776 SUMMARY OF CUKRENT RESEARCHES RELATING TO 



and ductility when submitted to liquid ail temperatures, and the effect 

 of nickel upon iron when alloyed with it is to diminish the tendency of 

 the latter metal to become brittle at low temperatures. Microscopic 

 examination of etched specimens at - 1H2° C. did not give any indica- 

 tions of changes in micro-structure caused by the low temperature. An 

 exceUent feature of the paper is the comprehensive bibliography appended. 



In the discussion on this paper W. F. Barrett gave details of the 

 singula]' electric, magnetic, and thermo-electric properties of Hadfield's 

 iron-manganese-nickel alloy. H. le Chatelier and L. Dumas advanced 

 hypotheses explaining the great differences in the effect of low tempera- 

 tures on different iron alloys. F. Osmond disputed Hadfield's conclusions 

 as to the allotropic theory. 



The Types of Structure and the Critical Ranges on Heating and 

 Cooling of High Speed Tool Steels under Varying Thermal Treat- 

 ment.* — H. C. H. Carpenter has obtained cooling and heating curves of 

 16 specimens of steel, containing one, two, or three of the alloy metals 

 chromium, tungsten, and molybdenum in varying percentages. The 

 carbon varied from 0*25 p.c. to 1*81 p.c. For iron-chromium-carbon 

 alloys the author concludes that (1) the initial temperature from which 

 the metal is cooled is almost without influence on the position of the 

 critical point, and (2) increase of chromium tends to raise the critical 

 point. He also considers that, contrary to the widely accepted belief,, 

 the presence of chromium hastens instead of retarding the transformation 

 of hardening carbon into annealing carbon during cooling. The action 

 of tungsten and molybdenum in high-speed steels is to hinder or prevent 

 the changes which result in a softening of the alloy, and to impart a 

 high resistance to tempering. The steels examined, when cooled from 

 temperatures not higher than 000° C, pass through a critical change at 

 about 700° C. If the initial temperature is raised, the same rate of 

 cooling being maintained, the critical change is usually split into two or 

 more parts and spread over a range of temperature from 700° to 300° or 

 400° C, or even lower. Molybdenum is more active than tungsten in 

 promoting this split. When suitably treated, the alloys useful as high- 

 speed steels have a polyhedral or " austenitic " structure. 



Heat Treatment and Fatigue of Steel.f — F. Rogers has carried 

 out a large number of mechanical tests (tensile and fatigue) on samples 

 of three steels containing respectively 0"14 p.c, 0"27 p.c, and 0*82 p.c 

 carbon, heat treated in different ways. The alternating stress machine 

 was of the Wohler type, the fatigue tests carried out on it exhibited 

 great irregularities. The author concludes that overheating lowers the 

 elastic limit greatly, while increasing Young's modulus, these two effects 

 both tending to reduce the resilience of the steel enormously. Steel 

 fatigued beyond a certain limit cannot be restored by heat treatment 

 alone. Microscopic examination of polished and strained specimens 

 demonstrated that fatigue cracks tend to select a path through ferrite. 



The Elastic Properties of Steel at High Temperatures.^- B. 

 Hopkinson and F. Rogers have found that with rise of temperature, up 



* Journ. Iron and Steel Inst., lxvii. (1905) pp. 433-73 (14 pis.). 



t Tom. eit.. pp. 486-94. 



X Proc. Roy. Soc, Ser. A, No. 7<! (1905) pp. 419-25 (3 figs.). 



