nails always protrude more or less into the inside of the joint, and materially impede the current of water. The mouths of the joints are also turned outwards, and form a shoulder, as at B. The intention of this, is probably to assist in securing the leather in its place, and to prevent the cording from slipping. The effects of it are as follows:-First, from the leather being strained over this projection, it becomes liable to be cut by every accidental injury, and very soon cracks and gives way, when a portion must be cut off and a fresh seizing made; second, the leather being stretched over the projection, does not fit the other part of the joint, and must be loose or filled up with pieces of leather, or, as is sometimes done, with brown paper; third, the irregularity of the calibre of the conduit which this shoulder occasions diminishes the performance of the engine.

Fig. 4, plate 4, is the coupling-joint adopted in Edinburgh. The furrows at the tying place are shallow, but their edges present a powerful obstacle to the slipping of the leather. No screw-nails are employed, nor is there any shoulder, as at B; there is therefore no impediment to, or variation in the velocity of the current, as the calibres of the coupling-joints and of the hose are so nearly uniform. It will be seen also, that, as the cording projects above the leather, this latter can never be injured by falls or rubbing on the ground. The slipping of the outer turns of the cord is easily prevented, by a few bits of copper-wire being first laid under

them, and then turned back and confined by the following turns of the cord.

Another great advantage attending the joints used here, is the manner in which their screws are finished. On examining the figure minutely, it will be observed, that the male-screw ends in a cylinder of the diameter of the bottom of its thread, consequently of the diameter of the top of the thread of the female-screw. The effect of this is, that, when the screws are brought together, the cylindric portion serves as a guide to the threads, and the most inexperienced person cannot fail to make them catch fair, at the first trial. The advantage of this in the circumstances attending fires is obvious. The same precaution is observed in the screws of the fire-cocks, and generally in all parts of the apparatus.*

These joints, although requiring three or four turns to close them up, yet as it is only the ring D which requires to be turned, it can easily be done with the hand alone without the use of keys. Although, when the whole length of hose has been jointed, it may be as well to send a man with a pair of wrenches to set the joints firm; this, however, is by no means absolutely necessary; if the joints are kept in proper order, a man can secure them sufficiently with the hand.

There is also a facility in taking turns out of the hose, which no other but a swivel-joint affords.

* This form of joint was introduced by Mr Robison.

By slackening a single turn any twist may be taken out, without undoing the joint or stopping the engine, while, from the number of turns required to close the joints, there is no chance of the screw being by any accident undone. In order to prevent the threads from being easily damaged, they should be of a pretty large size, not more than five or six to the inch. For the same reason also the thread should be a little rounded.

As it sometimes happens that the screws are damaged by falling on the street, or by heavy bodies striking them, whenever the hose have been used, the joints should be tried by a steel gauge-screw, to be kept for that purpose. This ought to be particularly attended to, as, on arriving at a fire, it is rather an awkward time to discover that a joint has been damaged, while the delay thus occasioned may be attended with very serious consequences.

The joints in plate 4, fig. 4, are 23 of an inch in diameter, which may appear large for an engine of sixinch barrels. They are made of this size, in order that, when the water-mains are large enough to admit them, the fire-cocks may be 2 inches in diameter. The only disadvantage of the large size of hose is a trifling addition to the weight and expense, as the engine always works more easily with large hose.

Two Rolls of Sheepskin.-These are simply four or five stripes of sheepskin, each about three or four inches broad. When a leak occurs in a length of hose, which cannot be easily replaced at

the time, one or more pieces of sheepskin are wrapt tightly over the leak, and tied firmly with a piece of cord. This is but an indifferent method of mending, but I do not know of any other which can be so readily applied with the same effect. If another length of hose can be substituted for the leaky one, it is better to do so; but that is not always at hand, nor does it always happen that time can be spared for the purpose.

Two Balls of Cord.-These are used for tying the sheepskin and various other purposes.

Two Lengths of Suction-pipe.-These are generally made of leather sewed tightly over a spiral worm of hoop-iron, about three quarters of an inch broad, a piece of tarred canvass being placed between the worm and the leather. They are usually made from six to eight feet long, with a copper rose screwed on the farther end, to prevent as much as possible any mud or dirt from getting into the engine with the water. It is of advantage to carry two lengths of suction-pipe, as, when one is too short, they can be joined to reach the water; if one is damaged, the other will still be serviceable.

I would recommend, however, that the suctionpipes be made with the leather rivetted, instead of being sewed. When a sewed suction-pipe becomes dry, the engine does not work well until the expanded threads contract, by being again wetted. The suction-pipes are more troublesome to rivet than the common hose, and are done here in the following manner :-After the joints are fixed on

the spiral worm, and it is covered with the tarred canvass, an iron mandrel longer than the worm is put through it, the edge being rounded to the circle of the inside of the worm. The projecting ends of the mandrel are supported to allow the worm to lie quite clear. One end of the mandrel has a check, that the brass joint may not prevent the worm from lying flat on the mandrel. The leather is then put over the worm, and the rivets being put into one side, a small thin mandrel is laid over the canvass, and the rivets struck down upon it. If the small mandrel be not used, the heads of the rivets are apt to lie unequally on the worm.

Four Wrenches for Coupling-joints.-These are for tightening the coupling-joints, when that cannot be sufficiently done by hand. When the hose are all put together, a man is sent along the whole line, with a pair of wrenches, to tighten such of the coupling-joints as require it. The wrenches are generally made as in plate 4, figs. 7 and 8; the hole at A is made to fit the nob on the coupling-joint, and when used, are placed, one on the nob of the male, and another on the nob of the female-screw, so as to pull them in opposite directions.

Two Directors or Jet-pipes.-These are taper copper tubes, having brass nozles. They are screwed to the end of the hose, and are generally made as in plate 4, fig. 2. The female-screw at A is made to fit the coupling-joints of the hose, and in order to afford greater facility in clearing it out,