THE PROPAGATION OF THE DISTURBANCE. 131 
The hodograph of the “large waves” of Professor Milne in the earlier observations does 
not refer to the same surface waves as those which are here tabulated as regular waves, 
but to the time of maximum displacements on the seismograms. The position of the maxi- 
mum is very largely dependent upon the proper period of the recording pendulum, and 
the instruments whose records we have of the California earthquake differed so greatly 
in this respect that it is not possible to identify as a maximum any characteristic part of 
the disturbance, except for a limited number of seismograms. 
In his very interesting memoirs on the propagation of earthquake motion, Prof. G. B. 
Rizzo gives hodographs of the two Calabrian earthquakes of September 8, 1905, and 
October 23, 1907. The former was a severe earthquake and was recorded all over the 
world. ‘The latter was much smaller and satisfactory observations were only obtained up 
to distances of about 22°. The hodographs of the first and second preliminary tremors 
agree very well with my curves, except about 20° and in the immediate neighborhood of 
the origin, where Professor Rizzo has made his curve convex upward to represent the 
assumed changes in surface velocity; and he has measured his times from the estimated 
time of arrival of the disturbance at the epicenter, whereas I have measured time from the 
actual time of occurrence of the shock at the focus, and have assumed the velocity of 
7.2 km. per second for short distances from it. 
Professor Omori, in his very complete account of the seismograph records of the Kangra 
earthquake of April 4, 1905,’ gives the hodographs of the first and second preliminary 
tremors and a later phase of the principal part; the latter, however, does not correspond 
to the regular waves which I have recorded. The hodographs of the first two phases 
correspond fairly well with those of the California earthquake?’ up to distances of about 60° 
for the first preliminary tremors and 90° for the second preliminary tremors; but beyond 
they diverge greatly. It is rather curious that the observations of the Indian earthquake 
are most numerous between 37° and 60°, in which interval there is but one observation of 
the California earthquake; whereas, between 70° and 100°, where the great majority 
of the observations of the California earthquake lie, there are but four very unsatisfactory 
observations of the Indian earthquake. The cause of the disagreement between the 
observations at the greater distances is very evident. All the observations of the Indian 
earthquakes at distances greater than 60° are made with instruments of very low magnify- 
ing power, and it is hardly possible that the true beginning of the disturbance has been 
recorded. With regard to the second preliminary tremors it is a question of the interpre- 
tation of the seismograms. Of the four observations which Professor Omori uses beyond 
90° three are from Bosch-Omori 10 kg. instruments, and I think it quite impossible from 
an examination of their seismograms to determine where the second preliminary tremors 
really began. The other record, at Wellington, was made by a Milne pendulum, and the 
time I take to mark the arrival of the second preliminary tremors is nearly 10 minutes 
earlier than that taken by Professor Omori, and is between 2 and 3 minutes later than my 
curve would lead us to expect. The record at Christchurch, 0.7° nearer the origin, is 2.5 
minutes earlier. The Milne seismograms from Victoria, Toronto, Baltimore, and Christ- 
church (from 97.7° to 115°) are not used by Professor Omori in making his hodograph of 
the second preliminary tremors, tho he reproduces them among his plates. As I read 
them, the times of arrival of the second preliminary tremor are in fair agreement with 
my curves and are from 8 to 12 minutes earlier than the times adopted by Professor 
Omori for similar distances. He is thus led to nearly linear hodographs of the first and 
second preliminary tremors, and consequently to a linear relation between the distance 
of an earthquake origin and the duration of the first preliminary tremors. 
1 Report on the Great Indian Earthquake of 1905. Pub. Earthquake Investigation Committee in 
Foreign Languages, Nos. 23 and 24, F 
2 Professor Omori also gives the velocities obtained from the California earthquake in the same report, 
but he has taken the time of the shock a half minute too early. 
