114 THE FRESH- WATER LOCHS OF SCOTLAND 
warmer portions of the lake, and for a time these currents will be 
vigorous. But when these currents have cooled down the surface of 
the open water sufficiently, and have slackened in consequence, a 
portion of the water round about the ice-fringe becomes frozen before 
it can reach the warmer water in the lake, and when this stage is 
reached freezing proceeds rapidly. 
In many of the Scottish lakes, however, the shores are steep, and 
freezing takes place over the whole surface at once, without the 
formation of a shore fringe. Freezing in such lakes usually takes 
place during a calm night in spring with a sharp frost, and an open 
sky which favours radiation and evaporation from the lake. The 
surface is at first covered with a network of ice-crystals, and then by 
a thin sheet of ice. One fact about this mode of freezing, which at 
first sight is remarkable, is that lakes that have never had ice on them 
during winter — that is, say, up to the end of February — may be covered 
with a sheet of ice in a single night during spring. The reason for 
this is that in spring there are frequently calm nights with intense 
frosts — spring frosts, as they are called — which are favourable for 
the rapid formation of ice. Evaporation is very active, owing to the 
relative dryness of the atmosphere in spring, and on a clear, dry night 
a wet surface may be from five to ten degrees lower than a dry one. 
In addition to this, all through the winter the water in a lake is 
continually losing heat, and it is in early spring, before heating of the 
water begins to take place, that its temperature is lowest and freezing 
takes place most readily. In the spring of 1908 observations were 
made in Loch Garry, and though at no time was the loch completely 
covered with ice, on several occasions during March large portions of 
the surface were covered with ice in a single night — with a thickness 
of as much as half an inch. 
But freezing of this nature does not necessarily take place uniformly 
over the surface of the lake. Ice forms in irregular patches stretching 
long arms out into the lake. Sometimes there may be open water at 
the shores, and ice over the deepest parts of the lake. Frequently ice 
forms about the mouths of small streams, for the waters entering the 
lake may be very nearly at freezing point, and will float on the surface 
of the water in the lake and freeze readily. But this is not enough 
to account for the irregular formation of ice. Isolated patches occur 
well out from the shores, and not connected with them at all. This 
may partly arise because some parts of the lake are more sheltered 
than others from local breezes, which disturb the surface and produce 
mixing of the surface layer with the warmer water below, so prevent- 
ing freezing. 
There is another possible explanation of the patchy formation of 
ice. Everyone who is familiar with the appearance of the surface 
