ACTION OF WATER ON IRON. 271 



46. It has been considered by most authors that ^'chilled" or 

 white cast iron contains less combined carbon than the black 

 or gray varieties ; this, however, appears to be a mistake ; yet 

 it is undoubtedly true that it does contain much less carbon or 

 graphite in a suspended or uncombined state, and that the latter 

 is mechanically expressed and locally deposited in much the 

 same way, as the author has endeavoured to show in another 

 place, — that the contemporaneous quartz veins in granite have 

 been formed from the residual quartz existing in that rock, over 

 and above that which was necessary to the atomic composition 

 of its constituent minerals. It is also analogous to the ob- 

 servation made by Pelletier respecting the combination of 

 phosphorus and silver, viz. that this metal holds more phos- 

 phorus in combination, or rather in suspension, while in fusion 

 than when solid, for at the moment of congelation of fused 

 phosphuret of silver it exudes a quantity of phosphorus, which 

 takes fire, unless the whole be suddenly cooled in water. — 

 [Ami. de Chun. vol. xiii. p. 110.) 



47. By two remarkable properties may the whole of the 

 metals be divided into as many classes, namely, those which 

 pass from the fluid state of fusion instantaneously to the solid ; 

 and those which assume the latter state by passing through an 

 intermediate condition of pastiness. 



The former property is exemplified in all the metals which 

 crystallize best, as bismuth, zinc, arsenic, &c.; and the latter in 

 potassium, sodium, iron, platina, &c. The power of being 

 welded is entirely due to this latter condition of intermediate 

 pastiness between fluidity and solidity, and hence it is properly 

 not confined merely to metals, for wax, tallow, resins, camphor, 

 caoutchouc, glass, and most vitrifactions, have strictly the 

 welding property. 



But it results from this that those bodies which can be 

 welded can scarcely ever be crystallized by fusion and slow 

 cooling, because that pasty and viscid condition they assume 

 before solidification forbids the freedom of motion to their 

 molecules, which is essential to their crystalline arrangement. 



But in "chilling" cast iron bj^ sudden cooling, time is not 

 given it to assume the viscid or pasty state; its particles are 

 compelled to pass ])er saltiim from a liquid to a crystalline solid, 

 and, e converso, by continued cementation at a temperature ap- 

 proaching its fusing point, "chilled" cast iron may be brought 

 back again to the ordinary grain of common cast iron. By this 

 instantaneous change, the crystals of iron in forming, in obe- 

 dience to the general law, reject and throw out the uncombined 

 carbon as heterogeneous, which itself also assumes the cry- 



