Nine Hundred Flowers and Four Per Cent Fruit Set
Encik R owns 40 mature durian trees of a single clone on 0.6 ha. In April each tree produces about 900 flowers, opening between 19:00 and 22:00. For five consecutive nights there was heavy rain from 18:00 until midnight, and daytime temperatures reached 34 °C. Fruit set was about 4% of flowers. In a neighbouring orchard 2 km away, planted with four different clones, the trees flowered in the same week and set 28% of flowers. Encik R sprayed ethephon at flowering and gibberellic acid two weeks later; neither treatment changed the set. He then bagged 10 inflorescences in muslin before anthesis: none set fruit. He hand-pollinated another 10 bagged inflorescences at 21:00 with pollen collected from a different clone: 7 set fruit.
1. Explain why the bagged inflorescences set no fruit while hand cross-pollination restored set in 7 out of 10.
Read the worked answer
What bagging does. The muslin excludes the natural pollen vectors. Durian flowers open in the evening and are pollinated mainly by nectar-feeding bats, with moths contributing, so a bagged inflorescence receives no pollen at all. No pollen means no pollen tube, no fertilisation and therefore no fruit.
Why cross-pollen is needed. Durian is largely self-incompatible: pollen tubes from the same clone are arrested in the style, so even if self-pollen landed, fertilisation would fail. The four-clone orchard 2 km away had compatible pollen available and set 28%, which already points to pollen supply rather than tree nutrition or disease.
The double fertilisation link. A successful fruit requires two fusion events: one sperm with the egg to form the zygote, and the second with the polar nuclei to form the triploid endosperm. Only then does the ovule become a seed. Developing seeds are the source of auxins and cytokinins that establish the young fruit as a sink and prevent the formation of an abscission layer. Where ovules are unfertilised, hormone levels fall and the flower is shed within days.
Why the hand pollination worked. The 7 out of 10 result shows that the flowers were receptive at 21:00, that the stigma and style were functional, and that the trees had the resources to retain fruit. It isolates the limiting factor as pollen transfer. The rain is the likely cause: it suppresses bat and moth flight, and it can wash pollen from anthers and dilute or dislodge it from stigmas. Daytime temperatures of 34 °C would also shorten stigma receptivity, narrowing the window in which pollination must occur.
2. Explain why the ethephon and gibberellic acid sprays failed to improve fruit set.
Read the worked answer
Ethephon. Ethephon breaks down to release ethylene, whose normal roles include ripening and the promotion of abscission. Sprayed at flowering, it is at best irrelevant and at worst counterproductive: it encourages the very abscission the grower wants to prevent. It does not supply pollen and cannot make an unfertilised ovule develop.
Gibberellic acid. GA can induce parthenocarpy in some species, such as grape and citrus, where it substitutes for the auxin normally supplied by seeds and allows fruit to develop without fertilisation. It cannot do that here. Durian fruit development is tied to the presence of fertilised ovules; GA cannot create an embryo or endosperm, and in a self-incompatible flower with no compatible pollen there is no seed to act as a sink. A single spray is also unlikely to reproduce the sustained, rising hormone output of developing seeds.
Timing. The GA was applied two weeks after anthesis. By then the flowers that had not been fertilised had already formed abscission layers and were being shed, so the spray was aimed at fruit that was already committed to drop.
The underlying error. Hormones were being used to treat a pollination problem, which is a misdiagnosis. The comparison with the mixed orchard, the bagging result and the hand pollination all locate the failure at pollen transfer. A fair test of GA would be to hand-pollinate all flowers, then apply GA to half and compare retention, so that pollination is held constant and only the hormone varies.
3. Design an investigation to identify the best polliniser clone and to test whether the problem is pollination or tree nutrition. Then give the practical recommendation for the orchard.
Read the worked answer
Design. Choose 40 inflorescences on 20 trees, two per tree, and allocate treatments randomly within each tree so that tree-to-tree variation is blocked rather than confounded.
Treatments: (1) open pollination, untreated control; (2) bagged, no pollen added; (3) bagged, self-pollen applied by hand; (4) bagged, pollen applied by hand from each candidate clone in turn, one treatment per clone. Bag at the bud stage, pollinate at 21:00 when stigmas are receptive, and remove bags after 48 hours.
Measurements: number of flowers per inflorescence, number retaining fruit at 2 weeks and at 6 weeks, and number of mature fruit harvested per inflorescence. Record nightly rainfall and temperature so that weather can be accounted for.
Interpretation: if all hand-pollinated treatments set far better than the bagged controls, pollen transfer is limiting. If cross-pollen sets significantly more than self-pollen, self-incompatibility is confirmed. The clone giving the highest set is the best polliniser. If even hand-pollinated flowers set poorly, the limitation lies elsewhere, for example in carbohydrate supply, water stress or boron deficiency, and leaf and soil analysis plus a thinning trial would be needed.
Recommendation. Establish polliniser rows of at least three other clones within the block, since the neighbouring mixed orchard set seven times as many fruit, and hand-pollinate the most valuable inflorescences during wet spells. Avoid ethephon at flowering and avoid evening insecticide sprays during anthesis, because the pollinators are nocturnal. Converting the result into a management trial, comparing set in blocks with and without polliniser trees over two seasons, would confirm the benefit before the whole estate is replanted.