1880 .] 
71 
[Nichols. 
the bottom of the pond, and take the several means of the two 
observations we have the following : — 
December 
19 
3°. 8 
Three days before freezing. 
January 
30 
2°. 8 
February 
6 
3°. 9 
u 
17 
2°. 9 
d 
18 
2°. 9 
These figures show conclusively that there was no appreciable 
warming of the mass of the water after the first observations 
were made through the ice, and as we, unfortunately, do not 
know the bottom temperature just after the ice had formed, we 
are unable to say whether any warming of the bottom took place. 
It would appear that the warming of the water from the heat of 
the ground and from the entrance of ground-water was counter- 
balanced by the cooling from above. On this point, I hope to 
obtain further data myself during the coming winter, but, in any 
event, a natural pond is not the same thing as a large body of water 
in an impervious basin. There is generally water entering the pond 
through the ground, in winter at a higher, and in summer at a 
lower temperature, and this must influence the temperature of the 
mass of water to a greater or less extent. In the case of an 
impervious basin with a constant surface temperature, in contact 
with the ice, of 0°, we should expect the temperature of the lower 
layers to fall, owing to conduction upwards to the colder layers 
and to the ice, if the cooling from 4° to the point at which ice 
forms is really due to conduction. The whole truth in the matter 
can hardly be reached except by the accumulation of a consider- 
able number of observations made on different ponds and under 
different circumstances. The observations most difficult to make 
at any distance from shore, and the ones which are at the same 
time of very great value, are those taken the day before ice forms, 
and those made immediately after the coating of ice has become 
continuous. 
There is one other point on which a larger number of observa- 
tions is desirable. I refer to the depth to which the diurnal 
changes of temperature extend under different circumstances. 
The curves of temperature in Mystic Pond show several times 
when a few successive days of warm or cold weather produce an 
effect on the water, even at a depth of seventy-five feet. Note 
for example, the rise of temperature observed on November 15, 
