Uarieiiing and Tewpey'titg of Suet. 381 



revenhelefs formed it Into a fpoon, by rolling, hammering, and afterwards polifhing iu 

 It was then cxpofed to a low red heat, in a common fire, and became bliftered all over. 



The above fafts poffefs the utility which attends unfuccefsful experiments, namely, that 

 the narration may fave others from a repetition of the labour. But, philofophically fpeaking, 

 there are no new experiments which would be untuccefsful if we thoroughly underftood them. 

 I think the decay and recovery of ignition in this is a curious inftance of what feems to be a 

 general law of the congelation of fluids. It is probable that all fluids, as well as water, are 

 capable of being cooled below their freezing point, ^nd afterwards become hotter by the 

 cfcape of latent heat when they congeal. Thus water cooled below 32 degrees is fuddenly 

 railed to that temperature the inftant ice is formed, becaufe the ice gives out the diffei-ence 

 of the heat which was latent in the fluid ftate. And fo it appears to have happened with, 

 the fluid metallic compound in thefe experiments. As its temperature was diminifiied, it 

 became lefs luminous or ignited ; but, at the inftant of congelation, that portion of heat which 

 had been employed in maintaining the fluid ftate, was extricated, and became employed in 

 raifing the temperature. 1 he effeifb of this incrcafe was feen in the greater emilTion of light, 

 and the boiling of the nitre. 



2. Tempering of Steel. 

 INSTRUMENTS of fceel are required to be hard, in order that they mav penetrate and 

 ^divide the fubftances intended to be cut; and tenacious, that they may not break during the 

 operation. The hardeft fteel is the moft brittle ; for which reafon, though hardnefs would 

 in every cafe be a defirable quality, yet, for the fake of tenacity, it is in many inftances 

 neceffary to diminifh it. We fee, therefore, that there muft be a precife mean between 

 too foft and too brittle, which will be beft fuited for the refp&Slive purpofes to be ac- 

 complilhed. A fpiing muft be tenacious, and need not be very hard. A knife for cutting 

 leather, and foft fubftances, muft be fomewhat harder than a fpring. Pen-knives and 

 razors muft be ftill harder ; and files and tools for working metal muft be hardeft of all, 

 though even in thefe care muft be taken not to deftroy their tenacity by making them too 

 hard. Steel is hardened by ignition, and fubfequent plunging in water. The ch'ef art bf 

 this procefs confifts in covering the fteel with fome mixture, which fliall prevent its bein? 

 degraded in the fire to the ftate of iron. The file-makers ufe the grounds of beer mixed 

 with common fait. Others ufe the cementing mixture. No greater heat is to be ufed 

 with any fteel, than by experience is found fufficient to produce hardnefs at leaft equal to 

 that of a file. More heat would render the grain coarfe and open. Urine is thought to be 

 better for quenching the fteel than water, probably becaufe it may be a better conductor of 

 heat, and perhaps on account of its phofplioric ingredient, which is now with juftice fup- 

 pofed • to be an eflcntial part of fteel. When fteel is not intended to pofll'fs the utmoft 

 hardnefs, it is after«'ar<ls foftcned by the application of a lower degree of heat. Thii 

 operation is called tempering. 'I'he greatcft difticulty confifts in applying the proper de- 

 gree of heat uniformly over the whole mafs. '1 he cpmmon method is to judge by the 

 colour afFumcd by the clean furface of the fteel when thus heated. The lieat in'.;y be ap- 

 plied by the fire, or a pan of charcoal, or t)ie flame of a candle or lamp, or by laying the 



' riiilufopliic.i! Journai. 



piece- 



