When laying out a line of railway, the natural features of the country will necessitate the introduction of curves, and the question for consideration will be whether they are to be made of small or large radius. In some cases sharp curves are inevitable, except by incurring 50 enormous works which would not appear to offer any corresponding prospective recompense. In others the curves may be made of easy radius, at a comparative moderate extra outlay, if the character of the line and description of traffic to be accommodated will warrant the expenditure. For main through lines, with heavy, high-speed traffic, it is advisable to have the curves of large radius, so as to avoid the necessity of reducing speed when passing round them. Although a high-class fast train may be allowed to run round an 80 chain (5280 feet) curve at almost unrestricted speed, safety demands that there should be a reduction of speed on curves of 40 or 30 chains radius, and a very much greater reduction for curves of 20 chains radius and under. A sharp curve will in some places form a greater check to fast trains than a length of moderately steep gradient on a straight line. In the former the trains running in either direction must slow down for some distance before reaching the curve, round which they should pass at greatly reduced speed, and then some distance must be run before they can attain their full speed again. On the other hand, with a rising gradient, on a fairly straight line, the acquired momentum of the train will materially assist in ascending the incline, and although the speed may be slackened as the train advances, there may not be any very great diminution in the running before the gradient is passed, and average level line reached again. A reduced rate of running must be maintained round curves of small radius, for, however substantial the works and permanent way, and however well devised and constructed the rolling-stock, there is an element of danger ever present when passing round sharp curves at anything more than moderate speed. In the great rush for fast through trains this point is very apt to be overlooked, and too little time allowed for the running. Even with the fastest trains on any line there are some portions of the route which must be traversed with greater caution and less speed than others, either on account of sharp curves or of gradients; and if those who are entrusted with the preparations of the time tables do not possess the technical information necessary to deal properly with the question of relative speeds, there is the strong probability that the programme prepared may be one both difficult and dangerous to fulfil. The spirit of rivalry is a strong incentive to fast running, but prudence and common sense should indicate that record speeds should only be attempted on 51 the straight or favourable portions of the line. There is, unfortunately, the growing tendency to run faster and faster round the