126 Mr. Meikle on the 



author assumes, be always the same in the same medium. For, 

 to attain the same final velocity, the circumstances must be 

 similar to those of a weight descending an inclined plane of a 

 given height ; where, abstracting from friction or other resis- 

 tance, the accelerating force is inversely as the plane's length. 

 But, in the case before us, the law of the force accelerating the 

 cyhnder must be of a very opposite description 5 for, as we 

 shall afterwards see, in order that the velocity of sound, as de- 

 duced by this sort of investigation, may be independent of the 

 intensity, or of the degree of condensation, the elasticity of the 

 air would require to be either independent of, or to vary in- 

 versely as, the density, which are alike absurd ; but here the 

 elasticity is supposed to vary directly as the | power of the 

 density. 



That the above are not the only serious charges which may 

 be brought against Mr. Ivory's investigation, will appear from 

 the following extracts ; to which I shall subjoin some remarks, 

 for the purpose of pointing out a few more of the tacit assump- 

 tions and undefined steps, which are not unfrequent, and for 

 setting their merits and mutual relations, which are sometimes 

 curious, in a proper point of view : — 



'• Conceive a slender horizontal tube of an indefinite length, 

 containing air in a state of equilibrium ; and let x, reckoned from 

 a fixed point in the axis of the tube, be the distance of a small 

 cylinder of air within the tube, the thickness (length) of which 

 is equal to dx. Suppose now that the cylinder is pushed forward 

 by some force to the distance x+z from the fixed point, and 

 that it occupies the length dx+dz in the axis*. It is to be 



* It is not, however, this movement of the cylinder over the space z 

 that is considered in the sequel of the investigation ; but its retracing of it 

 occasioned by the natural tendency of the air to regain its equilibrium, 

 and which accelerates the cylinder back over the space z towards the 

 assumed point from which the distance x + z was reckoned. A concus- 

 sion or tremor is thus produced in the air, and propagated from atom to 

 atom along the line x ; and it is conceived that this tremor or sound moves 

 uniformly along x with the velocity, whatever that be, which the cylinder 

 has acquired during its acceleration over the line z. This supposed uniform 

 velocity of the cylinder projected along x is further conceived to be the same 

 with the velocity it happens to have, whenever its density equals the mean 

 actual density of the medium. If so, how does this consist with the well 

 known fact, that the series of aerial vibrations conducting sound through 

 the atmosphere always get feebler and feebler as they become more distant 



