528 



GEOLOGY. 



projecting them backwards to their underground crossings, a first 

 approximation to the location of the focus is reached (the Hues EF', 

 Fig. 446). From the nature of the case, the observations of the angles 

 of emergence cannot be very accurate, but an effort is made to limit 

 the error by making the number of observations great. 



Two systematic corrections are to be applied to all such estimates, 

 the one for varying elasticity and density, and the other for varying 

 continuity. Both reduce the estimated depth. In making the correction 

 for varying elasticity, it must be noted that the velocity of vibrations 

 varies directly as the square root of the elasticity, and inversely as the 

 square root of the density. The velocity is also accelerated by increase 

 of temperature. The elasticity, temperature, and density all increase 

 with depth. Theoretically, the increase of velocity due to the in- 





Fig. 446. — Diagram illustrating by closed curves the different rates of propagation of 

 seismic tremors from a focus F, and, by lines normal to these, the changing directions 

 of propagation of the wave-front. It will be seen that the paths of propagation curve 

 upwards in approaching the surface. If the lines of emergence, as at E and E, be 

 projected backwards, as to F\ the points of crossing will be below the true focus. 



creasing elasticity and temperature of increasing depths, overbalances 

 the retardation due to increasing density, and recent observations on the 

 transmission of seismic waves through deep chords of the earth have 

 confirmed this conclusion. The path of the vibration will, therefore, 

 be curved toward the surface, as pointed out by Schmidt and illus- 

 trated in Fig. 446, taken from his discussion.' From this it is clear 



» Nature, October 24, 1895. 



