418 H. B. Dewing, 
observing the rule carefully at the stronger pauses only. In this 
connection it is remarked very justly that the condition of the 
text is a matter of great importance in the case of any writer 
whose clausule are studied. 
In the reading of clausulz such words as the article, conjunctions 
or adverbs (Hilfsworter der Sprache) are treated either as accented 
or unaccented according to the situation in which they stand. So 
the following cases are cited from Synesius as regular: xate§aviorn 
tot madous. qv xal avéyeoov. Moeooneoovtas ovx yveyxev. Tatety vm0 Aiwvos. 
ddeapovv toy Iodewv. avdayou tov Adyou. pact thy xapdiay. Enthadéctae ue 
oléy te. The cases cited would excite little scepticism, but there 
was need to define very carefully exactly what “ Hilfsworter” can 
be so treated, because scholars could not readily agree on just 
what words in Greek do bear a weak spoken accent. 
Meyer makes much of the secondary accent. There must be one 
present in such cases, he says, as émavte grdotewoduevos, in the very 
nature of things, and the use of the secondary accent to mark the 
thesis in accentual poetry shows this factor to have been a real 
one in spoken Greek, just as it is in any language where accent 
is a matter of stress. This is reasonable enough, but Meyer intro- 
duces the secondary accent into his clausulz in a quite arbitrary 
way. For example: “ou Shi méol a Evuooas. TLVL pelotipotmeeee 
Duniraniccey: av pavnvee. The uncommon occurrence of the form 
énuwécae mooehouevos is explained as due to the presence of a secon- 
dary accent on the middle one of the three unaccented syllables, 
thus making the clausula equivalent to ——— ——(——). Whether 
this be the correct explanation or not, it is one of the weak points 
of Meyer’s statement that he does not mark off this form of clausula 
from the rest as irregular. That these cases with three syllables 
in arsi! do occur is justified by a supposed “ falscher Nebenaccent” 
which stands on the first of the three syllables instead of on the 
second, as it normally would, thus: pied évecuévos. But this pro- 
cedure can not deserve serious consideration. 
As a result of admitting free play of the secondary accent in 
clausule Meyer goes to the extreme of counting perfectly normal 
only that form which has two syllables in arsi. He is too reckless 
in abandoning written accents and reading this form into all such 

1 Arsis is used here and in the following to denote all the syllables 
without ictus standing between two spoken accents, ‘zes?s to denote the 
two stressed syllables between which are included the syllables forming 
the arsis. 
