members being seriously weakened, and the encasement being split off by rust. It is not so important when the concrete slab merely rests on the tops of the stringers without encasing them; for the destructive effects are then limited almost entirely to the slab itself and to the tops of the stringer flanges.

The question as to whether reinforced-concrete viaducts should be water-proofed is somewhat more complicated than that for steel bridges, owing to the fact that there is not the same concurrence of opinion as to the injurious effect of water on the structure. It may be stated, though, that the water-proofing of concrete viaducts has lately become a common practice with the railroads of this country and with many of the private consulting engineers, as well as with a number of state, county, and municipal officials. Certainly, those engineers who include water-proofing in their specifications are actuated by the same desire to produce creditable work, the same loyalty to the interest of their clients, as those who do not. This being the case, on what theory and for what reasons do they consider it justifiable to make this addition to the cost of the structure? Just why should a reinforced-concrete viaduct be water-proofed?

That there is virtue in water-proofing a flat slab, even though well pitched for drainage, is shown by the experience of the country with concrete roads. It cannot be denied that such roads, well blanketed with asphalt, give longer and better service, with less cracking and other evidence of disintegration, than those not so blanketed. While this is partly due to the fact that mechanical wear is eliminated, the protection of the concrete from the action of frost and freezing water is an important factor. The drainage problem on a road is much more simple than that on a bridge floor. The standard width of road is generally but 18', so that the area to be drained is only 9' wide-the distance from the crown to the curb. The areas on a bridge floor are generally much larger; and since the water must be conducted to small down spouts, instead of to an open trench, as is the case with a road, the problem in bridgework exists in a much more aggravated form. If, therefore, it is found that water-proofing a concrete road by means of an impervious blanket of asphalt protects the road, prevents its disintegration, and prolongs its life, how much more urgent is the need for water-proofing a bridge floor?

A brief analysis of the more important of the disintegrating effects of water penetrations on concrete will serve to bring out the reasons why from an economic standpoint—i.e., from the standpoint of preservation and consequent reduction in both annual-replacement reserves and maintenance charges,-the water-proofing of concrete bridges is profitable.

Water is a universal solvent, affecting, of course, some materials more than others. There is in concrete some soluble matter; and, if water be permitted to pass completely through, such soluble matter is gradually removed by a leaching process. This is shown by the fact that water, after it has passed through concrete, will invariably give alkaline reaction,