Resistance of certain Amahjams with Temperature. 455 



iron and nickel, the two most magnetic metals, should lose 

 their most characteristic property when alloyed together, and 

 that this should be restored by taking them to low tempera- 

 tures or by mechanical actions, such as hammering or turning, 

 hence M. Guillaume concludes that in the non-magnetic 

 state we have a true chemical combination of iron and nickel, 

 which would account for the absence of the characteristic 

 property of each metal when uncombined, and that we render 

 the alloy magnetic by breaking down this combination by 

 some means or other. Hence in the magnetic state we have 

 uncombined iron and nickel with probably a compound also 

 of the two. In the case of the reversible alloys he considers 

 that any dissociation which has taken place through cooling 

 disappears nearly entirely when the temperature is again 

 raised to the same point. It would appear to follow from 

 this that these alloys can take up a stable state at any tem- 

 perature, but the irreversible alloys are in unstable equilibrium 

 when they are nonmagnetic at ordinary temperatures, and 

 hence when dissociation has once started it continues until a 

 stable condition is reached. 



Such is the theory that M. Guillaume advances to account 

 for the changes, briefly given above, in the magnetic condition 

 of the nickel-steels as well as for their variations in volume, 

 and it is seen that it is very similar to the one given in the 

 earlier part of this section to account for the changes in the 

 electrical resistance of zinc amalgams, if we take into account 

 that mercury may exist in the liquid condition between the 

 crystals of the amalgam, and so would possess the power of 

 dissolving zinc or zinc-mercury compounds, and hence con- 

 verting what might be an evolution into an absorption of 

 heat. For a combination of zinc and mercury with solution 

 of this in excess of mercury accounts for the absorption of 

 heat at 36°, a compound being formed which is fairly stable, 

 or in which the dissociation brought about by cooling is slow, 

 thus explaining the non-reversibility of this part of the curve. 

 At a temperature near 70° further combination and solution 

 require a further absorption of heat. As is seen from curve E 

 (fig. 3) the compound then formed is reversible for a consider- 

 able range of temperature, but whether it partially dissociates 

 on cooling and the dissociated met ds recombine on heating, as 

 Guillaume considers to take place in the reversible nickel- 

 steels, or whether the compound persists unaltered down to a 

 temperature near 25° and then begins to dissociate, cannot be 

 known certainly from the experiments, although the abrupt 

 manner in which the resistance changes near this temperature 

 would favour the latter view, sinee it seems most probable that 



