Remove Ter. Andr. 668

Dating from a time before iambic shortening was well understood, it is now unnecessary.

Co-authored-by: Benjamin Victor <[email protected]>

latin.csv

@@ -2665,7 +2665,6 @@ Ter. Andr. 386,ha(n)c,hoc,Paul Wessner,1902,Flor. Maruc. C 224 (F),"'given this
 Ter. Andr. 509,h. mihi p.,hoc posterius,Gabriele Faerno,1565,,
 Ter. Andr. 533–4,CH optato aduenis aliquot,optato aduenis. CH aliquot,Dziatzko,1884,Paris. lat. 10304 (p),'the manuscripts' indications of parts are themselves conjectural' (B. Victor)
 Ter. Andr. 665,scelus,o s.,Andreas Spengel,1875,ΠbLp,
-Ter. Andr. 668,h. c.,consilium hoc,Fleckeisen,,Vat. lat. 3305 (S),
 Ter. Andr. 686,hem,ehem,Bentley,1726,,
 Ter. Andr. 784,anne h. t. o.,an haec tu omnia,Bentley,1726,Vat. BAV Arch. Cap. S. Pietro H 19 (B),"'B's place in the history of the text makes it unlikely that it alone inherited the original form ""anne""; it is likelier that B's scribe unconsciously substituted ""anne"" for ""an"", these two particles being functionally equivalent.' (B. Victor)"
 Ter. Andr. 898,u. amittere,uis mittere,Gabriele Faerno,1565,Bonon. Bibl. Inst. Filol. class. 945 (b),