We report our experience of the incidence of false-positive (i.e., clinically positive and histologically negative) and false-negative (i.e., clinically negative and histologically positive) neck nodes and define their relationship to the primary site, T-stage, and the histologic grade of the primary lesion. Clinical and pathologic records of 255 patients with squamous cell carcinoma (SCC) of the head and neck who underwent 266 radical neck dissections were reviewed. We did not find any relationship between the T-stage of the primary tumor and the incidence of the false-positive and false-negative neck nodes. Regarding the primary site, the benefit of an elective neck dissection is likely to be maximum in oropharyngeal SCC and minimum in supraglottic SCC. The incidence of false-negative neck nodes was lower in well differentiated primary lesions.