, p. 1). Porch and passage together cover about 200 m² (Salamis of Cyprus 2019, p. 132). The chamber covers under 8 m². The space made for the funeral was some twenty-five times the size of the space made for the dead.

Four steps led to the stomion, the chamber doorway (Barnett 1976, p. 1). The sources do not say which way they ran. The published plan has a line of steps along the outer edge of the porch, so we draw them going down 60 cm from the passage floor to the paving; the other reading would be four steps up from the paving to the threshold. Over the chamber stood a small mound of earth. David Ussishkin proposed a monumental building above the tomb; Karageorghis rejected it (Barnett 1976, p. 1), and we draw the low mound.

No record of the building survives, and no tool marks are reported, so the sequence here is ours, by analogy with the other built tombs at Salamis. A pit was dug for chamber, porch and passage together. The chamber’s “two very large blocks” we read as a U-shaped block forming the walls and a single gabled block for the roof; they could also be two halves split along the axis. The blocks were lowered in first. The porch and passage walls were then laid dry against the sides of the pit, the porch paved, the passage floored with lime concrete, the steps set, and the pit filled back round the chamber and heaped into the mound.

Our estimate of the digging is about 345 m³ of spoil: some 212 m³ from the wedge of the passage, 75 m³ from the porch and 60 m³ from the chamber pit. With iron picks and baskets, at two or three cubic metres a person a day, that is 116–173 person-days of work. The dressed stone for the walls and façade comes to some 53 m³, and the roof block alone to about 17 m³, roughly 40 tonnes of limestone. Moving that block would have been the largest single task: it needed rollers, levers and an earth ramp down into the pit. Quarrying, dressing, hauling and setting about 80 m³ of stone would have taken several hundred person-days more.

The width of the passage was set by what happened in it. It had to take a vehicle and its team, turn them, and leave room for the animals to be killed in their yokes (Karageorghis 1982, p. 135). And it was dug more than once. At the second funeral it was dug out again through the filling of the first, narrower towards the entrance, which by our estimate meant moving another 250 m³ of earth, and putting it back.

Whatever was put in the chamber is gone. It was robbed in antiquity, reused for Roman coffins in the 2nd century AD, and robbed again, and nothing of its first contents survives (Karageorghis 1969, pp. 76–77). Of the person buried at the second funeral Karageorghis wrote that there is “no evidence concerning the nature of the second burial” (1969, p. 98). The chamber is where that person was most probably laid, so the model shows a skeleton on its floor, drawn faint in the plates and marked as our speculation in the 3D; a switch empties the room.

The doorway and its blocking are also ours. Neither is described in what has been published of the tomb, so we draw a doorway 1.00 m wide and 1.40 m high through the 70 cm façade, closed by a single limestone slab, on the pattern of other Salamis tombs. The Roman niches are excavated: one each in the north and south walls, and two in the west wall that cut right through it into the earth behind (Karageorghis 1969, pp. 76–77).

One object raises the question of what else the chamber held. The ivory bed found dismantled in the porch measures 1.89 by 1.11 m, and it would fit into the chamber, just. Whether it stood there with the body or served outside in the passage is open (Karageorghis and Rehren 2017, p. 2); the model sets it in the porch by default.

Two vehicles came down the passage at the first funeral. Chariot B was a war chariot for a team of four. Its two poles were 2.90 m long, and so was its axle; its two wheels had ten spokes each, measured 90 cm across, and had no tyres, only ten pairs of iron nails round each rim. The box, 68 cm wide, 85 cm deep and 44 cm high, was of plaited osier with a partition inside and a bronze hand-loop on top. Bronze caps closed the pole ends, two oval bronze discs hung behind the poles showing a winged lion over a fallen enemy, and the axle ends were capped with sphinx heads whose eyes were inlaid in white. Each axle cap was held by an iron linch-pin 56 cm long, topped by a hollow bronze figure of an armed warrior, eyes and cuirass inlaid in blue, with a rattle inside it (Karageorghis 1969, pp. 78–80; Pilides and Papadimitriou 2012, cat. 238).

The second vehicle, hearse Γ, had a flat body nailed with bronze, with five hollow bronze lion heads at its four corners and the middle of its front, socketed to take the posts of a canopy, and 60 bronze rivets along its sides (Karageorghis 1969, pp. 80–81). Its wheels and pole were not recovered. Reconstructed, its body measures only 0.71 by 0.51 m, too small to carry an adult laid out; it could have carried a child, remains already burnt, or the gifts and food for the feast (Salamis of Cyprus 2019, p. 132).

Six animals drew the two vehicles, and all six were killed in the passage. Karageorghis wrote “quadrupeds (horses?)”, because their skeletons had been broken up (1969, p. 78), but their harness was made for horses: four bronze breastplates, 50 cm wide, embossed with 23 figures (griffins, sphinxes, scorpion-men, winged figures, a sacred tree and a winged disc); six crested head-bands, four of them embossed with lions, cobras and naked figures; two side-pendants 58 cm long whose discs show a winged naked goddess holding lions; and blinkers, some plain, some embossed with a lion attacking a bull or a sphinx over a fallen figure (Karageorghis 1969, pp. 87–88). The harness confirms the count: six head-bands for six heads, and four breastplates for the four horses of the chariot.

The arms in the passage, probably from the first funeral, were few: a large bronze spearhead and the hemispherical boss of a silver shield, its outer face gilded (Karageorghis 1969, p. 89). No sword was found. Where the two vehicles stood that day is our reconstruction, with both facing the chamber, chariot B in the north half and the hearse in the south; only the places they were moved to later are recorded.

By the north wall of the porch stood the largest bronze object in the tomb. The griffin cauldron is a bowl of beaten bronze 94 cm across at its widest, with eight cast griffin heads, 23 cm high, and four double-faced bird-men, 36 cm high, fixed to it; it sits on a tripod of iron rods ending in bronze hoofs, and the whole stands 1.25 m tall (Karageorghis 1969, pp. 89–90). Some modern accounts give it ten heads. It has twelve attachments: eight griffins and four bird-men. A second cauldron lay beside it, about 75 cm high when whole, on a conical foot, with three bull heads facing inwards under each of its two handles and a head of the Egyptian goddess Hathor below a winged disc on the handle plates (Karageorghis 1969, pp. 90–91).

Next to them lay a pair of iron firedogs, 1.10 m long, their ends shaped as the prow and stern of a ship, and a bundle of twelve iron spits, 1.50 m long, bound by two rings with a loop for carrying (Karageorghis 1969, p. 91). Firedogs and spits like these come from warrior graves at Palaepaphos and at Argos in Greece (Karageorghis 1969, pp. 91–92). They are roasting gear, and the food was found: dishes holding chicken bones, eggshells and fish bones, and large numbers of murex and other sea shells (Karageorghis 1969, p. 97).

The pottery is all Cypriot, White Painted and Bichrome, and all of one date, at the turn from the Cypro-Geometric to the Cypro-Archaic period (Karageorghis 1969, p. 97). Large amphorae 80 to 90 cm high, painted or plain, stood against both side walls of the passage. There were medium jars like those from Tomb 3, bowls and dishes of every size, and the sixty or so jugs of plain clay coated with tin so that they shone like silver, in two shapes, one with a sack-shaped body and mushroom rim, the other with a tapering neck and trefoil mouth. Several dozen mortaria, thick-walled bowls with a ring base used for grinding, were also found in the passage, the earliest of the type known in Cyprus (Karageorghis 1973, pp. 13, 116, 121, cited in Villing and Schlotzhauer 2006, p. 45).

The full catalogue counts these vessels. The amphorae, jars, bowls and dishes, mortaria and shells in the model are our estimates, set out in the table further on.

Tomb 79 held the only Iron Age thrones and bed to survive from Cyprus. The ivory throne, chair Γ, now in the Cyprus Museum, was a wooden chair sheathed in ivory plaques pegged to the frame: eleven bands of guilloche, a running pattern of twisted strands, on its curved back, two friezes of palmettes, and a plaque of gold sheet worked as scales across the top. It stands 90 cm high, with a seat 50 by 59 cm, which was of wood or leather straps under a cushion; traces of the cushion were found on the passage floor (Karageorghis 1969, pp. 92–93). It is the one throne found where it had been placed, against the west wall at the bottom of the passage, north of the porch (Feldman 2014, p. 194).

Two openwork ivory plaques were found near it, one with a winged sphinx in the double crown of Egypt (16.4 cm high), the other with a composite lotus (16 cm), both gilded and inlaid in cells with blue and brown. Karageorghis set them, tentatively, under the arms of the throne. Marian Feldman points out that they were found apart, the sphinx about 70 cm north of the throne and the lotus on top of the heap in the porch corner, and that they are finer work than the throne’s own plaques, which may make them later additions (Feldman 2014, pp. 194–195). The model shows them on the floor, where they were found; a switch puts them under the arms. The blue is glass: analysis identifies it as cobalt-blue Egyptian glass made with mineral natron (Karageorghis and Rehren 2017, pp. 1–3).

A second throne was sheathed in silver, with six silver rings on the seat and gilded silver studs along the front stretcher. It was 75 cm high as found; the museum reconstruction stands 105 cm. A silver-sheathed stool with hoof feet stood next to it. A third throne, Δ, faced with ivory plaques and rows of small ivory discs, had been moved to the south side of the passage, shedding its discs (Karageorghis 1969, p. 93). Of the other three stools, one was almost complete and two survived only as impressions in the soil (Barnett 1976, p. 1).

The bed, 1.89 m long and 1.11 m wide, had a wooden frame banded in dowelled ivory and stood 89 cm high to the top of its headboard, which had three carved bands: six plaques of the Egyptian god Heh, four of palmettes inlaid with blue glass, six of confronted sphinxes. Seventeen gilded ivory flowers on gilded iron nails decorated it (Karageorghis 1969, pp. 94–95). There was also a small table, of which one solid ivory leg survives, 36.5 cm high, shaped as an S with a sphinx and a feline paw, nearly identical to a leg from Nimrud; at least two other legs were destroyed by the Roman workmen who dug in front of the chamber. Three ivory objects shaped as rows of drooping petals, the largest 31 cm high, may be incense burners; an iron knife had an ivory handle ending in a feline paw (Karageorghis 1969, pp. 95–96).

The only human remains from either funeral are a few bones. In 1969 Karageorghis reported fragments of a human skull and other bones inside one of the large amphorae standing against the passage walls (pp. 97–98). Neither the amphora nor its position is specified, so the model puts them in an amphora against the north wall, among the first funeral’s pottery. The final report says more. In Richard Barnett’s summary of it, the passage held traces of a pyre with burnt human bones, “the ashes having been probably placed in the chamber” (Barnett 1976, p. 1).

The two accounts may describe one rite: a body burnt in the passage and the remains gathered, partly into an amphora. The small hearse fits a cremation, since a platform 0.71 by 0.51 m could carry burnt bones and could not carry an adult laid out (Salamis of Cyprus 2019, p. 132). The model follows the excavator’s first account by default, with the bones in the amphora, and a switch lights a pyre on the passage floor instead. No age or sex for this person has been published.

The rite reminded Karageorghis of Homer, and the subtitle of his 1969 book, Homeric, Hellenistic and Roman, says so. At the funeral of Patroclus in the Iliad, Achilles leans jars of honey and oil against the bier and throws four horses onto the pyre (Iliad 23.170–172); the bones are gathered into a golden bowl (23.243), and Hector’s, in the last book, into a golden box (24.795). The horses, the jars and the gathered bones all have counterparts in Tomb 79.

The second funeral came soon after the first, soon enough for chariot B to be wheeled aside in one piece, its wood still sound (Karageorghis 1969, pp. 78, 97). The diggers did not empty the old passage. They cut a narrower one down through its filling to the doorway (Barnett 1976, p. 1). Its width is not recorded. We draw it 4.5 m wide at the floor, room for chariot Δ’s 2.10 m axle and a pair of horses abreast, and open it from wall to wall for the 10 m nearest the porch, where the vehicles were pushed against the south wall and the gear heaped in the north-west corner.

What they found at the bottom they rearranged. The skeletons of the six animals were broken and their bones thrown back into the new filling. Their bronze harness was stripped off and thrown on a heap in the north-west corner of the porch, together with one of the two warrior linch-pins, the pieces of the dismantled bed, ivories and other bronzes; the other linch-pin stayed on the floor by the chariot (Karageorghis 1969, pp. 78, 87; Karageorghis and Rehren 2017, p. 2). Chariot B was pushed to the south wall, its poles against the stone, and the hearse, by then without wheels or pole, was set behind it to the east. Throne Δ was carried to the south side. The tin-coated jugs were collected and put into the griffin cauldron (Karageorghis 1969, p. 97).

The clearance is why the find plan looks as it does: a dense band of objects along the north side of the passage and the vehicles parked to the south. It also means that most of the first funeral was found where the second funeral’s workers put it, which the 3D model shows in its “Through time” mode, step by step.

The excavators could tell the two funerals apart on the floor itself. The bowls and dishes of the first lay in tiny fragments scattered over the concrete, sealed by a layer of soil; those of the second stood on top of that layer (Karageorghis 1969, p. 97). How the first set came to be broken is not recorded. They may have been smashed at the first funeral, or trodden underfoot during the clearance; we draw them as a carpet of sherds in eight patches. The soil layer between the two we draw 12 cm thick, our estimate.

The final report also records four piles of pottery holding the remains of food (Barnett 1976, p. 2). Neither their positions nor the funeral they belong to is given. We assign them to the second funeral, whose bowls stood above the soil layer, and spread them along the axis of the narrower passage, where its mourners could reach them.

Chariot Δ was a two-horse chariot. Its box, 90 cm wide and 72 cm long, had two compartments and a rear plank held by removable iron pins; its axle was 2.10 m long, its ten-spoke wheels were crushed in the ground, and its 2.80 m pole curved up in front of the box to a yoke 1.64 m wide. Four bronze standards shaped as flowers, 45 to 50 cm high, stood on the yoke. Stones had been set round the wheels (Karageorghis 1969, pp. 81–85; Pilides and Papadimitriou 2012, cat. 237).

Its two horses were found where they were killed, still yoked. They wore bronze blinkers, crested head-bands and breastplates embossed with a crescent, and each had one side-pendant, on its outer flank, with a large beetle, the Egyptian scarab, on the disc. Their bits were iron (Karageorghis 1969, p. 85). Hearse A, of the same type as the hearse from Tomb 2, lay near the surface at the outer end of the passage with its own two horses, poorly preserved and damaged by later looters; their harness was not all in place (Karageorghis 1969, pp. 81, 85–86).

The passage was then filled again, with soil that held the scattered bones of the first funeral’s teams. Of the dead person at this funeral, as of their rite, nothing is known (Karageorghis 1969, p. 98).

The final catalogue has 1,107 numbers (Barnett 1976, p. 2), and they are not a count of objects: a single number can cover a whole vehicle and its fittings, with sub-numbers running past sixty, while a pot has a number of its own. The catalogue’s breakdown of the pottery, which is most of the deposit, has not yet been entered in this model, so the metal, ivory, furniture, vehicles and animals below follow the published text, and the pottery and shells are our estimates, set to fill a deposit known to be very large. The 3D model shows 806 objects in 143 groups, of which 431 are estimated counts.

The estimates follow the find plan and the text. The plan of the first funeral shows its pottery as one dense band along the north side of the passage, from the façade about 10 m out, and none on the south side, so the first funeral’s amphorae stand in two rows against the north wall there, with its jars, mortaria and the zones of trodden bowls and dishes beside them. The second funeral is not on the published plan: its amphorae stand against the walls wherever the first deposit and the parked vehicles leave room, and further out along the sides of the narrower second passage. They are drawn to show how full the passage was. When the catalogue’s figures are entered, these numbers will change.

Robbers reached the chamber in antiquity, and took what it held. They did not touch the passage, which is why its contents survive (Barnett 1976, p. 1). We place the robbery before the Roman reuse, though its date is not known.

In the 2nd century AD the chamber was used again. The Roman diggers built a rough stone staircase down to the doorway through the filling of the porch, which destroyed at least two legs of the ivory table, and they cut niches in the north and south walls and two more through the west wall to take clay coffins, with lamps and pottery beside them (Karageorghis 1969, pp. 76–77, 96). We give the chamber four Roman burials, one per niche; their number is our reading.

The roof block, showing above the ground, was seen by Max Ohnefalsch-Richter and published in 1885 by Salomon Reinach, who judged that the chamber had been entered through the niches in its west wall (Karageorghis 1969, p. 101). Karageorghis thought that 19th-century looters “must have found only the Roman burials” (1969, p. 76). Pits dug by looters near the outer end of the passage, at a date nobody knows, broke into hearse A and its horses.

The Department of Antiquities excavated the tomb in 1966. The passage alone took three months.

The measurements of chamber, porch and passage are excavated, and so is the long elevation of the north side, published with the find plan. Excavated too are the objects and animals described above, the places where the clearance left them, the soil layer between the two funerals, and the Roman niches and staircase. Each object in the 3D model has a card that says how its place is known, and the view “Where positions come from” colours every object by the basis of its position: read from the published plan, taken from the excavator’s text, set by the practice of the cemetery, or an estimated count. Of the 804 objects the model shows by default, 52 are placed from the find plan, 214 where the excavator’s words put them, 59 by our reconstruction, and 479 belong to groups whose number is our estimate.

The rest is our reconstruction, and it is drawn to be seen as such: the doorway and its blocking slab; the direction of the four steps; the width of the second passage; where the vehicles and animals stood at the first funeral; which amphora held the bones; the skeleton in the chamber at the second funeral; the number of Roman burials; the size of the mound; and the ground itself, drawn as the sandy clay of the Salamis plain, because the rock under the tomb is not described. Where the scholars disagree, the excavator’s reading is the default, and the other is a switch: bones in an amphora or a pyre; plaques on the floor or under the throne arms; the bed in the porch or in the chamber. The steps are always drawn going down.

Some readings are not offered. The monumental building over the tomb was rejected by the excavator. The silver throne is drawn at its 75 cm height as found, not at the museum’s 105 cm. The vehicle letters follow Karageorghis and the Cyprus Museum’s catalogue (Pilides and Papadimitriou 2012, cat. 237–239), not the letters printed in Barnett’s review, which do not agree with them.

Karageorghis read the Salamis tombs through Homer, as funerals of the kind the Iliad describes, held by rulers who knew the poems or the customs behind them. That the practices took place is excavated; the Homeric meaning is his interpretation. Later scholars have read the same evidence as a display of power by a new kind of ruler. David Rupp, in 1988, saw the royal tombs as statements of authority by kings establishing themselves, and dated Tomb 79’s first burial earlier, to about 760 to 740 BC, against Karageorghis’s 710 to 690 (Rupp 1988). Nicholas Blackwell has compared the royal tombs with the rock-cut tombs of the Cellarka cemetery, where the ordinary citizens of Salamis were buried (Blackwell 2010).

The dates matter for the politics. In 707 BC the Assyrian king Sargon II set up a stele at Kition, on the south coast, recording the submission of the kings of Cyprus. On either dating, Tomb 79’s first funeral falls before or around the start of Assyrian overlordship (Feldman 2014, p. 199). The tomb’s goods point east: the ivories have their closest matches at Nimrud, the bronzes carry sphinxes, griffins, Hathor and the scarab, and the blue glass is Egyptian. The pottery is local, and the roasting spits and firedogs have their parallels in Cyprus and in Greece.

Whichever reading is preferred, the people of the second funeral handled the first one practically. They broke up the animals and threw the bones back in the filling. They stripped the bronze breastplates from the horses and heaped them in a corner with the pieces of the bed. They kept the chariot whole and rolled it out of the way, to the south wall, with its poles against the stone, where it was found in 1966.