alone there are 676,000 wells supplying water for irrigation, nearly all of which is lifted by men or animals.

The extension of pumping depends on whether or not it can be made to pay. If it can be, it will be a most influential factor in extending the area of settlement in the semiarid region. The irrigation of small tracts in conjunction with larger areas cultivated by dry farming will make it possible to grow more diversified and higher priced products, to maintain fruit and shade trees, thus beautifying the landscape and rendering the social and industrial conditions of farm life far more attractive. This kind of irrigation will pay as an insurance where the expense would otherwise be prohibitive.

The pumping investigations are dealing with two problems. One is the most efficient type of machinery for lifting large volumes of water from streams or canals. The other is the development of cheap, simple methods of lifting water for the irrigation of small areas and for lifting water from sources where the supply which can be delivered in any given unit of time is restricted. California, Arizona, the rice districts, and the sugar plantations of our insular possessions are the most promising fields for the development of irrigation through the construction of large pumping works, either by corporate capital or farmers' cooperative associations. Here the fundamental problem is efficiency. First cost and payment for mechanical skill in operation are subordinate considerations.

The semiarid region is the field for experiment to determine the best type of individual pumping plant. Here the investigations are being directed to determine the possibilities of windmills, gas engines, and electricity as motive power. The results of windmill irrigation at Garden City, Kans., Stockton, Cal., and a number of other localities where they have been extensively tried, render it desirable that the full possibilities of this form of power be determined, and this is being done by both field and laboratory tests in cooperation with the State Agricultural College of Iowa.

IRRIGATION IN THE HUMID PORTIONS OF THE UNITED STATES.

In 1903 the legislature of Wisconsin appropriated $2,500 a year for two years for an investigation of cranberry irrigation by the Wisconsin Experiment Station. This Office has been cooperating with that station in studying the conditions needed to make this industry profitable. Cranberry growing is largely dependent on such control of water as will permit of fields being flooded quickly and drained thoroughly. The work carried on during the past season has included the collection of information about the amount of water used, losses by seepage and evaporation, the effect of standing water of different temperatures on the berries and vines, and the influence of irrigation and drainage in preventing injury from frost.

In cooperation with the New Jersey Experiment Station similar studies are being made in the cranberry districts of New Jersey.

In other sections of the humid States the Office has given advice about irrigation methods, about the selection and installation of pumping machinery, and has continued its collection of data to determine the profits of irrigation in the East. Last year was an unfavorable one because of the excessive rainfall.

LEGAL AND ECONOMIC QUESTIONS.

At the request of the Modesto and Turlock irrigation districts in California this Office began in 1904 a study of the legal and economic questions connected with canal management, the purpose in this case being to aid the farmers in these two districts to perfect a system of administration which would prevent waste of water and controversies between farmers. As there are 270,000 acres of irrigable land in these districts, and the canals will in time serve many thousand irrigators, the working out of an effective distribution system is an important and somewhat complicated matter. As a first step in working out such a system records of the quantities of water delivered to each user were kept during 1904 by an agent of this Office, and on his recommendation the districts have adopted a system for keeping records of the flow of water distributed to each user and the time during which the water is used. These studies are to be continued in California, Washington, and Oregon in sections where the land under large canals is being brought under cultivation.

The most conspicuous weakness of our present irrigation development grows out of our failure to recognize that irrigation is an organized industry, that the relations between water users should be clearly defined, and that the same administrative regulations for the distribution of water from a river or large canal system are needed as are required in the transmission of goods by railways and express companies. In this respect our systems are far inferior to those of older irrigated districts.

The investigations of methods of distributing water from rivers in Colorado, Wyoming, and Nebraska show that there are great dangers of water monopoly and a necessity for more effective public control, and that a lack of a strict enforcement of priorities causes great waste in the use of water and costly and vicious litigation between farmers.

In cooperation with the State engineer's office a study was made of the practical working of the Idaho State law. An agent of this Office was appointed water commissioner, and in that position dealt practically with the problems of distribution. One of the greatest difficulties in just division of water was the lack of proper measuring devices. Another was the rendering of decrees by the court giving

to appropriators excessive amounts of water and recognizing the validity of sales of these surplus appropriations. In cases where sales were made it resulted in injustice to subsequent appropriators, and where they were not made there was a constant temptation on the part of the irrigator to use water excessively, in order to divert all that was decreed to him and hence maintain his right thereto until it could be sold.

One feature which needs to be studied is the methods of organizing farmers under laterals in order to lessen the expense of canal management and the losses of time and water in irrigation by a satisfactory system of rotation such as prevails in Italy and is described in Bulletin 144 of this Office.

The growing diversion of water from streams in Louisiana to irrigate rice has called attention to the inadequacy of the State water laws, and an agent of this Office has been for two years past studying the legal and economic problems of this State for the purpose of recommending additional legislation.

The chief of this investigation has been called upon to advise State commissions with respect to irrigation legislation, and is now detailed as an engineer to aid the Government in the litigation over the Arkansas River, in which the State of Kansas, the State of Colorado, and the United States Government are all parties.

DRAINAGE.

The drainage investigations carried on during 1904 dealt with the engineering questions incident to the location and construction of field drains and the legal and economic problems which have to be solved where united action of many landowners is a necessity. Personal examinations of conditions, and in some cases surveys, have been made by engineers connected with the Office in order to render practical assistance and promote the best practice. The importance of drainage for irrigated land is found to be more urgent as these investigations proceed.

Examination of saturated land in Cache, Emery, and Washington counties, Utah, was made in 1904, and soil measurements in the Fresno district, California; Sunnyside, Wash.; Truckee and Carson valleys in Nevada, and Yellowstone Valley, Montana. By means of records of the rise and fall of soil water, obtained through the sinking of wells in the irrigated districts, the influence of excessive irrigation on the water plane has been studied and information obtained which will furnish a working basis for fixing the dimensions of drainage works to relieve the swamped areas. The season of 1904 demonstrated the practical value of these drainage studies, in every instance the works constructed according to the recommendations of the Office having proved effective, and considerable

bodies of land which had been rendered worthless have been in this way restored to fertility. In one instance a tract, which in 1993 produced no crop because of the rise of seepage water, in 1904, after being drained, produced a crop which sold for $600 an acre.

In company with a representative of the Utah Experiment Station, an examination of the swamped areas of a number of Utah counties was made by Mr. C. G. Elliott in the spring of 1904. This showed that drainage is one of the most necessary irrigation improvements in Utah, there being scarcely an irrigated valley in the State in which some of the best land has not become too wet for cultivation or from which only uncertain crops of inferior value can be obtained. In a number of instances attempts to drain these swamped areas have proven unsuccessful. The reasons for this are explained in a subsequent portion of this report. In order to determine how best to overcome the obstacles encountered, an experiment in drainage has been inaugurated in Cache County, where tile drains have been put down and their effects on the removal of seepage water are being studied.

During the past year there has been a marked extension of drainage studies in the humid portions of the United States. These have included examinations and reports on the drainage problems of districts in Nebraska, Iowa, South Dakota, Wisconsin, and Illinois, and in cooperation with the engineering department of Purdue University a comprehensive inquiry into the results of tile drainage in the Mississippi Valley during the past twenty-five years has been inaugurated. Requests have been received that this investigation be made to include problems connected with the construction of dredged drain ditches and the best methods of maintenance and repair. Should means be provided, it is intended to undertake this in connection with a study of the best means of reclaiming the Kankakee marshes of Indiana and some of the swamped areas in other States of the Mississippi Valley.

During May and June of 1904 an investigation was made of the drainage conditions of the Illinois River bottom lands, extending from Peoria to the mouth of the river. These lands vary from 1 to 3 miles in width and lie from 8 to 20 feet above low water, while from 1897 to 1904 the high-water line varied from 13.7 to 19.9 feet, during four years reaching the 18-foot mark. When cultivated the higher parts of the bottom yield crops about two years out of three, but the lower parts flood so frequently that it is not profitable to cultivate them; but if these lands are protected by dikes and properly drained they produce large crops of corn and wheat. On account of the slight fall of the river, the level surface of the bottoms, and their slight elevation above low water the reclamation of these lands is expensive and difficult. Near Pekin, Havana, and Beardstown

organizations have been formed under the provisions of the State levee and drainage laws and attempts made to reclaim tracts of this land by means of levees, drain ditches, and pumps. The results of these improvements have not been satisfactory, because the levees were not built high enough and in some instances the pumping stations were not properly located. The investigation was to aid in determining the nature of the improvements required. The reports show that the complete reclamation of these lands is feasible and can be accomplished at a profit, the estimated annual net profit for the districts examined being $2.26 per acre. As the problems to be dealt with and the nature of the work necessary to success are now better understood by landowners, more satisfactory results will doubtless be obtained in the future than were realized in the first attempts.

OTHER BRANCHES OF RURAL ENGINEERING.

The studies of the use of power in pumping and irrigation shows that many millions of dollars are spent by farmers each year in the purchase of this class of machinery alone. They also show how important it is that our schools and colleges give increased attention to the training of farmers in these matters. Through lack of knowledge of mechanical principles and skill in the use and repair of machinery, it breaks down when it should not, and wears out before the natural term of its service is completed. The cost of this machinery to the farmer is one of his heaviest outlays, and the gain to the country as a whole, by increasing its life through better care or more skillful operation, would in the aggregate amount to millions of dollars each year. The difficulty of providing for this is greater to-day than ever before because of the more complicated types of machinery used. The traction engine, the steam plow, and the automobile, the machinery used in dairies, in the cultivation and harvesting of rice, in the growing of sugar beets, and the manufacture of beet sugar, are illustrations of the momentous changes in the character of farm machines which have taken place in the past fifty years.

Nor is this need of increased knowledge of rural engineering confined to farm machinery. The changes which are taking place in the character of farm buildings make a knowledge of structural design, of ventilation, and of the cost and durability of different materials all matters of interest to the individual farmer and of great importance to the country as a whole. That this is being recognized practically is shown by the large number of inquiries on these matters which come to the Department, and which are referred to this Office for answer. It is also shown by the increased attention being paid in the agricultural colleges to courses of instruction in rural engineering. It is one of the functions of this Office to aid these institutions in increasing the efficiency of their different lines of