Seismological Investigations. 101 



thrown upon their ultimate origin and the magnitude of the 

 energy involved, while the study of velocities of transmission, 

 co-seismal and iso-seismal lines, may afford valuable informa- 

 tion concerning the nature and condition of the rocks within 

 the disturbed area. . Admitting the greater importance of a 

 knowledge of these facts it follows that time measurements 

 should first receive attention, and that the seismic chrono- 

 graph is the instrument to be used. 



The seismograph or seismometer is designed primarily for 

 the purpose of recording or measuring the actual motion of 

 that part of the earth to which the instrument is attached 

 during the transit of one or many waves. In reality this 

 motion is extremely complex. Undoubtedly the emerging 

 wave is modified very greatly in its character by the lack of 

 homogeneity in the material through which it last travels, as 

 well as by the fact that this material differs immensely in 

 elasticity and density from that in which it has in the main 

 existed. For this reason it is believed that however accurately 

 the motion may be resolved into three components at right 

 angles to each other by a perfectly operating seismograph, but 

 little information would be afforded as to the position of the 

 origin, or the amplitude of vibration and amount of accelera- 

 tion of any point in the earth, other than that at which the 

 instrument is located. It is clear, however, that while these 

 considerations may seriously affect the integrity of the record 

 of a seismograph, they will have little influence upon the 

 actual transmissive time, that is to say, while the character of a 

 wave may be greatly altered upon emergence into a non- compact- 

 ed, non-homogeneous material, the time of its arrival at a given 

 point cannot be greatly altered, even if its velocity in this material 

 is much less or greater than the mean, for the reason that it is 

 subjected to this modifying influence for a comparatively brief 

 period. It is true that the seismograph, in addition to register- 

 ing the motion of the earth particle, may and generally does 

 record the epoch of the passage of a wave, and it affords the 

 advantage of distinguishing one wave from another. If only 

 very short distances are used for the determination of veloci- 

 ties this would be of decided value, were it not that experi- 

 ence* seems to prove that what is the maximum wave at one 

 point may not be the maximum at other points very near, so 

 that it is by no means certain that a particular wave can be 

 identified at different stations, even if they are not widely 

 separated. These considerations, together with the very great 

 expense of seismographic equipment and the greater difficulty 

 of maintaining them in constant working order in a country 

 where earthquakes are infrequent, compel the admission that 

 * Milne, in Trans. Seis. Soc. of Japan, vol. x. 



