During an 8 wk period (September 23 to July 30) 23 patients with histories of late-summer allergic rhinitis received, on a random basis, alternating 2 wk courses of either chlorpheniramine (anti-H1), 12 to 48 mg/day, plus placebo, or anti-H1 in the same dose plus cimetidine (anti-H2), 300 mg t.i.d. Symptom (SX) and medication (MED) scores were recorded twice daily. After the season the immediate (ICR) and late (LCR) reactions to allergens and histamine were measured in six of these patients while on: (1) no drugs, (2) anti-H1 plus placebo, or (3) anti-H1 plus anti-H2 in the doses employed during the season. Mean weekly SX scores were significantly lower on anti-H1 plus anti-H2 compared with anti-H1 plus placebo during the sixth week of the study (p less than 0.001). MED scores were lower on the anti-H1 and anti-H2 combination than on anti-H1 alone during weeks four and five (p less than 0.02). In the six patients studied after the season, the ICRs were consistently smaller with the anti-H1 and anti-H2 combination, but the differences were only significant for the histamine-induced wheal (p less than 0.01). In this study of allergic rhinitis there appeared to be a small but statistically significant additive effect of anti-H2 to the clinical response to anti-H1. No additional side effects were noted.