288 REPORTS ON THE STATE OE" SCIENCE, — 1916. 



of a range of stress somewhat below the above-mentioned limiting raflge 

 do not cause an increment of hysteresis measurable by the same apparatus 

 as used for the tests cited in this note. 



The large increase of hysteresis due to repetitions of a range slightly 

 exceeding this limiting range cannot be regarded as increased elastic 

 hysteresis for two reasons : — 



(1) Because on subsequent application of much less ranges of stress 

 the hysteresis retains an augmented value which appears to be much 

 more than what can be regarded as elastic hysteresis, and (2) the large 

 increase of range of strain is not independent of the speed of cycle ; for, 

 as previously shown by the author^ (see also Test No. 8), a reduction 

 of frequency of cycle gives an increase of range of strain, and vice versa ; 

 whereas Rowett* has found that the area of the elastic hysteresis loop 

 is the same at low and high speeds within 5 per cent. 



At this limiting range of stress there appears to be a definite impair- 

 ment of elasticity with repetition of cycle, and the increased hysteresis 

 is most probably the coarser form of hysteresis believed to be due to 

 crystalline slipping. 



Appendix III. 



On the Fatigue Resistance of Mild Steel under Various Conditions of 

 Stress Distribution. 



By Dr. T. E. Stanton and Mr. R. G. Batson. 



The material on which the experiments described in this Report were 

 made was a special sample of mild steel procured for the Committee by 

 Dr. F. Rogers. The ordinary mechanical properties of the steel have been 

 investigated fairly completely, and the results of the tests are given in the 

 Report for 1915. It should be mentioned that the specimens used were 

 prepared from the 1-5/16" bar, and were not heat-treated before 

 testing. The results of a tensile test on the bar used give results which 

 were practically identical with those obtained by Mr. Cook (see Report 

 1915, p. 160), and were :— 



Yield Stress . . . . 13 5 tons per sq. inch. 

 Maximum Stress . . . 22-4 ,, ,, ., 



I 



% Extension (j=^3 5) . 49 



% Contraction of Area at 



Fracture .... 70-8 

 Modulus of Elasticity . . 297 x 10'^ lb. per sq. inch. 



The scheme of experiments was the determination of the fatigue 

 resistance of solid cylindrical specimens subject to rapid alternations 

 of a combined bending and twisting moment of given value and such 

 that the ratio of bending moment to twisting moment could have any 



' Proc. Boyal Soc. A, vol 92. * Proc. Boyal $oc. A, vol. 89. 



