A very important, and we believe an unexpected feature of the Fell system was immediately developed in the experiments; M. Desbriere calls special attention to it. The centre rail not only aids the train in going round curves, but also adds largely to its safety. Each vehicle is provided with four horizontal pulleys, each of about eight inches diameter, playing around vertical axes fixed upon the frame of the vehicle, and placed two and two at each of its extremities, and on each side of the centre rail. By the tightening of these pulleys, the flanges of the wheels, instead of gliding along the outer rails, press strongly against them, and thus the train is able to overcome curves unaccomplishable by any other means. This arrangement of parts also renders it impossible for either engine or vehicles to get off the line; all the more important, seeing that by the terms of the concession, the portion of the roadway ceded to Mr. Fell is, for the most part, on the outside, that is, the side nearest to the precipice, the bottom of which is, in many places, eleven or twelve hundred feet deeper than the roadway. Before proceeding to extract from the Reports of the French Commissioners to their government their opinions upon the Fell system, it is desirable to give some general deductions which Captain Tyler makes from what he witnessed, as bearing upon the important general question of railways over mountain passes. Hitherto the immense cost of the construction of such railways, and the immense cost of working them, have proved all but an effective barrier to their adoption. The cost of construction divides itself into two portions, and we will illustrate our meaning in the following manner:—Let us take A and B as the bases respectively on each side of a mountain. Were we to take these two points and to measure the interval between them, according to the every day application of the words “as the crow flies,” we must first take our crow perpendicularly up from A, to the exact height of the highest point which a person would be obliged to ascend, in order to cross the mountain, and then having drawn our imaginary tape to the point perpendicularly above B, we have the distance between them “as the crow flies.” But this distance in no way represents the actual distance that a person has to trudge up the mountain, and then to trudge down again in order that he may get, _matériellement et physiquement_, from one side of it to the other. The road which he has gone along can only be constructed with gradients that man and beast can accomplish. Exceed those gradients and the road might as well not be constructed. 1 in 12, or 440 feet in an English mile, is about as much as ought to, or, indeed, can be accomplished in this way. Yet, even to attain a mountain road with this gradient, it is necessary to make a great many turns and twists, to tunnel here, to raise an embankment there, to span a gorge sometimes several hundred