Rubber tree (Hevea brasiliensis) is a member of the family Euphorbiaceae, sub-family articulated crotonoids. The 10 species of the genus Hevea have 2n = 36 chromosomes and there has been a suggestion of allotetraploidy with x = 9, but genome mapping of hundreds of genetic markers revealed only disomic segregations in H. brasiliensis. Restriction fragment length polymorphism (RFLP) mapping allowed revealing 10 duplicated chromosome segments, but there is no clear evidence of multivalent pairings during meiosis. Consequently, Hevea species have to be considered as diploids with 2n = 2x = 36. Status of the genome of Manihot, a genus phylogenetically close to Hevea and belonging to the same botanical sub-family, is very similar with 2n = 36 chromosomes for all Manihot species, including cassava (Manihot esculenta). In contrast, despite these similarities, the nuclear genomes of rubber tree and cassava differ greatly in size, with a haploid genome (1 C) of 2,100 mega bases (Mb) in H. brasiliensis and of 800 Mb in M. esculenta. It can be postulated that this genome size difference is mainly attributable to the amount of repetitive DNA which constitutes probably more than 64 % of the Hevea nuclear genome, and that conversely the single/low copy fraction, encompassing the transcribed regions, probably represents less than 750 Mb in this species. Therefore, whole genome sequencing of rubber tree will be a tremendous task. New DNA sequencing technologies theoretically open to the possibility obtaining a draft sequence of the whole genome even for large genome species. Obviously, it would be useful to concentrate DNA sequencing effort on the low single/copy fraction in order to avoid wasting time and money in repetitive DNA sequencing. But, even in this case, a whole genome sequencing work would first produce a huge number of unarranged small contigs, and after this first step, a great amount of work in physical mapping will be necessary. Moreover, sequence contig assembly is