Following the injection of B16 melanoma cells into the portal veins of mice, all animals developed liver tumors, but only 16% developed lung tumors. Portal vein injections of radiolabelled B16 and Walker 256 cancer cells into mice and rats, respectively, revealed that all of the cells were temporarily arrested in the liver and most were then slowly released. Bioassays indicated that of 8 X 10(4) B16 cells released from the liver over 24 h after portal vein injections of 10(5) cells, only approximately 1% were delivered to the lungs in a viable state. Experiments made with radiolabelled B16 cells showed that transit through either the liver or lungs following portal vein or tail vein injections, respectively, resulted in massive death of cancer cells. It is suggested that the death of most circulating cancer cells passing through the first organ encountered after leaving their primary tumor, serves to severely limit their further direct spread to other organs. It is therefore expected that metastases to these other organs would, to a large extent, be generated by cancer cells from metastases in the "first organs" as distinct from direct seeding from cancer cells released from the primary tumor. If the results of the present experiments have general application, they serve to emphasize the importance of metastasis of metastases in the natural history of the spread of cancer. In a previous publication (Weiss, 1980), studies of lung-to-liver traffic of cancer cells in rats revealed that, after tail vein injections, most Walker 256 cells were temporarily arrested in the pulmonary vasculature and then slowly released. A large proportion of the released cancer cells were dead or lethally injured on release from the lungs, and this "first organ processing" apparently accounted for the comparative rarity of extrapulmonary tumors following tail vein injections. In this communication, the concept of "first organ processing" of circulating cancer cells is further examined with respect to the liver-to-lung traffic of B16 melanoma and Walker 256 cells injected into the portal veins of mice or rats respectively. Both of these cell types grow well in the lungs following tail-vein injection.