In all popular discussions of death rates or of the probable duration of human life one finds the phrase “according to the mortality table.” Sometimes it even appears that a writer is assuming that a definite law of mortality has been discovered from which deviations are not to be expected and which has been embodied in the mortality table. Every serious student of the subject knows that this is far from the truth and that there are many mortality tables, each representing the mortality which was experienced by a specified group of lives during a specified time.

To the actuary, any mortality table is simply a mathematical tool with which to study such problems as population growth or change, or to calculate the monetary values of benefits dependent on life contingencies. It is a tool in the same sense that a gauge is a tool for a machinist. In some parts of a machinist's work only the most accurate precision gauges are acceptable, and similarly in calculating premiums for non-participating insurance or rates for life annuities, the actuary must use the most accurate and up-to-date mortality table available. On the other hand, a machinist may use a less exact gauge for other types of work where conditions are suitable. For certain purposes such as calculating policy liabilities, the actuary has a corresponding choice, for, as will appear later in the paper, reserves by all mortality tables are remarkably similar.