PERIODS AND AMPLITUDES. 137 
At Mount Hamilton (1.16° or 129 km. from the origin and 35 km. from the fault) the 
3 component Ewing instrument indicated amplitudes, both in the north-south and 
east-west directions, greater than 40 mm. 
At Carson City (2.62° or 291 km.) the horizontal amplitude was about 11 mm. in all 
directions. 
Beyond the megaseismic district.— We have collected in table 19 the periods and the 
greatest earth-amplitudes at all the stations for which we have sufficient data to deter- 
mine these quantities. In a few cases they are taken directly from published reports. 
At many stations there was so close a correspondence between the period of the vibra- 
tions and that of the pendulum during the very strong motion that it was impossible to 
make any determination of the earth-amplitude. 
It will be seen that the periods of vibration during the regular waves were, in general, 
not very far from 30 seconds, tho in a few cases they were 10 or 12 seconds less, and in 
a few 10 or 20 seconds more. During the principal part the periods were principally 
between 17 and 25 seconds. 
Where we have determinations of the earth-amplitude during both the regular waves 
and principal part at the same station, the former seems to be somewhat the larger, altho 
the instrumental record on the seismogram is almost always larger during the principal 
part. This is due to the variations in the magnifying power of the instrument on account 
of difference in periods. 
Altho the amplitudes do not diminish regularly with the distance from the origin, never- 
theless with the exception of a few abnormal values, which are not understood, there is in 
general a reduction of amplitude with the distance. In the megaseismic region we found 
that these amplitudes were 50 mm. or more; at distances of 30° to 50° they have dimin- 
ished to about 5 mm., and we must go as far as 100° or so to find amplitudes less than 
1mm. We see, therefore, that the great world-shaking earthquakes cause movements 
of the earth at great distances which are by no means inconsiderable, and the only reason 
why they are not felt is that the period is very long, and, therefore, the movement too 
slow to make them evident to our senses. 
In attempting to determine the depth of the fault (page 13) we were led to assume 
that the energy is sent out from the fault-plane proportionally to the cosine of the angle 
between the direction of propagation and the normal. Altho this will probably hold 
approximately in the neighborhood of the fault, it does not hold at a distance, where the 
distribution of the energy, so far as we can tell from the altogether unsatisfactory deter- 
minations that could be made, is entirely independent of the direction from the origin. 
For instance, Sitka and Tacubaya, whose directions make angles 16° and 29°, respectively, 
with the direction of the fault, have apparently instrumental amplitudes similar to those 
of the stations in the eastern part of North America, whose direction is nearly at right 
angles to the fault-plane. Pilar, Argentina, and the Cape of Good Hope, whose directions 
make angles of 12° and 51° with the fault, gave very small records, whereas Mauritius 
(nearly 35°) gave a much larger record. Calcutta, Kodaikanal, and Bombay (5° to 16°) 
also gave much larger records. 
In looking over the table of earth-amplitudes, to compare the results between stations 
at about the same distances from the origin, but in different directions, we find the 
irregularities so great that no satisfactory conclusions can be drawn. We notice, how- 
ever, that Zi-ka-wei had about the same amplitude during the principal part as Carlo- 
forte and Sarajevo; and that the amplitudes at Sofia, Catania, and Manila do not differ 
greatly during the regular waves; but these comparisons carry very little conviction with 
them, because of the great variations between the amplitudes at various European sta- 
tions, which do not differ greatly in their distances from the focus nor in their directions 
from it. 
