Varro: On the Latin Language (De Lingua Latina)

Having set forth this example, we add that we have included distinctions, abbreviations, and the distribution of lines in the apparatus only in those places where the matter seemed memorable for some reason or where the words absolutely had to be presented. For example, a distinction was not to be entirely neglected, because quite often the readings which learned men had corrected by conjecture were actually read correctly in the book itself, whereas in other cases a significant error was observed.

Very little need be said about abbreviations: for even those which Spengel in his preface, volume 2, page 5 brings forward as "less common" are, in reality, common. And it is strange that the same man, at V 69 (p. 22, 11), noted that the abbreviation q q = quoque also/too, added in the margin by F², is not found elsewhere: for it returns immediately at 28, 5 in the very verse and in many other places (such as 137, 14; 139, 12; 146, 3; 154, 6; 168, 14; 179, 13; 183, 17). No more notable is dr = dicitur it is said, or similar (but rather that at 37, 17 we ventured to think de = dicunt they say [see notes], and similarly at 40, 9 de = dicam I will say [but cf. notes], and at 107, 14 di = dicam). Furthermore, see the notes at 128, 2.

Accents appear quite often, which we deemed worth noting in a few places, such as quaré wherefore (or -eˢ) — which stands out immediately on the first page of the codex (above v. 27) — because of an error by Keil (at 93, 14, who derived from it either quarem or quare? why?): this accent is almost always preserved, unless qua re by what thing/wherefore is separated. Cf. also at 119, 3; 162, 12; 188, 28. Likewise, at 142, 5 q̃ is read (= qui who, i.e., quomodo how). Furthermore, at 144, 24 sq., one could easily link analogîas to ἀναλογίας analogies, but it can also be repeated from faulty pronunciation.

We have observed the corrections of the ancient scribe more accurately than our predecessors, the frequency of which carries some weight for judging the authority of the codex and the necessity of the critical art.1 Cf., among other places to be cited in part later, at 4, 14. 15; 5, 7. 9; 6, 3. 13. 30; 7, 6. 12; 8, 9. 11; 11, 1; 12, 2; 13, 9. 17; 17, 13. 15 sq.; 19, 15. 20; 20, 11; 22, 11; 24, 13. 27; 25, 13 (?); 27, 15. 20; 29, 2. 7. 13; 30, 3. 13; 31, 8; 32, 2; 34, 4. 7. 9; 35, 3. 4. 15; 36, 4; 79, 19; 80, 13. 14; 81, 7; 83, 1. 12. 14; 85, 9. 11; 87, 7; 88, 26; 89, 16; 91, 2; 93, 18; 94, 2. 14 sq. 16; 95, 18; 96, 10. 12; 97, 1. 21; 98, 2. 18; 99, 6. 12; 100, 10; 101, 2. 14; 103, 2. 9; 104, 3. 15; 106, 8; 107, 2. 6; 108, 12; 110, 7; 111, 11 sq.; 113, 2. 7 sq.; 117, 6. 9; 119, 3.

1) Sometimes a correction made long ago lies hidden in what appears to be continuous script, such as 5, 22 (cuiusque of each); 18, 10 (uelterāina?); 43, 11 sq. (eee); 44, 1 (circumo around); 45, 20. 22 (see app.); 51, 10 (cf. note); 86, 11 (didesp.); 96, 3 (cf. note); 114, 1 (see app.); 135, 9 (simileli); 148, 5 (cconum); 148, 10 sq. (cf. note); 160, 5 (cf. note); 162, 26 (cf. note); 175, 19; 184, 18; 187, 5 (casisus case); 190, 15 (see app.).