The comparative analysis of Ig class switch recombination in a priori IgG/IgA-expressing myelomas and hybridomas, in switch variants and in activated normal B cells shows the following characteristics of class switch recombination in activated B cells: It is prevented during most of B cell ontogeny. It happens on both IgH loci of activated and switched B cells. The recombination is programmed in that on both IgH loci of switched cells the same switch regions recombine with Smu. This is true at least for the IgG1 pathway. IgM-expressing cells show no class switch recombination on the inactive IgH locus. Thus, physiological class switch recombination is a programmed rather than a random event and is controlled as such. The initial stages of class switching and the molecules involved in these are largely unclear: What is the nature of the protection of switch regions and how is this protection abrogated? Do specific recombinases exist? What is the role of large transcription units? Is the specificity of class switch recombination a result of specific “opening” of the DNA for transcription? Do all B cells use the same switch mechanism? What is the role of switch factors (such as lymphokines)? These and more questions await answers and although a variety of switch scenarios could be discussed at present a detailed speculation seems premature.