Blood from various commercial sources used for transfusion at two New York City hospitals, samples from a large commercial blood bank, and samples from the city's major volunteer blood source, Were tested for SH antigen by the Ouchterlony technic. The different commercial sources averaged an antigen positive incidence of 1.2 per 100 units, whereas the volunteer source had less than one positive per thousand. The relative risk of transfusion hepatitis between the two sources was approximately 12: 1. This is in accord with follow‐up studies of patients. Improvements in the sensitivity of the antigen test and exclusion of positive units would reduce the absolute risk of posttransfusion hepatitis but would not be expected to alter the relative risk. Because of this, volunteer blood sources should be considered preferable to commercial sources, at least in large urban area