restrained from going too fast and stopped by this recharging device, the hand-brake being used only when stopping on a grade, and to hold the car while standing. Thus it will be seen that the energy of the ordinary brake employed by other systems to hold the car from going faster than desired, or to stop the car, is by this system returned to the battery to be used over again, or to the line to reduce the amount of power generated in the power house. This is saving the cake that is also eaten. The amount of energy so returned to the batteries on the Beverly and Danvers street railway is determined by careful measurement to be about fourteen per cent of that taken out, and after reduction for efficiency of batteries, loss in conductors, etc., leaves about nine or ten per cent, as the increased mileage due to the charging back device. The motion of the car is by its use made much more agreeable, and its control more effective.

Had the car that recently ran into another on Warren Street been fitted with this simple recharging attachment, the collision would not have taken place, notwithstanding the fact that the trolley was off the wire.

The two curves shown in Figs. 3 and 4, taken on different electric railroads, showing in a graphic method the amount of horse-power necessary to propel electric cars, are interesting. The curve in Fig. 3 was made from data taken on a car on a level road, using the ordinary overhead single trolley system, with the motor armature geared to the wheels by the ordinary two sets of gears open to the dust gathered from the road. The curve in Fig. 4 was formed from data taken on the Beverly and Danvers Railroad, which is very hilly, on a car equipped with this system. As will be seen from the first curve, it took about twelve horse-power to start the car at first, the gear being new; and at a certain part of the road consumed 20.1 horsepower. The next start took 21.4 horse-power, while the third start took 33.5 horse-power, presumably owing to the gear getting dirty, making an average throughout of 9.8 horse-power over the whole run. In the lower curve on the Beverly road the car was started with from 5 to 10 horse-power, and running at the same speed as in the former test the highest horse-power consumed was 16.1, while, where the curve passes below the horizontal or "no-current "line, considerable saving was effected by charging back, as much as 19.8 in one case being returned to the battery. The average horse-power in this case

is put down at 6, but as a matter of fact it is about 5.54. The saving over the whole line by charging back is about fourteen per cent, as stated above. The saving in such a case is not at first easily seen, but we think a careful study of the above carves will prove that there is yet much to be improved in the ordinary method of running electric cars, and they will serve as an explanation of the acknowledged and unnecessarily high power which today is regularly being consumed on electric railroads.

The methods that I have tried to describe to you perhaps seem too simple and too easy of accomplishment to be sufficiently effective, but a serious attempt to do it I think will convince the most sceptical that it is not so easy as one would be inclined off-hand to assert, and it is a fact that a sufficient amount of power is saved to enable the storage battery in its present stage of development to do the work. This is the important consideration, and what it has been sought to accomplish. I do not doubt that the frequent recharging of the batteries while in service increases their endurance, and no doubt the constant agitation of the solution enables them to withstand heavier drains than would be the case with stationary work.

This system was applied last summer as the motive power to a transfer table at the car shops of the Fitchburg Railroad at East Fitchburg. The current is taken from the same dynamo that furnishes light for the shops, and is carried to and from the motor by two wires and trolleys.

MEETING 402.

Experiments with Alternating Currents.

BY PROF. ELIHU THOMSON.

The 402nd meeting of the SOCIETY OF ARTS was held at the Institute on Thursday, April 10th, at 8 P.M., Prof. C. R. Cross in the chair.

After the reading of the records of the previous meeting, the chairman introduced Prof. Elihu Thomson, of Lynn, who read a paper on "Experiments with Alternating Currents."

PROF. THOMSON said: The study of alternating currents and the effects of such currents in producing fields of magnetic influence has been greatly stimulated by the industrial development taking place with alternating currents in electric lighting. In particular, the phenomena occurring as a result of induction and self-induction have opened to us very many interesting fields for study and investigation. The consideration of the action of displacement of phase due to induction or self-induction, as the result of a retardation or lag brought about by such induction or self-induction, has been particularly interesting and fruitful. Much light has been thrown upon the more obscure actions occurring in ordinary electrical apparatus using continuous currents by the analogous but more pronounced effects obtained with alternating currents. The subject of losses due to magnetic friction or hysteresis has been and is receiving careful study in the hands of some of the ablest electricians. The revival of the almost forgotten idea that the static spark or Leyden jar discharge is an alternating discharge at a very high rate or speed of reversals has not only assisted in our general understanding of electrical actions, but has borne fruit, it may be truly said, in the experiments proving that light and radiation are phases of electrical action,- not, as I have seen seriously discussed that light and electricity are one and the same thing, but that light and radiant heat are related to the science of electrical undulations or vibrations. It becomes simply, then, a widening of the electrical field to cover light and radiation, not a question of identity in all respects.

In the same way I anticipate eventually that we may learn by experimental research, coupled with theoretical considerations, that conduction of electrical current is not different from electrolysis in essence, except that the interchange of atoms in molecules of the conductor replaces that occurring in the electrolyte, and where the conductor is a solid a restriction of the decomposing and reforming molecules to definite positions occurs, while in the electrolyte the newly formed molecules are freer to move, and may therefore take new positions.

I anticipate, further, that we may learn that the warming of a body absorbing light or radiant energy is a result of almost infinitesimal closed electric circuits, just as the warming of a copper plate exposed to magnetic waves is due to electric currents on a larger scale. It may be possible also that we may learn to regard atomic move

ments, or transfers of atoms from molecules to other molecules in chemical actions, as brought about by electrical strains; perhaps to find that the dispersion of light by refracting media is in reality closely related to such strains affecting the atoms or atomic structure, as distinguished from the molecular structure; or it is a kind of atomic interference phenomenon. Already we are compelled to admit that the atomic structure of heated gases is such as to enable the atoms to give out definitely repeated electric strains, that is, definite wave lengths of light or color,- that the atoms are, as it were, like tuning forks of definite pitch, which, instead of setting up sound waves, can give forth electric waves.

It is not impossible that the science of chemistry itself may be adjudged to be a department of electricity. Whether gravitation and other molecular forces and properties will eventually be found to be so closely related to electrical actions is difficult to say. Perhaps we may be able to make the broader statement that the properties of the universal ether are electrical, and all the phenomena of the physical universe are closely related to, or exemplify the properties of, the ether.

I have extended this opening statement to some length, chiefly because I did not wish to miss the opportunity of furnishing, if possible, some food for thought, and because what we are to deal with in our experiments here are some instances of phenomena which depend directly on the ether motions, and not on the air, that is, they depend upon waves in the ether differing from those of light and radiation by being of extremely slow rate comparatively; but a very few hundred a second instead of millions of millions.

If we pass an alternating current through a coil of wire surrounding an iron wire core, or even if such core be absent, we will obtain an alternating magnetic field around the coil and core. The polarity or direction of the magnetic lines in such field is reversing a number of times per second, equal to the number of changes in the direction of the current in the coil. Now, if we immerse in such field in the proper way a closed coil of wire, a band of conducting metal, a plate of metal, or in fact any substance which conducts electricity, it will be the seat of currents of alternating direction corresponding to the inducing currents in the coil. The coil, band, or plate becomes a secondary circuit for the inducing coil. The ether waves or magnetic waves produce