Birge—Work of Wind in Warming a Lake. 
349 
TABLE 1. 
Work of wind and calories of wind-distributed heat for Lake Mendota. 
The direct curve of work. 
A 
B 
C 
E 
E 
F 
Depth, m. 
Temp. 
RTXZ 
1 -D 
Product, G. cm. 
Calories 
0-1 
24.0 
4800 
0.002680 
12.86 
1920 
1-2 
23.4 
13500 
2530 
34.15 
1750 
2-3 
23.0 
21000 
2440 
51.24 
1600 
3-4 
22.7 
27700 
2360 
65.37 
1480 
4-5 
22.6 
34200 
2340 
80.03 244.65 
1410 8160 
5-6 
22.4 
39600 
2300 
91.08 
1320 
6-7 
22.3 
44900 
2270 
101.92 
1260 
7-8 
21.5 
50300 
2090 
105.13 
1170 
8-9 
20.0 
54400 
1770 
96.29 
1020 
9-10 
18.0 
58900 
1380 
81.28 475.70 
870 5640 
10-11 
16.0 
60900 
1030 
62.73 
700 
11-12 
15.1 
62100 
889 
55.21 
600 
12-13 
14.6 
63800 
814 
51.93 
540 
13-14 
14.2 
64800 
757 
49.05 
490 
14-15 
14.0 
63800 
720 
46.51 265.43 
440 2770 
15-16 
13.7 
65100 
688 
44.79 
410 
16-17 
13.6 
62700 
674 
42.26 
360 
17-18 
13.4 
56000 
647 
36.23 
300 
18-19 
13.2 
50000 
621 
31.00 
250 
19-20 
13.1 
41000 
609 
24.97 179.25 
190 1510 
20-21 
13.1 
27900 
609 
16.99 
124 
21-22 
13.1 
21900 
609 
13.34 
93 
22-23 
13.0 
15600 
596 
9.30 
62 
23-24 
13.0 
7600 
0.000596 
4.52 44.15 
13 290 
1209.18 1209.18 
18370 
The results of the computation show that the heat gained 
above 4° by Lake Mendota in 1910 amounted to about 
18400 cal. per sq. cm. of surface, and that it required about 
1209 g. cm. of work to distribute this heat through the lake. 
The table also shows the amount of heat given to each one- 
meter stratum of the lake and the amount of work needed to 
transport it there. Both these sums are also stated in terms 
of a square centimeter of surface of the lake. 
The results of this table are shown graphically in plate VI. 
6. COMPUTATION BY LARGER INTERVALS. 
If accurate results are desired the computation should 
be by single meters either for the full depth of the lake or at 
least to a depth well below the thermocline. Such a compu¬ 
tation is necessary if, for instance, diagrams are to be pre¬ 
pared similar to plate VII. But if general results only are 
needed the process can be much shortened by using larger 
intervals of depth; but if this method is followed the result 
rarely will if ever, be exactly the same as that from the 
