198 WRITINGb OF JOSEPH HENRY. [1855- 



which has been expelled from the higher stratum will be 

 caught by that next below, and in this way we shall have a 

 series of air-bubbles extending downwards to the surface of 

 the unfrozen water. 



Accustomed as we are to see bubbles of air rise in the 

 water, it would appear at first sight that the bubbles seen in 

 ice come up from the water below; but from actual observa- 

 tion in the manner we have described, it is clearly proved 

 that the bubbles are composed of air which had been ab- 

 sorbed at the surface of the water and expelled downward 

 from stratum to stratum in the process of freezing. 



The ice then over a lake or pond consists of crystallized 

 water, of which the axis of crystallization is at right angles 

 to the surface and the principal cleavage in the same direc- 

 tion. It results from this that in the thawing of the ice in 

 spring it tends to resolve itself into innumerable prismatic 

 crystals at right angles to the surface, and is liable to be dis- 

 integrated by a strong wind in a single night, thus produc- 

 ing the phenomena of a sudden disappearance of ice over a 

 large surface, a fact which has been erroneously attributed 

 to its sinking, an evident impossibility, since the minutest 

 portion of crystallized water is specifically lighter than the 

 same substance in a liquid form. General Totten several 

 years ago arrived at the same conclusion as to the sudden 

 disappearance of ice which I have demonstrated in the ex- 

 periments before mentioned. 



Ice before it tends to give way becomes pervious to water, 

 which is readily transmitted through the interstices of the 

 crystals; hence those who are accustomed to travel on frozen 

 lakes or rivers are aware of the fact that so long as the water 

 of the melted snow does not pass through the surface of the 

 ice underneath, it is safe and in a sound condition, though 

 we must be careful not to confound this water with that 

 forced up by hydrostatic pressure from below, on account of 

 the bending downwards of the whole field. 



A simple method has been proposed for determining the 

 relative severity of different winters, b}' observing the thick- 

 ness of ice. For this purpose a shallow vessel of water is 



