242 
DR. HAROLD JEFFREYS ON TIDAL FRICTION IN SHALLOW SEAS. 
shore is fairly open and regular in outline, like most of the coast of Africa, it is not 
possible to find the dissipation along it, for there are no data to shoAv how far out 
the currents extend. In partly enclosed regions, however, it is frequently possible to 
interpolate between the records made on opposite sides. A serious difficulty may 
arise if the depth of the sea is very different at different points within it, for this 
may destroy the possibility of interpolation, and therefore we must always examine 
the soundings for any great variation. Ordinarily we should not expect much 
variation in the velocities, for such places are intermediate in character between 
narrow channels and the open shore, and therefore the currents in them may be 
expected to show some increase in shallow water, but not so much as would be caused 
by a proportional decrease in depth along a narrow bay or in approaching the open 
shore. In shallow water also friction may, and often does, neutralize the magnifica¬ 
tion that would occur in its absence. 
One other fact may be noted. In shallow bays the difficulties of navigation 
may be great, and navigators avoid them if possible by choosing a harbour 
near the entrance. Thus observations of currents are most numerous about the 
entrances, and often at the very places where the currents, and consequently the 
dissipation, are greatest there are insufficient observations to give a satisfactory 
estimate. 
Great care must be taken in dealing with observations among islands, straits, and 
shoals. When the passage of a tidal current is obstructed by a shallow of small 
horizontal extent, part of it goes round the shoal and part over the top. The 
influence of this on the main current is of course small, but on the top of the shoal and 
in its immediate neighbourhood the velocities may be much increased, for much the 
same reason as accounts for the greater speed of a river where it is shallow. On the 
other hand the increase in the influence of friction may greatly reduce the currents, 
and shoals often afford in this way an important shelter from tidal currents to the 
deeper water behind them. This is particularly noticeable at some points on the Korean 
side of the Yellow Sea. Thus observations of currents taken at lightships and buoys 
over shoals whose dimensions are all much smaller than those of the main bay or channel 
must be regarded as giving no reliable estimate of the main current. Small islands 
also require examination before the records obtained are accepted. If one is surrounded 
by a shoal they are of course untrustworthy ; but if deep soundings are found 
within a few miles of it, they will probably give a very good idea of the main current, 
which will, especially in a wide channel, be fully as useful in our investigation as the 
results of observations at the sides. Straits are in a different position. When the 
tides in two seas or even oceans are in widely different phases, a large head may be 
produced between the two ends of a strait connecting them, so that a swift tidal 
current will flow along the strait. In no circumstances, however, can this give any 
indication of the currents in the seas, for it is produced by the tide heights, and not 
directly by the currents. Such currents may attain very great velocities, as in the 
