^{page 47 note 2} Earlier studies which I have found of great assistance are: on the development of the pelta as an ornamental feature in sculpture, Germania, vi (1922), 24–31Google Scholar (G. Hock, Römische Inschrift aus Kastell Obernburg a. M.); and on the zoomorphic treatment of the terminals, Jahreshefte des österreichischen archäologischen Institutes in Wien, xxi–xxii (1922–4), 229–50Google Scholar and Taf. III–V (Josef Zingerle, Kyknos-Relief in Wien). Zingerle's view that the occurrence of the zoomorphic pelta on the northern frontiers of the Empire is attributable to Celtic influence was followed by Lehmann-Hartleben in his discussion of the pelta-shaped standard finial found at Dura-Europos in 1935–6 (The Excavations at Dura-Europos, Preliminary Report of the Ninth Season of Work, 1935–6, Part 1 (1944), pp. 187 ff.). Certainly the Celts showed a fondness for both pelta and griffin but the borrowing was probably by them from classical sources, as with the lyre and palmette motifs, rather than vice versa.

^{page 47 note 3} Arch. Camb. cxiii (1964), 83Google Scholar and pl. iii.

^{page 47 note 4} J.R.S. lv (1965), 200Google Scholar, 223 and pl. xix, 1, 2.

^{page 47 note 5} P.S.A.S. xcvi (1962–3), 196Google Scholar and xcvii (1963–4), 202 and pl. x.

^{page 47 note 6} Sir George Macdonald, The Roman Wall in Scotland (2nd ed. 1934), ch. x.

^{page 47 note 7} Ibid., p. 412 and pl. iii.

^{page 48 note 1} Macdonald, op. cit., p. 393.

^{page 48 note 2} e.g. ibid., pl. LXVI, I.

^{page 48 note 3} Daremberg et Saglio, Dictionnaire des Antiquités, s.v. clipeus; Pauly–Wissowa, Real-Encyclopädie, s.v. Pelte; Snodgrass, A. M., Arms and Armour of the Greeks (1967), pp. 78–79Google Scholar.

^{page 48 note 4} Herodotus, vii, 75.

^{page 48 note 5} A useful list appears in Daremberg et Saglio, s.v. Amazones; cf. von Bothmer, Dietrich, Amazons in Greek Art (1957Google Scholar), passim.

^{page 48 note 6} Lippold, Georg, Griechische Schilde (Münchener archaäologische Studien, 1909), pp. 494–5Google Scholar, figs. 26, 29.

^{page 48 note 7} Ibid., figs. 27, 28.

^{page 48 note 8} Ibid., pp. 498 f. and figs. 31, 32.

^{page 48 note 9} Magnesia am Mäander, Taf. xii ff. (the temple of Artemis Leucophiyene).

^{page 48 note 10} Milet: Ergebnisse II, 1908 (Das Rathaus von Milet), Taf. xv.

^{page 48 note 11} Richter, Gisela M. A., Ancient Italy (1955)Google Scholar, fig. 205.

^{page 49 note 1} von Massow, W., Die Grabmäler von Neumagen (1932)Google Scholar, Taf. 46, 240a.

^{page 49 note 2} Daremberg et Saglio, op. cit., s.v. clipeus.

^{page 49 note 3} Cf. two from Ostia (? Pliny the Elder and Younger) kindly brought to my notice by Miss Toynbee (Le Arti, 1941–2, pp. 172–80 and pi. LII).

^{page 49 note 4} Ibid., pl. LIII, fig. 13; Grabar, André, The Beginnings of Christian Art (1967)Google Scholar, pl. 292.

^{page 49 note 5} Daremberg et Saglio, loc. cit., and figs. 1668, 1669.

^{page 49 note 6} Cf. Vergil, Georgics, ii. 388–9; tibique oscilla ex alta suspendunt mollia pinu. For a full discussion, cf. Daremberg et Saglio, s.v. oscillum; Pauly–Wissowa, s.v. oscilla.

^{page 49 note 7} La mosaïque gréco-romaine (Centre National de la Recherche Scientifique, Paris, 1965)Google Scholar, passim.

^{page 49 note 8} The fullest account is in Pauly–Wissowa, s.v. Gryps, although this confines itself to the areas round the eastern Mediterranean; an accessible account, which also deals with the spread to Siberia and Mongolia, appears in Bolton, J. D. P., Aristeas of Proconnesus (1962), pp. 85–93Google Scholar.

^{page 50 note 1} Minns, E. H., Scythians and Greeks (1913), pp. 266Google Scholar ff.

^{page 50 note 2} Ibid., pp. 298, 343 (fig. 249), 391, 423 (fig. 315, 629) (pl. v, 18); Rostovzeff, M., Iranians and Greeks in South Russia (1922), p. 37Google Scholar.

^{page 50 note 3} Rostovtzeff, M., ‘The Animal Style in South Russia and China’ (Princeton Monographs in Art and Archaeology, xiv (1929)), pp. 6–9Google Scholar.

^{page 50 note 4} The literature is plentiful but the better-known pieces are reproduced by most scholars: for Iranian griffins (to as late as the Sassanian period), cf. J. A. Potratz, Die Skythen in Südruβland (1963), pls. 26–32; for Scytho-Siberian grypomachies, cf. Borovka, Gregory, Scythian Art (1928), pls. 46, 50, 73Google Scholar; for the Dnieper burials, E. H. Minns, op. cit., ch. viii; for the penetration of the Iranian and Greek animal style into south Russia and its further spread eastwards, M. Rostovtzeff, opp. citt. and Le Centre de I'Asie, la Russe, la Chine et le style animal (1929); and Minns, E. H., ‘The Art of the Northern Nomads’ (Proceedings of the British Academy, xxviii).Google Scholar

^{page 50 note 5} Minns suggests (Scythians and Greeks, p. 78) that the Scythian tribes possibly had the same feeling for deer or griffins that the early Turks had for the wolf.

^{page 50 note 6} J. D. P. Bolton, op. cit., is the most recent and accessible treatment.

^{page 50 note 7} Herodotus iii, 116; iv, 13, 27.

^{page 51 note 1} Unless these are Arimaspian women; cf. Daremberg et Saglio, op. cit., s.v. Arimaspi.

^{page 51 note 2} Herodotus, iv, 110–16.

^{page 51 note 3} Walters, H. B., Catalogue of the Terracottas in the British Museum (1903), p. xviiiGoogle Scholar.

^{page 51 note 4} Ibid., pp. 403–5, D611–17; cf. Campana, Antiche opere in plastica, pls. LXXV–LXXX.

^{page 51 note 5} M. Rostovtzeff, Iranians and Greeks in South Russia, pls. xix, xxi, 1; J. D. P. Bolton, op. cit., pp. 5ff., 89 ff., claims that the first grypomachy in Greek art appears on the mirror from the Kelermes barrow near the river Kuban dated c. 575 B.C; it includes a scene of two savages (? Arimaspi) fighting a griffin.

^{page 51 note 6} e.g. the bow represented on the statue of Apollo in his temple at Cyrene (Smith, A. H., Catalogue of Greek Sculpture in the British Museum, ii (1900), no. 1380)Google Scholar; note, too, that the pelta is used as a repetitive motif on the god's lyre.

^{page 51 note 7} Two extremes in date are provided by the bronze, protome from Olympia of the seventh century B.C. (P. Demargne, Naissance de l'art grec (1964), figs. 480, 481) and the bronze furniture leg from Persia of the third/fourth century A.D. (J. G. H. Potratz, op. cit., Taf. 32).

^{page 52 note 1} Cf. the fine sarcophagus in the Fitz William Museum of the early second century A.D. (L. Budde and R. Nicholls, Catalogue of the Greek and Roman Sculpture in the Fitz William Museum, Cambridge (1964), pl. 52.no. 160), and the cippus in the British Museum of similar date (A. H. Smith, op. cit. iii, fig. 52.no. 2350); for a pair of griffins guarding an urn, cf. Archaeologia Belgica, 42 (1958), pl. XXVII, no. 36 (from Buzenol, citing references to Rhineland parallels).

^{page 52 note 2} Garcia y Bellido, Antonio, Esculturas romanas de España y Portugal (Madrid, 1949), pp. 432–3Google Scholar and pl. 313 (nos. 440 and 441).

^{page 52 note 3} Espérandieu, E., Recueil général des bas-reliefs de la Gaule romaine, i (1907), 415Google Scholar.

^{page 52 note 4} Ibid., 376.

^{page 52 note 5} Toynbee, J. M. C., Art of the Romans (1965), pl. 58 and p. 254Google Scholar; Le Arti, 1941–2, pp. 163–71 and pls. LI, LII.

^{page 52 note 6} I am greatly indebted to Miss Toynbee for providing me with this fine enlargement and allowing me to publish it.

^{page 52 note 7} Espérandieu, op. cit., no. 731.

^{page 52 note 8} Ibid., nos. 717, 727, 732, 745.

^{page 52 note 9} Ibid., no. 1406.

^{page 52 note 10} Ibid., no. 879.

^{page 53 note 1} R. Eggers, Eine syrische Göttertrias (in A. Betz and G. Moro (eds.), Römische Antike und frühes Christentum (1962), pp. 189–93 = Wiener Studien, liv (1936), 183 ff.Google Scholar). It was first published by Valentin Kuzsinszky in Aquincum, Ausgrabungen und Funde (1934), p. 99 and Abb. 4.

^{page 53 note 2} O.R.L. lii, 11.

^{page 53 note 3} Zingerle, op. cit. (on p. 47, n. 2), 229–31 and Taf. iii, iv; recently republished by Neumann, A., Die Skulpturen des Stadtgebietes von Vindobona (Vienna, 1967)Google Scholar, Taf. xxv, xxvi and 25, 26, who does not relate them (I am indebted to Miss Toynbee for this latter reference).

^{page 53 note 4} Zingerle, op. cit., 244 and Abb. 84.

^{page 53 note 5} G. Hock, op. cit. (on p. 47, n. 2).

^{page 53 note 6} Espérandieu 4990 (sarcophagus from Trier), 5200 (fragment of funerary inscription from Neumagen), 5518 (sarcophagus from Königshafen), and 5631 (sarcophagus, perhaps from Avolsheim).

^{page 53 note 7} O.R.L. liii and liii1, pi. vii, fig. 1.

^{page 53 note 8} M.Z., xiv, 30 = C.I.L. xiii, 6956.

^{page 53 note 9} Collingwood, R. G. and Wright, R. P., The Roman Inscriptions of Britain, i (1965)Google Scholar.

^{page 54 note 1} J. D. P. Bolton, op. cit., p. 88.

^{page 55 note 1} P.S.A.S. xcvii, 203.

^{page 55 note 2} In The Romans in Redesdale (A History of Northumberland, xv (1940), 134Google Scholar).

^{page 55 note 3} Arch. Aeliana, 4th series, viii (1931), 191–3Google Scholar.

^{page 55 note 4} Haverfield in A History of Northumberland, x, 471, no. 5; J. Collingwood Bruce, Handbook to the Roman Wall (12th ed., 1966 by I. A. Richmond), pp. 72–73.

^{page 56 note 1} Brought independently to my notice by Miss Vera Evison and Mr. John Cowen; I am indebted to Miss Evison for further references.

^{page 56 note 2} Lapidarium Septentrionale, p. 338, no. 650.

^{page 56 note 3} C.W.², xxxix (1939), 19–30Google Scholar.

^{page 57 note 1} Not ‘goose-head terminals’ as described in R.C.H.M., Eburacum: Roman York (1962), p. 126Google Scholar, no. 91.

^{page 57 note 2} J.R.S. xlvii (1957), 227.