Examination of the genomic sequence of the telomere region makes it possible to understand the evolution of the structure of chromosomal ends. We compared the genomic sequences of 14 chromosomal ends of rice, Oryza sativa, L., on the basis of the variation in TTTAGGG repeats. In the proximal telomere repeats, nucleotide substitution occurred more frequently than in the more distal repeats. The most significant diversity was observed at the 1st, 2nd, or 3rd position of TTTAGGG, suggesting that T has been a target of mutation preferentially. Copies of ATTAGGG, CTTAGGG, GTTAGGG, TTCAGGG, TTGAGGG, or TATAGGG were arrayed in tandem, or the same subtypes were located close to each other. The substituted variants were accumulated in chromosomes 2L, 3L, 7L, and 10S but not in the ends of the other chromosomes. In contrast, deletion variants, almost all of which were TTTAGGG to TTAGGG, were dispersed over approximately 4.9% of the sequenced telomere repeats. In summary, the rice proximal telomeric arrays were composed of blocks of at least 6 types of substituted variants and the canonical sequence in a chromosome-specific manner. These results suggest that the variants might arise from the rapid expansion of a single mutation rather than from the gradual accumulation of random mutations.