428 ON THE FORMATION OF ICE 



from uniting with one another cnnfiisedly, in such a manner as to generate that 

 kind of spongy ice through which M. Hugi so readily thrust the oars of his 

 bateau. 



"Having reached this point, the reader will, perhaps, ask why I do not offer 

 the preceding as the complete exphnatiop of the formation of the grundcis of 

 the Germans, the bottom ice of our watermen. The following is my answer ; 



" We are still in want of observations to prove that nowhere does this sort of 

 ice show itself until the whole mass of the liquid has descended to zero. It is 

 not certain that the small needles of ice floating on the liq[uid, of which M. Knight 

 makes mention, and which may have acquired, at least at their surface, a tem- 

 perature much below zero, do not play an important part in the phenomenon^ 

 and one which I have wholly neglected ; that, for instance, of imparting cold- 

 ness to the pebbles which cover the bed of the river when the current carries 

 them down to it. Might it not be possible even that these floating filaments 

 were the principal elements of the future spongy ice ?" 



In 1836 Gay Lussac {Annates de C'kimie et de Pkysiqve, t. Ixiii) recurs to this 

 theory of the ice at the bottom of the water. He regards with doubt the first 

 explanation of Arago, and adheres rather to the hypothesis of the floating fila- 

 ments of which Arago speaks in the last place : " The spongy ice," he says, 

 "which is observed on the bed cf certain rivers of rapid current has its origin in 

 the small and innumerable flakes of ice which drift on top of such rivers in very 

 cold weather, and whose surface in contact with the air is somewhat below zero. 

 The submersion of these flakes takes place by means of the current which 

 sweeps them along in its movements. Their adhesion, whether to the extraneous 

 bodies which cover the bed of the river or to one another, is occasioned by the 

 congelation of the film of water at the points of contact, owing to the greater 

 coldness of the flakes, while their permanence at the bottom of the water, with- 

 out other enlargement than through accumulation of numbers, is explained by 

 the constancy of the temperature at zero, which itself could not exist except by 

 the efi'ect of a current sufiiciently rapid." 



In 1847 Dr. Plieninger, of Stuttgard, published in the annual report of the 

 Wurtemberg Society of Natural Sciences a note, accompanied by an account of 

 the experiments made on the Danube by M. Lenke, of Ulm. Three troughs of 

 deal plank were immersed in that river, one of which was made of smoothly 

 planed boards, the second of rough unplaned boards, while the third had been 

 rendered still more uneven by hewing. It was found that in the first no ice 

 was formed; in the second there was a deposition of minute crystals ; and in the 

 third, considerable groups and accumulations of ice. The results of these ex- 

 periments seem to me unfavorable to the views taken by Gay Lussac ; their 

 explanation will be hereafter adverted to. 



The following are the experiments which I made in 1829, and which I have 

 since, at different times, repeated. For this purpose I employed three kettles 

 used for smelting, of about one metre in diameter, which I filled with water. In 

 order to judge of the influence of extraneous bodies, I placed at the bottom of 

 one of these kettles bits of wood and cast iron ; in another a little water had been 

 allowed to congeal ; in the third there was nothing. These extraneous bodies 

 exerted no sensible influence on the congelation. At the moment of commencing 

 the experiment the circumambient air was at — 2 degrees ; it became colder 

 during the night ; the water was at 0°, and was immediately covered with films 

 of ice which crossed one another at 30, '60, and 120 degrees, forming very soon 

 an entire crust of ice over the surface. The following day I broke this crust, 

 which was from 35 to 4U millimetres in thickness ; I decanted the water from 

 the kettles, and found the bottom and sides lined with a continuous stratum of 

 ice having a thickness of 20 to 25 millimetres. The surface was smooth, ofi'ering 

 but here and there slight rugosities ; to these rugosities tufts of crystals of ice 

 were found to have attached themselves. 



