the same when or if necessary substantially as herein described and in part illustrated in the accompanying drawings.

"Seventhly. The method herein before described and illustrated in the accompanying drawings of combining the disc armature or insulated sectors with the continuous disc constituting the brake."

The remaining five claims were for-the horizontal and vertical mercury

commutators, grooving the pole pieces, diminishing friction at the bearings as described, and the meter as a whole as described.

This was an action for infringement.

Evidence was given of the plaintiffs' and defendants' meters. The working of the former can be best understood by a reference to the diagram of the current connections as shewn above. The form here

indicated is one with the saturated magnet described in the specification (ante p. 473). Only four sectors are shown; the action is precisely the same with sixteen. The current enters from the dynamo, or main, along the conductor, passes through the mercury trough 4, and commutatorj, thence along to the sector at x, where it divides into two branches, flows outwards as indicated by the arrows, thence by the conductors, h, to the outer edges of the other two sectors, along them reuniting at y, and thence by the commutator, k, through m and wire s to the lamp circuit. The current flowing inwards between the poles, f, g, of the constant magnet produces a force tending to turn the armature round on the axis a. As the sectors succeed each other, the commutator keeps the currents always flowing in the same direction relative to the magnetic field. The force or torque tending to rotate the armature is proportional to the strengths of the magnetic field and current. As the former is constant, the torque is proportional to the current, i.e. to

E
R'

where E is the voltage and R the resistance of the circuit. In the form of meter here shown the magnetic field was kept constant by the electro magnet being kept saturated; the magnetizing current being a shunt one in the conductor e, e. Unless some resisting force or brake were provided the armature would revolve with an acceleration until the resisting forces became equal to the driving force, or the armature split, or the mechanism otherwise broke down. To enable the device to work as a meter, a brake, consisting of a copper disc, b (attached in this form of meter to the back of the armature d), revolving between the poles, f, g, of the magnet, was used. On the known principles of Arago's Rotations, Foucault currents were set up of a strength proportional to the intensity of the magnetic field and the velocity of rotation. In a constant field this brake acted with a force proportional to the velocity of rotation. On a current of any given strength, therefore, being passed through the armature, rotation and acceleration ensued until equilibrium was established, the velocity of rotation being then proportional to the current passing, and thus by suitable recording mechanism the meter would show the total of ampère-hours.1 It did not measure energy. Evidence was also given to the effect that this meter was the first practical meter in which the difficulty arising from friction was overcome. This was accomplished by making the driving (and necessarily braking) force large in comparison to friction and by reducing the latter by the adoption of anti-friction wheels and, in some forms of meter, by the mercury commutator.

1 The result indicated would be

dt. Hookham's meter would give the same result on a circuit of 200 volts pressure through two lamps in series, as on one of 100 volts through one similar lamp. But on the same circuit (while R is constant), the current being proportional to the pressure changes in the latter would be appreciated. Thus, other things being equal, if the pressure fell from 100 to 99 volts on a consumer's circuit the corresponding indications on the meter would show a decrease of 1 per cent. in the ampère-hours recorded.

Evidence of previous knowledge was given in order to show that the specification, if it included claims covering the defendant's meter, would be invalid. It was proved that the conditions of the problem were well known :-the use of magnets with a copper disc (such as those used in ammeters, or in Abel's meter, No. 10237 of 1884); the necessity for constancy in the magnetic field of the brake, and for increasing the brakepower by multiplying the poles, which gave better effects when of narrow area; the necessity for obviating the effect of friction, which would prevent a meter starting with a small current, and when started would prevent a full record being made. Magnets of all shapes and sizes were known, including those that were used by the defendants. Numerous publications were referred to, of which the most important were: Messrs. Ayrton and Perry's Specification (No. 2642 of 1882) and papers by Marcel Deprez and F. Upperborn on Siemens' Energy Measurer.

The specification of Messrs. Ayrton and Perry disclosed a combination consisting of a small dynamo-electric machine and suitable brake, that is, one in which the force resisting rotation was proportional to the velocity of rotation. The movable armature coil and fixed coils consisted, one of a high-resistance and the other a low-resistance coil. It was immaterial which was the movable coil. The coil of high resistance was connected as a shunt to the points on the circuit between which the energy to be measured was used, the low-resistance coil carried the current used by the consumer. The torque was proportional to the product of the two currents, that used by the consumer and the shunt current. But, as the latter passed through a constant resistance, the current in it was proportional to the voltage on the whole circuit. Hence the torque varied as the product of voltage and current, and the dials would record the total energy that passed, that is, the true integral of current into electromotive force. But it was proved that this invention was not a practical meter, because the difficulty caused by the force of friction which was relatively considerable, was not overcome.

The paper by Marcel Deprez in La Lumière Electrique (January, 1884) dealt with the action of Siemens' Energy Measurer (No. 2210 of 1883). That device consisted of a small dynamo-electric machine of the Siemens' type, in which the consumer's current passed through fixed coils of low resistance and the shunt current through the movable bobbin of high resistance.

1 If R and be the resistances of the consumer's and shunt circuits respectively, E the E E. E.M.F. of the total circuit, then the currents in the armature and fixed coil will be and R E E E2 but as is constant this will vary as R or rR'

The torque will vary as the product

X

E2

R

or as EC; C being the consumer's current. With proper calibration the meter will show

Cdt, i.e. the total amount of energy or watts used. It will not only take into account variations of E, but, unlike Hookham's, a circuit of two lamps in series will have recorded double the amount of energy as in one lamp on a circuit of half the voltage.

In it the resisting force did not consist of a magnetic brake, but one formed of vanes revolving in a liquid. M. Deprez demonstrated rigorously that this instrument could not measure energy unless a brake were used which developed a resistance proportional to the velocity of rotation and not to its square, as did the vanes. He suggested "a copper disc revolving between the limbs of a permanent magnet" in which currents would be developed according to the well-known law, and which he had formerly applied in his magnetic-speed indicator. He also suggested reversing the Siemens' arrangement by putting the consumer's current through the armature and the shunt through a fixed high-resistance coil.

In a paper (16th February, 1884), F. Upperborn arrived at the same result as Marcel Deprez, and called attention to another consideration, namely the importance of relieving "more delicate measuring instruments from the performance of their own inherent work. But if this cannot be effected, then care must be taken that the measuring and measured forces are great in comparison with the forces coming otherwise into play. If we apply this law to the instrument under consideration, it results that one must then endeavour to diminish the frictional resistance as much as possible, and to make as great as possible the force consumption of the Faraday disc."

The alleged infringement consisted in the use, on a circuit alleged to be of constant potential, of a meter invented by Elihu Thomson. A diagram of it is given (post, p. 480), the left-hand half being shown in front elevation and the right hand in sectional. Its action will be best understood by reference to the diagrams shown (post, p. 481), in which the same letters are used for the corresponding parts. The current enters at A from the main, divides into two at B, one passing through the circuit and fixed coils L, L, and the lamps or consumer's circuit to D. The other, a shunt current, passing through the circuit S, commutator C, and movable armature and commutator to D, and thence to the return main. The commutator, C, was specially devised to get rid of friction, the contact-pieces were tipped with silver and of slight make, the shunt-current being small. The brake was a copper disc F on the same spindle, revolving between the poles of magnets G, of which at first three and subsequently two were used. In the improved form of meter an additional mode of overcoming the initial friction and eliminating error due to running friction was adopted, namely, the employment of a starting coil. This starting coil consisted of an extra coil of a few turns introduced into the shunt circuit, as shown in the lower diagram (post, p. 481). The magnetic field produced by the shunt-current always flowing through it tended, but was insufficient, to produce rotation of the armature, until at least sufficient current for one lamp was used in the circuit L, L. The armature was connected with a suitable train of mechanism and dials, on which was recorded the energy consumed, i.e. watts.1

1 The principles of this meter and their mathematical expression are the same as those shown in connection with Messrs Ayrton and Perry's in the note on the preceding page.

It was

proved that the magnets used in the defendants' meter were the constant, but constancy was not secured by reason of their shape and

proportions, but by a process of artificial ageing.

It was contended on behalf of the plaintiffs,' the specification should

be construed widely, being for the first successful meter made in accordance with known principles. That the invention consisted in a combination of five elements performing certain functions :-(1) An electro-motor of any ordinary form, in which the magnet must be constant unless when 1 Argument as finally presented in the House of Lords.