ON MUSICAL INSTRUMENTS. 405 



the notes by literal characters; the rules of counterpoint were formed by 

 degrees from the experience of the ecclesiastical musicians; and early in the 

 eleventh century, Guido of Arezzo, otherwise called Aretin the monk, 

 introduced, together with some improvements in the theory and practice of 

 music, a new method of naming the notes by syllables. 



Some curious experiments on sound may be found in the works of Bacon, 

 but they added very little to the true theory of aeustics, and some of them are 

 not perfectly accurate, Galileo rediscovered what was well known to Aris- 

 totle, respecting the nature of sound; for the words of Aristotle had been 

 so much misunderstood and misinterpreted, that he could have profited but 

 little by them. His cotemporaries Mersenne and Kircher made a variety of 

 very ingenious experiments and observations, on sound and on soundin<i- 

 bodies, many of them unknown to authors of later date. The theory of 

 the ancient music was very accurately investigated, in the middle of the 

 17th century, by Meibomius : our countryman Wallis, also, besides employ- 

 ing much learning and penetration in the illustration of the ancient 

 music, observed some insulated facts in harmonics which were new and 

 interesting. 



Sir Isaac Newton's propositions respecting the velocity of the propagation 

 of sound were the beginning of all the more accurate investigations relating 

 to aeustics. It must not be denied that these propositions contain some very 

 inconclusive reasoning respecting the nature of the motions constituting 

 sound, by which the determination of a particular case is erroneously extended 

 into a general solution of the problem. The velocity is, however, truly cal- 

 culated, because it is in fact independent of the particular nature of the vibra- 

 tion, and all that is wanting to generalise the proposition is the remark, that 

 if the velocity of sound is the same in all cases, it must be such as the calculation 

 . indicates. An error nearly similar was committed by Brook Taylor, who in the 

 year 1714 investigated the time occupied by the vibration of a string or 

 chord upon a particular supposition, which he co:>sidered as a necessary 

 condition, but which in fact confined the inquiry to a limited case. It 

 happensjhowever, that the same determination of the frequency of vibration is 

 equally true in all possible cases. Sauveur obtained, about the same time 



