articles of small value. Equally salutary results might however be obtain ed by modifications of the charges on the other principal division of the commerce of the country-that which is rendered by its value an object for the competition of rival works. But the examination of the subject with reference to the latter division would, from its complication, be much less susceptible of receiving a popular form. This and other views of the subject, which have always to be considered in any attempt to establish a correct tariff, are examined in considerable detail in my "Essay on the Laws of Trade," where the methods to be adopted to obtain the greatest revenue which the work can possibly extract from the commerce of the country is fully exposed. It is not possible to repeat here, in the narrow space which we can now appropriate to the subject, even the most important of the principles to be regarded in the administration of our public works, which I have there attempted to develope. The glance which is here directed to the question is necessarily confined to a very few prominent points.

To establish a tariff of toll for all articles, on sound principles, requires a certain intimacy with the statistics of the line, and a proper acquaintance with the laws by which the tonnage and revenue are governed. This knowledge cannot be obtained intuitively; and a correct tariff cannot be devised, as those on all our improvements are, by the mere conjectures of the parties to whose discretion such subjects are usually referred.

SECTION Y.-General Laws of Trade.

18. There are some facts of a general character relating to this subject which are susceptible of a most rigorous demonstration, that may be here profitably repeated. It has already been shown that a great loss of trade and revenue is sustained in the management of public works by the adoption of a uniform rate of assessment. In the examples adduced, this results from over charges, which, under such regulations, invariably have place in some part of the line. It may be shown by a legitimate course of argument, that however we depart from the charge which will yield the greatest revenue there will be an increase or diminution of tonnage, and, of course, always a decrease of revenue. If the departure be an overcharge, the tonnage will be reduced a quantity directly proportional to the value of the overcharge, and the revenue proportional to the square of that departure.

Nothing can prove more conclusively the danger of submitting so important a subject as the preparation of a tariff, to the uncertain guide of conjecture. For, if we err ten mills in the charge which we establish for any article, the loss will be one hundred times greater than if we err but one mill.

19. It is usual to assume that the charge for toll should be proportional to the distance the article is carried that it should be greater for a great distance than a short one. But we have already seen that, on the contrary in most cases, the greater the distance the article is carried the less should be the aggregate toll upon it. (Art. 13.)

20. It is common to suppose either that the tariff of toll is independent of the cost of freight, or that the higher the expenses of carriage the greater should be the charge for toll. But, on the contrary, the fact is susceptible of general and easy proof, that the higher the charge for freight on the line, the lower must be the toll; and also that any increase of the cost of freight will at the same time diminish the toll or profit on the article, and increase the whole tax for its transportation.

21. The charge for toll is also assumed, in the ordinary establishment of tariffs, to be independent of the mode by which the trade approaches the Jine; and, for many articles, this is true; but, for others, it is an ascertained

fact, susceptbile of easy demonstration, that if they are brought to the work by a common turnpike a higher toll should be charged for their passage on the improvement than if the same articles reach the work by a railroad, and, a fortiori, than if brought by a canal,

22. Where the object is to obtain the greatest possible revenue, it is a general law, susceptible of satisfactory proof, that the charge for toll should not exceed half that charge which would exclude the trade from the line. It may be shown, however, that, although a higher charge than this can never be advantageously adopted, it may frequently be reduced to three-eights of the sum which would cause the exclusion of the trade, with a very beneficial effect on the tonnage, and without leading to any sacrifice on the score of revenne.

23. Another fact, which may be derived immediately from the preceding is, that where the most judicious charge is levied, the tonnage of the line will be one-half of the tonnage which would be obtained if no toll at all were exacted.

24. The charge for toll has already been shown to depend in part on the price of freight; and it is an established law that if the cost of freight be increased or diminished by any modifications of the work, or the system of transportation adopted on it, the toll must be increased just half as much as the freight is diminished, or diminished just half as much as the freight is increased. And it is further susceptible of demonstration, that the increase of revenue which will follow a general reduction of the charge for freight will at least be equal to the arithmetical mean between the values of the tonnage before and after the reduction, multiplied by the amount that the freight is reduced. In other words, if the trade of the line be one hundred thousand tons, and the freight be from any cause reduced $1 per ton, and the tonnage thereby increased ten thousand tons, the revenue of the work will be increased more than $105,000; and this result will have place simultaneously with a reduction of the whole tax on the trade to the amount of $50,000, and the corresponding augmentation of the tonnage.

25. The fact is not usually recognised, that the toll, and tonnage, and revenue, are all more or less dependent on the length of the line of the improvement. It may, nevertheless, be easily proved by a general demonstration, that the tonnage, the charge which may be levied per ton per mile, and consequently the revenue will all receive an increase by a reduction of the length of the line of transportation. So that if the toll be judiciously established on any given line, and any material change of location afterward be made, by which the distance, or cost of freight is diminished, there must be a certain increase of the charge for toll, from which an augmentation of revenue will necessarily result. The value of this increase of revenue is equal to the whole annual tonnage of the line, multiplied by the actual cost of freight through the distance saved--and, considering only the value of the trade, it is therefore worth, to reduce the length of the line, the capital of which the interest is equal to that sum.

26. In the arrangement of the charges in a tarriff, there is no subject of greater importance, for some articles, than the positive, and for others, than the relative, value of the mart. From the positive value of the article, is determined the tax which one division of the trade can sustain; and from the relative value is in part deduced the proper toll on all commodities for which other works are competitors.

A permanent change of the relative standing of the mart, ought, therefore, to superinduce a modification of the toll. And it may be easily shown that if the relative value of the market-by which is intended its value compared with that of the rival mart-be increased any given sum, we shall find

the corresponding increase of toll per mile proper to be made, by divving that increase by twice the length of the improvement in miles. If, for instance, the value of Philadelphia as a mart for tobacco, compared with the value of New Orleans in reference to the same article, be, from any cause, increased $4 per hogshead, and the distance from Philadelphia to Pittsburg be four hundred miles, then, I say, the toll on tobacco on the whole central line of the Pennsylvania improvement, should be increased a half-cent per ton per mile.

It is also easy to demonstrate that, at the same time, the revenue will be augmented by any increase of the relative value of the market, an amount obtained by multiplying the original tonnage added to half the increase of tonnage consequent on an improvement of the market, by the increased value of the tonnage at the market.

It is shown, in fact, that whatever circumstance occurs to increase the value of the property sent along an improvement, to the holder of the property, will cause a certain increase of tonnage, of which the measure can be obtained, and a simultaneous augmentation of revenue equal to the whole increase of its value; and that whatever unnecessary tax is levied on the trade, is at least so much deducted from the revenue of the improvement. If, for instance, the engineer in making the location permit his line to be one mile longer than is essential, he will thereby cause to the State or company an annual tax, or equivalent loss of dividend, equal to the whole annual expense of transporting the whole trade of the country through that mile. He incurs, at the same time, the responsibility of reducing the tonnage of the work, and of injuring, to a certain extent, both the country that supplies the trade, and the emporium which receives it.

If he embarrass the line by an unnecessary grade, or any other impediment which involves a similar increase of the charge for freight, he is responsible for the same result; and, before adopting such a measure, is bound to compare the value of the difficulties to be avoided, with that of these inevitable consequences to the future trade.

28. It is an error, and a very frequent one, to suppose that the toll is in any manner dependent on the cost of the work; or ought, under any circumstances, to be directly proportional to the value of the article. It is, however, not uncommon to assume, that it ought to bear some relation to the cost of construction; and there are tariffs of toll in the data for the calculation of which the value of the commodity is in principal, if not the only element. It is not a little singular that, after encountering an expenseof hundreds of millions for our public works, the tariffs by which they areto be sustained should be entirely ruled by considerations which, however plausible in a superficial view, have no legitimate relation to the question.

For one division of the trade, it is not the positive value of the commodi-ty, but the difference between the market value and the cost of production; and for the other, the difference between the value of the article at the mart of the improvement and at that of its rival, by which the toll is influenced. And this influence is only partial. The value of the article, taken in any sense, is only a part of the data by which the true charge must be determined.

29. It is by no means the intention here to attempt a general analysis of this most important subject. My object, as already announced, has been to show that, under certain circumstances, a marked increase of trade and revenue may be obtained on all our public works, by simply reducing the charges; and that such modifications of the tax levied on the community is rendered imperative by the first principles of equity. If I have succeeded in making this truth perfectly apparent, the design of these pages will be fully accomplished.

I scarce hope to have satisfied the reader in a discussion so brief and popular, that to enable a company to take the full advantage of its position, and obtain the highest degree of success of which their enterprise is susceptible, demands a careful and close investigation of the laws of trade in reference to every branch of the subject. To appreciate the importance of this course, requires that the mind should have investigated such questions sufficiently to estimate the consequences of its neglect. To know the value of establishing the most correct tariff, we need to know what we are likely to lose by the adoption of a conjeetural or empirical one.

This subject, though usually taken under other auspices, is peculiarly a professional study. An intimate acquaintance with the principles which govern the trade of an improvement, is a department of knowledge which is essential to the engineer in the location of his line, and the establishment of many of the plans of his work. His first duties, when properly discharged, compel him to become the most intimately acquainted with the productions and statistics of the country, and the information acquired in the accomplishment of these labors, ought to be turned to account in directing the future administration of the line.

The preparation of the tariff is, or ought to be, the peculiar province of the engineer. For, though there may be many considerations of policy, which should have a certain weight with those whose final action is requisite to carry his recommendation into effect and which may frequently render it advisable to modify the charges which an a priori investigation may indicate to be proper, where the questions of trade and revenue only are under consideration,-still, it is not less important that the tariff should first be correct in itself, that some estimate may be made of the effect that will be felt when such political modifications come to be admitted. And, withal, it is difficult to conceive what motives of policy should induce any material departure from those limits which the administration of justice, the promotion of trade, and the augmentation of the revenue simultaneously recommend.

44

Uniformity in the width of track" finds as little favor on the other, as on this side of the water.

When the contracts along the line are fairly at work, one of the first knotty points which the directors will have to decide on, is, the width of the rails, their shape, the length of their bearing, and the form of their chair. The width between the rails has only lately become a subject of dispute, nearly all the railways prior to the Great Western, having been laid down 4 feet 8 inches apart. Mr. Brunel has extended it to seven feet, the Irish railway commission recommend six feet two inches, several of the Scottish railways are laid down at five feet six inches, in fact, the variations run from four and a half to seven feet.

The question of the stability of the carriages on the railway may be left out of consideration in looking at this matter, because the machinery will always require sufficient space between the wheels to insure this. Now, as 4 feet 8 inches are found to be enough for the good performance of an engine, which with 5 feet wheel, will go on a level upwards of 69 miles an hour, as with 5 feet wheels, Marshal Slout, on his visit to Liverpool just after the Queen's coronation, was taken over 103 miles of favorable ground on the Junction Railway within 10 minutes, & as an engine has gone 60 miles Grand an nour on the London and Birmingham Railway, up an inclined plane, is it wise or prudent to make any change at all, and will any additional

speed, which may be gained by increasing the width of the rails and the diameter of the wheels compensate for the greater expense of the outlay which will constantly be required to keep the road in order on account of the increased weight? This will receive light from the experiments on the Great Western, but will not be fully decided until it be tried on the Irish or some other railways, as Mr. Brunel's rails are altogether different froin most others in use. The plan recommended by the Irish railway commissioners, of putting the rails farther apart but not widening the carriages, merely making the wheels run outside the bodies, is a good one in some respects; but it would add to the expense of the works considerably, and the result would be exceedingly questionable.

It must not be forgotten, that, where a different width from that in common use is adopted, the railway on which it is used becomes isolated. None but its own carriages can travel on it, and they can travel on no other line. This alone will, in most cases, be a serious objection. For our own parts we should say, let well alone; wait for more experience; we are yet infants among railways, and we ought not to innovate on that which has proved to do well, until we become giants. The majority of opinions, however, are beginning to lean towards some increase in the width, although there is every diversity in the quantity which practical men think necessary. Cer tainly the machinery under the boiler is compressed into its minimum space, and more room for it would be a great advantage, if it does not induce an incommensurate loss in other ways.

With respect to the form of the rail, it can be proved that a fish belly has greater strength, weight for weight, than any other. A 60 lb. fish belly at three feet bearings, rolled with a lower web, would be the best form of all; and this has been effected, as the original Liverpool and Manchester rails had partially this shape. The question, however, must be looked at in conjunction with the length between the supports. We have given below. those forms most approved of in practice, and have added that in use on the Great Western Railway, which is however light, and does not stand well, three feet having been the original distance of the bearings.

Fig. 1 is the old Liverpool and Manchester rail, laid down at three feet bearings; weight thirty-three lbs. yer yard, with square joints. This rail.

was rolled with a lateral swell at the bottom, which on one side was con. tinued the whole length, but on the other did not quite reach the chair.One side of the chair was cast with a cavity, into which the lateral swell fitted, and the opposite side had a nearly similar opening, in which was driven an iron key, shaped like a wedge, which, entering in a longitudinal direction, not only forced the swell into the cavity which was formed to receive it, but by this means, at the same time, kept the rail down in the chair. Fig. 2. Losh's patent rail, in which he sought to gain a still more pow erful mode of keeping the rail down in the chair, by having his key tapered vertically as well as longitudinally, so as to act as a wedge downwards, as well as in the direction of its length; whilst, at the same time, the neces