Within two or three weeks after completion, the following defects had appeared: 1. The inner edge of the rails was con siderably depressed, owing to the compression of the crushed stone under the rail. 2. The face of the rail was concave except where the gage-ties were attached, owing to the compression of the wood under the rail and to the spreading apart of the lower edges of the rail. 3. The rails were bowed horizontally outward, between the gage-ties, owing to the wedging action of the crushed stone between the rails. 4. Pieces of broken stone were continually loosened by the horses' feet and kicked upon the face of the rail, where they were ground to powder.
At this stage, sawed railroad ties were inserted 3 feet apart, and both flanges of the rails were spiked to the ties. After these repairs, the rails maintained their position reasonably well, but deflected between the ties; and bad ruts soon formed adjacent to the rails, and the crushed stone was ground up to such an extent that the rail was almost hidden by dust. It is conceded that this steel wheelway has not been a success. It has been taken up.
any successful wheelway must have a permanently hard surface between the rails; and consequently the tractive resistance of a wheclway should be compared, not with an earth road but with a road having the same surface as that between the rails of the wheelway.
Tests made by the author on the steel wheelway described in § 396 (see Table 8, page 29 ) show that the tractive resistance of such a road under the most favorable condition is more than that of a good brick or macadam surface; and that under ordinary conditions it is about the same as that of a new plank road, or an ordinary brick pavement with concrete foundation, or a specially dressed granite-block pavement. The force of traction on the steel wheelway is surprisingly large, and is doubtless due to the deflection of the rail as a whole and also of its surface immediately under the wheel, the wheel being continually compelled to climb a grade. In the light of these tests, it must be concluded that unless the rail is very rigid and is kept clean, the reduction of trad tive power will not be very great. If the wheelway is made rigid, the cost of construction will be considerably increased; and if the rail is to be kept clean, the space between the rails and adjacent to them must be covered with a practically indestructible surface which must be frequently swept.
In considering the relative tractive power required on different road surfaces, it should not be forgotten that the disadvantages of a grade increase as the tractive resistance decreases. For ex ample, if on a steel wheelway the resistance is 20 pounds a ton, a 1 per cent grade will double the resistance; while on a macadam road having a tractive resistance of 40 pounds a ton, a 1 per cent grade will increase the resistance only one half; and if the tractive force required on the macadam is 80 pounds a ton, a 1 per cent grade will increase the resistance only one fourth. In other words, in the above illustration a horse can draw only half as much on a level steel wheelway as up a 1 per cent grade, while on the first macadam road he can draw two thirds and on the second eight tenths as much up the 1 per cent grade as on the level. There fore the advantage of steel wheelways in decreasing the force of traction can be obtained only when the wheelway is level or nearly so.