Our Knowledge of the Fundamental Constants of Physics and Chemistry in 1965

Experimental data bearing on the precision determination of the numerical values of the fundamental physical constants are reviewed, with particular emphasis being placed on the identification and isolation of discrepancies and inconsistencies. The purpose of the analysis is to present a consistent set of values of the fundamental constants and to present a careful and complete description of the steps taken to reach this end. The Introduction discusses the significance of such an analysis and indicates the general method of approach. The indispensability of local unit systems and conversion factors connecting them, in order to avoid a sacrifice of precision peculiar to different metrological techniques, is emphasized. The point is stressed that conversion constants introduce the danger of ignoring error-statistical correlations between physically measured quantities, and the effects of such correlations on the assignment of errors is discussed. All available sources of experimental information relative to the necessary input data are presented, and changes in definitions of units since our last review are discussed. After the available stochastic input data have been reviewed and the less reliable items eliminated, the third section examines the remainder for mutual compatibility by means of an analysis of variance in which special criteria for recognizing the incompatibility of a datum are developed, using the analogy of the energy of internal strain introduced in overdetermined mechanical structures. Tables of least-squares adjusted values of fundamental constants and conversion factors of physics and chemistry based on the 1963 adjustment are given. Research pertinent to the constants which has been completed or published subsequent to the 1963 "recommended" adjustment is discussed, and the effect of these on our knowledge of the numerical values of the fundamental constants is presented.