j05 TRANSMISSION OF SOUND THROUGH LONG TUBE». 



sensibly in unison. The first an<l more rapid was trans- 

 mitted by the substance of the cylinder, the second by the 

 air. Strokes of a hammer on the last cylinder likewise pro- 

 duced this transmission. We observed attentively with 

 half second chronometers the intervals between tRe two 

 sounds transmitted. We even employed successively sexa- 

 gesimal and decimal watches, to vary the numbers observed. 

 Thus we found 



Difference; in I" ^ ^ observations • • • 0-5'27" . Bell. 



Telocity of the o~2 0'555 " . Hammer 



Widandihe .^^ ,; 0-544" . Bell. 



■>•• 



53 observations. Mean 0542" 



Velocity of 

 trarsiiii-ision 

 through the 

 solid calcu- 

 btc4L 



The interval given by the hammer, and by the bell, ap- 

 peared to us absolutely the same, without any sensible dif- 

 ference. For this reason we have united them in the same 

 mean. Their tones however were very diiFerent. Thus in 

 solid bodies, as in the air, the tone makes no diflperence in 

 velocity of the sound. 



The temperature of the air during the experiment was 

 ll''[5l-8° F.}. The barometer was' about 0'76 [29-9 in.]. 

 Jn similar circumstances the velocity of sound in the air is 

 34084 rnet. [372*487 yds] according to the experiments of 

 the academy, which give 334*02 met. [365*034 yds] for the 

 velocity under the same pressure, and at the temperature 

 of melting ice. For the distance of 197*27 [215*587 yds] 

 therefore, that at which the experiment was made, the time 



of transmission of the sound by the air was 0*579" 



The interval observed between the two sounds was** 0*542" 



Difference, or time of its transmis. thro' the metal • • 0*037' 



We do not pretend to give this small differenoe as exact, 

 since the slightest errour would have a considerable influence 

 on it, but it proves, that the transmission w^s not absolutely, 

 instan'rancous. 

 f rf et rtf e^tpe- 'f he second set of experiments was made by Messrs. Bou- 

 vard and Malus on twice the former number of cylinders, or a 



length 



iinv^nu, on a 



