Developing lymphocytes must be tested at critical ated IgMhiIgDlo/JHSAhi cells expressing higher levels of checkpoints to ascertain if they have made in-frame B220 and Bcl-2 are also found in the bone marrow and rearrangements of immune receptor genes and to deterare sometimes called transitional B cells (Carsetti et al., mine if individual cells express potentially useful antigen 1995). Newly formed B cells in the spleen are phenotypireceptors. Pre-B cells carrying in-frame rearrangements cally similar to these immature and transitional B cells at the immunoglobulin heavy chain locus (Rajewsky, in the bone marrow but express higher levels of B220 1996) are positively selected. Not every pre-B cell exand Bcl-2. They rapidly differentiate into recirculating pressing a productively rearranged immunoglobulin IgM+IgDhiCD23+ follicular B cells, which continue to exheavy chain gene may undergo selection and expansion; press high levels of HSA for a few days, but subsequently selection mediated by the pre-B receptor may be based mature into naive HSAint/lo cells. These latter cells constion reading frame choice as well as on considerations tute the major pool of long-lived naive B lymphocytes. of the apparent usefulness of certain Ig heavy chains in The BCR, as will be discussed below, is required for the humoral repertoire. Some of the known selection the development of naive B cells. Does this requirement and migration events during the development of naive B imply that this receptor is involved in a positive selection cells are depicted in Figure 1. Space limitations, coupled event or merely in a selection-neutral maturation prowith an attempt to provide a fairly comprehensive list cess? There are two broad pieces of evidence that are of mutations that affect cell selection events (Tables 1, frequently considered in support of the notion that im-2, and 3), have made it impossible for me to cite many mature B cells might be positively selected. It has been of the contributions I originally wished to. I apologize in suggested that the repertoire of newly formed peripheral particular to the many whose work is mentioned but not B cells is more limited than the repertoire of bone marcited. row B cells. Some studies have also suggested that only Newly formed B lymphocytes that bear self-reactive a small proportion of newly formed B cells may emerge antigen receptors may be silenced or eliminated in the in the periphery, indirectly implying that only properly bone marrow. They may be clonally deleted (Nemazee selected cells find their way out of the bone marrow or and Burki, 1989; Hartley et al., 1991), anergized by a receive signals that permit survival in the periphery. reprogramming event that does not involve further anti-A comparison of VH gene usage in B220+IgMJ pre-B gen receptor gene rearrangement (Goodnow et al., cells in the bone marrow with IgD+ mature B cells in the 1988), or given the opportunity to reform themselves spleen (Gu et al., 1991) suggested that there is a broader by generating new antigen receptors that are not self- BCR repertoire in the bone marrow as compared to the reactive (Radic et al., 1993; Tiegs et al., 1993; Pelanda periphery. This study raised the possibility that positive et al., 1997). While there is little doubt about the need or negative selection events, or both, occurring either to silence self-reactive B and T cells, why newly formed in the bone marrow or the periphery, may skew the naive naive lymphocytes need to be positively selected is less B cell repertoire during development. A small proportion obvious. of the skewing observed in this particular study might An attractive explanation for why we require positive have been …