The purpose of this study was to determine the measures that photosensitive use to control their sun exposure. Each week from March until September 1995, 30 patients with polymorphic light eruption (PLE) and 17 patients with chronic actinic dermatitis (CAD) returned a set of reply paid postcards on which they recorded information about their outdoor behaviour and symptoms. The principal differences between the two groups were that CAD patients had a much greater incidence of symptoms despite making more use of protective measures such as covering arms, wearing hats and applying sunscreen, than patients with PLE. And that as summer approached the PLE patients spent more time outdoors, whereas there was less seasonal variation in this respect among CAD patients. Tentative conclusions drawn from mathematical modelling indicated that the incidence of rash on a particular day was influenced by ambient ultraviolet radiation and length of time spent outdoors. There were indications that wearing a hat and keeping the arms covered offered some protection, whereas use of sunscreen may actually increase the likelihood of symptoms.