348

the first place the number of hours required for transformers to attain a steady temperature was much greater than Mr. Mordey would have us believe, for he (Mr. Crompton) had kept them running for 30 hours, and still the temperature was not constant. He had also kept both a "Hedgehog " and a closed circuit transformer of equal output on for 20 hours, the final temperature of the former being 120° F., whilst that of the latter was 212° F. His experience with the "Hedgehog' type in the lighting of Chelmsford had been satisfactory. Referring to "all-day efficiency," he said most people speak of the efficiency of transformers being 90 per cent. at half load, as if this was their normal working condition. This load was far above the average, even during lighting hours. It might be approximately right for shops and other places where, the probable load is readily ascertained, but for private houses it was much too high. Taking the case of a customer whose lighting bill, at 8d. per unit, amounts to £20 a year, he said such a house would require a 50 to 60-light transformer, and the average consumption throughout the year would be 70 watts. The transformer would thus be working, on an average, at th its full load. It was curious to notice that in the transformer mentioned by Mr. Mordey (see Table III.), the iron losses at light loads were about 69 watts, and thus the yearly efficiency would become very near indeed to 50 per cent., the number he gave three years ago. He would have been glad if Mr. Mordey had given the statistics of some of the transformer stations with which he was connected, in the shape of the ratio of energy paid for, to energy sent out from the station. This, he said, was the important point, and at present it appeared that transformer plants were using about 24 times as much coal, per unit sold, as direct systems. All this extra coal, he submitted, was spent in heating the transformers. Banking transformers and distributing at low pressure would diminish this loss, but after all, what did this come to? Simply that you put down a high pressure line with its expensive insulation and a transformer at the end of it, whereas with continuous currents a rather heavy feeder with cheap insulation was all that was required. Transformers, he said, were not so easy to manage as was imagined, and were, in this respect, much better than secondary batteries. On the question of "storage he could not agree with Mr. Swinburne's opinions, for in stations subject to uncertain loads, the employment of accumulators to the extent of 20 or 30 per cent. of the total output enabled the working expenses to be reduced immensely. In his opinion, the reason why storage is not used in America was that the loads were fairly steady, and a satisfactory storage cell was not to be had there.

Mr. ADDENBROOKE remarked that Mr. Swinburne was under a wrong impression when he stated that probably no existing company intended changing their system of supply, for he (Mr. Addenbrooke) well remembered that when Mr. Ferranti took out his first patents for transformers, the subject of sub-stations and low pressure distribution was thoroughly appreciated by the inventor. At that time, however, it was impossible to found sub-stations, but it had always been the intention of Mr. Ferranti, as well as of the House-to-House Company, to adopt this system whenever they could. The cutting out or in of transformers at such stations could be effected either automatically or by hand on the lamp-lighter principle, and with either method a difference of 1 or 2 per cent. in the efficiency of the transformer would be a matter of little importance.

Mr. EVERSHED regretted that the diagrams belonging to the paper were not issued with the proofs, for without them it was very difficult to follow the reasoning. He described as admirable the explanation of how alternating dynamos act as motors, and said the action was much easier to understand on the armature reaction theory than by the older ideas about self-induction. Referring to the compound machines mentioned on slip 7, where the magnets are excited by a redressed alternating current, he enquired how the sparking could be got over. On slip 8, the author says cast iron magnets would not respond quickly to armature reactions, but he (Mr. Evershed) failed to see why this should be so, for the high specific resistance of the iron would prevent Foucault currents almost as well as laminated wrought iron. In conjunction with Mr. Vignoles, he had made many experiments on the loss by hysteresis in different samples of iron, and had never found a sample give more than 25 per cent. more loss than Ewing's specimen; only one sample gave this value, and it was an ordinary bar taken out of a smith's shop. In most cases, the losses are about 5 per cent. more than that found by Ewing. On the subject of efficiency of transformers, he said Mr. Swinburne had not taken a good design of closed circuit converter with which to compare his "Hedgehog."

Mr. SWINBURNE here remarked that in his B.A. paper he had given the best proportions of copper and iron for the closed circuit form, but in this paper he had taken an average of the various forms in the market.

Mr. EVERSHED said that to be anything like a good design there should be much more copper in it, about 2,000 turns in the primary instead of the 400 given. This would reduce the iron loss to 40 watts instead of 120, and the exciting current would be reduced to 08 instead of 3 ampère required by the open circuit type. Nevertheless, if good condensers could be obtained, the large exciting current would not be a serious defect. The subject of open versus closed circuit transformers had occupied his attention for some time, and he found that if you cut the iron circuit of one of the latter form, its efficiency is increased. Such being invariably the case, he said, why not " go the whole hog or none," and do away altogether with the iron core. Recently he had worked out the dimensions of a 10 H.P. transformer made on this plan, and found that at a frequency of 70 it would require two tons of copper. If, however, the frequency was raised to 210, the weight was reduced to 200 lbs., and the saving resulting from the absence of a core amounts to 20 per cent. per annum on the cost of the transformer. Difficulties might be experienced in making alternators for such high frequencies, but so far as coreless transformers were concerned, high frequency was decidedly advantageous.

[MARCH 13, 1891.

Physical Society, February 27th, 1891.

Prof. W. E. AYRTON, F.R.S., President, in the chair.

Prof. A. Gray, M.A., was elected a member of the Society. The following communications were read:-" Proof of the Generality of certain Formulæ, published for a Special Case by Mr. Blakesley. Tests of a Transformer," by Prof. W. E. AYRTON, F.R.S., and Mr. J. F. TAYLOR.

In 1888 Mr. Blakesley published a number of formulæ relating to the measurement of power, &c., in alternating current circuits by means of electro-dynamometers, one of which had its two coils independent, and placed in different circuits. These formulæ were deduced on certain assumptions, the chief ones being that the currents and magnetisations varied harmonically, and that the magnetic stress in the iron was proportional to the ampère turns. The present paper shows that the above assumptions are not necessary to the truth of the resulting formula. To take the case of a transformer, let alternating current ammeters be placed in the primary and secondary circuits, and a direct reading" split dynamometer have a coil in each circuit. Let D, D, and D be the respective readings of these instruments, then whatever be the law of variation of the currents,

The power lost in heating the iron core was also shown to be

· (D-D). Other formula, such as that given by Mr. Blakesley,

expressing the primary volts in terms of the dynamometer readings, are shown to be true generally.

=

On the subject of transformer magnetisations, the authors state that it is desirable not to speak merely of the ampère-turns, but also of the self-induction, for they have reason to believe that the magnetising value of an ampère-turn varies. Defining the self induction d N of the secondary by the equation L, , N being the total finx d As through the secondary, they show geometrically (assuming harmonic variations) that if L, diminishes the efficiency increases, and at the same time, the phase angle, e, between the primary and secondary currents, and the magnetic lag, p, decreases. On comparing this theoretical deduction with the results of experiment, they find a good agreement, as will be seen from the following table :

Similar results were obtained in different sets of experiments. It was also noticed that as the secondary current is increased, the efficiency rises to a maximum and then diminishes, whilst L,,, and . diminish and then rise again; the current which gives the maximum efficiency coinciding almost exactly with that which gives minimum values to L., 0, and 4.

The methods employed in making the tests are described in the paper, and the formulæ used in working out the results are there demonstrated. Amongst the facts deducible from the experiments are the following:

1. With constant frequency the ratios primary volts rise.

V. D. and Vp Dy

increase as the

MARCH 13, 1891.]

3. With constant secondary current, L. decreases as the primary volts are increased.

4. With small constant secondary currents the efficiency diminishes as the frequency increases.

5. For large secondary currents the efficiency is approximately independent of the frequency.

6. Shunting the secondary with a condenser increases the efficiency."Further contributions to Dynamometry," by T. H. BLAKESLY, M.A. The object of Mr. Blakesley's paper was in the first place to show what sort of physical quantities could be advantageously evaluated by using electro-dynamometers of two coils of low resistance In circuits conveying electric currents. The meaning of a dynamometer reading was explained to be the mean value of the product of two currents, either steady or undergoing any periodic variations with sufficient rapidity. In mathematical language, such an instrument measured

T

C1 C2 dt, where c1 and c2 were the instantaneous

values of the currents in the two coils, including of course the common case where these currents are identical. Any physical quantity whose value was such a product c, c2 multiplied into something which was independent of the time and which therefore on integration came outside the integrator, was well adapted to have its mean value given by such instruments. Power was such a quantity, being merely Current) x resistance. The square of an E.M.F. was another such quantity, but he did not wish to restrict the method to such evaluations. It follows that any quantity whose instantaneous values can be expressed by terms each quadratic in current, and whose other factor was independent of time, could have its mean value expressed in dynamometer readings. In addition, the particular place and mode of coupling of the dynamometers was indicated by the instantaneous equation, as well as the factor to be applied to each reading. Thus the equations are made to indicate the practical arrangement to be adopted, and the use to be made of the observations in each case. Examples were given for the cases of transformers in series and parallel, and special applications of the method were suggested in the measurement of the power employed in such diverse apparatus as voltameters subject to direct or variable currents of any sort, welding machines, parallel generators, tuning-fork circuits, vacuum discharges and imperfect condensers. For parallel generators, the power of each could be separately estimated, and in the case of electric welders, the the power employed in the welding circuit was shown to be measurable without introducing any resistance whatever into that circuit.

Mr. J. SWINBURNE said the authors' assumption that there is no back or forward E.M.F. in the primary and secondary circuits of di transformers except that due to where i is the total induction in d t

the core, was unwarrantable, for in all real transformers there was a "drop" due to waste field, and this made the split dynamometer method useless. It makes the full load efficiencies too high and this, he thought, accounted for the extraordinary results obtained by Prof. Ayrton and Mr. Taylor. If a dynamometer be used at all, it should, he said, be used as a wattmeter, the moving coil of one turn being joined in series with a non-inductive resistance and put as a shunt to the primary. The power absorbed by the instrument itself should be then determined, and the power given out by the secondary measured by the same instrument, if the secondary be not non-inductive. Any errors due to self-induction in a wattmeter are, he said, equally present when it is called a split dynamometer, and in addition to this, the wattmeter as a split dynamometer precludes the possibility of measuring power.

Mr. MORDEY said the results obtained by Prof. Ayrton and Mr. Taylor confirmed experiments he had made himself by an entirely different method, for he found that the losses in the iron decreased considerably as the secondary current increased, and this gave increased efficiency. In his experiments he kept the load constant until the transformer attained a steady temperature, and then substituted a direct current for the alternating one, varying its strength until the same steady temperature was maintained. The power thus supplied a measure of the loss in the transformer under the working condition. A 6 kilowatt transformer tested by this method gave a loss of 110 watts at no load, and at full load 205. Of this 205, 176 was accounted for by the loss in the copper coils, leaving only 29 watts as the iron losses at full load. Figures which he quoted from Prof. Ayrton and Mr. Taylor's paper showed the same general results.

A "Note on Electrostatic Wattmeters," by Mr. J. SWINBURNE, and a paper on "Interference with Alternating Currents," by Prof. W. E. AYRTON, F.R.S., and Dr. SUMPNER were postponed.

NEW PATENTS-1891.

3,239. "Improvements in underground conduits for conductors of electricity." J. S. RAWORTH, T. O. CALLENDER, and C. E. WEBBER. Dated February 23rd.

3,262. 66 An improvement in conductors for electrical glow lamps." E. A. GIMINGHAM, of the Edison and Swan United Electric Light Company, Ld. Dated February 23rd.

3,265. "A new or improved process of electro painting." H. F. GHANDINGER. Dated February 23rd.

349

3,336. "Improvements relating to welding, brazing, or otherwise joining metal strips, bars, rods, and the like by electricity, and to apparatus therefor." H. H. LAKE. (Communicated by H. Lemp and L. M. Schmidt, United States.) Dated February 24th. (Complete.) 3,338. "Improvements in telegraphy." J. A. PARKER. February 24th. (Complete.)

Dated

3,349. Improvements in electric welding apparatus." M. W. DEWEY. Dated February 24th. (Complete.)

3,364. "Improvements relating to the manufacture of incandescent electric lamps." H. H. LAKE. (Communicated by W. E. Nickerson and A. Berrenberg, United States.) Dated February 24th. (Complete.) 3,373. "Improvements in and connected with medical batteries." G. F. WEBB and J. A. CRISP. Dated February 24th.

3,383. "Improvements in dynamo-electric machines." J. A KINGDON. Dated February 24th.

3,398.

"An improved electrical burglar alarm." A. E. SPENCER. Dated February 25th.

3,404. "Improvements in switches for telephone circuits." C. H. ELLIOT and E. F. FURTADO. Dated February 25th.

3,417. "An improved galvanic battery." W. H. MUNNS. municated by G. A. Smith, Canada.) Dated February 25th.

(Com

3,426. "Improvements in brushes and brush holders for electric machines." H. J. DOWSING. Dated February 25th.

3,449. Improvements in telephones." P. RABBIDGE. February 25th. (Complete.)

Dated

8,241. Improvements in electric lamp hangers." C. L. JEFFERS. Dated May 27th. 8d. The object of the invention is to support an electric lamp in such a manner at the outer end of a horizontal arm reaching from the top of a mast, placed at the edge or curbstone of a sidewalk, over and toward the centre of a street, that the electric current can be disconnected by the operator, and the lamp drawn in toward the mast and lowered down the front of the mast to the hand of the operator, thus leaving the mast, the horizontal arm, and the wires carrying the electric current fixed in their original position during the operation of trimming the lamp, and consequently not interfering in any manner with the occupation or use of the street. Another object of the invention is to so construct the mast and arm that, while obtaining the necessary strength, stiffness, and durability, I have at the same time avoided all unnecessary weight, and by the peculiar construction of the mast to permit of the drawing in of the lamp without occupying any more room for appliances used for such purpose than what is required for simply the mast, and the lamp itself, the hood and other connections being firmly attached to the 9 claims.

arm.

350

CORRESPONDENCE.

Electric Transmission of Power over Long Distances. It is possible some of the ideas that occurred to me whilst listening to Mr. Kapp's very instructive Cantor lectures may be of interest to your readers.

With regard to automatic regulation of the speed of the motor, with constant speed of the generator, I would point out that the latter condition is much more difficult to attain than the former. Constant speed within 10 per cent. between empty running and full load can easily be obtained with series machines having curves that will not result in pulsations of the current being set up, and the load in an ordinary factory does not vary very greatly except at certain stated times, such as intervals for meals, when the turbines must be regulated by hand, as most of them have no automatic regulator. For electric purposes turbine regulators are usually worthless, as with ordinary falls they act much too slowly to be serviceable. In some instances that I am aware of the load has been suddenly taken off, owing to the circuit being broken from various causes, but no damage has ever been done although the increase of speed is rather alarming. These were all cases of series machines; shunt machines would have burnt up very probably, even if Gramme-wound. Out of 30 or 40 sets of plant designed by Mr. Brown, which I have set in action for the Maschinenfabrik Oerlikon, only two or three have been shunt, and these were cases of the distribution of power where series machines could not have been employed. Small motors can, of course, be run in series if controlled by centrifugal governors, as in America, but where one machine is very large and others small, there are difficulties in the way. Series machines, besides requiring no resistance in series to start them, if only one is driven from its own turbine, spark less with a varying load than shunt machines. With reference to the employment by Mr. Dobrovolsky of a non-inductive shunt across the ends of the series magnet wire, I do not wish to take up your space with personal claims, further than to say I mounted a permanent non-inductive shunt on the switchboard of an installation belonging to a Mr. Boller, in the neighbourhood of Zurich, about 1 years ago, as near as I can remember, after having found that shunting the generator with a German silver resistance belonging to an arc lamp would stop the pulsations in the current, but that unwinding the magnet wire, so as to reduce the ampère turns, would not. Then as to multipolar dynamos and motors. Although the armature reaction is less in a multipolar machine than in a two-pole, when the dynamo has to give many volts, and these are concentrated in, say, a sixth of the commutator, the sections on each side of the neutral point have a considerable difference of tension between them, and a small movement of the brushes may cause as much sparking as a much larger movement in a twopole machine. With regard to the Schaffhausen plant, I do not think that Mr. Kapp mentioned the convenient way very employed for testing the two primary machines up to their full power (300 H.P. each). One machine was driven from the other as a motor, and the power absorbed by turning the water on to the backward revolving turbine.

Posthumous Scientific Honours.

Leslie Miller.

The remarks which you passed on M. Silvanus Thompson's criticisms are quite sufficient and "cogent," so I do not believe it is necessary for me to show by any argumentation that I did not play a part in a "comedy of errors when bestowing upon an electrician a praise which was deserved by his brother, perhaps to a greater degreee. But in case you do not believe it is time to stop the controversy, I should presume to say, to our friend, that I see really a great difference between a horse-shoe loadstone and a horseshoe electromagnet, but not any difference of horse-shoism This understanding of horse-xhoism is precisely the ground upon which is based my opposition to his claims in favour of Sturgeon. According to my own opinion, the invention of this mutual corroboration of the poles by their vicinity, is to be attributed exclusively to the first man who horse-shoed a loadstone, and not to the one who, many years afterwards,

[MARCH 13, 1891.

horse-shoed an electromagnet, who in fact is to be considered as being the first horse-shoer of a loadstone, and the creator of horse-shoe form. I suppose that it is quite impossible even to so careful a historian as M. Silvanus Thompson may be, to ascertain. I suspect that this great discovery was done by a series of happy chances and gradual approximations, the work of a real evolution, and a kind of scientific Darwinism.

It must be noted, moreover, that in 1820 or 1821, Arago himself published an essay in Annales de Chemie et de Physique, where he proves by experiments that the nature of the poles excited in a bar of iron or steel, is determined by the direction of the helix, and can be varied at will with the same current.

Consequently every element for the construction of the first horse-shoe electromagnet was ready before 1829, and Mr. Sturgeon had nothing to invent, for using these facts for his purpose in case he has not been anticipated.

But in the meanwhile I should be quite sorry if an impres sion was created upon the minds of your readers that I am denying any merit to Sturgeon. This active and independent editor of the first electrical paper ever published, was the first scientist able to understand fully the possibility of creating an immense force of mechanical energy with the newly discovered apparatus. He saw a great and startling fact which had escaped the notice of Ampère and Arago. Consequently he deserves unquestionably to be ranked amongst those "celebrated" workers in electricity, who from the time of James I., up to the present year of the Victorian era, raised your land so far above any other in the annals of the allimportant science, won for themselves the admiration of the world, and rendered the knowledge of Shakespearian idiom almost an obligation to any followers of Faraday and Sir William Thomson's researches or investigations.

Paris, March 9th, 1891.

W. de Fonvielle.

I annex below a copy of a letter I have written to-day to the members of the Council of the Institution of Electrical Engineers, and I shall feel obliged if you will allow it to appear in your next issue, in order that it may come under the notice of the members of the Electrical Institute generally. S. Alfred Varley.

March 9th, 1891.

[COPY]

To the Members of the Council of the Society of Electrical Engineers. Gentlemen,-I beg to call your attention to a letter which appeared in the last issue of the ELECTRICAL REVIEW, signed Silvanus Thompson, and headed, "A Silly Joke."

Dr. Thompson is quite welcome to abuse me as much as he pleases, I am quite capable of taking care of myself, but M. de Fonvielle is necessarily not so favourably circumstanced for doing so.

I feel sure the Council will be of opinion that M. de Fonvielle, writing to an English journal in a language which is not his native tongue, has a right to be protected from being held up to ridicule by a physicist who not only is a member of the Council of Electrical Engineers, but who also holds a professorial chair.

I enclose cuttings from the ELECTRICAL REVIEW containing the letter of M, de Fonvielle, and also that headed, "A Silly Joke," signed Silvanus Thompson. Together with the cuttings I enclose a reprint of an article written by myself, and I would direct attention to the extract from the Philosophical Magazine of January, 1821, which will be found on page 2.

The extract referred to will, I think, be considered to demonstrate that the main contention of M. de Fonvielle is perfectly sound. The object I have in writing this letter will be accomplished by its simple acknowledgment.

I am, Gentlemen, yours obediently,

power could be obtained by using bars that were long in comparison with their breadth." The date when this magnet was constructed is not given, but a horseshoe magnet was used in connection with the star-wheel motor, described in the Phil. Mag. of March, 1822, as will be seen by a reference to the engraving accompanying Barlow's description. Permanent steel horseshoe magnets were, threfore, certainly in existence in 1822, and probably very much earlier.

The extract which I gave from Mr. Hatchett's communication in the January number of the Philosophical Magazine of 1821, headed "Electro-magnetic Experiments of Oersted and Ampère," clearly demonstrates that before January, 1821, Arago had not only magnetised iron by placing it in the interior of a helix forming part of a voltaic circuit, but that he had at that period also demonstrated that the magnetism developed was directly as the quantity of the electric current flowing through the convolutions of the helix.

Now, as horseshoe magnets existed certainly early in 1822, and bar electro-magnets were produced before 1821, M. De Fonvielle very fairly contends that the making a horseshoe electro-magnet in 1825 by Mr. Sturgeon does not entitle him to be considered the first to have made electro-magnets, as is more than suggested in the Cantor Lectures delivered by Dr. Silvanus Thompson.

I pointed out in my article on "Posthumous Scientific Honours that Mr. Sturgeon's letter of 1825, to the Secretary of the Society of Arts, was accompanied by three certificates, signed by Profs. Christie and Gregory and also Mr. Barlow. The certificates of Christie and Gregory are somewhat lengthy, and describe the special features and claims to novelty urged on behalf of Mr. Sturgeon. Mr. Barlow's certificate is a short one; he simply says he has read the certificates of Christie and Gregory, and fully endorses what is there stated.

Now, neither in the certificates nor in Mr. Sturgeon's letter is there any reference whatever to the horseshoe electromagnet which formed part of Sturgeon's apparatus, and this goes to demonstrate that neither Sturgeon himself nor Messrs. Barlow, Gregory and Christie considered that the horseshoe electro-magnet, forming part of the apparatus rewarded by the Society of Arts, was a new scientific discovery.

Sturgeon, as an original and also as an industrious scientific worker, is entitled to every credit for what he did. At the same time, there is no necessity to credit him for what was really done previously by Ampère and Arago.

To Ampère, Arago, and Sturgeon it can now be a matter of no importance to whom the credit of the discovery of the electro-magnet be given; but it is of some importance to France herself that the distinguished men she claims as her sons should not be deprived of the scientific credit which is fairly their due.

I have read with considerable interest Mr. King's letter in your number of March 6th. If Mr. King had thought for a moment he would not have imagined that I could have had any hand in the writing of the article, the tone of which he complains of. As a matter of fact the statements about the Vienna installation, which make Mr. King attribute the authorship to me, astonished me not a little, and even if I had not seen Mr. King's letter, I should have asked for further information on my own behalf. It is now some years since I have had anything to do with the working at Vienna. During the first few months of working the difficulty mentioned in your article of charging the four batteries in series and simultaneously taking current at different rates from each of them, had been noticed, and had been satisfactorily dealt with by alterations in the switchboard, and I heard from the engineers in charge that the efficiency of the whole system was considerably over 80 per cent. I do not know whether later experience confirms this, but at any rate I am quite certain that I could again carry out the Vienna system combined with accumulators, and obtain a high efficiency, provided that at times of maximum demand at least two-thirds

351

of the current comes direct from the dynamos, and only one-third is taken from the accumulators. It is on this point that I join issue with Mr. King and those working with him. I have always admired the extreme ingenuity of the system employed at Chelsea, but I have felt that far too large a portion of the current passes through the accumulators to enable a fair efficiency to be arrived at. Further than this, that the large proportion of accumulator plant that is necessary renders the system very costly to put down and maintain.

I fully agree with the remarks in Mr. King's last paragraph but one, to the effect that the addition of storage plant can reduce both the coal and wages bill, but it must be so employed that the saving under these heads must not be over-balanced by the expenditure under the other head of increased cost of upkeep.

My own experience and calculations show me that the best economical advantages are obtained when the storage plant. is only of sufficient capacity to supply a fifth part of the total 24 hours output of a long December day and night. Such a proportion is, I believe, less than a fourth of that contemplated by Mr. King even when he employs direct current transformers as an adjunct. R. E. Crompton.

March 10th, 1891.

[The mention on p. 309 of last issue of Mr. Crompton's name in connection with the subject was a passing allusion of our own, to what might possibly be running in Mr. King's mind and for which he was not in any way responsible.EDS. ELEC. REV.]

In his letter appearing in your last issue, Mr. Frazer states the life of E.P.S. secondary batteries on tramcars to be 10 months, or 300 discharges. As a matter of fact, the life of the ordinary positive traction type plates is about 400 discharges more or less, according to the degree of care in handling. The life of the negative plates is for effective work about 750 discharges, but I have not up to the present time seen a plate which has actually broken down from fair wear and tear. When in February, 1890, the positives of the first set of batteries at Barking Road required renewal, it was decided by the Traction Company to renew both positive and negative plates, but it was afterwards evident upon inspection of old plates sent in that the majority of the negatives would have certainly run for another year.

The experience of the past year from February, 1889, to February, 1890, has proved that the maintenance rate for E.P.S. batteries on the Barking Road line, with all the manifest disadvantages of a small installation, has not exceeded one penny per car mile, and this with batteries of the old traction type.

Crucial experiments have now satisfied me that the positives of the "K" type traction plates will endure in actual work from 800 to 1,000 discharges, and that the maintenance of batteries will hereafter be fully covered by two-thirds of a penny per car mile. Frank King.

"Government Electric Pinnaces." Advertisements appear in this morning's daily papers announcing the "launch of the first Government pinnace, at Chiswick, 1.30 p.m." We must guess whether the makers of this pinnace intend to convey the impression that the Government officials had long neglected to inform themselves of the progress of electric propulsion, or whether the said boat builders regard the launching of an electric pinnace as an historical event. Since the successful trial trips of the launch Electricity, in 1882, on the Thames, and the equally successful Voyages of the Volta, from London to Calais and back, in September, 1886, quite a number of electric boats have been constructed on precisely the same lines, and they have been brought under the notice of the British Government from time to time. Those who have taken an interest in the discussions of papers on electric launches, read at the British Association, the Society of Arts, and the Institution of Naval Architects, will recollect that members of the Royal Navy were frequently to the front with substantial praise,

352

strongly recommending this novel mode of propulsion for special purposes.

The Lords Commissioners of the Admiralty, however, were rather slow in adopting the recommendations of admirals and naval constructors, and as late as February, 1885, in a reply to the petition of a commander who wanted an electric boat, included the following comments: "That it is not advisable to make the trial until further progress has been made by private companies, and until greater power is obtained than at present, with the same or less weights than are required for steam power." In July of the same year, notwithstanding the above decision, I received an order from the Lords Commissioners of the Admiralty for the supply of an electric launch, motor and accessories. This was subsequently used by Commander C. Robinson, R.N., of H.M.S. Vernon, for the propulsion of a 22 feet pinnace at Portsmouth. The equipment comprised a 2 H.P. motor, a 14-inch forged steel screw, shafting, switches, and 48 accumulators. Commander Robinson's series of experiments with this boat did not reveal any advantages as regards speed over steamboats, and the question of electric propulsion was shelved by the Navy for the time being. In the following year the Italian Government ordered an electric launch 37 feet long, of 6 feet 4 inches beam, steel hull, to give a speed of not less than 7 statute miles per hour for 5 hours' continuous run. launch was constructed by Messrs. Yarrow & Co., and I supplied the electrical equipment. During the official trial trip" over the measured mile" below Gravesend, the contract speed was exceeded by nearly half a mile per hour, to the complete satisfaction of all concerned. The weight of existing accumulators precludes successful competition against first-rate steam launches, as far as speed is concerned, and in this respect no improvements whatever have been made since. 1882, but the electric boat has several well known advantages which ought to make it valuable in special naval services such as have been instituted by the Italian Government, viz., the laying and removal of torpedoes in harbour and coast defences. A. Reckenzaun.

Mr. Varley's Blunder and his Way out of it.

This

It is all very well of Mr. Varley, after having been found out in his blunder, and after having tried to wriggle out of it by slandering the late Prof. Fleming Jenkin, to assume the air of injured innocence, and declare he will not henceforth notice anything that I may write. As for his nonsense about his 1866 machine being the first self-exciting dynamo constructed, what on earth has that, or my refusal to admit into the rank of established facts a claim that stands solely upon Mr. Varley's own assertions, to do with his present blunder or his insinuations against Prof. Jenkin? In vain does he trail this very ancient red herring across the track. He neither admits his blunder nor repents of his slander, which is quite a sufficient excuse why I should no longer waste a single word more on either him or his truly self-exciting pre

[MARCH 13, 1891.

have always understood that for a current to flow, a difference of potential was necessary; and in endeavouring to apply that rule to the case in point, I was somewhat staggered to find that if there were any difference of potential between the points indicated by arrows, the current flowed in a very curious manner. Let us suppose that the potential at one point be 10; then it must be lower at the next succeeding point, and so on (see cut). What I want to know is this: Why doesn't the current flow direct from 10 to 6?

March 6th, 1891.

"Potential."

In the first place, our correspondent is wrong in saying a P.D. is necessary for a current; all we want is an E.M.F. The following may be the best way of placing the matter. Let p and q be two closed circuits, and let the magnet, n 8,

be moving in the direction of arrow. Then the lines through P will first increase and then decrease, causing first a current in one direction and then the other; similarly for Q. If, now, we suppose a large number of such rings, we finally get to the case considered.-EDS. ELEC. REV.

J. MACDONALD.-You are quite correct. The conditions of correspondence will apply equally to the person who commences a controversy and the one who replies.-[EDS. ELEC. REV.]