THE CANTOR LECTURE, 1890 . 
45 
together, constitute a very full statement of the whole matter. Faraday 
considered that he had proved that each demagnetic line of force con¬ 
stitutes a closed curve; that the path of these closed curves depended 
on the magnetic conductivity of the masses disposed in proximity; that 
the lines of magnetic force were strictly analogous to the lines of 
electric flow in an electric circuit. He spoke of a magnet surrounded 
by air being like unto a voltaic battery immersed in water or other 
electrolyte. He even saw the existence of a power, analogous to that of 
electro-motive force in electric circuits, though the name, “ magneto¬ 
motive force,” is of more recent origin. The notion of magnetic 
conductivity is to be found in Maxwell's great treatise (vol. II., p. 51), 
but is only briefly mentioned. 
In recent years the notion of the magnetic circuit has been vigor¬ 
ously taken up by the desiguers of dynamo-machines, who, indeed, 
base the calculation of their designs upon this all-important principle. 
Having this, they need no laws of inverse squares of distances, no 
magnetic moments, none of the elaborate expressions for surface dis¬ 
tribution of magnetism, none of the ancient paraphernalia of the last 
century. The simple law of the magnetic circuit and a knowledge of 
the properties of iron are practically all they need. About four years 
ago much was done by Mr. Gisbert Kapp and by Drs. J. and E. 
Hopkinson in the application of these considerations to the design of 
dynamo-machines, which previously had been a matter of empirical 
practice. To this end the formulae of Professor Forbes for calculating 
magnetic leakage, and the researches of Professors Ayrton and Perry 
on magnetic shunts, contributed a not unimportant share. As the 
result of the advances made at that time, the subject of dynamo design 
was reduced to an exact science. 
It is the aim and object of the present course of lectures to show how 
the same considerations w r hich have been applied with such great success 
to the subject of the design of dynamo-electric machines may be 
applied tp the study of the electro-magnet. 
Historical Sketch. 
The effect which an electric current, flowing in a wire, can exercise 
upon a neighbouring compass needle was discovered by Oersted in 
1820. This first announcement of the possession of magnetic properties 
by an electric current was followed speedily by the researches of 
Ampere, Arago, Davy, and by the devices of several other experi¬ 
menters, including De La Hive's floating battery and coil, Schweigger's 
multiplier, Cumming's galvanometer, Faraday's apparatus for rotation 
of a permanent magnet, Marsh's vibrating pendulum, 1 and Barlow's 
rotating star-wheel. But it was not until 1825 that the electro-magnet 
was invented. Davey had, indeed, in 1821, surrounded with temporary 
coils of wire the steel needles upon which he was experimenting, and 
had shown that the flow of electricity around the coil could confer 
magnetic power upon the steel needles. But from this experiment it 
1 In the original memoir, already quoted, I described Marsh as “ the Arsenal chemist.” Pro¬ 
fessor Peter Barlow was, however, the Arsenal chemist, and Marsh was Barlow’s assistant. In 
March 1822 (Philosophical Magazine) Professor Barlow described his own star wheel electro¬ 
motor, and mentioned Marsh very favourably—crediting him with original work, and regretting 
that “ Marsh was not in ar position where his great abilities would receive higher pecuniary 
recognition.” (sic.) 
