The San Francisco Earthquake 



297 



ton appear at a on the record. These are 

 the so-called preliminary tremors, which, 

 either because they traveled faster or 

 along a shorter path — perhaps for both 

 reasons — -reached Washington not less 

 than six minutes earlier than the stronger 

 motion, the record of which begins at b. 

 About seven minutes later the needle was 

 deflected off the record sheet by still more 

 violent motion, but returned again at 

 8:35, and thereafter recorded vibrations 

 of continually diminishing force for about 

 four hours. 



SPEED OF PROPAGATION 



In the study of earth vibrations several 

 distinct classes of waves are recognized, 

 such as longitudinal, transverse, surface 

 waves, distortional waves, etc. These ap- 

 pear to travel along different paths and 

 at different velocities, and a full state- 

 ment of the question of the speed of 

 propagation is very complex. In a gen- 

 eral way, however, it may be said that the 

 so-called preliminary tremors travel by 

 the shortest path between points — that is, 

 through the earth along the chord rather 

 than along the surface of the sphere. In 

 the present case the great circle distance 

 from San Francisco to Washington is 

 about 2,435 miles, whereas the distance 

 through the crust is about 40 miles 

 shorter, and the straight-line path cuts 

 below the surface of the earth about 186 

 miles at its deepest point. At still greater 

 distances from San Francisco than Wash- 

 ington it is plain that the straight-line 

 path cuts still deeper and deeper below 

 the surface, and is also proportionately 

 shorter and shorter than distances on the 

 surface. Furthermore, because of its 

 greater density and the enormous superin- 

 cumbent pressure, the elastic properties 

 of the deep-lying substance of the earth 

 seems to propagate vibrations with higher 

 and higher velocities the deeper the path. 

 These considerations lead to the conclu- 

 sion that if earthquake vibrations follow 

 the path of the chord the speed of propa- 

 gation should not be constant for all dis- 

 tances from the origin, but should be 

 Greater the greater the distance from the 



origin. This has generally been found to 

 be true in the case of the preliminary 

 tremors, and will doubtless be shown in 

 this earthquake when accurate reports 

 from numerous stations are examined. 

 From San Francisco to Washington the 

 speed along the chord is found to be 5.4 

 miles (8.7 kilometers) per second. This 

 is based on Professor Davidson's time at 

 San Francisco, viz, 5:12 a. m. This re- 

 sult is perhaps a trifle faster than we 

 might expect. 



The strong waves do not seem to fol- 

 low the path of the chord, but rather 

 travel along the surface at a slower rate, 

 which is nearly constant for all distances. 

 In the present case the velocity is 3.1 

 miles per second for the first strong 

 waves, or as low as 2.2 miles per second 

 for the maximum waves. Both of these 

 speeds, however, are a little high, per- 

 haps. 



AMPLITUDE AND PERIOD OE THE WAVES 



The recording needle of the seismo- 

 graph traces a ten-fold magnification of 

 the movement of the earth. On this 

 basis the actual maximum displacement 

 of the ground at Washington was nearly 

 one-half inch each side of its position of 

 rest. This may seem an astonishingly 

 large movement, especially in view of the 

 fact that no one felt any motion whatso- 

 ever. The explanation, however, is found 

 in the fact that the motion is very slow 

 and without shock, so that buildings and 

 objects are moved to and fro without the 

 slightest strain or injury. 



THE SEISMOGRAPH 



The record at Washington was made 

 by the aid of the instrument shown on 

 page 293. The massive lead weight C is 

 suspended by the wires W and the hori- 

 zontal strut so as to swing very freely 

 about a nearly vertical axis from D to B. 

 During an earthquake this heavy mass 

 remains almost completely at rest with 

 respect to lateral motions, and thus. pro- 

 vides a steady point in reference to which 

 the motion of the ground can be meas- 

 ured and registered. For this purpose a 



