42 M. A. Kundt's Acoustic Experiments. 



of the tube to the nearest node and sometimes in the opposite 

 direction. If the ring, after it had moved in the direction from 

 the end towards the node, was taken off and put on again 

 reversed, it moved from the node towards the end. It was 

 found that the direction in which the cork travelled depended 

 upon its shape, so that a somewhat conical cork inside the tube 

 always moved in the direction from the larger to the smaller 

 basis, while a cork ring of a somewhat conical section, placed 

 round the tube, always moved from the smaller sectional surface 

 towards the larger one. The degree of conicity need in either case 

 be but small. The energy of the motion is materially increased 

 if a number of notches be made in a somewhat conical cork, so 

 that it presents to the glass a series of conical surfaces all in the 

 same direction. It is even sufficient to file the cork roughly, 

 so that all the inequalities are in the same direction. 



Other materials may be used instead of cork — stoppers and 

 rings of vulcanized and of uuvulcanized caoutchouc, or rolls of 

 list ; only the latter are apt to prevent the tube from sounding. 



Perfectly cylindrical smooth stoppers or rings of cork or caout- 

 chouc, on the contrary, move either not at all, or, like sand, to 

 the nearest transverse node. This is also the case with conical 

 and even roughly filed stoppers and rings of wood, metal, and, in 

 general, substances of small elasticity. If any aperture (for in- 

 stance a triangle) be cut with a knife in a sheet of paper, the cut 

 surface will be nearly always conical; and when placed on a lon- 

 gitudinally vibrating rod, the paper moves in accordance with 

 the law given. If the conicity is removed by smoothing the 

 paper, it moves at most to the next transverse node. 



Corresponding movements take place on longitudinally vibra- 

 ting bars. Strips of plate glass 5 or 6 feet in length, were clamped 

 and rubbed longitudinally, and the body to be investigated placed 

 upon them. Wood, metal, and glass were observed to be passive. 

 This was also the case with perfectly flat plane surfaces of more 

 elastic bodies ; if, however, they were roughened on the under 

 surface, and the roughnesses were all in the same direction, they 

 moved in the opposite direction to, or away from, the nearest lon- 

 gitudinal node. Most energetic was the motion of a piece of cork 

 having a section of about a square inch, notched like a saw on 

 the under surface, so that one surface of each tooth was normal 

 and the other oblique to the base. The force with which this 

 motion is effected is evident from the following experiment. 

 Such a piece of cork as has just been described was placed upon 

 a plate of glass, loaded with a weight of 200 grammes, and at- 

 tached, by means of a string which passed over a pulley, to a 

 scale-pan. When the strip was set in vibration, the cork moved 

 in the direction from the vertical to the oblique section, even 



