Water is the most valuable of the mineral resources. The study of ground waters is therefore clearly within the field of economic geology and constitutes an important part of the work of the geological surveys, both State and national, as defined by law. In the spring of 1911 the investigation of the ground waters of North Dakota was begun by the North Dakota Geological Survey, and the work was assigned to the author of this paper. During each of the three summers 1911, 1912, and 1913 several weeks were devoted by the author to the field work of a general survey. A report on the ground waters of the State was then prepared by him and was transmitted by the director of the North Dakota Geological Survey to the State printing commission for publication. However, owing to lack of available funds the report was not published.

A portion of the summer of 1914 was given to a study of artesian conditions in the Souris River Basin. Since that time considerable work has been done in connection with detailed surveys made for a number of cities and villages in an effort to obtain the best available supply of water for public use.

During the summer of 1920 arrangements were made by the United States Geological Survey with the North Dakota Geological Survey whereby the author completed the work as fully as possible by correspondence and brought the report up to date.

In the spring of 1921 samples of water from 196 sources were collected by the author and J. H. Buchanan and were sent to the United States Geological Survey for analysis. Most of these samples were analyzed by H. B. Riffenburg, who has used the analyses for a description of the chemical character of ground waters in the State. In addition to the analyses of samples collected in connection with the preparation of this report, over 700 partial analyses from different sources were examined. These analyses are not given in this paper, because the location of many of the wells was not stated definitely, and most of the analyses were incomplete. They were useful, however, in confirming the conclusions based on the analyses that are printed, particularly for counties where only a few samples were collected for this report.