The optimum mitogen concentration and time required for using whole blood from dogs in a microassay were determined, and this test then was compared with a standard lymphocyte-stimulation microtest, using gradient-isolated lymphocytes, 2 different mitogens (phytohemagglutinin and concanavalin A), and 2 different culture media. Statistical analysis of the data from 10 dogs showed that whole blood was significantly more reactive than were gradient-isolated lymphocytes (P less than 0.05). Waymouth's medium was significantly better than RPMI 1640 (P less than 0.001), and concanavalin A was significantly more mitogenic than phytohemagglutinin (P less than 0.001). The interaction between lymphocyte source and mitogens was the only one of the various interactions that was significant at P less than 0.05.