Complications related to lymphangiography have been minor and easily treated in most instances. The possibility exists, however, that the passage of contrast material through cancerous lymph nodes may facilitate metastasis. This possibility became apparent to us after we viewed histologic preparations of lymph nodes removed from patients subjected to lymphangiography. The histopathology of these nodes revealed marked inflammatory changes, acute and chronic; replacement of portions of lymph node parenchyma by oil droplets; and decreased number of germinal centers (Figs. 1 and 2). Similar findings have been reported by Ravel (4), who studied a large series of nodes removed at varying intervals following lymphangiography. Our study was undertaken to determine whether lymphangiography affected tumor metastasis by any one or a combination of the following factors: (a) facilitation of metastasis from the marked alteration in lymph node architecture, (b) increase in blood-borne metastasis through lymphovenous shunts during or following the procedure, and (c) facilitation of metastasis within the lymphatic system from the increased intralymphatic pressure during the procedure. Materials and Methods One hundred three white female New Zealand rabbits, two to three months old and weighing 5–6 pounds, were used in the study. They were housed individually in air-conditioned quarters and given water and Rockland rabbit pellets freely. A modification of the technic of Zeidman and Buss (8) for tumor preparation and infusion was employed. The tumor used was the transplantable V–2 rabbit carcinoma which has been carried in our laboratory for the past three years. Tumor-cell suspensions were prepared by homogenizing small pieces of tumor tissue in physiologic saline. The homogenate was then passed through a sieve into a sterile test tube and packed in ice until used. A cell-viability count was made by Schrek's method (6). These suspensions were prepared to contain 1 × 106 ± 200,000 viable tumor cells per cubic centimeter. The animals were divided into three groups. In Group A (control animals), V–2 tumor suspension was infused intralymphatically into the afferent popliteal vessel of the left hind leg. Group B animals were subjected to direct lymphangiography with 1 cc of Ethiodol3 ten days prior to tumor cell infusion. Group C rabbits were subjected to direct lymphangiography ten days after tumor cell infusion. The rabbits were anesthetized with intravenous sodium pentobarbital. The medial side of the left hind leg was shaved and prepared with tincture of benzalkonium chloride An oblique incision over the medial aspect of the lower left hind leg was used to expose the afferent lymphatics. These vessels were usually found alongside a tributary of the great saphenous vein. To visualize the lymphatics 0.1 cc Direct Sky Blue 4 per cent4 was injected into the foot pad just before the procedure.