^{page 1 note l} In I 246 f. ‘neque enim arcano de pectore fallax | Tantalus… periit’ Baehrens proposed arcani, ‘quod qnominus ad fallax referamus, impedimur collocatione uerborum’ says Mr Klotz. One may steal a horse while another may not look over the hedge.

^{page 2 note 1} Mr Klotz's kindly willingness to explain what he does not understand is less of a help than a hindrance. Examples follow at II 637 f. and IV 2 f.; and there are others at II 327, IV 318, V 702 (the most amazing of all), IX 19.

^{page 4 note 1} Wernsdorf proposes a change; Sudhaus imagines the expression to be a personification characteristic of ‘our Stoic author’, and is so completely at a loss for a parallel that he compares Iuu. X 193–5 ‘rugas, | quales… in uetula scalpit iam mater simia bucca’.

^{page 8 note 1} anhela with tollit is not, as the scholiast says, ‘aestuosa’, but pictures the day panting up the steep of heaven.

^{page 8 note 2} Lucan's errors in this department, though not so many as Scaliger made them out to be, are sometimes monstrous, and one of them (X 210–8) involves the dogstar.

^{page 9 note 1} Mr Klotz, with the old editors, reads dominumue, against the best MS; Mueller, Kohlmann, Wilkins, and Garrod are right.

^{page 12 note 1} What I then said against his explanation of Ovid's BeῙides and Lycurgῑdes was afterwards confirmed by the emergence of κοδρείδης and λαγειδης from papyri of Callimachus.

^{page 14 note 1} I do not add his punctuation of silu. II i 129, which, though equally vicious, had more excuse until I analysed the sentence in Class. Rev. 1906 p. 41.