Objective Evaluate palpation, ultrasonography (USG), and computed tomography (CT) with respect to their value for comparative determination of metastatic lymph nodes in head and neck cancer. Study design A prospective study was conducted in a tertiary referral hospital. During September 1996 and April 1999, clinical investigation, USG, and CT were performed preoperatively on 48 patients who would undergo neck dissection for primary head and neck malignancy. The presence of metastatic lymph nodes in pathologic specimens were compared with the findings of palpation, CT, and USG. Results Twenty-two of the patients had positive lymph nodes on pathologic investigation. The sensitivity, specificity, negative predictive value, positive predictive value, and accuracy for palpation are 64%, 85%, 74%, 78%, and 75% respectively. Respective values for USG and CT for sensitivity, specificity, negative predictive value, positive predictive value, and accuracy are 72%, 96%, 80%, 94%, and 85% and 81%, 96%, 85%, 90%, and 87%. Conclusion All of the pretreatment evaluation methods, palpation, CT, and USG yield results that are significantly different from the histopathologic results, suggesting that no pretreatment study can accurately assess the requirement to histopathologically stage the neck. USG findings are more correlated with the pathologic findings than palpation, but CT gives the most effective and reliable results when it is combined with USG in neck staging. Preoperative USG and CT scanning of the neck by an experienced radiologist is essential and useful for diagnosis, staging, and therapy choices. Significance Statistical analysis was done with χ 2 test. The difference between histologic findings and palpation is significant ( P = 0.0005). The difference between histologic findings and USG and CT is significant ( P = 0.0001 and P = 0.0001).