Feb. 34, 1888.] 



SCIENTIFIC NE^AAS. 



177 



NOTES ON SOUND. 



A RECENT article in the Evening Standard re- 

 corded with considerable detail a number of well- 

 authenticated instances of the travelling of the sound of 

 artillery over very great distances. The noise of the 

 guns at Waterloo is said to have been heard by Sir 

 Edmund Head at Hythe in Kent, at Walmer, and near 

 Aldborough in Suffolk, the distance in the first case 

 being no less than no miles. The bombardment of 

 Antwerp in 1833 was heard on the beach at Southwold 

 in Suffolk, and stories are told of the sound of firing 

 ■having been heard over distances of 180 and even 200 

 miles, but some corroborative testimony seems to be 

 needed. The frightful volcanic convulsion in the Tom- 

 boro mountain, in the island of Sumbawa, in 18 15, is said 

 to have been heard at Sumatra at the distance of 900 miles. 



of this, and in attempting to calculate the velocity of 

 sound, he found that his result was one-ninth too small, 

 and with the readiness with which the scientific men of 

 those days generally invented some ingenious hypo- 

 thesis to assist them in such difficulties, he suggested 

 that the molecules of air were round hard globes whose 

 diameters were one-ninth of their distance from each 

 other. A simple answer to this is, that air could not, 

 in that case, be compressed to less than one-ninth of its 

 bulk. When the effect of this heating and cooling was 

 discovered by Laplace, it was found that the discrepancy 

 which Newton observed was exactly accounted for. For 

 a small amount of compression, the pressure of the air 

 increases at the same rate as that at which its volume 

 decreases, but for great changes this proportion ceases 

 to be correct, and it is found that for a given diminution 

 of volume, the pressure increases faster. Hence the 



Fig. 3. 



Although the propagation of sound in the air has been 

 carefully studied, and the whole of our knowledge of the 

 laws which govern it may be expressed by a few simple 

 formulae, it does not appear that any attempt has been 

 made to find the theoretical connection between the 

 limit of aiidibility and the magnitude of the disturbance 

 which produces the sound. There is reason to suppose 

 that a loud noise will travel rather further than a feebler 

 one, for the following reasons : sound is the result of 

 the transmission of a series of alternate conditions of 

 condensation and rarefaction of the air, just as the waves 

 on the surface of water are due to the up and down 

 movement of the liquid, or the waves of a cornfield to 

 the swaying of the separate ears. It is not the air 

 which is travelling, but the conditions of pressure. Now, 

 when air is compressed it becomes heated, and con- 

 versely, when it is rarified its temperature falls. The 

 changes, take place so rapidly that there is no time for 

 the heat to be conducted away. Newton was not aware 



mass of the air traversed by the sound will be greater in 

 those parts than in those where the pressure is small ; 

 that is, the condensed portions will travel faster than 

 the rarefied portions. A rather curious result follows, 

 namely, that at certain points the condensation will 

 have caught up the rarefaction, and the result will be 

 silence, but at a greater distance, the sound will increase 

 in strength, for the condensation can travel quite inde- 

 pendently of the rarefaction. 



It is often said that sound rises. Under ordinary cir- 

 cumstances this is not the case, for it is propagated 

 uniformly in all directions, the waves of condensation 

 and rarefaction being spherical, sphere within sphere, 

 like the coats of an onion ; the source of sound being 

 at the centre. The direction of the sound is also in a 

 straight line from the source. One reason why sounds 

 below the observer are often remarkably distinct, as, for ex- 

 ample to persons in a balloon, is that the earth beneath 

 acts as a reflector extended to an immense distance all 



