On RYTHMICAL MEASURES. 73 



and we are often infenfibly led by the fentiments or by the 

 flructure of the words, to execute the duftus rythmicus, or occa- 

 fionally to accelerate or retard the movement. 



In the ancient Greek and Latin languages, the proportion 

 betwixt the longhand the fhort fyllables is better afcertained, more 

 diftinctly marked, and more fenfibly felt, than it is in the mo- 

 dern languages of Europe. Thofe languages are, therefore, 

 fufceptible of a more perfect and a more diverfified rythm. 

 The poets who have written in them have availed themfelves 

 of this advantage, and have left us in their works a variety of 

 very pleafing rythmical meafures. Critics have difcovered the 

 rules by which thefe meafures were conflructed ; and, in order 

 to explain them with greater eafe, have eftablilhed and defined 

 different fmall fcales of long and fhort fyllables, which have 

 been called metrical feet, and by which the different meafures 

 have been parcelled out into their component parts. The 

 names and the nature of thefe feet are generally known, and 

 need not, in this place, be explained. I fhall only, at this 

 time, make two obfervations upon them. The flrft is, that 

 they feem, in general, to mark what I have formerly called An- 

 gle intervals or units of time, and thefe not very large. Any- 

 one may find, when he recites a verfe, that he can eafily pro- 

 nounce two feet in a fecond ; but that he can hardly draw 

 them out, fo as that each of them fhall occupy the time of a 

 fecond. The other obfervation is, that it does not appear to 

 have been the intention of the perfons who defined and applied 

 thefe feet, to divide verfes by them always into equal intervals 

 of time. The formation of them proceeds upon the fuppofi- 

 tion, that when a perfon recites a poem, he pronounces every 

 fhort fyllable in one determined fpace of time, and every long 

 fyllable in a fpace of time exactly double of that in which he 

 pronounces a fhort one. The real duration, therefore, of any 

 one foot will be to the real duration of any other foot exactly 

 in proportion to the number of thefe fmalleft times contained 



Vol. II. K. in 



