of Solid Bodies by Internal Resistances. 163 



the cylinder, the volume of air between the glass and the bag is 

 powerfully agitated by the sides of the latter ; but as this volume 

 of air does not communicate with the external air, and as, more- 

 over, it cannot appreciably agitate the thick glass sides, none of 

 the motion imparted is transmitted to the surrounding air. 



If while the bag is freely suspended (the musical box oscilla- 

 ting at some depth) the ear be moved along the sides, a great 

 decrease of the sound is observed as the ear is moved upwards. 

 This experiment gives the reason why so little sound reaches the 

 air through the surface of the water. Suppose a vessel, open at 

 the top and with absolutely rigid sides, filled with water, and that 

 in any part whatever an impulse be given to the water (by a sound- 

 ing body for instance), this impulse will be transmitted in all di- 

 rections, but, owing to reflection from the rigid sides, escape prin- 

 cipally at the free surface. But if the sides are pliable they will 

 give way laterally, and in proportion to this pliability will impart 

 a certain amount of motion to the air; hence, if the vessel is 

 deep enough and the impulse be produced in a deep part, only a 

 small amount of motion will reach the free surface of the water. 



The conductors to be investigated are suitably connected at one 

 end with the musical box suspended in water; the other end is 

 fastened to a sounding-board, to which the ear is applied. 



In this way the sound of the box was transmitted to the ear 

 through a caoutchouc rod 460 millims. in length and 6 mii- 

 lims. in diameter : the deep accompaniment of the small piece 

 of music which the box played was exclusively heard. If various 

 parts of the rod be examined^ it will be found that only near the 

 source of sound is any thing to be perceived of the highest notes. 



To investigate whether the resistance of the air has an influ- 

 ence on these phenomena, the musical box was suspended by a 

 strip of caoutchouc in a vacuous vessel . Through the point of sus- 

 pension and the vessel deep tones only reached the support ; 

 when the caoutchouc rod was replaced by a lead tube, high as 

 well as deep tones were heard by an observer who applied his ear 

 against the support. 



Hence it must be concluded that the weakening of the tones 

 by conduction through caoutchouc does not arise from the resist- 

 ance of the air, but has its foundation in the nature of the caout- 

 chouc itself. 



Rods of wood, steel, glass, lead, wax, of the same dimensions as 

 that of caoutchouc, propagated high and deep tones in sensibly 

 equal proportion. Even in the conduction of sound through a 

 slightly stretched copper wire0'2 millim.in diameter and 30metres 

 in length, no difference could be discriminated in the propagation of 

 high and of deep sounds. But when a lead wire 1^ millim. in dia- 

 meter andll metres in length was interposed between the source of 



M2 



