388 Prof. Bischoff' on the Temperature of' 



tinually exerts a cooling influence upon it, and reduces its tem- 

 perature below that which it would otherwise have from its lati- 

 tude and elevation above the sea. But since this surface is coa- 

 nected with strata of a higher temperature by means of a con- 

 ducting medium, heat must, according to the conducting power 

 of those strata, be continually communicated to the cooling mass. 

 In the case of the glaciers this heat is employed in melting the 

 snow and ice, but in lakes it is carried to the surface by the 

 ascending particles of water. Here it either assists in promoting 

 the evaporation of the water, and thus becomes latent, as in the 

 glaciers, or it raises the temperature at the surface. Whether 

 both these take place at the same time, which it is most probable 

 they do, might be ascertained by corresponding observations on 

 the temperature of the surface of a lake and of the earth in its 

 immediate vicinity. Such observations, which it would be ne- 

 cessary to continue for a whole year, in order to obtain the 

 mean, would therefore decide a very interesting question.* 



" The annual meteorological report of Wurtemberg for 1831, contains the 

 thermometrical observations in the Lake of Constance and the surrounding 

 atmosphere, made by Dr Dihlmann, in July 1831, daily at two o'clock p. sr. 

 The mean of the former is 77°-25, that of the latter 79°.02. As the observations 

 were made at the time of the daily maximum, the true mean temperature of 

 the Lake of Constance must be lower than 77°25. According to observations 

 which I made during a journey by steam to Sfrasburg, on the 19th and 20th of 

 August 1835, on the temperature of the Rhine, the greatest daily difference was 

 1°.8 to 2°25. The true mean temperature of the Lake of Constance in July 1 831, 

 may therefore be estimated at about 76°.12. But from the mean temperature 

 of the air a far greater deduction must be made. At Stutlgard, whose climate 

 must be very nearly the same as that of the neighbourhood of the Lake of 

 Constance, the mean temperature in July 1831 was C7°-32. According to this 

 the mean temperature of the Lake of Constance must in that month have sur- 

 passed that of the surrounding atmosphere by about 9°00. In the colder sea- 

 sons this difference is probably not less considerable, as the water which is 

 cooled by the air at the surface continually sinks, whilst the ascending cur- 

 rents must take place to neavl}' the same extent in winter as in summer. 

 From this it may therefore be concluded, that the mean temperature of the 

 surface of the Lake of Constance is higher than that of the soil in its vicinity. 



Ten observations on the temperature of the Lake of Constance, made by 

 Dihlmann in June, July, and August 1828 (from the|13th June to the 20th 

 August), in the afternoon and evening, gave a temperature 5°.85 higher than 

 the mean temperature of the air on those days, and only 2°.47 lower than the 

 temperature of the air in the shade, at two o'clock in the afternoon. Similar 



