The sequence organization, evolution, and transcription of cloned repetitive DNA elements in sea urchin were studied. Total S. purpuratus DNA was reassociated such that only the reiterated sequences were duplex. The flanking single-stranded DNA was digested with S1 nuclease and synthetic Eco RI sites were ligated to the ends. These molecules were cloned in RSF2124 and recombinants were selected by their Amp+Col- phenotype. These clones ranged in reiteration frequency from 3 to 12,000 copies per haploid genome. The mean intrafamilial divergence was as low as 4[degrees]C for some families which could be attributed entirely to polymorphism. Other families contained members that showed greater divergence, some as high as 25°C. Nine randomly selected sequences were found to be transcribed in oocyte RNA, gastrula hnRNA, and intestine hnRNA. The concentrations of these transcripts are tissue specific and unlike single copy sequences both strands were represented in the RNAs. The transcripts were longer than the repetitive elements themselves indicating linkage to single copy sequences. The number of copies per haploid genome was determined in S. franciscanus and L. pictus in addition to the parent species. The ratio of the reiteration frequency in S. purpuratus to S. franciscanus ranged from about 20 to 1. The copies remaining in S. franciscanus and L. pictus genomes were, however, conserved relative to average single copy DNA sequence. The plasmid clones were used to select individual family members from total genomic [lambda] libraries and their characteristics were investigated. Families 2034 and 2108 were found to be members of the long repetitive sequence class. The 2034 elements exist in clustered arrays while the 2108 family is dispersed. The 2109 elements were shown to be largely of the short repetitive class, however, about 10% of the members occur in long repetitive regions. The interspersed elements were found to be flanked by low reiteration frequency or single copy sequences. Thus, a large number of genomic regions are physically linked via the 2109 repetitive sequences. The implications of these findings with respect to evolution and gene regulation are many-fold and are discussed in detail in the following chapters.