were: surface, 16.7°; bottom, 11.9°; and mean air 

 temperature, 18. 0°. The average number of clear 

 days per month was 11. 2, which is normal for this 

 period (table 4). As the normal mean aii temper- 

 ature during this period is 18. 3°, the amount of heat 

 received by the lake this summer was probably close 

 to normal. The annual heat budget, as derived 

 from the maximum -minimum means, can not be de- 

 termined, as the survey did not cover the period of 

 minimum temperature. 



Bilge, Juday, and March (1928) in their compre- 

 hensive investigations of Lake Mendota have shown 

 that approximately 8 percent of the heat budget of 

 that lake is absorbed by the bottom mud. This heat 

 is returned to the water mass in winter and in early 

 spring. Lake Mendota is comparatively small, but 

 its maximum depth (24. 5 meters) is the same as that 

 of the Great Plain of Lake Erie. As Lake Mendota has 

 a much smaller percentage of water over 20 meters 

 than Lake Erie and a different thermal cycle, no 

 direct comparison is attempted. Nevertheless, the 

 absorption and subsequent release of heat by the upper 

 stratum of bottom deposits must be an appreciable 

 factor in the thermal cycle of Lake Erie. When bot- 

 tom water of the marginal zone reaches a tempera- 

 ture of 1° or 2° C. , it is slowly warmed by the bot- 

 tom and becomes heavier as it approaches 4° (tem- 

 perature of maximum density). This heavier water 

 tends to move to greater depths; and in areas such as 

 those that surround the Deep Hole, where the slopes 

 are steep, the movement may be of importance in 

 preventing physical stagnation in the deep waters of 

 the lake during the cold months. 



From available data, the following estimate is 

 made of the probable time that maximum mean 

 temperatures were reached: 



Air July 15-30 



Surface water August 10-20 



10-meter level August 20-30 



Bottom water Sept. 20-Oct. 5 



Surface of bottom mud. . . October 15-30 

 The highest surface temperature observed in 1929 was 

 22. 3° C. at station 4-38 on July 12. The day was 

 calm and smooth and the air temperature was 24°. 



To aid in visualizing the thermal changes that 

 took place during the season, temperature graphs 

 have been prepared of the water columns at stations 

 15 and 40. Figure 14 shows the graphs for station 15, 

 the deepest part of the lake. It will be seen from this 



figure that the water column was practically 

 homothermous on May 24. By June 11, vernal 

 warming had penetrated to a depth of 30 meters, 

 and on June 26 a thermocline was evidenced be- 

 tween 5 and 10 meters. The surface, at 18° C. , 

 had warmed 12° in a month. The strong winds that 

 swept this area during the first week of July broke 

 down the thermocline above 10 meters; and on July 

 8 the column of water was divided into two fairly 

 even gradients, from the surface to 20 meters and 

 from 20 meters to the bottom. The wave action due 

 to strong winds lowered the temperature from the 

 surface to 7 meters, and raised it from 7 meters to 

 20 meters (crossing of the curves for June 26 and 

 July 8). 



The closely parallel curves for July 25 to Sep- 

 tember 7 depict the normal thermal condition of 

 the area for this time of the year. The upper layers 

 are thoroughly mixed down to 20 meters, where a 

 broad discontinuity layer is encountered between 

 that depth and 30 meters; the temperature drop in this 

 10 -meter stratum is about 10° C. From 30 meters 

 to the bottom (60 meters), the gradient is uniform. 

 The bottom temperature gradually increased from 

 4. 5° on May 24 to 5. 5° on September 7. The tem- 

 perature here probably never exceeds 6. 0°. 



The graphs for station 40 (fig. 15) are based on 

 table 6-L. The cycle of the thermocline is of suf- 

 ficient interest to warrant the careful inspection of 

 these curves. Station 40, as may be seen from 

 figure 1, is in the deepest part of the Great Plain, 

 and is typical for this large area. 



On May 17, the water was cold and homother- 

 mous from a depth of 3 meters to the bottom and 

 the surface was only 1° C. warmer than the rest of 

 the column. On June 17, the bottom temperature 

 had risen 2° and the surface 9°, and the vertical 

 gradient was steep in the upper 10 meters. By July 

 12 continued surface warming and convectional 

 mixing changed the upper portion of the vertical 

 gradient to the extent that the differential was now 

 only 5° between the surface and 19 meters. The 

 thermocline was first observed on this date when the 

 temperature dropped 3. 3° from 19 meters to 20 

 meters. August 17 found the upper water mass well 

 mixed; the warm-water column with less than a degree 

 of differential reached from the surface to the ther- 

 mocline at the 20 -meter level. A difference in 

 temperature of 6. 0° was now found between 20 and 



53 



