erosion extends over the circumference of to the rear of the last position. (3) At the rear of the cone-the gun being taken as a smooth-bore, with the powderchamber larger than the bore, and connected with it by a cone.

the bore, at the positions of the two zones, but the top is always most injured. Other things being equal, the erosion de velops as much more rapidly as the calibre is greater, and more quickly in bronze than in steel. Tool-marks or defects in metal in the upper part only of the powder-chamber are increased rapid ly. After prolonged firing, an increase in the diameter of the bore is found chiefly in the shot-chamber, where the greatest erosion occurs. At the same time the ring of the projectile advances further up the cone, joining the powderchamber to the bore.

The length of service of a gun is almost always limited by the loss of velocity and accuracy occasioned by erosion.

The wear of the rings, almost nothing at the commencement of firing, increases with the service of the gun.

For the first round of each series in a gun, the wear of the ring is about double that of the succeeding rounds (Report 826, 1879).

Chapter III. is a theoretical study of the mode of production of erosion in the bores of guns made since 1870.

The author states that when a gun is fired, it expands under the action of the powder-gas, which exerts a certain tension on the interior of the bore in a circular sense. When the gun is in one piece, this expansion of the bore is proportional to the calibre and interior pressure; according to the formula of General Virgile it is in inverse ratio of the modulus of elasticity, and diminishes when the thickness of the tube increases. But large guns are composed of several pieces, which, if they have the same modulus of elasticity, require a fresh formula to express the relations of the circular tension and interior pressure on the bore, owing to the shrinkage employed to build up the gun. This formula is given, and the author, with the object of proving that the erosion commences in the upper part of the bore at the point occupied by the driving ring of the projectile when the maximum pressure occurs, and that all things being equal, it is greater as the calibre increases, proceeds to consider the ring of the projectile in three positions in the bore: (1) In the forward part of the cone at the commencement of the bore. (2) A little

The second zone of erosion is next dealt with, this being a consequence of the first. The theory that the principal zone is caused by blows from the projectile, is questioned, as in this case the lands should suffer most, whereas it is the grooves that are most injured; it is not denied that blows are given by the projectile, but it is suggested that these will be reduced, as also the escape of gas, by increased forcing of the rotating rings. In considering the wear of the driving rings, on the hypothesis that the walls of the gun act as a file, for helicoid rifling the author gives a formula showing the wear to be proportional to the tangent of the final inclination of the grooves and the square of the initial velocity, and considers it advantageous to increase the number of the grooves. To diminish the wear as much as possible, and to preserve the ballistic properties of the gun, the author suggests a parabolic form of rifling, in which the pressure will be little at the commencement of movement, that is, at the parts most eroded; but in choosing a form of rifling, the wear, and also the maximum circular pressure exerted by the lands on the ring, must be taken into account.

Suggesting an interior design for a gun, the author adopts for the shotchamber a cone sufficiently long for the ring to be tightly in contact at the moment of maximum pressure. The angle of this cone has been approximately determined by experiment, the length should be a little more than the distance, which in guns of the same calibre already proved, separates the initial position of the driving-ring and the mass of the erosion. The suggested modifications entail an increase in the diameter of the fillet of the driving-ring, which must be fixed by experiment.

The simplest solution from the manufacturing point of view consists in making a second cone to follow the cone of supports for the ring of a projectile, this cone having a greater inclination, the commencement of the grooves being well in advance of the driving-ring of the projectile when sent home.

SOME POINTS IN ELECTRICAL DISTRIBUTION.

BY PROFESSOR GEORGE FORBES.

From the "Journal of the Society of Arts."

In a parallel system, each lamp is attached by wires to the two main conductors, and thus to the two terminals of the dynamo. This is its characteristic feature, viz., that the lamps are all metallically connected with the terminals of the machines.

WHEN I had the honor, just a year the use of secondary batteries and secago, to give a course of Cantor Lectures ondary generators. Each of these systo this Society, on the distribution of tems is capable of being varied in the electricity, the subject had not till then details of its application, and each is been treated in a systematic manner. I specially applicable to particular cases. had then occasion to speak in strong terms of the methods habitually adopted, and subsequent experience has confirmed me in the justness of my criticisms. I gave a fairly complete account of the practical systems of laying electric mains which were at the disposal of the engineer when designing a system of dis- In the series system, the conductor tribution; but in the course of only three going through the district has its two lectures it was impossible to do justice ends connected one to each pole of the equally to all systems. When I was this dynamo. But wherever there is a lamp year asked to read a paper before the the conductor is severed, and each terSociety, I eagerly availed myself of the minal of the lamp is attached to one of opportunity to take up the straggling the severed ends. threads, and say a few words on the subjects which were previously but inadequately treated.

In the parallel-series arrangement the district is divided into sections. The first section and the last section have Allow me to remind you that, in my each a conductor attached to one terminal lectures, I described five different ways of the dynamo. The other conductor of of laying the mains, each one depending the first section extends also through the upon a different method of connecting second section, but no farther. Another the dynamo machines and lamps. Three conductor extends through the second of these may be called direct systems, and third sections, another through the and two indirect. Each system has spe- third and fourth, and so on. In each cial advantages when applied to suitable section the two wires from each lamp are cases, and when an engineer is designing attached to the two conductors of that a scheme of distribution he carefully section. considers the pros and cons of each system as applied to the special case, and very often he has to go through the whole calculations of the scheme according to several different systems. And in very special cases. here let me repeat, what I have often is at present only used for arc lights, stated before, that no amount of labor sometimes with the insertion of a few and expense in these preliminary calcula- glow lamps in parallel to replace an arc tions should be grudged, as the ultimate lamp. saving effected by the careful selection of a suitable and economical system may easily range to 50 per cent. or more on the total cost of the installation.

The first, or parallel system, is generally most suitable for small installations, but the cost of conductors becomes extravagant when the area is extended, except The series system

The third system is admirably suited. in every case where the number of lamps in use at any time is constant, and may perhaps be modified so as to be applicable to other cases. Smaller conductors can be used on this system than on the first, and this effects a very great economy.

arc, series, and multiple series. I prefer the terms now advocated.

Having recalled these points to your economical principles. This regulation notice, I need say no more, as I treated must be accomplished by altering the these three systems and their modifica- length of the filament in the lamp. This tions very fully in my Cantor Lectures. has never yet been done, but if inventors I thought I had spoken strongly once realize the importance of the object enough about the enormous cost of the to be attained, I feel sure that we shall main conductors, and the necessity for not have long to wait for the solution of selecting that size of conductor which is the problem. Now I am delighted to be the most economical. We must not use able to inform you that during the past too small a conductor, for then the waste year at least one inventor has been applyof energy in heating the conductors ing himself to the placing of glow lamps would be enormous. Nor must we u-e in series. Mr. Bernstein, who has paid too large a conductor, for this means so much attention to the system of glow buried capital. We must select that size lamps, has worked out the problem of which balances one cause of loss against using lamps of low pressure and large the other, so as to give the maximum current in series. He has boldly taken economy. extreme measures, and proposes to use lamps of 6 volts and 10 ampères. And here he has got over a difficulty I referred to in my Cantor Lectures. I said that there was no economical way of preventing a broken lamp from extinguishing all the others on the series. Passing part of the current through a shunt whose resistance is 100 times that of the lamp would be very expensive if we had in use lamps of the ordinary type, but with a lamp whose resistance is a fraction of an ohm, the expense would be trifling. When the lamp breaks, the current in the shunt is increased so as to release a contact which short-circuits both the lamp and the shunt.

I find, however, that, while this axiom has been accepted as such in this country, some writers from America still try to uphold the fallacious views which have hitherto been in vogue here. I have no ticed more than one criticism from America, by people asserting themselves to be connected in some way with some of the Edison companies, who protest against my statement that the three-wire system effects an economy of copper of only 25 per cent. over a two-wire system. The difference between us is that I start with the axiom that we are to use the most economical size of conductor. My critics start with the assumption that they shall use a conductor which shall I still think, however, that the loss of waste exactly 10 per cent. of the energy. energy by heat conduction at the terIn this way they use a conductor which minals may be very great in lamps of so is not the most economical, and they low a resistance; Mr. Bernstein tells me show a saving of 62 per cent. of copper it is ten per cent. Still I can well foresee over the two-wire system, but with a pos- that, even if a very sensible loss of energy sible extra waste of energy of 62 per occurred, the economy and convenience cent. I said in my Cantor Lectures, and of using such a system might be amply I repeat it this evening, that the manipu- sufficient to warrant its adoption in many lation of figures may suit American cases. financiers, but it is not electrical engineering.

I congratulate Mr. Bernstein on his work in the past, and would urge him to develop this idea more fully, and I can assure him that I, for one, will give the fullest consideration to the system when preparing the plans for electric light installations.

Before proceeding to discuss the indirect methods of distributing electricity, which is the main point I wish to touch upon to-night, I wish to supplement my remarks on series lighting by a statement of what has been done in the last year. I should like also to take this opportuI expressed a hope that more attention nity of congratulating other makers of elecmight be given to the lighting of houses trical appliances on the advances which by glow lamps in series. I further have been made since the date of my showed that it was only by means of a constant current flowing through glow lamps in series that the regulation of each lamp could possibly be effected on

Cantor Lectures. I should like especially to congratulate Mr. Swan on the improvements he has made in the manufacture of glow lamps; the Messrs. Siemens

I have many of these appliances on the table, which any members of the audience may examine at the conclusion of the paper.

and Messrs. Crompton on their success-rent and the friction be proportional to the ful construction of very large dynamos; square of the velocity, then the number Messrs. Elwell and Parker on their great of turns completed in any time is a diimprovements in the details of construc- rect measure of the total quantity of tion of dynamos and motors; and the electricity which has flowed through the Electrical Power and Storage Company meter in that time. The marked imon the continued improvements in their provements introduced into this meter by secondary batteries. I would equally Mr. Ferranti consist in the perfection of congratulate Messrs. Willans and Cromp- the magnetic circuit, and the introducton for the continued improvements in tion of an almost frictionless mechanism steam-engine governors, for keeping the for correcting the revolutions, and showcurrent from a dynamo or the pressure at ing them on dials like a gas-meter. any point in the circuit constant; likewise Mr. Ferranti for having brought out an electric current meter which promises to fulfil all the wants of engineers and the public in this direction; and last, In referring to my Cantor lectures, let but by no means least, Messrs. Wood- me make a few remarks about a double house and Rawson for various new de- table which I introduced to assist engivices which will go far to assist in de- neers in calculating the size of mains resigning and carrying out many schemes quired in any installation. These tables of lighting; and I would specially men- have been copied into a number of techtion their automatic cut-out, adjustable nical journals and books of reference, but to any strength of current, which, though in nearly every case there is no word of not so new as some other devices, has explanation how to use them. I have only been brought prominently to my found these tables of such paramount notice during the past year, and whose convenience, that I think it right to call simplicity and reliability give me great attention to them once more. The ecoconfidence in employing it. I must not nomical size of main depends upon three fail to include here, as a subject for con- variable factors-the price of laying down gratulation, the rising esteem in which a ton of copper mains, the interest to be secondary generators are held, and the charged on the capital sunk in mains, improvements which have been made in and the value of an electrical horsetheir manufacture. power for the number of hours that it is to be utilized during the year. My table is divided into two parts. Look along the line in the second part, which relates to the percentage on capital which you propose to use, until you come to the column relating to that price of copper which you propose to use, and you then find a number printed which we may call the auxiliary number. Now take the first part of the table and look along the line which relates to that value of a yearhorse-power which you propose to use, until you find the auxiliary number; the heading of that column of figures tells you the size of main per 1,000 ampères which is most economical to use according to the data you have employed.

When I add to this list the admirable lamp-holder of Mr. Alfred Swan, Varley's adjustable carbon resistances, and other small but useful accessories, I think you will agree with me that the facilities at the disposal of the engineer for designing a satisfactory scheme of electric lighting, and which have been added to his store during the past year, are very abundant.

About one of these appliances I should like to add a few words :

First, I may say that the Electric Lighting Act was rendered abortive, if for no other reason, simply for the want of an electric-current meter which was in all points satisfactory. In the year 1882, a patent was taken out for a meter in which the current passing round an electro-magnet created a field where mercury was placed, which rotated when the current passed through the mercury from the center to the exterior of the field. If the magnetism be proportional to the cur

I will now remind you of the position in which we were left after a full discussion in the Cantor Lectures of the three direct systems of distribution. The paraliel system, or the three-wire system, is admirably adapted in cases where the farthest lamp of the district is within

two hundred yards of the central station. So soon as you go to greater distances, the cost begins to increase in an alarming ratio. Hence this system would be applicable to a large district only by having engines and dynamos put down at a very large number of small stations. Now, a station is not worked economically unless the engine-driver and stoker are fully employed, and the maximum economy cannot be reached with less than about 1,000 horse-power. Thus, the parallel system is only applicable to cities so densely populated as to have a consumption of about 10,000 16-candle power lamps in a radius of two hundred yards. But in most districts of this class space is valuable, and a large number of such stations would be very expensive. We must then, in the parallel system of distribution, be content to work at a cost which is above that of the ideal of economy. There are many cases where the great simplicity of such a system does actually make it worth while to sacrifice this ideal of economy, and when an engineer is consulted on such matters, it is his duty to estimate the exact loss from not strictly following the ideal economical rules before adopting or reject ing this system. In order to make a parallel system effective, the central station must use distributing boxes, the methods of placing which are described in my Cantor Lectures.

at the conclusion that at this time there is no parallel-series system of distribution which is quite satisfactory. I do not say that such a system cannot be devised, but I do say that at this moment no such system is available to the electrical engineer, except in the case where the same quantity of electrical energy is consumed in any district at all hours during which the current is being used. There are some cases where it will pay to insert equivalent resistances when lamps are put out. These cases are rare, but they must not be lost sight of by the engineer.

Having in the Cantor Lectures discussed all the direct systems of distribution, I will complete my outline of this part of the engineer's profession by an examination of the two indirect methods which I formerly touched upon only lightly, viz., secondary batteries and secondary generators.

SECONDARY BATTERIES.

During the electric boom (as Americans would call it) of 1882, the secondary battery took a prominent part, not so much for what it had done as from what was expected it would. It was said that since no one would think of sending gas from the retorts without storing it in a gasholder, so the same must be done with electricity. Every one will admit that it is most desirable that this should be The cure for all these evils would be done, and I am glad to say that secondfound if a thoroughly satisfactory system ary batteries have been improved to such on the parallel-series plan were known. an extent that there is every reason to In that case, high tension electricity hope that this will be done, always promight be used, involving smaller conduct- vided that the companies who supply Thus, the central station might be them, do not expect the public to repay at a distance, where land is cheap, and it them for the hundreds of thousands of might be of any size, so that all the works pounds which may have been thrown would be in the same place, as they are away in the past. in gas distribution. Here the only departure from the ideal of economy lies in the length of the conductors to the district which has the supply. But these conductors are comparatively small, and no engineer would recommend the adoption of such a system unless he found that the economy of having the works out of the expensive district counterbal anced the extra expense in mains.

ors.

Now I regret to be compelled to say that, after discussing all the ingenious plans on the parallel-series system proposed by Mr. Edison and others, I arrived

I have been given to test about a dozen different kinds of secondary batteries, of which I have found some four or five to be good, and the others mostly worthless. I can say this, that at the present moment secondary batteries can be supplied by certain makers which are to be relied on with careful usage, and which give us 70 to 80 per cent. return for the energy expended.

Now, it might be asked, what are the advantages of storing the energy if you lose 25 per cent. by doing so? The first advantage is that a breakdown in the ma