manufactured goods used by all of us. So states one of the first pages of this book. Because of the experience gained in industry the author was invited to deliver four lectures before the Graduate School of the Department of Agriculture with the thought that much was to be gained by correlating the development of statistical methods of research as applied in agriculture with the statistical methods of control in industry. The book contains the four lectures edited for publication.
Central to the theme of the four lectures is the concept of the act of control, which consists of the three components (a) the act of specifying the end to be attained, (b) the act of striving to attain the end specified, and (c) the act of judging whether the end has been attained. In mass production these are called specification, production and inspection. In scientific method they may be likened to hypothesis, experiment, and test of hypothesis.
Chapter I shows that the nature of the problem of judging whether a state of statistical control exists is essentially one of testing the hypothesis that assignable causes have been eliminated. Chapter I I takes up the practical problem of establishing tolerance limits that will make possible the most efficient use of raw materials and pieceparts. The problem of providing an efficient running quality report in mass production is discussed in chapter III. The fourth chapter takes up the simplest type of problem of specifying in an operationally verifiable way a state of statistical control of a single quality characteristic.
The treatment refers to applications on measurements and practical situations which go far in the way of assistance to clarity. It can be recommended to those interested in this field of endeavor.
R. H. 0PPERMANN.