Strain improvement has played a major role in the commercial production of microbial products such as antibiotics and enzymes.  Classical strain improvement using random mutagenesis techniques have been successfully employed to improve the productivity of many industrial microbial cultures.  Since it is a very labor-intensive process, many companies have shifted to use new genetic techniques for their strain improvement programs.  These new genetic techniques have been recently reviewed by Adrio and Demain (FEMS Microbiol Rev 30: 187-214, 2005) which include genome-based strain reconstruction, metabolic engineering, directed evolution, DNA shuffling, and whole genome shuffling.  However, classical strain improvement for increasing productivity of microbial secondary mtabolites and enzymes during the early phase of the development projects is still very effective when there is little knowledge on the genetic and the biosynthetic pathway of the producing microorganism.  This is especially true for the improvement of titers and the generation of blocked mutants of highly potent anticancer agents esperamicin and dynemicin produced by Actinomadura verrucosospora and Micromonospora chersina, respectively.  Furthermore, continuous culture can be used in combination with the classical strain improvement for the selection of desirable mutant strains.  Improved production of antibiotic patulin by Penicillium urticae and fructofuransidase inulinase by Kluyveromyces fragilis will be presented.