Interest in planting Jatropha curcas L. for the production of biodiesel is growing exponentially. The properties of the crop and its oil have persuaded investors to consider J. curcas oil as a substitute for fossil fuels. However, this plant is still undomesticated, basic agronomic properties are not thoroughly understood and the environmental effects on growth have not been investigated. This thesis investigated different approaches that may contribute to improving the productivity of this plant. Seed germination and methods of propagation are usually the first consideration in any plant development programme. The effects of aerosol smoke, smoke water, potassium nitrate, naphthalene acetic acid and indole-3-butyric acid on germination and seedling growth of J. curcas were investigated. Seed coat removal accelerated water imbibition and germination occurred within 48 h. Seeds exposed to aerosol smoke failed to germinate over the whole study period of three months. There were no significant differences in total germination between the treatments and the untreated control (intactand shelled-seed). However, shelled-seeds had a shorter mean germination time. The seedlings were subsequently sown in trays under shade house conditions and different seedling growth traits measured after three months. Smoke water, potassium nitrate and naphthalene acetic acid produced significantly heavier seedlings with longer stems and roots, wider stems and a higher vigour index compared to the control treatments. Smoke water, potassium nitrate and naphthalene acetic acid stimulated seedling growth and vigour of J. curcas. This opens the possibility of applying these treatments to produce quality seedlings for large scale planting and accelerated plant establishment in production orchards. Effective pollination is a prerequisite for many crops to increase seed-set and fruit production. Experiments were conducted to determine factors that could influence seed production in this potential biofuel seed crop. Controlled pollination experiments showed that plants required pollinator visits for seed production and were genetically selfx compatible. Pollen-supplementation did not lead to increased fruit set, suggesting that seed production in the study population was not pollen-limited. Both male and female flowers produced nectar and were highly attractive to honeybees. These insects were effective pollinators of J. curcas, as shown by experiments in which flowers exposed to single or multiple visits by honeybees set significantly more fruit than those from which visits were precluded. Pollinator-mediated self-pollination led to marginally lower levels of seed production relative to cross-pollination. Progeny from selfed plants had significantly shorter roots than progeny of outcrossed plants. However, in general, there was little evidence of inbreeding depression. The present results provide empirical evidence that honeybees are effective pollinators of J. curcas. Fruit arising from selfpollination were almost as numerous and as large as those arising from crosspollination, suggesting that promotion of cross-pollination does not have to be a priority in orchard management for fruit yield. Manipulation of pollen development and function is of vital importance for crop development and improvement. Experiments were conducted to investigate pollen viability, in vitro pollen germination and in vivo pollen tube growth in J. curcas. Light and fluorescence microscopy were employed to examine the different developmental stages. It was possible to determine pollen viability and distinguish between fresh and dead pollen using 2,3,5-triphenyltetrazolium chloride (TTC). Pollen germination was significantly higher in an agar-based medium composed of sucrose, boric acid and calcium nitrate compared with the control treatment (distilled water). Supplementation of IAA to the different media significantly increased pollen germination and pollen length compared with the control treatment. Pollen from hermaphrodite flowers had a lower viability, lower germination rates and shorter pollen tubes, with abnormal shapes, compared to the pollen from male flowers. Pollen tubes from both self- and crosspollinated flowers entered the ovary within 8 hours after pollination (HAP). However, at 6 HAP, the pollen tube length and growth rate were significantly higher in crosscompared to self-pollinated pollen. Our results suggest that TTC is a reliable test for pollen viability; boric acid, calcium nitrate, sucrose and addition of IAA are essential and beneficial for pollen germination in this plant. Pollen germination and pollen tube growth were not inhibited, nor interfered with, as a result of self-pollination treatments. During, both types of pollination, fertility is maintained as evidenced by ovule penetration by pollen tubes. This suggests that type of pollination has no influence on the success of fertilization in J. curcas. Manual pruning is one of the major management practices in commercial plantations of J. curcas, resulting in production of more branches and thus increased potential for more inflorescences leading to a higher seed yield. Experiments were conducted to determine the response of J. curcas plants to manual pruning under summer and winter conditions. The results showed that manual pruning under both conditions significantly increased the number of branches per plant. However, there were no significant differences in number of branches between winter and summer manual pruning. Winter pruning, however, had a significantly wider crown diameter compared to the control and summer pruning. Both treatments produced significantly less fruits/per plant in the subsequent season compared to the un-pruned control. This study revealed that winter and summer manual pruning may be suitable practice to promote branching. Manual pruning, however, is time consuming, labour intensive and expensive. A study was conducted to determine the potential of different plant growth regulators (PGRs) to increase the number of lateral branches of J. curcas plants. A single foliar application of BA (benzyladenine) at 12 mmol l-1 significantly increased branches in both the pot (4) and field (13.2) trials compared to manual pruning (MP) (1.8 and 5.7 respectively) and control (no new branches) plants. In the field, treatment with TIBA (2,3,5-triiodobenzoic acid) (1 mmol l-1) significantly increased the number of branches (15.9) after seven months from application. Of all the PGRs examined, DK (Dikegulac) (2,3:4,6-di-Oisopropylidene- 2-keto-L-gulonic acid) at 2 mmol l-1 produced the maximum number of branches (18) in the field seven months after application. Concentrations of 2 and 3 mmol I-1 of MH (Maleic hydrazide) (1,2-dihydro-3,6-pyridazinedione, coline salt) significantly increased the number of branches, four and seven months after spraying in both the pot trial in the shade house and field respectively. Under field conditions J. curcas plants responded better to all the PGRs (DK < TIBA < BA < MH) when treated once, with insignificant variations of other growth parameters. This study indicates that a single foliar application of PGRs under field conditions can be an alternative method to MP for increasing the number of lateral branches of J. curcas plants. The field chemical pruning experiment was continued to determine the potential subsequent effects of the different PGRs on seed production. In the subsequent year following the single foliar application, the parameters of flowering, fruit set, fruit characteristics, total oil content and free fatty acid (FFA) content were evaluated. Number of flowers per plant and number of fruits per bunch were significantly affected by the different treatments. However, there were no variations in the degree of fruit set. A single foliar application of BA (6-benzylaminopurine) produced more flowers per plant, more fruits per bunch, heavier and bigger fruits and seeds with more oil compared to MP (manual pruning). TIBA (2,3,5-Triiodobenzoic acid) produced significantly more flowers per plant and heavier fruits compared to the control and MP treatments. However, it produced significantly bigger fruits with more seeds and a higher oil content than MP. DK (Dikegulac) (2,3:4,6-di-O-isopropylidene-2-keto-Lgulonic acid) produced more flowers per plant and seeds with high oil content compared to the control and MP. However, it produced more fruit per bunch and more seeds per fruit compared to MP. MH (Maleic hydrazide) produced more flowers per plant, heavier and bigger fruits with numerous, heavier and oil rich seeds compared to the control and MP. This study indicates that foliar application of PGRs can be used in J. curcas to increase seed production and improve fruit quality. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.