¹ ‘On the Text of the Periplus Maris Erythraei’, Mnemosyne 1975, p. 293 ff. The present paper is the result of a δεύτερος πλοῦς through the same material.

² Unless otherwise stated, the bibliography quoted by me is contained in Frisk, H., Le périple de la mer Erythrée, Göteborg 1927 (Göt. Högsk, Årsskr. 1927, 1Google Scholar), to which I refer the reader for the sake of brevity.

³ This syntactical coalescence is generally considered to be due to the fact that both the infinitive and the participle were on their way to extinction. Other analogous phenomena, whereby infinitives are mixed with constructions entailing a verbum finitum, are mentioned by Aalto, P., Studien zur Gesch. des Infin. im Griech. (Helsinki 1953), p. 74Google Scholar (‘Vermischung’), p. 98 (‘Kontamination’): a beautiful example, I should like to add, occurs in Heliod. i 24.2 (final τοῦ μὴ γίνεσθαι and final καταναγκασθείη). Even those who do not regard the employment of the infinitive and of the participle alongside each other as due to ‘Verwechselung’ and to attenuated ‘Sprachgefühl’ must concede that in later Greek prose infinitives and participles are used in parallel (cf. Weierholt, , Stud. Sprachgebr. Malal., p. 76 ff.Google Scholar).

⁴ Ἀγκύρας means ‘the cables of the anchors’, exactly as at §43 ἀποκόπτειν τὰς ἀγκύρας.

⁵ Scil. by the current (cf. §46, τὰ προληφθέντα πλοῑα).

⁶ Cf. Blass-Debrunner, , Gramm, neut. Griech. ¹¹, §164Google Scholar, 2 (‘der Partitiv … wird auch als Subjekt oder Objekt verwendet’); Mayser, , Gramm. Pap. II, 2Google Scholar, Zweite Halfte, Erste Lieferung, §84, 2.

⁷ On such an employment of the dative in later Greek prose cf. e.g. Mann, , Sprachgebr. Xen. Eph., p. 15Google Scholar.

⁸ Παρ' ὅλον is a form parallel to παρὀλου (cf. LSJ, s.v. παρόλου), just as there exist καθ' ὅλον and καθόλου, δι' ὅλον and διόλου.

⁹ On the meaning ‘mole’, ‘pier’ of χῶμα cf. LSJ, s.v. I, 4. Pollux (ix 34) tells us that the χῶμα is amongst τὰ περὶ τοὺς λιμένας μέρη where merchandise is laden; the author of the Periplus is talking about the Μόσχαλιμήν, where the frankincense is laden. The ‘heaps’ (χώμασι) of frankincense which Fabricius wants to force into the text are an intrusion, all the more absurd as χῶμα, so far as I know, can only denote a mound of earth, soil. Cf. Preisigke, Wört. Pap.. s.v. χῶμα (‘aufgeschüttete Erde, Damm’).

Χώματι is a ‘datif local’ denoting the place where the lading (ἐμβολή) of the return cargo is carried out (ὅρμος … λιβάνου πρὸς ἐμβολήν). Cf. LSJ, s.v. ἐμβολή, 3. An exactly parallel case of ‘datif local’ involving ships in harbour occurs at §44 (Frisk, p. 59): ῥυμουλκοῦσιν αὐτà σταθμοῑς ἤδητεταγμένοις ‘they berth them at fixed quays’.

¹⁰ Κείμενος, used absolutely and referring to structures, means ‘ruinous’, ‘lying in ruins’ (e.g. Ox. Pap. 1287, 17 κειμένη οἰκία). The χῶμα is an οἰκοδόμημα (cf. Thes., s.v. χῶμα, 1790, A), i.e. it is a structure built for the storage of the goods which are to be loaded and unloaded. The author means that the structure was lying in ruins, so that the frankincense could not be kept under lock and key, and, moreover, the place was unguarded, so that anybody might have stolen the frankincense, had it not been for the vigilance of the local god.

¹¹ Note that ‘le participe coordiné avec un adjectif’ (in this case κειμένῳ connected by καί with ἀφυλάκτῳ) is typical of the author's style (cf. Frisk, p. 63).

¹² ‘Dans le Périple, ἀπό tend à supplanter ἐκ' (Frisk, p. 73, n. 1.).

¹³ For this type of parenthesis—the most common in Greek prose—cf. Mayser, , Gramm. Pap. II 3, p. 186 ff.Google Scholar, §168. It is of course also frequent in poetry, cf. e.g. Seelbach, , Die Epigr. des Mnas. und des Theodor., p. 81Google Scholar.

¹⁴ To be more precise, the author is describing not the European lizard, which is skinny and fatless because it must run quickly in order to catch insects, but the large lizard called σκίγκος (cf. LSJ, s.v., and Keller, , Antike Tierwelt, vol. II, p. 275 ff.Google Scholar: ‘Waran’, ‘Dornechse’), a kind of crocodile whose ‘Fleisch und Fett’ (Der Grosse Brehm, Berlin 1964, vol. IV, p. 344) are greedily eaten by the natives.