Chap. XXV.

ON STONE BRIDGES.

1507

But when, at the end of the last century, it became the custom to erect large archesstruck from three or more centres, or those of the arc of a circle of one great radius,the practice of giving a thickness to the piers sufficient to resist the thrust of the archeswas abandoned: for had it been adhered to, it would have been necessary to constructenormous piers, which, with the inconvenience of giving to the bridge an exceedinglyheavy appearance, would have added the more important one of considerable expense,and much obstruction to the passage of the water. This latter is the principal ob-jection against the piers being thick; the resistance which they oppose to the current,and the contraction which they cause, depending in great measure on their width.Supposing all other things equal, a bridge in which the piers are very thick would havea greater inclination to fall than one in which they are small; and for this reason, that it isnecessary to augment the surface of the water-way, and consequently the length of thebridge: but the importance of this objection is diminished by observing, that yffienthe dimensions of a pier are considerable, it is difficult for it to be carried awayby a single flood, and it is generally possible to arrest the progress of the damage:a narrow pier does not present the same properties, unless it is carried on largefootings, and the advantage in the point of economy is almost nothing, as the savingin the masonry is of little consequence, either in the piers or in the foundation, on accountof the wide sets-off on which it is necessary to place them; but there is a considerableincrease in the expense of the centres, as all the arches must be centred at the same time,instead of the timber employed in the two first arches serving for all the others;this inconvenience is always important, and it will become more so, as the scarcity oftimber is more felt. In cases in which there is no danger in carrying the arches on verythin piers, as when established on a rock, and there can be no fear of either being carriedaway, the increase of the price of the centres may oblige a total rejection of the use ofpiers, and a proscription consequently of very flattened arches, which absolutely requirethem. Arches composed of the arcs of a circle flattened very much should always becarried on piers qf little thickness, and one of these arrangements necessarily inducesthe other. The inconvenience resulting from the use of piers is thus united to the otherreasons opposed to the general adoption of very flat arches for the construction of bridges.When we cannot give to all the piers the thickness they should have for the office of abut-ments, and when the great length of the bridge and the nature of the ground createsome alarm for the solidity of the edifice, an arrangement may be adopted which woulddiminish the dangers resulting from the use of thin piers.

Bridges are divided into several parts, by constructing at intervals piers capable of sup-porting the thrust of the arches; then, if an intermediate pier be carried away, it willcertainly induce the fall of some arches, but there will not be a necessity for reconstructingthe whole bridge ; if the bridge were established on one general platform, there would beno danger of accident; nevertheless as it may happen that some part of the platform maybe carried away, it is proper to place the resistance of the piers in equilibrio with thethrust.

When the piers serve as abutments, their thickness is determined after the prin-ciples already shown; when their object is only to carry the weight of the arches, theresistance of the stone with which they are constructed is the principal consideration towhich attention must be paid, and the only one which can be submitted to calculation ; itleads to dimensions much below those in use. M. Perronet has remarked that thepiers of the bridge .at Neuilly would support the mass with which they are charged, iftheir thickness were reduced to 13 inches : he admitted certainly that the stone would thencarry a weight under which it was crushed in a previous experiment; these piersare about 14 feet in thickness. According to this example, we must- not determine thethickness of a pier with relation only to the strength of the stone, but also to variouscircumstances which it is not possible to submit to calculation, such as the nature of theconstruction, which may be entirely in freestone, or some part in moellon or rubble(rough stone); the force of the shocks to which the pier is exposed either from the ice,from trees, or other bodies which the river may carry down, and above all in the mannerin which the pier has its foundations laid. There is no doubt that when the con-siderable weight which a pier has to support is spread over a great surface, moreconfidence should be felt in its foundation, whether in relation to the accidents, whichare less dangerous and more easy to repair, or to the unequal settlements which may bemanifested in it; and for that reason we cannot dispense with a large base for a pier thebody of which is not of great thickness.

The form of piers , and above all that of their starlings, is important, and whenthe piers are very large, have great influence on the solidity and duration of a bridge;when their form is imperfect, it is one of'the causes of the destruction which mani-fests itself, and always increases that which may arise from others. It appears from thefirst view, that if the water-way of a bridge is regulated in such a manner that the relationbetween the resistance of the soil and the mean velocity of the. current be not sensibly

5 d 2-