406 Proceedings of the Royal Irish Academy. 
place (cf. xtv. 5. 1, Neque enim sum admiratus, &c., H, where enim 
makes no sense) ; it does not appear in either Pal. Sext. or Erf. 
xvi. 1.3. Nos ita te desideramus ut amemus. So nearly all 
mss. : ‘‘ My longing for you is conditioned by my love.” But H has a 
line under desideramus, and wideamus written above it. The fact that 
wuideamus appears in no other ms. leads us to suppose that it is some 
sort of a gloss, perhaps wide (= uedere), supplementary of desideramus, 
which the copyist of H supposed to be a variant. 
xvi. 4.1. H had we; but the ¢ is deliberately scratched out. It 
is very rare that an originally right reading in H is altered. 
xv. 5.1. Volebam ad te Marionem remittere quem cum melius- 
cule tibi esset ad me mitteres M; Volebam ad te Marionem muittere 
quem cum meliuscule tibi esset ad me remitteres H, Hittorp. ; remittere 
is read by Erf. and Pal. Sext. in the first clause, but in the second 
both read remitteres. The reading of H, Hittorp. is certainly right ; 
and the corruptions can be readily accounted for by the proximity of 
-nem and me. 
xvi. 6.1. Tertiam ad te hance epistolam scripsi eodem die M; 
scripsi sed si eodem die H, Erf. Sed is written thus (S,). It is often 
confused with se. I think the s¢ of H, Erf. arose from dittographia ; 
but that sedis sound. To suppose a double dittographia of the -s? in 
scripst is violent. ‘‘ This is the third letter I have written to you— 
aye, and on the same day.” The use of sed = ‘‘ aye, and” is mostly 
post-Ciceronian ; but it is found in Cic. Orat. 97, hic est enim cuius 
ornatum dicendi et copiam admiratae gentes eloquentiam in ciuitatibus 
piurimum ualere passae sunt, sed hance eloquentiam quae cursu magno 
sonituque ferretur. 
xvi. 7.1. Nemo nos amat qui te non diligat. So most mss., 
rightly. For nemo qui non (or quin) always takes the subjunctive. 
But H and Hittorp. erroneously read diligit. rf. has diligat. 
xvi. 8.2. Ego certe singulos eius versus singula testimonia puto 
H, Erf. It had been already adopted by Cratander. M erroneously 
adds evus again before testimonia, for which Orelli has conjectured 
dAnbeias, and Klotz uerrtatis. 
xvi. 9.2. Inde austro lenissimo, caelo sereno, nocte illa et die 
postero, &c. So M and most mss. But H and Erf. read, nocte e¢ dre 
illa et die postera. We can only suppose that the common collocation 
nocte et die, added to the close proximity of e¢ die, led to this super- 
fluous addition. 
xvr. 9.3. me cui iussisset curaturum. Ecum et mulum Brun- 
disii tibi reliqui M; curaturum (curatum Erf.)} metum et mulum, &c., 
Pal. Sext., Erf., H (by first hand); curaturum medicum et mulum 
H (by second hand). The reading of M is doubtless right ; and the 
correct but unusual spelling, ecwm (for eqwwm), led to the various inte- 
resting corruptions of the other mss. 
