but its growing scarcity will certainly make the reinforced-concrete shells of such constructions more and more popular among bridge builders as the years go by. In using it, the time required to let the concrete set and harden is liable to cause delay in the sinking; and the removal of the forms is often troublesome.

This last statement leads to the thought that there is coming up soon the economic question of steel versus timber for concrete forms. At present the former material is not much employed for this purpose, but its adoption therefor is on the increase, especially when the forms have to be used a number of times. After timber has been utilized for this purpose three or four times it becomes badly broken up and unfit for further service, while the steel forms with care can be used an indefinitely great number of times. A combination of the two materials might be employed to advantage, the steel being interposed between the concrete and the timber girders, thus avoiding injury to the latter and permitting them to be used over and over again.

In respect to the depth below extreme low water to which it is economic to carry the shaft of a pier, there is great difference of opinion amongst engineers. The author generally locates the plane of division between shaft and base at an elevation of two feet below the lowest-recorded water-level, thus providing against exposure to the air of water-soaked timber, even in seasons of abnormal drought. Such treatment for a short time would probably do no harm, but the exposure of the crib to vision is not pleasing. Those who claim economy for locating this division plane far below the water's surface do so on the plea that it requires less material. This is true enough, but the unit price of the portion of the shaft below low-water is far higher than that above the same, and generally somewhat greater than that of the top of the crib itself. For this there are several reasons, viz.:

First. In order to build it, a fairly-water-tight, removable cofferdam has to be constructed, which is certainly more expensive than the simple crib-top.

Second. This cofferdam has to be kept pumped clear of water until after the shaft is built.

Third. The form-work below low-water is expensive and adds materially to the unit cost of the shaft-concrete.

Fourth. Where the shafts are carried down deep, more allowance has to be made for possible error of position; and to do this would involve the enlarging of the area of the base, thus increasing the total cost.

The condition sometimes exists which calls for the least possible obstruction of the waterway, and then it often becomes necessary to carry the shaft down to the bottom of the channel, irrespective of the extra cost of the piers. In such cases it will be found that the unit value of the shaft-concrete will be high, and that, as far as mere cost is concerned, it would have been better to carry the cribs up to near low-water mark.