TEMPERATURE OF SCOTTISH LAKES 
117 
over the water. The result is that the ice to windward is rough and 
white, while that grown on the other side of the sheet is clear and 
crystalline. 
Much interesting information about the phenomena accompanying 
the formation of ice is contained in Dr von Cholnoky's " Das Eis 
Balatonsees,"" ^ to which reference is made. Lake Balaton is pre- 
eminently suitable for a study of ice-formation, as its shallowness 
makes it respond readily to variations in atmospheric temperature. 
During the winter of 1892-3 the ice attained a thickness of about 
18 inches. 
The DiscoNTiNuri'Y Layer 
The general character of the discontinuity layer has already been 
described, but without reference to its cause. Both in Loch Garry 
and in Loch Ness the growth of the discontinuity was carefully 
observed, and it was found that whenever the atmospheric conditions 
were such that there was no accession of heat to the lake, there was 
a tendency for the formation of a layer at the surface of uniform 
temperature, and a consequent discontinuity at the bottom of this 
uniform layer. There are two principal causes for the formation of 
the uniform layer: — (1) Owing to the cooling of the surface-water 
during cold days and during the night, vertical convection currents 
arise which equalise the temperature of the water ; (2) these vertical 
currents are not appreciable to any great depth — probably to not 
more than 10 feet. But the circulation produced by the wind 
deepens this layer and equalises the temperature to considerable 
depths. 
The time of year at which the discontinuity makes its appearance 
varies with different lakes. In the great and deep tropical lakes, 
such as Loch Ness, it does not appear so early as in small lakes of the 
temperate class, for in these small lakes the surface temperature 
rises much more rapidly than in large lakes, and in consequence the 
time at which the lake ceases to gain heat arrives sooner than in 
large lakes. Thus in Loch Garry the discontinuity was distinct in 
July, and even in June, whereas in Loch Ness it was not distinct till 
about a month later. Varying seasons must also be taken into 
account with regard to the period at which the discontinuity appears. 
The growth of the discontinuity in Loch Garry at a depth of 
from 50 to 75 feet is shown by the following observations, made in 
June, at a time when, owing to dull and cold weather, the rate of 
increase of heat in the loch was arrested, and there was an actual 
decrease : — 
1 ResuUate der wissenschaftlichen Erforschung des Balatonsees, Bel. i. Th. 5, 
sec. iv. 
