244 SIXTH RKPORT — 1830. 



obtained this ratio from his own experiments. In the next place 

 he establishes, contrary to an opinion previously expressed by 

 M. Biot, that with gases very different in their physical proper- 

 ties, such as hydrogen gas and carbonic acid gas, the nodal 

 sections are exactly in the same positions in the same tube. This 

 is an important fact with reference to the theory of vibrations 

 of aeriform fluids in tubes, from which it readily follows that 

 the relative velocities of propagation of different elastic fluids 

 may be inferred from the musical notes they give out from the 

 same tube ; and taking the ratio of the specific heats of air to be 

 that determined by direct observations on sound, the ratios for 

 the other fluids will be inunediately deduced from these veloci- 

 ties. Representing in general the ratio of the specific heats by 

 1 + /, the quantity/ is taken by Dulong to be the measure of 

 the thermometric effect produced by sudden and equal changes 

 of density of the several fluids ; then assuming the thermometric 

 effects thus developed to be inversely as the specific heats under 

 a constant volume, he is furnished with numbers to express these 

 specific heats, that of air being expressed by unity. Hence by 

 means of the ratios of the specific heats obtained as above men- 

 tioned, numbers expressive of the specific heats under a constant 

 pressure are also arrived at, that of air being again taken for the 

 unit. These last numbers, compax'ed with those which Berard 

 and Delaroche* obtained by direct experiment, are found to 

 agree with considerable accuracy. In concluding this part of 

 the subject I cannot forbear remarking, in the words of Dulong, 

 " how much science owes to the natural philosophers who direct 

 their labours towards giving more and more precision to the de- 

 termination of the numerical coefficients which become theo- 

 retical elements of constant use. ' ' Such are the numerical measure 

 af the force of gravity ; the ratio of the density of mercury to 

 that of air ; the coefficients of the dilatation of mercury and of the 

 gases ; the ratios of the densities of elastic fluids ; the actual 

 velocity of sound in air. All these constants, together with the 

 exact length of the aerial vibrations corresponding to a given 

 musical note, have been employed in arriving at the principal 

 conclusion contained in the valuable memoir which has been the 

 subject of the preceding remarks. 



Propagation of Sound through Liquids. — The experiments 

 of Canton, of CErsted, and Colladon and Sturm have ciscertained 

 the degree in which water is compressible, and proved that for 

 small changes of volume the compi-essions are proportional to 

 the compressing forces. This law seems to indicate thcit the 



* Annales de Chim'te, fom. Ixxxv. pp. 72 and 113. 



