PART I. : TEMPERATURE OBSERVATIONS IN THE MADUSEE, POMERANIA. 625 



compartments, the temperature changes in the upper layer being directly due to wind 

 currents and other meteorological factors, while below the discontinuity the changes are 

 due to a seiche or oscillation. This is essentially true. A glance at fig. 2 will show 

 it true in the Madiisee, for while the surface isotherms are subject to aperiodic fluctuations, 

 the lower isotherms fluctuate periodically. Except in high storms the wind-produced 

 currents are only felt to a limited depth. Of course an alteration in the disposition of 

 isotherms in the surface layer alters the density distribution not only in the surface 

 layer but all over the lake, and so the effect of wind currents is indirectly felt, and it 

 is indeed these alterations in the density distribution of the surface layer which cause 

 the temperature seiche. The wind currents seldom penetrate below the discontinuity.. 

 The temperature oscillations and other waves at the layer of discontinuity are the most 

 important phenomena to be observed and discussed. It is believed from experimental 

 observations that there are often waves propagated along the surface of separation 

 between the warm and cold water in a lake, and it should be possible to detect these by 

 direct observation. 



§ 13. Before passing on to the theory of the temperature oscillations, there are one 

 or two points about the temperatures observed which call for attention. At Werben, 

 where Professor Halbfass observed, it was common to find that at the bottom of the 

 lake there was a rise in temperature of '1° or '2° C. Similar inversions of temperature 

 were not observed at Seelow or Moritzfelde, though they have been recorded in other 

 lakes where there is an accumulation of decomposing organic matter at the bottom. 

 Observations just at the bottom were not taken so often at Seelow and Moritzfelde as at 

 Werben, but when they were taken no inversion was observed. Possible reasons for 

 the presence of inversions at Werben and their absence at the other stations are : — 

 (l) At Werben there is a very gradual shelving of the lake bottom, and decomposing 

 matter may accumulate there more quickly than at Moritzfelde, where the slope of the 

 bottom is steeper ; (2) currents may prevent accumulation of organic matter on the 

 relatively steep slopes at Seelow and Moritzfelde ; (3) at Seelow there is a stream 

 entering the lake which brings down inorganic matter, and the muds there are of a 

 different character from other parts of the lake : this stream may carry a supply of 

 oxygen to the bottom of the lake and favour rapid decomposition ; (4) the observations 

 at Seelow were made in deeper water than at Werben, and deposits of organic matter 

 would therefore probably be less. 



§ 14. Elsewhere* attention has been drawn to the very rapid temperature changes 

 which are found at the discontinuity layer. An example of this was found at Moritzfelde 

 on 12th August. The observations on that date were so erratic that some of them 

 (those at noon) were omitted in drawing the diagram in Plate I. The following are 

 the observations made on that day t : — 



* Trans. Roy. Soc. Edin., xlv. (ii.) p. 430. 



t Hours are numbered continuously from midnight to midnight. Figures appearing after the point denote 

 minutes after the hour. 



