As a part of studies of streptococcal infections of the upper respiratory tract, sera from 142 individuals (mean age 11.9 years) with signs and symptoms on tonsillitis or pharyngitis and their family contacts were studied to quantitatively evaluate the sensitivity and specificity of the Streptozyme test. We also compared this agglutination test for streptococcal extracellular antibodies with the antistreptolysin O and antideoxyribonuclease B tests. Using different lots of Streptozyme reagent we found evidence suggesting variation in the strength of the reagent from lot to lot. Evaluation of the group specificity of the Streptozyme test suggests that it may not be specific for antibodies to extracellular products of group A streptococci, since a response to this test was seen in individuals with only non-group A strains isolated from their upper respiratory tracts. In patients with group A streptococci isolated from the upper respiratory tract, the quantitative sensitivity of the antibody response as measured by the Streptozyme test is comparable to, but no greater than, the ASO or anti-DNase B tests. Our analysis of upper limits of normal for the Streptozyme test in the study population indicates that previous recommendations for values for elevated titers are too low, especially for predominantly pediatric populations, so that previous reports of patients demonstrating an antibody response to Streptozyme-measured antibodies may include a significant number of patients with "normal" levels. This study indicates the need for both further documentation of standardization of the reagents used in this agglutination test and additional quantitative studies of the response measured by Streptozyme.