TEMPERATURE OF SCOTTISH LAKES 
121 
For the most part, brine solutions were used — a dense solution 
representing the coldest water in the loch, and water of less salinity 
representing the warmer and lighter layers. 
The results of these experiments may be stated as follows : — 
(1) During the period of the year when the differences of tempera- 
ture in the lake are small, there is a vertical circulation of all the 
water due to currents of wind, and the return current which supplies 
water to take the place of water driven along by the wind at the 
surface is appreciable to the bottom of the lake. This was also 
thought to be the case from temperature observations in Loch Ness 
and other lakes, which show^ that in winter, during stormy weather, 
the isothermals of the lake may be nearly vertical, necessitating the 
existence of a current reaching to the bottom of the lake, though 
probably a very slow one ; and indeed the uniformity of temperature 
in a lake at certain seasons of the year is, I think, sufficient proof 
that there is a circulation of the water. 
DIRECTION OF WIND 
ab - discontinuity lays it 
c - surface current 
d - primary return current 
e - secondary surface current 
f - secondary return current 
Fig. 45. 
(2) When the temperature gradient in a lake becomes marked 
the density gradient is also marked, and produces its effect on the 
current systems. It is thought, a 'priori^ and from the observations in 
the experimental tank, that during this period the depth at which 
the return current is felt grows gradually less as the difference in 
density increases, and that by the time the discontinuity layer is 
formed the return current is strongest in the neighbourhood of the 
discontinuity. 
(3) The observations in the experimental tank also indicated that 
below this return current at the discontinuity, which I call the primary 
return current, there is a secondary return current in the opposite 
direction to the primary return current, and induced by it, but in the 
same direction as the surface current. The system of currents which 
is thought to exist is shown in fig. 45.^ Such a secondary return 
current would necessarily be very slow. Temperature observations 
^ Recently J. W. Sandstrom has described a similar system of currents in his 
"Dynamischen Versuchen mit Meerwasser," Ann. der Hydrographie, Heft i., 1908. 
