260 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1953 



its overtones. This has been well shown by a former student of 

 mine, Professor Moklitar, who recorded the sound pressure in the 

 neighborhood of a beating reed. The oscillograms on the left of 

 plate 2, figure 1, show the sound of the reed in its "boot" without the 

 pipe, and the ones on the right those when the reed was loosely coupled 

 to the pipe — a cylindrical tube 45 cm. long and 3.6 cm. diameter — at 

 the same pressures. Without detailed analysis it is easy by visual 

 inspection of the records to see how the overtones of the uncoupled 

 reed intrude as the wind pressure is raised. With the pipe added, 

 however, these raucous overtones are smoothed off. 



The transient sounds of a reed pipe, like those of the violin, are 

 usually smaller copies of the final steady state (pi. 2, fig. 2, 5) and do 

 not comprise extraneous harmonics. 



Bells. — The scientific interest in the sound of bells is centered on the 

 remarkable fact that while, from theoretical considerations, it ought 

 to be one of the most complex of musical instruments, yet through the 

 ages the craft of the bell founder has made the sound of the church 

 bell pleasantly simple in the sense that most of the overtones produced 

 by striking it at the right place are harmonic or quasiharmonic. This 

 he does mainly by varying the thickness and moving the sound bow 

 (the place at which the surface changes from convex to concave). 



The first six partial tones of bells in a carillon are usually adjusted 

 to lie as follows : 



1. Hum note ; the lowest. 



2. Fundamental ; one octave above hum note. 



3. Minor third ; above fundamental. 



4. Fifth ; above fundamental. 



5. Nominal ; one octave above fundamental. 



6. Harmonic decime; a major third above normal. 



(It will be noted that this use of the term "fundamental" is unortho- 

 dox.) Roughly speaking the pitch of the hum note is inversely pro- 

 portional to the circumference at the sound bow. 



A peculiar feature of bell timbre is the "strike note," the tone most 

 prominent to the ear and the one which is intended when the pitch 

 of the bell is named. It seems that this is a beat note formed between 

 two of the higher partials, falling about one octave below the fifth 

 partial or, as some maintain, the fifth partial itself wrongly judged 

 by the ear to be an octave lower than it really is. 



Whereas the nominal is most apparent immediately after striking, 

 it disappears quickly leaving components 1 and 3 as the strongest. 

 After some 10 seconds, only the hum note remains. I have a record, 

 too long to reproduce in full, given to me by the late Prof. Taber 

 Jones, of the sound of a bell at various instants after it was struck. 



