130 REPORT OF THE CALIFORNIA EARTHQUAKE COMMISSION. 
San Fernando at 135 59™, and probably are indicated by the strong motion a minute 
earlier; the major are waves arriving at Wellington at 15> 52™ appear at Coimbra at 
145 01™, and cause the strong motion at San Fernando at 145 04™. The direct waves 
at Wellington at 145 07" and 145 12.5" are due at San Fernando at 15> 41™ and 
155 50™ and are undoubtedly represented by the strong swelling about 155 43™; they are 
due at Coimbra at 155 43™ and 155 53™; these are weak parts of the curve, but prob- 
ably the swellings a few minutes earlier than these times represent the waves we 
are considering. 
We must conclude, from the foregoing survey, that altho the strong motion arriving by 
the major arc makes itself evident at some stations, perhaps on account of synchronism 
of its period and that of the recording instrument, at other stations it can not be detected. 
The small time scale of the Milne seismograms is much better adapted for identifying the 
major arc waves than the open time scale of other instruments. 
EQUALITY OF VELOCITIES ALONG DIFFERENT PATHS. 
As already pointed out, all the distant stations had instruments of low magnifying 
power and apparently were too late by various amounts in recording the shock; so we 
must confine our attention to stations less than 100° distant. On comparing the times of 
arrival of the various phases (given in table 7, page 116) at stations nearly equally distant, 
we can not find any differences, greater than the errors of observation, which might be 
dependent upon the direction of the station from the origin; and this applies to all three 
phases of the motion. ‘Thus, Honolulu receives the second preliminary tremors a little 
earlier than the observations at stations in the east of North America would lead us to 
expect (see hodograph, plate 2), but the first preliminary tremors arrive at the expected 
time. The paths to Honolulu and these stations are totally different, the first being under 
the Pacific and the other across the continent of North America, as shown in plate 1. 
The Japanese, on the one hand, and the British and Scandinavian stations, on the other, 
are about equally distant from the origin; the path to the former lies under the deep 
Pacific, that to the latter across North America, Greenland, and under the shallow North 
Sea; but we do not find a greater difference between the times of arrival at these two groups 
of stations than we do between the individual stations of the same group. 
Irkutsk and Jurjew are at practically the same distance from the origin; the path to 
Irkutsk passes under the Pacific, across Alaska and northeastern Asia; the path to Jurjew 
crosses North America and Greenland and continues under the North Sea; yet the times 
of arrival at the two stations are within a very few seconds of each other. 
We conclude, therefore, that the velocity of propagation is independent of the position 
of the projection of the path on the earth’s surface; or, at least, is too little affected by it 
to be detected by our observations. 
COMPARISON OF THE HODOGRAPHS OF THE CALIFORNIA EARTHQUAKE WITH 
OTHER OBSERVATIONS. 
When we compare the hodographs obtained from the California earthquake with those 
given by Professor Milne in 1902 ' and with those of Professor Oldham, 1900,” we find that 
our times of arrival of the first and second preliminary tremors are, for the greater part of 
the curves, about 2 minutes earlier. This appears.to be due to lack of accuracy in the 
earlier observations, and a glance at the earlier diagrams will show that the curves are 
drawn from observations differing greatly among themselves. 
1 Report Seis. Com. B. A. A. S., 1902. 
On the Propagation of Earthquake Motion to Great Distances. Phil. Trans. R. S., 1900-1901, 
vol. 194, pp. 135-174. 
