110 BELL SYSTEM TECHNICAL JOURNAL 



Early in the investigations it was found that heat-treated permalloy 

 is sensitive to strain, and the routine measurements were so con- 

 ducted as to avoid this disturbing effect. Separate investigations 

 of the effects of strain upon permeability and electrical conductivity 

 in straight samples, and of the converse effects of magnetization 

 upon dimensions and conductivity were also undertaken. While 

 these studies are not yet complete it can be stated that all these effects 

 are large in comparison with the corresponding effects in hitherto 

 available magnetic materials. So long as the elastic limit of the 

 material is not exceeded the effects due to strain are reproducible and 

 disappear when the strain is relieved. The effects of magnetization, 

 however, show the expected hysteretic properties. As an example of 

 the magnitude of the effects producible it may be stated that between 

 its value in the unstrained condition and about one-tenth that value 

 the initial permeability of a heat-treated strip of certain of these 

 materials can, by the mere variation of strain, be adjusted to any 

 value we may for the moment desire. The range through which 

 the conductivity can similarly be adjusted by strain is much narrower, 

 the maximum reduction being about 2 per cent, which, however, is 

 a large effect compared with that found in other metals. 



The effect of magnetization in reducing conductivity is as much 

 as 2 per cent for fields of the order of one gauss. This makes it easy, 

 for example, to measure the earth's magnetic field to within about 1 

 per cent by finding the strength of the opposing field necessary to 

 give a permalloy strip its maximum conductivity. It will be noted 

 that the conductivity change which we have mentioned as attainable 

 by magnetization is the same as that attainable by elastic strain. 

 This is no mere coincidence, for we find that the maximum change 

 due to either cause alone is not further increased by superposition of 

 the other, although the effects of small tensions and magnetizing fields 

 are additive. This suggests, of course, that both causes ultimately 

 produce the same change in the mechanism responsible for conduction. 

 Since the effect of tension upon permeability is in some of these 

 cases so marked it seemed surprising that the only reported study 6 

 of the converse effect, that is of magnetostriction, indicated a zero 

 value within the permalloy range. It appeared advisable therefore 

 to study the magnetostriction of the series of alloys here available. 

 Preliminary results indicate that under usual conditions of experi- 

 ment, heat-treated 78.5 per cent nickel alloy exhibits larger mag- 

 netostriction than does iron. 



6 K. Honda and K. Kido, Tohoku Univ. Sci. Rep., 9, 221-232, (1920). It should be 

 noted, however, that their alloys had received different treatments than ours. 



