Mr. R. H. M. Boscinquet on the Theory of Sound, 221 



rection =*6R, we find 394, a difference of two vibrations. 

 Unless this was due to an alteration of the fork in the mean- 

 time, I am confident that an error of this amount could not 

 have taken place. 



The general conclusion I draw is, that the value '55 R for 

 an open end is about the mean of the values I obtain under 

 various circumstances, and may be used provisionally until 

 improved observations enable us to go further. 



The 5-in. iron tube of 2 in. diameter was also used for the 

 determination of the effect of a flange. Two grooves were 

 turned on the outside edges of the tube, and two wooden cir- 

 cular flanges about 7 in. in diameter fitted, with faces flush 

 with the ends of the tube. The interval between the note with 

 and without the flanges I estimated at about 1| mean semi- 

 tone, or about 11 : 12 in ratio. The reduced length of the 

 tube being just 6 in., the change introduced was equivalent to 

 \ an inch — '5 R for two flanged ends, or "25 R for each flanged 

 end. 



The knowledge of the correction for the open end of a pipe 

 may be applied usefully to obtain an approximate standard 

 pitch for practical purposes. If we form a cylinder of gas- 

 pipe, and measure the length, diameter, and temperature, we 

 obtain the vibration-number from the formula 



n=1087 1 + -^^1^ 



2(Z + -55D)' 



where 32 + ^ is the temperature (Fahr.), I the length, and D 

 the diameter. The note is obtained very clearly by blowing 

 short discontinuous jets of air across the end. The pipe must 

 be left to acquire the temperature of the room, and handled 

 quickly with gloves, or, better, placed in a stand. A pipe 25 

 inches long and J inch in diameter gives a medium modern c\ 

 A wider pipe of the same length gives a somewhat lower c' , 

 The note can easily be got correctly in this way within say 

 two vibrations. 



The c" forks sold vary in pitch through about a semitone ; 

 modern forks being generally about 544, and most of those 

 sold as 512 being from 520 to 530. Compared with such 

 forks, notes obtained as above may be regarded as very accu- 

 rate. 



I have made one determination of the correction for a square 

 pipe open at both ends. The pipe was made accurately of the 

 same length as the iron pipe with the flanges ; the side of the 

 square was equal to the diameter of the iron pipe ; and the edges 

 were thinned off outside so as to leave nothing in the nature of 

 a flange. The note is exactly that of the cylinder with flanges ; 



