M. A. Kundt's Acoustic Experiments. 43 



when there was a weight of 200 grammes in the scale-pan. A 

 notched cork in a tube of 6 millims. diameter could draw a 

 weight of 70 grammes. 



The author assigns as the cause of these motions the oblique 

 propagation of the impulse of the transverse vibrations, produced 

 by the unevennesses of the moving body, although the motion 

 travels over the transverse nodes. As the motion ceases at the 

 longitudinal nodes, the longitudinal vibration may exert some 

 inriuence, but cannot produce the motion. 



If, in order to eliminate the influence of the longitudinal vi- 

 bration, a glass rod be suspended by threads in two places, and 

 if by striking it or by drawing across it a violin-bow it be made 

 to vibrate, a notched piece of cork moves from one end to the 

 other in the direction indicated by the notching of the cork. 



The following experiment served for the direct proof of the 

 action of the oblique propagation of an impulse : — Through a 

 block about a cubic inch in volume an oblique aperture was bored 

 about 10 millims. in diameter and at an angle of about 60° to 

 the base. The aperture was closed at the top by a metal plate 

 in which was a small hole ; through this passed a wire fixed below 

 to a piece of wood which accurately fitted in the aperture. 

 Round the wire there was a spiral spring which pressed against 

 the plate at the top and against the piece of wood in the aper- 

 ture. When this apparatus was placed on a sounding rod and 

 loaded by a weight, the spiral in the aperture was thereby com- 

 pressed ; and it moved, either on a longitudinally or transversely 

 sounding rod, always in the direction in which the spring was 

 inclined. 



That the motion oversteps the transverse nodes may be partly 

 explained by the circumstance that the nodal lines are not fixed, 

 but that the sand in a tube is often moved more than half an inch 

 while it is sounding, and partly that the contact does not take 

 place either at a mathematical point or in a mathematical line. 

 The motion at the transversal nodes is also essentially slower 

 than between them, and on the longitudinal nodes it entirely 

 ceases. On transversely sounding rods the moving force of the 

 shifting body is smaller than upon longitudinally sounding bars. 



The author has made a second series of experiments, on the 

 transference of the motion of longitudinally sounding bodies to 

 the air, and on a method based on this principle for determining 

 the velocity of sound in solid bodies and in gases. 



If some lycopodium powder be distributed in a glass tube 

 about 4 feet in length and | of an inch wide, open at both ends, 

 and if the tube, while held horizontally, be set in longitudinal 

 vibration, the powder collects at the bottom in places which 

 correspond to the spiral nodal lines. If, however, the tube be 



