Rivers should therefore have a fair chance to justify their care, without favor and on economic grounds alone; and those that do not survive this test should be abandoned in favor of the more efficient carriers.

It is believed by many who have watched the unequal contest between rail and river that when the day shall arrive (not now far distant, it is hoped) on which unjust competition shall be ended, the railroads will find in their search for increased efficiency that the rivers are their greatest friends, whose cheap carrying power they cannot overlook. Already some steps in this direction have been taken. The increase of traffic through the St. Mary's River of late years has largely resulted from the development of lake transportation under the supervision of the railroads. Should this same interest some day extend to our rivers, we shall again see our streams busy with freight.

The local and often disconnected character of river work in the past, even when it has underlying it an inseparable and intimate relationship, has already been mentioned as being a disadvantage. My second desirable governmental policy is greater coördination of work.

In the early days of government aid to our rivers many projects of merit were undertaken, and were carried along by appropriations made every few years. As conditions changed from time to time the sums appropriated for each varied, the amounts often being dependent on the state of the country's finances and often on the local pressure with which the work was urged. It thus happened that completion was seldom provided for in the beginning, but only a continuation of improvement; and years were sometimes consumed in doing a work of several seasons. If large enough sums were provided, contracts were usually let; but in every case when the funds were exhausted the progress of the work was arrested until another act could supply more money. These periodical delays were often enough to more than neutralize all the efforts of the engineers toward economy, and frequently disheartened those citizens who were interested in the project as a business enterprise. There seems no doubt among those accustomed to good business methods that any engineering construction when once determined upon should be finished or guaranteed all the funds necessary to push it to completion. Otherwise there is inevitable loss in the idle invested capital and in deterioration of the structures. This disconnected method of

carrying on our river work was followed for many years, and is still the rule for many projects. But in the course of time considerable dissatisfaction with the delay in the execution of some adopted projects arose among the people of the localities to be benefited, and many engineers publicly discussed the wasteful methods in vogue. The delays could not be ascribed to the engi neers or to the high cost of the work, but rather to the disad vantages arising from the methods of supplying funds.

At length, however, improved methods were determined on. under which many of the causes of dissatisfaction could be ob viated. The first step toward a new dispensation was the adop tion of what is known as the "counting-contract system." B this system a work can be carried forward through several years under a single contract, the appropriations being made at prope times, each in an annual bill; the total being kept within a authorized sum. In this way, instead of waiting for the usus. river appropriations, contracts can be let for the full amount of the authorization, although the act making provision therefor may appropriate only a small part of the total in immediately available funds. This was the first step, and an important one, toward the better coördination of work.

A new plan under which annual instead of biennial appropria tions will be made is now proposed. This will be another grea: assistance, if consistently followed, and will prevent many delays

But these improvements did not change the practice of adopting local projects based on local needs. Each proposed plan was examined and reported on, and if finally adopted was executed more or less independently of its relationship to other parts of the system. Nearly all the rivers of the interior part of this country are tributary to one trunk stream, and as such are to some extent interrelated. To enable boats to be easily trans ferred from one river to another, the depths of channels and width and length of locks should in some measure correspond, in order that streams belonging to the same system and possessing similar physical characteristics may afford easy interchange of

boats.

Until recently there has never been an official body charged with the duty of testing the relative necessities of connecting streams in advance of the adoption of projects for federal improvement. Furthermore, our navigable streams vary widely in importance, and there has been no authorized way of arranging

them in proper order so that the most useful should receive the earliest attention. Sometimes tributaries were selected for improvement before the trunk stream was ready, and less worthy streams were occasionally provided for in advance of more useful ones. The manifest defects of such a method became more and more apparent as the volume of work grew, and finally of necessity led to another important step in the direction of better coordination. This was the provision in 1902 for a Board of Engineers for Rivers and Harbors, which was to sit as a board of review and report on the projects submitted to it from the local districts in obedience to legislative directions. This at last gave some measure of coördination, and was one of the most important steps ever taken by the government in the history of this class of work. Now a single body can view the entire system and correlate its parts, whenever a river project is examined under governmental direction. The board can test the worthiness and relative importance of each project and determine the necessities of the stream, all in relation to the needs of the country as a whole. It may withhold approval if not found to meet its tests, or modify previous recommendations in order to bring the proposed project to its own proper rank.

But an important provision is still lacking. This board can not propose new works or arrange a plan of execution beyond what is submitted to it. It resembles a court which can decide a case only when duly presented, but cannot initiate any action of its own. It would appear to be in the direction of progress to endow this board with more ample powers; and it does not seem a vain prophecy to predict that this will eventually be done, if only by degrees.

There has been a demand among those interested in this class of work for a comprehensive plan of construction applicable over the whole interior of the country. Here is the means, already at hand. There have been criticisms that no new plans can be presented for government adoption which will be so designed as to keep all the system in harmony. Here is the organization, ready for instructions.

The second policy, therefore, that I would suggest is greater coordination of projects, to be secured by enlarging from time to time the functions and responsibility of the present duly constituted river and harbor board, to enable it to prepare a comprehensive plan of improvement of all our navigable rivers, having

due regard to their present needs and prospective usefulness; to suggest the more worthy streams for earliest work, to arrange the distribution of available funds within such limits that those works once begun may be completed within a reasonable time, and to see to it that the less important shall wait until the more useful ones are earning suitable dividends for the public by accommodating an adequate volume of traffic.

This body should also be expected to propose new work and new lines of investigation and experiment, and should be endowed with all the vigor and initiative necessary thereto. It should be required to present to Congress periodically a list of new works to be undertaken, together with a studied scheme of construction that would give the best and quickest returns for the money er pended. This method of solving a difficult problem will probably not be satisfactory to those who desire a new organization, such as the establishment of some bureau of public works, or a complete upheavel of present methods; but it presents the advantage of being ready and trustworthy, requiring no new adjustment of duties, but only an amplification of authority; not revolutionary, but in the direction of present development; economical, and on the whole worthy of a trial whenever the present plans are to be improved.

The third policy which is suggested herein is coöperation. By coöperation is not meant the loose partnerships between the government and private individuals or corporations which so often result in one-sided benefits, but it intends an assistance on the part of the communities and localities to be benefited, in order to render more useful the channels constructed by the government. Coöperation may be tendered in many ways and many works are now being carried on with the financial assistance of the state, city, or home community. But reference is made here to only one class of coöperation; one that is usually the most necessary, that is, that involving the construction of terminals, landings, and warehouses.

It has often been asserted that an important reason for some river work was the temporary prosperity induced by the large expenditure of money in the vicinity. Some have also said that the reduction in freight rates on the competing railroads was the main object to be accomplished by these expensive structures; but all must finally admit that unless the improved waterway comes to be used enough for river traffic to pay back to the public

a benefit equivalent to a reasonable dividend on the investment, the work will be an economic failure. Channel development must rely mainly on its usefulness to the public as a means to easier or cheaper transportation, or it cannot long receive popular support.

The government in its efforts for easier communication is concerned primarily with channels, but landings are equally necessary if the channels are to be used. Rivers must have transfer facilities for exchanging freight with rail lines or other rivers, storage facilities and warehouses must be provided, and some efficient means of loading and unloading vessels must be furnished to complete the work. Local communities can coöperate in this, and can demonstrate at the same time their sincerity in their demands for governmental aid. They can coöperate with the government by constructing or providing for terminals. A channel without terminals is like a railroad without sidings or freight yards. An engineering periodical recently stated that the terminal charges against railroad freight between Philadelphia and New York were fourteen times the cost of transportation alone, and that the terminal charges at Chicago and New York about equaled the cost of transportation between these cities. The enormous importance of this feature as an element of rail freight cost is generally recognized, but its bearing on river improvement is only just beginning to be understood. The construction of terminal facilities has not yet been undertaken by the government, and may never be; but without terminals and adequate landings much of the money spent on channels is wasted.

It was hoped that when good river channels were provided suitable terminals would immediately follow, but this has been true only in a few instances. The early method of making landings almost anywhere along the stream to pick up freight or put it ashore has passed; for the delay and expense of handling freight up and down steep banks and the lack of security and shelter without warehouses, are disadvantages that weigh heavily against the river. It seems hardly necessary to insist that towns should build and equip terminals and warehouses in advance of any possibility of their use, but they can at least purchase water frontage and definitely promise aid to anyone who will construct greater facilities whenever they may be needed. Assurances of meeting these requirements when they arise can be given. Even with these facilities provided, some public organization is nec