by some electrical engineers. There is another way of mildly explaining these differences. It is that a number of minor, younger, or country firms are extremely anxious to get a piece of work which may prove a good stepping-stone to something in the future. Business push of this kind is commendable, but if the work is undertaken at a loss, the system is a hazardous one. Men nowadays are kept tolerably well acquainted with what this or that work costs, and commencing at a low price and gradually putting up the price is apt to lead to loss instead of gain of business. There must have been some undercurrent of reason to account for the different figures. It may be news to some of our electrical friends to hear that many authorities keep a "black-list," and once a firm's name gets upon the list all chance of obtaining work from that particular source is gone. We have assumed that each contractor tendered for the best work and the best materials, otherwise other suggestions might be made as to the differences. A large part of the electrical work of the future will be in connection with local authorities, where prices cannot be kept secret as with private individuals-hence it will be well to pay closer attention to the proper methods of making out tenders.

THE FINANCE OF THE LONDON COUNTY COUNCIL.

We all know what happens when a ship is sinking the cry is sauve qui peut. A sinking ship and a senile Council have much in common. For the latter, the coming election looms ominously in view. There are cases, however, in which bravado takes the place of wisdom, and this seems to be a feature in many of the acts of the expiring London County Council. The reason may be that many of the members know what the future has in store for them. They will pass into a limbo whence the electors will not recall them. Not long since we recorded the peculiar action of the Council in regard to the St. Pancras loan. We are glad to see that St. Pancras has been independent of the Council. It will be remembered the Finance Committee recommended the loan; the Council though admonished as to importance of time, referred back the report. Again the committee recommended the loan, and the Council agreed to the report, but meanwhile St. Pancras altered its mind and now said to the Council, No, thank you. The granting of a loan by the County Council is no favour to St. Pancras. The latter is not a supplicant nor to be treated as such, and has quite rightly snubbed the Council. It was part of the duty of the late Metropolitan Board of Works to loan out money to local authorities; it is part of the duty of the County Council to do the same. There is no favour conferred by so doing-rather doing rather it is a right. Similarly the County Council are riding the high horse with the School Board, but here again their action is landing them in ridicule. Perhaps the constituencies will not have forgotten these matters

some four months hence, and will plainly tell the new councillors that they are elected for certain business purposes, and not for the purpose of airing fads or throwing obstacles in the way of local improvements.

CORRESPONDENCE.

"One man's word is no man's word
Justice needs that both be heard.'

THE ELECTRIC OMNIBUS.

SIR,-In reply to W, "Where is the electric omnibus," I beg to suggest to him that the duty placed on a locomotive engine is so heavy that the said engine is shut out of the market. However, here is the law a locomotive engine may travel along the road on condition that an attendant is attached to it, and also that a man walks in front carrying a RED FLAG (size of flag ad lib), the pace not to be more than four miles per hour. Fancy a hansom cab subject to those conditions, "four miles per hour and man with red flag in front." Oh! thou free and enlightened country. Now the way out of the difficulty is this: the general election is close at hand; let each of us interested in electricity interview his respective candidate, and get a promise from him that he will assist in getting the law amended. It not being a party question, it will apply to both sections of members of Parliament.-Yours, etc., C. MARSON,

Member of the Vestry of St. Mary's, Battersea. 22, Swaby-road, Northcote-road, Clapham Junction, December 13th, 1891.

[Mr. Marson has evidently not read his Electrical Engineer, or he would have learnt that the legal aspect of the case is not as he states. A license can be obtained for electric vehicles, and, in fact. an electrical omnibus has often traversed the principal London streets.-ED. E. E.]

D.P. ACCUMULATOR CELLS.

SIR, We notice in your last issue you publish the paper read by Mr. Robertson at the Society of Arts without the discussion, and as the description of our cell is likely in its present form to create a bad impression, we trust you will allow us space to correct the statement that our positive plates are of the Faure type. As pointed out in the discussion by Mr. Drake, the plates, both positive and negative, are of the Planté type, no oxides or salts of lead being employed in their construction.-Yours, etc.,

London, S.W., 11th Dec., 1891.

DRAKE AND GORHAM.

DREHSTROM.

SIR,-Under the above heading you published last week a paragraph stating that I am about to make tests with a multiphase current set. This is misleading. The set in question has been very kindly lent me by the Berlin alternating currents at the School of Military Engineering Electrical Company for illustrating one of my lectures on at Chatham, but I have received no instructions to make tests with it.-Yours, etc., GISBERT KAPP.

Westminster, Dec. 15, 1891.

GWYNNE AND CO.'S DISCLAIMER. SIR, Our attention has been called to a letter which

appeared on page 563 of your last issue, signed by Messrs. Gwynne and Co., of Victoria-embankment, informing you that they were not shareholders in our company. This disclaimer is quite correct so far as the firm of Gwynne and Co. is concerned, and we are somewhat surprised that they thought it worth while to rush into print to refute a matter which in no way concerns their firm unless it was done by them with some ulterior object, and we would remind these gentlemen that there are more "Gwynnes" than one

engaged in the engineering world. However, to put this matter right, from a financial point of view, and so far as our company is concerned, we beg to state that the particular Mr. Gwynne who is interested in the success and welfare of our company is Mr. James Eglinton Anderson Gwynne, J.P., of 97, Harley-street and Folkington Manor, Polegate, Sussex, and the owner of the Pilsen Electric Works, who, moreover, since the date of the first announcement in your 66 City Notes," has also taken over our Mr. Arthur Shippey's entire interest in the company, also Mr. Shippey's vested interest in the Curtiss, Crocker-Wheeler, and Waterhouse electric systems; and we have no doubt in due course that Mr. Gwynne will financially handle the above valuable group of inventions in the manner they deserve to be developed in the European markets, and this probably without the assistance or intervention of the firm alluded to. Knowing that your journal has a known reputation for inserting authentic information, and that it also aims in carrying out as far as possible your well-known motto for correspondents, we will thank you to insert this letter in your next issue, and by so doing you will oblige. Yours, etc., SHIPPEY BROTHERS, LIMITED, (Per William B. Hammond.) 13 and 14, Cheapside, E.C., Dec. 17, 1891.

UNDERGROUND LIGHTING MAINS IN PARIS.* BY E. DIEUDONNÉ.

(Concluded from page 512.)

Mains of the Compagnie de la Place Clichy.-The system of mains adopted by this company is essentially different

the Société Alsacienne de Constructions Mécaniques. The author of these articles had the good fortune to see the manufacture of this class of electric cables at the works of Messrs. Siemens themselves at Charlottenberg, near Berlin. The process of insulation is extremely carefully carried out, and the insulated cables are covered with lead. They are classed ordinarily into four different types, corresponding to various inherent requirements of employment combined with economy:

(a) Conductors with lead sheath, unprotected with exterior envelope.

(b) Conductors with lead sheath, covered with a serving of asphalted jute.

(c) Conductors with lead sheath, covered with layer of insulation, again wrapped by two iron strips wound spirally and simultaneously around the cable, Fig. 57. The whole is finally covered with tarred jute.

(d) Lastly, those in which the external wrapping is of iron wire.

Among these four types of cable the mains engineer will exercise a judicious selection. When the section of the conductor does not exceed 30 square mm., a single copper wire is used; for larger sections the core is formed of stranded wires. The insulation is composed of a serving of cotton or jute, highly dessicated in vacuum, and then impregnated with an insulating compound the description of which is not divulged. The proprietors of the process affirm its superiority to indiarubber, guttapercha, and similar materials, a claim which must so far remain for us a simple allegation. The thickness of this insulation depends

in many ways from all the others laid in Paris. Here there are no conduits of cast iron, concrete, earthenware-and no bare cables. The cables are all heavily insulated, and simply laid at the bottom of a trench under the pavement on a layer of sand. [It is perhaps as well for English readers to note that the subsoil of Paris is of an entirely different nature to that usually experienced in the damp soil of London, consisting usually of light, dry, stony earth, like crushed stone.] The laying of the cables becomes in this case a very rapid operation, the time of interfering with the pathway is considerably diminished as compared with the time required for the other systems described. To open the trench, lay the cables, and fill up again quickly, are the three phases of this system. The junctions and connections are made, under a portable shelter, in cast-iron boxes, as afterwards described.

The cables employed by the Compagnie de la Place Clichy are of the Siemens type, manufactured at Belfort by

*From L'Electricien.

FIG. 60.

upon the voltage it is to withstand. Amongst the strands forming the cable is an insulated strand of smaller section, which acts as a pilot wire to measure the difference of potential at the extremity of the cable.

These mains, although sufficiently armoured with their iron sheathing, when laid in their trenches and covered with a layer of sand are, nevertheless, liable to various eventualities; such as a violent stroke from a pickaxe, if there is nothing to call the attention of the labourer to their presence in the ground. It is thought sufficient generally to lay over the trench, just below the pavement, a galvanised iron wire meshing, which would serve to attract the attention of any workman taking up the flags. In many cases this would appear insufficient. Occasionally it would be better instead of the metallic meshing, to use creosoted deal boards. When passing under the roadway the cables are threaded through iron pipes.

The distribution in the district of the Place Clichy is on the five-wire system, supplied by one or several groups of

separately joined. After the junction-box is bolted up, it is filled with melted insulation through a hole in the cover. The junctions of the copper are not soldered.

Fig. 60 shows a tee junction-box based upon the same method of connection.

At points where the conductors ramify, distributing-boxes are installed of the type of those shown in Fig. 61, which gives the example of a right-angle connection with the addition of two feeders. These are made of cast iron, with hermetically sealed covers; junction-pieces connect the cables which come into the box through lateral stuffingboxes, at which point the cables are bared. The end of the cable is fitted into a socket fixed by pointed set screws; and the whole is covered with an indiarubber sheath. The space left free around the cable is filled up with insulating material, poured in at several times to allow for contraction The centre of the box is occupied by metallic bars by which any cable can be put into communication with any other. Cut-outs are inserted between the mains and these bars. The cover itself of the box is surmounted by a further hollow and elevated lid, the upper surface of which is level with the footpath.

The cables are carried in concrete conduits 45 cm. (17in.) wide and 30 cm. (11in.) deep, Fig. 63, on wood frames, B, which are furnished with enamelled metallic hooks. These double frames are embedded in the concrete at distances of about one metre. The culvert is closed by slabs of the same material, the joints filled up with loam, after which the path is relaid above them. As in the other systems already described, the crossing of the streets is carried out by vaulted galleries ending in vertical manholes.

The high-tension mains in which alternating currents of 2,400 volts are carried, are also placed in concrete conduits, but instead of being supported by hooks, they are run in wood casing in order to avoid all accidental contact. Fig. 63 shows the position of the cables.

The specification of these will be of interest and is herewith given as follows:

1. Low-Tension Mains.-Metallic core of tinned copper, a layer of pure indiarubber, a layer of mixed indiarubber 2 mm. thick; two ribbons of indiarubber tape, vulcanised; a serving of bitumenised braid.

2. High-Tension Mains.-Metallic core of tinned copper; two layers of pure indiarubber 1 mm. thick; several layers of mixed indiarubber of 4 mm. thick; a layer of resinous hemp 3 mm. thick; two ribbons of cotton tape.

The only difference between the two cables is in the thickness of insulation, as seen in Fig. 64. The insulation required by the specification was 300 megohms per kilometre (about 190 per mile) for the low-tension, and 8,000 megohms per kilometre (about 5,000 per mile) for the high-tension cables at 24deg. C.

Along a portion of the route the high-tension main, passing near the telephone wires, it has been necessary to avoid induction noises by employing concentric cables.

Braid

FIG. 63.

Fig. 62 shows both the arrangement of tee junctionboxes, and the distributing-box to which the house service mains are connected. There are several different types of junction-boxes, the form varying according to requirements; but the system of construction and laying of the mains is always based upon the same general method.

The Company of the Place Clichy, while confident of the good insulation of its armoured cables, has multiplied its precautionary measures at those points at which leakages and faults mostly occur in underground mains-namely, at the junctions and branchings-so as to thoroughly assure the necessary safety of working.

Municipal Station of the Halles Centrales.-The technical commission which directed the erection of the Paris municipal station at the Central Markets, decided to use both continuous currents and alternate currents with transformers. According to the choice of one or of the other system, the mains used have been varied. They will therefore be here considered separately.

The distribution of the continuous current in the market and the neighbouring streets is carried out by means of the three wire system and feeders. The difference of potential between the extreme wires does not exceed 220 volts.

Within the markets none of the mains are underground except the feeders; the distributing cables are run along the walls of the stalls, protected by wood casing.

FIG. 64.

Pure Indiarubber

We have thus given, in the course of a somewhat extended review, with abundance of illustrations, and in as short and direct manner as possible, the various systems of underground mains adopted by the electric light companies of Paris. Which system will prove in all respects the best, time alone can be trusted to decide.

THE MEASUREMENT OF THE RESISTANCE OF CONDUCTORS CONTAINING DISTURBING E.M.F.'S.*

BY ROLLO APPLEYARD.

The principal practical case in which we have a conductor containing a disturbing E. M.F. is a telegraphic circuit terminating in earth-plates. These earth-plates, from various causes, act as a battery, and the resistance of the line appears too small or too great, depending on the direction of the natural currents with respect to the working current. In order that the position of a fault may be accurately found, it is desirable to know the resistance of the earth-plates and the value of the natural E. M.F. As it was with earth-plates that my first experiments were made, I propose in this paper to give you the results obtained with them in full. For the sake of clearness, it has been found necessary to refer to much that is elementary. In this I must ask for your kind indulgence.

Let us first be sure of what we mean by the resistance of an earth-plate. Taken alone, the term does not quite explain itself. Paper read before the City and Guilds Old Students' Associa

tion.

Schwendler ("Testing Instructions," vol. ii.) defines it as the istance offered by our planet between the earth-plate and other earth-plate of very great size, situated at a very great stance. It is there also shown that if we accept this definition are justified in assuming the resistance between any two earthates as equal to the sum of their respective resistances. If then we are given three earth-plates of unknown resistances, y, and z, it is possible, by connecting them in pairs, to find their dividual resistances. Let

The second experiments were made with Wheatstone's bridge in its usual form. The difficulty was precisely the same. The E.M.F.'s in the "unknown "arm of the bridge run themselves down through the other branches. The accuracy depends here, as before, upon e keeping constant.

The results were, if anything, rather worse than by the tangent galvanometer, the smallest resistance nearly always appearing with a negative sign. The wonder is that such methods still find a place in manuals on telegraphy. Schwendler's book suggests that these tests are best carried out on Sunday. The fact that mine were made on all days of the week may explain their inconsistency.

If a nearer approach to accuracy was to be obtained it became evident that a means must be found for diminishing, as far as possible, the time during which the plates are on circuit at the bridge. The following method suggested itself:

Let m, n, p, q, represent the four corners of the bridge; a and b the ratio arms; d'the variable resistance; and x the unknown resistance which contains the disturbing E. M.F. At q, the meeting point of the battery branch with x and d, separate the three branches, and bring a connection from each to the upper three

This gives us a. The values of b and c are similarly arrived at, and these results are said to give the earth-plate resistances. A series of tests was made by this method with three "earths," of which was the resistance of the gas-pipes at the place in question; y, an earth-plate to a lightning conductor; and z, a copper earth-plate of ordinary type. The following values for the individual resistances were obtained at different times within a few weeks.

FIG. 3.

springs of a five-contact key. The lower springs, insulated from the three above, are for the completion of the galvanometer circuit. The conductor whose resistance is to be determined is at present on open circuit. Now adjust the arm, d, to some assumed value of x, and verify it by a single tap of the key. The E.M.F. in the a branch cannot appreciably vary during that short interval, and by a few trials the correct value of d is determined. The testing battery is now reversed, and a new value, d1, obtained for the variable arm. For general purposes, it will be found sufficiently accurate to take the mean of d and d' as the required resistance of x. If the disturbing E.M.F., however, is high, we must use the whole formula for the resistance of the bridge arm and make the necessary correction. In this case let EE.M.F. of testing battery.

E.M.F. in the x branch.

e

= the resistance in the six branches.

=

the corresponding currents in a, b, d, x, f, g.

E

Schwendler's tangent galvanometer, as made for the Indian telegraphic department, was employed, with two Minotto cells connected in parallel.

The absurdity of resistance with a negative sign should sufficiently show the fallacy of this method; and it will appear the more inconsistent when I refer you presently to results which make the resistances nearly constant throughout the same period of time. The variations are not due, as is generally assumed, to changes in the actual resistance of the plates themselves, but to the faulty method. We cannot take it for granted that the E.M.F.'s remain unchanged when the circuits are once completed. e in equations (3) and (4) is not constant.

FIG. 4.

Then when G O there is the same difference of potentials between the ends of a as between the ends of b. A b B a Considering point m, X=A

..

a

B = A

b

", ", D= B.. D≈ A. D= A/

, p, F=A+B.·. F= A

(4)