Because the indepen- dence of polymorphisms resolved by clones GM727, GM798 and GM842 is likely to be violated by the tight linkage between the segregating restriction sites, “single locus” statistics for each presented throughout the paper (with the exception of allele numbers) represent means calculated over the multiple polymorphisms revealed by the same clone. If linkage disequi- librium between these variable sites is complete, this amounts to using information provided by only one polymorphism per cDNA clone. If the multiple sites are in linkage equilibrium, this averaging provides a summary statistic for each structural gene region. Because the detection of a polymorphic restriction site by a specific cDNA clone is unaffected by linkage, statistics on allele numbers are based on all 17 RFLPs. The treatment of linkage disequilibrium will be addressed in a future paper
(G. H. POCSON, K. A. MESA and R. G. BOUTILIER, unpublished observations).