Ten middle-aged patients (7 males and 3 females) suffering allergic rhinitis and endonasal polyposis, were polypectomized and studied with allergoimmunologic techniques. Clinical evaluation preceded laboratory examinations consisting in routine data and electrophoretic study of serum proteins. These examinations were performed before and after one-year hyposensitization treatment. One patient revealed a hypogammaglobulinemia and received 800 mg of commercial human gammaglobulin every 28 days, during 3 months; subsequent proteinograms showed acceptable improvement of the initial value. Skin tests resulted positive in all patients particularly to house dust, micotic (Candida albicans, Trycophyton and Alternaria) and bacterial antigens (Gram positive and negative bacteria). Specific hyposensitization was performed throughout one year with weekly subcutaneous injections of the corresponding antigens. No autovaccine of nasal exudates was employed. All the polyps were divided into two parts; one half for histological study and the other half stored at -20 degrees C for protein determination using immunoelectrophoresis, polyacrilamide gel and radial immunodiffusion. Immunoelectrophoresis was performed with polyps homo-genate against an antihuman serum and showed two precipitin lines at alfa-1 and beta-1 globulin sites. Polyacrilamide gel revealed more lines in those runs belonging to post-vaccinated polyps than in those belonging to the prevaccinated group. Immunoquant polyps were checked against anti-A, anti-G and anti-M; only IgA was detected in polyps obtained at any time. These apparently contradictory results may be due to the different sensitivity of the techniques applied as well as the small amount of polyps homogenate and its low protein content. New polyps reappeared in three patients at six, ten and twelve months, respectively, from the beginning of the treatment. These patients were again polypectomized and histological and immunological studies were repeated. They were fixed in Bouin, and paraffin sections of 6 to 10 microns in thickness, were submitted to hematoxilineosin, P.A.S. and toluidin blue techniques. These preparations were microscopically evaluated with a Carl Zeiss Photomicroscope with magnifications of 100, 250, 400 and 1,000 times. An important difference between the original and the recurrent histological images was seen. Meanwhile the former polyps were vascularly rich and cellularly poor, the latter appeared inversely composed. An abundant infiltrate of plasmocytes, lymphocytes and histocytes was the rule; it was regularly disposed but a preference to collect around the vessels was observed. The vessels seemed to be lesser in number and size though this observation must be checked with further investigations. The absence of eosinophils led us to support--in this case--the participation of an infectious mechanism in an allergic filed with high sensitivity to bacterial and inhalant antigens...