Although not 'ideal', xenograft models of human hematopoiesis have proved to be extremely useful experimental systems for the study of the in vivo engraftment/proliferation potential of human hematopoietic stem/progenitor cells (HSC). Among these, the human/sheep xenograft model is unique in several respects. This model takes advantage of fetal immunologic immaturity and developing 'homing' spaces in the fetal bone marrow to obtain donor HSC engraftment in normal recipients without marrow conditioning. This avoids both the possible stromal abnormalities associated with irradiation- and/or chemotherapy-induced immunodeficiency which may limit human HSC engraftment or long-term maintenance of human hematopoiesis, and the use of genetically deficient hosts which may result in restricted donor cell expression. In the human/sheep xenograft model, human HSC (1) colonizes the bone marrow, (2) persists for many years, (3) is capable of multilineage differentiation, (4) retains its ability to respond to human cytokines, and (5) retains its ability to engraft/differentiate in secondary recipients. Another unique aspect of the sheep model is its large size; this permits repeated evaluation of human cell activity in the same chimeric sheep over long periods. The model appears to discriminate between the different populations of human HSC and exhibits a degree of sensitivity that suggests this in utero approach may serve as a biologically relevant model for the identification, characterization, and study of the in vivo potential of human HSC from available sources.