^{page 318 note 1} The following conjectures should have been assigned to their true authors thus: ii 6 41 seducel Birt, iii 18 24 atrocis Leo, iv 1 81 (fallitur … luppiter) Tyrrell, iv 3 55 Cravqidos Bergk.

^{page 318 note 2} At iii 1 27 he rejects the words eunabula parui as interpolated, but in a note of twenty lines he does not even mention the one decisive argument which proves them so.

^{page 319 note 1} I do not know what he means by saying ‘It may reasonably be objected [to quamuis in 33] that we should require non minus, not nee minus’, nee is indispensable and non would be inadmissible. He adds ‘the presence of quamuis might be explained on the hypothesis that fertur had been accidentally omitted’; and at iii 14 19, desiring to read capere arma with N, he says ‘supposing capere to have been accidentally omitted (as perhaps in L), arma would easily be expanded into armata’. I wonder what the patrons of N would think if anyone invoked these hypothetical accidents to save the credit of another MS. Fortunately no one ever does.