of Edinburgh, Session 1885 - 86 . 
425 
Table XII. 
Depth. 
Dates of intersecting Curves. 
Mean Date. 
5 
18th August. 
5th September. 
27 th August. 
9 
18 th August. 
22nd September. 
5th September. 
14 
5th September. 
22nd September. 
14 th September. 
16 
5th September. 
15th October. 
26th September. 
19i 
22nd September. 
15th October. 
4th October. 
23 
22nd September. 
14th November. 
19th October. 
65 
14th October. 
14th November. 
31st October. 
As the autumn of 1885 has been a cold one, the above mean 
dates may be taken as the earliest dates of maximum temperature at 
the depths indicated. 
If we compare the mean temperature of the whole column of 100 
fathoms of water at Inversnaid on the 5th and the 22nd September, 
we have on the 5th the mean temperature 44 0, 54 F., and on the 
22nd September 44° '52, the difference being 0°*02 F. 'Remember- 
ing that heat is lost more quickly in autumn than it is gained in 
spring, and considering that these temperatures are nearly identical, 
we shall not be far wrong if we put the epoch of maximum 
heat in the water of the deepest part of the lake as occurring some 
time in the third week of September. This brings it very near the 
date of the equinox, and it seems natural to expect the heat to 
accumulate in the water so long as the day is longer than the night, 
and to decrease so soon as the conditions are reversed. This con- 
clusion is supported by the observations of Fischer, Forster, and 
Brunner, who made a most interesting series of observations on the 
distribution of temperature in the Lake of Thun, in Switzerland, 
during the years 1848 and 1849. They found hardly any increase 
of heat between 3rd February and 28th March, but after the latter 
date the influx of heat was very rapid. It is probable, therefore, 
that the temperature of the bottom water in our deepest lakes 
depends chiefly on the temperature of the air between the preced- 
ing autumnal and vernal equinoxes. Two causes combine, namely, 
the greater meridian altitude of the sun and the greater length of 
the day in the summer than in the winter half year. The latter 
cause has the effect that the water is exposed to heating for a 
greater portion of the twenty-four hours than it is to cooling, while 
the former cause ensures a greater supply of heat per minute during 
the day in summer than in winter. 
