only two. The circumstances of the case determine how many through leads are necessary, and also what the main source of electricity must be.

In every case, however (except in omnibus trains), we use the double system of main and auxiliary lamps with main and auxiliary supply of electricity.

TUESDAY, AUGUST 25.

The following Papers were read :—

1. An Electrical Parcel Exchange System. By A. R. BENNETT, M.I.E.E.

The congested state of the streets in many of the large towns, notably in the City of London, invites reflection as to whether it is not possible to devise means by which vehicular traffic may to a certain extent be diminished. To avoid absolute blocking of the thoroughfares, it is now necessary to forbid the collection or delivery of goods in certain localities during business hours, and trade certainly suffers under such restrictions, while warehouses have to be of larger capacity than would be needed were the free receipt and despatch of goods permissible. The author develops a scheme by which parcels and small packages may be freely interchanged between the various buildings of a town by means of miniature electric railways laid preferably, but not necessarily, underground, in pipes or culverts. Such pipes may be laid along the principal thoroughfares, communicating to the right and left by means of spurs or sidings with the premises of the subscribers to the system. In imitation of a telephone exchange, the pipes converge at one or more central stations, where operators having sole control of the traffic are on duty, and where are situated the dynamos and other apparatus. The scheme may be worked out in various ways, but the author proposes, by preference, a rectangular tube carrying two tracks or lines of rails, one above the other, the lower being used for the down, and the upper for the up, traffic. On the rails run trucks fitted with electro-motors, deriving propelling current from a parallel conductor laid between the two tracks, so that on the down journey trucks gather current by a collector pressing on the under, and on the up journey by one pressing against the upper, surface of the conductor. By dividing one of the rails of each track into insulated sections, the operators, by watching miniature semaphore signals placed in the central station, and worked by electro-magnets, are enabled to tell on which section a truck is, and to follow its progress out and home with the greatest exactitude. The sidings into the premises served are connected to the main line by switches resembling those of an ordinary railway, which are normally, by means of springs, kept in their position for through traffic, but which by the agency of electricity the operator at the Central Station can put over so as to connect with the sidings. The tracks enter the premises one above the other, but if there is room available, they then diverge and effect a junction, so that trucks can be shifted from the down to the up line without lifting them off the rails. On entering a siding a truck automatically signals the operator that it is clear of the main line, and on running into the premises it is brought up by means of catches and springs, which are depressed in one direction only, and which also serve to announce its arrival by ringing electric bells, and prevent its being returned by error or design into the tube on the wrong track. The connections of the motor are so arranged that a truck, even if placed on the wrong line, would not move backwards so as to cause a collision. The up track is blocked at the sending end, so that, although a truck may be placed on the up siding ready for despatch, it cannot obtain propelling current until the operator has a clear road for it. He then electrically removes the block, and gives the truck current by which it is brought into the Central. The operators have thus complete control over the movements of trucks. When a truck, or train of trucks, is intended by one subscriber for another, it is despatched in the first instance to the Central,

where, on reading the address, the operator forwards it to the siding of the consignee. Subscribers may telephone or otherwise communicate with the Central about despatch and receipt of trucks, but it is not necessary to do so, since trucks can be delivered into subscribers' sidings, and even unloaded automatically, and then withdrawn again without any attention on the subscriber's part, while trucks placed for despatch by subscribers can be brought into the Central at stated intervals if the operators make it a rule to tap all sidings occasionally for unannounced traffic. Goods could therefore be delivered during the night, and empty trucks sent into the various sidings for next day's traffic. If desired, collisions could be prevented between trucks in motion in the same direction by automatic blocks. In the event of a truck through any accident stopping in the tube, the semaphore connected with the section it is on will remain continuously at danger, and the operator will know that such a stoppage has taken place, together with its position, and take steps to remove it. For instance, he could send an empty truck forward, the speed of which can be reduced to a minimum as it approaches the disabled truck by modifying the propelling current until it strikes against the disabled vehicle, when, full current being turned on, both trucks can be forwarded to the consignee or shunted at the next convenient siding. Should trucks by mistake be delivered to the wrong siding, the subscriber would transfer them to the up track, and return them to the Central. The paper claims that such a system would prove of immense service if existing between the chief and branch post-offices of a city, between railway goods stations, parcel-receiving offices, and large business establishments, &c. The delivery of letters, telegrams, and parcels could be effected by the Post Office to subscribers without the aid of postmen, while matter for despatch by post and telegrams, together with the money to defray the charges thereon, could be forwarded by subscribers to the Post Office. For Post Office work the system would simply be a great development of the existing pneumatic tubes. Hotels and restaurants could telephone for and obtain in a few minutes viands they may be short of, and enable their customers to choose wine not only from the cellar of the establishment, but from those of every wine merchant on the system. It is not contended that such a system would pay if constructed specially for parcel work, although the surprising developments of the last decade scarcely permit of limits being assigned to the possible developments of the next; but the author assumes that the construction of subways beneath all the chief thoroughfares of large towns for the purpose of containing electric light and power leads, telephone wires, pipes for gas, fresh water, sea water, hydraulic power, compressed air, and other adjuncts of our complex civilisation, will shortly become an absolute necessity. A beginning in that direction has been made under the auspices of limited companies in some American cities, and we must sooner or later follow suit. Then when that time arrives an electrical parcel exchange could be carried out effectively and economically as part of the scheme. Our footpaths and carriage-ways will eventually be laid upon the lids of huge boxes, through which well-lighted pathways, affording crossings and short cuts for passengers at congested spots, may even he carried.

2. The Bénier Hot-Air Engine. By M. BÉNIER.

The question of hot-air or caloric engines has much interested the scientific and engineering world for many years. It has generally been admitted that the discovery of a really good hot-air engine would be of the greatest importance from economical and other considerations-amongst other advantages, boilers, with attending expense and danger, being entirely dispensed with.

Appended to this notice are illustrated drawings of the hot-air motor invented by Messrs. Bénier Frères. A considerable number of these engines are already in use in France and elsewhere on the Continent for industrial, electric-lighting, and other purposes. Several have been supplied to the French Government for use in lighthouses and fog-horn lightships. In the engine illustrated the air passes through the fire itself directly into the combustion-chamber. With this type of engine a much greater initial pressure can be obtained than in engines using a

separate combustion-chamber, or where the air is heated through an intervening metallic diaphragm. The drawing up of the grit and ashes is completely prevented in the present motor, this latter feature forming an important part of the invention. As will be seen by the drawings, the engine is constructed on the beam principle, and the combustion-chamber is really a prolongation of the working cylinder. The piston (or plunger) is of considerable length, the upper part only being made to fit the cylinder. The lower part of the piston is of slightly less diameter, consequently an annular space is formed between it and the cylinder. This space is connected with the main air-supply, which is controlled by a valve operated by

a connecting-rod and cam-lever worked from a cam on the crank-shaft of the engine. The air-pump is placed in the centre of the machine, immediately beneath the beam-standard, and is operated by a rod attached to the rocking-beam, and this is connected by a rod to the crank-shaft. Owing to the position of the beam, pump, and connecting-rods, the piston of the air-pump is at the outer end of its stroke when the working piston, on its return stroke, has reached a middle position. During the last half of the return stroke of the working piston the air-piston is pushed inwards, and compresses the charge of air previously drawn in until it has reached the middle of the stroke, at which moment the working piston is at the

end of its stroke. The air-valve, operated by the cam as already mentioned, has communicating passages with the air-pump, the furnace or combustion-chamber, and the annular air or packing space in the main cylinder. Consequently, the compressed air is forced partly through the fire and combustion-chamber, and partly into the annular air-space, the flow of air continuing during the time the air-piston performs the second half of the stroke. Meantime, the main piston receives its charge from the combustion-chamber, and cold compressed air passes into the annular space, and practically acts as a packing, effectually preventing grit and dust rising from the fire to the working faces of the cylinder. When the air-pump has finished its stroke, the air-valve is closed, and the air in the working cylinder is allowed to expand for the remainder of the stroke.

The cylinder is kept cool by means of a circulating-water jacket.

The bottom of the combustion-chamber is hinged, and the fuel is coke. As the combustion takes place under pressure, an air-valve, working automatically, is employed for feeding the fire.

The consumption of coke is about 3 lbs. (one kilogramme and a half) per brake horse power and per hour.

3. On the Internal and External Work of Evaporation.

By W. WORBY BEAUMONT, M. Inst.C.E.

Several of the most interesting problems in connection with the steam-engine turn upon the view that is taken of the mode of employment of the heat equivalent of the external work of evaporation.

=

When steam is generated under constant pressure external work is performed - PV, P being the pressure and V the volume generated. It may therefore, in accordance with the thermodynamic conceptions, be assumed that more heat is

used in the generation of steam under constant pressure than under constant

PV volume, the extra quantity of heat being U= J being Joule's equivalent, and J

U the heat units.

The author suggests the following explanation of the way in which the heat equivalent of the external work of evaporation is used. If heat flows out of steam when mechanical work is done by it during its formation, it must be supposed that steam is cooled by the outflow. If heat flows out and if cooling follows, the corresponding condensation or liquefaction takes place, and a further supply of heat is demanded to re-evaporate steam so liquefied; or, what is the same thing, the further supply of heat is used in continuously preventing the liquefaction from reaching more than the incipient stage. The action here sketched is readily conceived if for the purpose of explanation the evaporation and the external work be supposed to take place per saltum. Suppose a piston, immediately over the water in a simple evaporating vessel, to have been moved by the steam through a small distance A. Then heat corresponding to the work done in moving the piston through A will have flowed out of the steam, and this quantity of heat Q being gone, condensation must have taken place in order that the temperature T and pressure P of the remaining steam may be unaffected (L being latent heat of evaporation). Now before the piston can be again moved through a further similar distance A1, that quantity Q must be restored by a further demand on the source of heat, and if Q1 be the quantity of heat required to produce the volume of steam V, then the total quantity of heat Q2 required to move the piston through distance A' will be Q2 Q1 + Q, in order that volume V1 may be produced and the condensed steam Q be re-evaporated.

Now if A be taken as less than any assignable distance or the process assumed continuous, then evaporation and incipient liquefaction may be supposed to be contemporaneous, and Q and Q' will be supplied contemporaneously.

In this way it appears to the author that an explanation can be found of the mode of conversion of heat into the external work of evaporation under constant pressure, or of conversion of heat into the work performed by a steam-engine during the admission part of the stroke, or, more correctly speaking, the work done by the steam on the piston during admission. If this be a true statement of the actual mode of employment of the heat converted into the mechanical work of a steam-engine during the admission part of the stroke, then it follows that liquefaction takes place during admission, which must be sufficient to represent the mechanical work done. This being so, the question arises, To what extent will this liquefaction result in water or suspended moisture in the cylinder of a steamengine? The outflow of heat and corresponding liquefaction may be supposed to take place at the moving wall or piston, and in the hypothetic case supposed the liquefied steam may be assumed to be re-evaporated by the steam or water immediately below, which in its turn demands and receives more heat for its resuscitation from the source. In the case of the steam-engine cylinder, however, it is open to question whether the killed molecules in the cylinder or next the piston are resuscitated by the incoming steam, which follows up the movement of the piston. If they are not, then liquefaction will take place in the steam-engine cylinder during admission as a result of the performance of work, although the work is the external work of the evaporation which is performed in the boiler. The heat required for evaporation is that of Regnault's tables, but under the assumption here explained (when the liquefaction takes place in the cylinder and the resulting water does not return without loss of heat to the boiler), the heat required to raise the temperature of the quantity of feed water to the temperature of evaporation must be

U
L

added, because in order that one pound of steam may be supplied to the cylinder as steam at cut off, the extra quantity of feed water must be supplied to the

U
L

boiler. The quantity of water actually evaporated in the production of one pound