460 BELL SYSTEM TECHNICAL JOURNAL 



of a segregation such as other evidence leads me to think must occur. 

 Perhaps the reason for this is that the different constituent metals 

 are so closely related that small changes in the structure cannot be 

 detected by this means, or perhaps the reason is that the size of the 

 groups of atoms making up the constituents is too small to be detected 

 by X-ray methods. 



While the variation in the permeability and in the other magnetic 

 characteristics which could be affected by heat treatment may be 

 explained by the theory of segregation, this theory does not explain 

 why these alloys have such high permeabilities at low magnetizing 

 forces. 



Nor does this theory explain the unexpected magnetic charac- 

 teristics, such as high saturation values of induction, or the low 

 electrical resistance which characterize a large proportion of the 

 alloys in the iron-cobalt series. It has been suggested by Weiss ^ 

 that when the saturation values of an alloy are higher than they are 

 for the constituent metals, it is an indication of the existence of an 

 intermetallic compound. On this ground he has accounted for the 

 high saturation values he found for an iron-cobalt alloy containing 

 34 per cent cobalt. In our investigation, which was not carried up 

 to the high magnetizing forces used by Weiss, the 50 per cent cobalt 

 alloy gave us as high flux densities as any in the series for magnetizing 

 forces upwards of 1,500 gauss. 



It has been found in the study of intermetallic compounds that one 

 indication of their existence is a low resistivity. It is generally believed 

 that if an alloy has lower resistivity than any of its constituent metals, 

 an intermetallic compound exists. In our measurements of the 

 resistivity of the iron-cobalt series we found that the alloys with lowest 

 resistance were those containing between 25 per cent and 60 per cent 

 iron. There is a rather abrupt decrease in the resistivity as the 

 percentage of iron increases beyond 25 per cent. Beyond 50 per cent 

 iron there is a gradual increase with a maximum at about 85 per cent 

 iron. From these measurements we would conclude that if an inter- 

 metallic compound exists, it is of a higher cobalt percentage than that 

 suggested by Weiss. From our measurements of the resistivity of 

 the alloys in this series the most probable intermetallic compound 

 would be one containing approximately 66 per cent cobalt of the 

 chemical formula FeCo2. 



The data which I have presented are those for the compositions of 

 iron, nickel and cobalt. In addition to these alloys, we have studied 

 the effects of adding non-magnetic elements to numerous alloys of 



^ Transactions of the Faraday Society, Vol. 8, p. 149. 



