To investigate the molecular basis for the phenotypic diversity and evolutionary relationships within Oryza sativa, we performed large-scale end sequencing and chromosomal in silico mapping (mapping by sequence homology) of rice BAC (bacterial artificial chromosome) clones derived from an indica rice cultivar, ‘Kasalath’. In total, 78,427 high-quality BAC-end sequences (BESs) showing an average read-length of 482 bp with a total of 37.8 Mb of genomic sequences were obtained from 47,194 BAC clones. After removal of those BESs containing repetitive sequences and use of the high-quality genomic sequence of the japonica rice cultivar ‘Nipponbare’ as a reference standard, 12,170 clones with paired BESs were mapped in silico to the 12 chromosomes. These clones consisted of 450 contigs and showed a total physical length of 308.5 Mb, indicating a coverage of the rice genome of about 80%. Confirmation of the chromosomal positions of the Kasalath BAC clones mapped on chromosome 1 using specific DNA markers revealed that the map accuracy was extremely high: at least 94.8%. In general, the genomic composition and structure of all the chromosomes were highly conserved between the two subspecies. However, we still found evidence suggesting the existence of different components in the genome of the two subspecies within large chromosomal fragments, especially on chromosomes 11 and 12. Extensive analysis of the genomic sequences of Nipponbare and Kasalath revealed a frequency of 0.71% for single nucleotide polymorphisms (SNPs) and 1.23 sites per kilobase for indels (1–16 bp length), respectively. The BAC-based Kasalath map could become an invaluable resource, not only for the isolation of genes, but also for conducting extensive analyses of genomic sequences for comparative genomic studies within or between species.