4: r Peirce — Magnetic Properties at High Excitations. 



the test piece. Usually under strong magnetic excitation, 

 between the jaws of the yoke, there was a sensible leakage of 

 lines of induction through the surface of the specimen into the 

 air, and the field in the air about the rod was far from uniform 

 in any available portion. We found eventually, however, that 

 if a piece of the rod of about 80 millimeters free length, with 

 tapered ends, was inserted into holes in the ends of the conical 

 jaws represented in figure 3, the lines of force in the air just 

 about the specimen near its center were for a considerable 

 distance practically parallel to the axis of the rod and that the 

 value of II in the air in this region was sensibly equal to the 

 value of the same quantity in the rod. 



Fig. 3. 



After a specimen of this standard length had been accurately 

 fitted to the jaws by Mr. Thompson, the central portion of the 

 iron rod was given a very thin coat of shellac varnish and two 

 test coils, each consisting of twenty turns of very fine well 

 insulated wire, were wound side by side in a single layer over 

 the rod and these extended over rather more than a centimeter 

 of the length of the specimen near its center. These coils were 

 first tested against each other to find out whether they were 

 practically alike, and then — if this condition was satisfied — both 

 together in series formed the inner test coil (K). The outer 

 test coil (L) was wound in a single layer on a very thin 

 shell of boxwood which had been seasoning for many years. 

 After corrections had been made for the thickness of the wire 

 of the test coils and of its insulation, it was possible .to com- 

 pute from the measured change of induction flux through 



