The Claitstila in Amniianus Marcellinus. 225 



only does it occur freely, but it is the rule rather than the exception. 

 At least we may safely assert it to be the rule in the words Trai- 

 anus, Pompeiiis, plebeitts, Aquileia, Troia, Fonteius, for the clausulae 

 cited, in which / counts as a syllable, are the only clausulae in 

 which they occur.^ But in the clausula xit amnt in flammam (11. 

 131. 4) aiimt has consonantal i, eius often has it (e. g. I. 235. 2; 

 264. 6; II. 20. 22; 35. 28; 91. 3; 99. 1), and mains occurs several 

 times with consonantal i (I. 80. 25; II. 45. 31; 136. 7), but not to 

 my knowledge with syllabic /. Furthermore, intervocalic i seems to 

 be always (except in Traiamis) consonantal when it is pre-tonic; thus 



aiebat I. 94 5; II. 175. 14 akbant I. 318. 22 



maiora I. 23. 28 maiore I. 90. 2 maiores II. 145. 18 



projecit I. 116. 28 



eiectat 11. 19. 30 



seiunctum I. 119. 25 

 Since the syllabication of i regularly appears in certain words, it 

 follows that we are dealing not so much with a metrical expedient 

 on the part of Ammianus as with a characteristic feature of his 

 ordinary pronunciation. Is it peculiar to him, or does he share it 

 with his times ? Lack of positive information as to the pronunciation 

 of his period precludes our answering this question definitely, but 

 it may be pointed out that Ammianus was a Greek, and that pre- 

 cisely this peculiarity was noted by Consentius as a fault in the 

 pronunciation of Latin-speaking Greeks.- 



Coming now to the instances of the syllabication of n, we may 

 remark first that the word behia is always a trisyllable in Ammianus 

 (I. 15. 11 ; 283. 30; II. 32. 13; 46. 31). In this, of course, he agrees 

 with classical usage. 



After 5 in suesco, snadeo and the like u is always syllabic. 



suesco I. 89. 15 adsuesco L 137. 19; 11. 232. 28 



adsuetus, consuetus I. 11. 20; 88.8; 113.1; 136.31; 271.2; 

 IL 112. 19 etc. 



adsuefactus I. 269. 26; II. 38. 16 



suadeo L 124. 23; 308. 31 ; 318. 17 ; H. 247. 8; 262. 21 



suasorum I. 219. 15 



A fairly extensive search in the poets for an instance of the quadri- 

 syllabic scansion of Pompehis has not proven fruitful. For Trota see 

 Sen. Troad. 824, 853. 



' Except those in which the word containing i comes last and which 

 are consequently indecisive ; for example, ventttm est Aqm'leiam (I. 49. 

 31) may be read either as III or IV. 



- Consentius K. V. 394. 11 ff. : Lindsay p. 45. 



