.

It commences at the terminus of the Canta Gallo Railway, crosses the Serra at an elevation of 3,000 feet above the Canta Gallo line, and ter minates at the town of Novo Friburgo, a distance of twenty miles. In some of its principal features this railway resembles the summit line of the Mont Cenis, the gradients for the passage of the Serra over a distance of ten miles being principally from one in twenty to one in twelve, and the curves by which the line winds round the spurs or counter forts of the mountain being, for a considerable portion of it, from 40 to 100 meters radius. The narrow gauge of 1.1 meters has also been adopted. In other features, however, there is an important difference between these two center-rail lines. The concession for the Mont Cenis was but temporary, terminating on the completion of the great tunnel, and the railway is laid on the existing public road, whereas the Canta Gallo line will be permanent, and the works will be so constructed as to be specially adapted to its requirements. It will not have to contend with the diffi culties of an Alpine climate; and, profiting by the experience of two years' working on the Mont Cenis, it will have the advantage of important improvements which have been made in the engines, carriages, and permanent way during that period. Consequently, the Canta Gallo, and other similar lines now being or about to be commenced, have the interest of marking an important development of the capabilities and advantages of the center-rail system as applied to the construction and working of mountain railways.

In the new engines for the Canta Gallo line, it is proposed to dispense with the toothed wheels, and substitute for them a system of direct driv ing by connecting rods. The power of adhesion will also be considerably increased. These new engines will have the advantage of being able to run at a speed of from twenty to thirty miles an hour upon the ordinary gradients of the line, and of taking their loads up the mountain_section at a diminished speed of from eight to ten miles an hour. In an economic point of view, the result of the application of the center-rail system to the Canta Gallo Railway will be as follows: The cost of construction, assuming it to be as estimated, about £300,000, would be at least doubled if made on gradients upon which ordinary engines could work. In this case the cost of traction and maintenance for a center-rail line will not be greater than for a line with ordinary gradients passing over the same country. The clear saving, therefore, effected by employing the center-rail system is at least £300,000, and the construction of a valuable line of railway has been rendered possible, which would otherwise have been commercially and financially impracticable. Mr. Fell mentioned a somewhat similar line of railway under consideration by the Indian government, from the port of Karwar to Hoobie, in the southern Mahratta country, both/by way of the Arbyle and the Kyga Ghats. The distance is seventy miles, and it is proposed to employ the center rail for a length of about ten miles upon gradients of one in twenty for the passage of the Ghat, by which a saving would be effected of about £500,000. The cost at the present time of the transport of cotton and other produce over the ninety miles is stated to be £235,000 per annum, and there is in addition the disadvantage of not being able to con vey the whole crop to the port of shipment before the rainy season sets in a large portion of it has consequently to be housed and kept until that is over. Negotiations are going on with the government local authori ties and people interested for the construction of center-rail lines in Italy from the Adriatic to Macerata, and crossing the Apennines to Fo ligno from Florence to Faenza, and for three branch railways in the Neapolitan States; in France, from Chambery to St. Andre du Gaz

and Lyons direct, crossing the Col de l'Epine; in Switzerland, for the passage of the Simplon; and in Spain, for lines from Leon to Corunna and Gion. The concession for the Mont Cenis Railway expires on the opening of the tunnel line; and when that period arrives, it has been proposed to remove it to one of the neighboring mountain passes, where it would have a permanent life. At the time the concessions were granted, it was considered that the line would be worked for ten, or at least seven years. The progress of the great tunnel has, however, been so much accelerated, that it is stated the tunnel line may possibly be opened for traffic by the end of 1871. In that case, and taking into a count the difficulties of all kinds with which the enterprise has had to contend, the Mont Cenis Railway can only be regarded as an experimental line and the pioneer of a system destined to confer the benefits of cheap and safe communication between many countries separated by mountain ranges hitherto impassable by railways and locomotive engines. Of still wider importance and application is the so-called narrowgauge system. The center rail may be said to involve this, since the localities in which it is likely to be employed are generally such as exclude the broader gauges by reason of their cost in grading; but the narrow-gauge principle itself is, of course, quite independent of the other. At the meeting of the British Association already referred to, Mr. R. F. Fairlie, civil engineer, read a paper on the gauge for the "railways of the future." I quote the substance of it, as reported in the newspapers at the time:

The object of this paper was to advance a new argument in favor of the use of a narrow gauge in the construction of railways, founded upon a comparison of the amount of weight hauled, for the same amount of paying traffic, over a railway of 3-foot gauge and a railway of the English "narrow," or 4-foot 84-inch gauge. Although maintaining that the principle of his argument applied to passenger traffic, and that the cost of working a railway, or, in other words, the proportion of non-paying to paying weight, (as far as this is independent of management,) is increased exactly in proportion as the rails are farther apart, because a ton of materials disposed upon a narrow gauge is stronger as regards its carrying power than the same weight when spread over a wider basis, the author on the present occasion went into detail only with regard to the conveyance of goods; and he selected the London and Northwestern Raiway as his illustration of the effects of the 4-foot 84-inch gauge, on the ground that its management is so good that the defects in its working must be wholly traceable to its construction. He undertook to show that this line, if made of a 3-foot gauge, would accommodate the whole of its present goods traffic as well as at present, and would do so at half the present cost, with half the present tonnage and motive power, and with half the present wear and tear of rails, so that the expense now being incurred for the construction of a third track would be rendered unnecessary. Assuming that the present goods traffic, independently of minerals, amounts to ten millions of tons per annum, and that the non-paying weight of trucks by which these goods are hauled amounts to the low estimate of forty millions of tons more, (seventy millions being nearer the truth,) there results a total gross weight hauled by the locomotives of fifty millions of tons at an average speed of twenty-five miles an hour. The earnings for the goods traffic on this line are 68. 3d. per train-mile, which, at an average rate all round of 14d. per ton per mile, would give about 50 tons as the paying weight and 225 tons as the gross weight hauled per train-mile. Dividing these 225 tons into the fifty millions, gives 196,089 trains, which, being divided by 312 working-days of a year, gives 626 merchandise trains over all parts of the Northwestern Railway in the twentyfour hours. The company's balance-sheet shows that each net ton produces about 48. 3d., which, at 14d. per ton per mile, makes the average distance traversed by each ton to be about thirty-eight miles; so that as each ton of the total weight hauled runs thirty-eight miles, and the entire length of line worked is one thousand four hundred and thirty-two miles, it follows that there must be, on an average, thirty-seven merchandise trains distributed over the total length. Dividing by this number the total number of trains per day of twenty-four hours, gives an average of over seventeen trains per day running on each mile of the line. Having reached this conclusion, it becomes possible to see how it would affect the question if the gauge of the line were 3 feet instead of 4 feet 8 inches. In the first place, the same or a greater speed could be maintained, say up to thirty-five or forty miles an hour. On the 4-foot 84-inch gauge the proportion of non-paying to paying load has been taken at four to one,

although it has proved largely in excess of this. The wagons employed average four tons in weight, so that on this reckoning each wagon carries one ton for every mile it runs. The wagons for a line of 3-foot gauge weigh each one ton, and carry a mari mum load of three tons. Supposing that the same number of wagons and trains were run on the narrow gauge as on the broad, it follows that the average one ton of mer chandise now carried would easily be taken in a wagon weighing one ton instead of four tons, and that the gross load passing over the line for one year would be only twenty millions of tons instead of fifty millions; while the same amount of paying weight would be carried in either case; that is, the small wagons which are capable of carrying three times the weight of goods now actually carried in a four-ton wagon would only have to carry one-third of that quantity, and would produce the same paying load as the heavier wagons, and, as the haulage cost is precisely the same whether the tons hauled consist of paying or non-paying load, it follows that this expense would be reduced to two-fifths of what it now is. If the same number of trains were to run per day, the weight of each would be reduced from 225 tons to 102 tons; or, if the same gross weight of train was employed, the number of trains per day would be reduced from 626 to 250. If there should be sufficient traffic to load the narrow-gauge wagons in such a way as to require the same number and weight of trains that are now worked, the result would be that without increasing by one penny the cost of haulage and of the permanent way of expenses, the 3-foot gauge would carry a paying load of twentyfive millions of tons as against the ten millions now carried. Here, then, we have established the fact that, as far as capacity goes, the narrow gauge is superior to the broad one. The former can produce twenty-five millions net out of a gross tonnage of fifty millions; while the latter, to produce the same result, if continued to be worked as it now is, would require that one hundred and twenty-five million tons should be hauled, and that at an increased cost in the same proportion of one hundred and twenty-five millions to fifty millions. The rest of the paper was devoted to an appli cation of these figures to the question of the best gauge for Indian and colonial railways, and to the argument that such railways might be made cheaply and efficiently on a 3-foot gauge, so as to charge a reasonable tariff and to afford a satisfactory return.

I think Mr. Fairlie has pushed the argument in behalf of the narrow gauge further than an impartial judgment will follow. It is scarcely fair to take actual working results on one hand and sanguine expectations on the other as the basis of comparison, and to ignore all considerations other than those of paying and non-paying weight. But there is no doubt, whatever may be the gauge of the "railways of the future," that the narrow gauges will play an important part, and that their economi cal advantages will be more closely studied than heretofore. Especially in the United States, where the peculiarly American system of pushing railroads in advance of settlement and traffic has been so vigorously and successfully followed, this subject possesses a special interest.

There are few questions of more practical and pressing importance at the present day than the best means to be adopted for extending our railroad system (carrying with it as it does fresh life and energy into all the districts which it penetrates) into the Territories and other parts of this vast Union, where the traffic to be expected, at any rate, for years to come, cannot be such as to warrant a large capital expenditure.

The following information was kindly furnished me by Sir Charles Fox & Sons, the well-known civil engineers of London, who have for many years been actively engaged in practically working out a similar problem in Australia, India, and Canada.

Their object has been to construct railroads which, while very economical in first cost, should be substantially built and equipped, and therefore operated and maintained at a moderate percentage of the gross income.

Mr. Carl Pihl, the engineer of the government railway of Norway, has also for some years been carrying out very successfully a system almost identical with that under review.

The question of gauge is one which requires to be determined after careful investigation of the circumstances in each case. Where a standardnational gauge exists, caution must of course be exercised in introducing any diversity; and yet it may, upon examination, be found that to adhere in

all cases to the gauge suitable for trunk-roads would be to check, nay, even to prevent, the construction of many tributary or branch roads, in themselves most necessary for the development of the country. Thus, in the vast empire of India, where the trunk-roads have the 5-foot 6inch gauge, the government is seriously and favorably entertaining the question of adopting a much narrower gauge for the tributary roads, to open up the country.

With the above reservation, Sir Charles Fox & Sons, and others interested in this question, have found, after an experience spread over many years, that the most economical gauge for such tributary roads which can be used with advantage is that of 3 feet 6 inches. A broader gauge than this is, in their opinion, for speeds of twenty-five miles an hour and moderate traffic, quite unnecessary, and of course involves additional outlay, especially if the country is of a hilly nature. To use, on the other hand, a narrower gauge than 3 feet 6 inches is likewise objectionable, and especially so where wood is the fuel chiefly employed, as on very narrow gauges it becomes impossible to use either boilers or fire-boxes of such dimensions as to give satisfactory results.

The following are given as examples of railroads which have been built, or are now in progress of construction, upon this gauge:

Examples and cost in gold, including in each case stations, rolling-stock, engineering, and all management expenses.

1st. The Queensland Railways, Australia. Length, two hundred and twenty-two miles; gauge, 3 feet 6 inches. Wages: Skilled laborer, $2 50 to $3 10; ordinary, $1 50 to $1 75. Average cost per mile, $32,000. 2d. The railway from Conyeveram to Arconnur, India. Length, nineteen miles; gauge, 3 feet 6 inches. (Land and portion of road-bed given by government.) Materials chiefly sent out from England. Rails, 35 pounds; iron. Average cost per mile, $19,000.

3d. The Toronto, Grey and Bruce, and the Toronto and Nipissing Railways, Canada. Length of first section, one hundred and ninetythree miles; gauge, 3 feet 6 inches. Wages: Ordinary laborer, $1 to $1 50. Average cost per mile, $14,150.

4th. The government railways, Norway, (constructed by Carl Pihl, civil engineer.) Length, one hundred and six miles; gauge, 3 feet 6 inches. (Rails and many other materials sent out from England.) First, through easy country, $15,900 per mile; second, through heavy country, $23,700 to $26,150 per mile.

It will be seen that, the two first of these principles being conceded, it at once becomes possible to construct a thoroughly substantial track with rails not weighing more than from 30 to 40 pounds per linear yard, provided that the ties are laid sufficiently close, the rails well fished at the joints, and an ample supply of ballast provided.

The speed of twenty-five miles an hour is found in practice to be more than sufficient for tributary roads. A load of four tons per wheel is sufficient to enable the passenger and freight cars to be of ample dimensions for convenience of traffic.

The passenger-cars of latest design are of the usual American type, 32 feet long exclusive of platforms, and 8 feet 6 inches wide, carrying very comfortably thirty-two passengers. Their center of gravity being very low, they run with great steadiness. The box-cars are 15 feet long and 8 feet 6 inches wide. The platform-cars are 24 feet long and 8 feet 6 inches wide, and carry ten tons, their own weight being only five tons. It will thus be seen that the non-paying load, or dead weight, is re

duced from the usual proportion of one-half to one-third of the gross weight; and from this results an important saving in operating expenses. The locomotives are generally of two types.

Type 1, or freight-engine, has 11-inch to 12-inch cylinders, six coupled wheels, 3 feet 3 inches in diameter, and a four-wheeled bogie, and weighs about twenty tons, exclusive of tender.

Type 2, or passenger-engine, has 10-inch to 11-inch cylinders, four coupled wheels, 3 feet 3 inches in diameter, and a four-wheeled bogie, and weighs about sixteen tons, exclusive of tender.

The maximum grades vary from 100 to 132 feet to the mile. The sharp. est curves are from 500 to 330 feet radius. The grading has generally a width in cuts of 15 feet, and on banks of 12 feet. The bridges are sometimes of iron, with masonry substructure; sometimes altogether of tim ber. The depots are of neat but economical design. The railroads are generally fenced throughout.

On the Canadian roads careful arrangements are made to protect the track from snow, and to provide in every detail against the effects of frost.

The cost of the Canadian roads, viz, about $14,000 per mile, may be taken to fairly represent the probable outlay (including every expense of right of way, management, &c.) for the building and equipping of a road of this gauge through an undulating and well-settled country, involving considerable expenses for right of way, say, 15,000 cubic yards of grading per mile, frequent road-crossings, fencing throughout, and a fair proportion of bridges, depots, side-tracks, and an ample equipment for a considerable traffic.

On the other hand, in the great prairie country of the West, where right of way, grading, and bridging are at the minimum, where 36-pound rails would be ample, where fencing is not necessary, and where the provision for depots, side-tracks, and equipments could be materially reduced, a well-constructed and substantial railroad of the 3-foot 6-inch gauge may be completed and equipped ready for operating, with an expenditure not exceeding from $8,000 to $10,000 per mile, and on such a road, as both grades and curves would be easy, trains carrying a net load of 100 tons of freight could be operated with facility at a speed of from twenty to twenty-five miles per hour.

OPERATING.

So far as the gauge is concerned, the cost of traction would be very much the same per train-mile, and per ton as on railroads of the ordi nary 4-foot 83-inch gauge, were it not reduced by the much more favora ble proportion which the net or paying load bears to the gross weight of the train. Moreover, in consequence of the lower speeds and the lightrolling loads there is on these light railways a most important saving in wear and tear, both of tracks and equipment, and the total cost of operating is therefore considerably reduced.

I earnestly commend this subject to those in the West who are interested in the opening out and settlement of districts not yet supplied with railroads, and refer them for more detailed information to Sir Charles Fox & Sons, 6 Delahay street, London.

Another variety of so-called narrow-gauge railways should rather be ranked with the suspension tramways, of which the wire tramways at the Brown, Griffith, and Stevens mines in Clear Creek County, Colorado, described in my last report, (page 372,) are examples. At the meeting of the British Association, to which reference has already been made, draw