

HEAT. 51 



cles of air sufficient to form foci for the melting points of ice 

 might be dissolved by the water as soon as they came in con- 

 tact with it. Be this as it may, the existence of these points 

 throughout the ice, where it gives way to the heat of the solar 

 beam, if it does not prove actual vacuous or aeriform spaces 

 to exist in ice, proves that it is not homogeneous, that its 

 structure is probably definitely crystalline, and that the 

 matter composing it is in different degrees of aggregation, so 

 that its mean specific gravity might well be less than that of 

 water. 



"We cannot examine piecemeal the ultimate structure of 

 matter, but in addition to the fact that the bodies which 

 evince this peculiarity are bodies which, when solidified, ex- 

 hibit a very marked crystalline character, there are experi- 

 ments which show that water between the point of maximum 

 density and its point of solidification polarises light circularly ; 

 showing, if these experiments be correct, a structural altera- 

 tion in water, and one analogous to that possessed by certain 

 crystalline solids, and to that possessed by water itself, where 

 it is forcibly made to assume a polarised condition by the in- 

 fluence of magnetism. 



The accuracy of these results has, however, been doubted, 

 and the experiments have not succeeded when repeated by 

 very experienced hands. Whether this be so or not, and 

 whether the above explanation of the exception to the other- 

 wise invariable effect of expansion by heat be or be not re- 

 garded as admissible, must be left to the judgment of each 

 individual who thinks upon the subject ; at all events, no 

 theory of heat yet proposed removes the difficulty, and there- 

 fore it equally opposes every other view of the phenom- 

 ena of. heat, as it does that which I have here consid- 

 ered, and which regards heat as communicable expansive 

 force. 



As certain bodies expand in freezing, and indeed, under 

 some circumstances, before they arrive at the temperature at 



