Back in 1962, when I had by accident become the supervisor of Roberto Celis in Argentina, it occurred to me that antibody diversity might arise from the joining by disulphide bridges of a variety of small polypeptides in combinatorial patterns.