212: ; REPORT—1883. 
The cause of these movements is unknown. Rossi makes the sugges-- 
tion that they may be due to a variation in volcanic activity beneath the’ 
surface of the ground, which increases with a barometrical depression. 
They may, however, be attributed to a complexity of causes acting on 
the surface of our earth, At the time of a high wind the movements of 
houses and trees may set the surface of a considerable area into a state of: 
tremor. When they are observed without a wind they may occasionally 
be due to an irregularity in the increase or decrease in atmospheric 
pressure. In a typhoon I have observed the needle of an Grdinany, aneroid 
to move backward and forward through a range of from +1, to 7%, of 
an inch. This motion was irregular, having a period of from one to ten 
seconds. Small but rapidly succeeding variations in atmospheric pres- 
sure, even very much smaller than those just quoted, indicate that tbe 
surface of the ground is being subjected to and relieved from stresses, in 
every probability, competent to produce the oscillations observed in the 
pendulum. 
A second set of observations has been recording the motions of the 
bubbles of two delicate levels placed beneath glass covers on the same 
column with the tromometer. One of these is placed N. and S. and the 
other E. and W. The variation in temperature in the room seldom ex- 
ceeds 1° or 2° F', per day. These levels have shown continual movements. 
At present the N. and S. level bas a diurnal backward and forward 
motion of about three divisions. One division equals about 2” of are. 
As an example of the larger movements which have been recorded, I may 
state that the bubble of the N. and 8. level moved, from March 25 to 
May 4, through twenty-nine divisions. The direction of the deflection of 
the tromometer pendulum has a general correspondence with these larger 
movements. A curious phenomenon which bas been observed in the 
levels is that accompanying a barometrical depression—there is a slight 
surging in the bubbles. The surge, which has an amplitude of from ‘25 
to ‘5 of a division isirregular, having a period of from 1 to 5 or 6 seconds. 
This motion, inasmuch as it is different from the effects produced by 
alterations in temperature, and as it accords with the microseismic move- 
ments of the tromometer, I am inclined to attribute to a true earth- 
pulsation. 
Another phenomenon indicative of the existence of earth- pulsations 
—by which I mean motions which may have an amplitude equal to that of 
an earthquake, but which are not perceived on account of the slowness 
of their period—is the slow surge-like motion in a level, which continues 
for fully three or four minutes after all sensible motion of an earthquake 
has disappeared. This surging, as it dies out, closely accords with the 
surge observed at the time of a barometrical depression. 
This last observation is supplementary to observations made on an 
earthquake with a seismograph. . The records from a seismograph show 
that a moderately strong disturbance sometimes commences as a series of 
tremors with a frequency of from 4 to 6 per second. These movements 
are so small in amplitude that, unless an observer is favourably situated, 
they are passed by unnoticed. While they continue, however, I have 
heard pheasants scream, and it has been noticed that frogs cease their 
croaking. Immediately after the tremors we get the shock of the earth- 
quake, some of the vibrations of which have occasionally been performed 
so rapidly that I have failed to measure their duration. It is not unlikely 
that this portion of an earthquake may take place so suddenly that rocky 
strata in the immediate vicinity of the origin have not time for elastic 
