Reported here are the results of a large-scale trial conducted under the guidance of Prof. Giovanni Motta, which I coordinated. A total of 656 patients with nasal hyperreactivity were recruited in 51 Italian centres (18 in the north of Italy, 20 in central Italy and 13 in the south). The trial's results were as follows. 1) A clear clinical prevalence of sensitization to different allergens in the different areas of Italy, which could be roughly classified thus: a) in Northern Italy birch and grasses were in the main species; b) in Central Italy mites prevailed but oleaceae were also significant; c) in Southern Italy parietaria and oleaceae were the prevailing species. 2) Most cases were sensitive to several allergens, although a considerable proportion (22%) actually presented reactions only to one allergen and those responding account of the patients sensitive to only one allergen and those responding to a main allergen, the proportion of patients reacting clinically to only one allergen rises to 64%. 4) The specific nasal provocation test (sNPT) offers specificity comparable to in vivo diagnostic methods such as the prick test, and in vitro methods such as RAST, but is much more sensitive. 5) The sNPT can be done in any season. 6) The sNPT is highly specific below a threshold value of nasal reactivity, which can be identified for each allergen studied and expressed in Allergenic Units. 7) In patients in whom the prick test shows multiple sensitivity, the specific NPT identifies the allergen presumably responsible for the nasal reactions (main allergen). In the light of these findings double-blind specific immunotherapy was started, to last 1 year, in 107 patients (49 given placebo and 58 active treatment), with nasal allergy to grasses, parietaria and mites. The results of this treatment were as follows: 74.1% of patients presented a reduction in nasal resistance, measured by dynamic anterior rhinomanometry, indicating relief of nasal obstruction; mucociliary transport time became normal in 81% of patients, meaning that rhinorrhoea had become less marked; there was significant rise in the nasal reactivity threshold in 74.1% of patients, illustrating the degree of desensitization achieved; nasal IgA increased by 62.5% of patients and IgG in 55.2% indicating improvement in the local immunological picture. No such improvements were detectable in the patients given placebo. In conclusion, therefore, the findings of this trial in allergic rhinitis underline that the specific NPT proved more sensitive than other in vivo and in vitro diagnostic methods and is unquestionably a fundamental investigational approach for assessing nasal allergies, identifying the allergens causing the symptoms and setting up rational local immunotherapy. The sNPT also showed the efficacy of specific intranasal immunotherapy.