We have undertaken a double blind placebo controlled study of the effect of nasal beclomethasone on the tendency to wheeze in 20 unselected hay fever sufferers, half with a history of previous seasonal wheezing. We found no difference between either bronchial hyperresponsiveness, as measured by methacholine challenge, home-monitored PEFR, nor recorded wheeze nor cough between treated and placebo groups although the numbers were small. All were allowed the antihistamine cetirizine hydrochloride 10 mg daily. Eighteen out of the 19 patients had either bronchial hyperresponsiveness (PD20 methacholine < 8 mumol or a > 2 doubling dose change in their PD20 during the pollen season). We have shown a significant positive correlation between a hay fever score (HFS) (created by taking the sum of the home scored; nasal discharge, nasal blockage, eye irritation, sneeze and antihistamine use) and peak seasonal specific IgE to mixed grass pollen (Spearman correlation coefficient 0.5 P < 0.02). There was also a positive correlation between the rise in specific IgE from pre to peak season and the HFS, correlation coefficient 0.6 P = 0.03).