io8 On RYTHMICAL MEASURES. 



of the Phaleucian form may be confidered as fhort couplets of 

 the fame kind. 



A farther deviation from regularity is when fuch unequal 

 timed feet are admitted into the fame line. The mixed iambic 

 and trochaic verfes of the ancients furnifh us with the moft 

 remarkable inftance of this. I formerly ventured to exprefs a 

 conjecture, that in the performance of the mufic, which was 

 adapted to verfes of this kind, the ancients, by fome means of 

 accommodation, of which they were not diftinctly confcious, 

 might occafionally exprefs as equal thofe contiguous feet, which, 

 according to rule and fyftem, were unequal ; in other words, 

 that they fometimes departed from the proportion of two to 

 one, which they eftablifhed as fubufting betwixt the long and 

 fhort fyllables of words. It feems ftill more probable, that 

 they did this when reciting fuch verfes. It is indeed difficult 

 for us to conceive how they could do otherwife. In finging, 

 they might be aflifted in exprefTmg thofe unequal meafures with 

 accuracy, and even, in fome degree, conftrained to do fo, by 

 feeing the arns and thefis of each foot diftinclly marked, and 

 hearing the tingle times uniformly (truck ; but they could not 

 always have the fame amftance, when reciting. The time of a 

 fhort fyllable might be counted and parcelled, when expreffed 

 in the continuous and more extended tones of mufic j but this 

 could fcarcely be done with eafe and certainty in common 

 fpeech ; and, without this, it is not eafy to difcover, how the 

 proportions of thofe unequal feet could be accurately expreffed 

 or perceived. I fhould, therefore, be apt to fuppofe, that the 

 propenfity to equal parcels or meafures of fyllables would pre- 

 vail, as it might be gratified almoft infenfibly, and as there ap- 

 pears to be nothing of fufficient force to counteract it. It is 

 often difficult to determine exactly the proportional quantity of 

 contiguous fyllables, or to lay down any particular proportion 

 as invariably fubufting betwixt them. We can fometimes arti- 

 culate three, perhaps even four fyllables, in our own language, 



in 



