10 EXPANSION OF LIQUIDS. 



point of the liquid, the anomaly disappears. This is the reason 

 why the interesting fact we have been discussing cannot be 

 observed in the case of sea water. 



There is a solid body which presents the only other known 

 parallel case of progressive contraction by heat ; this is Rose's 

 fusible metal, which is an alloy of 



2 parts by weight of Bismuth 



1 part Lead 



1 yy 55 y> ?? 1m. 



A bar of this metal expands progressively, like other bodies, 

 till it attains the temperature of 111; it then rapidly contracts 

 by the continued addition of heat, and at 156 attains its 

 maximum density, occupying less space than it does at the 

 freezing point of water. It afterwards progressively expands 

 melting at 201. It may be remarked, however, of this body, 

 that it is a chemical compound, of a kind in which a change of 

 constitution is very likely to occur from a change in temperature j 

 and that it cannot, therefore, be fairly compared with water. 



The dilatation which water undergoes below 40 has been 

 supposed to be connected with its sudden increase of volume 

 in freezing, for ice is specifically lighter than water, in the 

 proportion of 92 to 100. The water, it is said, may begin to 

 pass partially into the solid form at 40, although the change is 

 not complete till the temperature sinks to 32. But such an 

 assumption is altogether gratuitous, and improbable in the 

 extreme. 



The extraordinary irregularity in the dilatation of water by 

 heat is not only curious in itself, but also of the utmost 

 consequence in the economy of nature. When the cold season 

 sets in, the surface of our rivers and lakes is cooled by the con- 

 tact of the cold air and other causes. The superficial water so 

 cooled, sinks and gives place to warmer water from below, 

 which, chilled in its turn, sinks in like manner. The progress 

 of cooling in the lake goes on with considerable rapidity, so long 

 as the cold water descends and exposes that not hitherto cooled. 

 But this circulation, which accelerates the cooling of a mass of 

 water in so extraordinary a degree, ceases entirely when the 

 whole water has been cooled down to the temperature of 40, 

 which is still 8 degrees above the freezing point. Thereafter 

 the chilled surface water expands as it loses its heat, and re- 

 mains on the top, from its lightness, while the cold is very 



