Current trends in the fight against malaria suggest that further progress will be difficult with the use of insecticide-based control measures alone. Without major reductions in the burden of malaria registered in the past few years, the use of additional interventions with synergistic effects on the current standard measures is required. Currently, interest in employing Larval Source Management (LSM) as a complementary tool is growing as it has shown to significantly reduce larval densities and consequently adult populations in settings where it has been applied along other interventions. LSM is commonly executed via 1) habitat modification, which includes physical transformation of a larval habitat through draining, filling and land levelling and 2) larviciding, commonly using an endotoxin-producing bacterial larvicide, Bacillus thuringiensis var. israelensis (Bti). Knowledge on the ecology of anopheline larval habitats is therefore important as it informs where LSM should be targeted. Also, knowledge about community acceptance and participation in LSM is important as it affects the scalability and future sustainability of the intervention. The study described in this thesis focused on the potential of community-led LSM in Malawi. Chapter 2 describes the habitat ecology of malaria vectors in the Majete area, southern Malawi. In this area, anopheline larvae develop in habitats with little silt, surrounded by bare-grounds and occupied by culicine larvae. I conclude that larval control should be directed towards such anopheline-productive habitats which sustain malaria transmission. In Chapter 3, I investigated whether application of Bti induces discrimination of treated sites by gravid females seeking oviposition sites. I found that treatment of the sites with the bacterial larvicide does not repel ovipositing females from laying eggs in such sites. This finding implies that the female mosquitoes did not detect the presence of the larvicide in aquatic sites. In Chapter 4, we explored whether application of lower doses (sublethal) of Bti in larval habitats can negatively affect fitness parameters of malaria vectors and hence their ability to successfully transmit malaria. Sublethal Bti doses are likely to occur when applications are done under field conditions, especially by local communities who may lack the desired expertise in comparison with trained experts. Immature and adult life history parameters, including larval survival, adult longevity, wing size and oviposition of An. coluzzii, an important African malaria vector, were assessed in a laboratory setting. Our results show that larval densities were reduced when exposed to the sublethal doses. When exposed to Bti LC70 as larvae, the proportional hazard rate for mortality as adult females was about three times higher than in the control group. At the same LC70 dose rate, the mean wing length of the adult females increased by 12% compared to that of the control group. These findings are valuable as they demonstrate that larval exposure to Bti, even at lower doses, reduces the longevity of emerging adults which also reduces their vectorial capacity as they may not live long enough to effectively transmit the malaria parasite. In Chapter 5, we assessed whether communities would accept and are willing to participate in community-led LSM activities. Specifically, we explored factors that would motivate community acceptance and participation in LSM. Our results show that community involvement in LSM as an additional tool for malaria control increases local awareness of malaria as a health problem, its risk factors and control strategies. The results also show that specially trained members of the community easily accepted the intervention and were more willing to participate in the associated activities than the rest of the community. Further, the findings highlight the need to make activities less demanding in terms of time and labour. It was also observed that the community needs incentives to participate in community-led interventions but though critical, forms and modes of delivery of incentives need to be further studied. In Chapter 6, we investigated whether community involvement in LSM is feasible and can result in reduced larval vector densities. Our results showed that groups from the community, which received tailored training from the research team, participated more actively in the LSM activities than the rest of the community. Also, larviciding using Bti was the more preferred component of LSM by the community than habitat modification. Interestingly, application of Bti reduced larval densities in intervention villages. The findings of this study suggest that community involvement in LSM is only feasible when the community understands their malaria risk factors and control methods. Also, the study demonstrates that community involvement in application of Bti has the potential to reduce larval densities but should be implemented after proper training of the spraying teams. In Chapter 7, the key findings of this research and the implications for community-led LSM in Malawi are addressed and recommendations for future investigations are provided. In conclusion, the results of the research described in this thesis show that participation of communities in LSM is feasible and can reduce the malaria burden via reduced larval densities.