The objectives of this study were to establish the prevalence of respiratory, eye, nose and throat symptoms of likely work-relation in workers exposed to colophony solder flux fumes and to assess their lung function. A cross-sectional study was conducted in four medium-sized electronics firms in which control measures to capture solder flux fume were absent or visibly ineffective. All female solders and women working adjacent to soldering stations completed an administered questionnaire concerning symptoms, work history and current soldering frequency. Measurements were made of their forced vital capacity (FVC) and forced expiratory volume in one second (FEV1) during the course of a working shift, using a Vitallograph-Compact portable spirometer. Using weekly hours of soldering as a crude index of current exposure, workers were classified into high (> or = 37 h/wk) and low (< or = 20 h/wk) exposure groups, and their health responses were compared in the analysis. Individuals with symptoms suggestive of work-related asthma were also asked to provide serial peak flow measurements over a further 2-week period, and adequate returns were charted and read by two physicians experienced in the diagnosis of occupational asthma. Data were collected on 152 female workers (overall participation rate = 97%). Symptoms of recurrent, persistent wheeze and/or chest tightness were reported by 75 (49%) of interviewees; 36 (24%) gave a history typical of occupational asthma and six more (4%) a history of pre-existing asthma worsened at work. Twenty-one (14%) of the workforce complained of recurrent breathlessness on moderate exertion; 41 workers (27%) had work-related symptoms of the nose or throat and 25 (16%) had work-related eye symptoms. The odds ratios for 'all wheeze', shortness of breath, and work-related eye, nose and chest symptoms were all significantly greater (raised about 4-5 fold) in women who soldered > or = 37 h/wk when compared with those soldering < or = 20 h/wk. After adjustment by logistic regression for atopy, age and smoking status even higher risk estimates were generally obtained. The odds ratios (OR) and 95% confidence intervals (CI) for high vs. low were: for 'all wheeze', OR = 7.2, CI = 2.5-20.7; for work-related eye symptoms, OR = 5.2, CI = 1.4-19.8; for work-related nasal symptoms, OR = 4.0, CI = 1.4-11.1 and for occupational asthma symptoms, OR = 5.2, CI = 1.4-14.2. Mean FEV1 and FVC percentage difference from expected were slightly lower in full-time solderers than in part-time solderers, but the differences were not significant. Thirty-seven of the 51 workers (73%) who were asked to carry out serial peak flow measurements completed an adequate return: 27 of these records confirmed the presence of asthma, and in all of the cases the history suggested onset post-dating employment in soldering. Eleven peak flow records were indicative of occupational asthma. The health problems associated with colophony solder flux were documented over 18 years ago, but are still clearly apparent in situations where adequate control has not been achieved.