The Clatisula iji Aiitiiiiaiius Marcelliuits. 183 



all words of four syllal)les in Latin must have had a secondary accent 

 on the first syllable, the cadence in i) types would seem to be collibus db- 

 dide'riint, militibus obscurissimis. If we might believe that the sec- 

 ondary accent in polysyllables always fell on the second syllable 

 before the main accent [exisiimabat, malivolentia), the effect of the 

 ;' types would appear to be the same. Since, however, it is doubtful 

 if we can make this assumption, we must admit the possibility of 

 a slight discrepancy in cadence between different examples of the 

 Y types (perhaps magna dece'rne'batur, but altiora tncditatnrnni). This 

 discrepancy, if genuine, cannot have been very great.' 



Turning now to the three-word clausulae we find that types in 

 three words are made by the use of an atonic monosyllable or dis- 

 syllable, and are frequent. Theoretically, a monosyllable may be 

 introduced either between the accents or after the last accent. In 

 the first case the possible types are {i y and ;' d in forms I and II, 

 y 6 and 6 t in forms III and IV. In III and IV both of the two 

 possible types occur with equal frequency. Both in I and 11 the 

 type ;' cV is common, while the type li y is extremely rare, for the 

 same reason that accounts for the rarity of the simple types I /3 

 and II t^. As to the introduction of a monosyllable after the last 

 accent, it is virtually impossible in I and III, where the monosyllable 

 would have to be preceded by a trisyllabic oxytone to produce the 

 cadence in a three-word clausula.^ Consequently it is not surprising 

 that we find no examples. On the other hand it is quite possible 

 in II and IV, and frequently occurs, giving in II the types (^ l, (haec 

 prosecutus est), y l, {vh'us exustus est) and d ^ {cadavera tracti sunt), 

 in IV the type 6 0^ {fidentius absolntus est)? 



While nearly all the instances of the three-word clausula are 

 occasioned by the employment of atonic monosyllables in the way 

 just indicated, there are a few cases produced by the use of a 

 lightly accented dissyllable. Naturally instances of forms I and II 

 made in this way are very rare. There is no case of I [i (V in the 

 table, but I note one good exam])le of it at the end of a sentence 

 (XL 124. 12 adhiic apud Persas). Of II (i 6 there is one instance 



■ See p. 221 f. 



''■ Fictitious examples are «//;/f Thebais est, (I), optima Thebais est (III). 

 The csesura "would be somewhat more likely to occur in a four-word clau- 

 sula, {usiis est, ars est), but does not. 



^ Cf. II. 9. 9 Poliorcete's appellatiis est, a sentence-ending, and two cases 

 of IV y b- "which I find outside the collections : I. 279. 27 parare adsiiefacti 

 stmt ; 290. 30 sumil exaniniati siint. 



