The main target of mammography in asymptomatic women is the early diagnosis, or rather the identification, of non-palpable breast cancers. Doubtful or suspicious findings on conventional mammograms with no clinical evidence call for radiologic or other complementary imaging techniques to assess the exact lesion nature. Direct magnification and US are targeted techniques to employ as additional investigations after conventional mammography. Fifty consecutive patients were referred to our department of radiology for the preoperative localization of non-palpable breast lesions previously identified on conventional mammograms. The diagnostic or complementary roles of direct magnification and of US were thus investigated. US was always repeated during the preoperative localization; a 10-MHz immersion sectorial probe was used. Magnification was performed if absent or poor in conventional mammograms. The contribution of each technique to conventional mammography was graded as valuable (A), medium (B), or null (C). The lesions were grouped according to their structure: microcalcifications (a), nodules (b), scars (c), and complex lesions (a+b, a+c, b+c, ecc.). Six cases are included in our series which had been diagnosed as questionable or suspicious on previous mammograms. In our department, they were diagnosed as benign. Two of them were operated on because biopsy was required by the gynecologist and the other underwent stereotaxic FNB: negative cytology was considered the final diagnosis. Forty-six histologic and 4 cytologic examinations diagnosed 25 malignant and 25 benign lesions. Direct magnification was of great value in all cases, whereas US was useless in microcalcifications and useful in nodular or complex lesions, especially those with a nodular component. However, the incidence of US false-negatives was high, even in very suspicious cases on mammography, which suggests that US negativity cannot be considered an adequate sign to rule malignancy out.