Abstract
With the exception of appendices 2 and 3, the following is based on the after dinner talk given by Professor John W. Tukey at, the first meeting of the Section of the Physical and Engineering Sciences of the American Statistical Association held in New York City on May 26, 1955. This talk was repeated at a later date before a dinner meeting of the Metropolitan Section of the American Society for Quality Control. On both occasions considerable discussion ensued. The talk is published here both for the record, and in the hope that some readers may be stimulated to prepare written rejoinders.

This publication has 0 references indexed in Scilit: