TRANSACTIONS OF SECTION A. 921 
‘change in the inclination of the surface of the ground relatively to the vertical. 
Movements of this class have been examined by d’Abbadie and Plantamour, and 
‘also by G. H. and H. Darwin, who have given the results of their observations to 
the British Association in two Reports on the Lunar Disturbance of Gravity 
(1882-3). Perhaps the most convenient name for these movements is earth-tiltings. 
They are measured by what may be called the equibrium method. A pendulum 
suspended in a viscous fluid is employed to show by its equilibrium position the 
true direction of the vertical, and that is compared with the direction of a line 
which is fixed relatively to the surface of the ground ; or, instead of a pendulum, 
a dish of mercury or a pair of spirit-levels are employed to define a truly horizontal 
surface, and the tilting of the earth’s surface relatively to that is observed. 
This method is practicable only when the displacements of the surface have so 
great a vertical amplitude in comparison with their horizontal wave-length that 
the slope of the wave is sensible; and further, only when the changes of slope 
occur slowly enough to put the inertia of the pendulum or fluid out of account. 
On the other hand, the inertia method is applicable only when the displacements 
have so short a period in comparison with their amplitude that the acceleration of 
the ground during the greater part of the motion is large relatively to the frictional 
resistance of the suspended mass. Between ordinary earthquakes and tremors on 
the one hand, capable of observation by the inertia method, and slow earth-tiltings 
on the other, capable of observation by the equilibrium method, it is at least 
possible that there may lie many movements not reducible to either type. For 
example, if successive upheaval and subsidence of small amplitude were to occur 
with a very long horizontal wave-length, and with a period of say one or two 
minutes or more, it would be practically impossible even to detect its existence by 
either of the methods named, unless by chance it were repeated several times with 
uniform period in the presence of a very frictionless vibrator whose free period 
happened to agree nearly with the period of the disturbance; even then no 
measurement of its amount could be made. We are in fact forced to classify earth- 
movements under the two heads which have been named—not because there is any 
necessary discontinuity between the two, but because they must be treated by two 
entirely distinct modes of observation. 
For the measurement of palpable earthquakes by the inertia method the writer 
has devised many instruments which have been successfully applied to the regis- 
tration of Japanese earthquakes, and which are described in a memoir on earth- 
quake measurement published in 1883 by the University of Tokio. He has not 
attempted in any case to give the astatically suspended mass three degrees of 
freedom, and nothing would be gained by doing so. An instrument with two 
degrees of freedom is now exhibited to the Association. It consists of an ordinary 
pendulum coupled with an inverted pendulum in such a manner that the two bobs 
move together in any horizontal direction. This combination of a stable with an 
unstable mass can be adjusted to give any desired degree of astaticism. In 
practice it is convenient to allow the joint mass to have a free period of from five 
to ten seconds—the period of ordinary earthquake waves being much less than this, 
A long and light lever pivoted to the frame of the instrument at one point and to 
the steady mass at another forms a registering index, by which a magnified trace 
of the earth’s horizontal movement is deposited on a fixed plate of smoked glass 
with the least possible friction. 
In another instrument two separate components of horizontal motion are deter- 
mined each by a horizontal pendulum tilted slightly forwards to give a small 
degree of stability, and furnished with a multiplying pointer. In this instrument - 
the pointers trace the successive movements of the earth on a plate of smoked 
glass, which is kept revolving uniformly by clockwork. The velocity and accelera- 
tion of the movements are deducible from the records. This is the standard form 
of seismograph employed by the writer; and to make the information it gives 
complete, another instrument for registering on the same plate the vertical motion 
of the ground is added. 
The vertical motion seismograph is a horizontal lever, supported on a horizontal 
fixed axis, and carrying at one end a heavy mass. A spring, attached to a fixed 
5 
