406 On the Formation of Bottom Ice. 



the ice mass appears quickly to increase in size, and when its 

 "body becomes considerable, it rises to the surface of the water, 

 often bringing up with it stones and other objects, which 

 plainly proves that it has been formed on the bottom. Some- 

 times, however, the ice remains fast on the bottom, and gra- 

 dually grows to a large mass till it reaches the top surface of 

 the water. As far as we at present know, this bottom ice 

 forms only in the beginning of winter — never in the depths of 

 winter or in the spring, when the surface of the water is covered 

 with a tolerably thick coating of ice. 



This phenomenon has often been observed in Sweden, as 

 well as in other lands. The blocking up of the streams at 

 Motalla and Norkopping is not uncommon, and it has been 

 observed that this mass of bottom ice consists of innumerable 

 small oblong plates piled one above the other. In many of the 

 streams which run through the iron districts of Wermeland 

 this bottom ice, or, as it is called here, l ' krafis," often forms 

 in the beginning of winter, and dams up the whole stream ; 

 but it is occasionally hindered by stones and other inequalities 

 being cleared from the bed of the river, so that the water can 

 flow in a smooth, unbroken current. 



Water which is at a temperature below the freezing point 

 is brought to freezing through shaking, but this only takes 

 place where the particles which lie close together receive an 

 uneven motion, so that their relative momentum is disturbed. 

 If it happens during the shaking that the relative position of 

 the particles to each other is not disturbed, no freezing takes 

 place. This has been proved by experiment with a glass cylin- 

 der ; and moreover, a globe of glass, which is filled with water 

 below freezing point, can be shaken to any degree without the 

 water freezing into ice; consequently, the upper layer of a 

 mass of water in a river can be at a temperature under freezing 

 point ; yet, on account of its even motion, it will not become 

 ice, and can even rush down a waterfall to the very bottom of 

 the river, without any very great portion of it becoming frozen 

 together. On the bottom of the river, however, this continual 

 even motion is hindered by upright objects, which cause a 

 whirling motion, and which, through their contact with the 

 water, promote its congealment into ice. In order, in some 

 measure, to hinder the formation of bottom ice, which is often 

 very pernicious, the bed of the stream should be continually 

 cleaned, and all inequalities removed. 



We can hardly, however, consider that this phenomenon is 

 even sufficiently understood or properly studied, so that any 

 new observations respecting the formation of bottom ice will 

 be welcomed; and those which follow on the formation of ice in 

 the Cattegat are very interesting, as proving a fact which is 



