150 Dr. Barton and Mr. Garrett on Vibration Curves 



curves were projected on to a screen and made visible to a 

 small audience. 



At this point the desirability appeared of making simul- 

 taneous photographs of the motion of the string together with 

 that of one of the resonant bodies. This would show whether 

 the variations of the latter' s motion were traceable to bad 

 bowing of the string. It would also exhibit the relations of 

 amplitudes and phases of the two motions. 



On surveying the possible ground to be covered in experi- 

 ments of this kind, it was seen that there was too much for a 

 single article of moderate dimensions. To exhaust the field 

 the following variables occurred to us as needing successive 

 treatment : — Pitch of string, its material and dimensions, 

 place of bowing, place of touching (if any) to elicit an 

 "harmonic,'" place of observing string, place of observing 

 belly in length and width, motion of bridge vertically and 

 horizontally, motion of membrane covering one of the holes 

 of the sound-box. All these would need repeating if the 

 string were excited by plucking, and again if it were struck. 

 Then for the violin there would be eight such sets, one for 

 each of the four strings if bowed and one for each if plucked. 

 This w f ould entail in all considerably over a hundred sets of 

 observations. 



It was accordingly decided to confine our attention at the 

 outset to a metal string on a monochord excited by bowing, 

 and to deal with simultaneous photographic records of the 

 motion of string and belly. These form the subject of the 

 present paper. It might seem more logical to observe 

 the bridge before the belly. But the importance of obtaining, 

 as early as possible, simultaneous records, led us to prefer 

 the other alternative. For it seemed desirable to observe 

 both belly and string near their middle points ; because the 

 belly might naturally be expected to have the greatest 

 amplitude there, and the string near, but not quite at, its 

 middle point shows all the Fourier components of its motion. 

 And these places of observation of belly and string being near 

 together, facilitated the focussing of the two corresponding 

 images on the same plate. Whereas to do the same for the 

 bridge and the middle of the string would entail much labour 

 for which at the outset there appeared no compensating 

 advantage. 



Experimental Arrangements. 



The monochord used bears no name, but is a fine well- 

 seasoned instrument which has been at this College for twenty 

 years. It was securely mounted horizontally on one table 



