june-^, 1875] 



NATURE 



91 



ments or radicles produced by the electrolysis could not 

 be obtained separately. Pixii is said to have applied a 

 commutator to his machine in order to obviate this defect. 

 An important modification of Pixii's machine was soon 

 after made by Paxton, who caused the armature to re- 

 volve instead of the permanent magnet. According to 

 the character of the current required, armatures with 

 longer or shorter wires were employed. A large machine 

 of this construction, exhibited some years ago at the 

 Polytechnic Institution in London, was capable of 

 igniting a short platinum wire. In Clarke's machine 

 the position of the armature was altered and an im- 

 proved commutator applied. Siemens afterwards, by 

 giving the armature a cylindrical form, rendered it more 

 compact and better fitted for rapid rotation. 



Siemens' armature was happily applied by Wilde, in 

 1866, to the construction of a machine of extraordinary 

 power. Starting from a small magneto- electric machine 

 provided with six steel magnets, each weighing one 

 pound, and capable of carrying about ten times their 

 weight, Wilde transmitted the direct current from this 

 machine through the coils of an electro-magnet provided 

 like the former with a Siemens' armature, and the direct 

 current from the latter he in like manner transmitted 

 through the coils of another large electro-magnet— so 

 large, indeed, that its armature alone weighed above one 

 third of a ton. This was also provided with a Siemens' 

 armature. When the machine was in full action it 

 melted a rod of iron 15 inches in length and a quarter 

 of an inch in diameter, and gave the most brilliant illu- 

 minating effects when the discharge took place between 

 carbon points. As nearly as could be estimated, the 

 mechanical force absorbed in producing these results 

 was from eight to ten-horse power. Wilde's machines 

 have been successfully employed by Messrs. Elkington 

 for the precipitation of copper and other metals, and he 

 has lately proposed some important modifications to 

 adapt them to the production of the electric light. 



Some years before Wilde's experiments were published, 

 Holmes had constructed on the Saxton principle a powerful 

 magneto-electric machine, which has been successfully used 

 at Dungeness and other lighthouses, and machines differ- 

 ing little from Holmes's are employed in some of the 

 French lighthouses. In Holmes's original machine forty- 

 eight pairs of compound bar-magnets were arranged for 

 the armatures (160 in number) to revolve between the 

 poles of the magnets, and by a system of commutators 

 the current was obtained always in the same direction. 

 Its amount, however, varied at almost indefinitely short 

 intervals from a maximum to one-half of that amount. 

 In practice these variations were wholly inappreciable. 



The first suggestion of a magneto-electric machine 

 capable of giving a continuous current always in the 

 same direction is due to Dr. A. Pacinotti, of Florence. In 

 the nineteenth volume of " II Nuovo Cimento," which was 

 published in 1865,* Pacinotti describes the model of an 

 electro-magnetic machine constructed, some tima before, 

 under his direction, for the Cabinet of Technological 

 Physics in the University of Pisa, whose essential feature 

 was a novel form of armature to which he gave the name of 

 " transversal electro-magnet." This armature was formed 

 of a toothed iron ring, in m (Fig. i), capable of rotating 

 on a vertical axis, M M, and having the spaces between 

 the teeth occupied by helices of copper wire covered with 

 silk. The wire of the helices was always wound in the 

 same direction round the ring, and the terminal end of 

 each helix was brought into metallic connection with the 

 adjoining end of the wire of the succeeding helix. From 

 these junctions connecting wires were carried down 

 parallel to the axis of the machine, and united to insu- 

 lated plates of brass, of which a double row, as shown in 

 Fig. I, were inserted in a wooden cylinder, c, which was 



* The date on the title-page of the volume is 1863, but it contains % letter 

 dated Rome, Jan. 19, 1865 



itself firmly attached to the lower part of the axis. The 

 current entered through the successive brass plates as 

 they came into contact with a small metallic roller, k^ 

 which was in communication with one pole of a vol- 

 taic battery. At the point of junction with the wires 

 of the helices, the current from the battery divided into 

 two parts, which respectively traversed in opposite direc- 

 tions the connected helices, each through a semi-diameter 

 of the ring, and finally left the machine on the opposite 

 side by a second roller, k, which was in connection with 

 the other pole of the battery. When the connections 

 were made, the iron ring began to rotate round its axis 

 with considerable force. In a trial in which the current 

 was supplied by four small elements of Bunsen, a weight 

 of several kilogrammes was raised. In the apparatus as 

 actually constructed, the poles of the electro-magnet were 

 enlarged by the addition of two segments of soft iron, 

 A A, B B (Fig. 2), which extended over the greater part of 

 the iron ring. The details of the construction of the 

 transversal electro-magnet will be easily understood from 

 the plan given in Fig. 2. 



Towards the end of the paper to which I have already re- 

 ferred, Pacinotti shows that the iron ring armature, or trans- 

 versal electro-magnet, may be applied to reverse the con- 

 ditions just described, and to obtain continuous electrical 

 currents, always in the same direction, from a magnet, 

 whether a permanent one, or an electro-magnet. As the 

 original paper has not, as far as I know, been translated 

 into English, and the scientific journal in which it was 

 published is little known in this country, I will not make 

 any apology for giving the following extract without 

 abridgment. 



" If we substitute for the electro-magnet A B (Fig. i) a 

 permanent magnet, and make the transversal electro- 

 magnet revolve, we shall have a magneto-electric machine 

 which will give an induced continuous current always in 

 the same direction. To find the most suitable positions 

 on the commutator from which to collect the induced 

 current, let us observe that in presence of the poles of the 

 fixed magnet, there are formed, by influence, at the extre- 

 mities of a diameter, opposite poles on the moveable 

 electro-magnet. These poles, N s (Fig. 2), maintain a fixed 

 position when the transversal electro-magnet rotates upon 

 its axis ; and therefore, in respect to the magnetism and 

 consequently to the induced currents, we may consider, 

 or suppose, that the helices of copper wire move round 

 on the ring magnet while the ring itself remains at rest. 

 To study the induced currents which are developed in 

 these helices, let us take one of them in the various posi- 

 tions it can assume. From the pole N (Fig. 3) advancing 

 towards the pole s, there will be developed a direct cur- 

 rent in one direction till the middle point a is reached ; 

 on passing this point the current will assume an opposite 

 direction. Proceeding from s towards N, the current will 

 maintain the same direction which it had from a to s, till 

 the middle point b is reached ; after passing b the direc- 

 tion will be again changed, and will now be the same 

 which it had from N to a. Now, since all the helices 

 communicate with one another, the electro-motive forces 

 will be collected in one given direction, and will give to 

 the entire current the course indicated by the arrows in 

 Fig. 3 ; * and for collecting it, the most suitable positions 

 will be rt: ^ ; that is to say, the rollers on the commutator 

 should be placed at right angles with the line of magne- 

 tism of the electro-magnet." 



Pacinotti does not appear to have constructed specially 

 a magneto-electric machine on the above principle, 

 but he states that he verified the correctness of his 

 views by turning the iron ring in the electro-magnetic 

 machine, and observing the direction of the currents 

 when a galvanometer was introduced into the circuit. 



* Fig. 3, as given in the text, is an exact fac-simile of the corresponding 

 figure in the original. It is obvious from the figure itself, as well as Irom the 

 text, that there is an error in the engraving, and that the arrow between s and 

 b should point towards s and n4 towards b. 



