It has been thought that metastases are clonal and originate from rare cells in primary tumors that are heterogeneous in genotype and phenotype. Recent studies using DNA array analysis challenge this hypothesis and suggest the genetic background of the host is the important determinant of metastatic potential implying that metastases are not necessarily clonal. Previous methods to determine clonality of metastasis used karyotype or molecular analysis that were complicated, thereby limiting the number of metastatic colonies analyzed and the conclusions that could be drawn. We describe here the use of green fluorescent protein-labeled or red fluorescent protein-labeled HT-1080 human fibrosarcoma cells to determine clonality by simple fluorescence visualization of metastatic colonies after mixed implantation of the red and green fluorescent cells. Resulting pure red or pure green colonies were scored as clonal, whereas mixed yellow colonies were scored as nonclonal. In a spontaneous metastasis model originating from footpad injection in severe combined immunodeficient mice, 95% of the resulting lung colonies were either pure green or pure red, indicating monoclonal origin, whereas 5% were of mixed color, indicating polyclonal origin. In an experimental lung metastasis model established by tail vein injection in severe combined immunodeficient mice, clonality of lung metastasis was dependent on cell number. With a minimum cell number injected, almost all (96%) colonies were pure red or green and therefore monoclonal. When a large number of cells were injected, almost all (87%) colonies were mixed color and therefore heteroclonal. We conclude that spontaneous metastasis may be clonal because they are rare events, thereby supporting the rare-cell clonal origin of metastasis hypothesis. The clonality of the experimental metastasis model depended on the number of input cells. The simple fluorescence method of determining clonality of metastases described here can allow large-scale clonal analysis in numerous types of metastatic models.