Abstract
In a previous paper I argued that any enthusiasm for modifying general lighting practice to take account of the impact of light exposure on human health would be premature. Since that paper was written enthusiasm for this course of action has waned somewhat. This is not because enhancing human health is considered a bad thing but rather because of the evergrowing physiological complexity that has been revealed and the increasing concern that circadian disruption might be involved in the development of various cancers. But there has been progress. There are now models of the spectral sensitivity of the circadian system and some estimates of the threshold amount of white light required to influence it. There is also available equipment for measuring and recording circadian light exposure, in the field, over several days. This equipment offers the possibility of identifying lighting conditions that lead to circadian disruption. As a result of these developments, the use of light exposure as a means of enhancing health for at least some people remains a possibility but not a certainty. Until a full understanding of the nonvisual effects of light exposure is available, the best approach to lighting for human health is to try to replicate the conditions under which mankind evolved—daylight by day and little light at night.