In a group of 33 typical hay fever patients with apparent conjunctivitis a double-blind clinical trial was carried out during the summer of 1978. One group of patients used 2% cromoglycate eye drops and the other group used placebo eye drops, both in sterile unit-dose packages without benzalconium hydrochloride. Symptoms and use of other symptomatic treatment were recorded daily. The mean daily scores were compared with the daily pollen counts. The daily records of symptoms were somewhat lower in the cromoglycate than in the placebo group, though not significantly so. Mean daily scores of symptoms correlated with the pollen counts fairly well. Need of symptomatic medication did not differ in either groups and had no relation to the pollen counts. Although a beneficial effect had been expected, cromoglycate eye drops were considered ineffective in this study.