(1) The. joint should be so constructed that the power required for its formation may be applied to the best advantage, and its application should be independent of special skill on the part of the workman.

Instead of filling the joint with lead, it has been suggested to use some alloy which expands in cooling. But this would still be open to the objection of losing its hold after compression due to the expansion of the spigot by heat, or to the longitudinal expansion of the soil pipe by the same agency, and, moreover, most of the alloys that expand on cooling are too expensive. Old type metal is perhaps one of the cheapest alloys which has been suggested for this purpose, but the antimony it contains renders it too hard and brittle. Belvidere bronze or Spence's metal, an alloy of iron and sulphur, has been tried in England. It is also brittle, and has the additional disadvantage of requiring the pipes to be heated to insure its running all around the hub.

Mr. Gerhard, in his "Sanitary Engineering of Buildings," gives formulae for a quick and for a slow setting iron cement to take the place of lead. The former is composed of 98 parts fine cast iron borings, 1 part flowers of sulphur, 1 part sal ammoniac.

to be mixed with boiling water before use. The latter is composed of 197 parts fine cast iron borings,

1 part of flowers of sulphur.

2 parts sal ammoniac.

The writer finds the rust joint rarely satisfactory, being brittle and soluble in water.

More recently, a patented compound of steel and iron, with other ingredients, called "Smooth on Joints," has been introduced, which is made up by the plumber into a paste without using heat and used in the pipe joints, where it soon solidifies and hardens. This compound, it is true, possesses the great incidental advantage that it does away with the necessity of plumbers' melting pots and furnaces in non-fireproof buildings, and therefore removes the danger of fire, but it is not waterproof, a very serious defect.

The writer has made a great many experiments with rust joints and a great variety of cement joints made with various compounds. They are usually unsatisfactory, especially when made by ordinary mechanics after published formulae, either on account of their great solubility in water, their gradual disintegration after standing for considerable lengths of time, their great brittleness, or their slow setting.

We find that the ordinary hand caulked hub and spigot joint possesses only one of all the desiderata necessary for an ideal joint. It is unreliable, incapable of resisting the effects of expansion and contraction, or heavy strains, requires unusual skill in its formation, affords every opportunity for fraud by covering up the traces which might lead to its detection, and it is expensive. It is compact, and has some facility in conforming to the irregularities of the structure for which it is intended, but even in this respect it leaves much to be desired.

If now we examine the form of this joint from a scientific standpoint with a view to determining the principles which give rise to its failure, we shall arrive at the following conclusion: Its inability to resist changes of temperature is due to the fact that the bell and spigot are so placed relatively to each other and to the interior of the pipe that they cannot be equally affected by changes of temperature, from within or from without; while the packing is so placed that when the pipe is heated, say from within, it is obliged at every point of its circumference to receive the full compression due to the excess of expansion of the entire diameter of the spigot over that of the bell. In other words, the smallest dimension of the packing ring is obliged to receive nearly the entire variation of size of the largest dimension of the spigot. The thickness of the ring is reduced by the expansion of nearly the entire diameter of the spigot, and owing to the absence of the property of elasticity in lead, the reduction in the thickness of packing ring is permanent while the alteration in the size of the spigot is only temporary. From this consideration we deduce the following laws with regard to the form of pipe joints:

(2) No bell and spigot joint having packing between the bell and spigot is capable of withstanding the effects of the lateral expansion and contraction of the pipes unless the packing has an elasticity sufficient to restore the alterations in its thickness produced by the difference of expansion and contraction of the entire diameter of the spigot over that of the bell.

(3) No doubt even a defective joint will in time take up under certain conditions enough sewage and rust to enable it to withstand a very limited amount of air and water pressure and appear sound. But such a joint is not a lead but a dirt or grease joint, which is a highly unscientific and undesirable formation. It is no recommendation to a pipe joint that it can only be made air and water tight after it has become filled with sewage, and it is well known that on all buildings where steam or very hot water are liable to be run through the piping, even dirt will not render the joints tight.

Fig 444a.

Fig 444a.

Fig. 44 5.

Fig. 44 5.

Fig. 445 represents a form of bell and spigot joint devised with a view to keep the lead from being drawn out of the socket by the longitudinal expansion and contraction of the pipe. The socket has internal annular grooves for the purpose, and a pour hole for the lead. A sleeve of lead is also pushed down to the bottom of the socket to keep the yarn or jute out of the pipes. These refinements fail to remove even the minor objections to the bell and spigot joint, and increase the cost and labor. The groove in the socket increases the cost of casting without in any way preventing the caulking 301frorn being worked loose by the expansion and contraction of the pipe. It undoubtedly, however, retards the lead from protruding from the bell under the influence of a longitudinal play of the pipes.