Adequate sampling is important to minimize the risk of false-negative fine-needle biopsy (FNB) diagnoses. This article reports a prospective study of the following criteria for adequacy of sampling for FNB benign diagnoses: at least six aspirates, at least two of which each contain at least six clusters of benign cells. Of 888 patients studied with FNB, 60 had FNB diagnoses of malignancy; 58 were surgically confirmed, and there was one false positive. Seven of eight nodules with diagnoses of suspected malignancy were excised; three were malignant. Fifty-seven nodules had diagnoses of cellular or Hurthle-cell adenomas; 35 were excised, and four were malignant. All 33 nodules excised after benign diagnoses (including Hashimoto's thyroiditis) were benign. Five of 23 FNBs with small numbers of benign cells were malignant, but false-negative FNB diagnoses were avoided because of adherence to the proposed criteria for adequacy of sampling.