at the back; and walls tapering longitudinally ought to have their breadth reduced in the same manner, in order to prevent refuse pieces of bricks being inserted in every course.

A great defect in the laying of bricks is the not mortaring the joints of the outside bricks to their full breadth, and the headers to their full length; an omission which allows the wet and frost to penetrate into the interior of the wall; producing those evil consequences so often observable in the interior of dwelling-houses.

The disposition of bricks in a wall ought to be so regulated that the extremities of no two laying contiguous to each other shall coincide, but, that the extremities of each brick shall stretch half way along the sides of those adjoining to it; and that all the bricks in the same course, or layer, shall lay in the same direction. The best method of securing this regular connection exists in what is termed English bond, which consists of one course of stretchers and one of headers alternately; the stretchers being the bricks which have their lengths disposed in the wall longitudinally, and the headers those having their lengths laid in the thickness of the wall. Every alternate outside header, on both sides of the wall, being only half the length of a brick, in order to connect the external heading bricks with those in the interior of the same course; thus rendering the half bricks available in producing bond. Adjoining the head of each quoin-brick, and diagonally through the courses to the interior angles of each returning or cross-wall, pieces, one quarter of the length of a brick, are necessary to prevent the occurrence of what are termed straight joints on the face of the wall, and also to preserve the continuity of bond at those intersections.

It is almost superfluous to say that bricks forming the frustrum of a wedge-usually termed radiating bricks are the best adapted for constructing arches ; common bricks, having their beds in parallel planes, cannot strictly be considered as constituting any part of arches. But, there being generally a great discrepancy in the thickness of common bricks, by carefully selecting them, and procuring bricks from different kilns, we obtain them of various thicknesses; hence by laying the thin bricks at the intrados, the thicker ones behind, and those of the greatest thickness at the extrados, we obtainwithout additional mortar-arches of great strength and durability. This is the method commonly resorted to, in practice, and which is found to answer the purpose even under embankments of great altitude.

But, when arches are of small diameter, the above method becomes inapplicable; and if radiating bricks are not used, the arches are either composed of distinct rings of brick-work, or extra rows of bricks are introduced into the interior and back part of the arches, and bonded together, at intervals, where any two rings of brickwork happen to coincide.

In arching with distinct rings of brick-work, it is indispensably necessary to use every precaution, such as avoiding extra thickness of mortar, irregular keying of arches, &c., in order to give to each distinct ring its due proportion of pressure, so that they shall act in unison with each other; and, comparatively speaking, the component parts thereof constitute only one ring.

When connecting two rings together, as adverted to in a former paragraph-especially should many courses of bricks intervene between the places where they are bonded-we would recommend, in

general, the bonding them together with two single rows of headers, laid immediately upon each other; the upper headers crossing those underneath half the breadth of a brick, longitudinally; we urge this, conceiving that the thickness of one brick is often inadequate to bond two rings securely; for, when the arch becomes loaded, a trifling deviation of form takes place, which has a great tendency to snap in two the bond bricks at those places.

The pressure against the retaining walls of bridges erected in deep excavations, is immense, often altering their position, fracturing, and ultimately rendering the structures very insecure. In such situations it will be prudent to dispense with the building of retaining walls, and erect the bridges with piers and arches, as shown in the subjoined elevation; which not only greatly re

duces the pressure against the before-mentioned walls, but renders the structures superior in appearance, and is accomplished also without incurring any additional expenditure.

In erecting such bridges, it is a good rule to observe that the height of the arches must not exceed that of the piers: where the latter are of great elevation, semicircular arches may be intro

duced to advantage; but where bridges are erected in shallow cuttings, the lower kind of arches are preferable; indeed, piers of any elevation may consistently be surmounted with the latter.

When a communication is required underneath an embankment of great altitude, it is desirable to erect the arch nearly to the extreme breadth of the embankment, to avoid a necessity for retaining or wing walls; these being very apt to be fractured by the lateral pressure of the mound when subsiding, and not unfrequently thrown over when a disruption takes place in the mound.

O

In circular tunnels, we have frequently observed the masonry of the upper part of the arches seriously injured by the superincumbent earth, especially underneath the centre of embankments; a circumstance obviously caused by the pressure on the arches being unequal. Serious defects arise also from inverts having too little curvature; indeed, so much so, that in some cases, inverts have risen up in the middle, which allowed the bottom part of the side walls of the tunnels to be forced in. We have been thus induced to believe

that by making those erections of an oval or eggshape, similar to that represented in the preceding transverse section, the unequal pressure on the quarters before alluded to, and on the sides of the tunnels, would be materially diminished if not entirely avoided; and perhaps cause the resultant of the different pressures to become nearly concentrical. We have therefore adopted this plan underneath embankments, some of which were upwards of sixty feet in altitude, and the results were in accordance with our anticipations.

T