A WMNOLOGICAI. STUDY OF THE FINGER TAKES. 
557 
the lake as compared with a mean temperature derived from numerous sets of observa- 
tions, but we have done much work with this end in view. In the New York lakes, 
series of three or four sets of observations were taken along the axes of three lakes. 
The results are shown in the following table. In each lake the stations are named from 
south to north; the mean temperature is computed to the depth of the water at the 
end stations; and the result from the center station is placed in italics. 
Table VI. — Mean Temperatures op Takes as Computed from the Several Series of 
Observations, and from their Mean. 
Lakes. 
Year. 
A. 
B. 
c. 
D. 
Mean. 
Canandaigua 
1910 
1911 
I9II 
II. 94 
12. 17 
15 - 34 
12 . 00 
12. 27 
15 - 13 
12. SO 
12 . II 
14. 80 
12 . II 
12. 00 
15. 07 
Do 
II. 96 
14. 82 
Explanation of table.— The first series is computed to the depth of 50 meters; the second, to 40 meters; the third, to 30 
meters, this being the depth of the lake at the end stations. Figures in italics indicate the center station. 
Ju/y 
^3 
S'? 
30 
3/ 
2 
3 
a 1 
, ^ 
1 
1 _ 
•* * 
• 
• 
* 1 1 
..... 
* . 
• 
•• 
..1.- 
• • 
1 
Ll_ 
•• 
, *• 
• • 
••i -i 
Fig. 7. — Mean temperature of Green Eake, Wis. From observations at center of lake in 1911. Computed to depth of 60 meters, 
the maximum depth at the center. All temperatures lie between 11°, or slightly below, and 12°. The position of the dots 
along the horizontal axis of the diagram indicates the day and hour of the observation. 
The above table shows that the mean temperature deduced from the center series 
differed by less than i per cent from that derived from three or four series. Similar 
results have been reached in more numerous cases in the Wisconsin lakes. In Lake 
Mendota in 1911, 1912, and 1913, observations were regularly taken at 10 to 12 different 
stations so placed as to give the distribution of heat in this lake, which is 9 kilometers 
long, 6 kilometers wide, and 24 meters deep. In series taken during these years at 
approximately regular intervals of time on 36 dates from June to September, the maxi- 
mum departure of the mean temperature derived from the 10 or 12 series from that 
derived from the single series at the center was less than 3 per cent and this was 
reached only once; and the mean departure was less than i per cent. The maximum 
differences were in June, when the lake was warming, and the distribution of heat was 
more irregular than later. In July, August, and September, the maximum departure 
was about 1.5 per cent and the mean 0.8 per cent. 
In five similar series taken in Green Lake, Wisconsin, the maximum departure of 
the mean temperature derived from the middle series and that derived from all was 
5 per cent and the mean was less than 2 per cent. 
From July 26 to August 5, 1911, observations to determine the temperature seiche 
were made on Green Lake. During this work 83 series of temperatures were taken at 
the middle station. The results are shown in figure 7. 
60289° — Bull. 32 — 14 36 
