Induction of parthenocarpic fruit set in watermelon by pollination with bottle gourd (Lagenaria siceraria (Molina) Standl.) pollen

• Seedless watermelon can be obtained by pollinating with bottle gourd pollen.
• Fruit set after pollinating with bottle gourd pollen was higher than expected.
• Parthenocarpic fruit were almost the same as control fruit except for fruit shape.
• Parthenocarpy resulting from pollination with bottle gourd pollen was not pseudogamy.

Production of seedless watermelon (Citrullus lanatus (Thunb.) Matsum. and Nakai) using soft X-ray irradiated pollen has the advantage that this method can be used with diploid cultivars; however, pistillate flowers must be covered with a bag before flowering to prevent normal pollination by insects. Therefore, we attempted to prevent normal pollen penetration by pollinating pistillate watermelon flowers with intergeneric pollen before anthesis. In the course of conducting these studies, we found a new method to produce seedless watermelon by pollinating with bottle gourd (Lagenaria siceraria (Molina) Standl.) pollen. When the pollen of bottle gourd was used to pollinate pistillate watermelon flowers, the fruit set rate was 57.1%. On the other hand, the fruit set rate by pollinating with watermelon pollen was 65.0%. Although the fruit set rate using bottle gourd pollen was slightly lower than the fruit set rate obtained when watermelon pollen was used, this percentage was higher than expected. Preservation of pollen collected from staminate flowers of bottle gourd was difficult, and fruit did not set. In contrast, fruit developed when bottle gourd pollen from staminate flowers stored at 4 °C for 72 h was used to pollinate watermelon. All parthenocarpic fruits produced by pollination with bottle gourd pollen were deformed (triangular or oblong shaped). There was a significant difference in the ratio of fruit length to fruit width; however, fruit weight, rind thickness, flesh color and Brix in the parthenocarpic fruit were almost the same as control fruit. There were no normal seeds except for small, white empty seeds in the fruit obtained from pollinating with bottle gourd pollen. Bottle gourd pollen tubes did not reach the ovules of watermelon ovaries. Therefore, we concluded that parthenocarpy resulting from pollination with bottle gourd pollen was stimulative parthenocarpy, not pseudo parthenocarpy (pseudogamy).