470 Proceedings of the Royal Society 
60° F., the time being that of low water on a calm day without 
wind. The river is flowing out by a narrow channel about two 
feet deep over the bar. The tide gradually rises, and opposes an 
increasing resistance to the outflowing current, the salt water rises 
over the bar and creeps up the bottom of the channel, raising the 
level and so gradually widening the opening by covering sand- 
banks. At about quarter flood the river begins to rise near the 
mouth from the damming back of the fresh water, then suddenly 
(see observations III, on p. 466) a thin wedge of brackish water 
is found inserting itself between the bottom and the fresh water ; 
a few minutes later there is a layer of salt water a foot deep 
topped by an equal stratum of brackish, and as time proceeds 
the salinity at the bottom remains constant, but extends higher, 
and passes through a brackish zone into the river water, about a 
foot or eighteen inches of which remains hardly touched by salt, or 
even quite fresh, and flows out to sea slowly, tearing itself across 
the layers of denser water flowing at a lower level more slowly but 
resistlessly upwards. About an hour before full tide there is a 
state of balance ; the wedge of salt water has pushed its brackish 
edge to the highest point, and the enfeebled force of the tide only 
suffices to hold back the dammed up water above. This con- 
tinues for some time ; then there is a turn, the salt wedge is forced 
slowly out, the dammed back water rushes down unchecked, keep- 
ing up the level of the river near the mouth for some time, while 
the salt and finally brackish water withdraws along the bottom ; and 
by half ebb the river down to the bar is entirely fresh, and rapidly 
returning to its low-water level. 
The diagrams on Plate XIV. may be explained on the supposition 
that water containing less than 0'5 per cent, of salts in solution may 
be considered fresh, while when the percentage of salt is greater, 
but does not exceed 2 per cent., it is brackish, and when more than 
2 per cent, of salt is present it is thoroughly salt. In order to be 
truly representative, the shades of colour in these figures should 
have the margins blended, but reference to the curves (Plate XIII.) 
will show that if the abrupt transitions did not take place at points 
corresponding to 0*5 and 2 per cent, of salt respectively, they did 
take place very near those positions. The diagrams show how with 
the rising tide the layers of “ fresh ” and “ brackish ” water remain 
