when correlated with records of discharge, show the variations in character of the water with changes in discharge and seasons. They also furnish the basis for classifying the water for irrigation, domestic, and various industrial uses.

For the publication of the results of stream measurements the country has been divided into twelve primary drainage basins, and an annual progress report is published separately for each of these basins. For the publication of the data as to quality of water, utilization of streams, and other special data, no fixed territorial basis is used. Most of the data concerning either surface or ground water are now published in the Survey's series of "Water Supply Papers." Occasionally the study of the water resources of an area for a special purpose, as municipal water supply, results in a publication covering both surface and ground waters in the area studied. The Survey's investigations of surface waters are put to numerous and important uses. They furnish valuable, and concerning some streams, absolutely indispensable data needed in the consideration and development of irrigation, power, water supply, navigation, and flood protection projects. Their results have been of high value in connection with the irrigation of the arid lands of the West. The location of reclamation projects not only by the national government but by state and private enterprise has for the greatest part been guided by the data gathered by the Survey. Less well known but hardly less important is the part which the Survey's findings have played in aiding power development on the streams of the country.

It is difficult to define in terms of time or of magnitude the entire project of surveying the surface waters of the country, and it is indeed to be questioned whether the work may properly be regarded as definitely terminable, for it is not yet apparent that a point will be reached when the collection of stream measurements and the analyses of water will become unnecessary; on the contrary it is believed that this work, like the collection of climatic data, must be continued indefinitely.

With these limitations in mind, it may be stated that stream measurements covering a period of years are now available for 4,000 points on the streams of the country. The collection of data as to the quality of surface waters has made less progress. Reports including data on the quality of the surface waters in nearly every state have been published, though only six reports cover the quality of surface waters of entire states, and these reports consider only the principal streams. In general, it may be said that most of the principal streams of the country have been investigated for their quality, but these streams constitute a very small percentage of the total surface water supplies of the country.

Ground Waters. The location of ground waters is essentially a geologic problem and is determined principally by geologic methods. It involves an examination of the strata in the region explored and the determination of the position, character, and capacity of those that are water-bearing. These geologic methods give results that are in the nature of forecasts. Drilling is necessary to determine with certainty the yield, head, and quality of the waters sought, but until the fiscal year 1917-18, the Survey did no test drilling for this purpose. The appropriation act for that year provided that of the $175,000 appropriated for water resources investigations, $25,000 might "be used to test the existence of artesian and other underground water supplies suitable for irrigation in the arid and semiarid regions by boring wells." It will be noted that this provision includes only supplies suitable for irrigation. Under this authorization the Survey has begun drilling in a desert area in Nevada where the geologic conditions are believed to be favorable, and has already discovered a large supply of water that will be available for irrigation.

The determination of the quantity of ground water in any given area is difficult. In some areas it is practicable to measure the flow of the ground water by electrical apparatus inserted in test holes or to make estimates based on the ob

served fluctuations of the water table, which represent the filling or emptying of the underground reservoir. In other areas it is practicable, by observation of surface conditions, to determine approximately the quantity of water that annually enters or is discharged from the underground reservoir. These observations comprise stream measurements to ascertain the quantities of surface waters that percolate into the ground, and measurements of the discharge of ground water through springs and by evaporation from soil and plants. These methods are laborious and difficult and none of them are likely to afford precise results, but they are nevertheless valuable, because they give a tangible basis for estimating the supplies of ground water.

In addition to determining the quantity and quality of the waters a survey of ground waters discloses important factors affecting the practicability of recovering them, such as hydrostatic pressure and the nature of the rock to be encountered in sinking wells. The investigation of these elements is involved in the study of the local geology, which is supplemented by examinations and tests of the existing wells in the area surveyed.

The data gathered as to sources of ground water are naturally of value chiefly in the arid and semiarid regions of the West, where such supplies are needed both for domestic use and for irrigation, but the importance of deep ground water supplies over the whole country is being emphasized. It is highly desirable to obtain supplies of domestic water from sources other than the shallow wells, some of them open, that are sunk near many houses. The water obtained from deep wells has percolated through sands and other material for so great a distance that its impurities have been removed by filtration, and it possesses a sanitary value that cannot well be overestimated, for it is free from the bacteria that cause typhoid fever and the protozoa that cause amoebic dysentery, and its use obviates the necessity for shallow wells that may

serve as a breeding place for Anopheles, the mosquito to which malarial infection is due.

Altogether the Survey has accumulated a vast amount of accurate data of both scientific and economic value on the ground waters of the United States. A total of 474 papers has been published relating in some way to ground water, of which 271 papers relate primarily to this subject. Many of these papers include detailed maps showing the ground-water conditions in specific areas. In addition to the data already published there is a great quantity on file at the Geological Survey still unpublished.

About 75 per cent of the area of the United States (excluding Alaska and the Insular Territories) has been covered by ground-water surveys of some sort, and about 25 per cent has been covered by what may be called standard groundwater surveys. The total area covered by intensive surveys is very small, amounting to only a fraction of 1 per cent of the total area of the United States.

The Survey for some years has published in its water supply papers notes on desert watering places, but in 1917 a specific appropriation of $10,000 was granted by Congress "for discovering, developing, protecting, and rendering more accessible, springs, streams, and water holes on arid public lands of the United States; for erecting and maintaining suitable and durable monuments and sign-boards; and for providing convenient and ready means, apparatus, and appliances by which water may be brought to the earth's surface." The Survey has devoted the funds thus appropriated to prosecuting more vigorously than has hitherto been possible the survey of watering places, and to preparing guides and maps, and erecting sign-posts directing to them.

The total area sufficiently arid to require maps and guides to desert watering places is about 570,000 square miles, or nearly one-fifth of the area of the United States, exclusive of Alaska and the Insular Territories. Of this area about 75,000 square miles have been surveyed since July 1, 1917,

and additional areas of approximately the same aggregate extent are covered by older guides to desert watering places. The region to be covered lies east of the Cascade and Sierra Nevada Mountains and west of the following cities: Boise, Twin Falls, Salt Lake City, Grand Junction, Santa Fé, Roswell, Pecos, and Fort Stockton. The area has been tentatively divided into twenty-six districts, for each of which a guide is ultimately to be prepared. The preparation of maps and guides and the erection of a fairly adequate number of sign-posts for the entire area will cost about $100,000. It is hoped that this work can be done within the next few years.

Investigation in Problems of Water Utilization and Control. From time to time the Survey undertakes the investigation of large problems of water utilization and control. The early investigations made by the Survey into the problem of irrigating the arid lands of the West, and the more systematic work of the Powell Irrigation Survey, in 1888-90, have already been mentioned. It is worthy of note that in his report for 1888, in discussing the problem of irrigation, Major Powell adverted also to the possible reclamation of the flood plains of the lower Mississippi valley by the control of the headwaters of that stream and its chief tributaries. A problem closely akin to this was brought expressly within the province of the Survey by the act of 1908, to which more extended reference will be made in the next section, imposing upon the Survey the duty of examining forest lands proposed to be purchased by the government for the prevention of floods on navigable streams.

While one or another problem of water utilization or control has thus from time to time been brought within the purview of the Survey's activities, no express authorization to investigate generally problems of this type has ever been conferred by statute. Such investigations as are currently carried on are conducted under the general authorization to investigate "the water resources of the United States."