Properties y Specific Resistance, and Hardness of Steel. 361 



nexions with the galvanometer being made, the application 

 of the ice wedge to the parts near the base generated a current 

 from thin to thick through warm, to parts near the apex a 

 current in the opposite direction; finally, between these a 

 position was found at which the application of the ice wedge 

 produced no current at all. This occurred at parts about 

 1*5 millim. in diameter. The other cone gave like results. 



4. When the bases of two similar cones of the same mate- 

 rial are connected with the galvanometer and their apices 

 brought into contact, upon warming the latter, a current in one 

 direction or another will be produced — this from the fact that 

 the points are rarely equally hard. Experiment shows, how- 

 ever, that by consecutively warming parts which lie symmetri- 

 cally to the right and left of the apices in contact, currents in 

 opposite directions are the effect. Herefrom it follows that 

 these currents originate each in a single cone. 



In endeavouring to generalize from these experiments, 

 attention must be paid to the following points : — a. The 

 maximum value of T. E. H. attainable is dependent on the 

 quantity of carbon contained in the steel. The thermoelec- 

 tric difference between rods of soft and suddenly cooled 

 wrought iron can, for instance, be neglected in comparison 

 with the corresponding difference of soft and hard steel. 

 b. T. E. H. is influenced by the temperature of the rod when 

 suddenly chilled (VII. c) as well as by the time of heating, 

 the latter affecting the composition, c. By the form of the 

 piece of steel and the method of sudden cooling, the internal 

 structure of the mass being thereby modified (VII. d). 

 d. Finally, we might add that in the time elapsing between 

 the removal of the rod out of the fire and the subsequent im- 

 mersion, the loss of heat by radiation will be relatively greater 

 in the case of thin than in the case of thick rods. 



With these facts in mind, we may conclude that the maxi- 

 mum values of T. E. H. attainable by glass-hardening rods of 

 the same composition increases as the thickness diminishes ; 

 that as this dimension continues to decrease a diameter is 

 reached at which the negative effects of decarbonization are 

 equal to and finally overcome the positive effects due to dimi- 

 nution of diameter. 



c. From the results contained in Table I. there follows : 

 the hardness of steel does not increase continuously with its tem- 

 perature at the moment of sudden cooling, but at a point 

 lying in dark-red heat the glass-hard state is suddenly attained. 

 From this point on, however, the degree of glasshardness 

 (measured thermoelectrically) continues to increase with the 

 temperature. This observation conduces to the conclusion 



