The melatonin rhythm is significantly attenuated in a wide range of human and animal tumor types. Since surgical removal of the tumor has been shown to restore this rhythm, we hypothesized that a plasma borne tumor-associated factor (TAMF) could be responsible. A population of mice were injected with tumor cells and sequentially killed and bled over the following 9 days, i.e., to the maximal state of tumor growth. Pooled serum from the different collection days was added to an established pineal organ culture system, and melatonin concentrations measured. A highly significant correlation between melatonin concentrations and the stage of tumor growth was seen with maximal inhibition occurring at day 9 (P < 0.01). These findings support our hypothesis and may help to explain the mechanism whereby melatonin rhythmicity is suppressed in cancer patients.