114 REPORT OF THE CALIFORNIA EARTHQUAKE COMMISSION. 
TaBLE 6.— Duration of the Principal Part as Recorded by Milne Instruments. 









STATION. Distance. | DurATION. STATION. DISTANCE. DouRATION. 
fe} ° 
Victoria 10.4 Kew. a oe 77.6 25 or 36 
Toronto 32.9 Inky take eee 80.8 28 or 60 
Honolulu 34.6 Coimbra ts. 81.4 24 
Baltimore . Sel San Fernando. . 85.3 54 
Paisley . 72.6 Calcutta. . . 1 WARE 31 
Edinburgh 73.0 Bombay i-semee 121.2 26 
Bidston . 74.8 ‘Perthicel: oe aco 132.4 60 

It is quite evident that no conclusion regarding the variation in duration of the prin- 
cipal part at different distances from the origin can be drawn from such data. 
But, altho we may be unable to recognize a progressive change in the duration of the 
principal part, nevertheless it is quite certain that all seismographs register a strong 
motion lasting much longer than the original shock. What has been called the violent 
shock, and which alone could have affected distant seismographs, did not last more than 
40 or 50 seconds; whereas the recorded principal part certainly lasted many minutes and 
in some cases an hour. This may be in part due to the synchronism of the periods of 
the waves and the instruments, but it can not be entirely explained in this way and it 
must be lookt upon as not yet understood. 
We have a little information regarding the prevalent direction of motion during the 
principal part. At Ottawa, Cheltenham, and Albany the longitudinal waves retained 
their intensity for a longer time than the transverse, tho we can not say which attained 
the greater maximum. At Rocca di Papa the longitudinal waves attained the greater 
maximum, but the durations of the two were about the same. At Messina the longitu- 
dinal waves were stronger and lasted longer than the transverse. On the other hand, 
the transverse waves lasted longer at Florence, and they had a greater maximum at Cag- 
giano. The observations are very meager and very inconsistent ; evidently more careful 
observations must be made to show to what extent the longitudinal and transverse waves 
are characteristic of different parts of the seismogram, how far this quality is different 
at different stations, to what variations it is subject, and what are their causes. 
The long tail portion of the seismogram is still a riddle; and altho we can hardly help 
considering it as in some way due to waves following different paths and to reflections, 
we shall see further on (page 124) that simple reflections will not explain it. One may 
easily be misled in attempting to correlate certain movements on different seismograms ; 
for instance, the last marked broadening of the trace of the Paisley, Edinburgh, and Bid- 
ston seismograms (sheet No. 1), occurring a few minutes before 14" 30”, is so similar that 
one would naturally suppose that they represent a special group of progressive waves; 
on determining the times of occurrence we find for its maximum, 14° 14.7, 145 20™ and 
145 26.1™ at the three stations respectively; the difference in time at Paisley and Bidston 
is 11.7 minutes and the difference in distance 252 km.; therefore the velocity of propaga- 
tion would be 22 km./min. But the time for them to reach Paisley from the focus, a 
distance of 8,060 km. would be 62.2 minutes, requiring a velocity of 130 km./min. 
These values are so different that we must regard this broadening of the trace as due to 
some accidental synchronism of periods at the three stations, and not to an objective 
characteristic of the disturbance itself. There are many difficulties in understanding the 
characteristics of the seismogram which have not been overcome; and it is not likely 
that we shall have a complete explanation of it until a large number of heavily damped 
seismographs are installed, whose records will correspond closely with the actual move- 
ments of the earth, and will not be materially affected by the peculiarities of the instru- 
ment itself. 
