106 Barrett, BROW^f & Hadpield— Ort ihe Electrical Conductivitij and 



Turning now to Plate VIII., we see from the curve for 1179 B, that 



when H is 45, B is a little over 14,000, 

 when H is 0, B is 9500, 

 when B is 0, H is 8. 



We will now proceed to give a tabulated statement of the results obtained for 

 these three magnetic properties, and will first take those carbon steels we have 

 tried. The B and H curves for this series were drawn, but are not given in the 

 plates, as there is no particular novelty about them, and are retained until the 

 series is more complete. A more detailed analysis of these and the other steels 

 is given under the respective groups in Part I. ; all that is here given is the 

 percentage of the principal element or elements added to the iron. The test rods 

 were, as already stated, 102 cms. long, and varied in diameter from 0*54 to 0'56 cm.* 



Unless otherwise stated, the determinations in all cases refer to specimens in 

 the annealed state, f 



Table II. — Carbon Steels. 



The specimen marked B is the best commercial iron, but not quite as low 

 carbon as the standard S2:)ecimen S.C.I. ; this latter not having been procured 

 until after most of the magnetic determinations were made. The values obtained 



* In rods of high permeability, having this ratio of length to diameter (about 200 to 1), a correction is 

 necessary for the values of the retentivity and permeability given in our tables and curves. Ewing has 

 shown {Phil. Trans., 1885, Part II.) that whilst the "on" curve — marked 1 in fig. 8 — even in a soft-iron 

 rod of 200 diameters is not much below the true value of that given by a much longer rod or ring ; the 

 " off " curve — marked 2 in fig. 8 — is notably different as it approaches the point where it cuts the vertical 

 axis at R, fig. 8. Here the demagnetising influence of the ends becomes apparent, and the result in soft- 

 iron is to reduce the residual induction, or retcntivihj, about 40 per cent., compared with an endless rod 

 erring. We have not applied this correction, as the amount varies with the permeability of each specimen, 

 A fair approximation to the true value can be found by inclining to the right, through a small angle, the 

 vertical axis, OB, in the curves. Plates III. to IX. (Sec Note A at end of this Paper.) 



f The absolute values of the maximum induction given in our Plates and Tables are a little too 

 high, as explained in Note B, p. 126. The word "retentivity" must be understood to he the residual 

 induction after our maximum field of 45 units lias been reduced to zero. 



