The complete sequence of pig switch mu (S mu) and the various domains of C mu, a total of ca 8 kilobases (kb) of sequence, are presented and comparatively analyzed. S mu starts about 2.5 kb downstream of the single pig JH gene and ends 0.4 kb upstream of C mu; the length of S mu is 3.2 kb. Three major pentameric repeats, gagct, gggct, and gggtt, comprise more than half of the entire S mu sequence. The two most frequent decamers are gagctgagct and gagctgggct. The longest single repeated sequence is 93 base pairs. The frequency of the various pentamers indicates that pig S mu is more similar to human S mu than to mouse S mu. The sequence of genomic pig C mu and its deduced amino acid sequence are similar to the C mu sequences published for other vertebrates and are most similar to those of sheep, another artiodactyl. The membrane and secreted tailpiece exons are highly conserved (ca > 90% similarity) and are actually identical to those of the sheep. The degree of similarity to other mammalian IgMs generally decreases from the carboxyl to the amino terminal domains, with the exception of C mu 2, which shows the lowest similarity among species (61% DNA, 44% aa). As in human C mu, we identified five potential N-linked glycosylation sites, two cysteines involved in forming interchain disulfide bonds, and the two additional cysteines involved in the linkage of IgM pentamers. This is only the third species for which a S mu sequence has been reported but the first from a species lacking the second virgin B-cell receptor, i.e., IgD. IgD in mice and humans lacks a switch region and its expression depends on the splicing of C mu-bearing mRNA with C delta. However, we found no evidence that the lack of C delta influenced the genomic structure of S mu, C mu, or C mu m.