558 
BULLETIN OF THE BUREAU OF FISHERIES. 
The diagram shows variations between the several observations; and also displays 
a slow but steady warming of the lake, as would be expected at that date. But for all 
purposes of a general discussion any one of these results might have been taken as 
representing the temperature of the lake quite as well as their mean, or as the mean of 
the several series of observations taken during this period at points along the whole 
length of the lake. 
This general result has been found to be true for observations taken at this time 
of year in all lakes during many years. It is therefore fair to conclude that the mean 
temperature of the water of a lake of simple form in late summer may be derived Jrom 
a single series of observations taken at or near the center of oscillation of the water. 
In the New York observations the least satisfactory series is that of Skaneateles 
Lake in 1911, when time did not permit me to go to the center of the lake. Seneca 
Lake offered the least favorable situation for taking the temperature in that the deepest 
water of this lake lies to the south of the center, about one-third of the distance from 
the south end. At so great a distance from the middle the water is subject to consid- 
erable oscillations. During 1910 five series of temperatures were taken from August 2 
to August 9. The mean temperature as deduced from the separate series ranged from 
7.37° to 8.05°, with a mean of 7.71° — a variation of about 4.5 per cent on each side 
of the mean. This variation was wholly due to variations in the apparent distribution 
of the heat and not to actual changes of temperature in the lake, since the temperature 
of the o-io meter layer was practically constant at 19.5° to 19.6°. The 10-20 meter 
stratum varied from 13.2° to 16.3° — enough to give a difference of 0.28° in the mean 
temperature. The 20-30 meter stratum varied from 7.3° to 10.2°, which would give 
a variation of about 0.25° in the mean temperature. Thus the oscillations in the stratum 
between 10 meters and 30 meters account for about 80 per cent of the range of the tem- 
perature, and relatively little is due to changes above or below the depths named. This 
variation is considerably greater than would be expected in case of the other Jakes in 
which temperatures were taken at the center, or several series were taken along the 
axis of the lake. 
The following table shows the mean temperatures of the lakes as deduced from the 
observations taken, both for summer and winter. 
Table VII. — Mean Temperature of the Water op the New York Lakes, as Observed in 
Summer and in Winter. 
Lakes. 
Tm®. 
1910. 
Tm'-. 
1911. 
Tm®, 
1911. 
Tm--. 
1912. 
TniL 
I 9 I 2 . 
19.32 
II. 07 
9. 26 
12. 17 
15 - 75 
13 - 59 
7. 71 
10. 10 
11. 90 
9. 99 
8. 94 
II. 48 
2. 23 
Otisco 
0. 83 
3 - 39 
1. 10 
2. 13 
12. 86 
7-35 
10. 84 
1 1. 42 
1.49 
13 - 93 
2-39 
I. 74 
® 10. 21 
II. 96 
o Oct. 18, and therefore below Tm®. See p. 565. 
