The function of B cells we have chosen for review is their maturation into antibody-producing plasma cells. Our own recent studies have centered on the factors that control the differentiation of mitogenstimulated B lymphocytes in human newborns. Human studies are necessary because the differences between the gestation times of experimental animals and man make extrapolation between species uncertain even with the use of age equivalence.1 While there is little evidence that healthy human fetuses are antigenically stimulated, responses to congenital infections do indicate that they have the capacity to respond before birth. Detailed information on the maturation of immune responsiveness in utero depends, however, on knowledge acquired from animal experiments. Relevant results from these are therefore briefly reviewed first. GENERATION OF ANTIGEN BINDING (IDIOTYPE) DIVERSITY Hypotheses to account for the means by which stem cells give rise to many clones carrying different individual specificities for antigen (idiotypes) have centered around the "germ line" and "somatic mutation" theories. These two may not be mutually exclusive. Convincing evidence for the germ line theory has come from sequence analysis of myeloma (monoclonal) proteins2 and, more recently, from hybridization studies.3 In the latter experiments the detection of DNA sequences coding for a myeloma-like λ light chain in mouse embryonic DNA indicated that some specificities are encoded in the genome. Experiments in mice4 and chickens5 have suggested that there is a genetically predetermined sequential expression of variable region genes during development which is followed by the appearance of a larger spectrum of specificities in the adult.