occurring during the detonation of explosives free to air, it will be prudent to avoid firing shots in the mine, even with charges considered the safest, at points where the mixture of air and firedamp is inflammable. The choice of explosives must be considered as diminishing danger, but not as absolutely suppressing it.

5. It is necessary to employ the explosives under conditions such as to develop from them the maximum useful work. Economy and security are in accordance to recommend this rule. To accomplish this the following conditions are necessary. The explosive must be rammed with care and the hole must be sufficiently deep. No void space must be left either in front, behind, or round the cartridge. The Bickford fuze must not be placed in contact with the explosive if it is used, and the dangers of the fuze are sufficiently great to make it desirable to replace it by some more certain mode of ignition.

The commissioners further remark that their conclusions led to abandoning the use of blasting-powder in mines where firedamp is known to exist, and even to place under suspicion ordinary dynamite, blasting-gelatine, ammonia dynamite, such as is actually manufactured. Of these, blasting-gelatine appears to be the most dangerous. The explosives which give greatest security are the binary mixtures of dynamite, gun-cotton, bi-nitrobenzine with nitrate of ammonia; but the best mode of manufacturing and protecting these mixtures from atmospheric moisture has still to be experimented on. The breaking up of the coal and rock by them. also requires practical study. The commission recommend the government to prepare sample cartridges and issue them to mining engineers who may be willing to conduct practical trials with them in ordinary work.

To the main report is added a supplementary one dealing more minutely with the conditions of the explosion of firedamp, and in the conclusions it is stated that the temperature of inflammation of firedamp is 500° Centigrade, but it is necessary that the action of this temperature should be prolonged to produce ignition. Because of this fact, and the almost instantaneous mixture of the products of combustion with the atmospheric air which causes them to cool rapidly, explosives in which the temperature of explosion is less than 2,200 Centigrade are incapable of inflaming firedamp mixtures when detonated under normal conditions.

The hole, however, must be carefully tamped, as the greater the imperfection of wadding, the greater the danger, and free explosion in air is the most dangerous of all. In the supplementary report the experiments with binary mixtures are detailed minutely, and it is again stated that nitrate of ammonia is the best substance to be used in connection with the explosives. The reports are illustrated with figures of the apparatus used, and numerous Tables are given.

D. C.

Experiments on Atmospheric Electricity.

By Dr. LEONHARD WEBER.

(Elektrotechnische Zeitschrift, 1889, p. 521.)

=

=

1

The Berlin Society of Electricians commissioned the Author to investigate experimentally the electrical condition of the atmosphere in various states of the weather. Three reports have already been made, of which the last dealt with the variation of the potential of the air with increasing elevation when the weather was clear, whilst the present report deals with the same subject when the sky is clouded. The Author finds that on clear days the potential rises with the height of the point of observation from the surface of the earth. He accepts Peltier's theory, according to which the earth contains a negative charge of electricity, and he finds that the potential V of a point h metres above the surface of the earth is related to the potential V1, of the earth itself, d V according to the equation V1 R, R being the radius of the dh earth. The potential found for points 350 metres above the surface of the earth was 96,400 V, and the mean value of the differential quotient was found to be 275 V per metre altitude. From these figures he deduces the potential of the earth as V1 = −1·72 × 109 V, or, in electrostatic measure, V1 5.8 x 106 C.G.S. units. The charge of the earth is in electrostatic measure 3.7 × 1015 units, and the density is 0.00072 units per square centimetre for smooth portions of its surface. For prominent points, such for instance as the top of the Eiffel Tower, the density is very much greater, and drops of water or other small particles of matter coming in contact with such points will receive a negative charge and be repelled with measurable force. The Author next assumes that there is electric radiation between bodies of different potential analogous to the radiation of heat between bodies of different temperatures, and inclines to the belief that such radiation may even take place between heavenly bodies. On these suppositions he finds that negative electricity is conveyed to the earth from the sun by radiation, whilst the earth dissipates constantly an equal amount into space. The highest and lowest parts of clouds, and the particles of matter floating in the atmosphere are most instrumental in the process of radiation, and the dust particles floating in the lower regions of the atmosphere assume a negative charge which they give up to ascending particles of vapour. Clouds may be considered as conductors, the lower side being positively, and the upper side negatively electrified. The total charge may be positive or negative, according to circumstances, and the Author thinks it probable that generally the former will be the condition of snow-clouds, and the latter that of rain-clouds. When a cloud of considerable vertical extent is sheared horizontally asunder by the action of air currents, there are formed two clouds

of opposite charge, which, if passing in succession through the zenith, give rise to the rapid changes of potential often observed during thunderstorms. In confirmation of these views the Author gives a series of Tables, containing the results of experiments made during 1888 with balloons and kites, the altitude varying from a few metres above the surface of the earth to about 400 metres. G. K.

New Type of Alternating-Current Motor. By F. J. PATTEN. Paper read before the American Institute of Electrical Engineers, New York, Sept. 10, 1889.

(The Electrical Engineer, New York, 1889, p. 424, 6 Figs.)

The Author describes a new type of motor for alternating currents designed by himself, and expresses the problem to be solved thus :

1. A machine that will start itself, independently of the speed of the generator or number of alternations of current per unit of

time.

2. A machine that has but one direction of rotation, and cannot reverse under any conditions of current alternation.

3. A machine that is not necessarily synchronous with the generator, revolution for revolution.

4. A machine in which reversals of current-direction do not produce corresponding reversals of magnetism in any iron part when the machine is in motion, at its normal speed and maximum efficiency.

5. A machine of simple form, having an ordinary continuous wound armature revolving in a single or two-pole field.

If a Gramme dynamo of the ordinary type be used as a motor, and a direct current supplied to the brushes, then the armature becomes polarised in a constant direction, and will turn in a constant direction, supposing the polarity of the field-magnets to remain constant; but if the continuous current be replaced by an alternating current, then the polarity of the armature, and therefore the tendency to rotate, is reversed with each alternation of current, supposing the field to remain constant; but it will be noted that the motion would be in the same direction still if the field were reversed by the same reversal of current. If, however, the field remain constant, and some method of reversing the brushes at each reversal of current be found, the polarity of the ring would remain constant; it is impracticable to reverse the brushes, but the same effect can be produced in the following manner. In an ordinary Gramme ring, the point between two coils is joined to the collector-bar immediately under it; but if coils 1, 3, 5, 7. . . be joined up in the usual way, and coils 2, 4, 6, 8 be joined to the bar diametri

cally opposite to the usual one, and we make the supposition that the ring shall turn through an arc equal to that covered by one bar of the collector during each alternation of current, a constant polarity will be maintained at the upper and lower points of the ring, without causing the brushes to change position mechanically. Its tendency to motion, then, in a constant field, would always be in the same direction.

The fundamental principle which underlies the construction of this type of machine is that, "The poles of any closed circuit may be maintained constant with an alternating current by causing opposite impulses to traverse the circuit in opposite directions." In order to obtain a shunt-current for the field-magnets, another collector is supplied, to which the ring is coupled up in the usual manner; and from the bars of this collector the connections are made to that just described, and by this means a constant field is obtained, and a constant polarity of ring, so that a constant direction of rotation is secured, provided the ring moves at such a speed that the brush touches the next collector-bar at each change of direction of current. Supposing that the machine does not move at such a speed as to fulfil the above requirement, then the polarity of ring and of the field both change together, and the machine becomes simply a direct-current machine on an alternating circuit, with a tendency always to rotate in the same direction. Assuming the machine to be self-starting, it will constantly gain in speed until the condition is fulfilled of one segment passing the brushes at each alternation, and it then becomes a synchronous alternating motor. The current then produces no reversals of magnetism, and there is a true alternating current in the armature circuit-producing, however, no reversal of armature polarity-and a current of constant direction in the field-magnet coils. Under these conditions the motor is self-regulating, moving at a constant speed and with a maximum rotary effort. It is not, however, essential that one bar should pass the brush at each alternation, as any number may be caused to do this, depending upon the speed required, and the number of coils upon the armature.

Thus groups of three coils may be treated as one coil; and supposing there were, say, twenty-four bars, then the machine would make one revolution for every eight alternations of current, and if connected to a circuit with 16,000 reversals per minute, its normal speed would be 2,000 revolutions per minute; and with forty-eight segments, in groups of three, it would be 1,000 per minute. There are blank segments insulating the groups of the inner collector, which are connected to the extremities of a rheostat which is placed inside the commutator, and is designed to offer a path for the alternating current, such as there may be, and prevent its absolute rupture at the period of change from one group of segments to the next; they also serve an important purpose in preventing a dangerous short circuit, which would be occasioned by the inner brush bridging two groups of segments oppositely connected. It follows as a matter of course, that, as the machine starts as a direct

current motor connected in an alternating circuit, rapid reversals of magnetism will at first be produced in all the iron cores; and these should be made of laminated iron, to prevent undue loss by heating at starting.

E. R. D.

The Distribution of Electricity by the Constant-Current System. By ALEX. BERNSTEIN.

(Electrotechnische Zeitschrift, 1889, p. 506. 7 Figs.)

Of the two systems of distribution, viz., that by means of constant current, with a varying electromotive force, and that by means of a constant electromotive force, with varying current, the Author gives the decided preference to the former. In this system, supposing glow-lamps to be used for lighting purposes, they are all arranged in series, the carbons being short and straight, and suitable for a current of 10 amperes, with a difference of potential at the terminals of each lamp of 6 volts, in the case of glow-lamps using the same number of watts, and working in parallel off leads, with a constant difference of potential between them, the usual type would be 100 volts, with a current of 0.6 ampere, but with such lamps the cross section of the main leads would need to be thirty-six times that required for the former kind, allowing for the same loss of power.

Another advantage is that, with the constant-current arrangement the difference of potential between any two points in the lamp circuit which might be touched at the same time by a person is very small, whereas with the parallel system the whole difference of potential may exist between two points close together. At first sight it would appear almost impossible to maintain a constant current in a circuit where the resistance must vary so frequently by lighting or extinguishing lamps, the electromotive force must be made to vary along with the varying resistance, and this may be done by varying the strength of the magnetic field, or by shifting the brushes. Neither of these appears to the Author to be at all satisfactory, and his method is to cause the speed of rotation of the armature to vary proportionally to the outside resistance. A series-wound dynamo is used, driven either direct or by belting, from a steam-engine, unprovided with the usual centrifugal governor. When the number of lamps in use is great, the speed of the engine and dynamo is high; but as the number is diminished so the speed falls. Supposing that at a given speed the dynamo is producing the proper current, with a certain number of lamps in circuit, this corresponds to a certain average steam-pressure acting on the piston; if now more lamps be lit, the outside resistance rises, the current falls, the dynamo turns more easily, and the speed of the engine rises until such a speed is reached that the balance is again restored; if a number of lamps are cut out, the speed in a