This study evaluated the effects of tillage system, rotational tillage sequences, and winter seed burial depth on weed seedling recruitment in irrigated cropping systems. Notill (NT) planting of vegetable crops reduced emergence of hairy nightshade by 77 to 99% and Powell amaranth emergence from 50 to 87% compared to conventional tillage (CT) and planting of crops. Cover crops suppressed weed emergence if soils were undisturbed but not if soils were tilled in the spring. Four years of NT vegetable crops reduced summer annual weed density by 48 to 79% at two sites. Rotational tillage sequences that alternated between NT and CT only reduced summer annual weed density if the shorter season crop of snap beans was paired with CT and sweet corn was paired with NT. Hairy nightshade density decreased by 83 to 90% if NT was paired with the longer season crop of sweet corn. Hairy nightshade seedling recruitment at 30.7 C was more than 15 times greater for seeds buried at 6, 13 and 25 cm than when buried at 1 cm in simulated NT. Recruitment potential was low in March and April but increased to a maximum in May and June. Germination rates for seeds buried at 1 cm were lower and mortality and dormancy greater than for seeds buried from 6 to 25 cm during the winter. Protecting the seeds buried at 1 cm from rainfall during the winter increased seedling recruitment from 0 to 2 of 10 buried seeds, but had a negligible effect on seed mortality and dormancy. Treatment of seeds buried at 25 cm with 1 cm soil temperature reduced recruitment from 4.8 to 2.3 of 10 buried seeds at 33.3 C, but did not significantly increase seed mortality or dormancy. Estimates of seed drift using electronic transponders found that 16% of the seeds at 5 cm moved to within the emergence zone for hairy nightshade. Seed drift coupled with the faster germination rate and lower mortality of seeds buried at 12 cm or below predicts that hairy nightshade recruitment will be optimized when soil is tilled in the spring. / Graduation date: 2005

Solanum mauritianum (bugweed, woolly nightshade) is a perennial tree native to South America that has invaded many countries including South Africa and New Zealand. In South Africa, after 143 years of naturalization, the plant is ranked as the country‟s sixth worst weed and has invaded 1.76 million ha. Invaded areas include agricultural lands, forest plantations, water courses and conservation areas, especially in the eastern higher rainfall regions. The success of the spread of this weed is due to its production of very high numbers of bird-dispersed seeds. Since conventional control methods are unsustainable in the long term, the weed has been targeted for classical biological control since 1984. Following exploration work in its native range, biological control experts recommended that agents that are able to limit the weed‟s reproductive potential would help to manage the spread and invasiveness of this weed. Anthonomous santacruzi, a flower-feeding weevil found throughout the native range of the weed, was imported and tested between 1998 and 2002. Following approval for its release in South Africa in 2007, a new colony was imported and propagated at the University of KwaZulu-Natal Pietermaritzburg. This study was initiated to investigate aspects that could influence the release and establishment of this agent. Three aspects were investigated namely: (1) reassessing the weevil's host range to confirm that the new colony is not different from the colony tested originally and to assess the risks associated with the release of the weevil in New Zealand; (2) surveying the arthropods associated with S. mauritianum in the field to identify groups of predators that could interfere with the establishment of the weevils as well as to investigate, through laboratory-based trials using spiders as surrogate, the impact of these predators on the survival and proliferation of the weevils; and (3) propagation and release of the weevil and monitoring of its establishment. Host-specificity tests revealed that the host range of new colony is not different from that of the originally tested culture. In no-choice trials, the weevils fed and reproduced on some non-target Solanaceae species but reverted back to S. mauritianum in the choice tests. Although the risks for releasing the weevils in New Zealand were calculated to be very low, additional evidence is needed to demonstrate this conclusively. Future research to provide this evidence includes open-field trials complemented with a chemical ecology study, to resolve the case of two species, a New Zealand native and South African native, which have shown higher risks in comparison to the other tested species. For arthropods associated with S. mauritianum in the field, Araneae (especially Thomisidae), Thysanoptera, Hemiptera (especially Miridae) and Hymenoptera (especially Formicidae) were identified as generalist predators that could interfere with the establishment of A. santacruzi. However, their numbers in the field appear to be too low to provide a major threat. Also, laboratory trials using spiders as a surrogate suggested that A. santacruzi populations can survive and reproduce in the presence of such predators. The weevils were released at four sites in KwaZulu-Natal and monitoring of three of these has confirmed establishment at the warmest site along the South Coast but not at the coldest site in the Midlands. Further releases in the province are intended to complement these promising results, while additional studies are intended to facilitate the weevil's release in New Zealand. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.