In an attempt to detect cancer within the tissue, patients were injected intravenously with 32P solution in a dose of 10 μCi/kg, 48hours later the radioactivity in the tissues wasdetermined using a catheter-type semiconductor radiation detector. The result of animal experiments revealed that the radioactive count was higher at the site of a tumor than at the inflammatory siteBy macro-and micro-autoradiography it was confirmed that 32P accumulated at the site of a tumor and that those beta rays were counted. The present study was conducted in cases with squamous cell carcinoma as a clinical study, and the relationship between the determined values and the histopathological results were compared. The results were as follows: 1) It was impossible to differentiate numerically the negative site from the positive site of cancer based on the radioactive count. 2) Significant differences were observed between the negative sites of cancer and the positive sites of cancer when the ratios of increases in radioactive counts were compared between the two sites, based on the count at the normal tissue. 3) It was possible to conclude that increase ratios below 0.8 are negative for cancer and above 3.8 are positive. However, it was notpossible to judge those between 0.8 and 3.8 by this method. Clinical findings and otherauxiliary diagnostic methods should be utilized for such cases. 4) This test method provides potential information when cancer is suspected macroscopically. 5) Of various factors influencing the radioactive count, the change in a tumor tissue appeared to be the main factor changing the uptake of 32P. 6) This test can be utilized in diagnosis of cancer, finding local recurrences and lymph node metastases.