² A single sarcophagus lies (about 100 m. away) above and to the right of the tomb, Money N27.

³ All these tombs are briefly discussed by David Money in his account of the sarcophagi of Balboura (Money, 34–5), and are shown on his figures 1 and 4 (N23–4, N25–6, SE12); for a fuller description of the minor tomb buildings of the area, see below pp. 63–7.

⁴ For the wide currency of this term in the surrounding areas, for tombs of this kind, see Kubrinska, J., Les monuments funéraires dans les inscriptions Grecques de l'Asie Mineure (Travaux du Centre d'arch. méditerr. Acad. Polonaise des Sciences, tome 5, Warsaw 1968) pp. 26–31Google Scholar.

⁵ Ibid. pp. 81–4. This term has a smaller geographical range than heroön, but is known in Cibyra, and is very common to the south in Lycia; from parallels there it seems the most suitable designation for our example.

⁶ There are also fragments of what is probably another sarcophagus worked from local stone, see below pp. 52–3.

⁷ On a chord of 0·525 m. a depth of curvature of 0·024 m. This yields a radius of about 1·45 m.

⁸ A half circumference of an arch of this radius would be 4·37 m.; enough archivolt blocks of this type are above ground for more than 7 m. of arch.

⁹ The measurement of the building from the back left-hand corner to the front of the step on which the stair cheek rests is 9·95 m. The tread of this step is 0·30 m, and that of the step above is 0·28 m. The distance from the back of the heroön to the front of the highest surviving step is therefore 9·95 m. –(0·30 m. + 0·28 m.) = 9·37 m. If we reconstruct the building with 5 treads of 0·30 m. (and estimate the tread of toichobate at 0·30 m.) This will give us 9·37 m. – (6 × 0·30 m) = 7·57 m. as the length of the heroön; 6 treads will give us 9·37 m.–( 7 × 0·30 m.) = 7·27 m. These calculations, it should be noted, were made entirely on the left hand of the stairway. The restored distances had to be adjusted somewhat on the right due to the irregularity of the plan of the building.

¹⁰ On a chord of 0·525 m. the fragments show a depth of curvature of around 0·018 m. This yields a calculated radius of 1·93 m., which would give a span of 3·86; but the curved surface is only roughly worked, so the figure is quite compatible with the (much lower) span of 3·5 m. estimated from the plan of the rear niche.

¹¹ An example of a similar tomb building without a porch is found at Ariassos: Cormack, S., AS 39 (1989), 31–40Google Scholar. Of the five built tombs known at Balboura traces of the upper structure only survive in four; and while all four show corner pilasters, none shows any evidence of a porch. This may also provide some slight support for the reconstruction proposed here.

¹² On a chord of 0·525 m. these archivolts possess a depth of curvature of 0·034–0·036 m. This yields a radius of 0·95–1·03 m.

¹³ The blocks do not show this cursory decoration for their whole length, however; if both blocks were set together the length of roughed out architrave and frieze would amount to about 2·70 m.

¹⁴ On this type of Asia Minor sarcophagus, see: Wiegartz, esp. 14–15; Strocka; Ferrari, G., Il commercio del sarcofagi asiatici (Rome 1966)Google Scholar. Wiegartz (14 n. 13) lists 15 such lids (or fragments of lids) which are preserved together with their sarcophagus chests, and a further seven cases where only a lid (or part of a lid) survives (ibid. 15, n. 14); Strocka (69–78) discusses some of these at greater length, and seeks to arrange them in chronological order (15 pieces described in brief catalogue entries; for references to the extant lids not discussed, see ibid, n 59). Strocka publishes three more additions to the series: one from Cumae (now in Pozzuoli), two in Naples; all three lids without the accompanying sarcophagus chest. To Strocka's list may be added another piece: a kline- lid, with the portrait heads of the deceased couple both preserved, now displayed in the garden of the archaeological museum at Burdur.

¹⁵ Cf. Wiegartz, 14.

¹⁶ The figure at the foot of the couch looks at first sight like a draped female figure, but comparison with other such lids shows that it is most likely another Eros.

¹⁷ E.g. Strocka, cat. nos. 5 (Rome K), 6 (Antalya N), 10 (Hierapolis A), 13 (Istanbul B), & 15 (Sagalassos A); also Cumae (fig. 1), Naples A (fig. 26) & Naples B (fig. 28), Pisa (n. 59); Burdur (see above n. 14).

¹⁸ E.g. Strocka, cat. nos. 5 (Rome K), 10 (Hierapolis A), 13 (Istanbul B), 15 (Sagalassos A).

¹⁹ In other cases when the wife extends her right hand in this way she holds a garland of flowers or fruit; e.g. Strocka, cat. nos. 5 (Rome K), 15 (Sagalassos A).

²⁰ The baby Eros at the head of the couch has a pose which seems related to that of several Putti found in this position on other kline-lids (e.g. Strocka, cat. nos. 4 “Istanbul G”, 5 “Rome K”, 6 “Antalya N”); but the figure has been reinterpreted as a sleeping Eros. The draped figure reclining at the foot of the kline, on the other hand, has no parallel at all among the surviving lids. This figure too seems to be represented as sleeping. In this respect these Balbouran Erotes are quite different from those found on other lids, since such figures are usually portrayed as awake, lively and cheerful.

²¹ Strocka, cat. no. 4 (Istanbul G); Wiegartz, 158: from the tomb of K1. Antonia Sabina at Sardis.

²² Cf. Strocka, 82.

²³ The kline-lid “Izmir O” (Strocka, cat. no. 2) provides a parallel of sorts for the “beam ends” becoming fused with the kline-frame, but there the kline-frame is represented in much greater detail. Only one Asiatic kline-lid shows a similar compression/omission of the kline-frame: “Hierapolis A” (Strocka, cat. no. 10), though this piece is unique in the series in that the distinctive “beam ends” are also omitted.

²⁴ See, for example, Naples A and B (Strocka, figs. 26, 28).

²⁵ In addition there were a number of (tiny) pieces of what look like human and animal figures scattered around. It looks as if the figures were smashed off and taken quite recently (the breaks on the fragments were still fresh)—probably when the tomb was wrecked by the illegal digging which forced an entrance into the hyposorion.

²⁶ The sarcophagi found elsewhere on the site normally have larger lids; see Money, 33.

²⁷ Such benches are fairly common at Balboura; Money, 30.

²⁸ Wiegartz (14, n. 13) lists every case in which an Asia Minor kline-lid survives together with its coffin; in each instance the kline-lid was combined with a columnar sarcophagus; in addition, there is only one known case of a columnar sarcophagus ever having a different kind of lid (ibid. 147, “Antalya M”). According to Wiegartz, the kline-lid should be regarded as an integral part of the Asiatic columnar sarcophagus, and even where only the lid survives a columnar sarcophagus chest can safely be assumed (ibid. 15, n. 14). Little remains of the relief decoration on the Balbouran chest, but enough to make it clear that there cannot have been any columns or pilasters framing the figures whose feet are preserved.

²⁹ A kline-lid with only two “beam ends” survives together with its chest in only one case (Strocka cat. no. 7, Rome K) and it is an Arkadensarkophag. Accordingly Strocka suggests (p. 82) that the other lids which have only two “beam ends”, like the three lids he publishes (Cumae, Naples A and Naples B), ought also to come from Arkadensarkophage. On this type of columnar sarcophagus, see Wiegartz, 16 (cf. “Typ mit Bogenarkade” on chart pl. 47), and Strocka, 82 ff.

³⁰ Konya Mus. Inv. 700, and a similar piece in the Sahip Ata mosque at Konya; both pieces published by Mansel, A. M., Belleten 18 (1954) 511–17Google Scholar, figs. 1–5 (German summary p. 518); for the former piece, see Koch, G. and Sichtermann, H., Römische Sarkophage (Munich 1983) 538Google Scholar and fig. 536.

³¹ A. M. Mansel (above n. 30) fig. 4.

³² Cf. sarcophagus chest C 7 at Balboura (Money, 35) which is c. 2·60 m. in length.

³³ Wiegartz, H., AA 86 (1971), 92Google Scholar (N. Himmelmann) fig. 4.

³⁴ Money, 40–1.

³⁵ See the statue bases from the exedras of Onesimos and Meleager: AS 38 (1988), 13 f.Google Scholar; AS 39 (1989), 46–8Google Scholar.

³⁶ Inan, J., IstMitt 27–8 (1977–1978), cat. no. 10, 281 f.Google Scholar; pl. 89, 1: almost certainly from the Sebasteion at Boubon; at present on display in the Cleveland Museum of Art.

³⁷ The nude bronze statue now in the Burdur Museum (Inan, ibid. cat. no. 5, 285–7; pls. 93, 94), likewise from the Sebasteion at Boubon, also shares this pose. All the other extant bronzes from the group assembled by Inan have a quite different stance.

³⁸ The architectural tradition is succinctly traced by Cahill, N., AJA 92 (1988) pp. 485–8CrossRefGoogle Scholar, with bibliography; see also Cormack, S., AS 39 (1989) 36–9Google Scholar, and Fedak, J., Monumental Tombs of the Hellenistic Age (Toronto 1990) 29–46; 65–102CrossRefGoogle Scholar.

³⁹ Here only built tombs of comparable size are listed:

Lycia: (1) Myra: Borchhardt, 61–3, fig. 10, pl. 30–1: two vaulted hyposoria, entrances at the back; (2) Saracik (ancient Apollonia?): Petersen-von Luschan, 151–3, fig. 67, 69–72: barrel vaulted hyposorion, entrance on the right; (3) Sidyma: Benndorf-Niemann, 78 & pl. 22 (left—not right as stated in the text) and Fellows (1841) 155 pl. 16; entrance to the (unexplored) hyposorion at the back; (4) Patara: Texier, III 197 & pl. 189; (5) Faralia nr. Pinara: TAM II, 247Google Scholar: entrance on the left; (6) Lydae: Bent, J. T., JHS 9 (1888) 84CrossRefGoogle Scholar; Roos, P., OpAth 9 (1969) 77–80Google Scholar: two tombs, each with two vaulted hyposoria; entrances to the hyposoria of the northern tomb are on the right; those of the southern tomb are at the back; (7) Xanthos: Fellows (1852) 503; Rodenwaldt, G., JHS 53 (1933) 181–4CrossRefGoogle Scholar: vaulted hyposorion, entered through a shaft to the left of the doorway;

Pisidia: (8) Ariassos: Cormack, S., AS 39 1989; 31–40Google Scholar: vaulted hyposorion, entrance on the right (assuming the east wall is the front);

Pamphylia: (9) GelÇik (ancient Etenna?): (tomb building I) Swoboda, 114–16 & figs. 25–7: hyposorion below the large (walled) pronaos, no entrance visible;

Rough Cilicia: (10) Djambazli, : MAMA III, 34–5Google Scholar & fig. 54, with pl. 17: fig. 53: vaulted hyposorion, entrance at the back; (11) Elaiussa Sebaste: Machatschek, T 11: 98–100, pl. 52–4, figs. 68–9: vaulted hyposorion with entrance on the left; (12) Dösene (ancient Imbrogion?): tomb with double storied faÇade: MAMA III, 25f. pl. 12, fig. 37 (right)Google Scholar; Machatschek, A., Mél Mansel, 255–9, figs. 54–6Google Scholar; Machatschek suspected there was a lower burial chamber, and this has apparently now been confirmed; the entrance is on the right, but blocked by fallen masonry and hence still unexplored (McDonagh, B., Blue Guide to Turkey, (London 1989), 524.)Google Scholar

One might also include the other double-storied tomb at Dösene (MAMA III, 25 & fig. 36Google Scholar; pl. 12: fig. 35) although this latter example is rather different from the others in having two cellas, one above the other; (for a report of what was apparently a similar building at Sidyma, see Fellows 1841, 153–4). For another two (possibly three) tombs with hyposoria in the area of Balboura, see the description below, pp. 63–7. Many more examples of tombs with burial chambers in the substructure are known from inscriptions.

Temple tombs of similar size and format (though without a hyposorion) are also fairly widespread in this part of Asia Minor:

Lycia: (13) Rhodiapolis, “the Heroön of Opramoas”: Petersen-von Luschan, 76–132; figs. 53–6, 63; (14) Sidyma: Benndorf-Niemann, 78 & pl. 22 (right—not left as stated in the text); (15) Patara: Texier, III pl. 185 ff.; (16) Xanthos: Coupel, P. & Demargne, P., Mél. Collart, 103–15Google Scholar;

Pisidia: (17) Selge: Machatschek-Schwartz, , DenkschrWien 152, 1981, 97–8, pl. 21 and fig. 71Google Scholar; (18) Cremna: Lanckoronski II, 181 fig. 142;

Pamphylia: (19) GelÇik: (tomb building II) Swoboda, 116–18, figs. 28–30;

Lykaonia: (20) Isaura: Swoboda, 138–43 fig. 69–72;

Rough Cilicia: (21) Djambazli, : MAMA III, 35–36Google Scholar, pl. 18: fig. 56; (22) Topalaryu Tseshme (two tombs): MAMA III, 30 pl. 14, figs. 44–5; (23) Ura: MAMA III, 84–5, fig. 110, with pl. 37Google Scholar: fig. 109; (24) Dösene (in addition to the two cases mentioned above, three more temple tombs): MAMA III, 23–6 figs. 31–6Google Scholar, with pls. 11–12; cf. also Machatschek, A., Mél. Mansel I, 251–61Google Scholar; Wegner, M., Mél. Mansel I, 575–83Google Scholar; (25) Elaiussa Sebaste: Machatschek, 96–7 (T10), pl. 51, fig. 67.

Several further examples, notably in Termessos Major in Pisidia, are discussed in the text below.

⁴⁰ For the agora at Oinoanda see Coulton, J. J., AS 36 (1986), 61–90Google Scholar, figs. 3, 4 & 8; cf. also the arches of the bath building Mk1 at Oinoanda (Severan: completed either 210/211 or 215/216: AS 36 (1986), 41–4Google Scholar); for the Exedra of Meleager, see AS 39 (1989) 42–62Google Scholar, fig. 9.

⁴¹ IGR III, 468Google Scholar, and Milner, N. P., AS 41 (1991), 46–9Google Scholar. For another building in the area with this type of frieze, see the “Temple of Ares” at Termessos, discussed below p. 56, no. 1.

⁴² See the comments of Coulton, J. J., AS 36 (1986), 67 & 86 f.Google Scholar

⁴³ Above n. 39, no. 16; the excavation and study of the heroön was begun in the 1950s, but has not been completed; the building received a provisional publication by Coupel, P. and Demargne, P. in Mél. Collart, 103–15Google Scholar.

⁴⁴ See Coupel, P. and Demargne, P., Mél. Collart, 106–7 figs. 4–5Google Scholar.

⁴⁵ For the Attic battle sarcophagus, see Demargne, P., Mél. Piganiol, 454–61Google Scholar; Koch, G. and Sichtermann, H., Römische Sarkophage (Munich 1983) 410Google Scholar no. 68; Koch and Sichtermann consider this genre of sarcophagus to have been created shortly before A.D. 200 and to have been popular throughout the first half of the third century (ibid. 408).

⁴⁶ See above n. 39, no. 1.

⁴⁷ For three more tomb buildings each with two hyposoria, see above n. 39 no. 6, and TAM II, 437Google Scholar.

⁴⁸ No sarcophagi remain in the tomb today, but the size of the niches (over 3 m. wide and 0·70 m. deep) would suggest that they were intended for sarcophagi (though a normal sarcophagus would protrude considerably from the niche). There are pedestals or benches still in place in front of these niches, however, which are much greater in depth than would be required merely to support a sarcophagus; the publishers of the tomb offer the plausible suggestion that the sarcophagi were set back in the niches, and the front of the supports then served as bench seats for the funerary cult.

⁴⁹ See above n. 39, no. 13; on the owner see Heberdey, R., Opramoas, Inschriften vom Heroon zu Rhodiapolis, (Wien 1897)Google Scholar. Opramoas held high office in the Lycian League under Antonius Pius. A heroön in Faralia (above n. 39 no. 5) is dated by its inscription to A.D. 146, but it is considerably smaller (3·13 × c. 5·5 m.) and has an unusual design somewhat different from the other temple tombs under consideration here.

⁵⁰ I.e. it stood on a high podium (no indication of any hyposorion) with an entrance stairway, and had a similar in antis porch. Although the cella seems to have had a vaulted interior no evidence survives of any sarcophagus niches inside. Most accessible short description of the building: TAM III, 905 and fig. 13Google Scholar.

⁵¹ The latest documents inscribed on the tomb are dated to the early 150s, which furnishes an approximate date for the building.

⁵² Lanckoronski II, 56: building N 2 “le petit temple corinthien” (see map opposite p. 23); 92–4; figs. 43–6; pl. 8.

⁵³ Lanckoronski II, 219, inscription no. 91; this is one of two inscribed statue bases which were found still in situ against the right wall of the podium. It formerly carried a bronze statue of a certain Hoples Obrimotou, priest of Dionysos, who is described in the inscription as having been sent on an embassy to the emperor Marcus Aurelius. Another statue base was found fallen at the front of the building, and once bore a bronze portrait of a son of this Hoples (another Hoples) who was apparently a priest of Ares (Lanckoronski II, 219 no. 89). Hoples' grandfather and great-grandfather are also known to have been priests of Ares (for the family tree, see ibid, stemma II p. 206). This led to the original interpretation of the building as the “temple of Ares”. However, since there is a sizable vaulted crypt within the podium (according to Lanckoronski, entered from the cella through the vault itself) one wonders whether this may not instead have been the heroön of this prominent local family, designed along the usual lines: temple tomb above, hyposorion below. The monument was built on a slope overlooking the agora, where Osbaras, Hoples' grandfather, had constructed a large stoa for his fellow citizens. For a similar temple tomb of this type, at Aizanoi, likewise situated within the agora of the city itself, see Naumann, R., IstMitt 23–4 (1973–1974) 183–95Google Scholar.

⁵⁴ See above n. 39 no. 6. The tomb possesses two hyposoria in the podium, and an entrance stairway, but here there seems to have been a columned porch.

⁵⁵ Walker, S., Catalogue of Roman Sarcophagi in the British Museum (= CSIR Great Britain 2.2 1990)Google Scholar nos. 63 and 65a & b.

⁵⁶ Heberdey-Wilberg, 205–7. This building stands at the centre of a large funerary enclosure.

⁵⁷ Heberdey and Wilberg thought that one of the two sarcophagi would have been set in the apse, but the interpretation of the inscription offered by Kubinska (above n. 4), 27–8 seems more plausible. Another prostyle-tetrastyle temple tomb in the same necropolis (Heberdey-Wilberg, 204–5, figs. 79–80) of about the same size (c. 10 × 6·5 m.) but this time undated, shows a similar design in the interior: a barrel vault with two relatively shallow arched niches in the side walls which, from their size (2·71 m. in length) were clearly intended to house sarcophagi (the standard dimensions for sarcophagi at Termessos are 2–2·30 × 0·90–1·10 m.: Heberdey-Wilberg, 207). There is a sarcophagus still in situ at the far end of the tomb, though not set in its own niche. Another similar interior arrangement, with two lateral arched sarcophagus niches, is found in the (unpublished) tomb of Licinia Flavilla at Oinoanda.

⁵⁸ The lettering of the inscription is very similar to that of the heroön of Ti. Kl. Agrippeina (Herberdey-Wilberg, 180–7, figs. 52–60), which is dated on the basis of an inscription in which her son Ouaros is named as priest of the emperor Hadrian (ibid. 186 f.).

⁵⁹ Keil, J., Öjh 26 (1930), Beibl. 7–12Google Scholar, figs. 2–4; for the original discovery of the tomb see Öjh 25, (1929), 45–51Google Scholar. The tomb also appears to have had an entrance stairway at the front.

⁶⁰ It carried two fragmentary inscriptions, one in Greek and one in Latin. The sarcophagus belonged to Tatiane's brother, [Q.] Aimilios Aristeides, epi[tropos], an imperial procurator (the restoration is confirmed by what remains of the Latin inscription “proc. Augg”).

⁶¹ Öjh 25, 1929 48–50, fig. 27Google Scholar: in the document, which takes the form of a letter, Tatiane grants permission for her brother Aimilios Aristeides and his wife to be buried with her in her heroön.

⁶² See above n. 30.

⁶³ See above n. 28.

⁶⁴ About 30 such lids are known (apart from the present example), but many consist only of smallish fragments, or remain unpublished (see above n. 14).

⁶⁵ For the attempt to organize the extant pieces in a chronological sequence, see Strocka, 69–78. For the peculiarities of the Balbouran lid, see above p. 50 f.

⁶⁶ According to Wiegartz, Asiatic columnar sarcophagi were produced from the 150s down to c. A.D. 270 (see the table, Wiegartz pl. 47); Strocka, on the other hand, believes that such sarcophagi go on being produced right down to the end of the third century and beyond (Strocka, 85–6).

⁶⁷ If Strocka is right that Kline-lids with only two “beam ends” belong to Arkadensarkophage, then this could have important implications for the date of the piece. Wiegartz has collected 20 fragments of such sarcophagi, but they divide into two groups of widely different dates. Nine pieces belong to an early group (A.D. 160–170), and 11 to a late group A.D. 245/50 to 265/70. (Wiegartz, chart on pl. 47). Strocka, however, in dating the lid of Rome K (Strocka, cat. no. 5), which belongs in Wiegartz's earlier group, suggests that this group should continue down to around A.D. 200, and perhaps later. Strocka adds several more pieces to Wiegartz's late group, emphasizing that during the second half of the third century the Arkadensarkophage outstripped the true columnar sarcophagus in popularity (Strocka, 82–5). Strocka also suggests that the completely smooth kline frame found on the lid from Cumae (and on our Balbouran piece) is an indication of late date (A.D. 270 or later; Strocka, 81–2; 85). At this stage in our knowledge, however—especially in view of the cursory carving of the Balbouran lid—it seems advisable not to try to infer too much about the date of our piece based solely on the number of “beam ends” indicated and the smooth Kline-frame.

⁶⁸ Above n. 39, no. 17; sarcophagus fragment, Machatschek-Schwartz, , Denkschr Wien 152 (1981), 97 fGoogle Scholar. figs. 72–3.

⁶⁹ Above n. 39 no. 9; for the sarcophagus fragment see Swoboda, 51 fig. 41; Wiegartz, 155 (GelÇik).

⁷⁰ Above n. 39 no. 6; for the sarcophagi, see Walker, S., Catalogue of Roman Sarcophagi in the British Museum (= CSIR Great Britain 2.2 1990) nos. 58, 59Google Scholar (southern heroön); 63, 65 A & B (northern heroön).

⁷¹ Above n. 39 no. 7; for the sarcophagi, see S. Walker (above n. 70), nos. 48, 50, 62, 67.

⁷² Above n. 39 no. 16; for the Attic sarcophagus, see above n. 45.

⁷³ Keil, J., Öjh 25, (1929), 45–51Google Scholar.

⁷⁴ Eichler, F., JdI 59–60 (1944–1945), 128–9Google Scholar, refers to this columnar sarcophagus as the Hauptsarkophag from the heroön; at first it was thought to have had a gable-lid, but this idea was later rejected: Rodenwaldt, G., JHS 53 (1933), 194CrossRefGoogle Scholar n. 38. Some scraps of a small figure of Eros are apparently all that is left of its kline-lid (see Eichler, loc. cit. 129, no. 3).

⁷⁵ Butler, H. C., Sardis I, 135–9Google Scholar; 170–4; Morey, C. F., Sardis V.I, 3–5Google Scholar. Morey suggests that a fragment of another columnar sarcophagus, likewise from Sardis, now in the Louvre, might also come from this tomb (ibid. 5; Wiegartz, 166 ‘Paris C’).

⁷⁶ Mansel, A. M., AA (1959), 376–402, figs. 7–43Google Scholar.

⁷⁷ Ibid. 384, (fig. 32). Mansel dates the tomb to the second half of the third century A.D. on the basis of its design and architectural ornamentation; Wiegartz, however, dates the sarcophagus fragment illustrated by Mansel to c. A.D. 195. It is conceivable that all of the remains may not date to the same period.

⁷⁸ Wiegartz, 171 “Side A”, “D bis …”

⁸⁰ Particularly when one considers that the great mausolea of the Hellenistic period, which stand close to the beginning of this tradition, had often contained Kline-shaped sarcophagi; e.g. the heroön of Kalydon: Dyggve, E., Poulsen, F. and Rhomaios, K., Das Heroon von Kalydon, (Copenhagen 1934)Google Scholar. And the mausoleum at Belevi actually has a Hellenistic ruler similarly represented reclining on a kline: Praschniker, C., Theuer, M. et al. , Forschungen in Ephesos 6 (Vienna 1979)Google Scholar.

⁸¹ The fact that the kline-lid was an indispensable part of the Asiatic columnar sarcophagus is regarded by Wiegartz as a strong argument for his interpretation of the sarcophagus type itself as a kind of miniature heroön for the honoured dead (Wiegartz, 24).

⁸² It cannot, however, have been an essential element, since several heroa are known to have had a full complement of limestone sarcophagi of local type: e.g. a double storied tomb at Dösene in Rough Cilicia has three sarcophagi all with lion-lids (above n. 39 no. 12; Machatschek, , Mél. Mansel. 257, fig. 55Google Scholar); a temple tomb in the south necropolis at Termessos (Lanckoronski II, 121–3, figs. 88–9; S 4) seems to have had three sarcophagi of local type with gable lids; and a rather similar tomb at Termessos (discussed above n. 55) has a Pisidian sarcophagus still in situ at the rear of the tomb, which being in the most important position, was probably the coffin of the tomb owner himself. We may also note the variety of sarcophagi found in the heroa at Lydae and Xanthos (above p. 59, nos. 3–6), and indeed in the East tomb at Balboura. Moreover, the simple undecorated sarcophagus now in the southern necropolis at Termessos, inscribed with the names of Ti. Kl. Markellos and Ti. Kl. Agrippeina, members of one of the most prominent families of Termessos (above n. 58), should also remind us that the assumption “elaborate and expensive tomb means elaborate and expensive sarcophagus” does not always apply (cf. Heberdey-Wilberg, 193 n. 5).

⁸³ For the family of Meleager, and for this M. Aurelius Thoantianos, see the inscriptions from the exedra of Meleager: Milner, N. P., AS 39 (1989), 49–61Google Scholar.

⁸⁴ IGR III, 468Google Scholar; cf. also IGR III, 473Google Scholar; the inscription of the agora gate is discussed by Milner, N. P. in AS 41 (1991), 46–9Google Scholar.

⁸⁵ So J. J. Coulton (pers. comm.); D. B. Money interprets W 2 and W 3 as separate “stereotaphs”.

⁸⁶ Cf. also Money, 35.

⁸⁷ A brief survey of the known examples from this part of Anatolia (collected above n. 39, and including only temple tombs of comparable size, described or mentioned in the scholarly literature) gives the following result: sites with only one such temple tomb recorded: Myra, Saraçik, Rhodiapolis, Ariassos, Selge, Ura, Isaura; cities with two; Lydae, Gelçik, Djambazli, Topalaryu Tsheshme. Some cities like Xanthos, Sidyma and Patara clearly had several, though for the most part they remain unpublished. Dösene, like Balboura, has five (all much more impressive examples than Balboura's), but there is no consensus as to the ancient settlement to which they belong (Machatschek, , Mél. Mansel 251–2Google Scholar, favours Imbrogion; Wegner, ibid., 575, Olba-Diokaisareia). Elaiussa Sebaste had eleven (many of them small, and of a distinctive design). The number of such temple tombs at Termessos in Pisidia, and the wide range of tomb types found there, makes it more comparable with major sites like Hierapolis, and puts it in a different class from the other cities under consideration here.

⁸⁸ Morey (above n. 75, 5; 16 f.) dates the tomb on the basis of its architectural ornament and the hairstyle of Sabina's portrait.

⁸⁹ In the inscription on her sarcophagus Kl. Antonia Sabina styles herself hypatike. Strictly taken this ought to mean she was the wife of a consular; but the term was sometimes more freely used merely to indicate membership of a consular family. For the inscription see Morey (above n. 75), 14 f.

⁹⁰ See above n. 76.

⁹¹ See the discussion of the verb periskoutloun in Kubinska (above n. 4), 27–8.

⁹² In terms of architectural ornament the Balbouran tomb is significantly plainer than the majority of the other examples listed above n. 39.

⁹³ A number of other temple tombs probably had statues set up inside or near them: at Patara Texier found one such tomb with a statue base still standing inside the cella (Texier III, 197); the tomb in Myra (above n. 39 no. 1) had two extra niches in the cella which seem intended for statues; tomb T11 in Elaiussa Sebaste (above n. 39 no. 11) has a large shallow niche in the wall opposite the entrance which is of a size suitable for a statue on a base (1·20 m. wide and 2·90 m. high); in the mausoleum at Ariassos (above n. 39 no. 8) the cuttings on the top step on the podium at the front of the building would be suitable for statue bases (this seems more likely than the alternative suggestion of freestanding columns.)

⁹⁴ Money 34.

⁹⁵ In this it is similar to a larger and better preserved tomb at Ariassos (Cormack, S., AS 39 (1989) 29–40Google Scholar).

⁹⁶ Cormack, S., AS 39 (1989) 36Google Scholar.

⁹⁷ Compare the tomb of Agathermos at Termessos, with a sarcophagus on a podium over 4 m. high (Lanckoronski II, 120, fig. 90), and, on a smaller scale, the monument Money W 1–3 in the west necropolis at Balboura, where the dimensions and position of the blocks suggests that W 2 and W 3 were not separate stereotaphs, but together formed a podium below the sarcophagus W 1.

⁹⁸ The species of the animal is unclear; in profile it looks most like a ram, but has a long curving tail, more appropriate to a dog. If a lion was intended, the likeness is poor. I have found no parallel for this motif, and the profile of the “mattress”, a sunk panel surrounded by a cyma reversa and flattened torus, is quite unlike that of normal Asiatic couch type lids. The piece is worn, and does not photograph well, but there is no doubt about its basic form.

⁹⁹ The hyposorion of the Ariassos mausoleum was vaulted (Cormack, S., AS 39 (1989) 32Google Scholar).

¹⁰⁰ Unchanged since 1985.

¹⁰¹ A threshold block lying well to the west of the main fall has pivot holes centred 1·27 m. apart, so cannot belong with the lintel.

¹⁰² This form is given on the Turkish 1:200,000 map (Elmalı sheet); but see also Bean, G. E., BSA 51 (1956) 141Google Scholar n. 1, and id., Lycian Turkey (1978) 166.

¹⁰³ It would have been invisible from the city except near the top of the acropolis.