300 The National Geographic Magazine 



this observatory also some preliminary 

 effects were noted. 



At the Sitka observatory this disturb- 

 ance was also recorded by the magnetic 

 instruments from 8 124 to 8 130, Eastern 

 time, somewhat later than the preliminary 

 tremors recorded on the seismograph at 

 the same observatory. 



It is of interest to point out that the 

 times recorded for the principal disturb- 

 ances at the latter two places, Baldwin 

 and Sitka, which happen to be nearly the 

 same distances from San Francisco, are 

 about the same. Furthermore, it is seen 

 that the disturbance at Cheltenham (see 

 Table II) begins at a proportionate in- 

 terval later. 



It is to be noticed that in each of the 

 three cases the disturbance on the mag- 

 netograph occurs at about the same time 

 that the greatest motion is being recorded 

 on the seismograph. 



The self-recording magnetic instru- 

 ments, or magnetographs, on which the 

 above-described effects were recorded, 

 consist of very small magnets about one 

 inch long and a third of an inch wide by 

 one-sixtieth of an inch thick, suspended 

 by a fine quartz fiber and carrying a tiny 

 mirror. The record is obtained by re- 

 flecting a spot of light from the mirror to 

 a sheet of sensitive photographic paper 

 wound on a cylinder, the latter being 

 revolved uniformly by means of a clock- 

 work. By this means a continuous record 

 of the movements of the magnets is ob- 

 tained on the paper. Two of these mag- 

 nets, each mounted at some distance from 

 the other, record respectively the declina- 

 tion or direction and the horizontal com- 

 ponent of the earth's magnetic intensity. 

 A third magnet, somewhat larger, 

 mounted on sharp points and moving in a 

 vertical plane, records the vertical inten- 

 sity of the earth's magnetic force. 



The question whether the earthquake 

 disturbs the magnets in a purely me- 



chanical way or by its action on the 

 earth's magnetism is by no means settled. 

 In fact it is only recently that this phe- 

 nomenon is being systematically studied. 

 Up to the present the results are con- 

 tradictory. At times the magnetic dis- 

 turbance is simultaneous with or actually 

 precedes the preliminary tremors. In 

 other cases, like the present one, it ac- 

 companies the principal portion of the 

 disturbance. In some cases of large 

 earthquakes no magnetic effect can be de- 

 tected, and in a few other cases, notably 

 March 21, 1904 (New England earth- 

 quake), the shock was recorded at Chel- 

 tenham by the magnetic instruments, but 

 was not recorded by the seismographs at 

 Toronto, Baltimore, or Washington. 



Table I 



SAN FRANCISCO EARTHQUAKE OE APRIL l8, 

 I906, AS RECORDED BY THE SEISMOGRAPHS 

 OF THE U. S. COAST AND GEODETIC SURVEY 

 AT SITKA, ALASKA, AND CHELTENHAM, MARY- 

 LAND 



(Eastern or 75th meridian time) 



Sitka, 

 Alaska. 



Chelten- 

 ham, 

 Maryland. 



h. m. s. 



Preliminary tremors be- 

 gan , 81656 8 19 24 



Principal disturbance at. 8 20 48 8 30 13 



Maximum disturbance, 



about 823 8 35 



Disturbance ended about 10 57 12 00 



Table II 



AS RECORDED ON THE MAGNETOGRAPHS 



(Eastern or 75th meridian time) 



h. m. h. tn. 



Sitka Alaska 8 24 to 8 30 a. m. 



Baldwin, Kansas . 8 24 to 8 31a.m. 



Cheltenham, Maryland. . 8 30 to 9 04 a. m. 



U. S. Coast and Geodetic Survey, 

 April 20 , 1906. 



