Highly efficient plant regeneration from protoplasts via somatic embryogenesis in citrus has allowed for expanding applications of somatic hybridization and cybridization to citrus variety improvement with increasing complexity. Citrus somatic hybrids are now possible from most desirable parental combinations, and hybrid plants have been produced from approximately 300 combinations to date, including more than 150 at the CREC. Applications for citrus scion improvement target the development of improved seedless fresh fruit varieties, and include symmetric somatic hybridization to produce superior alloteteraploid breeding parents (for use in interploid crosses to generate seedless triploids), haploid + diploid fusion for direct triploid production, and targeted cybridization to transfer mtCMS (cytoplasmic male sterility) from Satsuma mandarin to seedy but otherwise desirable diploid cultivars. The primary application for rootstock improvement is the production of symmetric somatic hybrids to combine complementary rootstocks without breaking up successful gene combinations. Rootstock somatic hybridization is providing opportunities for improving disease and insect resistance, soil adaptation, and tree size control. Wide somatic hybridization provides an opportunity for gene transfer from related species, including some that are sexually incompatible. Extensive field research on citrus somatic hybrid rootstocks combined with emerging molecular analyses of citrus has allowed for the development of additional strategies for rootstock improvement. These include rootstock breeding and selection at the tetraploid level using somatic hybrid parents, and the resynthesis of important rootstocks such as sour orange at the tetraploid level, via fusion of parents selected specifically to overcome the deficiencies of the original rootstock. Ongoing examples of each strategy will be provided, along with suggestions for extending the technology to other commodities.