533 WATER. 



tains it.* Lastly, M. Hermann has even determined qDantftatively, 

 carbonate of ammonia in the ferruginous waters of a turf- pit. The 

 water of the Nile is not exempt from it, judging at least from the 

 analysis of its mud. According to Regnault, 100 parts of this mud 

 dried in the air contain if 



Chloride of sodium, sulphate of soda, and 



carbonate of ammonia ) 



Organic matter 9 



Water 10 



Oxide of iron 6 



Silica 4 



Alumina 48 



Carbonate of lime 18 



Carbonate of magnesia 4 



100 



The beautiful synthetic experiments of M. Dumas demonstrate 

 that water is formed of : 



Oxygen 88.89 



Hydrogen 11.11 



When pure, it boils at a temperature of 212° F. under a barome 

 trie pressure of 30 (29.921) inches. It congeals at 32° F. 



All natural bodies dilate, augment in volume, by the action if 

 heat, and contract under diminution of temperature. Water is 

 amenable to this law between rather wide limits ; it deviates, how- 

 ever, and presents an anomaly as it approaches congelation. As 

 with all liquids, the density of water gradually increases in propor- 

 tion as it cools, until its temperature Is 39°. 38 F. Setting out from 

 this point the density diminishes, the liquid dilates more and more, 

 so that at 32° it occupies nearly the same volume that it did at 49°. 

 From this remarkable property, it results that during the most in- 

 tense cold the stagnant water which covers the meadows rarely at- 

 tains a lower temperature than 39", whereby the organs of plants 

 suffer no damage. 



Let us suppose, in fact, that at the beginning of winter a sheet of 

 stagnant water has a temperature of about 54°; in proportion as the 

 liquid at the surface cools, it becomes denser, descends, and is im- 

 mediately replaced by inferior layers, which rise in the ratio of their 

 less density ; but tliese new superior layers, subjected to the same 

 refrigerating cause, contract and descend alternately. There is 

 then established in the fluid molecules, movements of ascension and 

 descent, of which the result is the cooling of the entire mass. Let 

 us now admit that in virtue of this continued mingling of the cooled 

 superior layers with those below, the temperature of the sheet of 

 water is lowered to 39\38 : at this degree of the thermometer, the 

 water acquires its maximum density ; in parting with its heal it not 

 only contracts no more, but becomes lighter. If then a body of 

 stagnant water at a temperature of 39° is exposed to the chilling 

 action of tho atmosphere, the superior layer, far colder than the in- 

 ferior, will no longer descend, since it will become lighter as its 



* Llebig, Trait6 de Chimie, Introduction, p. cii. 

 t DMcripUon d« I'Egypt*, t. ii. p. 406. 



