| dc.description.abstract | Matooke and Mchare are two distinct subgroups of East African bananas (Musa spp.) that are endemic to the Great Lakes region of East Africa. In this region, they are pivotal in the food and income security of more than 70 million people. Productivity of these East African bananas is significantly constrained by pests and diseases that reduce yield and directly affect the livelihoods of banana farming communities in the region. While cultural methods of disease and pest control are widely applied, approaches that are tailored via genetic improvement are preferred because of their perceived long-term sustainability. However, conventional genetic improvement of bananas through hybridization is slow due to the complexity of factors that lead to sterility. Most of the popular banana cultivars are sterile and hardly produce seed after artificial and natural pollination. Therefore, the main objective of this study was to contribute towards the development of effective techniques for overcoming sterility in Matooke and Mchare bananas. The first study characterized flower opening time and behaviour in selected subgroups of bananas. A Nikon camera was set up to take pictures at 5-minute intervals through the flowering period in order to determine the optimum time of pollination. Timelapse movies were developed and played and the exact time of bract opening was recorded. The second study sought to improve pollen germination and ovule fertilization. Varying levels of glucose, fructose, sucrose, glucose plus fructose, and glucose plus fructose plus sucrose based pollen germination media were compared. Glucose at 30 g/L gave the highest pollen germination and was compared with diluted banana nectar. The effect of 30 g/L glucose pollen germination media on stigma receptivity was tested. Thirdly, seed set patterns were determined by making pollinations on East African bananas using wild banana ‘Calcutta 4’ between January 2016 and January 2019. Seed set per hand was recorded as well as fruits per hand and total hands per bunch. Weather attributes were also correlated with seed set before, during, and after pollination. Principal component analysis (PCA) was used to tease out the most influential weather attribute in regards to seed set. And lastly, determination of in vivo pollination techniques involved stigma receptivity enhancement, pollinating about a day before natural flowering opening, and evening pollination. Wild fertile banana ‘Calcutta 4’ was used as the pollen source. There were differences in the start time of opening of bracts subtending female flowers (p < 0.01) in a range of 14:54 – 20:19 hr. and with an average of 16:32 hr. Bracts subtending male flowers started opening later, between 16:53 – 01:03 hr. with an average of 18:54 hr. This implies that the pollination time between 07:00 – 10:00 hr. adopted by breeding programs globally is too late. At a concentration of 30 g/l, glucose performed better than sucrose and fructose (p < 0.001), and combinations as well as diluted banana nectar (p < 0.001) for pollen germination. Glucose at 30 g/l also increased ovule fertilization rates after its application on stigmas during pollination. Observed seed set in hands is not equal to expected seed set in different bunch size categories (p < 0.05, 0.01, 0.001). Enhancing stigma receptivity did not change seed set pattern. Weather, especially high daily temperature was significantly associated with seed set before, during, and after pollination in East African bananas. High temperature was required before and during pollination (p < 0.05 – p < 0.001, r = 0.172 – 0.488) while a low temperature was required after pollination (p < 0.05 – p < 0.001, r = -0.208 – -0.344). Use of glucose based pollen germination media significantly increased seed set (p < 0.05) in East African bananas with residual fertility but not in sterile cultivars. Pollinating in the evening did not increase seed set as ‘Calcutta 4’ flowers did not have fresh pollen. On the other hand, forcing flowers open and pollinating about a day before natural opening did not also increase seed set likely because of immature stigma tissues. In general, seed set increase was observed in the new pollination techniques of enhancing stigma receptivity. Seed set increased from 1.62 to 2.82, 14.37 to 20.11, and 1.25 to 5.02 seeds per 100 fruits in ‘Enzirabahima’, ‘Mshale’, and ‘Nshonowa’ respectively. The achieved increases are still a small fraction of potential seed set which can be as high as 30,000 seeds per 100 fruits. Stigma receptivity is not responsible for seed set pattern across hands as earlier reported. Data from this research lays a firm foundation for further research especially on overcoming sterility in bananas. | en |
