430 ON THE FORMATION OF ICE 



ductors. Ice, being specifically lighter than water, always rises and floats on 

 the surface when the ascensional force produced by its less weight has overcome 

 its adhesion to the bottom ; it sometimes even draws up with it from the bottom 

 of the water bodies still more ponderous than itself. The maximum density of 

 water not being at 0°, but at 4°. 44, causes all great masses of water which are 

 more or less tranquil, and even such as have a continuous but not whirling 

 movement, to maintain, at the bottom, a temperature above zero, even when the 

 water is at zero, or frozen on the surface. In winter, at a temperature of — 11 

 and — 15 degrees, the reservoir near the forge of Niederbronn, w^hich is but a 

 metre in depth, is covered with a stratum of ice 25 centimetres thick, and yet 

 the water which flows from it has a temperature of + 3 degrees. 



Owing to this admirable concurrence of circumstances, great masses 6f water 

 never freeze to the bottom ; and even when ice does form at the bottom of Avater, 

 it eventually becomes detached and floats on the surface. If it were otherwise, 

 all our seas, lakes, and great rivers would be but masses of ice which would never 

 be thoroughly melted. Yet we see also that whenever Vv'ater becomes cooled 

 to 0°, and finds a bottom also cooled to 0°, it congeals at the bottom as well as 

 at the surface It is necessary, then, in order that ice may be produced at the 

 bottom of water that the latter should be in motion, so that its inferior strata 

 shall be cooled to 0^ ; that this cold w^ater shall descend to the bottom of the 

 stream; that it shall chill the walls, and finally encounter, in the course of the 

 movement, some point of repose where its force of adhesion, its capacity of 

 crystallization, maybe exerted. 



To explain this apparent contradiction : A foreign bod}^ an obstacle placed in 

 the midst of a current of water, produces thei'e two different eftects ; on one hand, 

 it changes the direction of the liquid which strikes it, and gives to that liquid a 

 movement of rotation sufficiently strong sometimes to form vortices ; on the other 

 hand, the molecules of the liquid -which happen to be immediately behind the 

 obstacle pass into a state of repose, and there are points where they become 

 stationary and motionless. The conditions here are suitable for the formation of 

 ice at the bottom of rivers. The whirling movement produced by the obstacles 

 brings the w-ater cooled to 0° to the bottom of the bed of the river, and thus 

 chills the w^alls. Hence the molecules of water, nearly motionless behind the 

 obstacle, can there exercise their force of adhesion and crystallization. But, in 

 order to produce these effects, there must be cold both intense and of some 

 duration. 



The influence exerted by these obstacles is evident in the different experiments 

 ■ w^hich I have related; "we recognize it in the slight, asperities of the unplaned 

 and the roughened boards of Lenke, in the pebbles of the Rhine as observed by 

 Fargeaud, and the piers of the bridge of the Aar described by Hugh 



With Arago, then, I attribute the formation of ice at the bottom of water prin- 

 cipally to the obstacles which occur in the current ; but, in my view, tliese ob- 

 stacles are not solely resting points for the crystals, but they serve, on the one 

 hand, to augment the movement of rotation, the vortiginous movement by which 

 the Avater at a temperature of 0° is made to descend to the bottom of the river; 

 and, on the other, hand, they create stationary points in the midst of the move- 

 ment, where the crystallizing force can exert itself. 



I have satisfactorily established the influence of these foreign bodies in the 

 canal which supplies the works at Zinsweiler. During the winter of 1829 ice was 

 formed beneath the water wherever there were large stones, roots, or branches 

 of trees which were immersed in the canal. I caused the formation of this ice 

 to cease almost entirely by having those extraneous bodies removed. Hence the 

 propriety of removing, at least during great frosts, and as far as possible, all 

 grates and bars of iron from the neighborhood of canal locks and flood gates ; all 

 bodies, in short, which tend to communicate an eddying motion to the water. 



Upon the whole, then, it may be said : I. That water which is exposed to a 



