452 C. Barus and V. Strouhal — Strain-effect, etc. 



incidence of the resistance increments corresponding to like 

 volume increments, no matter whether the latter be the result 

 of temper or of temperature. The light which the new data 

 throw upon the internal condition of tempered steel — the tend- 

 ency to a fissured structure — shows, however, that this material 

 is not as well adapted for the quantitative evaluation of the 

 true relation between volume and resistance, as we supposed. 

 So long, moreover, as the meaning of the minimum resistance 

 and of certain carburation phenomena encountered during an- 

 nealing at red heat remains obscure, it is safest to give the said 

 relations no more than qualitative importance. The best means 

 to the end in question is probably to be sought for in the 

 compression of mercury. 



With this unimportant exception, the present results have 

 materially substantiated our earlier views at every essential 

 point. Hence we infer as before that "the annealing of steel, 

 considered physically, is at once referable to the category of 

 viscous phenomena. In the ordinary cases of viscosity meas- 

 urements, the phenomenon is evoked by sudden application of 

 stress (torsion, flexure, tension, volume-compression, etc.) under 

 conditions of constant viscosity; in the case of annealing by 

 sudden decrease of viscosity under conditions of initially con- 

 stant stress. Thermal expansion interferes with the purity of 

 these phenomena by destroying the conditions of existence of 

 the strain which accompanies hardness, and this in proportion 

 as the expansion is greater."* Again, irrespective of the mani- 

 festation of mere hardness, "the existence of the characteristic 

 strain in glass-hard steel is the cause of electrical effects 

 so enormous, that such additional effects which any change of 

 carburation may involve may be disregarded and the electrical 

 and magnetic results interpreted as due to variations in the 

 intensity of the said strain. "f 



This deduction applies of course to the first phase of the phe-. 

 nomena of annealing, since it is within these limits that the 

 strain in question is brought to vanish. With these results in 

 hand we may proceed justifiably toward a study of the ques- 

 tion, whether the conditions for the permanent retention of 

 magnetism in an iron-carburet are not the identical conditions 

 for the permanent retention of any strain. If we select the 

 temper-strain for comparison, we do it not merely because our 

 experience has familiarized us with this strain, but because of 

 the clear-cut beauty of its manifestations, and because of the 

 simplicity and pronounced character of the functions which 

 describe \\>.% 



Washington — Prague. April, 1886. 



* Bulletin, U.S. Geolog. Survey, No. 14. p. 196. f Ibid., p. 97. 



\ We desire to remark that the principal inferences of this paper have since 

 been substantiated by polariscopic evidence, and by an investigation of the 

 density of the consecutive shells of the P. R. drop. These results will be given 

 in our next paper, nearly ready. May, 1886. 



