^{page 165 note 1} This name has been adopted in preference to the building did not, in any case, lie outside the walls until Italian Terme extraurbane, as being more distinctive. The the second decade of the fourth century.

^{page 166 note 1} For the monticelli see Romanelli, P., Leptis Magna, Rome, 1925, pp. 72Google Scholar ff. To the west of the Wadi Lebdah they combine the functions of a flood-water channel and a defensive rampart. The corresponding earthworks, which complete the defensive circuit to the east, were identified from the air in 1945.

^{page 166 note 2} De Aedificiis, vi. 4.

^{page 166 note 3} One such blocking-wall can be seen in pl. XLVIII, beside the cisterns. A few yards to the south, at the angle of the insula, a second late wall completely blocks the street.

^{page 166 note 4} Many buildings were, however, demolished to provide building-material.

^{page 166 note 5} Ammianus, xxvii. 6. Traces of overturned walls and of extensive burning noted by the excavators along the north-west perimeter (Weekly Report of 28th June 1933) may perhaps be ascribed to these events.

^{page 167 note 1} For convenience the points of the compass are described throughout as if the building were exactly orientated.

^{page 167 note 2} Now considerably restored; but the sandstone blocks of the south wall are plainly visible in pl. xxxv, b, taken during excavation.

^{page 168 note 1} Memoirs of the American Academy in Rome, iii, 1919, pl. 77Google Scholar. Krencker, D. and Krüger, E., Die Trierer Kaiserthermen, Augsburg, 1929, pp. 260–2, fig. 391aGoogle Scholar.

^{page 168 note 2} The sockets for the wooden roofing of the service corridor can be seen in pl. xxxv, a, bottom left.

^{page 170 note 1} Perkins, J. B. Ward, ‘The Italian Element in Late Roman and Early Medieval Architecture’, from Proceedings of the British Academy, xxxiii, 1947–8, pp. 17–18Google Scholar and footnote 54. For a bibliography of the literature up to 1939, see Creswell, K. A. C., Early Muslim Architecture, ii, Oxford, 1940, p. 101Google Scholar.

^{page 172 note 1} Bartoccini, R., Le Terme di Lepcis, Bergamo, 1929, fig. 65Google Scholar.

^{page 177 note 1} Vitruvius, v. 10, testudines alveolorum. The functioning of this device is described in the next paragraph.

^{page 177 note 2} For permission to reproduce the testudo in the Stabian Baths, and for help in clearance and measurement, thanks are due to the Superintendent of Antiquities for Campania, Professor Maiuri, and to Dr.Olga Elia, Director of Excavations at Pompeii. For earlier schematic reproductions of this, and of the similar fittings at Boscoreale, see Daremberg-Saglio, s.v. Thermae, fig. 6877; Mau, A., Pompeii in Leben und Kunst, Leipzig, 1908, figs.97Google Scholar and 204; Fusch, G., Hypokausten-Heizungen, Hannover, 1910, figs. 64Google Scholar and 65. Krencker, D. and Krüger, E., in Die Trierer Kaiserthermen, Augsburg, 1929, pp. 209–12Google Scholar, figs. 285-9, discuss and illustrate the caldarium fittings of the Camp Baths at Lambaesis, which offer a close parallel to the Hunting Baths.

^{page 179 note 1} No other figured vault- or wall-mosaics have, as yet come to light in Roman Tripolitania. The fragments of wall-mosaic from the Hadrianic Baths at Lepcis( cini, R.Bartoc, Le Terme di Lepcis, 1929, figs. 90, 91)Google Scholar and those the small Baths, near the House of the Orpheus Mosaic (unpublished) show only floral and other purely of decorative motifs.

^{page 179 note 2} Cf., for example, the female Satyr suckling a fawn from the great frieze in the triclinium of the Villa dei Misteri Pompeii ( Maiuri, A., La Villa dei Misteri, pl. 5)Google Scholar.

^{page 180 note 1} For the popularity of such scenes in Tripolitania cf. (i) room 3, below; (ii) the well-known polychrome mosaic from the Villadel Nilo at Lepcis ( Africa Italiana, v, 1933, p. 6Google Scholar, fig. 3), now in the Lepcis Museum; (iii) the predominantly black-and-white mosaic from Fonduk Naggaza ( Rendiconti della Pontificia Accademia Romana di Archeologia, ser. 3,vi, 1930, p. 93Google Scholar, fig. 4), also in the Lepcis Museum; and (iv) two panels in the Four Seasons mosaic from the Villa di Dar Buc Amméra, near Zliten ( Aurigemma, S., I mosaici di Zliten, 1926, figs. 63, 71–4)Google Scholar, at present stored in the Sabratha Museum.

^{page 180 note 2} The account of the discovery contained in the Weekly Report of 8th January 1933 refers to a thick coat of painted plaster found overlying the mosaics at the time of their discovery.

^{page 186 note 1} Cf. the very similar stucco floral rosette, framed in a concave stucco medallion, from the vaulted ceiling of the frigidariumof the Theatre Baths, Sabratha (unpublished).

^{page 188 note 1} For a similar treatment of this ‘fried-egg’ motif in mosaic see the background of the Diana emblema from the Oceanus Baths, Sabratha (unpublished, early third-century ?), now in the Sabratha Museum, the floor of room 5 in the Office Baths, Sabratha (unpublished), and the central emblema of one of the mosaic pavements in a coastal villa, c. 2 miles east of Sabratha, which was partially excavated in the summer of 1948 (unpublished).

^{page 192 note 1} For a recent discussion of these buildings, with bibliography, see Perkins, J. B. Ward, ‘The Italian Element in Late Roman and. Early Medieval Architecture’, in Proceedings of the British Academy, xxxiii, 1947–8Google Scholar.

^{page 192 note 2} Die Antike, 1936, pls. 10, 11Google Scholar.

^{page 193 note 1} Bull. Arch. 1919, pl. xxGoogle Scholar; Krencker, D. and Krüger, E., Die Trierer Kaiserthermen, Augsburg, 1929, fig. 264Google Scholar.

^{page 193 note 2} Krencker, and Krüger, , op. cit., fig. 300Google Scholar, emending previous plans.

^{page 193 note 3} Thenae: Krencker, and Krüger, , op. cit., fig. 317Google Scholar. Khamissa: Bull. Arch., 1919, p. 59Google Scholar; Krencker, and Krüger, , op. cit., fig. 276Google Scholar.

^{page 193 note 4} Bâbiskâ: Butler, H. C., Ancient Architecture in Syria (Princeton Expedition to Syria, 1904-1905, part II, Leyden, 1907–20),section B, pp. 170–6.Google ScholarSerdjillâ: Butler, H. C., Architecture and other Arts (American Archaeological Expedition to Syria, 1899-1900, part II), New York, 1903, p. 165Google Scholar.

^{page 193 note 5} Butler, , Architecture and other Arts, pp. 384–90Google Scholar.

^{page 193 note 6} Butler, , Ancient Architecture, section A, pp. 260–5Google Scholar (Bosra), and pp. 439-40 (Sha'ârah).

^{page 193 note 7} Butler, , Ancient Architecture, section B, pp. 300–3Google Scholar.

^{page 193 note 8} Craswell, K. A. C., Early Muslim Architecture, Oxford, 1932, pp. 253–72Google Scholar (Qusayr 'Amra) and 273–6 (Hammâm as-Sarakh).

^{page 194 note 1} Fisher, C. S., ‘Bath C’, in Antioch-on-the-Orontes, vol. i, Excavations of 1932, Princeton, 1934, pp. 19–31Google Scholar, plan, pl. v; reproduced by Morey, C. R., The Mosaics of Antioch, 1938, p. 12Google Scholar.

^{page 194 note 2} Well-preserved examples can be seen at Sabratha, notably the Office Baths and the Theatre Baths, and perhaps also the Museum Baths.

^{page 194 note 3} Calza, G., ‘Il piazzale delle Corporazioni e la funzione commerciale di Ostia’, Bull. Arch. Com., 1915, pp. 178Google Scholar ff.; P. Romanelli, ‘Ricordi di tripolitani a Roma e in Italia’, ibid., 1928, pp. 71 ff.

^{page 194 note 4} Aurigemma, S., ‘L'elefante di Leptis Magna’, Africa Italiana, vii, 1940, pp. 67–86Google Scholar, citing also the literary evidence for the ivory-trade.

^{page 194 note 5} S. Aurigemma, loc. cit.

^{page 195 note 1} S. Aurigemma, loc. cit. It has been suggested, in view of the merchant-ships portrayed on the same base, that Porfyrius was an exporter of wild-beasts (Guidi ap. Rostovtzeff, , Social and Economic History of the Roman Empire, Italian ed., Firenze, , 1933, pl. lxvi; followed by Aurigemma, p. 84). But the inscription patently replaces an earlier text, to which the ships belongGoogle Scholar.

^{page 195 note 2} Bartoccini, R., Africa Italiana, ii, 1928, p. 48Google Scholar; S. Aurigemma, loc. cit.

^{page 195 note 3} This suggestion was made in the first instance by Professor Giacomo Caputo.