Analysis of literary data (for Saccharomyces cerevisiae, Caenorhabditis elegans, Arabidopsis thaliana, Homo sapiens, and some other Eucarya) and our data (for Schizosaccharomyces pombe) on the exon-intron organization of the genes encoding subunits of nuclear RNA polymerases showed that introns in the orthologous genes from different organisms are arranged nonrandomly, namely, their positions, if projected on the map of the comparison of the amino acid sequences of the orthologous subunits, not infrequently coincide in evolutionarily distant species. As a rule, intron positions correspond to the boundaries of the structurally conserved regions (domains) or to the sites of possible turns of the polypeptide chain. For example, introns flank the secondary structure elements in the Rpb8 subunit with the known three-dimensional structure or the structure-function modules in subunits Rpb10 and Rpc10. These facts are in agreement with the idea of the ancient origin of introns, and with the notion of evolution of ancient protein sequences through the assembly of their genes from short protoexons selected by the nature as far back as the RNA world times. Comparative analysis of the primary structures of the subunits of eukaryotic RNA polymerases allowed us to reveal a nuclear localization signal in subunit Rpb10 and some hypothetical archaeal homologues of subunit Rpc10.