Factors in a Ghyben-Herzberg System —Wentworth 
177 
the zone of mixing are to be regarded as 
short circuits which would produce poten¬ 
tial disturbance analogous to that in an elec¬ 
trical network. Moreover, the movement of 
saline water toward fresh, or vice versa, 
taking place during any phase would leave 
residues of great importance. It therefore 
appears that, increasingly, irregularity of per¬ 
meability would tend strongly toward efface- 
ment of the salt-water-fresh-water contact 
on which the Ghyben-Herzberg lens de¬ 
pends, just as would increase of general 
permeability. 
It appears that this variation in permeabil¬ 
ity, with some large openings going outside 
the favorable range of permeability, may be 
a very large factor in explaining the great 
differences in the Ghyben-Herzberg condi¬ 
tions on different coral limestone islands, or 
on different parts of the same island. Not 
only initial differences due to structure of 
the calcareous accumulation, but also fissures 
developed near sea level by the action of 
fresh water probably are important here. 
Such an interpretation has been suggested 
by the writer’s observations in the Marianas 
during early stages of military operations 
there and has also been emphasized by 
others. 2 
CAP ROCK 
It is not difficult to show why an effective 
cap rock is so very significant in permitting 
the growth of some of the larger Ghyben- 
Herzberg lenses. It is an essential part of 
that theory that in a steady condition of 
dynamic balance, with the thickness of the 
lens neither increasing nor decreasing, lines 
may be drawn from various points of the 
top of the lens (the water table) passing 
downward and outward in a wide curve to 
emerge in the ocean surface, along which 
hydrostatic pressures are in balance. If the 
2 Piper, Arthur M. Letter dated December 5, 
1946. 
position of the water table is changed by 
fluctuations of recharge or of loss, move¬ 
ment will tend to take place along these lines 
in accordance with the size of openings and 
length of path. It is evident that those paths 
Fig. 2. Schematic sections through the margin 
of a Ghyben-Herzberg lens, under the three con¬ 
ditions indicated in the text. For convenience in 
drawing, the mean line of balance is based on a 
ratio of 10 : 1, rather than the true ratio of ap¬ 
proximately 40 : 1. 
