Chromosome duplication and ploidy level determination in African nightshade Solanum villosum Miller By C. O. OJIEWO 1,2 , S. G. AGONG 2 , K. MURAKAMI 1 and M. MASUDA 1* 1 Faculty of Agriculture, Okayama University, 1-1-1 Tsushima Naka, Okayama, 700-8530, Japan 2 Department of Horticulture,Jomo Kenyatta University of Agriculture and Technology, P. O. Box 62000, Nairobi, Kenya (e-mail: mmasuda@cc.okayama-u.ac.jp) (Accepted 18 October 2005) SUMMARY Octoploids were induced in vivo from wild-type tetraploid Solanum nigrum ssp. villosum plants using colchicine sprays. Seedling survival rates and numbers of induced octoploids from 72 seedlings were 95.8%, 73.6% and 48.6%, and 4, 2 and 1, from 0.01%, 0.05% and 0.25% colchicine treatments, respectively.The applicability of pollen area and stomatal length, as indirect methods to determine ploidy level, was investigated. Further confirmatory tests involving direct chromosome counts (in root tip cells) and flow cytometric analysis revealed that pollen and stomatal cell size may not correlate accurately with ploidy level. Although octoploids generally had larger pollen and larger stomata, plants that were identified in the first generation (G 1 ) progeny on a large-pollen and large-stomata basis were not necessarily octoploids. In addition, a number of tetraploids also had “large” stomata or “large” pollen. Flow cytometric analysis revealed that this species exhibits polysomaty in leaf tissues (i.e., the tissues consist of cells with different ploidy levels), which could affect the size and morphology of both pollen grains and stomatal (guard) cells, thus explaining the inconsistency observed. We conclude that plant pollen and stomatal size can provide a good general guide to ploidy level determination in this species; but confirmatory tests, including direct chromosome scoring in root tip cells and flow cytometry in young leaves, are indispensable. A frican nightshade (Solanum nigrum L.) and related species are used widely as leafy herbs and vegetables, particularly in Africa and South East Asia (Edmonds and Chweya, 1997). Solanum nigrum is a complex of several morphogenetically distinct taxa distributed throughout most parts of the World. In Africa, the sub-taxon S. villosum is the most widespread and most popularly consumed. The nutritional composition of African nightshade is comparable or even superior to some exotic vegetables produced on a commercial scale (FAO, 1988). Leaf yields in these vegetables are, however, limited by prolific early flowering and excessive fruiting. Species belonging to the S. nigrum complex are predominantly autogamous, favouring the production of many small berries and seeds, which compete with leaves for photosynthates. Dry matter partitioning after anthesis is directed mainly towards pollen, fruit and seed formation and development, each demanding a high amount of energy. Elimination, or a marked reduction in berry and seed set appear to be a necessary step towards improving leaf yield. In Kenya, a few commercial-scale farmers try to circumvent this problem by manually deflowering plants. Mwafusi (1992) reported a 40% increase in leaf yield, with deflowered plants yielding 2,154 kg ha –1 compared to 1,539 kg ha –1 without deflowering. Whether done by hand or using chemicals, deflowering is expensive, time- consuming and labour-intensive, making it a serious constraint to full domestication and commercialisation of this crop. We recently proposed a strategy to induce and introduce new male-sterile varieties, postulating that by eliminating the reproductive function and subsequent fruits, which are the major assimilate sink, the male- sterile mutants should be able to use the energy no longer allocated to pollen, and eventually to berry and seed, to produce additional leaves (Ojiewo et al., 2005). We are currently testing the yield potential of induced male-sterile mutants against their wild-type counterparts. Another strategy with high potential for leaf yield improvement in these vegetables is “heteroploidy”. Heteroploids are organisms or cells with a chromosome number that is not an even multiple of the basic chromosome number for that species. During meiosis of heteroploids, heterovalents are formed (Cassani and Caton, 1985). During the following anaphase I, the chromosomes are distributed into both daughter cells. Only in rare cases, does one get exactly double the amount (2n) of the simple set (n). Generally, both receive incomplete sets (aneuploidy), resulting in an imbalance in chromosome composition, leading to lethality. Therefore, with few exceptions, heteroploidy causes sterility or highly reduced fertility of both pollen and ovules. Wild-type S. nigrum ssp. villosum is a natural tetraploid (2n = 4x = 48). Chromosome duplication would therefore yield an octoploid mutant (2n = 8x = 96). A back-cross of the octoploid to the wild-type parent would then result in a hexaploid (2n = 6x = 72), which could be further back-crossed to produce a pentaploid (2n = 5x = 60). Alternatively, a cross between the hexaploid and the *Author for correspondence. Journal of Horticultural Science & Biotechnology (2006) 81 (2) 183–188