2B 
MR. J. MUIR ON THE TEMPERING OF IRON 
have been expected at 59|- tons liad only restoration of elasticity been effected by 
tlie heating to slightly over 300° C. Another specimen of the same steel also 
showed that 350° C. was not sufficiently high a temperature to effect any softening 
of the material when it was in a condition having an elastic range to 52 tons 
per sq. inch. A temperature of about 300° C. may therefore be taken as the lowest 
which could be shown with the steel in question to produce tempering from the 
condition of hardness produced by tensile overstrain. 
It is further shown, by a comparison of Cffirves 9 and 11 and of Curves 12 to 16 
in Diagram 6, that a definite annealing temperature corresjDonds to a fairly definite 
position of the yield-point. This simple relation, however, is not shown to have held 
throughout all the experiments of Diagram 6. Curve No. 5 shows that 360” C. 
has sufficed to lower the yield-point to 47 tons per sq. inch, while in Curve No. 10 
no yieldqjoint is shown iqj to the stress of 51 tons, although the material had 
previously been heated to 400° C. 
Further experiments were made to test how far a definite annealing temperature- 
corresponded to a definite position of the yield-point. A specimen of the same steel 
as employed in the experiments described above was primarily annealed by being 
heated to 780° C., and slowly cooled, and was then overstrained by simply loading to 
tlie primary yield-point, which occurred at 29 tons per sq. incli. The specimen was 
then heated to 375° C., and after cooling loaded to 35 tons per sq. inch without a 
yield-point being passed. The yield-point, raised in conse(|uence of recovery from 
overstrain, would have been expected at about 36 tons per sq. inch. The specimen 
was then heated successively to 400, 450, 500, and 550° C., aud still the yield-point 
was found to l)e above 35 tons of stress. A temperature of 580° C. was next tried, 
and on testing the specimen after cooling, a yield-point was obtained at 33 tons per 
sq. inch. This result with a specimen which had been but slightly hardened by 
tensile overstrain, is practically in agreement witli tbe results shown in Diagram 6, 
where after severe overstraining, and some tempering from tlie condition of hardness, 
a temperature of 600° C. is sliown to have brought the yield-point to about 35 tons 
per sq. incli. 
Anotlier annealed specimen of the same steel rod was largely overstrained by 
carrying the |)rimary loading far beyond the yield-point, which occurred at slightly 
over 28 tons per sq. inch. The load was steadily increased until a stress of 38 tons 
per sq. inch of original section was attained. This corresponded to a stress of about 
40 tons on the actual reduced section of the bar. The extension produced by this 
loading was 0’22 of an inch on 4 inches of length, llecovery of elasticity was then 
efiected by heating the specimen to 300° C. When cooled and tested, the specimen 
was found to bear a load of 44 tons per sq. inch without yielding. The specimen 
was then heated to 400” C., and after cooling, a load of 44 tons per sq. inch was 
again applied without a yield-point being passed. A temperature of 500° C. was 
next tried, and it was observed on testing the specimen that large yielding began 
