Hot and Thermal Springs. 389 



So long as the temperature of the bottom of a lake be above 

 38°. 75, currents of water will continue to rise, but they will rise 

 more feebly, and with a smaller increase of heat, the nearer the 

 temperature of the bottom approaches that of the maximum den- 

 sity of water. If in higher latitudes, or in greater elevations, the 

 temperature of the bottom of a lake does not exceed 38° 75, an 

 equihbrium will take place : the bottom of the lake will neither re- 

 ceive heat from the water nor communicate heat to it, and the 

 lower beds of water will never be warmer than 38°.75. If, lastly, 

 the temperature of the bottom of the lake fall below 38°.75, the 

 water which was cooled by the air at the surface will sink to the 

 bottom, and there become still more cooled by coming into equi- 



difFerences had been observed the preceding year. See Schweigger, Seidel's 

 Jahrb. der Chemie, &c. &c. vol. lix. p. 32. Nineteen observations on the 

 temperature of the Lake of Constance, between the 23d May and 30th July 

 1830, taken at three o'clock p.m., gave a mean temperature which was 3°.60 

 lower than the mean temperature of the air during the same period. 

 Schweigger Seidel's Jahrb. vol. Ixv. p. 57- There is, therefore, no doubt 

 that the mean temperature of the air during this period was also lower 

 than that of the surface of the lake. During frosts the mean temperature 

 of lakes must be higher than that of the air, as the temperature of the 

 water cannot fall below 32°, whilst that of the air descends several degrees 

 below 32°. During the severe cold in January and February 1830, the Lake 

 of Constance was frozen over: a phenomenon of very rare occurrence. 

 Scheigger Seidel, as above, p. 3C. 5000 feet from the shore, the temperature 

 of the air was 72°.50 ; 2000 feet from it, — 15°.25 ; that of the water itself, near 

 the ice, varied on diflferent days from 32° to + 33°.35. Atler the 31st Janu- 

 ary the lake froze almost entirely over, a small circle only, opposite to the 

 Friedrichshafen, where the lake has the greatest depth, remaining open, but 

 covered with islands of ice. Up to the end of January the thickness of the 

 ice near Uttweil was found to be : 



10,000 feet distant from the shore, . CJ inches. 



16,000 ... ... .. . 5 



20,000 ... .. ... . 4 



30,000 ... ... ... . 34 



If, at the commencement of the frost, the surface of the lake becomes 

 gradually cooled down, and thus gives rise to descending currents of water, 

 which continue until the temperature of the surface reaches 38°.75 ; tliis period 

 will naturally happen later in the deeper parts, than where the depth is 

 less considerable ; for there is a greater quantity of water, of a temperature 

 above 38°.7o in those parts, and so long as water above 38°.75 exists below, 

 the descending currents will continue to take place. So that the deeper the 

 lake is, the later will its surface be reduced to 38°.75, and consequently also to 

 the freezing point. If the depth of the lake should be found to increase in the 



C C 2 



