FLOATING ISLANDS 



3°7 



Another View of Sadawga Lake, showing the main floating island (1) and three 

 smaller floating islands which have been broken off from it and have drifted ashore. 



the sphagnum plants retained vitality. Generations of these succeeding each 

 other contributed to the weight of the ring and finally pressed it down upon the 

 bottom of the pond, forming the anchored atoll of the present. 



[The essential requirements in the formation of atolls are] a small parent 

 pond, height and regularity of banks, regular and gentle slope of the bottom, 

 suitable original littoral vegetation, small lateral pressure and tension of winter 

 ice, and comparatively prompt anchoring of the bottom. 



III. Periodic Islands 



As previously explained, periodic islands may rise without the for- 

 mation of gas, but as gas usually escapes from them after they have 

 risen, irrespective of the force causing them to rise, it is a factor which 

 demands attention. Under the average floating island there is little 

 gas, owing to the loose texture of the peat. Consequently, in order to 

 have gas support islands it is necessary that an upper layer of the bog 

 be made of dense material, thus allowing the gas to escape but slowly. 

 The most favorable place for it to collect is between the layers of peat. 

 This is illustrated by experiments 2 in digging peat from a bog filled 

 with water. Holes were dug in the evening of one day and the next 

 morning they were found to be filled with peat masses in the form of 

 domes, cleft in the middle. These domes were formed of one of the 

 lower layers of the peat. The gas under this layer combined with lat- 

 eral pressure forced it up. 



Periodic islands usually rise in spring and sink in fall, owing to the 

 activity of gas-producing organisms in warm water. Some periodic 

 islands have been reported which rose for only a few days. Because of 

 the short time which most of these islands have been known, it has 

 been impossible to study them. Attention is now being given to one in 

 this country and it is to be hoped that through its study valuable in- 

 formation will be obtained regarding this type of island. 



2 Friih und Schroeter, "Die Moore der Schweiz," Beitr. geol. Schweiz 

 Geotech., ser. 3, Bern, 1904. 



