370 REPORT OF THE CALIFORNIA EARTHQUAKE COMMISSION. 
marigram at the Golden Gate. The marigram near Fort Point, for April 18, 1906, 
shows (fig. 67) a depression of the water-level in the Golden Gate at the time of the 
earthquake, or rather a little subsequent to that event. The amount of the depression 
was slightly in excess of 4 inches. The mari- 
gram shows a blurring of the pencil mark from 
the direct action of the earthquake agitation, 
and this bearing serves to give approximately % 
the time of the shock. It shows that the run- ( 
ning clock of the gage was probably too slow, Ay 
and that the depression of the water-surface 
did not begin instantaneously, but followed after % 
an interval which may have been from 9 to 10 
minutes. Before the shock the gage had had a £ 
small vertical movement, ascribed by the officers 7,8 
of the Coast and Geodetic Survey to an im- Y 
perfect oscillation across the Golden Gate. This | <~ Blurred b 
minor vertical movement continued during the =~ earthquake, 
drop in the level of the water after the shock. if 
The time for the lowering of the water was 
9 minutes, as near as can be read from the 
marigram. It immediately began to recover, 
and the record shows that the water level rose 
without minor oscillations, to the normal level 
within 7 minutes, the total interruption in the 
normal marigram curve due to this depression 
being 16 minutes. After full recovery to normal 
level, the depression was not followed by a com- 
plementary rise of the water-surface, and in 
this sense the movement was not periodic. ‘The 
minor oscillations referred to above ceased when 
the maximum depression was reached, and do 
not appear in their characteristic forms on the 
marigram curve for some hours after. They 
were replaced, however, after 6 o’clock, by 2 or 
3 oscillations having a period of about 40 to 45 
minutes and an amplitude of 1 to 2 inches. 
These probably correspond to oscillations in San 
Francisco Bay. ) 
The Tidal Division of the Coast and Geodetic 
Survey very kindly computed the time which 
would be required for a wave generated at the 
fault-line on the bottom of the Gulf to reach 
Fort Point, and found that it would require 9 
minutes, on the assumption that Fort Point is 
6 statute miles distant from the fault-trace in 
a direction normal to it. The position of the 
gage is, however, 1.3 miles distant from Fort 
Point within the Golden Gate, so that the time 
necessary for the wave to reach the gage would 
be somewhat longer. Now the time at which 
the gage began to fall is between 9 and 10 minutes 

a] 
4sSvog "Sf 94} JO SpIOSaT WHOA “QOGT ‘ST [dy “WV qT OF NW Vyns 
your T ‘Tet 
qnoqe Woy ‘oosPoURIg Ug ‘yUIOg Jog ye WeRIsE —*L9 9 
YUL T ‘TBoyaea :satvog *Laaang dyeposs) puv 
“mmoy T 
-wo0zItoy £400F T 
