380 
PRESTON. 
The preceding equation takes account of the matter lying 
above the sea-level. The effect of the sea-water would be 
to increase the deflection of the plumb-line at Kaupo, inas¬ 
much as the space below the level of the sea is filled with 
matter which is heavier on the north side than on the south. 
The influence of this matter would be that of a mountain 
having a density equal to the difference of the densities of 
the land and sea, or 
2.90 — 1 1.90 
5.576 — 5.58’ 
the value of 5.576 being taken from Harkness* and the 
value 2.90 results both from a study of the surface rocks 
and from a determination of the density of the mountain as 
a whole by means of the pendulum. If in equation (14) we 
should use d — instead of d, we would have 
Z) = 34".12|(i-i); 
so that the attraction calculated for the matter above the sea- 
level must be multiplied by —^—, or 0.655, to get the effect 
of the sea water, and the combined effect of land and sea 
would be 
D = 34".12 ~ X 1.655 = 56".47 -■ 
A A 
If we consider the ring between the 24th and 25th circles 
and calculate the attraction for the highest compartment, 
assuming 1.5 miles for h, we get for the result 0.01397 
whereas by neglecting K 1 we have 0.01430 a, giving a differ¬ 
ence of about one part in forty. As a matter of fact, there 
are many compartments where h is even greater and where 
the distance to the attracted point is still less, both of which 
circumstances would augment the attraction. It has there¬ 
fore been deemed advisable to consider the second power of 
the height in all compartments, although it has doubtless 
* Harkness (Prof. Wm.) The solar parallax and its related constants. 
4°. Washington, 1891. [Appendix III. Washington observations, 1885.] 
