112 REPORT OF THE CALIFORNIA EARTHQUAKE COMMISSION. 
of vibrations from various parts of the fault-plane make it impossible to recognize the 
second phase; but at Sitka (20.72° or 2,302 km.), and at more distant stations, the 
second phase is distinct. So far, therefore, as the observations of the California earth- 
quake are concerned, there is no reason to believe that the first two phases are not dis- 
tinct from their starting-point; and the reason this has not been recognized heretofore 
may be entirely due to the small time-scale of the instruments. . 
THE DIRECTION OF MOTION. 
Let us see how far the observed directions of motion support the above explanation 
of the elongation of the first two phases. The duplex seismographs of Berkeley and 
Mount Hamilton indicate the direction of the beginning of the motion; they show that 
the first movement of the ground at these stations was directed away from the origin. 
The extent of the fault-surface soon caused waves to come from many directions, so that 
the recorded movement became confused almost immediately; but at Mount Hamilton 
there were two longitudinal vibrations before other waves materially interfered with their 
direction. The seismogram of the three-component Ewing instrument shows, when 
we consider the arrangement of the recording pens, that the first and second preliminary 
tremors began there with a movement southeast and northwest, that is, along the 
direction of propagation. ‘These two were the only stations near the earthquake’s origin 
which yielded definite information regarding the direction of motion at the beginning of 
the shock. And of all the records at distant observatories there are comparatively few 
which throw light on this subject; because only a very few instruments were so 
oriented as to record separately the vibrations parallel with, and at right angles to, 
the course of the waves. The stations in the eastern part of the United States were 
well situated for this purpose, as the waves were moving almost directly eastward 
when they past them. Ottawa and Cheltenham each recorded the longitudinal 
waves (east component) about 13 seconds before the transverse, and the longitudinal 
waves also were somewhat stronger during the first preliminary tremors. In the second 
group, transverse waves (north component) were recorded at Cheltenham 9 seconds 
earlier than the longitudinal; and they seem very slightly stronger. The northern com- 
ponent of the second group in the Ottawa seismogram overlaps other parts and can not 
be clearly read; but it seems to be somewhat stronger than the eastern component. The 
Albany record does not yield definite results, and the other stations in this neighborhood 
only recorded one component of the motion. 
The waves arrived at the majority of the European observatories in a direction making 
angles between 30° and 40° with the meridian; and as by far the larger number of the 
instruments recorded either north-south or east-west motion, they would be affected about 
equally and would not distinguish between longitudinal and transverse waves. A few 
instruments, however, were oriented so as to make the distinction. The triple Ehlert 
instrument at Uccle began to record at the same moment with all three components, but 
the longitudinal waves (N. 60° W.) were stronger during the first preliminary tremors, and 
the transverse (N. 60° E.) during the second preliminary tremors. At Kremsmiinster 
the longitudinal waves (distributed between the two components, N. 18° W. and N. 73° 
W.) seem stronger during the first preliminary tremors, and the transverse (N. 47° E.) 
during the second preliminary tremors. At Rocca di Papa the longitudinal vibrations 
(NW.) in the second preliminary tremors were registered 42 seconds before the trans- 
verse (NE.), according to Professor Agamennone’s reading of the original record. At 
Messina, the transverse (NE.) vibrations in the second preliminary tremors were somewhat 
stronger than the longitudinal. 
