Accentual Cursus in Byzantine Greek Prose. 429 
forms is scarcely less marked than in the 0 and 4 forms: Demosthenes 
has 25.2°/, of 0 and 1 forms in the entire oration On the Crown 
(ed. Blass), while Zosimus III has 4.2 °/) of the same forms. If forms 0 
and 4 are irregular, then forms 3 and 6 are certainly so. It is there- 
fore evident that by including forms 3 and 5 as regular in such 
comparisons Litzica made all his conclusions useless. The law loses 
half its meaning by a comparison on this basis. It is rather one 
of the characteristic marks of the cursus rhythm that forms 3 and 5 
are avoided. 
It has been suggested by Paul Maas! that form 6, when it occurs, 
should be counted as regular as forms 2 or 4; in other words, that 
the law requires an even number of syllables between the last two 
accents. In the nature of things the 6 form is very rare (in Zosimus 
they scarcely exceed the 5 form in frequency), and it is difficult to 
find any basis for judgment without a very extensive collection of 
material. The 6 form does not exist in Latin, but there seem to 
be some cases in Greek; since they are so uncommon, they need 
not enter into a general discussion. Until they are certainly found 
to be regular it is safest to leave the law in a negative form, as 
given above. 
Now that the actual law has been seen through the written ac- 
cents, it is possible to work out the details of its application. Any 
satisfactory conclusion must be deduced from the facts themselves ; 
but before proceeding to a statement of such facts, some general 
observations may be in order. First of all, to take Zosimus’ History 
as typical of the rhythmical prose, the case may be stated as fol- 
lows: granted that Zosimus favored two particular forms in the 
arrangement of accents in his clausulz to such an extent as to make 
75 °/, of all cases conform absolutely to these forms, while five other 
possible forms are represented in only 25°/) of the total, it is rea- 
sonable to inquire whether the preferred forms may not be rec- 
ognized in cases where the written accent does not of itself make 
the preferred forms clear. A definite tendency toward a certain 
rhythm is apparent; can not this rhythm be found in apparently 
irregular forms without doing violence to the language ? 
We are dealing with a stress accent in Greek of this period, at 
least similar to the stress accent in English. If now, Zosimus even 
unconsciously sought the rhythm of the 2 and 4 forms, his procedure 
would be analogous to that of a person writing English verse. 
Here not every word has an accent, nor does the verse accent in 

* Rhythmisches zu der Kunstprosa des Const. Manass., Byz. Zschr. 1902. 
