Magnetic Permeability of various Alloys of Iron. 103 
point of maximum induction, and another series down gives the part 4 to where 
the curve cuts the axis OB again; the circuit is again broken and the current 
reversed, and the last series of steps up gives the part of the curve marked 5, 5. 
This completes the cycle, but another series of steps down was always taken, 
—i.c. repeating the portion 2 of the curve—and finally another series of steps up, 
corresponding to the upper part of 3, so as to leave the rod in a practically non 
magnetic state, thus saving time if we found that its cycle had to be repeated. 
In every case the whole cycle was gone over again, if by inspection of our 
observations we found any discrepancy between the upper and lower halves of 
the cycle, that is if the cyclic curve was 
not perfectly symmetrical above and below 
the axis OH. In the plates appended to 
this paper the position of the cycle below 
the axis of OH is therefore omitted. The 
initial curve 1, fig. 8, 1s shown in all the 
plates as a finer line than the continuous 
cycle of the particular specimen, but repre- 
sented in the same kind of dotted or 
continuous curve. For convenience of 
comparison the B and H curve obtained 
for our standard rod of iron is reproduced 
in all the plates. 
The residual induction of the specimen is 
expressed by the distance OR (fig. 8); that 
is to say, the induction which remains in a 
very long thin rod after the magnetising 
current is removed: for brevity we will 
call this the retentivity.* The coercive force 
is expressed in terms of H by the distance OC; that is to say, the magne- 
tising force required to demagnetise the specimen. The magnetic permeability 
of the specimen varies, of course, with the magnetising force employed, and is 
expressed by the ratio of the induction B to the magnetic field H at any given point. 
The hysteresis loss can be deduced from the areas enclosed by the B-H curves.t 
Fie. 8. 
* As the residual induction increases with the maximum induction the rato of these two quantities is, 
strictly speaking, the retentivity. The coercive force also rises with the maximum induction. 
+ For the measurement of most of these areas we are indebted to Mr. R. L. Wills, a former senior student 
in the Physical Laboratory of the Royal College of Science, Dublin, now working in the Cavendish 
Laboratory, Cambridge, as an 1851 Science Research Scholar. We must also acknowledge the useful 
assistance rendered to us in the laborious determination of these B and H curves by Mr. Wills, in con- 
junction with Mr. R.G. Allen, another Research student in the Physical Laboratory of the Royal College of 
Science, Dublin. 
