Thus facility of getting at all the parts of an engine for the purposes of repair, or even of examination, is one of the advantages which the broad possesses over the narrow gauge. § In many articles exposed to great or sudden force, and to much wear and tear, it is very desirable that if any breakage occur, it should happen at that point where the consequences would be the least dangerous to the persons using it, and the reparation of it least expensive. During a series of experiments made by the author in 1839, on the Great Western Railway, it was necessary, amongst a variety of other curves, to cause a pen to draw upon long rolls of paper the curve described by the centre of a carriage, projected on the plane of the road. When everything is in proper order, this line ought to be parallel to, and in the middle between, the two rails. But it is well known that instead of answering these conditions, it often describes a _serpentine_ curve, arising from that snake-like motion of a train which the carriages acquire by rolling alternately towards each rail, until they are checked by the flanges pressing against it. To accomplish the drawing of the line above-mentioned, it was necessary to have depending from the carriage, a very stout jointed wooden arm, terminating in an iron _shoe_ with a steel projection. This _shoe_ was, by a powerful spring, pressed close to the rail in the middle point between the two side wheels of the carriage, and by a communication with the pen the required curve was described. But such an apparatus was exposed to very rough work, and, in fact, was generally broken three or four times during each experimental journey. If the broken part had fallen between the wheel and the rail, it might have caused a serious accident. To prevent this the following precautions were taken— The wooden arm was strengthened with thin strips of iron, except at one part about an inch long. At this part of the wood a small notch was cut with a saw. The lower portion had a strong iron eye fixed into it, which was connected loosely to a hook by a rope passing through a hole in the middle of the carriage. Whenever the apparatus broke, it was always at the notch. The position of the loose rope holding the broken part was such, that the tendency was immediately to drag it into the middle of the road under the centre of the carriage. This at once removed it from interference with the wheels. The pen describing the curve soon gave notice by ceasing to move laterally, that the arm was broken; on which one of the assistants immediately took hold of the loose rope, and pulling the broken fragment close up to the bottom of the carriage, prevented the possibility of any further danger.