We have sequenced the variable heavy chain regions of a number of V_H4–21 encoded monoclonal IgM anti‐Rh(D) antibodies produced in response to deliberate immunization. These were compared with the sequences of similarly encoded IgM anti‐I cold agglutinins (CA) derived from patients with lympho‐proliferative diseases. The anti‐Rh(D) antibodies show evidence of clonal expansion and somatic diversification. Even though they are produced in response to an antigenic stimulus, they demonstrate limited hypermutation in the variable heavy chain (V_H) segments and there is no evidence of selective pressure acting on the complementarity determining regions (CDRs). The CA demonstrate a higher rate of mutation and yet this results in a lower ratio of replacement to silent mutations (R:S) in the CDRs than seen in the anti‐Rh(D) antibodies. It is not clear whether the different pattern of mutations seen in the CA is related to their auto‐reactivity or their tumour origin. In both groups of antibodies the region encoded by the V_H4–21 segment can be found in germline configuration at the amino‐acid level indicating that other V‐gene structures, i. e. light chains or CDR_H3s, are crucial to the generation of either specificity. A role of the CDR_H3 is indicated by the identification of a motif shared by four CAs and one Rh(D) antibody which also demonstrates CA activity independent of its anti‐Rh(D) specificity. Amongst the anti‐Rh(D) antibodies there seems to be an obligatory combination with V_L having closest homology to the DPL16 germline segment indicating this as particularly important in generation anti‐Rh(D) specificity.