go j\vw Procefs for making 



and, upon being touched by the file, felt perceptibly harde^. 

 The fraAure ot the piece thus hardened was of an open re- 

 gular grain, approaching fomewhat, in point of colour, to 

 malleable iron. A fra6ture from the unhardened part of the 

 bar was precifely that of bar iron made in the puddling pro- 

 cefs ; a dark blueifli fibre breaking Ihort acrofs, without tear- 

 ing out in groups. But before I allowed myfelf to draw any 

 conclufion from this experiment, 1 deemed it necefl'ary to 

 repeat it as follows : 



Exp. II. The quantity of iron now operated upon Gr^ii.s. 

 was - - - - 1331 



To which were added r-3d of the fame carbonate of 

 lime, or 443 grains; i-3d of Stourbridge clay (old 



pot), 443 grain^^- . ^ . 



This mixture was arranged as in No. 1. The cru- 

 cible was introduced into the furnace, and in 30 mi- 

 nutes I judged the fafion entire. A metallic button, 

 as in No. i, was found, beneath a dark green glafs, 

 which weighed - - - 1312 



Loft in fufion 19 

 equal to y^th part of the original quantity of iron introduced. 



The glafe obtained in this experiment, upon comparing 

 fragments of equal thicknefs, was found one fhade lighter 

 than in No. i. The upper furface of the metallic button 

 exhibited a more perfeft appi^oach to cryftallization than in 

 No. T. Tiiis fuppofes a more perfect fufion had been ef- 

 fefted. The under furface contained one large pit, the in- 

 terior furface of which was ftrongly marked with a Itreaky 

 cryftallization. The fra6lure was alto brighter, and the grain 

 more diftinftly formed than in No. i. 



A bar forged from one-half of the button drew into a 

 jfliape at a low red heat. At a white heal it Itood a few 

 blows; but, in returning through the intermediate heats, 

 diffipated like hot fand. A fragment well hammered, and 

 plunged into water at a low red heat, exhibited no (hale, 

 and bent eafily over the anvil. When plunged at a bright 

 red heat, a fair proportion of fliale turned up. The piece 

 then felt fimilarly hard, under the file, with No. I. The 

 hardened fracture was open, diftini^lly granulated, and bright. 

 A frafture of t;ie unhardened part of the bar was blueilh, 

 tough, without grain ; nearly two fhades lighter in the colour 

 than No. i, .This circumftance, along with others formerly 

 mentioned, was moft probably derived from the fuperior 

 degree of fufion which the produ^ft underwent. 



The fame experiment was feveral times repealed with 



^ nearly 



