^{page 11 note 1} This allowed for theoretical expenditure of £15,000 per annum. In the first year January-December 1974, a total of £19,130 (partly estimate) was spent; to this must be added £1,194–17 expenditure on the underwater survey of August 1973.

^{page 11 note 2} Carried out by a team under the direction of Messrs. J. Little and R. Yorke. Preliminary report (together with partial summary of the present excavation) published in The International Journal of Nautical Archaeology and Underwater Exploration, iv (1975), 85–101Google Scholar.

^{page 13 note 1} As some, most recently Cintas (below, n. 3), still argue that the Punic harbours of Carthage were elsewhere, it must be stated that the words ‘harbour’, ‘rectangular harbour’, or ‘circular harbour’ are used in the following account as the conventionally accepted titles for the two lagoons at Salammbo. Whether or not they are proved to be the ancient harbours must depend on the evidence set out in this and subsequent reports.

^{page 13 note 2} Libyca, viii, 96. The early identifications are referred to by de la Malle, M. Dureau, Recherches sur la topographie de Carthage (Paris, 1835), pp. 1–18.Google Scholar of fundamental importance is the plan published by Falbe, C. T. in Recherches sur l'emplacement Carthage (Paris, 1833).Google Scholar Useful bibliographies of subsequent research and discussion are given by Gsell, S. in Histoire ancienne de l'Afrique du Nord (Paris, 1920) ii, p. 38Google Scholar, n. 2, and, for the more recent material, by Cintas, P. in Le Port de Carthage (extrait du Manuel d'archéologie punique, ii, Paris, 1973), pp. 10–11.Google Scholar Cintas's publication is the most recent on the subject and rejects the identification of the circular and rectangular lagoons as the site of the Punic ports.

^{page 14 note 1} Falbe's plan (op. cit.) is the earliest reasonably accurate survey. For the underwater evidence see Roquefeuil, De, Comptes rendus de l'Academic des Inscriptions (1898), 20–39, 653–66Google Scholar; ibid. (1899), 19–38; Hantz, ibid. (1900), 53–78; also Little and Yorke, op. cit. Aerial survey by Baradez, J. in Karthago, ix (1958), 45 ff.Google Scholar Up-to-date maps at scales of 115000 and 1:2000 and vertical air photographs of the area can be obtained from the Service Topographique in Tunis.

^{page 14 note 2} Summarized most usefully by Carton in Documents pour servir à l'étude des ports et de l'enceinte de la Carthage punique (Paris, 1913Google Scholar: a reprint of articles in Revue tunisienne, vols. lxxxviii–xcvi) and, for the more recent material, by Cintas, , op. cit., pp. 73–8Google Scholar.

^{page 14 note 3} Fouilles à Carthage (Paris, 1861), pp. 87–118.Google Scholar

^{page 14 note 4} Bulletin archéologique du Comité (1909), pp. 51–3Google Scholar and pl. VI (accurate plan).

^{page 17 note 1} Revue tunisienne (1912), 37 ff.Google Scholar

^{page 17 note 2} This is confirmed by the underwater evidence. See Little and Yorke, op. cit.

^{page 17 note 3} Ibid. 88, 91.

^{page 17 note 4} Schematic section based on 1925 excavations published by Harden, D. B. in Iraq, iv (1937), 63Google Scholar, fig. 2. Sections of Cintas's 1947 excavations in Picard, C., Carthage (Paris, 1951,) pp. 24–5.Google ScholarSee also Harden, , The Phoenicians (Pelican, 1971), pp. 86–93Google Scholar for a useful summary, with bibliography, of this partly published site.

^{page 17 note 5} Op. cit. (above, p. 13, n. 2).

^{page 17 note 6} It has been added to by spoil from the present work.

^{page 19 note 1} Here, as elsewhere, the centre-to-centre measurement between lines of blocks is given.

^{page 19 note 2} Probably from the El Haouaria quarries, Cap Bon.

^{page 20 note 1} Op. cit., pp. 107–8.

^{page 20 note 2} Πρακτικὰ τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς ‘Εταιρίας (1885), pp. 63–8.Google Scholar The evidence for these and other Greek ship-sheds is summarized and discussed by Blackman, D. J. in Morrison, J. S. and Williams, R. J., Greek Oared Ships (C.U.P., 1968), pp. 181–6.Google Scholar I am indebted to Mr. Blackman for his help with the interpretation set out here.

^{page 20 note 3} American Journal of Archaeology, viii (1904), 227–37.Google Scholar

^{page 20 note 4} Blackman, , op. cit., p. 182Google Scholar, note.

^{page 20 note 5} Assuming the cisterns to be the same height as, for example, those found in the late Punic houses on the Byrsa, : Cahiers de Byrsa, v (1955), 31 ff.Google Scholar and ix (1960–1), 77 ff.

^{page 20 note 6} Beulé found two engaged column drums, similarly fluted and stuccoed, and having a diameter of 47 cm. One was on the island, the other beneath the (Roman) quay at the side of the harbour: op. cit., pp. 108–9.

^{page 22 note 1} Libyca, viii, 96.

^{page 22 note 2} Assuming the west wall, whose west side has not yet been established, to be of the same width as the south and north walls, i.e. 2·7 m.

^{page 25 note 1} Usually the vaulted substructure of a temple podium was built up from the precinct level, as, for example, in the central temple of the ‘forum’ at Sbeitla.

^{page 26 note 1} I am indebted to Miss A. Claridge for these and other marble identifications.

^{page 27 note 1} Op. cit., pl. iv and p. 105.

^{page 27 note 2} These may be the columns mentioned by Beulé as ornamenting the house of Khasnadar (op. cit., p. 105).

^{page 27 note 3} The piece is unfinished; a socket at the wrist still contains an iron dowel; and there is a socket on the inside of the forearm for the attachment of some metal object. Miss A. Claridge kindly made these observations.

^{page 29 note 1} Bartoccini, R., Il porto romano di Leptis Magna (Rome, 1958), pp. 59 ff.Google Scholar

^{page 29 note 2} For Roman lighthouses in general, see Daremberg, and Saglio, , Dictionnaire des antiquités (Paris, 1885)Google Scholar, under ‘Pharos’, and Thiersch, H., Pharos: antike Islam und Occident (Leipzig, 1909).Google Scholar For Dover, , see Arch. Journ. lxxxvi (1929), 29–46.Google Scholar

^{page 29 note 3} Thiersch, , op. cit., pl. 31, p. 24.Google Scholar

^{page 29 note 4} Bartoccini, , op. cit., pls. XXVII, XXVIII.Google Scholar

^{page 29 note 5} This was the highest level left by the earlier excavations.

^{page 29 note 6} The general ground level is assumed to be not less than 2–9 m. above sea level, on the evidence of Temple 2 (see fig. 9). Steps similar to those of F. 49 would rise c. 2 m. above this.

^{page 30 note 1} Centre-to-centre measurements of the foundations.

^{page 30 note 2} Using the numbering of Saumagne, C., Bulletin archéologique du Comité (1924), 133Google Scholar.

^{page 35 note 1} Revue tunisienne (1912), 37.Google Scholar

^{page 35 note 2} Saumagne, C., Bulletin archéologique du Comité (1924), 133.Google Scholar

^{page 36 note 1} I am indebted to Mr. D. Pringle for his help with the research for this account.

^{page 36 note 2} A human adult skull without its mandible was found in a pit of unknown date cut through a wall of Building 1.

^{page 36 note 3} Chronica Gallica, a. ccccxxv : ed. Mommsen, T., Chronica Minora, i (Berlin, 1892), 658Google Scholar.

^{page 37 note 1} De Aedificiis, vi, 5; De Bella Vandalico, i, 20–1, 23; ii, 26.

^{page 37 note 2} This can be inferred from the account in Vand. i, 20–1.

^{page 37 note 3} Vand. i, 20.

^{page 37 note 4} Ibid. ii, 26; Jed. vi, 5.

^{page 37 note 5} Description de l'Afrique Septentrionale, transl. de Slane, W. M. (Algiers and Paris, 1913, 2nd edn.), p. 94.Google Scholar

^{page 37 note 6} For example, Carton, , Revue tunisienne (1912), 41.Google Scholar

^{page 37 note 7} Second-century cemetery at Salammbo railway station, Carton, , op. cit., 38Google Scholar; ‘pagan’ cemetery just south of Koudiat el Heurma, ibid.; Christian burials on Koudiat el Hobsia, ibid. 40. See also Bordy, , Carte archéologique et topographique des ruines de Carthage (Paris, 1907)Google Scholar.

^{page 37 note 8} See Bordy, , op. cit. The wall was shown on Falbe's map (op. cit.) as PunicGoogle Scholar.

^{page 37 note 9} Carton, , Revue tunisienne (1911), 306–19Google Scholar; (1912), 55, fig. 1.

^{page 37 note 10} A wall about 5 m. wide and running east-west was revealed some 26 m. south of the circus by the geophysical survey carried out by a team of Polish archaeologists in 1972. It was interpreted by them as the Theodosian wall: Iciek, Andrzej, Jagodziński, Aleksander, Kolendo, Jerzy, and Przenioslo, Jacek, Carthage (Warsaw, 1974), p. 45 and fig. IIGoogle Scholar.

^{page 37 note 11} C.R.A.I. (1923), 306–11.Google Scholar

^{page 38 note 1} Falbe, , op. cit., pl. 1, nos. 77–8Google Scholar; M.E.F.R. xii (1892), 258–9Google Scholar; Bordy, , op. cit.Google Scholar

^{page 38 note 2} M.E.F.R. xii (1892), 259.Google Scholar

^{page 38 note 3} The Ilôt de l'Amirauté stelae all came from the robber trenches of Temples 1 and 2 or later levels; the Avenue Bourguiba stele was from the robber trench of the city wall.

^{page 38 note 4} Cf. Miedan, M. Hours in Cahiers de Byrsa, i (1951), 15 ff.Google Scholar

^{page 39 note 1} Cintas, P., Céramique punique (Paris, 1950), p. 69, pl. 11, 3b.Google Scholar

^{page 39 note 2} Cf. Cahiers de Byrsa, i(1951), 52–3.Google Scholar

^{page 39 note 3} Text of the inscriptions: C.I.S. i, 2, nos. 443 and 444. I am indebted to Mr. Mihalik for pointing this out.

^{page 39 note 4} , Beulé (op. cit.), p. 106.Google Scholar

^{page 39 note 5} Bulletin archéologique du Comité (1908) p. ccxlvGoogle Scholar; C.I.S. i, 2, nos. 3007–30, pl. LXII.

^{page 39 note 6} C.R.A.I. (1912), p. 284.Google Scholar

^{page 39 note 7} B.A.C. (1911), p. ccxxxvii.Google Scholar

^{page 39 note 8} Ibid. (1913), p. ccxliv; ibid. (1914), pp. cliii–cliv. For these earlier finds in general see Merlin, A., Catalogue du Musée Alaoui, Supplément II (1922), pp. 105–6Google Scholar, nos. 1267–1316; Picard, G.-C., Catalogue du Musée Alaoui, Collections puniques, i (1955), 215 fF.Google Scholar, nos. cb 798–869.

^{page 39 note 9} Mr. Mark Hassall kindly helped with these notes.

^{page 39 note 10} In this and the other measurements quoted, the height precedes the width.