M. A. Kundt's Acoustic Experiments. 47 



were measured with a pair of compasses, and the mean taken 

 of 5—10 measurements. 



I. Steel. — Three steel bars, of the same kind of steel as far as 

 could be ascertained, were used : one was 1002*7 millims. in 

 length and 10 millims. in thickness ; the two others were 5 mil- 

 lims. thick and 1001'7 millims. and 501 millims. long respec- 

 tively. Taking the the velocity of sound in air as unity, they 

 gave the velocities respectively 



15-325, 15-334, 15*343. 



II. Brass. — A brass rod, about a metre long and 5 millims. 

 thick, gave, in three successive experiments, 



10-88, 10-87, 10-86. 



A second, similar one gave 



1094, 



and, with a dust-tube not closed at the end, 



10-90. 



IIT. Glass. — A thin glass rod gave, in three successive expe- 

 riments, 



15-24, 15-25, 15-24. 



IV. Copper. — For a copper wire above a foot in length there 

 was obtained 



11-966 



Wertheim found for cast steel 14*961, for steel wire 15*108, 

 for copper 11*167, numbers which agree with the above as ac- 

 curately as can be expected from the difference of the materials. 

 In the case of gases the author did not use the external tube, 

 but filled different tubes with different gases and counted the 

 dust-figures produced. There is thus obtained an easily repro- 

 duced image of the various velocities of sound, whose ratio can 

 only be obtained approximately by counting, yet with sufficient 

 accuracy. 



On an entire glass wave the number of half waves of the gas 

 was — 



Carbonic acid ... 40 



Air 32 



Coal-gas 20 



Hydrogen .... 9 



Taking the velocity of sound in air as unity, we get the fol- 

 lowing velocities for different gases : — 



Carbonic acid .^ . . 0*8 



Coal-gas 1*6 



Hydrogen .... 3*6 



