^{page 291 note 1} R.Ph. xliv (1970), 203–5.Google Scholar

^{page 291 note 2} See most recently Schrijvers, P. H., études sur la poétique et la poésie de Lucrièce (1970), 294–6Google Scholar, who envisages the tradition as going back to the invasions of Alexander, quoting Meurig-Davies, E. L. B., Elephant Tactics, C.Q. xlv (1951), 53–5, who had referred to the wars against Pyrrhus and Carthage.Google Scholar

^{page 292 note 1} Cf. D. A. West's comment on this passage in The Imagery and Poetry of Lucretius, 21.

^{page 292 note 2} That the shells of the giant turtles of the Indian ocean are of exceptional dimensions and sturdiness is as true today as it was two thousand years ago, and doubtless they could be put to use as roof tiles, but I am unaware that any credible example is on record much more than a quarter as long as the fifteen-cubit shells described by Aelian (N.A. 16. 17) large enough for ‘quite a few people to dwell underneath’:,. Pliny (N.H. 9. 12. 35) also insists on the capacity of a single shell (‘tantae magnitudinis… uti singularum superficie habitabiles casas integant').

^{page 292 note 3} This fragment is overlooked in the fragments of Agatharchides printed in Müller, Jacoby, and G.G.M. For the reading of the manuscript (Keller gives), see Giannini's recent edition of the Paradoxographorum Graecorum Reliquiae, who tentatively proposes here. A comparison with parallel descriptions (abstracted from Agatharchides himself) in Phot. Bibl. 250(G.G.M. i. p. 139) and Diod. Sic. 3. 2I. 5.,, suggestsas a possible reading in Agatharchides (i.e. with top of the shell uppermost, opp.): for confusion, see Greg. Cor. p. 269.