ON SEISMOLOGICAI- INVESTIGATION. 



71 



velocities indicated in kilometres per second are from the isoseist of the 

 place indicated by its initial letter to the place beneath which it is written. 

 These latter places in the top line are also indicated by their initial letters 

 The letter O refers to a velocity measured between an origin and the 

 place named in the upper line 



As it is difficult to picture the directions of great circle paths outside 

 equatorial regions, these are shown with the above velocities and the 

 earthquakes to which they refer in the accompanying map (Plate II.). A 

 line not referred to in the above table is that for earthquakes numbered 

 36, 83, 100, and 119, which originated in Japan and travelled to the Isle of 

 Wight with an average velocity of 2 9 kms. per second. 



An inspection of the map (Plate I.) shows that the apparent velocities 

 over long paths are greater than those over short paths. Velocities across 

 the Pacific are apparently lower than those across the Atlantic, and those 

 across Northern Asia to Shide are lower than those across North 

 America to Shide. Between Mexico and Victoria along the stiike of 

 the chief North American anticline the velocity of transmission is the 

 same as that between Mexico and Toronto. Along paths terminating 

 at the Cape of Good Hope the rate of transmission has been high, whilst 

 on those terminating at Mauritius, excepting that referring to the long 

 path for earthquake 250, the velocity of propagation appears to be low. 



There does not appear to be any indication that direction of propaga- 

 tion is related to speed, and although earthquake 381 was larger than 250, 

 and 333 and 33S were larger than 337, we do not .seem to have any 

 definite evidence that velocity of propagation is connected with the 

 intensity of the initial disturbance. 



Taking the results of this investigation generally, we are hardly in a 

 position as yet to draw definite conclusions, and must wait for further 

 observations. 



10. Earthquake Echoes. 



In the British Association Report for 1899, p. 227, I drew attention 

 to the fact that in seismograms where a group of large vibrations corre- 

 sponding to a shock or shocks at an origin is pronounced, this is frequently 

 succeeded by a set of fairly similar movements. These latter impulses, 

 which may be repeated, but with decreasing intensity, many times, I pro- 

 visionally called earthquake echoes. Although earthquake repetitions (see 

 pp. 66-69) which succeed their primaries at very irregular intervals may 

 possibly be antipodean reflections of mass waves, they must not be con- 

 founded with the so-called echoes which succeed the maxima movements 

 at fairly regular intervals. 



The following table gives time intervals in minutes between a number 



