This index has been compiled with the aid of a computer program I developed using the dBase IV database application programming language. There are obvious advantages in using computers to help compile indexes but this program introduces the ability to typeset the output automatically.

The data contains typesetting markup code where appropriate, which, when coupled with similar markup code output by the program, results in a text file which the desktop publishing application (Corel Ventura 8) understands (this is very similar in concept to the approach taken by the Oxford English Dictionary). The result is what you see on the printed page, or in this case on the screen.

The program also fully integrates the index of each succesive volume with the cumulative index. As a new volume is completed, the index to it is added to the cumulative database. The program can generate just one year’s index, or a range: the same data are used to obtain the single volume index or an updated cumulative index. Thus, when Volume 48 is completed, it is no more work to output a new cumulative index to Volumes 21 to 48 than it is to produce just the index to Volume 48 (the limiting factor being the cost of publishing).

This computer program and the ideas behind it were described more fully in the article ‘Indexing Using a Computer’ in the Philatelic Quill, Vol. 13, No. 4, pp. 77–84 (October 1997).

I would like to thank Jean Alexander, Michael Brooks, Harry Dagnall, Robert Folkard, Chris Harman, Alan Huggins, Gordon Jessop, Graham Mark and Leslie Wilkinson for checking proofs of an earlier version of this index. Long telephone conversations with Jean, Harry and Graham brought forth much helpful advice and comments for which I am indebted.