An accurate prediction of the prognosis of melanoma is dependent on the ability to determine the presence or absence of malignant cells in lymph nodes. Our laboratory has developed a method for screening lymph node sections for the presence of micrometastatic disease. The nodes obtained from regional lymphadenectomies are dissected free from surrounding tissue and divided into halves. One-half is subjected to routine histopathological evaluation and the other one-half is disrupted into a single cell suspension and placed into cell culture. Lymph nodes from 19 patients with malignant melanoma were evaluated by tissue culture. Tumor cells were isolated in 125 of the 173 nodes examined. In contrast, only 14 of these nodes were initially found positive for melanoma by routine pathological evaluation. All lymph nodes positive for melanoma by pathological evaluation were also culture positive. Examination by cell culture has led to an upgrade in the stage of disease in 58% of the histologically node-negative patients. In addition, 43% of the patients who were histologically node negative but culture positive have recurred, suggesting some clinical correlation. These results indicate that the tissue culture method for detecting micrometastatic lymph node disease may prove more reliable and sensitive than current techniques for determining the presence of occult melanoma.