TIME OF EARTHQUAKE, RATE OF PUOl'AGATION. 39 
Kodaikanal Observatory. 
The following are the particulars supplied by the Director of 
the Kodaikanal Observatory : — 
Pielimi- 
uary 
Tremoi-s. 
Large 
Waves. 
Ma.xi- 
nmm. 
1 
End. 
n\u- 
iition. 
Ami)U- 
tude. 
Ul'.M.KRKS. 
If. M. 
H. M. 
H. 
11. M. 
II. M. 
mill. 
1918, .Tuly 8 
10 2G-4 
10 31-5 
12 2-8 
1 3fi-4 
140 
Very dostriic- 
tivo in Assam. 
Commencement 
of laigc waves 
abrnpt (not 
gradual). No 
p r e 1 i minary 
tremors. 
(Orcenwlch mean times). 
The seismogram is shown on Plate G, Fig, 2. 
Simla Observatory. 
The following information was given by Dr. G. T. Walker, 
C.S.I. , F.R.S., Director-General of Observatories in India : — 
" The earthquake was recorded on our two Omori-Ewing horizontal pen- 
dulum seismographs with masses of 50 kg. and natural periods of 45 seconds. 
There was also a shght local shock easily perceptible without instruments 
at two or three minutes to sixteen hours I.S.T. 
2. The instrumental records show the first preliminary tremors starting 
at 15 hours 55 minutes 20 seconds with periods of about 1^ seconds. At 
15 hours 57 minutes 45 seconds stronger oscillations are superposed, which, 
I guess, indicate yielding on some fault probably witliin two or three 
hundred miles of Simla. Then large oscillations began at 15 hours 59 
minutes 5 seconds, both the weights s\\inging backwards and forwards 
between the steps for about 2j minutes : after this the movement gradually 
subsided, lasting until 17 hours 10 minutes. At first there is considerable 
temptation to regard the second preHminary vibrations as beginning at 15 
hours 57 minutes 45 seconds and the main shock at 15 hours 59 minutes 
5 seconds 5 for in general the amplitude of the oscillations iu the main 
shock is much larger than in the second preliminary vibrations. But this 
interpretation would, I consider, be incorrect. In the first place these 
times would indicate that the origin was only 520 miles distant, and in 
such cases previous records show that the main shock has a smaller amplitude 
