﻿effect of Temper, in case of Steel. 283 



amount long before the precipitation of carbon is perceptible or 

 even when no appreciable precipitation occurs. If we regard c 

 and e and e' correlated, then we have in hand an example of 

 an exceedingly remarkable decomposition, which may be 

 regarded incipient in hard steel even at ordinary temperatures, 

 which is a certainly perceptible occurrence after annealing at 

 only 100°, and which becomes more and more definitely 

 marked and distinct as the temperature and the time of anneal- 

 ing increase. The anomalous character of this species of 

 decomposition when occurring in a rigid solid we have already 

 fully pointed oat.* To obtain further information it is neces- 

 sary to investigate the relations of the variable c minutely. 

 Finally, the critical difference between the thermo-electric and 

 the hydro-electric behavior of steel is well shown by construct- 

 ing h as a function of e'. 



We are not at present prepared to discuss the general ques- 

 tion further. Such discussion, to be satisfactory, presupposes 

 a knowledge of the relations of the variables in the table for 

 steel identical in kind and temper, exposed to the action of 

 temperature under identical conditions — a desideratum which 

 the tabulated treatment and the results do not, as yet, fully 

 supply. We therefore merely remark that if we had come upon 

 the hydro-electric results earlier, we should not have considered 

 it safe to dwell upon the exclusive importance of the strain- 

 effect during the first phase of the phenomena of annealing. 

 With this assertion we do not wish to imply that the views 

 of the preceding papers are to be modified. 

 Washington — Prague, June, 1886. 



* "On endeavoring to use the chemical hypothesis of tempering to account for 

 the phenomena of annealing, we at once encounter serious difficulties. On the 

 basis of this theory there must exist a fixed ratio of combined to uncombined 

 carbon for each temperature of the annealing bath. Moreover, for a given tem- 

 perature this ratio must be approached gradually (asymptotically) as time of expo- 

 sure is prolonged indefinitely; for different temperatures it must decrease as 

 temperature increases indefinitely, until finally a minimum value wholly inde- 

 pendent of temperature is asymptotically reached. The minimum of the ratio of 

 combined to uncombined carbon need not, of course, be zero. The last phase of 

 this decomposition is in many respects similar to the phenomena of dissociation. 

 The resemblance can, however, only be apparent, since continuous dissociation 

 has not been observed except in the case of gases. At least, to our knowledge, 

 there are no chemical examples in which solids are found to dissociate in accord- 

 ance with the laws to which gases, in virtue of their physical state, must conform, 

 and for this reason the explanation of the phenomena of anuealing given by the 

 chemical theory is remote and forced, and to be discarded. We gain no more by 

 adopting Matthiessen's hypothesis, which considers all iron-carburets as solidified 

 more or less thorough solutions of carbon in iron." — Bull. U. S. G. S., No. 14, 

 p. 94, 1885. 



