310 REPORT — 1851. 



times that in air ; but the velocity that would be deduced from the ittean 

 modulus of elasticity of cast iron gives 12'2 times that in air. 



Much more recently MM. Wertheim and Breguet have made some most 

 interesting experiments upon the rate of sound in malleable iron, namely, 

 in the wires of the electric telegraph on the Versailles Railway, with a range 

 of no less than 4050 metres. The method of experiment is remarkable for 

 its simplicity and accuracy. It consisted in placing an observer and assistant 

 at either end of the range, provided each with a chronometer, both of which 

 were accurately rated together, and also with a delicate stop-watch, by which 

 time could be noted to ygth of a second. At a predetermined moment the 

 observer struck a sharp blow of a hammer on the wire at one end ; the as- 

 sistant at the other noted the moment of arrival of the sound. This process 

 was repeated back from the other end, and so on. Thus the personal errors, 

 and those of the chronometers, became eliminated, and the length of range 

 practically doubled. 



Their results are, that sound travelled linearly in the wrought iron wire at 

 the rate of only 3485 metres=11433"936 feet per second. 



They then stretched two metres of the wire as a musical chord, and by 

 Chladni's method determined the velocity of transversal vibration to be 4634 

 inetres= 15233*721 feet per second. Hence the experimental velocity of the 

 sound-wave propagated longitudinally is much less than that of free or trans- 

 versal vibration, and still less than that deducible from the mean modulus of 

 wrought iron. The latter, however, would be abundantly uncertain in appli- 

 cation, as the state of annealing of the wires, or their density, &c., are not 

 given. (Comp. Rend. t. xxxii. p. 293.) 



Should this indicate a general fact, as seems highly probable, then we have 

 a theoretic cause, whose action coincides with that of want of homogeneity, 

 in producing retardation of the elastic waves of our experiments. 



In applying our results to any speculation as to the actual velocity of earth- 

 quake waves, the effects of the direction and depth from which they are prd* 

 pagated from the origin must be held in view. 



As we descend deeper in every rocky formation of the earth's crust, we 

 may expect to find the material not only denser (greater sp. gr.), but less 

 and less shattered ; and hence the velocity of the earth-wave will be greater 

 in the same formation in proportion as its path lies deeper. When an emergent 

 earth-wave reaches the surface, then its actual velocity, if transmitted from 

 a great depth and at no very large angle from the vertical, may be very much 

 greater than would be due to the surface rock ; for theory at present does 

 not admit of our determining how far into a second medium of lower elasti- 

 city the wave transmitted from another more elastic medium will penetrate 

 the former with a velocity due to the first medium, or in what way and to what 

 extent the velocity of the wave will be affected with reference to the distance 

 traversed beyond the junction of the media ; still less will theory yet enable 

 us to predict the effects of gradual change of elastic coefficient of the medium 

 through which the wave passes. 



Again, if we assume the formation homogeneous in depth, and vibration 

 propagated linearly in all directions alike from a centre of impulse; owing 

 to the compression of the mass from the effects of gravitation thus acting at 

 great depths, it is possible that the rate of wave transit may vary at different 

 levels as some function of the depth, bearing an analogy to the case of a liquid 

 confined in a tube, which Wertheim finds to transmit sound at a different rate 

 (its transversal vibrations being confined) from a boundless mass of the same 

 liquid. (Comp. Rend. t. xxviii. p. 151.) 



The discussion of many earthquake narratives in the progress of the cata- 



