376 
Cast Steel . 
Grs. 
Loss of weight in this experiment 58 
Exp . XIV. Swedish iron - - 1372 
Charcoal T-^h? or 14 grs. 
The button obtained weighed - 1312 
Lost in the fusion 60 
equal to ~-th part the original weight of the iron. The 
22 ? 
surface of this button was smooth, without crystallization. 
The under surface rough, and possessed of one large pit 
in the centre, faintly marked with the usual crystalline ap- 
pearance. The fracture presented regular light blue grains 
distinct, and more prominent than No* XIII. One-half 
of the button was drawn into a neat square bar, and prov- 
ed steel of an excellent quality. One end of the bar be- 
ing loose and shaled, welded tolerably well, and hardened 
afterwards with a low heat. In appreciating the quality 
of this result, it appeared to be that kind of steel suitable' 
for penknives, razors, &c. possessed of neither the ex- 
tremes of hardness nor of softnesss. 
Beyond the proportion of part of charcoal to iron, 
I continued the experiment till the proportion was reduc- 
ed to part. It would appear tedious to detail these 
experiments, the most interesting being already minutely 
described. In the same progressive manner, by dimi- 
nishing the dose of carbon, the metallic result approached 
more and more to the softness of malleable iron, though 
by no means possessed of all its properties. In this se- 
ries of experiments, iron presented with ^~th part its 
weight of charcoal was found to form very soft steel fit 
for making scissars, See. which, in a good workman’s 
hands, would have doubled, welded, and formed a very 
perfect point, afterwards hardening so as to display a beau- 
tiful close break of steel. By using the following precau- 
tion, it was even found capable of welding perfectly to 
