Cretaceous floor which has already been described. There are strong geologic reasons for believing that this Cretaceous floor in most places east of the foothills has a greater elevation than the underground water level in the great trough-like area trending north and south to the west of the "Hogback" and the high divide, so that the eastward motion of the water would be impossible, except here and there where there might be a channel or depression in the Cretaceous floor similar to those along the Arkansas and Platte rivers.

Whatever conclusion may ultimately be reached regarding the possibility of an eastward underground movement from the mountains themselves, the rate of the movement of the water through the sand is, after all, of the most importance. When water is being pumped from a well, it is not so important to know the amount of water 10, 50, or 100 miles away, as it is to know the rate of movement through the sand of the water immediately adjacent the well. The maximum supply that the well can furnish will be dependent upon the rate of the inflow to the well, and only remotely upon the sum total of the water over the whole area. The rate of movement is likewise of great importance in considering the total amount of available water on the plains as a whole.

Should the precipitation and the proportion of the precipitation which joins the ground water be greatly increased, unless the possible rate of underground movement were correspondingly increased, available masses of water would tend to pile up where the precipitation occurs, without sensibly affecting the supply in other parts of the country. Those who believe in a mountainous source for a part or the whole of the plains water must therefore be able to show that the rate of eastward movement is sufficiently rapid sensibly to increase the available water at different places on the plains. Otherwise the increased precipitation in the mountainous areas would be of no avail.

It is doubtful if laboratory tests are of much value in this discussion, because the conditions governing them are not applicable. The inclination of the surface, the porosity of the strata, the ease with which water may pass from one stratum into another as the strata become discontinued, are some of the questions which must be considered. In any experiment neglect of any one of these conditions, and probably others not yet fully understood, will cause the results obtained to be of little if any value in this consideration. We may have a mass of sand or gravel largely composed of coarse grains or pebbles, so that at first thought it would seem that water would flow freely through it. If, however, there is a small amount of silt or clay disseminated through the mass, this will have a most important influence. The coarse masses of gravel in such cases not only do no good, but do harm; for each grain, being impervious, renders just that much space

impervious. It is probable that the rate of flow through a heterogeneous bed of this character is governed by the finest materials in the bed. Investigations of the subject, therefore, must consider the detailed geologic conditions of the area studied.

SUMMARY.

In summarizing the foregoing discussions regarding the amount of water, it may be concluded that there is a surprisingly large amount of water in the Arkansas Valley, and a similarly surprisingly large amount on the uplands, although perhaps smaller than in the valley. The springs along Spring Creek are the best single evidence we have of the large amount on the uplands, and the pumping tests in the valley, particularly at Hutchinson, are the best tests we have of the amount in the valley. These, combined with the almost universal presence of water, and the meteorologic and geologic conditions already explained, furnish good reasons for believing that the amount of ground water under the area here described is sufficient to meet any demands that will likely be made upon it, even should industries spring up in the West which will use many times as much water as now seems probable. The greatest needs for western Kansas now, and for the whole plains area, are better methods of raising water from beneath the surface, and better methods in using it.

IRRIGATION DEVELOPMENT.

Irrigation is now practiced to a considerable extent in the Arkansas Valley and along the Cimarron River and Crooked Creek. Numerous small reservoirs and irrigation pumps are also established on the uplands, some of which have proved to be surprisingly successful. In the valleys the principal crops raised by irrigation are alfalfa and fruit, but on the uplands little is irrigated besides gardens and fruit. The water in the Arkansas River is so low during the greater part of the year that the canals are dry, but in times of freshets they are used to a considerable extent, and the thousands of acres of alfalfa and orchards and other crops are quite thoroughly irrigated once a year at least, and upon the average two or three times. This is sufficient to produce a moderate crop of alfalfa during a dry year, while in an ordinary season it will give from three to four cuttings of hay.

The acreage varies from year to year, depending upon many matters which affect the prosperity of the farmers in that part of the State. Several hundred acres are also irrigated in the valley from pumps. In most places the water is raised less than 12 feet, so that windmills are very efficient, a 14-foot wheel furnishing ample power to run an 8-inch pump that will throw from 5 to 8 gallons at a stroke. Irrigation from mills is on the increase, as it frequently happens that dry weather in the spring and early summer is disastrous before water from the river is available.