SOMACLONAL VARIATION IN CUCUMIS SATIVUS L.

Abstract

Hybrid seed of cucumber (Cucumis sativus) is expensive to produce. Production of `artificial seeds' through somatic embryogenesis may be a viable alternative. Somatic embryos were induced, multiplied on a semisolid medium for 8-10 weeks, and germinated on agar-gelled medium before transplanting to soil. It was then important to determine the extent of variation among plants derived from somatic embryos. The criteria for variation among regenerants of cultivar Clinton were; plant height, fruit shape, fruit weight and number (yield/plant), days to first female flower and variation in selected isozymes. All measurements were taken on greenhouse-grown plants. Some regenerants of Clinton were also planted in the field and they flowered and, qualitatively, bore fruit as well as the zygote-derived plants. When quantitative measurements were made, variation was greater than for plants from zygotic embryos, but the visual impact was that there was little variation amongst regenerants. Regenerants grew more slowly and tended to yield higher numbers of slightly smaller fruits than plants from zygotic embryos. Average yield per plant was higher for somatic embryo-derived plants. For cultivar Corona only morphology of plant and fruit was examined. One plant was especially visually mutant and unacceptable as a commercial plant. The defects were readily identifiable in the seedling stage. Preliminary evidence suggests that `artificial seeds' of cucumbers may be a viable proposition.