274 THE AMERICAN JOURNAL OF SOCIOLOGY
be only general statements. No one can recognize this more clearly than Dr. Underhill, who says, "It is at once clearly apparent that from such data results of any scientific value cannot be obtained, and emphasis is laid upon the fact that there has been no intention of drawing any conclusion the basis of which would necessitate exactness" (p. 319). In the investigation under review, the physi- ologist could not go back of the returns, and his results could not be more certain than the reports upon which he based them.
The sum-total of these difficulties should put students on their guard as to over-refinements in the interpretation of figures, origi- nally so uncertain, and should emphasize need of careful attention to method.
The reviewer desires once more to express his appreciation of the high grade of work done by Dr. Chapin and his colaborers who have done well, using a method that has in itself many serious diffi- culties and some positive limitations. He himself is under great obligations to the labor of these pioneers in this field. He recog- nizes the fact that this criticism of method is only negative, but hopes to make a positive contribution on this point at a later time.
T. J. Riley Washington University St. Louis
