149 
to be so small that it is difficult to distinguish it from 
variations arising from change of temperature. 
Annealing diminishes considerably the electrical resistance 
of puddled iron wire, and has a marked effect on steel, the 
exception in the case of sample 3a being probably due to 
temperature. 
The annealed samples were cut originally from the same 
piece as the bright, and the same was the case with the 
hardened and tempered, so they are identical in chemical 
composition. 
As the electrical resistance of steel wire seemed from 
these experiments to increase very rapidly, the writer 
determined to investigate it more closely. A set of seven 
samples of cast steel wire were prepared, all of them manu- 
factured in the same way, differing only in the amount of 
carbon and other impurities. These samples were drawn to 
No. 18 J and each coil divided, one half being hardened and 
tempered, then drawn to No. 20 and tested, the other half 
of each coil was annealed and drawn one hole to No. 20, 
then tested in the bright state and afterwards part was 
annealed and tested. The eighth sample was Bessemer 
steel drawn and tempered in the same way as the others. 
The tempering was done in melted lead at a constant tem- 
perature. 
The results are given in table B. 
The electrical resistances in this table, as in A, are the 
mean of two or more experiments differing very little 
amongst themselves ; the method of experiment and appa- 
ratus was the same as used before. 
The figures denoting breaking strain, elongation and 
torsion, as in table A are the mean of two experiments in 
