no On RYTHMICAL MEASURES. 



It is not always fafe indeed to reafbn from general principles 

 and analogy upon matters of fact, more efpecially when fuch 

 reafoning appears to be contradicted by pofitive teftimony. I 

 therefore offer thefe conjectures with diffidence. The accent 

 and manner of pronunciation of the ancients being now irre- 

 coverably loft, we have no means of having the matter fub- 

 jected to 'ienfible demonftration. All that we can fay is, that 

 if they did in fact pronounce thefe unequal feet in their juft 

 proportions, and were confcious of doing fo, they pofleffed a 

 power of combining very fmall intervals of time into unequal 

 parcels, to which perhaps no parallel can be found in modern 

 days. 



IV. 



