477 
of Edinburgh, Session 1885 - 86 . 
The observations Xa. show an entirely different set of conditions. 
The river was partially flooded, and the wind had been easterly for 
some days, thereby probably keeping the freshened surface water 
from spreading to the eastward. The result was that, within a mile 
of the shore there was an area, the surface density of which varied 
from 1 *01 2 to 1*024, the bottom water being scarcely below the 
normal (about 1*0256). The observations, Xb. and Xc., with the 
small hydrometer, show how the density increased steadily in a 
northerly direction from the river until the water was quite salt on 
the surface, and how, when the boat was turned towards Port 
Gordon, the salinity at first increased and then diminished as the 
fresh current sweeping to the eastward against the flood tide was 
entered on. The diagram, Plate XV. fig. 2, gives an idea of the 
manner in which these, combined with numerous other observations 
not recorded in detail, indicate the direction in which the river runs 
through the bay. The colouring shows the difference between the 
salt and brackish surface water, the former being represented by 
the darker shade. Density by the small hydrometer is marked at 
intervals along the course of the boat. 
It is evident, from the numerous observations of density made 
along the beach by means of the small hydrometer, that the shore 
water is much fresher on the east than on the west side of the river, 
and that on going eastward from the river the observer finds the 
density increase to a certain point (Xllla.), and then diminish, as 
if the current at first swept out to sea, and then curved back along 
the shore. This is indicated in a general manner in the diagram 
referred to above. The series XlVa. and XIV5. show the greater 
freshness on the east side very clearly, although on the day when 
those observations were made the salinity on the west side was con- 
siderably less than usual. The generally-accepted fact, which we 
in part verified, that much of the Spey water finds its way to the 
sea by percolating through the shingle, may account for the uni- 
formly low density along the shore. 
The phenomenon of percolation is presented on the small scale by 
the Tynet burn, near Port Gordon (Plate XV. fig. 2). It is a small 
stream flowing into a long, narrow deep pool without an outlet, one 
side of which is a shingle bank, sloping down to the sand. Its 
length is parallel to the shore line, and nearly at right angles to the 
