Immunoglobulin allotypes of the mouse.

morethanoneclassofheavychainflodd, 1963)couldonlyberationalized if these markers were present in the variable region. This has been the subject of muchcontroversy, as the presence of a genetic marker is indicative that the gene bearing it is present as only a single entity in the germ line, and hence an argument for somatic mutation as the mode of generation of antibody diversity. Sequence studies (Wilkinson, 1969; Mole et a/., 1971) on Aal and Aa3 variable regions did not resolve this controversy as multiple differences were found on comparison of the three allotypes (Fig. 1). Such structural complexity has promoted arguments over just which, if any, of the noted differences define a V-region genetic marker. This difficulty was overcome by following the inheritance of a single amino acid difference between Aal and Aa3 proteins by using a combined electrophoretic and chromatographic system (Mole, 1975). A breeding study showed this difference to be inherited in a simple Mendelian fashion. It was argued that this difference could not be preserved in a large number of variable-region genes. Rabbit a locus allotypes have also provided strong evidence for the translocation hypothesis, that is the sharing of the same V-region gene pool by all the class-specific Ggenes by some mode of gene transfer. Sequence studies on rabbit a-chain variable region (A. P. Johnstone & L. E. Mole, unpublished work) confirm this view as all the a locus sequence correlates found for y chain were found in the a chain. Sequence data on the variable-region allotypes controlled by the x and y loci are scant, but it is now known (A. P. Johnstone & L. E. Mole, unpublished work) that the Nterminal sequence of one, perhaps both, of these gene products is strongly homologous to the human VHIII variable-region subgroup.