CONTEMPORARY ADVANCES IN PHYSICS 



2>ii 



the /-vs.-// curve mounts much more swiftly than does the normal 

 curve for polycrystalline iron; the first segment is very short, and the 

 second passes into the third while the field is still low. The slope of the 

 first part of the curve, that is to say the initial susceptibility, is greatest 

 when the axis of the rod is a tetragonal axis of the crystal, less if it is a 

 digonal, least if it is a trigonal axis; though the differences (Fig. 5) are 

 not great. This is sometimes expressed by saying that iron is most 

 easily magnetized along the tetragonal axis, less so along the digonal 

 and least along the trigonal. Magnetization curves consisting of three 

 or four straight lines meeting at sharp corners have been observed by 

 two of the recent students of single crystals, but not by two others; I 

 infer they are still debatable. The saturation value of /, whether it be 



1600 



1400 



1200 



1000 



100 



200 



300 

 H (GAU55) 



Fig. 5 — Initial curves of a single crystal of iron, magnetized parallel to tetragonal 

 (100), digonal (110), or trigonal (ill) directions. (After W. L. Webster.) 



attained soon or late, seems always to be about the same — another of 

 the reasons for attaching a peculiar importance to it. Honda in fact 

 obtained the value 1707, which he confronts w\t\\ the 1706 given by 

 Weiss for polycrystalline iron; but this is an agreement which looks 

 too good to be true, or at least to be significant. 



The hysteresis-loop for a single crystal is so exceedingly narrow 

 that when it is plotted on any ordinary scale, its sides are too close 

 to be distinguished. Measurements upon rods composed of many 

 crystals, the average size of which varies from rod to rod, show that 

 the area of the hysteresis-loop decreases quite steadily as this average 



