^{page 233 note 1} C.Q.N.S. vii (1957), 129f.Google Scholar

^{page 233 note 2} The parallelism between Libanius and Zosimus is surely far more striking: (Zos.). Klein, W., of course, (Klio Beiheft xii [1914], 106f.), regarded this as proof that both were drawing on Magnus of Carrhae, but Norman himself has shown how unlikely it is that Libanius used Magnus (p. 130). If then the similarity between Libanius and Zosimus is due simply to the fact that both writers naturally used the same words in the same order to describe the same event, the parallelism between Zosimus and the Suidas fragment can hardly be pressed.Google Scholar

^{page 233 note 3} De perzische expeditie van keizer Julianus volgens Ammianus Marcellinus, Proefschrift, Groningen, 1959; at p. 10 he states Norman's conclusions without comment, but at p. 17 uses them to refute the contention of Maenchen-Helfen (A.J.Ph. lxxvi [1955], 392) that Eunapius used Ammianus for his description of the Huns.Google Scholar

^{page 233 note 4} C.Q. N.S. x (1960), 152 f.Google Scholar; cf. Vollebregt, J. C., Symbola in novam Eunapii Vitarum editionem (1929), p. IIIGoogle Scholar: it might also be observed apropos of that Photius ridiculed Eunapius' fondness for adjectives in (cf. the examples collected from the fragments by Cobet, , Mnemosyne x [1882], 38).Google Scholar Norman's article seems to have been unknown to the latest writer on Julian's Persian expedition, Dilleman, L., Syria xxxviii (1961), 87–158, as also was the work of Brok and Chalmers. Dilleman's detailed discussion of the sources adds little to the conclusions of Thompson, but is useful on points of topography.CrossRefGoogle Scholar

^{page 234 note 1} Mendelssohn, Whereas, Zosimus (1887), praef. xlvf.Google Scholar, had argued that Zosimus had abandoned Eunapius before this point. Cf. Thompson, E. A., The Historical Work of Ammianus Marcellinus (1947), p. 135, Chalmers, p. 154, Brok, p. 14.Google Scholar

^{page 234 note 2} Exsuperius, apparently a common name among soldiers (as pointed out by Linden-broghap. Wagner-Erfurdt [1808]iii, p. 73), is more likely to be correct: probably represents the commoner Exsuperantius (cf. Schwab, , Fleckeisen Jahrb. Supplb. xxiv [1898], 692): clearly the distinction between the names is not great. For a similar error (?) in Zosimus, cf. 3. 30. 4, where he gives as a tribune whom Ammianus calls Maximus (25. 6. 3).Google Scholar

^{page 235 note 1} Julian was evidently keen to revive these coronae, though he departed from the traditional qualifications for the awards. The corona obsidionalis was properly awarded to the general who lifted a siege: nor can the corona navalis Julian awarded shortly after wards have been won more veterum (Ammian. 24. 6. 16).

^{page 235 note 2} See Brok, , pp. 9f.Google Scholar, for a complete list of these writers and the other sources for the expedition. Another possibility is Julian himself: he would hardly have ha A the leisure to compose an account of his Persian as he had of his Gallic campaign, but might easily have touched on the incident in a letter— cf. the letter he sent to Libanius from Hiera-polis (Ep. 98, p. 180Google Scholar Bidez) describing in detail the early stages of the expedition. Suidas had access to letters of Julian that no longer survive: cf. Suidas s.v. and Bidez, , Œuvres de Julien i (2), p. xii.Google Scholar

^{page 235 note 3} While agreeing with Thompson (p. 31) that there is ‘no evidence whatever for this identification’ I do not see on what grounds Norman asserts that it is impossible and the author ‘cannot be Magnus’.

^{page 236 note 1} As demonstrated by Boor, de, Byz. Zeit. xxi (1912), 381 f., xxiii (1914/19), 1–127.Google Scholar

^{page 236 note 2} Krumbacher, , Gesch. der Byz. Literatur² (1897), p. 259.Google Scholar

^{page 236 note 3} Cf. Boor, de, Byz. Zeit. xxiii (1914/1919), 22 f. (Eunapius at p. 30) for Suidas' method when giving the author's name in short citations from the Excerpta.Google Scholar

^{page 234 note 1} The inaccuracy and unreliability of the excerptors can be demonstrated from the one occasion when they quote the same passage of Eunapius twice: once in the Exc. de Leg. (ed. Boor, de, 1903), p. 598Google Scholar, and again in the Exc. de Sent. (ed. Boissevain, , 1906), 91.Google Scholar The E.S. preserves an additional sentence (Cobet, which, Mnemosyne x [1882], 35 quite wrongly wished to delete as an interpolation: interpolators seldom indulge in such extravagant metaphors as or alliteration like and gives for the of the E.L., before cutting Eunapius short in the middle of a sentence, the rest of which is given in the E.L.Google Scholar

^{page 234 note 2} See my note ‘Priscus of Panium and Malalas, John in Suidas’Google Scholar, to appear in C.R. N.S. xiii (1963)Google Scholar, for two careless confusions by Suidas in the matter of proper names. Büttner-Wobst in the praef. to his ed. of the Exc. de Virt. et Vit., pp. xxix–xxxiii gives examples of the numerous omissions and transpositions that make it impossible to use Suidas' quotations from the Excerpta to reconstruct the original text of the Excerpta, in spite of the fact that he used a much earlier manuscript than any we now possess.Google Scholar