PEIRCE. — MAGNETIZATION IN IRON. 137 



finall;s' the I corresponding to // = 2600 fell to about 1700. One 

 specimen of wrought iron which we have annealed a great number of 

 times shows no permanent change in permeability although at one 

 stage this fell by more than one per cent temporarily^ and was restored 

 by the next annealing. 



According to my experience during the last few years with a good 

 many pieces of so called "Norway Iron," about one specimen in three 

 of those bought without care in the open market may be expected to 

 have a specific magnetism considerably above 1700. Different por- 

 tions of the same large bar may have very different permeabilities, 

 however, as one may readily believe after an examination of a series 

 of micrographs which always show a considerable amount of slag, 

 I believe that an occasional small piece such as would be used for an 

 isthmus might be found to have a specific magnetism three or four 

 percent above the best value to be found in a bar. I have myself 

 encountered two isthmuses which ga\'e 1790 and 1751 respectively^ 

 in spite of my best efforts to reduce what seemed to me at the time 

 impossibly large values. Some small specimens used by other ob- 

 servers have shown even greater \alues than this. In the case of a 

 rod a meter long and twehe millimeters in diameter, however, I have 

 never found an average value much above 1740. 



Much of the wrought iron to be had in the market, though very- 

 useful to blacksmiths, contains such an amount of slag that the con- 

 tinuity of the metal is seriously affected and the permeability of the 

 mass is not very high. Such are the specimens of "Farnley Iron," 

 marked here "F," the "Taylor Iron" and the "Best Refined Iron," 

 which show low values of / in moderate fields. It is possible to get in 

 the open market, "Norway Iron" of great purity. One specimen 

 which I used showed, upon analysis, no nickel, cobalt, manganese or 

 tungsten. It contained less than 0.03% of carbon, less than 0.047% 

 of phosphorus, less than 0.03% of silicon and less than 0.003% of 

 sulphur. This however does not compare in purity with the " Ameri- 

 can Ingot Iron, " which contains less than 0.03% of impurities all told, 

 and shows a very remarkal)le micrograph. 



Our specimens of this iron were very kindly furnished by Doctor 

 Percy W. Bridgman, who has been using this material in some of his 

 experiments upon the behavior of metals under very high pressures. 

 Plate II shows micrographs of two pieces. The first was in the nor- 

 mal state: the second had been exposed by Dr. Bridgman to a 

 hydrostatic pressure of 17000 atmospheres for about 16 hours! The 

 magnification is 120 diameters. 



