Hematoporphyrin derivative (HPD), a complex mixture of porphyrins has been used clinically as a tumor localizer both for diagnostic and therapeutic purposes. Relative lack of tumor specific uptake limits widespread clinical application. In an attempt to circumvent this problem, we studied 28 porphyrins with widely differing properties using a transplantable KHJJ mammary carcinoma in the mouse (Balb C) as a tumor model. Twenty hours after porphyrin (P) administration, the tumor, skin and gastrointestinal tract were excised and the latter lavaged with physiologic saline. Tissue porphyrin content was assessed visually by red U.V. fluorescence, and by quantitative fluorometric extraction, and photodynamic activity was evaluated in vivo using a tuneable dye laser emitting red light (615-640 nm). (1) Of the five porphyrins which were taken up by tumor tissue, i.e. HPD (photofrin), photofrin II, meso-tetra (4-carboxyphenyl) porphine (TCPP), tetra sodium-meso-tetra (4-sulfonatophenyl) porphine (TPSS), and uroporphyrin I (UROP I), skin and intestinal fluorescence also was marked with the notable exception of those mice receiving UROP I. (2) UROP I clearly was superior to HPD in this study in that the tumor to skin porphyrin content ratio was more than seven times greater than the ratio observed with both HPD and photofrin II, (3) As no measureable UROP was present in the gut, the tumor:intestinal P ratio under conditions of assay was for practical purposes infinity. (4) Photodynamic necrosis in the tumors of UROP I treated animals was similar to that seen following treatment with HPD. We conclude that UROP I appears the most specific tumor localizing porphyrin yet studied. This specificity suggests major potential clinical application both as a diagnostic marker for early mucosal cancer, and in photoradiation therapy. Moreover, the prolonged photocutaneous side effects as seen with HPD are unlikely.