Mouse peritoneal exudate cells induced by thioglycollate medium can form colonies in soft agar with a plating efficiency of about 5% (0.6%–10%). Cells from an unstimulated peritoneal cavity form no colonies or have a plating efficiency of less than 0.001 %. These colony‐forming cells from the peritoneal exudate are similar to bone marrow colony‐forming cells in vitro in that they both require a substance(s) present in conditioned medium from L‐cells or mouse embryo fibroblasts or the serum from endotoxin‐treated mice for the initiation and the continuation of their growth. However, peritoneal exudate colony‐forming cells have a much longer initial lag period (10–14 days) and can survive longer in the absence of L‐cell conditioned medium than bone marrow colony‐forming cells. Only mononuclear cells, presumably macrophages, are observed in peritoneal exudate colonies, whereas bone marrow cell colonies contain both polymorphonuclear cells and macrophages.