NEW INSTRUMENT FOR GRAVITY MEASUREMENTS 27 
the instrument and subtract the effect of elevation, which is one part in ten 
million for every foot. These differences in elevation can be determined to 
within three or four feet with the Paulin Altimeter, provided that it is used 
according to a particular system that has been worked out for this purpose, 
but if greater accuracy is desired the land must be surveyed. These facts 
make the use of the gravity balance much simpler than any other geophysical 
instrument except the magnetometer. 
Recent field work has shown that in reasonably good weather it is easy 
to make three stations per hour. The torsion balance makes three stations per 
day and the Government pendulum apparatus three or four stations per 
month. This measurement which takes only a few minutes appears to have 
four or five times the accuracy of a single run of the pendulum which takes 
twelve hours. The information obtained is exactly the same as with the pen- 
dulum, the total intensity of the gravitational field. For this reason the pres- 
ence of a heavy mass near by on the surface or at small depth has no effect on 
the reading; the presence of such a mass merely alters the direction of the 
Gravity gred:ent ca 
Gravity onemely v 
Ay as So ogansons | 
a Loe oa. 
a DO Zea 
Ly 
Y/; Uj We 
Fig. 4. Gradient and gravity curves for assumed structure. 

resultant force and as the instrument is set exactly in line with this direction 
the measurement is the same. These surface masses make so much trouble 
for the torsion balance operator that it seems to be hard for him to believe 
that they do not disturb this instrument. 
The relation between the gravity anomaly, which is measured directly 
by the gravity balance, and the gravity gradient, which is measured by the 
torsion balance, is very clearly shown in a diagram in Ambronn’s Elements of 
Geophysics, which he has taken from a paper by von Eotvos. Figure 4 is 
derived from this diagram; the heavy curve representing the values of the 
gravity gradient and the light curve the gravity anomaly caused by a mass 
of heavy rock at varying depths below the surface. It will be noted that the 
curve representing the magnitude of the anomaly follows very closely the 
outline of the rock mass below while the curve representing the magnitude of 
the gradient bears no resemblance whatever to this outline and requires some 
study to interpret its meaning. In actual field work these curves are not ob- 
tained but only a rather small number of points on them. A comparatively 
171 
