A variety of sources was accessed in seeking material for this special issue of Safety Science. The starting point was the conference on Risk Homeostasis and Risk Assessment, originally the idea of our then colleague at Aston, Tom Hoyes, and scheduled for October 1993. After the untimely death of Tom Hoyes, in May 1993, the editors of this special issue took on the task of organising the conference, which was eventually held at Aston University in March 1994. Earlier versions of some of the papers in this issue were presented at that conference. Around the same time, the first of a series of UK conferences on Risk was organised at Bolton Business School, and a few of the contributions here originate from this conference. In July 1994, at the 23rd International Congress of Applied Psychology in Madrid, there was a symposium on Risk Homeostasis at which presenters offered reviews of their work in this field. The offer was made at that symposium for presenters to prepare their papers for submission to this special issue and thus a number more came from this source. Finally, we made a small number of approaches to selected persons working in the field to submit papers. While a combination of the normal refereeing process, editorial decision-making and authors' personal choice have precluded us from publishing all those papers which were originally submitted, we are pleased to be able to include a number which are written by ex-colleagues and would-be students of Tom Hoyes, as well as from others who have worked in the field for many years. From the studies described in the paper by Glendon et al. ( 1996, this volume) it is clear that the work of Tom Hoyes, even in his very brief career, made a significant contribution to the field of RHT. We are very pleased to be able to dedicate this volume to his memory. We would also like to acknowledge the support that we have received from the editors of Safety Science, Andrew Hale and Richard Booth, and to thank our reviewers for their perceptive and helpful comments on the papers submitted - efforts which are otherwise unrecorded.