568 
PROFESSOR J. A. EWING ON EXPERIMENTAL 
force. We may, as an extreme case, applicable to a very soft iron wire strongly 
vibrated, take /x as 20,000 for any magnetising force of small intensity. Apply 
this to the case of a wire of ordinary size, say 4 mm. in diameter, and carrying a 
current of one-tenth of an ampere or O'Ol c.g.s. unit. At the circumference of the 
wire, where the magnetising force is greatest, its value is 
2C 
where C is the current and r the radius of the wire ; and this value is independent 
of the distribution of the current throughout the section, so long as that is 
symmetrical with regard to the axis. The magnetising force is therefore equal to 
2x0-01 , 
———=0T c.g.s. unit. 
With fA= 20,000 this gives 33 = 2000 as the value of the magnetic induction in the 
outermost layer of the metal—a magnetisation nearly half as great as that of the 
armature of a good modern dynamo ! Along with this circular magnetisation there is 
combined, in wires which run more or less north and south, a longitudinal magnetisa¬ 
tion due to the earth’s field, of an amount comparable with the above, so that the 
actual lines of induction form helices whose pitch, besides decreasing from the axis 
towards the circumference of the wire, shortens and lengthens with every fluctuation 
of current. 
In a circuit consisting of two long parallel straight aerial conductors, each of 
radius r and permeability /x, separated by a distance b, Maxwell’s formula (‘Elec¬ 
tricity and Magnetism,’ ii., § 685) gives for the self-induction per unit of length of the 
system the value 
7= 2l °g)r 2 +^ 
It is interesting to notice the comparative values which this quantity assumes when 
the conductors are, on the one hand, of non-magnetie metal, such as copper or 
phosphor-bronze, and on the other hand (again as an extreme case) of very soft 
annealed iron, kept in a state of vibration. To make a numerical comparison, we 
may take wires of 4 mm. diameter, separated by a distance of 20 centims. Then, for 
non-magnetic metal, 
L . 20 2 
7=2 log Q —-f 1 = 19-4. 
And for soft iron, under vibration, 
~ = 2 log I7-0 + 20,000 = 20,018. 
Hence, with the assumed data, the iron circuit has one thousand times the self- 
induction of the non-magnetic circuit. This throws light on the fact * that a notable 
© © 
* Preece, British Association (Aberdeen), 1885. 
